Early mobilisation after coronary angiography to reduce back pain

A thesis presented in partial fulfilment of the requirements for the degree of

Master of Nursing

at the

Eastern Institute of Technology

Taradale, New Zealand

Kelly Leigh Burn

2012

Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author.

ii

ABSTRACT

Background:

Coronary heart disease accounts for over 25,000 inpatient admissions and nearly

4,000 day-case admissions in New Zealand per year (Hay, 2004). Cardiac

catheterization, via the femoral artery, is a common procedure undertaken to assess

for and treat coronary heart disease (Chair, Li and Wong, 2004; Chandrasekar et al.,

2001; Wang, Redeker, Moreyra & Diamond, 2001). After the procedure, the patient

remains on bed rest (mainly supine with the affected leg straight) for at least a further

4-6 hours (Sabo, Chlan and Savik, 2008) in order to reduce the chances of

complications at the groin site (Chair, Taylor-Piliae, Lam and Chan, 2003). Due to this

enforced supine bed rest, immobilization and restricted positioning, patients frequently

experience back pain (Chair et al., 2003). Prolonged bed rest causes pressure to be

exerted continuously onto the same back muscles, causing muscle fatigue and

weakness. This fatigue causes back pain due to back spasms (Chair et al., 2004).

Objectives:

The aim of this thesis was to, via a Systematic Review and Meta-Analysis, ascertain

whether it is safe for nurses to mobilise patients out of bed four hours or earlier after a

femoral approach coronary angiogram without the use of a vascular closure device, in

order to reduce back pain whilst not increasing the risk of vascular complications at the

puncture site.

Methodology:

The research question was answered by completing a Systematic Review and Meta-

Analysis, which included fifteen studies; eleven randomised, two quasi-randomised and

two non-randomised controlled trials. The studies were collated, evaluated and

summarised using the Cochrane Collaboration’s framework. Each individual study was

also graded for quality of the evidence by allocating a level to it using the Grades of

Recommendation, Assessment, Development and Evaluation (GRADE) Working

Groups grades of evidence (Schunemann et al., 2008).

Selection criteria:

All randomised controlled trials (RCT’s), quasi-randomised controlled and non-

randomised controlled trials found in the literature search, that compared the safety of

iii

early mobilisation with vascular complications after a femoral approach diagnostic

coronary angiogram, were considered for inclusion in the review. In all, 15 studies from

around the world, published since 1996, were chosen.

Data collection and analysis:

Once studies had been obtained, they were vetted against the inclusion and exclusion

criteria for this review. The article was thoroughly scrutinised and if it was suitable went

for data collection. Three assessors decided inclusion or exclusion of each study. This

reduced the chance of selection bias in the review. Risk ratios and 95% confidence

intervals were calculated for all studies. A Meta-Analysis was then undertaken

including all of the diagnostic coronary angiogram studies, with separate analysis of

just the randomised controlled trials.

Main results:

Six studies looked at mobilisation at ≤ 2 hours after a diagnostic angiogram. There was

no significant difference overall in incidence of vascular complications in these studies

(RR 1.1591; 95% CI 0.7544-1.7809; P 0.5023). Ten studies looked at the safety of

mobilisation at ≤ 3 hours with no statistical significance in overall vascular

complications (RR 0.8430; 95% CI 0.7041-1.0094; P 0.0625). All 15 studies mobilised

their participants at ≤ 4 hours, again with no statistical significance in complications

(RR 0.8696; 95% CI 0.7399-1.0219; P 0.0891).

Authors’ conclusions:

The results from this study show no statistically significant difference in vascular

complications between the control groups and the early mobilisation out of bed groups

at ≤2, 3 or 4 hours post femoral approach coronary angiogram. Therefore, mobilisation

after coronary angiogram may be as safe at 1 ½ to 4 hours mobilisation as it is at 6

hours and may have a positive benefit of reducing back pain related to lying in bed.

Keywords:

coronary; cardiac; angiogram; angiography; catheterisation; back pain; discomfort;

mobilisation; ambulation

iv

ACKNOWLEDGEMENTS

Writing this thesis has been a true test of my willpower and motivation. Completing it

has been my life’s biggest challenge and greatest reward. This process could not have

been completed without the assistance of many people, who have guided, supported

and encouraged me.

Firstly, I would like to thank my supervisors Bob Marshall and Gill Scrymgeour for their

ongoing advice and guidance throughout the writing of the thesis.

My special thanks to my partner David as, without your love, support and expert

computer skills, this thesis would have never been written. Your patience and

encouragement whilst I completed this project has been greatly appreciated. Thanks

also to my parents Neil and Jenny who have not only been immensely supportive

throughout my entire post-graduate studies, but also provided a haven away from

home to concentrate on my thesis writing and sustenance to keep me going. Thanks

also to my sister Toni for her encouraging texts and emails.

I would like to acknowledge my Charge Nurse Manager Vicki Krog, who not only

supported me throughout the past year, but also enabled plenty of study time to write

my thesis. Also thanks to Marie Habowska for her ongoing support and encouragement

and Breen Lewis, who proved to be an excellent sounding board.

This thesis is dedicated to my nieces Lara and Allie - may it inspire you to realise that

you too can achieve whatever you set your minds to.

v

TABLE OF CONTENTS

Abstract ii

Acknowledgements iv

Table of Contents v

Figures vi

Tables vii

Appendices viii

Glossary of Terms ix

Chapter 1 - Introduction 1

Background 1

Researchers Interest 3

Research Question 4

Aims of the Research 4

Significance of this Research 4

Thesis Overview 5

Chapter 2 – Literature Review 7

Patient Factors Affecting Back Pain 7

Position Changes Whilst on Bed Rest 11

Radial Artery Approach Coronary Angiogram 13

Use of a Femoral Artery Vascular Closure Device 21

Early Mobilisation 27

Chapter 3 - Methodology 30

Evidence Based Practice 30

Evidence Based Practice and Nursing 31

Systematic Reviews and Evidence Based Practice 31

Thesis Methodology 33

Chapter 4 - Systematic Review 40

Review Protocol 98

Chapter 5 - Discussion 104

Chapter 6 - Conclusions 112

References 115

vi

FIGURES

Figure I Appendix I. Flowchart for inclusion / exclusion of studies 94

vii

TABLES

Table 1 Summary of Findings 45 Table 2 Risk of Bias in Included Studies 55

Table 3 Summary of Interventions in Early Mobilisation Trials 61

Table 4 Characteristics of Included Studies 69

Table 5 Characteristics of Excluded Studies 85

Table 6 ≤ 2 hours bed rest - total bleeding complications 86

Table 7 ≤ 2 hours bed rest - total vascular complications 87

Table 8 ≤ 3 hours bed rest - total bleeding complications 88

Table 9 ≤ 3 hours bed rest - total vascular complications 89

Table 10 ≤ 4 hours bed rest - total bleeding complications 90

Table 11 ≤ 4 hours bed rest - total vascular complications 91

Table 12 Results Randomised Controlled Trials Only 92

Table 13 Results – All Included Studies 93

Table 14 Risk of Bias Assessment Tool 101

Table 15 Data Collection Tool 101

viii

APPENDICES

Appendix I Flowchart for inclusion / exclusion of studies 94

Appendix II Search Strategies 95

Appendix III Criteria and definitions for risk of bias assessment 96

Appendix IV Systematic Review Protocol 98

Appendix V Research Notification Form 126

Appendix VI Low Risk Research Questionnaire 127

ix

GLOSSARY OF TERMS

Term Definition

Femoral Angiography Visualising the arterial system by placing a catheter into the right or left femoral artery

(Best, Pike, Grainger, Eastwood & Carroll, 2010)

Radial Approach Visualising the arterial system by placing a catheter into the radial artery (Caputo et al.,

2011).

Coronary Angiogram Injection of a radiopaque contrast media into the coronary arteries under fluoroscopy,

allowing visualisation of the coronary anatomy and pathologies such as atherosclerosis,

thrombosis and patency of any coronary artery bypass grafts (Asinas, 2010)

Manual Pressure Placing pressure with the fingers or hand on the femoral artery to maintain haemostasis

(Oxford University Press, 2012)

Mechanical Pressure Placement of a Femostop™ (or equivalent) device over a punctured artery to apply

pressure and provide haemostasis (Hoglund, Stenestrand, Todt & Johansson, 2011)

Vascular Closure Device A device deployed at the puncture site of an artery to seal the perforation. They are either

intravascular or extravascular collagen plug devices, suture based closure devices, staples

or clips (Narasimhan & Gabriel, 2011).

Back Pain Self-perceived unpleasant feeling on the rear surface of the human body from the

shoulders to the hips (Hoglund et al., 2011; Oxford University Press, 2012)

Bleeding Active bleeding from the arterial puncture site (Mah, Smith & Jensen, 1999)

Haematoma Visible and/or palpable lump under the skin which contains subcutaneous blood from the

arterial puncture site (Hoglund et al., 2011)

Pseudoaneurysm A dissection of the layers of the arterial wall, into which arterial blood enters (McCance &

Huether, 2002)

Early Mobilisation Decreasing the time a patient spends on bed rest from the traditional time a specific unit

allows (Best et al., 2010)

Systematic Review A collation of “all empirical evidence that fits pre-specified eligibility criteria in order to

answer a specific research question” (Green et al., 2008, p. 6).

Meta-Analysis The combining of statistical results from more than one study on a similar research issue,

to produce a pooled effect size and therefore to evaluate the effectiveness of the study

interventions (Acton, 2001)

Bias “A systematic error, or deviation from the truth, in results or inferences” (Higgins and

Altman, 2008, p. 188).

Evidence-based Practice The conscientious use of current best evidence from relevant, valid research to guide

practice decisions in the delivery of health care (Fineout-Overholt, Melnyk & Schultz, 2005;

Joanna Briggs Institute, 2012)

Heterogeneity Variability between studies in participants, interventions, outcomes, methods or

interventional effects which can lead to bias in the true interventional effect (Deeks, Higgins

and Altman, 2008)

Homogeneity Studies with similar participants, interventions, outcomes, methods or interventional effects

which when compared give meaningful results (Deeks et al., 2008)

1

Chapter 1

INTRODUCTION

The purpose of this thesis is to examine what factors a nurse can put in place to either

minimize or prevent back pain in adults on bed rest following a coronary angiography

procedure. The Systematic Review and Meta-Analysis incorporated in this thesis

ascertains whether it is safe for nurses to mobilise patients out of bed four hours or

earlier after a femoral approach coronary angiogram without the use of a vascular

closure device, in order to reduce back pain whilst not increasing the risk of vascular

complications at the puncture site. The information gained and presented in the

research findings will help medical professionals, working in the area of coronary

angiography, to put into place measures that may help to improve their patients

comfort post-angiogram whilst also maintaining their safety.

This chapter contains an introduction to the thesis and Systematic Review including a

background to coronary angiography and back pain. It then presents the research

question and the aim and significance of the research. Lastly, it contains an overview

of the entire thesis by chapter.

Background

Coronary heart disease

Coronary heart disease remains the leading cause of death in New Zealand at 21%,

with rates of cardiovascular disease mortality in Maori people two and a half times the

rate of other New Zealanders (Ministry of Health, 2011). It accounts for over 25,000

inpatient admissions and nearly 4,000 day-case admissions in New Zealand per year

(Hay, 2004). Although rates of death related to ischaemic heart disease are reducing

every year in New Zealand, the rate of reduction has progressively slowed, most likely

related to the increase of obesity and type 2 diabetes (Tobias, Sexton, Mann & Sharpe,

2006).

History of coronary angiography

Coronary angiography is a common procedure undertaken to assess for and treat

coronary heart disease (Chair et al., 2004; Chandrasekar et al., 2001; Wang et al.,

2001). The first cardiac catheterization was performed on a human being in 1929,

when Werner Forsmann cannulated himself and visualised his right atrium under

2

fluoroscopy. Left ventricle heart catheterisation followed in 1950, closely followed by

coronary artery catheterisation by F. Mason Sones in 1959 via the brachial route

(Bogart, 1998) and Dr Melvin Judkins in 1967 via the femoral artery approach

(Lehmann and Hotaling, 2005). Numbers of coronary angiograms increase every year

(Galli and Palatnik, 2005) and have now become one of the most frequently carried out

invasive procedures within hospitals (Leung, Hallani, Lo, Hopkins and Juergens, 2007).

Coronary angiography

Coronary angiography involves injection of a radiopaque contrast media into the

coronary arteries under fluoroscopy, allowing visualisation of the coronary anatomy

and therefore pathologies such as atherosclerosis, thrombosis and patency of any

coronary artery bypass grafts (Asinas, 2010). Trans-femoral puncture via a 5F to 8F

sheath is the most common approach, but the brachial and radial arteries can also be

used (Chair, Thompson and Li, 2007). After a diagnostic angiogram, coronary

angioplasty and stenting may be required, which can immediately relieve symptoms

and reduce the chances of a recurrence of ischaemic events in the future (Gallagher,

Trotter and Donoghue, 2010). Most patients post-angiogram are discharged within 24

hours, with an increasing trend towards same day discharge (Lauck, Johnson &

Ratner, 2005).

Post-angiography care

They are many differences in practices for care post-angiogram (Wang et al., 2001).

Arterial femoral sheaths after a diagnostic coronary angiogram are generally removed

immediately post-procedure (Lauck et al., 2005). After sheath removal, haemostasis

(the time from sheath removal to cessation of bleeding) is usually maintained with

manual compression, either with digital pressure or an adjunctive mechanical

compression device. Arterial vascular closure devices may also be used at the access

site to gain haemostasis and allow early mobilisation (Prada-Delgado et al., 2011). The

patient then remains on bed rest (mainly supine with the affected leg straight) for a

further 4-6 hours following the diagnostic procedure (Sabo et al., 2008). This is to

reduce the chances of complications at the groin site (Chair et al., 2003). These

complications can include groin or retroperitoneal haematoma, arterial bleeding or

pseudoaneurysm and may happen in up to 9% of cases (Fowlow, Price and Fung,

1995). At times this bed rest can be imposed until the following morning (Chair et al.,

2004; Koch et al., 1997).

3

Back pain post-procedure

Due to this enforced supine bed rest, immobilization and restricted positioning, patients

frequently experience back pain (Chair et al., 2003). Prolonged bed rest causes

pressure to be exerted continuously onto the same back muscles, causing muscle

fatigue and weakness. This fatigue causes back pain due to back spasms (Chair et al.,

2004). Keeling, Fisher, Haugh, Powers and Turner (2000) found in their study looking

at early mobilisation that 92% of participants required some sort of analgesia for back

pain. Wood et al. (1997) reported 42% of the patients in their control arm complained of

back pain at 4 hours and 14% in the 2-hour mobilisation group. Augustin, de Quadros

and Sarmento-Leite (2010) stated 40.8% of the patients in their sample of 347 patients

experienced back pain. From these figures, it is evident back pain is a significant issue

faced by patients on bed rest after a coronary angiogram.

Rezaei-Adaryani, Ahmadi and Asghari-Jafarabadi (2009) have suggested that bed rest

and positioning regimes after coronary angiogram are based on tradition rather than

research. The aim of this thesis and included Systematic Review is to ascertain what

factors a nurse can put in place to either minimize or prevent this back pain, focusing

mainly on early mobilisation.

Researchers Interest

As the researcher on this project, I have worked in the field of nursing for 17 years and

in the area of interventional radiology for 2½ years. I have often observed over this

time that enforced immobilisation and bed rest for my patients post-angiography leads

to discomfort and back pain. This research is important to me, as I feel there must be a

better way of doing things to improve the comfort of my patients. A comprehensive

literature search has suggested that sitting up in bed and mobilizing earlier improved

back pain with little increase in vascular complications. I will assess the literature to

determine if it is possible to make recommendations that will reduce patient’s

discomfort and back pain levels, whilst enabling their safe treatment post-angiography.

Unit’s current practice

The unit I work in at present has a policy of four hours bed rest (two hours supine with

affected leg straight then two hours sitting up 45° - 90°) post-diagnostic angiogram.

This all follows at least one hour lying flat during the actual procedure. I have tried

many different ways to relieve the back pain such as a pillow under the knees or a

4

back roll but nothing seems to work for any great length of time. Once the patient starts

mobilising out of bed, however, the back pain seems to quickly resolve.

Research Question

Is it safe for nurses to mobilise patients out of bed four hours or earlier after a femoral

approach coronary angiogram without the use of a vascular closure device, in order

to reduce back pain whilst not increasing the risk of vascular complications at the

puncture site?

Aims of the Research

The aim of this research is to perform a literature review to ascertain what factors can

be put in place by nursing staff caring for patients after a femoral approach coronary

angiogram to reduce back pain. These factors must also take into account patient

safety; thereby not increase the patient’s risk of arterial bleeding or haematoma at the

puncture site. A Systematic Review and Meta-Analysis will then be undertaken to

answer the above research question. Significance of this Research

The purpose of the literature review was to explore the available research to ascertain

what factors reduced or prevented back pain in patients who have a femoral artery

approach coronary angiogram. From this, the factor that was most effective in reducing

back pain after an angiogram was earlier mobilisation out of bed. Randomised, non-

randomised and quasi-randomised controlled trials identified in the literature search

were collated, evaluated and summarised by undertaking a Systematic Review and

Meta-Analysis using the Cochrane Collaboration framework.

Only one Meta-Analysis was located in the search for Systematic Reviews or Meta-

Analysis studies on the topic of early mobilisation post coronary angiography. Chair,

Fernandez, Lui, Lopez and Thompson (2008) carried out a Systematic Review / Meta-

Analysis entitled “The clinical effectiveness of length of bed rest for patients recovering

from trans-femoral diagnostic cardiac catheterization” (p. 352). This study looked at 18

studies involving diagnostic cardiac angiography from the period 1985 – 2007. It

involved only randomised controlled trials.

This Systematic Review aims to build on the Systematic Review by Chair et al. (2008),

including studies published after 2007. It will also contain non-randomised and quasi-

5

randomised controlled trials. A Cochrane Collaboration Systematic Review on this topic

was unable to be located.

It is already clear from the studies in the literature review that early mobilisation assists

in the reduction of back pain, whilst not putting the patient at increased risk of vascular

complications. The Systematic Review and Meta-Analysis will collate all of the

information from each study and provide recommendations as to the early mobilisation

of patients after a femoral approach coronary angiogram. It will guide nurses and

medical staff in their care of patients, to either reduce or prevent this back pain using

the evidence-based recommendations for care that will come as a result of this review.

The followers of the recommendations will have confidence, knowing they are not only

based on just one research trial but an evaluation of many.

Thesis Overview

Chapter One: Introduction

The first chapter in the thesis includes an introduction to the thesis and Systematic

Review. It provides a background to coronary angiography and back pain and presents

the research question, the aim and significance of the research.

Chapter Two: Literature Review

The second chapter contains a thorough literature review of all methods of potentially

decreasing back pain after a femoral approach diagnostic coronary angiogram. They

are discussed by their main themes including patient factors that can affect back pain,

position changes whilst on bed rest, using a radial artery approach instead of femoral,

using a femoral artery vascular closure device and early mobilisation after the coronary

angiogram. A discussion is then presented as to why early mobilisation was chosen

from these factors as the topic of the Systematic Review and Meta-Analysis.

Chapter Three: Methodology

This chapter looks at the methodology of the Systematic Review and Meta-Analysis. It

starts by introducing evidence-based practice, how evidence-based practice is

important to nursing practice and how the Systematic Review and Meta-Analysis is the

top level of evidence from which to base practice on. Next the research question is

presented, along with the methodology of this thesis – Systematic Review and Meta-

Analysis. The Cochrane Collaboration Systematic Review framework is presented

6

along with the ethical considerations required when completing a Systematic Review

as part of a nursing thesis.

Chapter Four: Systematic Review

The full Systematic Review is presented in this chapter. It follows the Cochrane

Collaboration framework and is presented as if it were a separate piece of work for

publication on the Cochrane Collaboration database. The Systematic Review Protocol,

which is the initial research proposal that would be sent to the Cochrane Collaboration,

sits in the Systematic Review as Appendix 4.

Chapter Five: Discussion

This discussion chapter ties together both the nursing thesis and the Systematic

Review. It discusses the main findings of the Systematic Review, that mobilisation after

coronary angiogram may be as safe at 1 ½ to 4 hours mobilisation as it is at 6 hours

and may have a positive benefit of reducing back pain related to lying in bed. The

chapter compares these results with the Systematic Review of Chair et al. (2008) and

the findings of other studies published on the topic.

Chapter Six: Conclusion

The final chapter provides a conclusion to the thesis and discusses how the findings of

this research could be included into the evidence-based practice of nurses and other

health professionals. It also presents suggestions for further research.

7

Chapter 2

LITERATURE REVIEW

Introduction

The previous chapter contained an introduction to the thesis and Systematic Review

including a background to coronary angiography and back pain. The research question

and the aim and significance of the research were presented, as was an overview of

the entire thesis by chapter.

Prior to carrying out the Systematic Review and Meta-Analysis, a literature search was

undertaken to research all methods of potentially decreasing back pain after a femoral

approach diagnostic coronary angiogram. This chapter presents the studies reviewed

in the literature search, discussed by their main themes including patient factors that

can affect back pain, position changes whilst on bed rest, using a radial artery

approach instead of femoral, using a femoral artery vascular closure device and early

mobilisation after the coronary angiogram.

Search Strategy

To locate studies that would provide information for the literature review, a computer

search of the databases MEDLINE, Wiley InterScience, CINAHL, PubMed, Proquest

Central, Google Scholar, Science Direct and Cochrane Database of Systematic

Reviews was carried out. Reference lists of already acquired research were also

searched. Search terms used for initially searching the databases were (coronary OR

cardiac) AND (angio* OR catheteri*) AND (radial OR femoral) AND (back pain OR

position* OR mobili*ation OR discomfort OR ambulation) OR (vascular AND closure).

Patient factors that affect back pain

It has been well documented that bed rest after coronary angiogram causes back pain

and is one of the most common complaints of patients (Hoglund et al., 2011; Koreny,

Riedmuller, Nikfardjam, Siostrzonek & Mullner, 2004; Lepper, 2004; Pollard et al.,

2003; Pornratanarangsi et al., 2010; Vlasic, 2004; Vaught & Ostrow, 2001). In a

qualitative study on patients’ responses to the angioplasty experience, patients

validated this by voicing their experiences of back pain, for example: “my back was

about ready to break” and “all I could do was cry… I wasn’t worried about my heart, just

my back” (Gulanick, Bliley, Perino & Keough, 1997, p. 28). However, not everyone gets

8

back pain and some experience more than others. If a person caring for a patient post-

angiogram is aware of the factors that could lead to an increase in back pain in

particular patients, they may be able to put interventions in place to prevent pain.

Age

Of the studies found in the literature search, one looked solely at patient factors that

may affect back pain levels when a patient is on bed rest after coronary angiogram.

Chair et al. (2004) carried out this prospective study alongside another randomised

controlled trial. It involved looking at the demographic information of 419 patients who

have undergone an angiogram and required a period of 8 – 24 hours bed rest. This

study found that younger patients experienced significantly more back pain than older

people.

Hoglund et al. (2011) found different results from Chair et al. (2004) in their study of the

early mobilisation (1.5 hours vs. 5 hours) of 104 participants and the effect of two

different timing methods on back pain levels. They found that patients over 70 years

tended to experience more back pain whilst on bed rest but this was not significant.

However, four hours after they began mobilising, when over 70 years of age, the

patients experienced back pain levels that were significantly higher statistically

(p<0.05) than the younger participants.

The age range was relatively similar between these two studies, with Chair et al. (2004)

having a mean age of 61.6 years and Hoglund et al. (2011) slightly higher at 63.7

years. The differences that were more obvious between these studies were the timing

differences of the back pain measurements and the length of time the participants

remained in bed. Hoglund et al. (2011) measured back pain at eight different time

points from immediately after coronary angiogram until a telephone interview 48 – 72

hours post discharge. Chair et al. (2004) measured back pain at six hours whilst the

participants were all still on bed rest and then the next morning when they were

mobilising out of bed. Chair et al. (2004) showed that in longer periods of bed rest (8 –

24 hours), younger people experienced higher levels of back pain and conversely, in

Hoglund et al. (2011), with shorter times to mobilisation (1.5 – 5 hours) older people

tend to experience a higher level of back pain, with this pain continuing even after a

period of mobilisation.

9

Another reason that older people could experience higher levels of back pain is that

they are more at risk of vascular complications after a femoral artery puncture

(Dumont, Keeling, Bourguignon, Sarembock & Turner, 2006; Waksman et al., 1995).

When a patient experiences a vascular complication, they are required to spend an

increased length of time on bed rest until haemostasis is reached. This can lead to an

increased risk of developing back pain (McCabe, McPherson, Lohse & Weaver, 2001).

Patient’s history of back pain

Chair et al. (2004) found that a history of back pain was not significantly related to back

pain in the six hours around bed rest and mobilisation following coronary angiography.

However, complaints of back pain the following morning from patients who had been

on bed rest for between eight and 24 hours overnight and had a history of back pain,

were significantly higher than those who did not have a history of back pain but had

experienced back pain whilst on bed rest. From the results of this study, it could be

concluded that a patient with a history of back pain should be mobilised as soon as

possible to reduce the risk of increasing back pain, whilst maintaining patient safety

from vascular complications.

Body mass index (BMI)

Hoglund et al. (2011) found that the lower the patient’s BMI, the more back pain

experienced. However, Chair et al. (2004) disagreed, finding an increased BMI

significantly increased back pain that they put down to the fact that persons who are

overweight generate a proportionally greater amount of force on their back muscles. As

stated earlier, the difference in these two studies is the length of time in bed. From the

results of Chair et al. (2004) a conclusion could drawn that the longer the bed rest time

and the higher the BMI, the more back pain may be experienced. In the shorter times

to mobilisation, back pain may be experienced more in the patients with a lower BMI

(Hoglund et al., 2011).

A person with an increased BMI has an increased chance of a vascular complication

after a femoral approach coronary angiogram (Gall, Tarique, Natarajan & Zaman,

2006; Waksman et al., 1995) potentially leading to a longer time on bed rest and

increased chance of back pain (Chair et al., 2004). Gall et al. (2006) reason that this

may be due to the difficulty in applying pressure on the insertion site in the femoral

artery post sheath removal due to the deeper location of the femoral artery pulse. Cox

et al. (2004) states that gaining femoral artery access itself is very difficult on patients

10

with an increased BMI, with potential for more than one arterial puncture leading to

difficulty in gaining haemostasis and an increased risk of vascular complication.

Gender

Fowlow et al. (1995) found that in their study of ambulation after a femoral approach

angiogram, that female patients experienced significantly higher back pain levels than

their male counterparts when mobilised at six to eight hours. Chair et al. (2004) found

no difference in the back pain levels between genders when mobilised between 8 and

24 hours.

In several studies it has also been found that females are more at risk of developing

vascular complications such as bleeding and haematoma than males (Dumont et al.,

2006; Eggebrecht et al., 2002; Gall et al., 2006; Gillane & Pollard, 2009; Mah et al.,

1999; Waksman et al., 1995) potentially leading to a longer bed rest requirement.

Hoglund et al. (2011) reason that this could be due to the smaller vasculature of

females, which can lead to multiple punctures of the artery and an increased chance of

a vascular complication.

Summary

Chair et al. (2004) suggests nurses have a key role in promoting patient comfort and,

when caring for patients post coronary angiogram, should take into account that some

patients may require more pain relief, position changes or back rubs than other people.

If health professionals caring for patients after femoral approach coronary angiogram

are aware of which patients have the potential to experience a higher level of back pain

then this can occur.

Actual information on patient factors that increase an individual’s chance of developing

back pain after coronary angiogram is somewhat limited. In this literature search, few

studies were found where specific information about patient characteristics and back

pain after coronary angiogram were collected and presented. From the literature, it was

ascertained that people caring for patients after femoral approach angiogram should

take into account that females may experience more back pain and younger people

and people with a low BMI on a shorter period of bed rest may experience a higher

level of back pain. Conversely, in units where patients remain on bed rest for an

extended period of time, older people, people with a history of back pain and patients

with an increased BMI may experience a higher level of back pain. People at higher

risk of vascular complications (older persons, people with an increased BMI and

11

females) are also at risk of higher levels of back pain due to a potentially increased

length of bed rest.

Most back pain resolves quickly on mobilisation and so the quicker the patient can be

mobilized, whilst maintaining their safety from vascular complications, the better for the

reduction of back pain (Chair et al., 2004). This can be seen in the Systematic Review

in Chapter 4.

Position changes whilst on bed rest

Changing a patient’s position in bed whilst on bed rest after a femoral approach

diagnostic coronary angiogram, whether it is rolling from side to side or back of bed

elevation to reduce or prevent back pain, is a well-researched procedure. This literature

review found five randomised controlled trials that solely looked at position changes

whilst on bed rest, without focusing on earlier mobilisation as well. Studies that also

involved early mobilisation after a femoral approach diagnostic coronary angiogram will

be presented later in the Systematic Review in Chapter 4.

Changing patients position

The study carried out by Yilmaz, Gurgun and Dramali (2006) involved a very complex

methodology. It included five different study groups with complex instructions on back

of bed elevation and alternating side to back lying after immediate sheath removal. The

control group maintained supine bed rest on their back until the next morning. Two

experimental groups had the head of the bed elevated to 30-45° (after one hour)

alternating side to back lying for 8 hours bed rest (with differing sandbag weights and

durations in each group). The other two groups remained on supine bed rest with

differing sandbag weights. Results of this study showed a significant decrease in back

pain in the two groups who were allowed to sit up and roll side to side compared to the

three groups who were not allowed to change position (p<0.001). However, the groups

on supine bed rest who had sandbags in place experienced significantly more back

pain and had no decrease in vascular complications when compared to the control

group without sand bags. Therefore, Yilmaz et al. (2006) suggests that sandbags do

not reduce vascular complications, can lead to increased back pain and therefore

should not be used.

12

Back pain was statistically significantly reduced (p<0.001) in favour of the experimental

over the control group in the study by May, Schlosser & Skytte (2008). In this study, the

femoral sheath was removed immediately, a sandbag was placed over the puncture

site, head of bed raised to 30°, with the experimental group allowed to move as freely

as the sandbag permitted. There was no significant difference in numbers of patients

experiencing bleeding or haematoma in the experimental vs. control group.

The participants in the experimental group of Chair et al. (2003) had their position

varied from supine, right side-lying and left side-lying hourly from two hours after

immediate sheath removal, whilst the control group were maintained supine on their

back for the same length of time (8 - 24 hours depending on the cardiologist). Back

pain measures at two, four and six hours and the next morning were significantly

reduced in the experimental group when compared to the control group (<0.001 at all

timeframes at and after two hours). There were no statistical differences between the

two groups in vascular complications (p=0.372).

Rezaei-Adaryani, Ahmadi, Mohamadi and Asghari-Jafarabadi (2009) ran a trial in Iran

where, in the two experimental groups, they elevated the back of the bed incrementally

by 15° every hour after immediate sheath removal until 45° was reached and then side

lying with 15° head elevation once the fifth hour was reached, with total bed rest for

eight hours. One of the experimental groups also had a thin pillow under one side,

which was changed from one side to the other every half an hour. The control group

remained supine on their back, with a sandbag in place at the femoral access point, for

eight hours. Back pain levels were significantly reduced in the experimental groups vs.

the control group after the third hour (p<0.05), without increase in vascular

complications (p=0.6 for bleeding, p=0.99 for haematoma).

The participants in the study by Pooler-Lunse, Barkman and Bock (1996) had the head

of their bed elevated to 45° once haemostasis was maintained after immediate sheath

removal. After 15 minutes at this level, the patient could raise the back of the bed

between flat and 45° as they wished. Results from this study showed a statistically

significant reduction in pain for the experimental group (p<0.02 overall, p<0.05 at three

and seven hours) over the control group who were supine for six hours and not mobile

around the bed. There were no differences between the groups in regards to vascular

complications.

13

Summary

All five studies, despite their varying methodologies, determined there was no

significant differences in bleeding rates or haematoma formation in those patients

allowed to change position in bed when compared to the patients in the control groups

who remained supine without moving privileges. It was also clearly shown in all of the

studies that there is a significant reduction in back pain when patients were allowed to

move in the bed as opposed to lying supine with no movement. This leads to the

conclusion that position change in bed, whether it is from side to side or back of bed

elevation, has a significant effect on reducing back pain after coronary angiogram with

no significant increase in vascular complications.

However, all of the studies still showed, that despite position changes within the bed,

patients in the experimental groups still had the potential to experience back pain.

There was no decrease in time to mobilisation in any of the studies meaning patients

remained dependent on nurses, required assistance with activities of daily living and

used more nursing resources (Chair et al., 2007; Wang et al., 2001). The repositioning

regimes in three of the studies were complicated and required planning and time to

remember where and when the patient should be repositioned.

It can therefore be concluded that repositioning patients in bed (using as simple a

regime as possible) may be a safe way to reduce back pain after coronary angiogram.

However, elements in the review suggest more can be done for these patients to

reduce back pain, in the form of earlier mobilisation, and this will be presented in the

Systematic Review in Chapter 4.

Radial artery approach coronary angiogram

Although the femoral artery remains the standard approach for coronary angiograms,

the radial artery approach is gaining recognition as an alternative access route (Sallam,

Al-Hadi, Rathinasekar & Chandy, 2009). When looking at preventing and reducing

back pain after a coronary angiogram, using the radial artery as the access route

makes sense. After a trans-radial approach coronary angiogram procedure, patients

may mobilise immediately potentially leading to an increased patient comfort and

satisfaction (Egred, 2011; Sallam et al., 2009) and possible earlier discharge (Agostoni

et al., 2004; Rosenstein et al., 2004). However, not every patient is suitable for a trans-

radial approach and not every operator is trained in this method or has the required

equipment available to use (Caputo et al., 2011).

14

History

The first radial artery access for a diagnostic coronary angiogram occurred in 1989,

when Lucien Campeau carried out the first radial approach coronary angiogram

followed closely in 1992 by Ferdinand Kiemeneij, who completed the first radial

approach PTCA (Agostoni et al., 2004; Caputo et al., 2011). Since then the trans-radial

approach has become more and more common due to its relatively low vascular

complication risk and potential for much earlier ambulation (From, Bell, Rihal & Gulati,

2011; Harrison & Grines, 2011). In some countries, radial access is now the dominant

access site replacing femoral access. Norway, Malaysia and Bulgaria have the highest

rates of radial access at 70-80%, with a worldwide average estimated at 20% (Caputo

et al., 2011; Harrison & Grines, 2011).

Benefits of radial artery access

The radial artery is considered to have several advantages over the femoral artery

approach (Agostoni et al., 2004; Sallam et al., 2009; Sciahbasi et al., 2011). These

include decreased vascular complications, no requirement to visualize the groin, ability

to avoid vessels damaged by peripheral vascular disease and easier access to the

artery, especially in patients with an increased BMI.

Decreased vascular complications

Vascular complication rates are much lower with trans-radial access when compared to

femoral artery access (Sallam et al., 2009; Egred, 2011). The radial artery is an easily

compressed artery, meaning that on removal of the arterial sheath, pressure can easily

be applied to the insertion point, leading to less chance of bleeding and vascular

complications (Agostoni et al., 2004; Sallam et al., 2009; Sciahbasi et al., 2011). A

study by Sallam et al. (2009) showed a vascular complication rate of 8.2% in their

femoral access group with 116 participants compared to no vascular complications in

their radial access group with 105 participants. Fewer vascular complications lead to a

decreased length of admission, with reduced hospital costs and improved clinical

outcomes (Caputo et al., 2011). The radial artery is away from other major blood

vessels and nerves and has alternative blood flow through the ulnar artery to the hand

in most cases (Caputo et al., 2011; Cevik, Cemil & Nugent, 2010; Rathore & Morris,

2008).

15

Age

As stated earlier, older people are more at risk of vascular complications after a

femoral artery puncture (Dumont et al., 2006; Waksman et al., 1995). In their study

looking at radial vs. femoral approach angiogram in 288 octogenarians, Louvard et al.

(2003) found a significant decrease in vascular complications with the radial approach

in this older population compared to femoral access. However, patients over 75 years

of age do have an increased radial access failure rate due to advanced vascular

disease, tortuosity of the subclavian artery and aorta, with potential calcification and

diffuse atherosclerosis, making catheter manipulation very difficult (Dehghani et al.,

2009).

Females

In their study on trans-radial vs. femoral artery approach coronary angiogram, Sallam

et al. (2009) found that in their predominantly Muslim population, the females much

preferred the trans-radial route, especially when menstruating, with many very reluctant

to go ahead via the trans-femoral route.

Peripheral vascular disease

Access via the femoral artery can be difficult in patients who have peripheral vascular

disease, especially in the increasingly more common aging population. Trans-radial

access reduces the chances of a vascular complication in patients with peripheral

vascular disease (Egred, 2011).

Increased BMI

As stated earlier, increased BMI has been reported to increase the chances of a

vascular complication in patients (Gall et al., 2006; Waksman et al., 1995). This may be

due to difficulty in applying pressure on the insertion site post sheath removal due to

the deeper location of the femoral artery pulse (Gall et al., 2006). In all patients,

including those with an increased BMI, the radial artery is superficial. Therefore it may

be easier to gain access via the radial route rather than femoral route in patients with

an increased BMI and also easier at the end of the procedure to gain haemostasis

(Cevik et al., 2010).

16

Limitations of radial artery access

Although the trans-radial approach is considered to have several major advantages

over the femoral artery approach (Agostoni et al., 2004; Sallam et al., 2009; Sciahbasi

et al., 2011), it also has limitations. These include a steep operator learning curve,

difficulty with radial artery anatomy, radial access complications, patients with coronary

bypass grafts and the potential requirement of an intra-aortic balloon pump, and

increased procedure length, radiation exposure and contrast usage.

Operator learning curve

Farman et al. (2011) discuss how, while the trans-radial approach for coronary

angiograms is being used more commonly around the world due to the benefits to the

patient, its acceptance amongst interventional cardiologists has been slow. The trans-

radial approach is more demanding than the trans-femoral approach and requires a

longer learning curve for the operator to gain the skills required. Accessing the radial

artery takes more time due to the technically more challenging approach (Rosenstein

et al., 2004; Sallam et al., 2009).

In the United States of America, the frequency of the trans-radial approach remains at

only 1.7% in the 2008 / 2009 year (Caputo et al., 2011), due to reasons such as a lack

of training, the steep learning curve required and the operator’s preference for the

femoral route. This also means fewer operators are able to teach the trans-radial

method (Harrison & Grines, 2011). Caputo et al. (2011) states trans-radial training

should ideally begin at the same time as the trans-femoral approach training begins so

that the interventional cardiology Fellow qualifies equally trained in both. They agree

with Harrison and Grines (2011) in that there are not enough interventional

cardiologists who are suitably qualified and skilled in radial approach angiograms to

train these Fellows.

Radial anatomy

Saito, Ikei, Hosokawa and Tanaka (1999) found in their study looking at radial artery

internal diameters vs. radial artery sheath outer diameters, that the inner diameter of

the radial artery was smaller than a 6F sheath in 14.3% of males and 27.4% of

females. Overstretching of the radial artery with the use of a ≥ 6F sheath in these

patients can lead to radial artery occlusion and artery spasm (Kanei et al., 2011).

17

In some procedures, a 7F sheath or greater may be necessary (Saito et al., 1999).

Sheath-less coronary catheters are now available, which can be used up to 8.5F (with

an outer diameter similar to a 6F sheath) in the radial artery without the requirement of

a sheath. However, these are not yet accessible in many countries including the United

States of America (Harrison & Grines, 2011).

Caputo et al. (2011) discusses how anatomical differences in some patients mean they

do not have dual circulation to the hand via a radial and ulnar artery. They state that a

negative (abnormal) Allen’s test is a contra-indication for trans-radial approach due to

the lack of arterial flow to the hand once a sheath and catheter are in the radial artery.

The Allen’s test is used to confirm anatomical suitability for a trans-radial approach.

Another test that can be used is placement of a pulse oximeter on the index finger, with

occlusion of the radial artery. A waveform suggests the patient has hand circulation via

the ulnar artery (Paul & Feeney, 2003).

In their Meta-Analysis of 12 radial vs. femoral coronary angiogram studies, Agostoni et

al. (2004) noted that all studies used the exclusion criteria of negative Allen’s test for

trans-radial approach and therefore these patients received a trans-femoral approach

angiogram. Gillane and Pollard (2009) state that between 10% and 27% of patients

may exhibit a negative Allen’s test making them ineligible for a trans-radial approach.

However, the validity of the Allen’s test prior to using radial access, and whether a

negative Allen’s test should lead to a radial approach not being used, is now being

questioned. Gilchrist (2006) and Hildick–Smith (2006) both believe that a negative

Allen’s test should not mean the radial artery is not used. Hildick-Smith (2006) explains

that many thousands of people, who have had a radial approach angiogram over the

15 years prior, would have shown false positive Allen’s tests due to incomplete

occlusion of the radial artery. Therefore, he concludes that with a 5% radial artery

occlusion expected on patients who receive a radial access coronary angiogram, there

should be a lot more examples of ischaemic hands, of which there are not. Hildick-

Smith (2006) and Gilchrist (2006) conclude that there is no evidence that a normal

Allen’s test is required for the safe undertaking of a radial access coronary angiogram.

Radial access complications

Radial artery perforation is a rare but potential risk for radial approach angiograms. It

can lead to compartment syndrome and acute hand ischaemia and has been reported

to occur in up to 1% of patients with radial access. Immediate recognition of this

18

problem and prompt action are required to limit damage to the limb (Kanei et al., 2011;

Sallam, Mehar and Al-Sekaiti, 2011).

In their study of radial vs. femoral approach coronary angiography, Sallam et al. (2009)

had a procedure failure rate of 17.1% in the radial access group due to complications

such as inability to gain access, radial artery spasm and subclavian artery tortuosity.

Caputo et al. (2011) believes that procedure failure via the radial approach should

decrease to less than 5% with a very skilled operator. From the findings of their study,

Lehmann et al. (2011) agree that experienced operators who perform more than 90%

of their angiograms via the trans-radial route have a failure rate of less than 5%.

Egred (2011) discusses how limitations and vascular access failure in the radial artery

approach can be due to anatomical variations such as tortuous arteries or stenosis, or

from persistent radial artery spasm. Kanei et al. (2011) state that 5 - 10% of cases can

experience radial artery spasm, mainly due to smaller arteries (especially in females),

multiple catheter changes and operator inexperience.

Agostoni et al. (2004) also discuss how the failure of procedures via a trans-radial

route can be due to the difficulties operators face cannulating the coronary ostia (where

the artery originates off the aorta) because it is more difficult to control and manipulate

the catheter via the radial access route. They give figures of a failure rate of the trans-

radial approach requiring crossover to a femoral approach as 1 in 14 patients and state

that all patients, despite the intended radial access route, should have a femoral

approach prepared just in case.

Patients with coronary artery bypass grafts

Visualisation of coronary artery bypass grafts is challenging from both the femoral and

radial route, due to where the coronary graft originates off the aorta and the lack of

catheter support from the aortic wall and aortic sinuses (Farooq, Mamas, Fath-

Ordoubadi & Fraser, 2011). Visualising coronary artery bypass grafts via the trans-

radial approach requires a very skilled operator, due to the increased difficulty in

gaining adequate support with the catheter via the radial artery. If an internal mammary

artery graft requires visualization, then a left sided trans-radial approach is required

(Caputo et al., 2011). The failure rate of the trans-radial approach attempting to

visualize coronary artery bypass grafts is seven times higher than with a femoral

19

approach attempting to do the same, due the technical difficulty of cannulating the

grafts (Farooq et al., 2011).

Requirement of an intra-aortic balloon pump

The trans-radial artery approach for coronary angiogram or PTCA does not allow for

the use of devices such as an intra-aortic balloon pump (IABP) (Agostoni et al., 2004;

Sallam et al., 2009). The IABP contains a balloon that is positioned in the aorta and

inflates on diastole. This increases the aortic diastolic pressure from the top of the

balloon back to the aortic valve during the diastolic period, which in turn increases

perfusion to the coronary arteries and improves myocardial oxygen delivery. On

deflation of the balloon during systole, the aortic pressure decreases and decreased

afterload occurs. With this comes decreased left ventricular workload and decreased

myocardial oxygen demands (Goldich, 2011).

Levy and Moussa (2011) and Gogo (2006) discuss how the IABP may be required in a

patient with cardiogenic shock to help maintain cardiovascular stability. It requires the

femoral artery route for insertion and a 7.5F sheath. Therefore, if a patient presents

with myocardial infarction, requiring a coronary angiogram, the risk of developing

cardiogenic shock before, during or after the procedure must be taken into account

when choosing femoral or radial access. If the patient requires an IABP, choosing the

femoral approach means access has already been gained for insertion of the pump if

required.

Increased procedure length, radiation exposure and contrast usage

Radial access for coronary angiogram often leads to an increased procedure time. This

has the potential to lead to increased radiation exposure and increased contrast dye

usage. In their study of radial vs. femoral access, Sallam et al. (2009) found that the

procedure time in the radial group was significantly longer than the femoral group (23.7

vs. 20.1 minutes). The measured radiation dose was also higher in the radial group,

because of this increased time, but not significantly so in this case. They suggested the

extra time was when gaining access, when fluoroscopy is not required. However,

Farman et al. (2011) found, looking at fluoroscopy times between radial and femoral

approach and between skill levels of operators, that the radial approach required

significantly more fluoroscopy time than the femoral. This was even more pronounced

with less skilled operators. This increased time may lead to an increased radiation

exposure, not only to the patient, but also to the operator.

20

Radiation exposure to the operator is also higher in the trans-radial approach due to

their closer proximity to the image intensifier. Therefore positioning of the arm

alongside the body and careful lead screen placement are vital to ensure the operator

is exposed to a smaller amount of radiation (Caputo et al., 2011). In their study

comparing operator radiation exposure in both radial and femoral access coronary

angiograms, Brasselet et al. (2008) found a significantly higher radiation exposure to

operators in the radial access group. They suggest weighing up the risk to the operator

with benefits to the patient when deciding which route to use, especially when a long

fluoroscopy time is expected, and always using lead screens and personal radiation

protective equipment.

An increased procedure length can also be significant in the patient experiencing a

myocardial infarction, where time is of the essence (Chodor et al., 2011). In their study

of radial artery access vs. femoral artery access in patients experiencing acute

myocardial infarction, Chodor et al. (2011) noted a significantly longer time from

hospital arrival to balloon inflation in the radial access group when compared to the

femoral access group. Overall times weren’t significantly different, which agreed with

the study by Sallam et al. (2009), where the delay was in sheath placement.

Summary

The radial access route has many advantages including its lack of vascular

complications (especially in patients with risk factors for bleeding post-procedure or

who are receiving anti-platelet medication or thrombolysis) and increased patient

comfort afterwards with the ability to mobilise straight away. However, the literature

suggests not every operator has the skills and expertise to carry out a coronary

angiogram via the radial artery. Some procedures such as visualising coronary artery

bypass grafts are difficult for even the most experienced operator when using trans-

radial access.

Not all patients are suitable for radial access. This may include patients with a negative

Allen’s test (although that concern is now disputed), physical deformities of the arms,

small arteries, those who present with actual or potential cardiogenic shock and may

require an IABP and those with a potentially long procedure ahead that may lead to a

high level of radiation exposure due to time (Brasselet et al., 2008). The longer time to

gain radial access may delay treatment in patients who present with myocardial

21

infarction, especially if access is unsuccessful and a switch to the femoral route is

necessary (Chodor et al., 2011).

The radial access route, although proven much safer, is still not without complications.

Up to 7% of radial access angiograms convert to a femoral artery approach due to

complications or difficulties (Agostoni et al., 2004), although this can drop to 5% with a

very experienced operator. Not all equipment to carry out a safe and effective

procedure via the radial route is available in every country (Harrison & Grines, 2011).

In summary, it would be of potential benefit to the patient if all coronary angiograms

could be carried out via the radial route, especially when trying to prevent or reduce

back pain post procedure. However, in up to 80% of cases worldwide (Caputo et al.,

2011), this is not done for many reasons as presented here.

Use of a femoral artery vascular closure device

Femoral artery vascular closure devices provide an alternative to manual or

mechanical methods of gaining haemostasis after a coronary angiogram (Nikolsky et

al., 2004). They were designed to avoid the need for manual compression and to

shorten bed rest times, therefore improving patient comfort and reducing back pain

(Adusumilli, Mah & Richardson, 2011; Biancari et al., 2010; Koreny et al., 2004).

Vascular closure devices were first developed in the mid-1990s, to overcome femoral

artery bleeding issues post-angiogram and angioplasty with the aim of decreasing the

rate of vascular complications (Dauerman, Applegate & Cohen, 2007; Sciahbasi et al.,

2009; Shroff, 2011). However, the rates of use have not increased as much as other

technologies such as drug-eluting stents (Dauerman et al., 2007).

There are several different types of vascular closure devices available, which can be

categorised as either intravascular or extravascular. The devices can be further

categorised into collagen plug devices, suture based closure devices and staples and

clips (Narasimhan & Gabriel, 2011).

Benefits of femoral artery vascular closure devices

There are several benefits to using vascular closure devices. These include earlier

haemostasis and earlier mobilisation, which can lead to reduced discomfort and back

pain. Vascular closure devices may also reduce vascular complications at the puncture

22

site, however there is some debate about this. Vascular complications with the use of

vascular closure device are discussed later in this section.

Earlier haemostasis

When a vascular closure device is used, the sheath is removed and the artery puncture

site immediately closed post-procedure, despite the patients’ coagulation status

(Chhatriwalla & Bhatt, 2006; Dauerman et al., 2007; Galli & Palatnik, 2005). Hamon

and Nolan (2008) state immediate removal of the sheath and use of vascular closure

devices after femoral artery access is beneficial in patients who have received

glycoprotein IIb/IIIa inhibitors and other medications that alter coagulation, without

having to wait for normalization of coagulation status.

Vascular closure devices reduce the time to haemostasis compared with manual

compression. Haemostasis should be immediate with placement of a vascular closure

device, whereas the time to haemostasis using manual or mechanical compression can

be 15 - 30 minutes after sheath removal (Dauerman et al., 2007). This also frees up a

nurse or doctor who would no longer have to apply manual compression.

Earlier mobilisation and reduced back pain

Possibly the most important benefit to the patient, after deployment of a vascular

closure device, is that they can ambulate much earlier than without the device

(Chhatriwalla & Bhatt, 2006; Galli & Palatnik, 2005). Caputo et al. (2011) and

Narasimhan and Gabriel (2011) explain how vascular closure devices have lead to

earlier ambulation and therefore improved patient comfort.

This literature search found three different studies where patients mobilised earlier after

coronary angiogram when a vascular closure device was deployed. In Sciahbasi et al.

(2009), patients in the manual compression group stayed on bed rest overnight, while

the group with vascular closure devices mobilised after three hours. Their study

involved 1492 participants who received a coronary angiogram. They compared radial

access, a femoral approach with use of vascular closure device and a femoral

approach with manual compression. When the discomfort of bed rest was measured

the day after their procedure, 55.8% of patients rated that their discomfort had been

above 5 / 10 in the manual compression group, whereas the group with vascular

closure devices rated their discomfort above 5 / 10 in 3.1% of cases for Starclose™

and 4.4% in AngioSeal™ cases. By mobilising the patients much earlier, because of

23

the vascular closure device, the patients discomfort levels were significantly decreased

(p<0.0001).

Hvelplund et al. (2011) carried out a study on 300 participants, after placement of an

AngioSeal™ device, where 144 mobilised immediately after either a coronary

angiogram (in fact walked out of the procedure room) and 156 mobilised at four hours

after bed rest with a compression bandage in place. Overall complications between the

groups were non-significant with control group 3.8% major complications versus 3.5%

in the immediate ambulation group, and 12.2% vs. 15.3% for minor complications

respectively. The authors conclude immediate ambulation after coronary angiograms is

safe with an AngioSeal™ device, and with routine use approximately 87% of patients

would be suitable for immediate ambulation.

Limitations of femoral artery vascular closure device

There are several limitations with the use of vascular closure devices. These include

contra-indications, device limitations, cost, a steep operator learning curve and

potential vascular complications with their use.

Contra-indications and device limitations

Contra-indications for the use of vascular closure devices include multiple arterial

punctures attempting to gain access (as the device will only close one of the

punctures), a femoral artery with a small diameter, femoral artery disease and sheath

insertion below the femoral artery bifurcation. Sheath insertion below the bifurcation

can occur in up to 13% of patients ruling them unsuitable for vascular closure device

placement (Dauerman et al., 2007). Applegate et al. (2010), Dehghani et al. (2010) and

Sciahbasi et al. (2009) state that in order to deploy a vascular closure device, a femoral

puncture must be more than 5mm above the femoral artery bifurcation, that the femoral

artery diameter is greater than 4mm and that there is no more than 40% calcification or

vascular stenosis in the femoral artery. If any of these occur, manual or mechanical

compression should be used. Most vascular closure devices can only be used if a ≤ 8F

sheath was used except in the case of the Perclose™ device (Azmoon et al., 2010).

Bangalore, Vidi, Liu, Shah and Resnic (2011) carried out a study deploying an extra-

vascular Starclose™ device with puncture sites at or within 3mm of the bifurcation

(which was 20% of their 1096 participants). They had non-significant numbers of

24

vascular complications when compared to access above the bifurcation with overall

vascular complication, both major and minor complications, at 1%. The authors

concluded that it was as safe to deploy a Starclose™ device at or near the bifurcation

as it was to deploy a device above the bifurcation. Therefore, it is perhaps the size of

the vessel and not the actual bifurcation itself that makes activation of the device

unsafe.

Cost

One downside to vascular closure devices is their cost (Chhatriwalla & Bhatt, 2006;

Galli & Palatnik, 2005). Dauerman et al. (2007) state the cost of the vascular closure

devices has dampened enthusiasm for their routine use. In New Zealand, the

Angioseal™ device (the vascular closure device most used in Capital and Coast Health

in Wellington) costs NZ$280 per patient (P. McKillop, personal communication, January

19, 2012).

Dauerman et al. (2007) suggests that cost effectiveness of vascular closure devices is

down to each individual interventional cardiology unit. Factors to take into account are

whether manual compression or devices such as Femostop™ to apply mechanical

compression are routinely used. If a hospital unit uses mechanical compression

routinely, which already comes at a cost, the step up to a vascular closure device use

would not be as great as those who use manual compression that doesn’t incur an

equipment cost. They suggest another factor to take into account is whether the

device, if no vascular complication occurs post-procedure, will mean a patient no

longer requires an overnight stay in hospital – another cost saving measure. Biancari et

al. (2010) agree and believe that the reduction of costs related to the use of vascular

closure devices is still to be demonstrated.

Operator learning curve

Since their introduction, most vascular closure devices have been through multiple

iterations to improve the devices. The operator learning curve to use each device is

steep (Dauerman et al. 2007), especially considering how complex placement of the

devices can be (Applegate et al., 2010). This can be seen in the study by Applegate et

al. (2006a) comparing rates of vascular complications between the first iteration of the

AngioSeal™ vascular closure device to the third iteration. The risk of a vascular

complication decreased 37%, which the authors put down to an increased operator

experience with using vascular closure devices (as well as improvements to the device

itself).

25

Malik (2008) reinforces that an adequate amount of training and management of

complications, to avoid an increase in vascular complications, must precede any new

device or iteration of a device. This training will take time, organisation and increased

cost to implement.

Vascular complications

There are significant complications that may arise with the use of vascular closure

devices that clinicians need to be aware of (Adusumilli et al., 2011). Complications after

placement of a vascular closure device can include femoral artery compromise,

laceration of the artery, bleeding, pseudoaneurysm, arteriovenous fistula, infection,

embolism and limb ischaemia (Dauerman et al., 2007).

Azmoon et al. (2010) also adds device malfunction requiring surgical intervention as a

complication. Ainslie, MacDonald and Smyth (2011) presented a case study of a 69-

year-old gentleman, who presented a month after coronary angiogram and placement

of a vascular closure device (AngioSeal™), with claudication to his right calf and

weaker pulse on his right leg. An MRI scan found a filling deficit to his right popliteal

artery. Vascular surgery found a collagen plug in his popliteal artery. When the

vascular plug was originally placed, haemostasis was not immediately gained.

Therefore, manual compression was required. The authors surmise that this manual

compression pushed the collagen plug into the artery, causing it to travel down to the

popliteal artery. Adusumilli et al. (2011) presented a very similar case in a 32-year-old

man who had the anchor, suture and collagen plug from an AngioSeal™ removed from

his right anterior tibial artery that was deployed two days earlier after a coronary

angiogram.

The impact of vascular closure devices on access site complications remains

controversial (Benninghoff, Amer & Klugherz, 2003). A review carried out by Dauerman

et al. (2007) looked at five different studies comparing manual compression to vascular

closure devices in diagnostic angiograms (Applegate et al., 2006a; Applegate et al.,

2006b; Arora, Matheny, Sepke & Resnic, 2006; Tavris et al., 2004; Tavris et al., 2005;).

Applegate et al. (2006a) showed no statistical difference in vascular complications after

diagnostic angiograms with vascular closure devices (1.2%) vs. manual compression

(1.4%). Similarly, Applegate et al. (2006b) found no differences with vascular closure

devices with respect to vascular complications (0.86%) vs. manual compression

(1.24%). Arora et al. (2006) however, did show a significant reduction in vascular

26

complications (58%) in diagnostic angiograms after vascular closure device usage vs.

manual compression. Tavris et al. (2005) had a similar finding with a statistically

significant reduction in vascular complications when a Vasoseal™ vascular closure

device was used post-diagnostic angiogram vs. manual and mechanical compression

groups (p<0.0001). An earlier study by Tavris et al. (2004) also looked at diagnostic

angiograms with vascular closure device use. Results in this study also showed a

statistical significance with a decrease in complications in diagnostic angiograms when

a suture based vascular closure device was used (p=0.0125) but not statistically

significant when a collagen plug device was used (p=0.10) when compared with

manual compression.

Sciahbasi et al. (2009) found that the radial approach had a significantly lower rate of

vascular complications (p=0.03). However, use of a femoral vascular closure device did

not significantly lower vascular complications in the femoral approach group

(Starclose™ 2.7%, Angioseal™ 3.9%) over manual compression group (2.9%) in their

study.

Applegate et al. (2010) found in their study of 575 participants after diagnostic

angiogram, that their vascular complication rate was only 0.2% after use of a vascular

closure device, with a deployment rate of 99.8%. This was using the new Evolution

AngioSeal™ that does not require the operator compaction of the anchor to collagen

sandwich, reducing operator variability as a cause of vascular complications.

In their study comparing manual compression to vascular closure devices post-

diagnostic angiogram, Lupi, Lazzero, Plebani, Sansa and Bongo (2011) found that

major vascular complication rates between the groups were not statistically significant

(1.1% in vascular closure device group vs. 0.4% in manual compression group).

However, when comparing minor complications between the groups, the vascular

closure device group had significantly more complications (5.5%) than the manual

compression group (2.7%; p=0.029).

Dauerman et al. (2007) conclude from their review that it is unsurprising that manual

compression remains the gold standard for achieving haemostasis. Studies on this

topic do not conclusively show that usage of a vascular closure device decreases the

risk of a vascular complication, and in some studies actually shows an increased risk.

Shroff (2011) agrees, stating that although vascular closure devices do reduce the time

to haemostasis and enable the patient to mobilise earlier, the vascular complication

27

rates after activation of these devices are similar to manual or mechanical compression

and in some cases even worse.

Summary

Femoral artery vascular closure devices are certainly a way to gain immediate

haemostasis and allow for earlier mobilisation (Chhatriwalla & Bhatt, 2006; Galli &

Palatnik, 2005). This earlier mobilisation has the potential benefit of improved patient

comfort and reduced back pain (Caputo et al., 2011; Narasimhan and Gabriel, 2011).

However, the devices have limitations. Not all patient’s anatomy is suitable for

placement of a device. Puncture sites can be situated where a device would not be

safe to place or can be too big for a device to safely seal the hole in the artery. The

devices are expensive, and depending on the usual practice of a unit, may not be cost

effective to use. Device operation can be difficult, with a steep operator learning curve

for placement of the device (Dauerman et al. 2007). If the operator is not sufficiently

skilled, this can lead to life or limb threatening consequences.

Many studies have been carried out on the rate of vascular complications after

placement of a vascular closure device. There is no conclusive evidence available that

says that the devices reduce complications, and in fact, some studies have shown an

increase in complications.

In summary, these devices are a good way to reduce back pain by allowing the patient

to mobilise earlier and a lot of cardiologists do use them in selected patients. However,

they cannot be used on everyone, and many operators choose not to use the devices

due to the increased costs, difficulty in using them and no clear benefit to the reduction

in vascular complications.

Early Mobilisation

Of all the factors found during the literature search, early mobilisation out of bed was

perhaps the most effective at reducing back pain due to prolonged bed rest. Back pain

severity increases the longer the patient remains on bed rest after coronary angiogram

and, conversely, the earlier the patient mobilises, the less back pain they experience

(Dabbs, Walsh, Beck, Demko & Kanaskie, 1999; Baum and Gantt, 1996; Vlasic, 2004;

Chair et al., 2003). Chair et al. (2007) recommends that to obtain optimal patient

comfort, the length of bed rest after a coronary angiogram should be minimized. It

28

seems logical that earlier ambulation could reduce some of the discomforts associated

with bed rest after a coronary angiogram, but this must be balanced against the need

for safety and prevention of vascular complications (Wang et al., 2001).

This literature review sourced 15 quantitative studies, published within the last 15

years, which directly researched early mobilisation of patients after femoral approach

diagnostic coronary angiography. The studies included 11 randomised controlled trials,

two quasi-randomised and two non-randomised trials, all with an intervention group

and control group.

All studies involved immediate sheath removal with no prerequisites except for no

bleeding or haematoma at site. Haemostasis was gained using either manual or

mechanical compression but did not include vascular closure devices. Occasionally, a

sandbag was utilized. Early mobilisation had varying time limits applied from 1.5 hours

to 4 hours compared to the control groups bed rest times of 3-24 hours. Medication use

varied between trials, but generally involved Aspirin, Heparin (and occasional reversal

with Protamine), Clopidogrel or Ticlodipine and low molecular weight Heparin. Warfarin

was usually stopped several days prior, and patients only included in trials if INR was

within normal limits.

Early mobilisation has many benefits to both the patient and unit. Patients who mobilize

earlier experience less back pain and discomfort. This literature review was specifically

looking at reduced back pain, which was evident in the four early mobilisation studies

that specifically measured and reported statistics on back pain and the six studies that

had anecdotal patient reports of decreased back pain. These results can be seen in the

Systematic Review in Chapter 4.

Early mobilisation proved to be very successful at reducing back pain after femoral

approach coronary angiogram, without increasing vascular complications, when

evaluated in the literature review. Earlier mobilisation also decreases urinary discomfort

due to the fact that using a bedpan or urine bottle whilst lying flat in bed can be difficult,

leading to urinary retention and discomfort (Chair et al., 2007). The intervention of early

mobilisation does not increase costs, and in fact could reduce costs (Best et al., 2010),

does not need any extra equipment or extra training, means an earlier regain of patient

independence and frees up nursing resources (Chair et al., 2007; Wang et al., 2001).

Koch et al. (1999) discuss how a patient who is mobilised sooner, without

complications, can be discharged earlier freeing up nursing resources and beds.

29

Because of these reasons, early mobilisation was chosen as the topic for the

Systematic Review in Chapter 4.

Conclusion

In this literature review, several different ways of reducing back pain after diagnostic

coronary angiogram were found. Knowing the patient factors that have the potential to

increase back pain, means the person caring for the patient may be able to pre-empt

and decrease the chances of that person developing back pain (perhaps prophylactic

pain relief etc.). The patient can be moved in bed or sit up in bed earlier to reduce their

back pain levels. A patient can be considered for a radial artery approach to their

coronary angiogram, meaning they can sit up and mobilise almost straight away after

the procedure. If a radial approach cannot be used, then a vascular closure device

could possibly be deployed after a femoral artery approach angiogram, to hasten

mobilisation.

The facts remain, however, that there are limitations to all of the above. The gold

standard and preferred access method around the world remains the femoral artery

(Hoglund et al., 2011) and vascular closure devices has not been widely adopted,

despite their availability (Dauerman et al., 2007). Of the five different ways to potentially

reduce back pain, only three (knowing the patient factors that have the potential to

increase back pain, moving the patient in bed or sitting them up in bed earlier and

earlier mobilisation) could be directly altered by a nurse without having to involve a

doctor. Earlier mobilisation proved to be the most successful way of the three at

reducing back pain when evaluated in the literature review. Decreasing the time to

mobilisation after femoral approach coronary angiogram, without vascular closure

device usage, seems the obvious solution to reducing back pain. But what is the

shortest time to mobilise these patients whilst maintaining their safety from vascular

complications? This is the question that the Systematic Review and Meta-Analysis in

Chapter 4 aims to answer.

This chapter has presented the studies reviewed in the literature search, discussed by

their main themes including patient factors that can affect back pain, position changes

whilst on bed rest, using a radial artery approach instead of femoral, using a femoral

artery vascular closure device and early mobilisation after the coronary angiogram. It

has shown that earlier mobilisation proved to be the most successful way that nurses

can use to reduce back pain when evaluated in the literature review. The next chapter

30

will discuss the decision to complete a Systematic Review on the topic of early

mobilisation after diagnostic coronary angiogram as the research component of this

nursing thesis. It will also discuss the Systematic Review process to be followed

(including ethical considerations).

31

Chapter 3

METHODOLOGY

The previous chapter explained how a thorough literature review revealed there were

five different ways of potentially decreasing patients back pain due to the bed rest

required after a femoral approach diagnostic coronary angiogram. As earlier

mobilisation proved to be very successful at reducing back pain after femoral approach

coronary angiogram, without increasing vascular complications, when evaluated in the

literature review, early mobilisation was chosen as the topic for further investigation

and research.

In this chapter, evidence-based practice (EBP) will be defined, the importance of the

nursing profession using EBP explained and how Systematic Reviews of randomised

controlled trials are the strongest level of evidence on which to base practice (Fineout-

Overholt et al., 2005). The decision to complete a Systematic Review on the topic of

early mobilisation after diagnostic coronary angiogram as the research component of

this nursing thesis will be discussed, an explanation given on the Systematic Review

process to be followed (including ethical considerations) and a discussion on how the

completion of this nursing based Systematic Review will be of benefit to nursing and

the evidence-based practice process.

Evidence-Based Practice

Evidence-based practice is the conscientious use of current best evidence from

relevant, valid research to guide practice decisions in the delivery of health care

(Fineout-Overholt et al., 2005; Joanna Briggs Institute, 2012). EBP involves finding,

critically appraising and applying scientific evidence to health care (Stevens, 2001)

through integration of the best available scientific evidence with experiential evidence

(Newhouse, 2007; Krugman 2012). The ultimate goal of EBP is to enhance the quality

of care provided to individuals, families, and communities and to improve overall health

outcomes. EBP also seeks to improve the performance of healthcare systems

(Newland, 2012).

Gawlinski (2008) and Cullen and Adams (2012) describe the process of evidence-

based practice, starting with the identification of a relevant clinical question that has

arisen from practice. From there, a thorough search takes place to identify the latest

research and evidence on the practice issue. The findings of this search are critiqued

32

and summarised and a proposal for change of practice is written. The evidence-based

practice change is implemented and the outcomes of the change evaluated and results

disseminated. This research process is carried out to generate a new knowledge base

with which to improve healthcare delivery, with the implementation of the empirical

evidence (Leasure, Stirlen & Thompson, 2008). Evidence-Based Practice and Nursing

Nurses play a pivotal role in the delivery of heath care forming the bulk of the clinical

health workforce (Flodgren, Rojas-Reyes, Cole & Foxcroft, 2012). They have the most

direct contact with patients and are in the best position to identify the gaps and

potential improvements in the delivery of health care (Newland, 2012).

Evidence-based practice is fundamental to the nursing profession (Newhouse, 2007).

Evidence-based nursing practice means using research and evidence to guide clinical

decision-making and not basing nursing care decisions on intuition or nursing tradition

(Leasure et al, 2008). Nursing practice based on research evidence has been shown to

improve health care quality and increase positive patient outcomes (Fink, Thompson &

Bonnes, 2005; Flodgren et al., 2012). It has been clearly shown that these outcomes

are at least 28% better when nursing care is based on rigorously designed research

than when care is based on tradition (Heater, Becker, & Olsen, 1988 as cited in

Fineout-Overholt et al., 2005). When given the necessary skills, any nurse can

participate in and be leaders for EBP initiatives (Newland, 2012).

By following the EBP process, nurses take ownership of their practices and transform

health care (Fineout-Overholt et al., 2005). By implementing EBP, nurses can be

assured that their patients are receiving safe care, there is a reduction in variations of

care from nurse to nurse or unit-to-unit, and positive patient outcomes are increased

(Fink et al., 2005; Leasure et al., 2008; Rickbeil & Simones, 2012). It has been shown

that nurses who are involved in applying evidence-based practice and practice

changes improve the quality of care. They develop professionally, think innovatively,

develop practice that is efficient, effective and improves patient outcomes, and they

become champions of change (Gawlinski and Becker 2012).

Systematic Reviews and Evidence-Based Practice

The establishment of evidence-based practice began in 1972, when a British

epidemiologist, Dr. Archie Cochrane, criticised the medical profession for “not providing

33

the public with rigorous systematic reviews of evidence from existing studies” (p. 336).

Because of Dr. Cochrane’s observations, the Cochrane Center was established in

1992, and followed a year later by the foundation of the Cochrane Collaboration

(Fineout-Overholt et al., 2005).

A hierarchy of evidence exists on which to base practice, ranked in order by the

strength of the evidence. Systematic Reviews of rigorous randomised controlled trials

sit at the top (level I evidence) and provide the strongest evidence. Level II evidence

includes at least one randomised controlled trial, level III a controlled trial with no

randomisation, a cohort or case control trial or a time series trial, down to level 4 which

includes expert opinions or descriptive studies (Leasure et al, 2008; Newhouse, 2007;

Nursing Research, 2012). Stevens (2001) describes how Systematic Reviews are

crucial in today’s paradigm of nursing following evidence-based practice. They have

been described by leaders in the field of EBP as “the most powerful and useful

evidence available” to guide practice (Stevens, 2001, p. 530).

Green et al. (2008) describe a Systematic Review as a collation of “all empirical

evidence that fits pre-specified eligibility criteria in order to answer a specific research

question” (p. 6). By using specific systematic methods with a view to minimizing bias,

trustworthy information is presented from which conclusions can be drawn and

decisions made.

A Systematic Review uses a transparent process to locate, evaluate and synthesise

the results of research using an explicitly defined process that has been prescribed in

advance. The process involves screening all available research for quality and

combining the findings of the studies. By following the Systematic Review guidelines,

the research process can be easily replicated and bias minimised (Campbell

Collaboration, 2012). Acton (2001) describe a Meta-Analysis as the combining of

statistical results from more than one study on a similar research issue, to produce a

pooled effect size and therefore to evaluate the effectiveness of the study interventions.

In simple terms, Systematic Reviews and Meta-Analysis are used to combine research

to provide important information for evidence-based practice. If the Systematic Review

or Meta-Analysis scientific process is not followed, combining the results of a number

of studies does not provide a reliable, scientifically derived evidence base for practice

(Stevens, 2001).

34

Thesis Methodology

Research Question

As mentioned in the previous chapter, a thorough literature review revealed there were

five different ways of potentially decreasing patients back pain due to the bed rest

required after a femoral approach diagnostic coronary angiogram. These were knowing

the patient factors that have the potential to increase back pain, position changes in the

bed whilst on bed rest, carrying out the angiogram through the radial artery instead of

the femoral allowing for earlier mobilisation, use of a femoral artery vascular closure

device (also allowing for earlier mobilisation), or simply allowing the patient to mobilise

earlier than the normal hospital unit protocol.

The research question asks whether it is safe for nurses to mobilise patients out of bed

four hours or earlier after a femoral approach coronary angiogram without the use of a

vascular closure device, in order to reduce back pain whilst not increasing the risk of

vascular complications at the puncture site and evolved from the literature review for

two reasons. Of the five different ways to potentially reduce back pain, only three could

be directly altered by a nurse without having to involve a Doctor. Then, because earlier

mobilisation proved to be the most successful way of the three at reducing back pain

when evaluated in the literature review, early mobilisation was chosen as the topic for

further investigation and research in the form of a Systematic Review.

Systematic Review and Meta-Analysis

As identified in Chapter 2, there were 15 studies already undertaken that directly

researched early mobilisation after femoral approach coronary angiogram. There was

one Systematic Review carried out on the topic, but it did not involve studies that were

published after 2007 (Chair et al., 2008). One of the 15 studies (Logemann et al., 1999)

identified that it included a Meta-Analysis of early mobilisation studies. However, it only

included a comparison of two studies to it’s own results and presented numbers and

percentages in a table and therefore was not a true Meta-Analysis.

Instead of carrying out further research on a topic that had already, clearly, been well

researched, a decision was made to perform a Systematic Review and Meta-Analysis.

By combining these 15 studies into a Systematic Review, the strongest and most

powerful evidence for which a nurse can base their practice on was produced. By

combining the statistical results from the studies into a Meta-Analysis, a pooled effect

35

and therefore important information for nurses and other health professionals was

found (Acton, 2001).

Systematic Review Framework

Through a literature search, several collaborations that are involved in producing

Systematic Reviews were located. The Joanna Briggs Institute, established in 1996, is

a research and development-based organisation situated at the University of Adelaide.

Although based in Australia, it involves an international collaboration of health

professionals including nursing, medical and allied health researchers. Their aim is to

provide reliable evidence with which health professionals can use to guide their clinical

decision-making. They have developed methods to appraise research, synthesise

evidence and produce Systematic Reviews and disseminate this information

internationally. They also promote evidence-based practice. This provides an

improvement of the health status of the global population, by enabling the health care

professional to deliver health care that is based on the best available evidence (Joanna

Briggs Institute, 2012).

The Cochrane Collaboration, founded in 1993, is an international organization with

over 15,000 contributors from more than 100 countries. The collaboration has the

primary aim of helping health care professionals make well-informed decisions about

care by preparing, maintaining and promoting the accessibility of Systematic Reviews.

These Systematic Reviews provide a reliable synthesis of the available evidence on a

given topic. Systematic reviews provide an accumulation of evidence, enabling the

health care professional to make care decisions considering all of the evidence on the

effect of an intervention (Cochrane Collaboration, 2008).

Two other institutions also prepare and provide a database of Systematic Reviews. The

Campbell Collaboration (2012) state they “prepare, maintain and disseminate

Systematic Reviews with a focus on education, crime and justice, and social welfare”

(para. 1). They don’t, however, have a health care focus. The Centre for Reviews and

Dissemination (2012) is part of the National Institute for Health Research in England.

They provide three databases and carry out Systematic Reviews on the effects of

health and social care interventions both in the United Kingdom and internationally.

When assessing all of the Systematic Review formats to be used as the framework for

the Systematic Review in this thesis, the Cochrane Collaboration was the organisation

36

that was found to be the most suitable due to it having a strong professional reputation,

a healthcare focus and a very clear structure to follow (outlined below).

The key characteristics of a Systematic Review using the Cochrane Collaboration

framework are a set of objectives with clearly stated eligibility for including and

excluding studies, a clear and easily duplicated methodology and a search strategy

that makes every attempt to locate all studies available to meet the inclusion criteria. All

studies must then be assessed for validity, including assessment for risk of bias. The

Review must be presented in a systematic way and the characteristics and findings of

the studies amalgamated (Green et al., 2008).

Search Strategy

The search strategy used in a Cochrane Systematic Review should be of a level that it

identifies every study that could possibly meet the study inclusion criteria (Green et al.,

2008). Lefebvre, Manheimer and Glanville (2008) discuss how the search for studies

will be determined by how the eligibility criteria are set. These criteria, in a Cochrane

Systematic review, include the types of study designs to be included, the participants to

be included, the types of interventions and control used and the types of outcomes to

be measured. The search strategy should be based around the main concepts of the

review. A typical search strategy to identify studies to be included in a Cochrane review

would include the population or health condition of interest, the intervention to be

evaluated and any filter to be added such as randomised trials.

Systematic Reviews using the Cochrane Collaboration framework typically focus

mainly on randomised controlled trials, due to the decreased chance of bias. However,

certain study designs are more suitable than others to answer specific research

questions. Also, restricting study types can lead to a smaller number of studies found,

whereas a more liberal criterion may lead to an increased number of studies but with

an increase in the chance of bias (O’Connor, Green & Higgins, 2008). A discussion on

the search strategy undertaken in this Systematic Review following the Cochrane

Systematic Review format is reported in the section entitled search methods. The

complete Systematic Review search strategy is outlined in Appendix II.

Data Extraction

Cochrane reviews need to be undertaken by more than one person. This is especially

important with the selection of studies and in the data extraction process. It is

37

suggested these reviewers work independently and then combine their data to

decrease the likelihood of errors. A third person may be required to help settle any

discrepancies between the original two reviewers (Green and Higgins, 2008).

Designing a data collection tool will be required, with ample time and thought invested

as it serves a very important function. More than one form may be required to collect

different types of data. This will need to be pilot tested using a representative sample of

the studies and then modified accordingly (Higgins and Deeks, 2008). The data

collection tool for this Systematic Review is included in Table 15 in Appendix IV (the

Systematic Review Protocol).

Risk of Bias and Quality Assessment of included studies

By rigorously assessing for risk of bias in the included studies, using the Cochrane

Collaboration Risk of Bias table (Higgins and Altman, 2008), more reliable results from

which conclusions and decisions can be made will be provided (Stevens, 2001).

Review authors will be required to make judgements about bias in studies, the overall

risk in a study and the relative importance of the different domains assessed (Higgins

and Altman, 2008).

The Cochrane Collaboration framework (Higgins and Altman, 2008) assesses six

different aspects of quantitative studies where bias can occur. Assessment of how the

participants in the study were allocated to either the intervention or control group using

a specified chance or random process (sequence generation) and that this allocation

has been concealed prior to assignment from both participants and investigators

(Allocation Sequence Concealment) is required. Evaluating whether blinding of both

participants and researchers from whether the intervention was received or not is

required, so that it can be seen that the intervention itself affected the outcome of the

study and not the knowledge of whether the intervention was received. Determining

whether a study has complete outcome data for all participants in the trial, including

those who withdrew from the study is necessary. If the data is not available, the study

should be assessed as to whether this is documented. Lastly, the study should be

assessed for selective outcome reporting. Often, studies with analyses where there are

significant differences between the interventional group and control group are more

likely to be reported than non-significant differences.

Other sources of bias were also addressed within the Systematic Review. Sensitivity

analysis involved analyzing the data as a whole and then re-analyzing only the studies

38

that were deemed to be of low risk of bias. In the case of this review, a secondary

analysis was undertaken that did not include the non-randomised or quasi-randomised

studies. Studies that show a positive outcome towards an intervention are more likely

to be published or published more quickly than a study that shows no benefit of the

interventional or a negative outcome from the intervention (Sterne, Egger & Moher,

2008). Therefore, studies were assessed for their overall result, whether it is positive,

negative or null and the time to publication from study completion was evaluated.

Duplication bias exists when a study is published more than once. Only one of the

publications was used in this review. Where there is more than one control group and

one interventional group, Sterne et al. (2008) recommends only including one

interventional group. If this does not occur, and a Meta-Analysis is undertaken, double

counting of the results of the control group can lead to bias. Funding bias was also

assessed in this review. Studies that have funding from companies who have a vested

interest in the result, such as pharmaceutical companies, can also lead to bias in the

results.

Critical Appraisal Tools

As stated earlier, a Systematic Review uses a transparent process to locate, evaluate

and synthesise the results of research using an explicitly defined process defined in

advance (Campbell Collaboration, 2012). The critical appraisal tool used as the

framework for this Systematic Review and Meta-Analysis was the Cochrane

Collaboration framework (Higgins and Green, 2008). This framework provides the

researcher with a clear guide to perform and present the review. Having a standard

review presentation format makes it easier for the reader of the review to follow the

process and results.

Assigning Levels of Evidence

As part of the Cochrane Review process, levels of quality of the evidence are allocated

to individual studies using the Grades of Recommendation, Assessment, Development

and Evaluation (GRADE) Working Groups grades of evidence. Using this tool, a

randomised controlled trial would be rated as high, a downgraded randomised trial as

moderate, a double downgraded trial or observational study as low and a triple

downgraded randomised trial or case report as very low. Further research would be

unlikely to change a reader’s confidence in the estimate effect if a trial was graded as a

high level study. Moderate quality studies would require further evidence to increase

confidence in the result of the particular trial. A reader would have little confidence in

39

the results of a study deemed to be of low or very low quality (Schunemann et al.,

2008). Each study in the Systematic Review has been through the GRADE process.

The levels of evidence of each study are presented in the Summary of Findings table

at the beginning of the Systematic Review. This is expanded on in the Quality of the

Evidence section in the Discussion of the Systematic Review.

Data Analysis and Synthesis

Meta-Analysis, combining the statistical results of at least two studies, is the most

common technique used in Cochrane Systematic Reviews. This Meta-Analysis focuses

on pair-wise comparisons such as comparing a new intervention to a control or two

new interventions. However, not all Systematic Reviews are suitable for Meta-Analysis.

If not, they will be presented as either a narrative or quantitative analysis. Deeks et al.

(2008) describe the effect, treatment effect or intervention effect of a study, as “the

contrast between the outcomes of two groups treated differently” (p. 244). However

they are analysed, synthesis of the studies should answer four questions. These

include the direction of the effect, the size of the effect, if the effect is consistent

between studies and the strength of the evidence of the effect. The first three questions

will be answered using statistics. The last question looking at the strength of the effect

will answered using judgements of the review author taking into account the study

design, risk of bias and statistical measures of uncertainty.

Studies brought together in a Systematic Review will be different in some way. This

difference is termed heterogeneity. Variability in studies can be in the participants,

interventions or outcomes (clinical heterogeneity), in the studies’ methods

(methodological heterogeneity) or in the interventional effects measured in the studies

(statistical heterogeneity). Heterogeneity, or differences between studies, can lead to

bias in the true interventional effect and therefore heterogeneity between studies must

be identified and measured. Only studies that are homogeneous (i.e. similar in their

participants, interventions, outcomes and study design) should be compared in a Meta-

Analysis, which will then provide meaningful results (Deeks et al., 2008).

Ethical Considerations

As this Systematic Review does not involve actual participant contact, ethics committee

approval was not required. However, a Research Notification Form (Appendix 5) and a

Low Risk Research Questionnaire (Appendix 6) were submitted to the Eastern Institute

40

of Technology’s Research Ethics and Approvals Committee. These forms were

accepted and the Committee agreed further ethical approval was not necessary.

Summary

This chapter defined evidence-based practice, the importance of the nursing profession

using EBP, and the decision to complete a Systematic Review on the topic of early

mobilisation after diagnostic coronary angiogram as the research component of this

nursing thesis. An explanation was given on the Systematic Review process to be

followed (including ethical considerations) and a discussion on how the completion of

this nursing based Systematic Review will be of benefit to nursing and the evidence-

based practice process was undertaken.

The next chapter includes the entire Systematic Review and Meta-Analysis, in the

format that would appear on the Cochrane Collaboration database of Systematic

Reviews. It follows the Cochrane Collaboration framework and is presented as if it were

a separate piece of work for publication on the Cochrane Collaboration database. The

Systematic Review Protocol, which is the initial research proposal that would be sent to

the Cochrane Collaboration, sits in the Systematic Review as Appendix IV.

41

Chapter 4

SYSTEMATIC REVIEW

Table of Contents

Abstract 42 Plain language summary 43 Summary of findings 45 Background 46 Objectives 47 Methods 47 Search methods for identification of studies 48 Data collection and analysis 49 Results 51 Effects of interventions 56 Discussion 62 Authors conclusions 64 Acknowledgements 65 Declarations of interest 65 Differences between protocol and review 65 References 65

Table 1 Summary of Findings 45 Table 2 Risk of Bias in Included Studies 55 Table 3 Summary of Interventions in Early Mobilisation Trials 61 Table 4 Characteristics of Included Studies 69 Table 5 Characteristics of Excluded Studies 85 Table 6 ≤ 2 hours bed rest - total bleeding complications 86 Table 7 ≤ 2 hours bed rest - total vascular complications 87 Table 8 ≤ 3 hours bed rest - total bleeding complications 88 Table 9 ≤ 3 hours bed rest - total vascular complications 89 Table 10 ≤ 4 hours bed rest - total bleeding complications 90 Table 11 ≤ 4 hours bed rest - total vascular complications 91 Table 12 Results Randomised Controlled Trials Only 92 Table 13 Results – All Included Studies 93 Table 14 Risk of Bias Assessment Tool 101 Table 15 Data Collection Tool 101

Appendix I Flowchart for Inclusion / Exclusion of Studies 94 Appendix II Search Strategies 95 Appendix III Criteria and Definitions for Risk of Bias Assessment 96 Appendix IV Systematic Review Protocol 98

42

Early mobilisation after femoral approach diagnostic coronary angiography to reduce back pain

Kelly L. Burn1, Bob Marshall2, Gill Scrymgeour2

1Interventional Cardiology, Capital and Coast Health District Health Board, Wellington,

New Zealand

2Eastern Institute of Technology, Taradale, Napier, Hawke’s Bay, New Zealand

Contact address: Kelly Burn, Interventional Cardiology, Wellington Hospital, Riddiford

Street, Newtown, Wellington, New Zealand. Kelly.Burn@ccdhb.org.nz

43

Abstract

Background:

Coronary heart disease accounts for over 25,000 inpatient admissions and nearly

4,000 day-case admissions in New Zealand per year (Hay, 2004). Cardiac

catheterization, via the femoral artery, is a common procedure undertaken to assess

for and treat coronary heart disease (Chair, Li and Wong, 2004; Chandrasekar et al.,

2001; Wang, Redeker, Moreyra & Diamond, 2001). After the procedure, the patient

remains on bed rest (mainly supine with the affected leg straight) for at least a further

4-6 hours (Sabo, Chlan and Savik, 2008) intended to reduce the chances of

complications at the groin site (Chair, Taylor-Piliae, Lam and Chan, 2003). Due to this

enforced supine bed rest, immobilization and restricted positioning, patients frequently

experience back pain (Chair et al., 2003). Prolonged bed rest causes pressure to be

exerted continuously onto the same back muscles, causing muscle fatigue and

weakness. This fatigue causes back pain due to back spasms (Chair et al., 2004).

Objectives:

To ascertain if it is safe for nurses to mobilise patients out of bed four hours or earlier

after a femoral approach coronary angiogram without the use of a vascular closure

device, in order to reduce back pain whilst not increasing the risk of vascular

complications at the puncture site.

Search strategy:

A computer search of the databases MEDLINE, Wiley InterScience, CINAHL, PubMed,

Proquest Central, Google Scholar, Science Direct and Cochrane Database of

Systematic Reviews was carried out. Reference lists of already acquired research were

also searched.

Selection criteria:

All randomised controlled trials (RCT’s), quasi-randomised and non-randomised

controlled trials found in the literature search, that compared the safety of early

mobilisation with vascular complications after femoral approach diagnostic coronary

angiogram, were considered for inclusion in the review. In all 15 studies from around

the world, published since 1996, were chosen.

44

Data collection and analysis:

Once studies had been obtained, they were vetted against the inclusion and exclusion

criteria for this review. The article was thoroughly scrutinised and if it was suitable was

included in the data collection. Three assessors evaluated the studies and reasons for

inclusion or exclusion. This reduced the chance of selection bias in the review. Risk

ratios and 95% confidence intervals were calculated for all studies.

Main results:

Six trials looked at mobilisation at 2 hours or earlier after a diagnostic angiogram.

There was no significant difference overall in incidence of bleeding, haematoma or

pseudoaneurysm in these studies, with results slightly favouring the control group (RR

1.1591; 95% CI 0.7544-1.7809; p = 0.5023). Ten studies looked at the safety of

mobilisation after ≤ 3 hours with no statistical significance in the overall figures, with

results favouring the experimental group (RR 0.8430; 95% CI 0.7041-1.0094; p =

0.0625). All fifteen studies observed mobilisation ≤ 4 hours, again with no statistical

significance in figures, again with results favouring the experimental group (RR 0.8696;

95% CI 0.7399-1.0219; p = 0.0891).

Authors’ conclusions:

The results from this study show no statistically significant difference in vascular

complications between the control and experimental groups who mobilised at less than

two, three or four hours post femoral approach coronary angiogram. Therefore,

mobilisation after coronary angiogram may be as safe at one and a half to four hours

mobilisation as it is at six hours and may have a positive benefit of reducing back pain

related to lying in bed.

Plain Language Summary

Getting out of bed earlier to reduce back pain after a procedure to look at the arteries

of the heart

A coronary angiogram is a procedure undertaken to visualise the arteries of the heart,

looking for any area of narrowing that may need treatment. It is commonly undertaken

via the femoral artery in the groin, which means the patient must remain on bed rest for

a period of time afterwards to decrease the chances of bleeding from the femoral artery.

This bed rest commonly leads to the patient experiencing back pain.

45

This review aims to find the least amount of time the patient must stay in bed whilst

keeping the risk of bleeding low. The earlier the patient gets out of bed, the less back

pain they will potentially experience. The 15 trials involved patients mobilising out of

bed between one and a half and four hours after the procedure. The review finds that it

is probably safe to get patients out of bed two hours after the procedure whilst keeping

the risk of bleeding low.

46

Table 1: Summary of findings:

Early mobilisation after femoral approach diagnostic coronary angiography to reduce back pain

Patients or population: Any adult patient undergoing a diagnostic coronary angiogram via a femoral artery approach

Settings: Cardiac catheterisation units Intervention: Earlier mobilisation after procedure than standard protocol

Comparison: Mobilisation after procedure at normal time

Outcomes Numbers (Experimental/Control)

Total Number Vascular Complications (Experimental/Control)

Risk Ratio (95% CI)

P-value

Quality of the evidence

≤2 vs. 3-6 hours bed rest 699 / 831

(6 studies)

39 / 40 1.1591

(0.7544 -1.7809)

0.5023

moderate 1, 4

≤3 vs. 4-6 hours bed rest 1910 / 2017

(10 studies)

190 / 238 0.8430

(0.7041-1.0094)

0.0625

moderate 1, 2, 4

≤4 vs. ≥6 hours bed rest 2237 / 2344

(15 studies)

239 / 288 0.8696

(0.7399-1.0219)

0.0891

moderate 1, 2, 3, 4

GRADE Working Group grades of evidence (Schunemann et al., 2008)

High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

1 Best et al. (2010) used a non-randomised controlled design and Roebuck et al. (2000) used quasi-randomised controlled design which may have increased risk of bias 2 Mah et al. (1999) used non-randomised controlled design which may have increased risk of bias 3 Wang et al. (2001) used quasi-randomised controlled design and may have increased risk of bias 4 None of the 15 included studies had complete allocation concealment or blinding which may increase bias

47

Background

Description of the condition

Coronary heart disease remains the leading cause of death in New Zealand at 21%,

with rates of cardiovascular disease mortality in Maori people two and a half times the

rate of other New Zealanders (Ministry of Health, 2011). It accounts for over 25,000

inpatient admissions and nearly 4,000 day-case admissions in New Zealand per year

(Hay, 2004). Although rates of death related to ischaemic heart disease are reducing

every year in New Zealand, the rate of reduction has progressively slowed, most likely

related to the increase of obesity and type 2 diabetes (Tobias, Sexton, Mann & Sharpe,

2006).

Description of the intervention

Coronary angiography involves injection of a radiopaque contrast media into the

coronary arteries under fluoroscopy allowing visualisation of the coronary anatomy and

therefore pathologies such as atherosclerosis, thrombosis and patency of any coronary

artery bypass grafts (Asinas, 2010). Trans-femoral puncture via a 5F to 8F sheath is

the most common approach, but the brachial and radial arteries can also be used

(Chair, Thompson and Li, 2007).

They are many differences in practices for care post-angiogram (Wang et al., 2001).

Arterial femoral sheaths post-diagnostic angiograms are generally removed

immediately post procedure (Lauck, Johnson and Ratner, 2005). After sheath removal,

haemostasis is most often maintained with manual compression using digital pressure

or an adjunctive mechanical compression device. The patient then remains on bed rest

(mainly supine with the affected leg straight) for a further 4-6 hours in a diagnostic

procedure (Sabo et al., 2008) intended to reduce the chances of vascular

complications at the groin site (Chair et al., 2003).

Why it is important to do this review

Due to this enforced supine bed rest, immobilization and restricted positioning, patients

frequently experience back pain (Chair et al., 2003). Prolonged bed rest causes

pressure to be exerted continuously onto the same back muscles, causing muscle

fatigue and weakness. This fatigue causes back pain due to back spasms (Chair et al.,

2004). It is evident back pain is a significant issue faced by patients on bed rest after a

coronary angiogram (Chair et al., 2007; Hoglund, Stenestrand, Todt & Johansson,

48

2011; Pollard et al., 2003; Wang et al., 2001; Wood et al., 1997). Rezaei-Adaryani,

Ahmadi and Asghari-Jafarabadi (2009) have suggested that bed rest and positioning

regimes after coronary angiogram are based on tradition rather than research. This

review aims to provide an evidence-based argument that time to mobilisation can be

decreased in patients post femoral approach diagnostic coronary angiogram with low

risk of vascular complications. This in turn will help to decrease or prevent the potential

for back pain occurring.

Objectives

The aim of this Systematic Review and Meta-Analysis to ascertain whether it is safe for

nurses to mobilise patients out of bed four hours or earlier after a femoral approach

coronary angiogram without the use of a vascular closure device, in order to reduce

back pain whilst not increasing the risk of vascular complications at the puncture site.

Methods

Types of studies

This review included randomised controlled trials (RCT), non-randomised controlled

trials and quasi-randomised controlled trials. There was no minimum number of trial

participants set and the study could come from any country as long as it was published

in English. There was a 15 year limit applied to the studies i.e. they needed to have

been published in 1996 or later to be included.

Types of Participants

The study included those with both male and female participants aged over 18 years.

They were of many different ethnicities and from many different countries. The

participants could include outpatients or acute inpatients.

Types of Interventions

The participants in the studies had received a femoral approach diagnostic coronary

angiogram with a sheath sized between 5F and 8F. The studies needed to test the

intervention of early mobilisation after femoral approach diagnostic coronary angiogram

whilst assessing levels of vascular complications. Secondary measurement of back

pain in each group of a trial was an advantage.

49

The study was not included if the participants received a diagnostic coronary

angiogram through an artery other than the femoral artery i.e. brachial or radial artery

or a simultaneous percutaneous coronary intervention (PCI). Studies where

participants received angiograms to other parts of the body not including the heart

where also excluded. Studies where patients were treated with a vascular closure

device were excluded.

Types of Outcome Measures

The primary outcome measures for this review were vascular complications including

bleeding, haematoma (including retroperitoneal haematoma) and pseudo-aneurysm.

For an intervention to have been successful, the rate of complications in both the

control and intervention group cannot be statistically significant or must favour the

experimental group, if there is a statistical significance.

The secondary outcome measures for this study included levels of back pain,

discomfort and patient satisfaction. For an intervention to have been successful, there

needed to be a statistically significant reduction in back pain or discomfort and a raised

level of patient satisfaction in the experimental group. Search methods for identification of studies

Electronic searches

For this study, a computer search of the databases available at the Eastern Institute of

Technology Twist Library and the Otago University Library were used. They included

MEDLINE, Wiley InterScience, CINAHL, PubMed, Proquest Central, Science Direct

and Cochrane Database of Systematic Reviews. The full search strategies for each

database are outlined in Appendix 2.

Other sources

Reference lists of already acquired research were also searched and the studies

obtained from the above databases. Several of the journals did not provide access to

older archive studies that still met my criteria and so these institutions were emailed

and provided the archived studies.

50

Data collection and analysis

Selection of studies

Through a thorough literature search, 709 studies were considered for use in this

review. After reading the abstracts of these articles, 18 were considered to be

potentially usable and the full-text articles were obtained. They were then vetted

against the inclusion and exclusion criteria for this review. Firstly, the abstract was read,

and if the study fit the criteria then the full article was sought. The article was

thoroughly scrutinised and if it was deemed suitable by all three assessors, it was

included in the data collection. If the article was not suitable, it was entered into the

Characteristics of Excluded Studies list and reasons why it was not included thoroughly

documented.

Data extraction and management

Relevant data for this review was extracted from the 15 included studies via a data

collection tool. This data included source information, the studies methodology, a

scrutinisation of the risk of bias within the study, information on patient demographics

and inclusion and exclusion criteria for participants. The interventions that each arm of

the study received and the results of these interventions were also gathered, as was

general information about the study such as conflicts of interest, ethics information,

funding sources and stated study limitations.

Risk of bias assessment

The risk of bias was assessed by all three of the review authors, with a consensus

used when disagreements occurred. The Cochrane Collaboration’s tool for assessing

bias (Higgins & Altman, 2008) was used to assess the risk of bias of the selected

randomised controlled trials. The criteria and their definitions are outlined in Appendix 3.

Each of the criteria was scored ‘yes’, ‘no’ or ‘unclear’, depending on the information

supplied in the report. The Risk of Bias summary table for this review is shown in the

Results section.

Data analysis

Separate analyses were carried out for the different time frames to mobilisation post-

angiogram comparing rates of vascular complications. Analyses were undertaken for

levels of back pain at each of these time frames, if they were measured in the studies.

51

Data synthesis

Studies were analysed for homogeneity looking at study participants, interventions and

outcome measures. Studies that had similar homogeneity were combined into the

three separate analyses based on time to mobilisation. Data on back pain measures

between groups was also analysed.

Measurements of intervention effect

For dichotomous data, the results were presented as risk ratio (RR) with 95%

confidence intervals (CI). For calculation of statistical significance, a two-tailed Fisher

Exact Test was used. In the case of a sample size too large for the Fisher test, a

Pearson Chi-Squared Test was used. Statistical significance was set at p ≤ 0.05.

Assessment of heterogeneity

Because all studies were looking at complication rates, they were considered to be

comparable and therefore did not require further assessment of heterogeneity.

Sensitivity analysis

After the initial analysis, each time frame was re-analysed including only studies

assessed as having a low risk of bias to find out if the risk of bias altered the initial

results. This meant each of the three timeframes had three separate analyses –

randomised controlled trials, non-randomised and quasi-randomised controlled trials

and all included trials.

Dealing with missing data

Two authors were contacted to provide missing data. In Mah, Smith and Jensen (1999)

it was unclear as to whether vascular complications occurred before or after

mobilisation. Jensen provided the additional information that confirmed all

complications occurred after mobilisation. The same information was missing from

Pollard et al (2003). However, they did not respond to an email sent. Mention was

made of this under Other Potential Bias in the Results section below.

Assessment of reporting bias

Publication bias: Of the 15 studies included in this systematic review, eight studies

showed a positive overall result towards early mobilisation (Bogart, Bogart, Rigden,

Jung & Liston, 1999; Chair et al., 2007; Farmanbar et al., 2008; Logemann, Luetman,

52

Kaliebe, Olson & Murdock, 1999; Mah et al., 1999; Roebuck, Jessop, Turner & Caplin,

2000; Singh, Kuganesan, Goode & Ricci, 1998; Wang et al., 2001; Wood et al., 1997)

and six studies showed a negative overall result towards early mobilisation (Baum and

Gantt, 1996; Best, Pike, Grainger, Eastwood & Carroll, 2010; Hoglund et al., 2011;

Keeling, Taylor, Nordt, Powers & Fisher, 1996; Lim et al., 1997; Pollard et al., 2003).

However, only one study showed statistical significance with a two-tailed Fishers Exact

Test of p = 0.00002, which was in favour of early mobilisation when the p-value of ≤

0.05 is used as the level of significance (Mah et al., 1999).

Time lag bias: Three of the 15 studies mentioned the study duration (Farmanbar et al.,

2008; Hoglund et al., 2011; Singh et al., 1998). The time of data collection from these

studies ranged from 6 months to one year. Each was published between one and two

years after data collection finished. The other 12 studies did not mention times, so time

lag bias could not be assessed.

Duplication bias: Only one study was found to have two publications of the same study.

In the case of Chair et al. (2007), her full Doctoral Thesis was published earlier as

Chair (2004). For the purpose of this systematic review, Chair et al. (2007) was used.

Studies with more than two interventional groups: Singh et al. (1998) contained a

control group (mobilisation at 6 hours) and two experimental groups (earlier

mobilisation at 3 hours and 4 hours). To decrease risk of bias due to double counting of

the control group figures in the overall Systematic Review, only the 3-hour group and

the comparison 6-hour control group were used in the analysis.

Funding bias: Of the 15 studies, five declared their funding sources (Best et al., 2010;

Logemann et al., 1999; Pollard et al., 2003; Singh et al., 1998; Wang et al., 2001).

None of the studies would be assessed as having funding bias, as all five involved

funding from research foundations, mainly from within the hospital the research was

set. The other ten studies either did not receive any funding or did not declare it.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies.

A total of 15 studies were identified that met the inclusion criteria. A further three were

excluded for reasons documented in the Characteristics of Excluded Studies table. The

53

included studies contained a total of 4581 participants, 2237 in experimental groups

and 2344 in control groups. The studies were published from 1996 – 2011.

Of these 15 studies, 11 were randomised controlled trials and four were quasi-

randomised or non-randomised with a control group. The Characteristics of Included

Studies table shows the summary of the methods, participants, interventions,

outcomes and risk of bias of each included study (Baum & Gantt, 1996; Best et al.,

2010; Bogart et al., 1999; Chair et al., 2007; Farmanbar et al., 2008; Hoglund et al.,

2011; Keeling et al., 1996; Lim et al., 1997; Logemann et al., 1999; Mah et al., 1999;

Pollard et al., 2003; Roebuck et al., 2000; Singh et al., 1998; Wood et al., 1997; Wang

et al., 2001). Inclusion criteria for all studies consisted of adult participants (both male

and female) who were to undergo diagnostic coronary angiogram via a femoral

approach. The extensive exclusion criteria from each study can be viewed in The

Characteristics of Included Studies table.

The studies reported on five different times to mobilisation in the experimental groups

that were condensed into three separate time analyses. These time analyses were ≤ 2

hours, ≤ 3 hours and ≤ 4 hours. The control groups mobilised at a time frame greater

than the experimental group (between 3 and 6 hours).

Four studies directly measured back pain in their study groups (Chair et al., 2007;

Hoglund et al., 2011; Wang et al., 2001; Wood et al., 1997). Six other studies

mentioned feedback from participants in groups with a shorter time to mobilisation,

about level of back pain, comfort or satisfaction, without directly measuring it (Baum

and Gantt, 1996; Best et al., 2010; Logemann et al., 1999; Mah et al., 1999; Pollard et

al., 2003; Roebuck et al., 2000). Logemann et al. (1999) gave participants a

satisfaction survey that asked patients if they experienced any pain and whether they

were satisfied with the procedure. It did not, however, measure solely back pain.

Roebuck et al. (2000) did not measure back pain specifically but randomly pulled

charts from 100 participants (50 per group) and compared intravenous analgesia rates

between the groups.

Seven studies were undertaken in the United States of America, three in Canada, two

in the United Kingdom and one each in Hong Kong, Sweden and Iran. All studies took

place in just one facility. Study length varied throughout the trials, from between 3

months and 24 months. Wang et al. (2001) had the lowest number of total participants

at 82 amongst the studies, with Mah et al. (1999) the highest with 880 total participants.

54

The participants in all studies included both males and females aged over 18 years.

Risk of bias in included studies

Details of the results of the risk of bias assessment of each individual included study

are shown in Table 2.

Randomisation - Of the 15 included trials, 11 used a methodology of a randomised

controlled trial. Only two of these described their randomisation process. Chair et al.

(2007) used a computer randomisation and Wood et al. (1997) used a dice throw. The

other nine randomised controlled trials did not mention by what method they achieved

randomisation (Baum et al., 1996; Bogart et al., 1999; Farmanbar et al., 2008; Hoglund

et al., 2011; Keeling et al., 1996; Lim et al., 1997; Logemann et al., 1999; Pollard et al.,

2003; Singh et al., 1998). There were two non-randomised controlled trials where the

control group was decided by a retrospective analysis of data, prior to a change in

protocol (Best et al., 2010; Mah et al., 1999). The experimental group was allocated the

new protocol for a similar length of time as the retrospective analysis. The two other

studies used a quasi-randomised method where the participants were allocated week

about to the experimental or control group (Roebuck et al., 2000; Wang et al., 2001).

Allocation – Two of the trials (Lim et al., 1997, Roebuck et al., 2000) made a statement

about allocation sequence concealment. In their studies, only the operator performing

the procedure was blinded from allocation. Wang et al. (2001) stated they blinded

participants until after the coronary angiogram had finished. None of the other 12 trials

stated whether they had allocation concealment.

Blinding – It would be difficult with the design of these studies to blind the participants

and personnel directly involved in the participants care from sheath removal until

discharge. However, medical staff performing the procedure and outcome assessors

could be blinded to reduce study bias. Pollard et al. (2003) declared they used an open

trial design with no blinding. Singh et al. (1998) mentioned they used a single blinding

method but not who was actually blinded. None of the other 14 studies mentioned

blinding of study participants, medical and nursing staff or outcome assessors.

Incomplete outcome data – All studies presented outcomes for all of the participants in

their trials. All trials presented figures for all patients at discharge. Not all trials followed

up patients after discharge. Two trials that had a loss to follow up after discharge

55

declared this (Logemann et al., 1999; Mah et al., 1999).

Other potential bias – It was unclear as to whether vascular complications occurred

before or after mobilisation in Pollard et al (2003). However, figures were given for the

same time periods for both the control and experimental group. There was no other

potential bias determined in any of the studies. In each study, the participants

remained in the groups to which they were allocated, the reports were free of selective

outcome reporting and no other risk for bias was established.

56

Table 2. Review authors’ judgments for each risk of bias item for each study.

Stu

dy ID

Stu

dy d

esig

n

Was

the

allo

catio

n se

quen

ce ra

ndom

ly

gene

rate

d?

Was

allo

catio

n ad

equa

tely

con

ceal

ed fr

om

both

par

ticip

ants

and

clin

icia

ns b

y ha

ving

a

secu

re s

ched

ule

of ra

ndom

izat

ion?

Wer

e pa

rtici

pant

s, c

linic

ians

and

out

com

e as

sess

ors

blin

ded

to th

e al

loca

ted

inte

rven

tion

adeq

uate

ly d

urin

g th

e st

udy?

Are

resu

lts re

porte

d fo

r eve

ryon

e w

ho

ente

red

the

trial

?

Was

inco

mpl

ete

outc

ome

data

ade

quat

ely

addr

esse

d?

Wer

e pa

tient

s an

alys

ed in

the

grou

ps to

w

hich

they

wer

e ra

ndom

ised

?

Are

repo

rts o

f the

stu

dy fr

ee o

f sel

ectiv

e ou

tcom

e re

porti

ng?

Was

the

stud

y ap

pare

ntly

free

of o

ther

pr

oble

ms

that

cou

ld p

ut it

at a

hig

h ris

k of

bi

as?

Baum 1996 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Best 2010 N-RCT ✗ ? ? ✔ ✔ ✔ ✔ ✔

Bogart 1999 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Chair 2007 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Farmanbar 2008 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Hoglund 2011 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Keeling 1996 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

Lim 1997 RCT ✔ ? ✗ ✔ ✔ ✔ ✔ ✔

Logemann 1999 RCT ✔ ? ✔ ✔ ✔ ✔ ✔ ✔

Mah 1999 N-RCT ✗ ? ? ✔ ✔ ✔ ✔ ✔

Pollard 2003 RCT ✔ ✗ ✗ ✔ ✔ ✔ ✔ ✔

Roebuck 2000 Q-R ? ? ✗ ✔ ✔ ✔ ✔ ✔

Singh 1998 RCT ✔ ? ✗ ✔ ✔ ✔ ✔ ✔

Wang 2001 Q-R ? ✗ ? ✔ ✔ ✔ ✔ ✔

Wood 1997 RCT ✔ ? ? ✔ ✔ ✔ ✔ ✔

✔ - Yes ✗ - No ? – Unclear (Explained in Appendix 3. Criteria and definitions for risk of bias assessment)

RCT – Randomised controlled trial N-RCT – Non-randomised controlled trial Q-R - Quasi-randomised controlled trial

57

Effects of interventions

≤ 2 hours bed rest post-diagnostic coronary angiogram

Six trials with 1530 total participants (699 experimental / 831 control) reported results

on mobilisation out of bed at or before 2 hours post femoral approach diagnostic

coronary angiography. Measures of bleeding, haematoma and pseudo-aneurysm were

presented and analysed. There were 78 incidences of bleeding or haematoma overall,

with 39 in both the experimental groups and control groups (RR 1.1888; 95% CI

0.7716-1.8318; p = 0.4310). There was only one pseudoaneurysm in any of the studies

in this time frame, occurring in the control group.

Overall, vascular complications occurred in 79 participants with 39 in the experimental

groups mobilised at or before 2 hours and 40 occurring in the control groups mobilised

between three and six hours post-procedure. These results favoured the control group

but not with any statistical significance (RR 1.1591; 95% CI 0.7544-1.7809; p = 0.5023).

Four of the trials used a randomised controlled trial methodology. Overall results of

these studies, when separated from the one trial with a quasi-randomised controlled

methodology and one with a non-randomised controlled trial methodology, showed

similar figures favouring the control group but again with no statistical significance (RR

1.0746; 95% CI 0.6073-1.9012; p = 0.8763).

Of the six studies, three showed total vascular complication rates that favoured the

control group and three the earlier mobilisation experimental group. However, none of

the overall vascular complication rates in any of the studies showed a statistical

significance.

≤ 3 hours bed rest post-diagnostic coronary angiogram

Out of the total 15 studies included in the review, 10 mobilised their participants out of

bed at or before three hours. These studies included a total of 3927 participants (1910

experimental / 2017 control). Again, measures of bleeding, haematoma and pseudo-

aneurysm were presented and analysed. There were 189 incidences of bleeding or

haematoma in the experimental groups mobilising at or before three hours and 234 in

the control groups mobilising after three hours (RR 0.8529; 95% CI 0.7116-1.0224; p =

0.0848). There was one pseudoaneurysm in the experimental groups and four in the

control groups.

58

Overall, vascular complications occurred in 428 participants. This included 190

mobilised at or before 3 hours and 238 in the control groups. The overall results

favoured the experimental group, although the figures were not quite statistically

significant (RR 0.8430; 95% CI 0.7041-1.0094; p = 0.0625).

Seven of the trials used a randomised controlled trial methodology. Overall results of

these studies, when separated from the one trial with a quasi-randomised controlled

methodology and two trials with a non-randomised controlled trial methodology,

showed figures that slightly favoured the control group but again with no statistical

significance (RR 1.0023; 95% CI 0.7818-1.2849; p = 1.000).

Of the ten studies, six showed total vascular complication rates that favoured the

earlier mobilisation experimental group and four the control group. One study showed

an overall statistical significance. Mah et al. (1999) showed, in their non-randomised

controlled trial, results that significantly favoured the experimental group who mobilised

out of bed at three hours (RR 0.5332; 95% CI 0.3961-0.7177; p = 0.00002). None of

the overall vascular complication rates in any of the other studies showed a statistical

significance.

≤4 hours bed rest post-diagnostic coronary angiogram

All of the 15 studies included in this review mobilised the participants in their

experimental groups out of bed at or before four hours. This included a total of 4281

participants (2237 experimental / 2344 control). There were 238 incidences of bleeding

or haematoma in the experimental group and one pseudoaneurysm. There were 282

incidences of bleeding or haematoma and six pseudoaneurysms in the control groups.

Overall bleeding results favoured the experimental groups but with no statistical

significance (RR 0.8843; 95% CI 0.7517-1.0404; p=0.1380).

Overall at this time frame, vascular complications occurred in 527 participants (239

experimental / 288 control). The overall vascular complication results favoured the

experimental group, although the figures were again not quite statistically significant

(RR 0.8696; 95% CI 0.7399-1.0219; p = 0.0891).

Eleven of the trials used a randomised controlled trial methodology. Overall results of

these studies, when separated from the two trials with a quasi-randomised controlled

methodology and the two trials with a non-randomised controlled trial methodology,

59

showed figures that slightly favoured the control group but again with no statistical

significance (RR 1.0361; 95% CI 0.8440-1.2719; p = 0.7401).

Within the fifteen studies included in the review, nine showed overall vascular

complication rates that favoured the early mobilisation experimental groups and six

showed results that favoured the control groups. Comparing mobilisation out of bed at

three hours versus five hours, Mah et al. (1999) was the only study to display statistical

significance in their results. As stated above, their non-randomised controlled trial

displayed results that significantly favoured the experimental group (RR 0.5332; 95%

CI 0.3961-0.7177; p = 0.00002). None of the overall vascular complication rates in any

of the other fifteen included studies showed a statistical significance.

Measured back pain

Four studies directly measured back pain in both the experimental and control groups

(Chair et al., 2007; Hoglund et al., 2011; Wang et al., 2001; Wood et al., 1997). There

was no standardized timing to gathering this information between the studies, so a

comparative analysis could not be done. However, tools used to measure the back

pain were similar. Chair et al. (2007) used a visual analogue scale 100mm long, left

anchor at 0mm no pain. A numeric rating scale of 0 – 10 (0 meant no pain or

discomfort or least possible satisfaction) was used by Hoglund et al. (2011) and Wang

et al. (2001). Wood et al. (1997) simply asked patients if they had back pain or leg

stiffness or not just prior to mobilisation.

Wood et al. (1997) found that 14% of participants complained of back pain or leg

stiffness in their early mobilisation group at 2 ½ hours compared to their 4 hour

mobilisation group with 42% of participants experiencing pain or stiffness. Wang et al.

(2001) however, got the opposite results. Their early mobilisation group of 4 hours

experienced more back pain than their 6-hour control group (average back pain 3.29

experimental group vs. 2.85 control group). This result may be the due to the inclusion

of three patients with chronic back pain in their experimental group who rated their

back pain 9 – 10, pushing the average figure of the experimental group much higher.

Hoglund et al. (2011) measured back pain at eight different intervals from sheath

removal to follow up 48 hours after discharge. At the end of bed rest, ten patients in the

experimental group rated their pain 1 – 4 and one patient 5 – 6. In the control group,

twenty patients rated their pain 1 – 4; seven patients rated 5 – 6 and four patients 7 – 8.

60

The study authors summarised their findings by stating that their early mobilisation

group at 1.5 hours experienced significantly less back pain at the end of bed rest with a

mean score of 1.5 / 10 compared to their control group who mobilised at five hours and

had a mean back pain score of 3.5 / 10. The result was statistically significant at

p<0.001, favouring the early mobilisation group.

Chair et al. (2007) measured back pain at four hours, eight hours and the next morning.

The average back pain level at four hours in the experimental group who mobilised at

this four-hour time was 0.97, whereas the control group who remained in bed for 12

hours had an average pain level of 1.55 at four hours. At eight hours, when the

experimental group was mobilising and the control group still on bed rest, the results

were 1.34 vs. 4.41 respectively. In the morning, when they were both mobilising,

results were 1.77 experimental vs. 4.01 control. The result between the groups at all

three time frames were statistically significant at p < 0.001, favouring the early

mobilisation group.

Six studies didn’t directly measure back pain, but did mention back pain or pain. Baum

and Gantt (1996) reported that participants in their early mobilisation group of two

hours, who had previously had an angiogram and mobilised at six hours, gave positive

feedback citing less discomfort. The nurses involved in the study of Best et al. (2010)

noted fewer reports of patient discomfort in their early mobilisation group of one and a

half hours, as did the nurses in the study by Roebuck et al. (2000) whose experimental

group mobilised at two hours.

Roebuck et al. (2000) also pulled 100 charts (50 per group) in their study and looked at

intravenous pain relief use. The observations about increased comfort by the nurses

where reinforced by figures of 2% intravenous pain relief in the control group versus

none in the experimental group. The limitation of this observation, however, is that the

pain relief was not necessarily for back pain only.

Logemann et al. (1999) noted lower levels of overall pain and higher levels of overall

satisfaction in participants in their experimental group of mobilisation at two hours

taken from the Patient Satisfaction Questionnaire their participants completed. There

were similar findings in the study by Pollard et al (2003) who stated that at all times

measured on the McGill pain questionnaire, the 2.5 hour early mobilisation group

experienced less pain and discomfort. Nurses caring for patients in the study by Mah et

al. (1999), with their two and a half to three hour ambulation experimental group, noted

61

that patients made positive comments about comfort levels in early ambulation group,

experiencing less discomfort than the control group.

Sensitivity analysis

All 15 studies were deemed to be at a low risk of bias. However, as Best et al. (2010),

Mah et al. (1999), Roebuck et al. (2000) and Wang et al. (2001) used a quasi-

randomised or non-randomised study design, further analyses of the time frames with

these studies removed were undertaken. Leaving out these studies did not change the

outcomes at each time frame.

62

Table 3: Summary of Interventions in Early Mobilisation Trials

Study ID Sheath Sizes

Method of Achieving

Haemostasis

Total Bed rest

Experimental

Total Bed rest Control

Bed elevation during bed rest

(time after gaining haemostasis)

Baum 1996 5-8F Not mentioned 2 4 ?

Best 2010 5-6F Manual or C-clamp 1.5 3-4 To 30° at 45 minutes

Bogart 1999 8F Manual 4 6 ?

Chair 2007 5-6F Manual or C-clamp 4 12-24 ?

Farmanbar 2008 ≤7F Manual then sandbag 2 6 No

Hoglund 2011 ≤6F Femostop 1.5 5 ?

Keeling 1996 7-8F Manual 4 6 ?

Lim 1997 6F Femostop 4 6 To 45° immediately

Logemann 1999 6F Femostop 2 6 To 45° at 4 hours

Mah 1999 7F Manual 3 5 To ≥30° at 1 hour

Pollard 2003 6F Manual 2.5 4.5

To 60° at 1 hour in experimental group To 60° at 4 hours in control group

Roebuck 2000 6F Femostop 2 4 To 30° immediately

Singh 1998 3x6 7F Manual or C-clamp 3 6 ?

Wang 2001 5-6F Manual then sandbag

4 6 ?

Wood 1997 6F Manual or C-clamp

2.5 4 ?

63

Discussion

Summary of main results

Fifteen studies were identified that met the inclusion criteria for this Systematic Review.

Of the studies, six mobilised participants at or before two hours including a total of

1530 participants. Ten of the fifteen studies mobilised their participants at or before 3

hors, including a total of 3927 participants. All fifteen studies included mobilised their

participants at or prior to four hours, with a grand total of 4281 participants overall. All

of the studies included an experimental and a control group. In the case of Singh et al.

(1998) there were two experimental groups, one at three hours and one at four hours.

To reduce the chance of bias, only the results of the three-hour group and the six-hour

control group were included. All of the studies used vascular complications as their

measure of safety related to earlier mobilisation. These vascular complications

included haematoma (and retro-peritoneal haematoma), bleeding and pseudo-

aneurysm. All 15 studies were assessed as having a moderate quality as per the

GRADE Working Group grades of evidence (Schunemann et al., 2008). The

downgrading was either due to a quasi-randomised or non-randomised study design,

or incomplete allocation concealment or blinding.

Results have been summarised into time frames relating to the early mobilisation.

None of the six trials in the ≤ 2-hour group showed statistical significance between the

early mobilisation group and the control group with measured vascular complications.

Overall results favoured the control group but not with any statistical significance (RR

1.1591; 95% CI 0.7544-1.7809; p = 0.5023). The randomised controlled trials, when

analysed separately, showed similar figures favouring the control group but with no

statistical significance (RR 1.0746; 95% CI 0.6073-1.9012; p = 0.8763). Overall results

in the ≤ 3-hour favoured the experimental group, although the figures were not quite

statistically significant (RR 0.8430; 95% CI 0.7041-1.0094; p = 0.0625). When the

seven randomised controlled trials were analysed separately, figures slightly favoured

the control group but with no statistical significance (RR 1.0023; 95% CI 0.7818-

1.2849; p = 1.000). At the ≤ 4-hour timeframe, the overall vascular complication results

favoured the experimental group, although the figures were not quite statistically

significant (RR 0.8696; 95% CI 0.7399-1.0219; p = 0.0891). The eleven randomised

controlled trials in this timeframe slightly favoured the control group but with no

statistical significance (RR 1.0361; 95% CI 0.8440-1.2719; p = 0.7401).

64

Four studies directly measured back pain in both their experimental and control groups

to evaluate whether earlier mobilisation decreased levels of back pain experienced by

participants (Chair et al., 2007; Hoglund et al., 2011; Wang et al., 2001; Wood et al.,

1997). Three of the studies showed clear levels of decreased back pain in their

experimental groups, two of which showed a statistical significance (Hoglund et al.

2011 and Chair et al. 2007 both showing p < 0.001 favouring earlier mobilisation).

Wood et al. (1997) showed results that favoured the later mobilisation group, however,

their results may have been biased due to the inclusion of three participants with

chronic back pain who elevated the overall pain measurements due to their

experiencing 9 – 10 / 10 back pain on bed rest. Six others studies mentioned anecdotal

comments from participants, with more favourable comments from the experimental

groups relating to levels of back pain.

Overall completeness and applicability of evidence

Early mobilisation at or before four hours after a femoral approach coronary angiogram

has been well studied, with the fifteen studies found here. For fifteen years,

researchers have been looking at mobilisation at two, three and four hours and getting

favourable results towards earlier mobilisation where there is no statistically significant

increase in vascular complications with the early mobilisation out of bed. Although

several of the studies were relatively small, and on their own provide limited information,

when added into this systematic review they add valuable statistics.

One point noticeable in the review was the lower level of women in the studies. Chair

et al. (2007) was the only study to include similar levels of women to men in each of

their groups (55.8% women control / 48.8% women experimental). In several studies

looking at risk factors for vascular complication after coronary angiogram, it has been

found that females are more at risk of developing vascular complications such as

bleeding and haematoma than males (Dumont, Keeling, Bourguignon, Sarembock &

Turner, 2006; Eggebrecht et al., 2002; Gall, Tarique, Natarajan & Zaman, 2006; Gillane

& Pollard, 2009; Mah et al., 1999). Hoglund et al. (2011) reason that this could be due

to the smaller vasculature of females, which can lead to multiple punctures of the artery

and an increased chance of a haematoma. Although the studies included in this review

are representative of the presenting patient population for coronary angiogram,

perhaps more study could be undertaken specifically on women to ensure that the

results from these studies are replicable to that population.

65

Quality of the evidence

All 15 studies were assessed as having a moderate quality as per the GRADE Working

Group grades of evidence (Schunemann et al., 2008). Four of the studies (Best et al.,

2010; Mah et al., 1999; Roebuck et al., 2000; Wang et al., 2001) used a quasi-

randomised or non-randomised study design by either carrying out a week about

allocation or retrospective / prospective analysis, which may increase the bias in the

studies. Because of this, they were graded as moderate quality. None of the 11

included randomised controlled trials had complete allocation concealment or blinding,

which may increase bias. Although a randomised controlled trial is usually graded as a

High quality rating on the using the GRADE levels of evidence, these 11 studies were

downgraded due to their lack of concealment or blinding.

Authors’ conclusions

Implications for practice

This review aimed to provide an evidence-based argument that time to mobilisation can

be decreased in patients post-femoral approach diagnostic coronary angiogram

maintaining a low risk of vascular complications. This in turn would potentially help to

decrease or prevent back pain occurring. The results from this study show no

statistically significant difference in vascular complications between the control groups

and the early mobilisation out of bed groups at ≤2, 3 or 4 hours post femoral approach

coronary angiogram. Therefore, mobilisation after femoral approach coronary

angiogram, without deployment of a vascular closure device, may be as safe at 1 ½ to

4 hours mobilisation as it is at 6 hours and may have a positive benefit of reducing

back pain related to lying in bed.

Implications for research

As stated earlier, further research on early mobilisation post femoral approach coronary

angiogram on a higher number of female patients would be of benefit, due to their

under-representation in these fifteen studies.

It would also be beneficial if any further studies on this topic were randomised

controlled trials. This lowers the risk of bias and ensures they can be included in a

Systematic Review in the future.

66

Acknowledgements

We would like to thank Dalice A. Sim (PhD), Statistical Consultant from the School of

Mathematics, Statistics and Operation Research at Victoria University, for her

assistance with the statistical analysis in this Systematic Review and Meta-Analysis.

Declarations of interest

There is no potential conflict of interest among the review authors

Differences between protocol and review

In the protocol it was stated that included studies looked at coronary angioplasty as

well as diagnostic coronary angiogram. Only five studies were identified in the literature

search that included coronary angioplasty. These studies proved to be too heterogenic

in their methodology to be suitable for comparison in a Systematic Review. Therefore

the inclusion criteria for the Systematic Review included only studies that looked at

femoral approach diagnostic coronary angiogram.

References

References to studies included in this review

Baum, R. A., & Gantt, D. S. (1996). Safety of decreasing bedrest after coronary angiography. Catheterization and Cardiovascular Diagnosis, 39, 230-233.

Best, D. G., Pike, R., Grainger, P., Eastwood, C. A., & Carroll, K. (2010). A prospective

study of early ambulation 90 minutes post-left heart catheterization using a retrospective comparison group. Canadian Journal of Cardiovascular Nursing, 20(4), 15-20.

Bogart, M. A., Bogart, D. B., Rigden, L. B., Jung, S. C., & Liston, M. J. (1999). A

prospective randomised trial of early ambulation following 8 french diagnostic cardiac catheterization. Catheterization and Cardiovascular Interventions, 47, 175-178.

Chair, S. Y., Thompson, D. R., & Li, S. K. (2007). The effect of ambulation after cardiac

catheterization on patient outcomes. Journal of Clinical Nursing, 16, 212-214. Farmanbar, R., Chinikar, M., Gozalian, M., Baghaie, M., Roshan, Z. A., &

Moghadamnia, M. (2008). The effect of post coronary angiography bed-rest time on vascular complications. The Journal of Tehran University Health Center, 4, 225-228.

Hoglund, J., Stenestrand, U., Todt, T., & Johansson, I. (2011). The effect of early

mobilisation for patient undergoing coronary angiography; a pilot study with focus on vascular complications and back pain. European Journal of Cardiovascular Nursing, 10(2), 130-136.

67

Keeling, A., Taylor, V., Nordt, L. A., Powers, E., & Fisher, C. (1996). Reducing time in bed after cardiac catheterization (TIBS II). American Journal of Critical Care, 5(4), 277-281.

Lim, R., Anderson, H., Walters, M. I., Kaye, G. C., Norell, M. S., & Caplin, J. L.

(1997). Femoral complications and bed rest duration after coronary arteriography. American Journal of Cardiology, 80, 222-223.

Logemann, T., Luetmer, P., Kaliebe, J., Olson, K., & Murdock, D. K. (1999). Two

versus six hours of bed rest following left-sided cardiac catheterization and a Meta-Analysis of early ambulation trials. American Journal of Cardiology, 84, 486-488.

Mah, J., Smith, H., & Jensen, L. (1999). Evaluation of 3-hour ambulation post

cardiac catheterization. Canadian Journal of Cardiovascular Nursing, 10(1-2), 23-30.

Pollard, S., Munks, K., Wales, C., Crossman, D., Cumberland, D., Oakley, G. &

Gunn, J. (2003). Position and Mobilisation Post Angiography Study (PAMPAS): a comparison of 4.5 hours and 2.5 hours bed rest. Heart, 89, 447-448.

Roebuck, A., Jessop, S., Turner, R., & Caplin, J. L. (2000). The safety of two-

hour versus four-hour bed rest after elective 6-french femoral cardiac catheterization. Coronary Health Care, 4, 169-173.

Singh, N., Kuganesan, K., Goode, E., & Ricci, A. J. (1998). The effect of early

ambulation on hematoma formation and vascular complications following 7 french diagnostic cardiac catheterization. Canadian Journal of Cardiology, 14(10), 1223-1227.

Wang, S., Redeker, N. S., Moreyra, A.E., & Diamond, M. R. (2001). Comparison

of comfort and local complications after cardiac catheterization. Clinical Nursing Research, 10(1), 29-39.

Wood, R. A., Lewis, B. K., Harber, D. R., Kovack, P. J., Bates, E. R., & Stomel,

R. J. (1997). Early ambulation following 6 french diagnostic left heart catheterization: a prospective randomized trial. Catheterization and Cardiovascular Diagnosis, 42, 8-10.

References to studies excluded from this review

Pooler-Lunse, C., Barkman, A., & Bock, B. F. (1996). Effects of modified positioning and mobilization on back pain and delayed bleeding in patients who had received heparin and undergone angiography: a pilot study. Heart & Lung, 25(2), 117-123.

Tagney, J., & Lackie, D. (2005). Bed-rest post-femoral arterial sheath removal – what is

safe practice? A clinical audit. Nursing in Critical Care, 10(4), 167-173. Tengiz, I., Ercan, E., Bozdemir, H., Durmaz, O., Gurgun, C., & Nalbantgil, I. (2003). Six

hour ambulation after elective coronary angioplasty and stenting with 7F guiding catheters and low dose heparin. Kardiology Polska, 58(2), 93-97.

68

Additional references

Asinas, M. (2010). Coronary angiography: clinical information. Retrieved from http://www.joannabriggs.edu.au

Chair, S. Y. (2004). Effects of early ambulation after cardiac catheterization (Doctor of

Philosophy thesis, University of Colorado, United States of America). Retrieved from http://search.proquest.com/docview/305044512/1370BC6239A5C0586B5

/1?accountid=34374

Chair, S. Y., Li, K. M., & Wong, S. W. (2004). Factors that affect back pain among Hong Kong Chinese patients after cardiac catheterization. European Journal of Cardiovascular Nursing, 3(4), 279-285.

Chair, S. Y., Taylor-Piliae, R. E., Lam, G., & Chan, S. (2003). Effect of positioning on back pain after coronary angiography. Journal of Advanced Nursing, 42(5), 470-478.

Chandrasekar, B., Doucet, S., Bilodeau, L., Crepeau, J., deGuise, P., Gregoire,

J., . . . Pasternac, A. (2001). Complications of cardiac catheterization in the current era: a single-center experience. Catheterization and Cardiovascular Interventions, 52, 289-295.

Dumont, C., Keeling, A., Bourguignon, C., Sarembock, I., & Turner, M. (2006).

Predictors of vascular complications post diagnostic cardiac catheterization and percutaneous coronary interventions. Dimensions in Critical Care Nursing, 25(3), 137-142.

Eggebrecht, H., Haude, M., Woertgen, U., Schmermund, A., Von Birgelen, C.,

Naber, C., . . . Erbel, R. (2002). Systematic use of a collagen-based vascular closure device immediately after cardiac catheterization procedures in 1317 consecutive patients. Catheterization and Cardiovascular Interventions, 57, 486-495.

Gall, S., Tarique, A., Natarajan, A., & Zaman, A. (2006). Rapid ambulation after

coronary angiography via femoral artery access: A Prospective Study of 1,000 Patients. Journal of Invasive Cardiology, 18(3).

Gillane, O., & Pollard, M. (2009). Early ambulation of patients post-angiography

with femoral puncture. British Journal of Cardiology, 16, 137-140. Hay, D. (2004). Cardiovascular Disease in New Zealand. The National Heart

Foundation of New Zealand. Higgins, J. P. T., & Altman, D. G. (2008). Chapter 8: Assessing risk of bias in included

studies. In J.P.T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 359-387). Chichester, England: John Wiley & Sons.

Lauck, S., Johnson, J. L., & Ratner, P. A. (2005). Self-care behaviour and factors

associated with patient outcomes following same-day discharge percutaneous coronary intervention. European Journal of Cardiovascular Nursing, 8, 190-199.

69

Ministry of Health. (2011). Provisional data: New Zealand mortality statistics: 2006 to 2009. Retrieved from http://www.health.govt.nz/publication/provisional-data-new-zealand-mortality-statistics-2006-2009

Rezaei-Adaryani, M., Ahmadi, F., & Asghari-Jafarabadi, M. (2009). The effect of

changing position and early ambulation after cardiac catheterization on patients' outcomes: A single-blind randomized controlled trial. International Journal of Nursing Studies, 46(8), 1047-1053.

Sabo, J., Chlan, L. L., & Savik, K. (2008). Relationships among patient characteristics,

comorbidities, and vascular complications post-percutaneous coronary intervention. Heart & Lung: The Journal of Acute and Critical Care, 37(3), 190-195.

Schunemann, H. J., Oxman, A. D., Vist, G. E., Higgins, J. P. T., Deeks, J. J., Glasziou,

P., & Guyatt, G. H. (2008). Chapter 12: Interpreting results and drawing conclusions. In J.P.T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 359-387). Chichester, England: John Wiley & Sons.

Tobias, M., Sexton, K., Mann, S., & Sharpe, N. (2006). How low can it go? Projecting

ischaemic heart disease mortality in New Zealand to 2015. New Zealand Medical Journal, 119(1232), 14-26.

70

Characteristics of Studies

Table 4: Characteristics of Included Studies (ordered by study ID)

Baum and Gantt (1996)

Aim of Study To assess the safety of allowing patients to ambulate after 2 hours of bed rest following

diagnostic coronary angiography.

Methods Study design:

Duration:

Randomised Controlled Study

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

United States of America

Single centre, Cardiology Department

205 total; 104 control; 101 experimental

Mean control 59.0 years / Mean experimental 58.0 years

Not mentioned

Diagnostic coronary Angiogram

Femoral approach

5-8F Catheter

80% received 2500iu heparin

Age > 75 years of age

Severe aortic stenosis (aortic valve area < 1 cm

Creatinine >2.5 mg/dl

Full dose anticoagulation (oral or intravenously)

Unstable angina

Obesity >4 standard deviations

Severe aortic regurgitation

Left ventricular dysfunction (moist rales on examination, gallop

or elevated pulmonary capillary wedge pressure (PCWP) >20

mmHg)

Brachial artery approach

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

2 hour bed rest period

Immediate ambulation in room

30 minutes - 200 feet

2 hours ambulation as desired

4 hour bed rest period

Immediate ambulation in room

30 minutes - 200 feet

2 hours ambulation as desired

Puncture site monitored through bed rest period

During ambulation in room then 30 minutes post ambulation

2 hours and 3 hours post ambulation

All participants received their allocated interventions

Outcomes Primary:

Bleeding

Haematoma

Risk of Bias Method of randomization:

Blinding:

Not mentioned

Not stated in study

71

Incomplete outcome data:

Outcome reporting:

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Best et al. (2010)

Aim of Study To determine safety of ambulating patients at 90 minutes sheath removal compared to the

current practice of ambulation at three to four hours post-sheath removal.

Methods Study design:

Duration:

Retrospective vs. Prospective Quasi-Experimental Trial

6 months retrospective and then 6 months prospective

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

Canada

Experimental single centre, Control all health centres in Alberta,

Canada via APPROACH database.

595 total; 402 control; 193 experimental

Mean control 62.1 years / Mean experimental 60.2 years

37.1% women control / 31.1% women experimental

Left heart diagnostic catheterisation

6F or less sheath

Over 19 years

Ability to ambulate post procedure

History of a bleeding disorder

Receiving anticoagulation by either the oral, intravenous or

subcutaneous route

Receiving IIb/IIIa platelet receptor inhibitors

Received thrombolytic therapy this admission

APTT greater than 38 seconds pre-procedure or platelets were

less than 130x109 /L pre-procedure

Prior femoral popliteal bypass

Percutaneous coronary intervention

Systolic BP >180mmHg or diastolic BP >110mmHg

Weight more than 350 pounds

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Immediate sheath removal

Manual compression or c-clamp

HOB elevation to 45° after 60 mins

90 minute ambulation

Immediate sheath removal

Manual compression or c-clamp

HOB elevation to 45° after 60 mins

3-4 hour ambulation

Not specifically mentioned

All participants received their allocated interventions

Outcomes Primary:

Bleeding was defined as access site bleed

Hematoma formation measured and categorized as <5cm, 5–

10cm or >10cm

AV fistula

72

Pseudoaneurysm

Retroperitoneal bleed

Vascular occlusion

Dissection

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not applicable, retrospective control group, prospective

experimental group

Not applicable

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Bogart et al. (1999)

Aim of Study To assess the safety of 4-hr ambulation after diagnostic cardiac catheterization with 8 French

sheaths and catheters.

Methods Study design:

Duration:

Randomised Controlled Trial

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

United States of America

Single centre, Cardiology Department

200 total; 100 control; 100 experimental

Mean control 60.0 years / Mean experimental 55.0 years

38.0% women control / 36.0% women experimental

Diagnostic cardiac catheterisation

Femoral approach

Geographic remoteness (more than 1h drive) from the laboratory

with inadequate or unreliable follow-up likely over the next 24 hr

An interventional therapeutic procedure

Transient cerebral ischemic episodes or recent stroke (less than

1 month before)

Severe systemic hypertension (systolic blood pressure greater

than 200 mm Hg and/or diastolic blood pressure greater than

100 mm Hg)

Severe peripheral vascular disease (femoral pulses 1 or less)

Age less than 21 years or greater than 75 years

Body surface area greater than 2.5 m2

Generalized debility or dementia

Frequent ventricular arrhythmias

Renal insufficiency (serum creatinine more than 2mg/dl)

Recent hospitalization for acute myocardial infarction (within past

7 days)

On intravenous heparin or with international normalized ratio

(INR) of >2

Chronic corticosteroid use

Non-compliance with the period of bed rest

Repeat procedure, same entry site within 1 week

Right heart catheterization

73

More than one arterial puncture to initiate the procedure

Thrombocytopenia (platelet count 100,000 mm3)

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Sheath removed when ACT less than 150

Mobilisation at 4 hours post sheath removal

Sheath removed when ACT less than 150

Mobilisation at 6 hours post sheath removal

The same researcher made a physical assessment of the groin

site upon ambulation and the following day in all patients.

Results documented at these times

All participants received their allocated interventions

Outcomes Primary:

Re-bleeding

Hematoma

Arterio-venous (AV) fistula

Pseudoaneurysm

Limb ischemia

Thrombosis of the femoral artery

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not mentioned

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Chair et al. (2007)

Aim of Study The main purpose of this study was to determine the effects of an early ambulation

intervention after cardiac catheterization on patient's back pain and incidence of vascular

complications, especially bleeding and haematoma formation

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

Hong Kong

Single centre study

86 total; 43 control; 43 experimental

Mean control 63.19 years / Mean experimental 62.70 years

55.8% women control / 48.8% women experimental

Coronary angiogram

Femoral approach

5-6F sheaths

Emergency cardiac catheterisation

Age <18 or >75 years old

Non-Chinese patients

Prior cardiac catheterisation

Known bleeding disorders

Anticoagulant therapy within 24 hours before procedure

Presence of back pain before procedure

74

Active bleeding from femoral site before sheath removal

Active bleeding from femoral site after sheath removal but before

ambulation

Medical complications such as angina and myocardial infarction

developed during the procedure

A systolic BP ≥ 180 mmHg and/or diastolic BP ≥ 110mmHg

before the procedure

Unable to ambulate independently before the procedure

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Sheath removed straight away

Mobilisation at 4 hours

Mobilised around bed for 2 minutes each hour for three hours

then mobilised freely

Sheath removed straight away

12-24 hours bed rest

Mobilised around bed for 2 minutes each hour for three hours

then mobilised freely

Puncture site checked regularly for first four hours then after

each period of ambulation

Back pain assessed at 4 hours, 8 hours and next morning

All participants received their allocated interventions

Outcomes Primary:

Secondary:

Haematoma

Bleeding

Back pain

Urinary discomfort

Puncture site pain

General well-being

Patient satisfaction

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Computer generated list

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Farmanbar et al. (2008)

Aim of Study To test the hypothesis that reducing time-to-ambulation from 6 hours to 2 hours would not

increase post-angiography vascular complications

Methods Study design:

Duration:

Randomised Controlled Trial

February 2006 to November 2006

Participants Country:

Setting:

Number participants:

Age:

Gender:

Iran

Post angiography wards in two different hospitals

120 total; 60 control; 60 experimental

Mean control 59.9 years / Mean experimental 60.17 years

41.7% women control / 38.3% women experimental

75

Inclusion criteria:

Exclusion criteria:

All patients of one interventionalist presenting for cardiac

angiography

Femoral approach

Single puncture of femoral artery mainly 7F sheath

Renal failure

Chronic obstructive pulmonary disease

Known pre-procedure bleeding disorder

Patient transfer to ICU post angiography

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Sheath removed immediately post procedure, then manual

compression of puncture site for minimum 10 minutes

Two 2.5kg sandbags placed, one removed at one hour the other

before mobilising

Mobilisation after 2 hours supine bed rest

Sheath removed immediately post procedure, then manual

compression of puncture site for minimum 10 minutes

Two 2.5kg sandbags placed, one removed at one hour the other

before mobilising

Mobilisation after 6 hours supine bed rest

Vital signs, pedal pulses and puncture site checked:

Prior to sheath removal

Every 15 minutes after haemostasis for 1 hour

Every 30 minutes for next hour

Every hour until ambulation

After ambulation every 15 minutes for 1 hour

Checked at discharge following day

All participants received their allocated interventions

Outcomes Primary:

Bleeding before ambulation

Small haematoma before ambulation (<5cmx5cm)

Large haematoma before ambulation (>5cmx5cm)

Post ambulation bleeding <24 hours

Post ambulation haematoma <24 hours

Late vascular complication (after 30 days)

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Hoglund et al. (2011)

Aim of Study To assess the safety, and perceived comfort, of early mobilisation after coronary angiography

with femoral approach in patients with dual antiplatelet therapy, comparing 1.5 and 5 hours of

bed rest

Methods Study design:

Duration:

Randomised controlled trial

January – June 2008

76

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

Sweden

Single centre catheterisation lab

104 total; 52 control; 52 experimental

Mean control 65.1 years / Mean experimental 63.7 years

17.0% women control / 24.0% women experimental

≥ 18 years of age

Coronary Angiography for valvular disorder, stable angina,

NSTEMI or ACS

Aspirin and Clopidogrel loading and low molecular weight

heparin included

Not willing to participate

Participation in pharmacological study

Language difficulties

Sheath size >6 F

Pre-treatment with heparin and/or abciximab

Underwent angioplasty

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Sheath removed immediately after angiography and Femostop

placed

Pressure remained for 1 hours then additional 1/2 hour bed rest

Mobilised after total supine bed rest 1 1/2 hours

Sheath removed immediately after angiography and Femostop

placed

Pressure remained for 3 hours then additional 2 hours bed rest

Mobilised after total supine bed rest 5 hours

Immediately after angiography

1 hour in bed rest

After release of compression

At the end of bed rest

After 2 hours mobilisation

After 4 hours mobilisation

12-24 hours after procedure

Telephone interview 48-72 hours post procedure

All participants received their allocated interventions

Outcomes Primary:

Secondary:

Haematoma

Bleeding

Back pain

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Keeling et al. (1996)

Aim of Study To determine if there would be a significant difference in the incidence of bleeding from

femoral artery insertion sites between cardiac catheterization patients who remained in bed for

77

4 hours and those who remained in bed for 6 hours after sheath removal

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

United States of America

Single centre acute cardiology unit

86 total; 43 control; 43 experimental

Mean control 58.0 years / Mean experimental 61.0 years

37.0% women control / 14.0% women experimental

Adults

Suspected or diagnosed coronary artery disease

Diagnostic cardiac catheterisation without PTCA

Femoral approach

Heparin or other anticoagulant or thrombolytic agents before the

procedure

Known bleeding disorders

Medical complications during the procedure

Those requiring admission post procedure

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Femoral artery sheath 7-8F

Immediate sheath removal with manual compression for average

20 mins

Sandbag placed (5lb) until mobilisation

Supine, leg straight, HOB allowed to 30°

Mobilised at 4 hours post sheath removal

As above but mobilised at 6 hours post sheath removal

After mobilisation and before discharge

All participants received their allocated interventions

Outcomes Primary:

Haematoma

Bleeding

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Lim et al. (1997)

Aim of Study To determine whether there is any difference in the femoral bleeding complication rate

between 4 and 6 hours of bed rest after elective coronary arteriography

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country:

Setting:

United States of America

Single centre tertiary cardiology department

78

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

200 total; 100 control; 100 experimental

Mean overall 59.0 years

% Women not mentioned

Left-sided diagnostic cardiac catheterization

Femoral approach

Peripheral vascular grafts

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Diagnostic angiogram via femoral approach

Immediate sheath removal

Femostop for 1 hour

Bed rest at 30-45° head up

Mobilised at 4 hours

As above then mobilised at 6 hours

From sheath removal to discharge

All participants received their allocated interventions

Outcomes Primary:

Bruising

Bleeding

Pseudoaneurysm

Blood transfusion

Need for surgical intervention

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

Operators only blinded from randomisation

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Logemann et al. (1999)

Aim of Study This study compares the groin complication rate after cardiac catheterization in patients

treated with the standard 6-hour duration of bed rest with that after a duration of only 2 hours

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

United States of America

Single centre regional heart centre

201 total; 96 control; 105 experimental

Mean control 62.0 years / Mean experimental 61.0 years

33.0% women control / 42.8% women experimental

Outpatient diagnostic cardiac catherisation

Femoral approach

Use of oral anticoagulants (except aspirin),

Age >80 years

Morbid obesity (>50% above ideal body weight)

History of bleeding disorder

Coronary disease detected at catheterization prompting

admission for treatment (coronary angioplasty or surgery)

79

Creatinine 2.0 mg/dl

Severe aortic valvular disease

Severe hypertension (systolic blood pressure >180 mm Hg or

diastolic blood pressure >105 mm Hg)

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

6F diagnostic femoral cardiac angiogram

Immediate sheath removal

Femostop pressure device for 30 minutes

Supine bed rest

Mobilised at 2 hours

Observed for 5 hours then discharged

As above except head of bed elevated at 4 hours to 45°

Mobilised at 6 hours

Discharged 1 hour after that

Observed from sheath removal until discharge 7 hours

afterwards

All participants received their allocated interventions

Outcomes Primary:

Bleeding

Haematoma

Pseudoaneurysm

Vascular repair

Blood transfusion

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Mah et al. (1999)

Aim of Study To evaluate the effects of 3-hour ambulation post cardiac catheterization with a 7 french

arterial catheter on bleeding, haematoma formation and vascular complications

Methods Study design:

Duration:

Retrospective / prospective controlled trial

Retrospective 8 months from start point for control group then

forward 7 months for experimental 3 hour group

80

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

Canada

Single centre large tertiary hospital

880 total; 472 control; 408 experimental

Mean control 61.55 years / Mean experimental 60.82 years

22.0% women control / 22.4% women experimental

All patients who underwent an elective diagnostic left heart

catheterization

Femoral access

Aortic and/or mitral stenosis, regurgitation or insufficiency

Full anticoagulation

Systolic BP >180mmHg and/or diastolic >110mmHg

PT INR > 2.2 if on Coumadin

Obesity: BMI >30

Catheterization via brachial artery

Use of Angio-seal

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

7F sheath, removed immediately

Minimum of 10 mins manual pressure

Supine for 1 hour

Then could lie on either side with HOB >30% as long as effected

leg straight

Mobilised at 3 hours

As above but mobilised at 5 hours

From sheath removal to discharge

All participants received their allocated interventions

Outcomes Primary:

Bleeding

Haematoma

AV fistula

Pseudoaneurysm

Thrombosis

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not applicable, retrospective control group, prospective

experimental group

Not applicable

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Pollard et al. (2003)

Aim of Study To examine the safety of early sit-up and mobilisation after routine cardiac catheterisation in

contemporary practice

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country:

Setting:

United Kingdom

Single centre

81

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

705 total; 362 control; 343 experimental

Not mentioned

Not mentioned

Presence of stable angina

Planned, elective diagnostic catheterisation

Successful single puncture of femoral artery

6F catheter

Age <18 years

Inability to give informed consent

Possible pregnancy

Participation in another study

Coronary angioplasty at same sitting

Heparin treatment

Warfarin with INR ≥2.0

Bleeding disorder

Previous surgery to iliac or femoral arteries

Right heart catheter performed at same sitting

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Sheath removed immediately with minimum of 10 minutes

manual compression

Supine then sit up after 1 hour

Mobilise after 2.5 hours

Sheath removed immediately with minimum of 10 minutes

manual compression

Supine then sit up after 4 hours

Mobilise after 4.5 hours

Levels of comfort and groin site

Pre procedure

30 minutes post

2 hours post

4 hours post

48 hours post

All participants received the allocated intervention

Outcomes Primary:

Secondary:

Bleeding

Haematoma

False Aneurysm

Transfusion

Surgical repair

Level of discomfort as per McGill Pain Questionnaire

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not stated in study

None – open label

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting except 50 patients withdrew from

study and not explained why

82

Roebuck et al. (2000)

Aim of Study The trial examined the safety of reducing bed rest from 4 h. to 2 h. after elective 6-French

cardiac catheterisation.

Methods Study design:

Duration:

Quasi-experimental controlled trial, week about allocation

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

United Kingdom

Single centre cardiothoracic referral centre

305 total; 117 control; 188 experimental

Mean control 61.0 years / Mean experimental 60.0 years

27.0% women control / 31.0% women experimental

Elective patients

6F femoral approach diagnostic cardiac catheterization

INR >1.5

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Diagnostic angiogram with 6F sheath

Immediate sheath removal then Femostop placement for 1 hour

Once Femostop off, back of bed elevated to 30°

Mobilised at 2 hours if no haematoma or bleeding

As above except mobilised at 4 hours

Hourly for first four hours, telephone follow up at 24 hours then

at one month

All participants received the allocated intervention

Outcomes Primary:

Haematoma

Bleeding

Transfusion requirement

Pseudoaneurysm

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Week about allocation to each group

Not stated in study

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Singh et al. (1998)

Aim of Study To assess the feasibility and safety of early ambulation 3 to 4 hours after diagnostic cardiac

catheterization

Methods Study design:

Duration:

Randomised controlled trial

April 1996 – August 1997

Participants Country:

Setting:

Number participants:

Age:

Canada

Single centre - tertiary care urban community hospital

874 total; 185 control; 336 experimental 1; 353 experimental 2

Mean overall 60.0 years

83

Gender:

Inclusion criteria:

Exclusion criteria:

29.0% women overall

7 french diagnostic cardiac angiogram

Femoral approach

Refused consent

Cardiologist refused

Patient involved in another study that could affect bleeding risk

Interventions Experimental group 1:

Experimental group 2:

Control group:

Follow-up:

Integrity of interventions:

7F femoral approach cardiac catheter

Sheath removed immediately

Manual compression or C clamp

Mobilised at 3 hours

As above then mobilised at 4 hours

As above then mobilised at 6 hours

Assessed after sheath removal and throughout stay until

discharge

Follow up phone call 24 hours after discharge

All participants received the allocated intervention

Outcomes Primary:

Haematoma

Blood transfusion requirement

Need for vascular repair

Thromboembolism

Local Infection

Pseudoaneurysm formation

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Not mentioned

Not mentioned

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Wang et al. (2001)

Aim of Study The following hypotheses were addressed: Patients who receive 4 versus 6 hours of bed rest

after left-heart catheterization will report no difference in the occurrence of hematoma or active

bleeding or loss of distal pulses due to clot formation at the puncture site, less back pain or

puncture-site pain, less numbness and tingling sensation in the procedure limb, and higher

satisfaction with the procedure care or time on bed rest

Methods Study design:

Duration:

Quasi-experimental controlled trial

Not mentioned

Participants Country:

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

United States of America

Single centre cardiology outpatient unit

82 total; 41 control; 41 experimental

Mean control 62.0 years / Mean experimental 58.7 years

26.8% women control / 34.0% women experimental

Admitted for left-heart catheterization

Able to speak English

84

Exclusion criteria:

Willing to participate in the study as evidenced by a signed

written consent

Had serum potassium level less than 5.5mEq/,

Serum prothrombin time (PT) level less than 15 sec

Blood hemoglobin level greater than 8.5gm/dL at preadmission

testing

Unstable angina during or after the procedure

Severe peripheral vascular disease

Hematoma immediately after catheterization in the

catheterization laboratory

Need for heparin bolus during or after the procedure

Requirement for right and left-heart catheterization

Pressure dressing to site post-procedure

Overnight stay

Pressure dressing required

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

Femoral approach

5-6F sheaths

Sheaths removed immediately then manual pressure applied

Once haemostasis, 12 pound sandbag applied

Supine for 2 hours

Bed raised to 45°

Mobilised after 4 hours

As above but mobilised after 6 hours

Every 15 minutes for the 1st hour

Every 30 minutes for 2 hours

Every hour until discharge

Numeric Rating Scale questionnaires given before mobilisation

All participants received the allocated intervention

Outcomes Primary:

Secondary:

Bleeding

Haematoma

Back pain

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

No randomisation, week about into control or experimental group

Not mentioned

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

Wood et al. (1997)

Aim of Study Early Ambulation Following 6 French Diagnostic Left Heart Catheterization comparing 2.5 and

4 hours

Methods Study design:

Duration:

Randomised controlled trial

Not mentioned

Participants Country: United States of America

85

Setting:

Number participants:

Age:

Gender:

Inclusion criteria:

Exclusion criteria:

Single centre community hospital

291 total; 167 control; 124 experimental

Mean control 60.8 years / Mean experimental 60.0 years

40.4% women control / 40.7% women experimental

Diagnostic coronary angiogram

Femoral approach with 6F sheath

Aspirin ok

Warfarin stopped 3 days prior

Outpatients

PTCA

Admission for CABG

Bleeding or haematoma prior to ambulation

Interventions Experimental group:

Control group:

Follow-up:

Integrity of interventions:

6F femoral sheath

Immediate sheath removal

Manual compression or C-clamp for minimum of 12 minutes

Patient mobilised at 2.5 hours

As above except mobilised at 4 hours

Immediately after ambulation patients assessed for haematoma,

bleeding and level of comfort whilst they had been in bed.

Follow up phone call within 48 hours of discharge

All participants received the allocated intervention

Outcomes Primary:

Secondary:

Haematoma

Bleeding

Pseudoaneurysm

Presence of back pain or leg stiffness

Risk of Bias Method of randomization:

Blinding:

Incomplete outcome data:

Outcome reporting:

Dice throw, (1,2,3 mobilised at 2.5 hours, 4,5,6 mobilised at 4

hours)

Not mentioned

Nil loss to follow up, all patients accounted for in data

Nil selective outcome reporting

86

Table 5: Characteristics of excluded studies

Study Reason for exclusion

Pooler-Lunse, Barkman &

Bock (1996)

This study looked mainly at modified positioning prior to mobilisation and has been

included the previous discussion on this topic. However, although it also discusses early

mobilisation, in this study that meant just standing beside the bed for two minutes and

not walking around the unit (as with all of the other early mobilisation studies). Therefore

it was excluded from the early mobilisation review.

Tagney and Lackie (2005) This study looked at both diagnostic angiogram and angioplasty. One third of participants

received a vascular closure device. The results were not displayed in such a way as to

ascertain only the vascular complications related to those who did not receive a vascular

closure device, nor did they differentiate between angiogram and angioplasty results.

Therefore, the study was not included in the review.

Tengiz et al. (2003) Despite an extensive search, including emailing the magazine itself and searching 3

different libraries, the full-text article of this study was not able to be located. Therefore it

could not be included in the review.

87

Data and analysis

Table 6: ≤ 2 hours bed rest post-diagnostic coronary angiogram - total bleeding complications (bleeding / haematoma)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Bleeding Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 5 1.6000 (0.5604-4.5685) 0.5547

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Total RCT’s 318 / 312 1-2 / 3-6 23 / 20 1.1283 (0.6327-2.0123) 0.7529

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Total Non-RCTs 381 / 519 16 / 19 1.1471 (0.5978-2.2012) 0.7286

Total All 699 / 831 1-2 / 3-6 39 / 39 1.1888 (0.7716-1.8318) 0.4310

88

Table 7: ≤ 2 hours bed rest post-diagnostic coronary angiogram - total vascular complications (total bleeding / pseudoaneurysm)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Vascular Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 6 1.3333 (0.4973-3.5749) 0.7749

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Total RCT’s 318 / 312 23 / 21 1.0746 (0.6073-1.9012) 0.8763

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Total Non-RCT’s 381 / 519 16 / 19 1.1471 (0.5978-2.2012) 0.7286

Total All 699 / 831 39 / 40 1.1591 (0.7544-1.7809) 0.5023

89

Table 8: ≤ 3 hours bed rest post-diagnostic coronary angiogram - total bleeding complications (bleeding / haematoma)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Bleeding Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 5 1.6000 (0.5604-4.5685) 0.5547

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Pollard 2003 343 / 362 2.5 / 4.5 69/55 1.324 (0.9595-1.8271) 0.0929

Singh 1998 336 / 185 3 / 6 21 / 12 0.9635 (0.4851-1.9137) 1

Wood 1997 124 / 167 2.5 / 4 7 / 13 0.7252 (0.2981-1.7642) 0.4961

Total RCT’s 1191 / 1026 120 / 100 1.0338 (0.8036-1.3299) 0.7913

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Mah 1999 408 / 472 3 / 5 53 / 115 0.5332 (0.3961-0.7177) 0.00002

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Total Non-RCT’s 789 / 991 69 / 134 0.6468 (0.4913-0.8514) 0.0016

Total All 1910 / 2017 189 / 234 0.8529 (0.7116-1.0224) 0.0848

90

Table 9: ≤ 3 hours bed rest post-diagnostic coronary angiogram - total vascular complications (total bleeding / pseudoaneurysm)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Vascular Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 6 1.3333 (0.4973-3.5749) 0.7749

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Pollard 2003 343 / 362 2.5 / 4.5 70 / 56 1.3192 (0.9591-1.8147) 0.0949

Singh 1998 336 / 185 3 / 6 21 / 13 0.8894 (0.4560-1.7347) 0.8532

Wood 1997 124 / 167 2.5 / 4 7 / 14 0.6734 (0.2801-1.6187) 0.4933

Total RCT’s 1191 / 1026 121 / 104 1.0023 (0.7818-1.2849) 1

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Mah 1999 408 / 472 3 / 5 53 / 115 0.5332 (0.3961-0.7177) 0.00002

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Total Non-RCT’s 789 / 991 69 / 134 0.6468 (0.4913-0.8514) 0.0016

Total All 1910 / 2017 190 / 238 0.8430 (0.7041-1.0094) 0.0625

91

Table 10: ≤ 4 hours bed rest post-diagnostic coronary angiogram - total bleeding complications (bleeding / haematoma)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Bleeding Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Bogart 1999 100 / 100 4 / 6 1 / 2 0.5 (0.0461-5.4265) 1

Chair 2007 43 / 43 4 / 12-24 0 / 1 0 (0-Not Defined) 1

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 5 1.6000 (0.5604-4.5685) 0.5547

Keeling 1996 43 / 43 4 / 6 1 / 0 Infinity (Not Defined -Infinity) 1

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Lim 1997 100 / 100 4 / 6 47 / 44 1.0682 (0.7884-1.4472) 0.7765

Pollard 2003 343 / 362 2.5 / 4.5 69/55 1.324 (0.9595-1.8271) 0.0929

Singh 1998 336 / 185 3 / 6 21 / 12 0.9635 (0.4851-1.9137) 1

Wood 1997 124 / 167 2.5 / 4 7 / 13 0.7252 (0.2981-1.7642) 0.4961

Total RCT’s 1407 / 1312 139 / 147 0.8817 (0.7081-1.098) 0.2598

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Mah 1999 408 / 472 3 / 5 53 / 115 0.5332 (0.3961-0.7177) 0.00002

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Wang 2001 41 / 41 4 / 6 0 / 1 0 (0- Not Defined) 1

Total Non-RCT’s 830 / 1032 69 / 135 0.6355 (0.4826-0.8369) 0.0011

Total All 2237 / 2344 238 / 282 0.8843 (0.7517-1.0404) 0.1380

92

Table 11: ≤ 4 hours bed rest post-diagnostic coronary angiogram – total vascular complications (total bleeding / pseudoaneurysm)

Study ID Numbers (Experimental/Control)

Hours Bed rest (Experimental/Control)

Total Number Vascular Complications (Experimental/Control)

Risk Ratio (95% CI) P-value

Baum 1996 101 / 104 2 / 4 5 / 3 1.7162 (0.4211-6.9939) 0.4938

Bogart 1999 100 / 100 4 / 6 1 / 3 0.3333 (0.0353-3.1505) 0.6212

Chair 2007 43 / 43 4 / 12-24 0 / 1 0 (0- Not Defined) 1

Farmanbar 2008 60 / 60 2 / 6 1 / 2 0.5000 (0.0466-5.3684) 1

Hoglund 2011 52 / 52 1.5 / 5 8 / 6 1.3333 (0.4973-3.5749) 0.7749

Keeling 1996 43 / 43 4 / 6 1 / 0 Infinity (Not Defined -Infinity) 1

Logemann 1999 105 / 96 2 / 6 9 / 10 0.8229 (0.3493-1.9387) 0.8100

Lim 1997 100 / 100 4 / 6 47 / 45 1.0444 (0.7734-1.4105) 0.8872

Pollard 2003 343 / 362 2.5 / 4.5 70 / 56 1.3192 (0.9591-1.8147) 0.0949

Singh 1998 336 / 185 3 / 6 21 / 13 0.8894 (0.4560-1.7347) 0.8532

Wood 1997 124 / 167 2.5 / 4 7 / 14 0.6734 (0.2801-1.6187) 0.4933

Total RCT’s 1407 / 1312 170 / 153 1.0361 (0.8440-1.2719) 0.7401

Best 2010 193 / 402 1.5 / 3-4 6 / 12 1.0415 (0.3968-2.7332) 1

Mah 1999 408 / 472 3 / 5 53 / 115 0.5332 (0.3961-0.7177) 0.00002

Roebuck 2000 188 / 117 2 / 4 10 / 7 0.8891 (0.3480-2.2712) 1

Wang 2001 41 / 41 4 / 6 0 / 1 0 (0- Not Defined) 1

Total Non-RCT’s 830 / 1032 69 / 135 0.6355 (0.4826-0.8369) 0.0011

Total All 2237 / 2344 239 / 288 0.8696 (0.7399-1.0219) 0.0891

93

Table 12: Results – Randomised Controlled Trials Only

94

Table 13: Results – All Included studies

95

Appendices

Appendix I. Flowchart for inclusion / exclusion of studies

Databases searched for relevant articles • Wiley InterScience (69 articles) • CINAHL (85 articles) • PubMed includes MEDLINE (94 articles) • Proquest Central (463 articles) • Science Direct (18 articles) • Cochrane Database of Systematic

Reviews (0 found)

Articles excluded after reading title and abstract of the articles (709 articles)

Full text articles sought for more detailed evaluation (18 articles)

Studies excluded from full text articles (3 articles)

Articles included in Systematic Review (15 articles)

Figure 1.

96

Appendix II. Search strategies

CINAHL, Science Direct, Proquest Central

(coronary OR cardiac) AND (angiogra* OR catheteri*) AND (back pain OR mobili* OR

discomfort OR ambulation)

Limits: in abstract or title; published 1996 – 2012; English

PubMed (includes MEDLINE)

1. clinical trial(Publication Type)

2. english(Language)

3. coronary(Title/Abstract)

4. cardiac(Title/Abstract)

5. (mobilisation(Title/Abstract)) OR mobilization(Title/Abstract)

6. ambulation(Title/Abstract)

7. "back pain"(Title/Abstract)

8. discomfort(Title/Abstract)

9. position*(Title/Abstract)

10. angiogram(Title/Abstract)

11. catheterisation(Title/Abstract) OR catheterization(Title/Abstract)

12. (#3 OR #4) AND #1

13. (#5 OR #6 OR #9) AND #1

14. (#7 OR #8) AND #1

15. (#10 OR #11) AND #1

16. #12 AND #15

17. #16 AND #14

18. #16 AND #13

19. #17 AND #18

20. #19 AND #2

21. #19 NOT #2

97

Appendix III. Criteria and definitions for risk of bias assessment

The Cochrane Collaboration’s tool for assessing bias (Higgins & Altman, 2008) was

used as the base for assessing bias in the studies includes in this review. A ‘Yes’

indicates a low risk of bias and a ‘No’ indicates a high risk of bias.

1. Was the allocation sequence randomly generated?

Does the study reveal how the participants were allocated to each group in the

trial? If a recognised randomisation method such as computer randomisation, dice

roll or coin toss was used, the study scored a ‘Yes’. If the authors declared that the

method was not randomised, such as a retrospective / prospective trial, the study

scored a ‘No’. If it was unclear as to whether the method was randomised or not

stated, the study scored as ‘Unclear’.

2. Was allocation adequately concealed from both participants and clinicians by

having a secure schedule of randomization?

Did anybody involved in the treatment of the participant have any idea of which

group the patient was to be randomised into before randomisation took place? If

concealment from randomisation took place, the study scored a ‘Yes’. If the study

stated that the allocation was known prior, the study scored a ‘No’. If it was not

mentioned or unclear, then the study was scored as ‘Unclear’.

3. Were participants, clinicians and outcome assessors blinded to the allocated

intervention adequately during the study?

To score a ‘Yes’, all of the participants, clinicians and outcome assessors had to be

blinded to the allocated intervention group. If only one or two were blinded then the

study scored a ‘No’. If it was unclear or not mentioned, the study scored as

‘Unclear’.

4. Are results reported for everyone who entered the trial?

Do the numbers given in the results equal the numbers of participants who entered

the trial? If they do, the study scores a ‘Yes’. If not, the study scores a ‘No’. If

unclear the study scored as ‘Unclear’.

5. Was incomplete outcome data adequately addressed?

Did the study mention all participants who left the trial and reasons for why they

left? Were their results presented alongside those who did have full data acquired?

98

If they did, the study scored a ‘Yes’. If there was data missing, the study scored a

‘No’. If unclear the study scored as ‘Unclear’.

6. Were patients analysed in the groups to which they were randomised?

Did the participants stay in the same groups that they were first randomised? If they

did, the study scored a ‘Yes’. If participants switched groups, the study scored a

‘No’. If it was unclear whether they stayed in the same group, the study scored as

‘Unclear’.

7. Are reports of the study free of selective outcome reporting?

Did the study provide all of the outcomes that they stated they would in the study

methodology? Was the same weight applied to all outcomes despite them being

positive or negative? If they were, then the study scored a ‘Yes’. If they were not,

the study scored a ‘No’. If it was unclear, the study scored as ‘Unclear’.

8. Was the study apparently free of other problems that could put it at a high risk of

bias?

Could any other sources of bias be ascertained from analyzing the study? If the

study was free of any further bias, it scored a ‘Yes’. If there was other bias

assessed, the study scored a ‘No’. If it was unclear about any matter mentioned in

the study that could lead to bias, the study was assessed as ‘Unclear’.

99

Appendix IV. Systematic Review Protocol

Early mobilisation after femoral approach diagnostic coronary angiography or angioplasty to reduce back pain

Kelly L. Burn1, Bob Marshall2, Gill Scrymgeour2

1Interventional Cardiology, Capital and Coast Health District Health Board, Wellington,

New Zealand

2Eastern Institute of Technology, Taradale, Napier, Hawke’s Bay, New Zealand

Contact address: Kelly Burn, Interventional Cardiology, Wellington Hospital, Riddiford

Street, Newtown, Wellington, New Zealand. kelly.burn@ccdhb.org.nz

Background

Coronary heart disease remains the leading cause of death in New Zealand at 22%,

with rates of disease in Maori and Pacific people over twice the rate of other New

Zealanders. It accounts for over 25,000 inpatient admissions and nearly 4,000 day-

case admissions in New Zealand per year (Hay, 2004). Cardiac catheterization is a

common procedure undertaken to assess for and treat coronary heart disease (Chair,

Li & Wong, 2004).

Coronary angiography and angioplasty / stenting is most commonly carried out via a

5F – 7F arterial sheath inserted into the patient’s femoral artery. Through this sheath,

interventional devices such as catheters, wires, balloons and stents are passed. To

prevent acute thrombosis in the coronary vessels, the patient may be administered

anticoagulation medication such as heparin. The femoral sheath is commonly removed

4-6 hours post-procedure and then the patient remains on bed rest for a further 4-6

hours (Sabo, Chlan & Savik, 2008) to reduce the chances of bleeding from the groin

site (Chair, Taylor-Piliae, Lam & Chan, 2003).

Due to this enforced supine bed rest, immobilization and restricted positioning,

patients’ frequently experience back pain (Chair et al., 2003). Prolonged bed rest

causes pressure to be exerted continuously onto the same back muscles, causing

muscle fatigue and weakness. This fatigue causes back pain due to back spasms

(Chair et al., 2004). Studies have shown that bed rest and positioning regimes are

100

based on tradition rather than research (Rezaei-Adaryani, Ahmadi & Asghari-

Jafarabadi, 2009). The aim of this systematic review is to ascertain the earliest time a

patient can mobilise after angiography to minimize or prevent this back pain, whilst

keeping the risk of vascular complications low.

The purpose of the systematic review is to collate quasi-randomised controlled trials,

non randomised controlled trials and randomized controlled trials looking at early

mobilisation to reduce back pain in patients who have a femoral artery approach

coronary angiogram or angioplasty. There is no Cochrane Collaboration systematic

review on this topic.

Objectives

The aim of this Systematic Review and Meta-Analysis to ascertain whether it is safe for

nurses to mobilise patients out of bed four hours or earlier after a femoral approach

coronary angiogram without the use of a vascular closure device, in order to reduce

back pain whilst not increasing the risk of vascular complications at the puncture site.

Methods

Types of studies

Both randomised and quasi-randomised trials, which contain a control group and at

least one experimental group, will be included. Studies with a qualitative methodology

or those that do not contain a control group will not be included. They must be

published in the last 15 years and be in English.

Types of participants and interventions

Studies that include adult patients over 18 years, both male and female, will be

included. The participants must have undergone a femoral approach coronary

angiogram or angioplasty.

Types of outcome measures

Studies that will be included in the review will have primary outcome measures of

vascular complications including bleeding, haematoma (including retroperitoneal

haematoma) and pseudo-aneurysm. It would also be preferable if the studies

measured as secondary outcomes levels of back pain, discomfort and satisfaction.

101

Search Methods

Databases that will be searched for studies will include CINAHL, PubMed, Proquest

and the Cochrane Database. Reference lists of already acquired research will also be

searched for studies. Search terms that will be used include (coronary OR cardiac)

AND (angio* OR catheteri*) AND (back pain OR position* OR mobili*ation OR

experiences OR discomfort). Search limits that will be placed will include studies

reported in the English language, only human research, involve adults over 18 years

old only and have been published in last 15 years.

Data collection and analysis

Once an exhaustive literature search has been undertaken, studies that have been

obtained will be vetted against the inclusion and exclusion criteria for this review. The

article will be thoroughly scrutinised by three individual assessors and if it was suitable

it will go for data collection. Data collection from each study will take place on the form

included as Table 15. Either the risk ratios or odds ratios and 95% confidence intervals

will be calculated for all studies. Studies will then be grouped into similar mobilisation

times and results collated. These results will be presented in both Tables and on a

Forest Plot graph. Conclusions will be drawn from these.

Assessment of bias in studies

The risk of bias will be assessed by all three of the review authors, with a consensus

used when disagreements occurred. The Cochrane Collaboration’s tool will be used for

assessing bias (Higgins & Altman, 2008) to assess the risk of bias of the selected

randomised and quasi-randomised controlled trials (Table 2). Each of the criteria will

be scored ‘yes’, ‘no’ or ‘unclear’, depending on the information supplied in the report. A

Risk of Bias summary table for the review will be completed.

Sources of support:

Dalice A. Sim (PhD), Statistical Consultant from the School of Mathematics, Statistics

and Operation Research at Victoria University will provide assistance with the

statistical analysis in this Systematic Review and Meta-Analysis.

102

Tables

Table 14: Risk of bias assessment tool Bias Quote: Comment: Was the allocation sequence randomly generated?

Was allocation adequately concealed from both participants and clinicians by having a secure schedule of randomization?

Were participants, clinicians and outcome assessors blinded to the allocated intervention adequately during the study?

Are results reported for everyone who entered the trial?

Was incomplete outcome data adequately addressed?

Were patients analysed in the groups to which they were randomised?

Are reports of the study free of selective outcome reporting?

Was the study apparently free of other problems that could put it at a high risk of bias?

Table 15: Data Collection Tool Source Study ID Report ID Review author ID Citation Eligibility Eligible for review Reason for exclusion. Methodology Research question / aim of study Country of study Setting of study/numbers of centres Study methodology Study duration Participants Control: Experimental

Group 1: Experimental Group 2:

Inclusion criteria Exclusion criteria Socio-demographics Ethnicity Total number participants Number of participants allocated to each group

Number of participants accounted for in results of each group

Age range / Mean age Gender distribution (% Women) Tool for comparison of demographic data between groups

103

Figure given for statistically significant differences in demographics between groups

Results of comparison of demographic data between groups

Interventions Control: Experimental Group 1:

Experimental Group 2:

Specific intervention and details. Were specified procedures or components of the intervention implemented as planned?

Outcomes Control: Experimental Group 1:

Experimental Group 2:

Follow up and time points (i) collected; (ii) reported.

Outcomes to be measured, with diagnostic criteria if relevant.

Unit of measurement (if relevant). For scales: upper and lower limits, and whether high or low score is good.

Tools used for result analysis between groups

Results Control: Experimental Group 1:

Experimental Group 2:

Dichotomous data (OR,RR,RD) Continuous data (MD, SMD) Estimate of effect with 95% confidence interval; P value; Standard error, Standard deviations.

Subgroup analyses. Miscellaneous Did the authors declare any other potential conflict of interest?

Funding source for study Ethics Approval Study limitations Key conclusions from the study authors

Miscellaneous comments from the study authors

References to other relevant studies Notes Miscellaneous comments by the review authors.

References

Chair, S. Y., Li, K. M., & Wong, S. W. (2004). Factors that affect back pain among Hong Kong Chinese patients after cardiac catheterization. European Journal of Cardiovascular Nursing, 3(4), 279-285.

Chair, S. Y., Taylor-Piliae, R. E., Lam, G., & Chan, S. (2003). Effect of positioning on back pain after coronary angiography. Journal of Advanced Nursing, 42(5), 470-478.

Hay, D. (2004). Cardiovascular Disease in New Zealand. The National Heart

Foundation of New Zealand

104

Higgins, J. P. T., & Altman, D. G. (2008). Chapter 8: Assessing risk of bias in included studies. In J.P.T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 359-387). Chichester, England: John Wiley & Sons.

Rezaei-Adaryani, M., Ahmadi, F., & Asghari-Jafarabadi, M. (2009). The effect of

changing position and early ambulation after cardiac catheterization on patients' outcomes: A single-blind randomized controlled trial. International Journal of Nursing Studies, 46(8), 1047-1053.

Sabo, J., Chlan, L. L., & Savik, K. (2008). Relationships among patient characteristics,

comorbidities, and vascular complications post-percutaneous coronary intervention. Heart & Lung: The Journal of Acute and Critical Care, 37(3), 190-195.

Declarations of interest

There is no potential conflict of interest among the review authors.

105

Chapter 5

DISCUSSION

The previous chapter contained the entire Systematic Review following the Cochrane

Collaboration framework. This chapter presents a discussion on the Systematic Review

findings and how these can impact on practice for nursing and other health professions

using an evidence-based approach.

The aim of this research was to ascertain what factors can be put in place by nursing

staff caring for patients after a femoral approach coronary angiogram to reduce back

pain, taking into account patient safety. When evaluated in the literature review, earlier

mobilisation proved to be the most successful intervention that a nurse can implement

to reduce back pain.

Literature review findings

The literature review found five different ways to potentially reduce back pain after a

diagnostic coronary angiogram procedure. By knowing the patient factors that have the

potential to increase back pain, a nurse can pre-empt the potential pain and administer

prophylactic pain relief or perhaps suggest to a doctor that a radial artery approach is

used or a vascular closure device is deployed post-procedure. By changing a patient’s

position whilst they are on bed rest such as side to side lying or back of bed elevation,

the nurse can help to prevent or relieve back pain after the angiogram procedure.

Medical staff could consider a radial artery approach to their coronary angiogram or

use a vascular closure device post-procedure, meaning the patient can sit up and

mobilise much quicker after their angiogram.

However, as discussed in the literature review in Chapter 2, there are limitations to all

of the above, and the gold standard and preferred access method around the world

remains the femoral artery (Hoglund et al., 2011). Vascular closure device use post-

femoral approach angiogram has not been widely adopted, despite their availability

(Dauerman et al., 2007). Only three of the five ways of preventing or reducing back

pain post angiography can be implemented by nursing staff. The decision to use a

radial artery approach or vascular closure device is made solely by the doctor involved.

Therefore, the Systematic Review and Meta-Analysis was undertaken to answer the

106

research question as to the safety of earlier mobilisation, four hours or earlier post

femoral approach coronary angiogram, without using a vascular closure device.

Early mobilisation

Earlier mobilisation has many benefits to both the patient and nursing staff. It proved to

be a very successful way of reducing back pain after femoral approach coronary

angiogram, without increasing vascular complications, when analysed in the literature

review and then when formally evaluated in the Systematic Review. Getting out of bed

sooner also decreases urinary discomfort due to the fact that using a bedpan or urine

bottle whilst lying flat in bed can be difficult, leading to urinary retention and discomfort

(Chair et al., 2007). The intervention of early mobilisation can reduce costs (Best et al.,

2010), does not need any extra equipment or extra training, means a patient regains

their independence earlier leading to increased patient satisfaction and a free up of

nursing resources (Chair et al., 2007; Wang et al., 2001). When patients mobilise

sooner, without complications, they can be discharged earlier also freeing up nursing

resources and beds with the potential of increased patient throughput (Koch et al.,

1999).

The results of this Systematic Review showed that earlier mobilisation after femoral

approach coronary angiogram, without deployment of a vascular closure device, has a

positive benefit of reducing back pain related to lying in bed, confirming the findings in

the literature review. Of the four studies that directly measured back pain, three

showed very favourable results in their early mobilisation groups, when measuring

back pain compared to the control groups (Chair et al., 2007; Hoglund et al., 2011;

Wood et al., 1997). Only Wang et al. (2001) found higher levels of pain in the early

mobilisation group, but that was due to the inclusion of three patients with chronic back

pain who experienced severe pain on bed rest. Six of the other studies did not directly

measure back pain, but mentioned anecdotal comments that nurses had received

during the studies from patients regarding decreased back pain in the earlier

mobilisation groups. It was clearly shown in the Systematic Review that back pain is

significantly reduced when patients mobilise earlier after femoral approach coronary

angiograms.

The results of the Systematic Review also clearly showed that mobilisation at one and

a half hours to four hours after femoral approach coronary angiogram, without

deployment of a vascular closure device, has levels of vascular complication rates that

107

are not significantly different from those when the patient mobilises at six hours or later.

Therefore, this Systematic Review found it was probably safe to mobilise patients as

early as one and a half hours after a femoral approach diagnostic angiogram.

Comparison to the findings by Chair et al. (2008)

Similar favourable results to earlier mobilisation were found in the Systematic Review

by Chair et al. (2008), which was presented using the Joanna Briggs Institute format

(Joanna Briggs Institute, 2012). They concluded that there was no evidence of

decreased bleeding and haematoma formation in patients who remained in bed for

longer than three hours compared to those who mobilised earlier at three hours. They

found a decrease in back pain in those who mobilised earlier than three hours, as did

this Systematic Review. Chair et al. (2008) also state that there is a possibility of

benefit to back pain reduction with early mobilisation at two hours post femoral

approach coronary angiogram. However, they suggest further study on levels of

vascular complications at this earlier time frame.

This Systematic Review included three studies that were published after the study by

Chair et al. (2008) and one non-randomised controlled trial published in 1999 that, due

to its non-randomisation, did not fit the inclusion criteria set out by Chair et al. (2008).

This study was by Mah et al. (1999) that gave very favourable results to mobilisation at

three hours vs. five hours, in favour of the early mobilisation group (p = 0.00002).

Although back pain was not directly measured in this study, nurses noted that patients

made positive comments about comfort levels in the early ambulation group,

experiencing less discomfort than the control group. Best et al. (2010) conducted a

retrospective / prospective non-randomised controlled trial, which gave non-significant

differences in vascular complications between mobilisation at one and a half hours vs.

three to four hours (RR 1.0415; 95% CI 0.3968-2.7332; P 1.000). Anecdotal evidence

from nurses in this study found more favourable comments from patients in the early

mobilisation group around back pain and discomfort. However, for the same reason as

the study by Mah et al. (1999), this study would not have been included in the

Systematic Review of Chair et al. (2008).

The studies of Farmanbar et al. (2008) and Hoglund et al. (2011), would meet the

inclusion criteria of Chair et al. (2008) and when added to their Systematic Review,

would add weight to the conclusion that mobilisation four hours or less after femoral

approach coronary angiogram is both safe and potentially effective at reducing back

108

pain. Farmanbar et al. (2008) found a non-significant rate of vascular complications

when patients were mobilised at two hours vs. six hours (p = 1.000), as did Hoglund et

al. (2011), who mobilised patients at one and a half hours vs. five hours (p = 0.7749).

By including these two studies in the Systematic Review in this thesis, the

recommendations by Chair et al. (2007) to further study levels of vascular

complications at the earlier time frame of two hours or less have been followed,

evaluated and reported. Therefore, the findings by Chair et al. (2007) that there was no

evidence of decreased bleeding and haematoma formation in patients who remained in

bed for longer than three hours compared to those who mobilised earlier at three

hours, and that it is safe to mobilise at this time, can also be applied to mobilisation at

two hours, using the results from these two further studies. Hoglund et al. (2011) also

adds weight to the suggestion that earlier mobilisation potentially decreases the

incidence of back pain. As mentioned earlier, they found participants in their early

mobilisation group of one and a half hours experienced significantly less back pain at

the end of bed rest with a mean score of 1.5 / 10 compared to their control group who

mobilised at five hours and had a mean back pain score of 3.5 / 10. The result was

statistically significant at p < 0.001, favouring the early mobilisation group. Farmanbar

et al. (2008) did not measure back pain or discomfort.

Strengths of the Systematic Review

From the literature review, it was evident that early mobilisation after femoral approach

diagnostic coronary angiogram had already been researched in the form of eleven

randomised controlled trials, two quasi-randomised and two non-randomised trials.

Therefore, instead of carrying out further research on a topic that had already, clearly,

been well researched, a decision was made to perform a Systematic Review and Meta-

Analysis.

The topic of early mobilisation to reduce back pain is extremely relevant to nurses

today working in cardiac catheterisation laboratories and the wards that care for these

patients post-procedure. Numbers of coronary angiograms are increasing every year

(Galli and Palatnik, 2005) and have become one of the most frequently carried out

invasive procedures within hospitals (Leung, Hallani, Lo, Hopkins and Juergens, 2007).

After the angiographic procedure, the patient can remain on bed rest for at least 4-6

hours (Sabo, Chlan and Savik, 2008). It is evident back pain is a significant issue faced

by these patients on bed rest after a coronary angiogram (Augustin, de Quadros and

Sarmento-Leite, 2010; Chair et al., 2007; Hoglund, Stenestrand, Todt & Johansson,

109

2011; Pollard et al., 2003; Powers & Turner, 2000; Wang et al., 2001; Wood et al.,

1997). Leasure et al. (2008), Newhouse (2007) and Nursing Research (2012) state that

a hierarchy of evidence exists on which to base practice, ranked in order by the

strength of the evidence and that Systematic Reviews of rigorous randomised control

trials provide the strongest evidence. By completing this Systematic Review, using the

Cochrane Collaboration format, a strong piece of evidence has been created with

which to guide evidence-based practice. By combining these 15 studies, the strongest

and most powerful evidence for which a nurse can base their practice on has been

produced.

The Systematic Review contained 15 quantitative studies, 11 randomised controlled

trials, two quasi-randomised and two non-randomised trials, all with an intervention

group and control group. Deeks et al. (2008) discuss how only studies that are

homogeneous (i.e. similar in their participants, interventions, outcomes and study

design) should be compared in a Systematic Review and Meta-Analysis to provide

meaningful results. The studies in this Systematic Review were analysed for

homogeneity looking at study participants, interventions and outcome measures. All of

the studies analysed early mobilisation after femoral approach diagnostic coronary

angiogram. They all removed their femoral sheaths immediately with no prerequisites

except for no bleeding or haematoma at site. Haemostasis was gained using either

manual or mechanical compression but did not include vascular closure devices.

Medications used in each trial were very similar. Therefore, the studies were deemed

to have homogeneity and therefore could be compared in a Systematic Review. With

an amalgamation of the 15 studies, which included a total of 4581 participants, strong

evidence was gained, with a low level of bias due to the homogeneity of the included

studies.

Results from the Systematic Review and Meta-Analysis clearly showed no statistically

significant differences in levels of vascular complications in any of the studies between

the early mobilisation and control groups except in the case of Mah et al. (1999) who

showed significance results in favour of the earlier mobilisation group (p = 0.00002).

Publication bias was assessed to check for publication of only positive trials. Of the 15

studies included in this Systematic Review, eight studies showed a positive overall

result towards early mobilisation and six studies showed a negative overall result

towards early mobilisation. Therefore, publication bias was clearly not an issue in this

Systematic Review. Results from this Systematic Review have clearly shown that

mobilisation after femoral approach coronary angiogram, without deployment of a

110

vascular closure device, is as safe at one and a half hours to four hours mobilisation as

it is at six hours and may have a positive benefit of reducing back pain in patients

related to lying in bed.

Weaknesses of the Systematic Review

As part of the Cochrane Review process, levels of quality are allocated to individual

studies GRADE Working Groups grades of evidence (Schunemann et al., 2008). All 15

studies were assessed as having a moderate quality, due to four of the studies using a

quasi-randomised or non-randomised study design, which may increase the bias in the

studies (Best et al., 2010; Mah et al., 1999; Roebuck et al., 2000; Wang et al., 2001).

None of the 11 included randomised controlled trials had complete allocation

concealment or blinding and only two described their randomisation process (Chair et

al., 2007; Wood et al., 1997), which can also increase bias. A randomised controlled

trial is usually graded as a High quality rating on the using the GRADE levels of

evidence. The 15 studies in the Systematic Review would have been rated as High

quality, and therefore provided stronger evidence, had the studies all been randomised

controlled trials that revealed that they had complete allocation concealment, blinding

of at least medical staff performing the procedure and outcome assessors to reduce

study bias and declared their randomisation process. To overcome the bias created by

the quasi-randomised and non-randomised trials, after the initial analysis of all included

studies, each time frame was re-analysed including only the randomised controlled

studies.

Criteria for study inclusion in this Systematic Review meant included studies were

required to be published in English (due to the study authors inability to read other

languages). This creates potential bias, as studies published in other languages, that

otherwise would have met the inclusion criteria and could of added valuable

information, were excluded. There was also a 15 year limit applied to the studies i.e.

they needed to have been published in 1996 or later to be included, which meant older

studies, that again may have met the inclusion criteria and also added to the overall

outcome, were excluded.

Originally, the Systematic Review was to include early mobilisation for participants who

had undergone a Percutaneous Transluminal Coronary Angioplasty (PTCA). This was

stated in the study protocol prior to beginning the Systematic Review. However, only

five studies were identified in the literature search that included coronary angioplasty.

111

These studies proved to be too heterogenic in their methodology to be suitable for

comparison in this Systematic Review. Therefore the inclusion criteria for the

Systematic Review included only studies that looked at diagnostic coronary angiogram.

The results and recommendations from this Systematic Review would not be suitable

for application to patients post PTCA. Recommendations for further research on this

topic are included in the next chapter.

The research question for the Systematic Review (whether it is safe for nurses to

mobilise patients out of bed four hours or earlier after a femoral approach coronary

angiogram without the use of a vascular closure device, in order to reduce back pain

whilst not increasing the risk of vascular complications at the puncture site) had two

aspects that required research. Firstly, the early mobilisation and vascular complication

rates, which all of the included studies clearly researched. However, the second part of

the question, as to whether the early mobilisation decreased back pain was only

formally assessed by four of the studies (Chair et al., 2007; Hoglund et al., 2011; Wang

et al., 2001; Wood et al., 1997). Six other studies mentioned feedback from participants

in groups with a shorter time to mobilisation, about level of back pain, comfort or

satisfaction, without directly measuring it. Recommendations for further research into

this are discussed in the next chapter.

During the data collection process, it was noted that in two studies, it was unclear as to

whether vascular complications occurred before or after mobilisation. Both study

authors were contacted, with Jensen replying on behalf of Mah, Smith and Jensen

(1999), stating the complications were all after mobilisation. However, in the other

study by Pollard et al (2003), contact was not made. A decision was made to include

the study, as figures were given for the same time periods for both the control and

experimental group, and Chair et al (2008) had also included the study, with the same

figures, in their Systematic Review. This may have added a level of bias to the overall

Systematic Review, if in fact the figures were from prior to sheath removal through to

mobilisation as well. This potential bias was declared under Other Potential Bias in the

Results section of the Systematic Review.

The full text could not be located on one study that appeared to completely meet the

inclusion criteria, when the abstract was assessed (Tengiz et al., 2003). Despite an

extensive search, including emailing the magazine itself and searching 3 different

libraries databases, the full-text article of this study was unable to be located. The

abstract itself did not give enough information to result in the inclusion of the study.

112

Therefore I could not include it in my review and it was noted in the Characteristics of

Excluded Studies section of the Systematic Review. By excluding this study, valuable

information may have been lost that may of strengthened the argument towards earlier

mobilisation, or conversely supported later mobilisation.

Evidence-based practice

As nursing is a science and a profession, using evidence-based practice is

fundamental. It should always be based on the best available evidence (Newhouse,

2007). As discussed in Chapter 3, a hierarchy of evidence exists on which to base

practice, ranked in order by the strength of the evidence. Systematic Reviews of

rigorous randomised control trials sit at the top (Level I evidence) and provide the

strongest evidence. (Leasure et al., 2008; Newhouse, 2007; Nursing Research, 2012).

They are “the most powerful and useful evidence available” to guide practice (Stevens,

2001, p. 530). This Systematic Review followed the Cochrane Collaborations rigorous

and transparent framework and therefore gives the nurse and other health

professionals confidence that the results can be safely implemented into their practice.

Nurses and other health care professionals have little time to evaluate and synthesise

many different studies and can potentially rely on past experience and intuition in

making practice decisions (Acton, 2001). The value of this Systematic Review is that it

entails 15 individual studies, which are collated into one article of evidence with which

to guide their practice. This evidence can then be used to support the development of

best practice guidelines and practice protocols within an organisation involved with

femoral approach coronary angiograms. Implementation of these protocols involving

earlier mobilisation can then take place as part of evidence-based practice and

evaluation of the effectiveness of the new intervention can take place (Stevens, 2001).

This chapter has provided a discussion on ways to reduce back pain after femoral

approach diagnostic coronary angiogram and the intervention of early mobilisation. It

has presented the results from the Systematic Review included in this thesis and

compared the results to those of Chair et al. (2008) who also completed an earlier

Systematic Review on this topic. It also shows what this Systematic Review adds to

what was already known from the Systematic Review by Chair et al. (2008). The

chapter has presented the strengths and weaknesses of the Systematic Review and

how using this Systematic Review as a basis for evidence-based practice will benefit

nursing. The following chapter will provide a summary of this thesis, including

implications for practice and further research.

113

Chapter 6

CONCLUSION

This final chapter will summarise the findings of the Systematic Review, introduce

implications for practice and further research and conclude the thesis.

This thesis has clearly shown that back pain is an issue for patients who are required

to remain on bed rest post-femoral approach coronary angiography. Although other

methods have been presented to reduce the back pain, early mobilisation prior to four

hours is safe, simple to implement, has a proven effect on back pain reduction and has

added benefits such as decreased urinary discomfort, reduction in costs and means a

patient regains their independence earlier leading to increased patient satisfaction and

a free up of nursing resources.

Implications for practice

The Systematic Review included in this thesis aimed to provide an evidence-based

argument that time to mobilisation can be decreased to four hours or earlier in patients

post-femoral approach diagnostic coronary angiogram. It also aimed to show a

decrease in the potential for back pain whilst maintaining a risk of vascular

complications the same or lower than patients who mobilise at the traditional longer

regime. The results from the Systematic Review showed no statistically significant

difference in vascular complications between the control groups and the early

mobilisation out of bed groups at ≤ 2, 3 or 4 hours post femoral approach coronary

angiogram. Therefore, mobilisation after femoral approach coronary angiogram,

without deployment of a vascular closure device, may be as safe at one and a half

hours to four hours mobilisation as it is at six hours and may have a positive benefit of

reducing back pain related to lying in bed.

Due to the results of this study, the Interventional Cardiology Unit at Capital and Coast

Health Board’s Wellington Regional Hospital has decided to start mobilising patients

after a femoral approach diagnostic coronary angiogram at three hours post-procedure,

with back of bed elevation after one hour. If this proves successful in this unit, the

decision will then be made as to whether to trial a change of practice to two hours to

mobilisation. Before the Systematic Review was compiled, this unit mobilised patients

at four hours, with back of bed elevation after two hours.

114

Implications for research

The Systematic Review included in this thesis contained both randomised controlled

trials and quasi-randomised / non-randomised controlled trials, with a separate analysis

of just the randomised controlled trials. Although non-randomised controlled trials can

add valuable information to evidence-based practice, it would be beneficial if any

further studies on this topic were randomised controlled trials. This lowers the risk of

bias and ensures they can be included in a Systematic Review in the future.

As stated in the Systematic Review, further research on early mobilisation post femoral

approach coronary angiogram on a higher number of female patients would be of

benefit. Female patients make up fewer numbers in presentation for coronary

angiography as was seen in all fifteen of the studies included in the Systematic Review.

Because of this, numbers were lower in female participants and therefore the results

would be more heavily weighted towards men. Therefore, conducting further research

on primarily female patients could lead to results that are more applicable to the female

coronary angiography population.

Initially, the Systematic Review in this thesis was to also include patients who had

undergone a femoral artery approach coronary angioplasty. However, only five studies

were identified in the literature search that included coronary angioplasty. These

studies proved to be too heterogenic in their methodology to be suitable for comparison

in a Systematic Review. Therefore the inclusion criteria for the Systematic Review

included only studies that looked at diagnostic coronary angiogram. Further research

needs to be undertaken on the timing of mobilisation and femoral approach coronary

angioplasty before it can be concluded safe to mobilise these patients earlier. The

methodology and inclusion / exclusion criteria in these studies should have enough

homogeneity that the studies could be compared in a Systematic Review and Meta-

Analysis.

Conclusion

In conclusion, back pain is an issue for patients who are required to remain on bed rest

post-femoral approach coronary angiography. The results of the Systematic Review in

this thesis clearly showed that mobilisation at one and a half hours to four hours after

femoral approach coronary angiogram, without deployment of a vascular closure

device, has levels of vascular complication rates that are not significantly different from

those when the patient mobilises at six hours or later, therefore is safe. This

115

Systematic Review and the earlier one carried out by Chair et al. (2008) both found

that back pain is significantly decreased with earlier mobilisation. This information can

now be accessed by nurses and other health professionals working in the area of

coronary angiography and be used to change practice and protocols. This will lead to

increased patient satisfaction whilst maintaining safe practice.

116

REFERENCES

Acton, G. J. (2001). Meta-Analysis: a tool for evidence-based practice. Advanced Critical Care, 12(4), 539-545.

Adusumilli, S., Mah, J., & Richardson, A. (2011). Acute leg ischemia secondary to

embolization of an Angio-seal device. Diagnostic Interventional Radiology, 17, 98-100.

Agostoni, P., Biondi-Zoccai, G. G. L., De Benedictus, L., Rigattieri, S., Turri, M.,

Anselmi, M., . . . Hamon, M. (2004). Radial versus femoral approach for percutaneous coronary diagnostic and interventional procedures. American Journal of Cardiology, 44(2), 349-356.

Ainslie, M., MacDonald, J., & Smyth, J. (2011). A pain in the leg following angiography.

Journal of Invasive Cardiology, 23(3), E58-E60. Applegate, R. J., Sacrinty, M. T., Kutcher, M. A., Baki, T. T., Gandhi, S. K., Santos, R.

M., & Little, W. C. (2006b). Propensity score analysis of vascular complications after diagnostic cardiac catheterization and percutaneous coronary intervention 1998-2003. Catheterization and Cardiovascular Interventions, 67, 556-562.

Applegate, R. J., Sacrinty, M., Kutcher, M. A., Gandhi, S. K., Baki, T. T., Santos, R. M.,

& Little, W. C. (2006a). Vascular complications with newer generations of Angioseal vascular closure devices. Journal of Interventional Cardiology, 19(1), 67-74.

Applegate, R. J., Turi, Z., Sachdev, N., Ahmed, A., Szyniszewski, A., Foster, M., . . .

Shavelle, D. (2010). The Angio-Seal Evolution registry: outcomes of a novel automated Angio-Seal vascular closure device. Journal of Invasive Cardiology, 22(9), 420-426.

Arora, N., Matheny, M. E., Sepke, C., & Resnic, F. S. (2006). A propensity analysis of

the risk of vascular complications after cardiac catheterization procedures with the use of vascular closure devices. American Heart Journal, 153(4), 606-611.

Asinas, M. (2010). Coronary angiography: clinical information. Retrieved from

http://www.joannabriggs.edu.au Augustin, A. C., de Quadros, A. S., & Sarmento-Leite, R. E. (2010). Early sheath

removal and ambulation in patients submitted to percutaneous coronary intervention: a randomised clinical trial. International Journal of Nursing Studies, 47, 939-945.

Azmoon, S., Pucillo, A. L., Aronow, W., Ebrahimi, R., Vozzolo, J., Rajdev, A., . . .

Hjemdahl, C. (2010). Vascular complications after percutaneous coronary intervention following hemostasis with the Mynx vascular closure device versus the AngioSeal vascular closure device. Journal of Invasive Cardiology, 22(4), 175-178.

Bangalore, S., Vidi, V. D., Liu, C. B., Shah, P. B., & Resnic, F. S. (2011). Efficacy and

safety of the nitonol clip-based vascular closure device (StarClose) for closure of common femoral arterial cannulation at or near the bifurcation: a propensity score-adjusted analysis. Journal of Invasive Cardiology, 23(5), 194-199.

117

Baum, R. A., & Gantt, D. S. (1996). Safety of decreasing bedrest after coronary angiography. Catheterization and Cardiovascular Diagnosis, 39, 230-233.

Benninghoff, M. G., Amer, H., & Klugherz, B. D. (2003). Complications following use of

vascular hemostasis devices. American Journal of Cardiology, 92(6), 17. Best, D. G., Pike, R., Grainger, P., Eastwood, C. A., & Carroll, K. (2010). A prospective

study of early ambulation 90 minutes post-left heart catheterization using a retrospective comparison group. Canadian Journal of Cardiovascular Nursing, 20(4), 15-20.

Biancari, F., D’Andrea, V., DiMarco, C., Savino, G., Tiozzo, V., & Catania, A. (2010).

Meta-Analysis of randomized trials on the efficacy of vascular closure devices after diagnostic angiography and angioplasty. American Heart Journal, 159, 518-531.

Bogart, M. A. (1999). Cardiac catheterization: the effects of early ambulation on

patient comfort and groin complications (Doctor of Philosophy thesis, University of Missouri-Colombia, United States of America). Retrieved from http://search.proquest.com/docview/304437913/1370FD6A7D410E8FB92/1?accountid=34374

Bogart, M. A., Bogart, D. B., Rigden, L. B., Jung, S. C., & Liston, M. J. (1999). A

prospective randomised trial of early ambulation following 8 french diagnostic cardiac catheterization. Catheterization and Cardiovascular Interventions, 47, 175-178.

Brasselet, C., Blanpain, T., Tassan-Mangina, S., Deschildre, A., Duval, S., Vitry, F., …

Metz, D. (2008). Comparison of operator radiation exposure with optimized radiation protection devices during coronary angiograms and ad hoc percutaneous coronary interventions by radial and femoral routes. European Heart Journal, 29, 63-70.

Campbell Collaboration. (2012). About us. Retrieved from

http://www.campbellcollaboration.org/about_us/index.php Caputo, R. P., Tremmel, J. A., Rao, S., Gilchrist, I. C., Pyne, C., Pancholy, S., . . .

Patel, T. (2011). Transradial arterial access for coronary and peripheral procedures: executive summary by the transradial committee of the SCAI. Catheterization and Cardiovascular Interventions, 78, 823-839.

Centre for Reviews and Dissemination. (2012). Our databases. Retrieved from

http://www.york.ac.uk/inst/crd/index_databases.htm Cevik, C., Cemil, I., & Nugent, K. (2010). Radial artery access as an emerging factor

for decreasing mortality in cardiovascular interventions. Journal of Interventional Cardiology, 23(1), 95-99.

Chair, S. Y. (2004). Effects of early ambulation after cardiac catheterization (Doctor of

Philosophy thesis, University of Colorado, United States of America). Retrieved from http://search.proquest.com/docview/305044512/1370BC6239A5C0586B5

/1?accountid=34374

118

Chair, S. Y., Fernandez, R., Lui, M. H., Lopez, V., & Thompson, D. R. (2008). The clinical effectiveness of length of bed rest for patients recovering from trans-femoral diagnostic cardiac catheterization. International Journal of Evidence Based Healthcare, 6, 352-390.

Chair, S. Y., Li, K. M., & Wong, S. W. (2004). Factors that affect back pain among

Hong Kong Chinese patients after cardiac catheterization. European Journal of Cardiovascular Nursing, 3(4), 279-285.

Chair, S. Y., Taylor-Piliae, R. E., Lam, G., & Chan, S. (2003). Effect of positioning on

back pain after coronary angiography. Journal of Advanced Nursing, 42(5), 470-478.

Chair, S. Y., Thompson, D. R., & Li, S. K. (2007). The effect of ambulation after cardiac

catheterization on patient outcomes. Journal of Clinical Nursing, 16, 212-214. Chandrasekar, B., Doucet, S., Bilodeau, L., Crepeau, J., deGuise, P., Gregoire, J., . . .

Pasternac, A. (2001). Complications of cardiac catheterization in the current era: a single-center experience. Catheterization and Cardiovascular Interventions, 52, 289-295.

Chhatriwalla, A. K., & Bhatt, D. L. (2006). Walk this way: early ambulation after cardiac

catheterization – good for the patient and the health care system. Mayo Clinic Proceedings, 81(12), 1535-1536.

Chodor, P., Kurek, T., Kowalczuk, A., Swierad, M., Was, T., Honisz, G., . . . Kalarus, Z.

(2011). Radial vs femoral approach with StarClose clip placement for primary percutaneous coronary intervention in patients with ST-elevation myocardial infarction. RADIAMI II: a prospective, randomised, single centre trial. Kardiologia Polska, 69(8), 763-771.

Cochrane Collaboration. (2008). About us. Retrieved from

http://www.cochrane.org/about-us Cox, N., Resnic, F. S., Popma, J. J., Simon, D. I., Eisenhauer, A. C., & Rogers, C.

(2004). Comparison of the risk of vascular complications associated with femoral and radial access coronary catheterization procedures in obese versus nonobese patients. American Journal of Cardiology, 94, 1174-1177.

Cullen, L., & Adams, S. L. (2012). Planning for implementation of evidence-based

practice. Journal of Nursing Administration, 42(4), 222-230. Dabbs, A. D., Walsh, R. M., Beck, J., Demko, S. L., & Kanaskie, M. L. (1999). Nursing

assessment of patient readiness for ambulation after cardiac catheterization. Medsurg Nursing, 8(5), 309-314.

Dauerman, H. L., Applegate, R. J., & Cohen, D. J. (2007). Vascular closure devices.

Journal of American College of Cardiology, 50(17), 1617-1626. Deeks, J. J., Higgins, J. P. T., and Altman, D. G. (2008). Chapter 9: Analysing data and

undertaking meta-analyses. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 243-296). Chichester, England: John Wiley & Sons.

119

Dehghani, P., Mohammad, A., Bajaj, R., Hong, T., Suen, C. M., Sharieff, W., . . . Cheema, A. N. (2009). Mechanism and predictors of failed transradial approach for percutaneous coronary interventions. Journal of American College of Cardiology Cardiovascular Interventions, 2(11), 1057-1064.

Dehghani, P., Mohammad, A., Marcuzzi, D., Hong, T., Holmes, K., Evangelista, M. J., .

. . Cheema, A. (2010). Efficacy and long-term safety of StarClose for haemostasis of arterial puncture sites distal to common femoral artery bifurcation after percutaneous coronary interventions. Journal of Invasive Cardiology, 22(11), 505-510.

Dumont, C., Keeling, A., Bourguignon, C., Sarembock, I., & Turner, M. (2006).

Predictors of vascular complications post diagnostic cardiac catheterization and percutaneous coronary interventions. Dimensions in Critical Care Nursing, 25(3), 137-142.

Eggebrecht, H., Haude, M., Woertgen, U., Schmermund, A., Von Birgelen, C., Naber,

C., . . . Erbel, R. (2002). Systematic use of a collagen-based vascular closure device immediately after cardiac catheterization procedures in 1317 consecutive patients. Catheterization and Cardiovascular Interventions, 57, 486-495.

Egred, M. (2011). Feasibility and safety of 7-F radial approach for complex PCI.

Journal of Interventional Cardiology, 24(5), 383-388. Farman, M. T., Khan, N. U., Sial, J. A., Saghir, T., Rizvi, S. N., & Zaman, K. S. (2011).

Comparison of fluoroscopy time during coronary angiography and interventions by radial and femoral routes - can we decrease the fluoroscopy time with increased experience? An observational study. Anadolu Kardiyol Derg, 11, 607-612.

Farmanbar, R., Chinikar, M., Gozalian, M., Baghaie, M., Roshan, Z. A., &

Moghadamnia, M. (2008). The effect of post coronary angiography bed-rest time on vascular complications. The Journal of Tehran University Health Center, 4, 225-228.

Farooq, V., Mamas, M. A., Fath-Ordoubadi, F., & Fraser, D. (2011). The use of a

catheter extension system as an aid during transradial percutaneous coronary intervention of coronary artery bypass grafts. Catheterization and Cardiovascular Interventions, 78. 847-863.

Fineout-Overholt, E., Melnyk, B. M., & Schultz, A. (2005). Transforming health care

from the inside out: advancing evidence-based practice in the 21st century. Journal of Professional Nursing, 21(6), 335-344.

Flodgren, G., Rojas-Reyes, M. X., Cole, N., & Foxcroft, D. R. (2012). Effectiveness of

organisational infrastructures to promote evidence-based nursing practice (review). Retrieved from http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD002212.pub2/full

Fowlow, B., Price, P., & Fung, T. (1995). Ambulation after sheath removal: a

comparison of 6 and 8 hours of bedrest after sheath removal in patients following a PTCA procedure. Heart and Lung, 24(1), 28-37.

120

From, A. M., Bell, M. R., Rihal, C.S., & Gulati, R. (2011). Minimally invasive transradial intervention using sheathless guiding catheters. Catheterization and Cardiovascular Interventions, 78, 866-871.

Gall, S., Tarique, A., Natarajan, A., & Zaman, A. (2006). Rapid ambulation after

coronary angiography via femoral artery access: a prospective study of 1,000 patients. Journal of Invasive Cardiology, 18(3), 106-108.

Gallagher, R., Trotter, R., & Donoghue, J. (2010). Preprocedural concerns and anxiety

assessment in patients undergoing coronary angiography and percutaneous coronary interventions. European Journal of Cardiovascular Nursing, 9, 38-44.

Galli, A., & Palatnik, A. (2005). Ask the experts. Critical Care Nurse, 25(2), 88-95. Gawlinski, A. (2008). The power of clinical nursing research: engage clinicians,

improve patient’s lives, and forge a professional legacy. American Journal of Critical Care, 17, 315-326.

Gawlinski, A., & Becker, E. (2012). Infusing research into practice. Journal for Nurses

in Staff Development, 28(2), 69-73. Gilchrist, I. (2006). Is the Allen’s test accurate for patients considered for transradial

coronary angiography? Journal of American College Cardiology, 48(6), 1287. Gillane, O., & Pollard, M. (2009). Early ambulation of patients post-angiography with

femoral puncture. British Journal of Cardiology, 16, 137-140. Gogo, P. B. (2006). The evaluation and management of cardiogenic shock. Critical

Pathways in Cardiology: A Journal of Evidence Based Medicine, 5(1), 1-6. Goldich, G. (2011). Getting in sync with intra-aortic balloon pump therapy. Nursing2011

Critical Care, 6(3). 14-19. Green, S., & Higgins, J. P. T. (2008). Chapter 2: Preparing a Cochrane review. In J. P.

T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 11-30). Chichester, England: John Wiley & Sons.

Green, S., Higgins, J. P. T., Alderson, P., Clarke, M., Mulrow, C. D., & Oxman, A. D.

(2008). Chapter 1: Introduction. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 3-9). Chichester, England: John Wiley & Sons.

Gulanick, M., Bliley, A., Perino, B., & Keough, V. (1997). Patients’ responses to the

angioplasty experience: a qualitative study. American Journal of Critical Care, 6(1), 25-32.

Hamon, M., & Nolan, J. (2008). Should radial artery access be the gold standard for

PCI? Heart, 94(12), 1530-1532. Harrison, W., & Grines, C. (2011). New equipment and techniques enhance

capabilities of trans-radial PCI. Journal of Interventional Cardiology, 24(5), 413-415.

Hay, D. (2004). Cardiovascular Disease in New Zealand. Retrieved from

www.heartfoundation.org.nz

121

Higgins, J. P. T., & Altman, D. G. (2008). Chapter 8: Assessing risk of bias in included studies. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 359-387). Chichester, England: John Wiley & Sons.

Higgins, J. P. T., & Deeks, J. J. (2008). Chapter 7: Selecting studies and collecting

data. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 151-185). Chichester, England: John Wiley & Sons.

Higgins, J. P. T., & Green, S. (2008). Chapter 4: Guide to the contents of a Cochrane

protocol and review. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 51-79). Chichester, England: John Wiley & Sons.

Hildick–Smith, D. (2006). Use of the Allen’s test and transradial catheterization. Journal

of American College Cardiology, 48(6), 1287. Hoglund, J., Stenestrand, U., Todt, T., & Johansson, I. (2011). The effect of early

mobilisation for patient undergoing coronary angiography; a pilot study with focus on vascular complications and back pain. European Journal of Cardiovascular Nursing, 10(2), 130-136.

Hvelplund, A., Jeger, R., Osterwalder, R., Bredahl, M., Madsen, J. K., Jensen, J. S., . .

. Galatius, S. (2011). The Angio-Seal femoral closure device allows immediate ambulation after coronary angiography and percutaneous coronary intervention. EuroIntervention, 7, 234-241.

Joanna Briggs Institute. (2012). About us. Retrieved from

http://www.joannabriggs.edu.au Kanei, Y., Kwan, T., Nakra, N., Liou, M., Huang, Y., Vales, L., . . . Saito, S. (2011).

Transradial cardiac catheterization: A review of access site complications. Catheterization and Cardiovascular Interventions, 78, 840-846.

Keeling, A. W., Fisher, C. A., Haugh, K. H., Powers, E. R., & Turner, M. S. (2000).

Reducing time in bed after percutaneous transluminal coronary angioplasty (TIBS III). American Journal of Critical Care, 9(3), 185-187.

Keeling, A., Taylor, V., Nordt, L. A., Powers, E., & Fisher, C. (1996). Reducing time in

bed after cardiac catheterization (TIBS II). American Journal of Critical Care, 5(4), 277-281.

Koch, K. T., Piek, J. J., de Winter, R. J., Mulder, K., Davi, G. K., & Lie, K. I. (1997).

Early ambulation after coronary angioplasty and stenting with six french guiding catheters and low-dose heparin. American Journal of Cardiology, 80(8), 1084-1086.

Koch, K. T., Piek, J. J., de Winter, R. J., Mulder, K., Schotborgh, C. E., Tijssen, J. G.

P., & Lie, K. I. (1999). Two hour ambulation after coronary angioplasty and stenting with 6 F guiding catheters and low dose heparin. Heart, 81, 53-56.

122

Koreny, M., Riedmuller, E., Nikfardjam, M., Siostrzonek, P., & Mullner, M. (2004). Arterial puncture closing devices compared with standard manual compression after cardiac catheterization. Journal of the American Medical Association, 291(3), 350-357.

Krugman, M. (2012). Evidence-based practice. Journal for Nurses in Staff

Development, 27(6), 310-312. Lauck, S., Johnson, J. L., & Ratner, P. A. (2005). Self-care behaviour and factors

associated with patient outcomes following same-day discharge percutaneous coronary intervention. European Journal of Cardiovascular Nursing, 8, 190-199.

Leasure, A. R. Stirlen, J., & Thompson, C. (2008). Barriers and facilitators to use of

evidence-based best practices. Dimensions of Critical Care Nursing, 27(2), 74-82.

Lefebvre, C., Manheimer, E., & Glanville, J. (2008). Chapter 6: Searching for studies.

In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 95-150). Chichester, England: John Wiley & Sons.

Lehmann, C., & Hotaling, M. (2005). Saving time, saving money: a time and motion

study with contrast management systems. Journal of Invasive Cardiology, 17(2), 118-121.

Lehmann, R., Ehrlich, R., Weber, V., De Rosa, S., Gotarda, M. N. B., Schachinger, V.,

. . . Fichtlscherer, S. (2011). Implementation of the transradial approach for coronary procedures is not associated with an elevated complication rate and elevated radiation patient exposure. Journal of Interventional Cardiology, 24(1), 56-64.

Lepper, B. (2004). Percutaneous coronary interventions. Journal of Cardiovascular

Nursing, 19(5), 346-353. Leung, D. Y., Hallani, H., Lo, S. T., Hopkins, A. P., & Juergens, C. P. (2007). How

compliant are we with guidleines for coronary angiography in clinical practice? Internal Medicine Journal, 37, 699-704.

Levy, M. S., & Moussa, I. D. (2011). Transradial PCI in cardiogenic shock, the final

frontier. Catheterization and cardiovascular interventions, 78, 886-887. Lim, R., Anderson, H., Walters, M. I., Kaye, G. C., Norell, M. S., & Caplin, J. L.

(1997). Femoral complications and bed rest duration after coronary arteriography. American Journal of Cardiology, 80, 222-223.

Logemann, T., Luetmer, P., Kaliebe, J., Olson, K., & Murdock, D. K. (1999). Two

versus six hours of bed rest following left-sided cardiac catheterization and a meta-analysis of early ambulation trials. American Journal of Cardiology, 84, 486-488.

Louvard, Y., Sabatier, R., Benamer, H., Garot, P., Hildick-Smith, D., Loubeyre, C., . . .

Hamon, M. (2003). A randomized comparison of transradial and transfemoral approaches for coronary angiography and PTCA in octogenarians. American Journal of Cardiology, 92(6), 17.

123

Lupi, A., Lazzero, M., Plebani, L., Sansa, M., & Bongo, A. S. (2011). Safety and efficacy of the new Angio-Seal Evolution closure device: A single-centre experience. Journal of Invasive Cardiology, 23(4), 150-155.

McCabe, P. J., McPherson, L. A., Lohse, C. M., & Weaver, A. L. (2001). Evaluation of

nursing care after diagnostic coronary angiography. American Journal of Critical Care, 10(5), 330-340.

McCance, K. L., & Huether, S. E. (2002). Pathophysiology – the biologic basis for

disease in adults & children (4th ed.). Missouri, United States of America: Mosby, Inc.

Mah, J., Smith, H., & Jensen, L. (1999). Evaluation of 3-hour ambulation post cardiac

catheterization. Canadian Journal of Cardiovascular Nursing, 10(1-2), 23-30. Malik, I. (2008). Closure devices for femoral punctures. Heart, 94(5), 547-548. May, O., Schlosser, H., & Skytte, L. (2008). A randomized trial assessing the influence

of lying still or being allowed to move in the observational period following coronary angiography using femoral approach. Journal of Interventional Cardiology, 21(4), 347-349.

Ministry of Health. (2011). Provisional data: New Zealand mortality statistics: 2006 to

2009. Retrieved from http://www.health.govt.nz/publication/provisional-data-new-zealand-mortality-statistics-2006-2009

Narasimhan, S., & Gabriel, P. P. (2011). Location, location, location: is the star really

close. Journal of Invasive Cardiology, 23(5), 200-201. Newhouse, R. P. (2007). Creating infrastructure supportive of evidence-based nursing

practice: leadership strategies. Worldviews on Evidence-based Nursing, 4(1), 21-29.

Newland, J. (2012). Reflecting on evidence-based practice: international nurses day.

The Nurse Practitioner, 37(5), 5. Nikolsky. E., Mehran, R., Halkin, A., Aymong, E. D., Mintz, G. S., Lasic, Z., . . .

Dangas, G. (2004). Vascular complications associated with arteriotomy closure devices in patients undergoing percutaneous coronary procedures. Journal of the American College of Cardiology, 44(6), 1200-1209.

Nursing Research. (2012). Evidence based nursing. Retrieved from

http://nursingplanet.com/research/evidence_based_nursing.html O’Connor, D., Green, S., & Higgins, J. P. T. (2008). Chapter 2: Defining the review

question and developing criteria for including studies. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 83-94). Chichester, England: John Wiley & Sons.

Oxford University Press. (2012). Definition for back. Retrieved from

http://oxforddictionaries.com/definition/back?q=back

124

Paul, B. S., & Feeney, C. M. (2003). Combining the modified Allen’s test and pulse oximetry for evaluating ulnar collateral circulation to the hand for radial artery catheterization in the ED patient. California Journal of Emergency Medicine, 4, 89-91.

Pollard, S., Munks, K., Wales, C., Crossman, D., Cumberland, D., Oakley, G. & Gunn,

J. (2003). Position and mobilisation post angiography study (PAMPAS): a comparison of 4.5 hours and 2.5 hours bed rest. Heart, 89, 447-448.

Pooler-Lunse, C., Barkman, A., & Bock, B. F. (1996). Effects of modified positioning

and mobilization on back pain and delayed bleeding in patients who had received heparin and undergone angiography: a pilot study. Heart & Lung, 25(2), 117-123.

Pornratanarangsi, S., Boonlert, S., Duangprateep, A., Wiratpintu, D., Waree, W.,

Tresukosol, D., & Panchavinnin, P. (2010). The effectiveness of Siriraj leg lock brace on back pain after percutaneous coronary intervention: PCI. Journal of Medical Association Thailand, 93(1), S35-42.

Prada-Delgado, O., Estevez-Loureiro, R., Calvino-Santos, R., Barge-Caballero, E.,

Salgado-Fernandez, J., Pinon-Esteban, P., . . . Castro-Beiras, A. (2011). Safety and efficacy of femoral vascular closure devices in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction. American Heart Journal, 161, 1207-1213.

Rathore, S., & Morris, J. L. (2008). The radial approach: is this the route to take?

Journal of Interventional Cardiology, 21(5), 375-379. Rezaei-Adaryani, M., Ahmadi, F., & Asghari-Jafarabadi, M. (2009). The effect of

changing position and early ambulation after cardiac catheterization on patients' outcomes: A single-blind randomized controlled trial. International Journal of Nursing Studies, 46(8), 1047-1053.

Rezaei-Adaryani, M., Ahmadi, F., Mohamadi, E., & Asghari-Jafarabadi, M. (2009). The

effect of three positioning methods on patient outcomes after cardiac catheterization. Journal of Advanced Nursing, 65(2), 417-424.

Rickbeil, P., & Simones, J. (2012). Overcoming barriers to implementing evidence-

based practice. Journal for Nurses in Staff Development, 28(2), 53-56. Roebuck, A., Jessop, S., Turner, R., & Caplin, J. L. (2000). The safety of two-

hour versus four-hour bed rest after elective 6-french femoral cardiac catheterization. Coronary Health Care, 4, 169-173.

Rosenstein, G., Cafri, C., Weinstein, J. M., Yeroslavtsev, S., Abuful, A., Ilia, R., &

Fuchs, S. (2004). Simple clinical risk stratification and the safety of ambulation two hours after 6 french diagnostic heart catheterization. Journal of Invasive Cardiology, 16(3), 126-128.

Sabo, J., Chlan, L. L., & Savik, K. (2008). Relationships among patient characteristics,

comorbidities, and vascular complications post-percutaneous coronary intervention. Heart & Lung: The Journal of Acute and Critical Care, 37(3), 190-195.

125

Saito, S., Ikei, H., Hosokawa, G., & Tanaka, S. (1999). Influence of the ratio between radial artery inner diameter and sheath outer diameter on radial artery flow after transradial coronary intervention. Catheterization and Cardiovascular Interventions, 46, 173-178.

Sallam, M., Al-Hadi, H., Rathinasekar, S., & Chandy, S. (2009). Comparative study of

the radial and femoral artery approaches for diagnostic coronary angiography. Sultan Qaboos University Medical Journal, 9(3), 272-278.

Sallam, M. M., Ali, M., & Al-Sekaiti, R. (2011). Management of radial artery perforation

complicating coronary intervention: a stepwise approach. Journal of Interventional Cardiology, 24(5), 401-406.

Schunemann, H. J., Oxman, A. D., Vist, G. E., Higgins, J. P. T., Deeks, J. J., Glasziou,

P., & Guyatt, G. H. (2008). Chapter 12: Interpreting results and drawing conclusions. In J.P.T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 359-387). Chichester, England: John Wiley & Sons.

Sciahbasi, A., Fischetti, D., Piccioli, A., Patrizi, R., Sperduti, I., Colonna, G., . . . Lioy, E.

(2009). Transradial access compared with femoral puncture closure devices in percutaneous coronary procedures. International Journal of Cardiology, 137, 199-205.

Sciahbasi, A., Mancone, M., Cortese, B., Pendenza, G., Romagnoli, E., Fischetti, D., . .

. Lioy, E. (2011). Transradial percutaneous coronary interventions using sheathless guiding catheters: a multicenter registry. Journal of Interventional Cardiology, 24(5), 407-412.

Shroff, A. (2011). Vascular access closure devices: in search of perfect closure.

Journal of Invasive Cardiology, 23(4), 150-155. Singh, N., Kuganesan, K., Goode, E., & Ricci, A. J. (1998). The effect of early

ambulation on hematoma formation and vascular complications following 7 french diagnostic cardiac catheterization. Canadian Journal of Cardiology, 14(10), 1223-1227.

Sterne, J. A. C., Egger, M., & Moher, D. (2008). Chapter 10: Addressing reporting bias.

In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (pp. 297-333). Chichester, England: John Wiley & Sons.

Stevens, K. (2001). Systematic reviews: the heart of evidence based practice. AACN

Clinical Issues, 12(4), 529-538. Tagney, J., & Lackie, D. (2005). Bed-rest post-femoral arterial sheath removal – what

is safe practice? A clinical audit. Nursing in Critical Care, 10(4), 167-173. Tavris, D. R., Albrecht-Gallauresi, B., Lin, B., Rich, S. E., Shaw, R. E., Weintraub, W.

S., . . . Hewitt, K. (2004). Risk of local adverse events following cardiac catheterization by hemostasis device use and gender. Journal of Invasive Cardiology, 16(9), 459-64.

126

Tavris, D. R., Dey, S., Albrecht-Gallauresi, B., Brindis, R. G., Shaw, R., Weintraub, W., & Mitchel, K. (2005). Risk of local adverse events following cardiac catheterization by hemostasis device use – phase II. Journal of Invasive Cardiology, 17(12), 644-50.

Tengiz, I., Ercan, E., Bozdemir, H., Durmaz, O., Gurgun, C., & Nalbantgil, I. (2003). Six

hour ambulation after elective coronary angioplasty and stenting with 7F guiding catheters and low dose heparin. Kardiology Polska, 58(2), 93-97.

Tobias, M., Sexton, K., Mann, S., & Sharpe, N. (2006). How low can it go? Projecting

ischaemic heart disease mortality in New Zealand to 2015. New Zealand Medical Journal, 119(1232), 14-26.

Vaught, K. B., & Ostrow, C. L. (2001). Bed rest after percutaneous transluminal

coronary angioplasty: How much is enough? Dimensions of Critical Care Nursing, 20(4), 46-50.

Vlasic, W. (2004). An evidence-based approach to reducing bed rest in the invasive

cardiology patient population. Evidence Based Nursing, 7, 100-101. Waksman, R., King, S. B., Douglas, J. S., Shen, Y., Ewing, H., Mueller, L., . . .

Weintraub, W. S. (1995). Predictors of groin complications after balloon and new-device coronary intervention. American Journal of Cardiology, 75, 886-889.

Wang, S., Redeker, N. S., Moreyra, A. E., & Diamond, M. R. (2001). Comparison of

comfort and local complications after cardiac catheterization. Clinical Nursing Research, 10(1), 29-39.

Wood, R. A., Lewis, B. K., Harber, D. R., Kovack, P. J., Bates, E. R., & Stomel, R. J.

(1997). Early ambulation following 6 french diagnostic left heart catheterization: a prospective randomized trial. Catheterization and Cardiovascular Diagnosis, 42, 8-10.

Yilmaz, E., Gurgun, C., & Dramali, A. (2006). Minimizing short-term complications in

patients who have undergone cardiac invasive procedure: a randomised controlled trial involving position change and sandbag. Anadolu Kardiyol Derg, 7, 390-396.

127

APPENDICES

Appendix V: Research Notification Form Research Notification Form 1. Applicant details

1.1 Name of Applicant(s) Kelly Burn

1.2 Position of Applicant(s) MN Candidate

1.3 School and Faculty EIT Hawkes Bay

1.4 Contact Phone 0211727848

1.5 Supervisor (if applicable) Bob Marshall / Gill Scrymgeour

1.6 Project Title What factors reduce back pain in adult patients on bed rest after coronary angiography, whilst avoiding vascular complications due to femoral artery puncture?

1.7 Project Start Date and Duration

2 May 2011 - 22 June 2012

2. Attach the completed Low Risk Research Questionnaire 3. If your project has been approved by another Ethics Committee, include your

application to them as well as their final approval letter.

128

Appendix VI: Low Risk Research Questionnaire

LOW RISK RESEARCH QUESTIONNAIRE (Part A and Part B of this questionnaire must both be completed) Name: Kelly Burn Project Title: What factors reduce back pain in adult patients on bed rest after

coronary angiography, whilst avoiding vascular complications due to femoral artery puncture? (A Systematic Review)

Part A The statements below are being used to determine the risk of your project causing physical or psychological harm to participants and whether the nature of the harm is minimal and no more than is normally encountered in daily life. The degree of risk will then be used to determine the appropriate approval procedure. If you are in any doubt you are encouraged to submit an application to EIT’s Research Ethics and Approvals Committee. Does your Project involve any of the following? (Please answer all questions. Please indicate either YES or NO for each question) Risk of Harm 1. Situations in which the researcher may be at risk of harm. NO

2. Use of questionnaire or interview, whether or not it is anonymous which might reasonably be expected to cause discomfort, embarrassment, or psychological or spiritual harm to the participants.

NO

3. Processes or results that are potentially disadvantageous to a person or group, such as the collection of information which may expose the person/group to discrimination. NO

4. Collection of information of illegal behaviour(s) gained during the research which could place the participants at risk of criminal or civil liability or be damaging to their financial standing, employability, professional or personal relationships.

NO

5. Collection of blood, body fluid, tissue samples or other samples. NO

6. Any form of exercise regime, physical examination, deprivation (e.g. sleep, dietary). NO

7. The administration of any form of drug, medicine (other than in the course of standard medical procedure), placebo. NO

8. Physical pain, beyond mild discomfort. NO

9. Any EIT teaching which involves the participation of EIT students for the demonstration of procedures or phenomena which have a potential for harm. NO

Informed and Voluntary Consent 10. Participants whose identity is known to the researcher who give oral consent rather

than written consent (if participants are anonymous, you may answer No). NO

129

11. Participants who are unable to give informed consent. NO

12. Research on your own students/pupils. NO

13. The participation of children (seven (7) years old or younger). NO

14. The participation of children under sixteen (16) years old where parental consent is not being sought. NO

15. Participants who are in a dependent situation, such as people with a disability, or residents of a hospital, nursing home or prison or patients highly dependent on medical care.

NO

16. Participants who are otherwise vulnerable. NO

17. The use of previously collected information or biological samples for which there was NO explicit consent for this research. NO

Privacy/Confidentiality Issue 18. Any evaluation of EIT services or organisational practices where information of a

personal nature may be collected and where participants may be identified. NO

Deception 19. Deception of the participants, including concealment and covert observations. NO

Conflict of Interest 20. Conflict of interest situation for the researcher (e.g. is the researcher also the

lecturer/teacher/treatment-provider/colleague or employer of the research participants or is there any other power relationship between the researcher and the research participants?)

NO

Compensation to Participants 21. Payments or other financial inducements (other than reasonable reimbursement of

travel expenses or time) to participants. NO

Procedural 22. A requirement by an outside organisation (e.g. a funding organisation or a journal in

which you wish to publish) for EIT’s Research Ethics and Approvals Committee approval.

NO

Part B The statements below are being used to determine if your project requires ethical approval by a Regional Health and Disability Ethics Committee. (http://www.hrc.govt.nz/assets/pdfs/policy/ReferralGuidelines.pdf) In situations where you are not sure whether the research needs approval by an HDEC, you should seek an opinion from the Administrator of the relevant HDEC. (http://www.ethicscommittees.health.govt.nz/) Include a copy of your written response from the Administrator with your application. Does your Project involve any of the following? (It is important that you answer all questions. Please circle either YES or NO for each question)

130

23. The use of staff or facilities of a health provider (e.g. DHB, PHO or health NGO). NO

24. Support, directly or indirectly, in full or in part, by health provider funds (e.g. DHB, PHO or health NGO). NO

25. Participants who are patients/clients of, or health information about an identifiable individual held by, an organisation providing health services (for example, general practice, physiotherapy, occupational therapy, sports medicine), disability services, or institutionalised care.

NO

26. Requirement for ethical approval to access health or disability information about an identifiable individual held by the Ministry of Health, or held by any public or private organisation whether or not that organisation is related to health.

NO

Determine the type of approval procedure to be used (choose one option):

If you answer YES to any of the questions 1 to 22 (Part A) and NO to all questions in Part B Prepare an application for the EIT C’tee using the RAD form.

If you answer YES to any of the questions 23 to 26 (Part B) Prepare an application using the Health & Disability Ethics Committee Application Form

If you answer NO to all of the questions Complete the Research Notification Form