CLINICAL STUDY

Medical Emergencies and Cardiopulmonary Arrestsin Interventional Radiology

Gregory Nadolski, MD, Amy Praestgaard, MS,Richard D. Shlansky-Goldberg, MD, Michael C. Soulen, MD,S. William Stavropoulos, MD, Scott O. Trerotola, MD, and

Cormac Farrelly, MD

ABSTRACT

Purpose: To evaluate the circumstances and determine the outcomes of medical emergencies (MEs) and cardiopulmonaryarrests (CPAs) in patients undergoing interventional radiology (IR) procedures.

Materials and Methods: Retrospective review of all MEs and CPAs that occurred between July 2006 and December 2011 wasperformed. Procedure type, technical outcome, complications, etiology and location of ME/CPA, event outcome, and posteventmortality were collected.

Results: A total of 58 events occurred during 38,927 procedures (0.15%). Complete records were available for 55 events (43 MEs,12 CPAs) in 53 patients (mean age, 63 y; 58.5% male) during 37 inpatient (27 MEs, 10 CPAs) and 18 outpatient (16 MEs, twoCPAs) encounters. Seven events (13%; six MEs, one CPA) occurred before the start of the procedure, and 18 (33%; 16 MEs, twoCPAs) occurred in the periprocedural holding area. Thirty-five procedures (64%) were completed successfully. Forty-two patients(76%) were alive at discharge, 37 (67%) at 1 month, 26 (47%) at 3 months, and 23 (42%) at 1 year. Procedural complications wereattributed as the main cause of 22 MEs (51%) and one CPA (8%; P ¼ .018). The relative risk (RR) of an ME or CPA occurringduring a hemodialysis access case versus all other cases was 5.2 (95% confidence interval ¼ 3.02–8.95; P o .0001).

Conclusions: Although the incidence of MEs/CPAs in patients undergoing IR procedures is low, the 1-year mortality ratefollowing these events is high. MEs are significantly more likely than CPAs to be directly attributed to a procedural complication.The RR of MEs/CPAs is significantly higher in hemodialysis access interventions.

ABBREVIATIONS

ASA = American Society of Anesthesiologists, CPA = cardiopulmonary arrest, ICU = intensive care unit, IR = interventional

radiology, ME = medical emergency, PAE = pulseless electrical activity, RR = relative risk

Studies in the United States and other countries haveshown serious adverse in-hospital events such as medicalemergencies (MEs) and cardiopulmonary arrest (CPA)

& SIR, 2013

J Vasc Interv Radiol 2013; 24:1779–1785

http://dx.doi.org/10.1016/j.jvir.2013.07.027

From the SIR 2013 Annual Meeting.

None of the authors have identified a conflict of interest.

From the Division of Interventional Radiology and Department of Radiology(G.N., R.D.S.-G., M.C.S., S.W.S., S.O.T.) and Biostatistics Analysis Center,Center for Clinical Epidemiology and Biostatistics, and Department ofEpidemiology and Biostatistics (A.P.), University of Pennsylvania MedicalCenter, Philadelphia, Pennsylvania; and Division of Interventional Radiology(C.F.), Mater Misericordiae University Hospital, Dublin, Ireland. Received May8, 2013; final version received and accepted July 30, 2013. Address

correspondence to G.N., Division of Interventional Radiology, University ofPennsylvania Medical Center, 1 Silverstein, 3400 Spruce St., Philadelphia,Pennsylvania 19104; E-mail: gregory.nadolski@uphs.upenn.edu

are not rare, often iatrogenic, and associated withmorbidity and death (1). Cardiac arrest occurs in asmany as five of every 1,000 adult hospital admissions (2).Despite aggressive postresuscitation management, in-hospital ME and CPA are associated with poor out-comes. In a review of 49,130 patients with in-hospitalCPAs, 84% died before hospital discharge or experiencedlong-term disability (3).Signs of deterioration usually precede ME and CPA,

and more than half of cases are preventable (1,4).Examination of MEs and CPAs can identify conditionsand trends associated with their occurrence and provideguidance for prevention of future occurrences (5).In the present study, the circumstances and outcomes

of MEs and CPAs in patients undergoing interventionalradiologic (IR) procedures in a single institution arecharacterized, with the goal of describing the outcomesand identifying the patients at most risk.

Nadolski et al ’ JVIR1780 ’ Medical Emergencies and Cardiopulmonary Arrests in IR

MATERIALS AND METHODS

Patient Selection and PopulationInstitutional review board approval was granted for thepresent Health Information Portability and Account-ability Act–compliant study. All MEs and CPAs occur-ring in IR between July 2006 and December 2011 werereviewed. The events were identified by retrospectivelyreviewing a prospectively recorded internal database ofall MEs and CPAs occurring in the institution. Addi-tional information regarding the identified patients wasobtained by examining the records from a dedicated IRdatabase (Hi-IQ; ConexSys, Lincoln, Rhode Island).Patient demographics, comorbidities, American Societyof Anesthesiologists (ASA) status classification, proce-dure type, sedation type and dosage, technical outcome,complications, etiology and location of ME/CPA, eventoutcome, and postevent mortality were all collected byreviewing the electronic medical record and informationwithin the IR database. The etiology of the ME/CPAand its relationship to the procedure were obtained byretrospective review of the records from the IR database(Hi-IQ, ConexSys).Etiology of the event was categorized as cardiac

arrhythmia, hypoxia/respiratory failure, oversedation,hemorrhage, air embolism, sepsis, contrast agent reac-tion, and other. Oversedation was defined as a decline incardiopulmonary function that occurred after the admin-istration of moderate sedation and resolved with admin-istration of reversal agents. Hypoxia/respiratory failureunrelated to sedation was defined as a decline incardiopulmonary function that did not respond toadministration of reversal agents. Hemorrhage wasdefined as a bleeding event at the site of vascular access,site of intervention, or remote (ie, retroperitoneal hem-orrhage) that resulted in a decline in cardiopulmonaryfunction. ASA status and history of adverse reaction tosedation was obtained by reviewing the proceduralrecord, which documents the previous adverse reactionto anesthesia and ASA status as determined by thesupervising physician based on the patient’s history,condition, and airway assessment. The frequency ofME/CPA and the relative risk (RR) of ME/CPA byprocedure type were calculated.A total of 58 events occurred during 38,927 proce-

dures. Complete records were available for analysis in 55events (43 MEs, 12 CPAs) in 53 patients (mean age, 63y; 59% male) during 37 inpatient (10 CPAs, 27 MEs) and18 outpatient (16 MEs, two CPAs) encounters. Comor-bidities were present in 92.5% of all patients (n ¼ 49; 37MEs, 12 CPAs). Postevent follow up at 1 year wasavailable for all 55 events. The three events for whichcomplete records were not available were excluded fromanalysis, and all calculations were based on the remain-ing 55 events. One patient had two MEs in IR duringthis time period, and another experienced a CPAfollowed by an ME in less then 1 month, both during

IR procedures. Three patients who had a ME/CPAindicated a previous complication from anesthesia ormedications administered during a procedure (history ofcontrast agent reaction, n ¼ 2; previous CPA fromanesthesia, n ¼ 1).

DefinitionsPatient monitoring in IR includes measurement andrecording of heart rate, blood pressure, respiratory rate,and pulse oximetry by a dedicated IR nursing staffmember. Capnography is not used for patient monitor-ing. Preprocedural monitoring occurs every 30 minutesuntil the patient enters the procedure room. Intraproce-dural monitoring occurs every 5 minutes. Postproceduralmonitoring occurs every 15 minutes for the first hour,every 30 minutes for the second hour, and then hourlythereafter.CPA is defined as the sudden cessation of cardiac or

pulmonary activity such that the individual becomesunresponsive, without normal breathing or signs ofspontaneous circulation. To prevent progression tosudden death, Advanced Cardiac Life Support is ini-tiated. This definition encompasses the 2006 AmericanCollege of Cardiology/American Heart Association/Heart Rhythm Society definition of sudden cardiacarrest (6).An ME is defined as abrupt change in an individual’s

clinical status without CPA. Such changes in mental,cardiac, or pulmonary function have been shown to beantecedents in the majority of in-hospital cardiac arrestand risk factors for increased mortality (7–9).In our institution, the following treatment algorithms

apply:When an ME occurs, rapid response teams are alerted

to the ME by activation of a rapid response system. Therapid response team consists of a dedicated group ofhealth care providers consisting of a respiratory thera-pist, two registered nurses, and an attending physicianstaffing an intensive care unit (ICU). Additionally, on-call medical and surgical house officers respond to therapid response system alert. All members of the team arealerted immediately by a pager system in addition to ahospital-wide intercom alert. The rapid response teamworks in concert with the IR nurses, fellow, and attend-ing physician caring for the patient to address the event.When a CPA occurs in the IR section, in addition to

the immediate initiation of Advanced Cardiac LifeSupport by the IR nurse, fellow, and attending physi-cian, concomitant activation of the code alert system isperformed. The CPA alert is responded to by a distinctcode call team composed of the same type of providersas the rapid response team, with the addition of ananesthesiology resident and attending physician to assistin airway management. Again, all members of the teamare alerted immediately by a pager system in additionto a hospital-wide intercom alert. The code team works

Volume 24 ’ Number 12 ’ December ’ 2013 1781

in concert with the IR nurses, fellow, and attendingphysician caring for the patient to address the event.

Statistical AnalysisDifferences in survival between MEs and CPAs wereanalyzed by the Kaplan–Meier method. Differences inlength of stay between groups was analyzed by Wilcoxonrank-sum test. Remaining differences between groupswere analyzed by Fisher exact test and χ2 test. In allinstances, statistical significance was defined by a Pvalue lower than .05.

RESULTS

Etiology, Care Level, and SurvivalEtiologies for events included cardiac arrhythmia, hypo-xia/respiratory failure, oversedation, hemorrhage, airembolism, sepsis, contrast agent reaction, and other.Other etiologies included hypotension of unspecifiedcause (n = 2), seizure (n = 2), panic attack (n = 1), fall(n = 1), and transfusion reaction (n = 1), which alloccurred during MEs. Ventricular tachycardia, brady-cardia, and internally paced rhythms were each encoun-tered once among CPAs. The frequency of a specificetiology for MEs, CPAs, and all events are listed inTable 1. Multivariate analysis demonstrated nosignificant differences in etiology of events between theME and CPA groups (P ¼ .054). During the 12 CPAs,the presenting cardiac rhythm was pulseless electricalactivity (PEA) or asystole in 75% of cases (n ¼ 9).Thirty-six events occurred during the procedure (65%;

nine CPAs, 27 MEs). Twelve events occurred followingthe procedure in the periprocedural holding area (22%;two CPAs, 10 MEs). Seven events (13%; one CPA, sixMEs) occurred before the start of the procedure, sixoccurring in the periprocedural holding area and one inthe procedure room before the beginning of the

Table 1 . Location and Etiology of Events

Detail ME (n ¼ 43)

Location

Procedure room 27 (63)

Holding area 16 (37)

Etiology

Arrhythmia 8 (19)

Hypoxia/respiratory failure* 5 (12)

Oversedation 9 (21)

Hemorrhage 5 (12)

Air embolism 1 (2)

Sepsis 5 (12)

Contrast agent reaction 3 (7)

Other 7 (16)

Values in parentheses are percentages.

CPA ¼ cardiopulmonary arrest, ME ¼ medical emergency.

*Not attributed to sedation.

intervention. The events that occurred before the startof the procedure were in two outpatients and fiveinpatients. In total, 18 events (33%; two CPAs, 16MEs) occurred in the periprocedural holding area, sixbefore and 12 after the procedure (Table 1). The initialmortality rate of these events was 2% (n ¼ 1). Theremaining patients survived the event and were alivewhen leaving the IR section. The patient who died in theIR section was a 62-year-old man who experienced PEAarrest during declotting of his hemodialysis access.The level of care following the event was increased in

70.9% of all events (n ¼ 39), 62.3% of MEs (n ¼ 27), and100% of CPAs (n ¼ 12; P ¼ .012). In all, 72.2% of theoutpatients who experienced an ME or CPA (n ¼ 18; 11MEs, two CPAs) were admitted. The remaining 27.8%of outpatients (n ¼ 5; all MEs) were evaluated andobserved in the emergency department and dischargedhome the same day. Following the event, 60% ofpatients (n ¼ 33; 23 MEs, 10 CPAs) were transferredto an ICU whereas 27.3% (n ¼ 15; all MEs) were able toreturn to a non-ICU floor. One patient died duringresuscitation (2% initial mortality rate) and one wasimmediately transferred to the operating room to beplaced on an extracorporeal membrane oxygenationcircuit as part of post-CPA resuscitation. The medianlengths of stay for MEs, CPAs, and all events were 9days (interquartile range, 2–21 d), 11 days (interquartilerange, 8.5–22.5 d), and 10 days (interquartile range, 3–22d), respectively (P ¼ .28).Forty-two patients (76.4%) were alive at discharge, 37

(67.3%) were alive at 1 month from the time of the event,26 (47.3%) were alive at 3 months, and 23 (41.8%) werealive at 1 year. Thirty-six patients (83.7%) who had anME were alive at discharge, 32 (74.4%) were alive at 1month from the time of the event, 22 (51.2%) were aliveat 3 months, and 20 (46.5%) were alive at 1 year. Sixpatients (50%) who had a CPA were alive at discharge,five (41.7%) were alive at 1 month from the time of the

CPA (n ¼ 12) All (N ¼ 55)

10 (83) 37 (67)

2 (17) 18 (33)

6 (50) 14 (26)

4 (33) 9 (16)

0 9 (16)

2 (17) 7 (13)

0 1 (2)

0 5 (9)

0 3 (6)

0 7 (13)

Nadolski et al ’ JVIR1782 ’ Medical Emergencies and Cardiopulmonary Arrests in IR

event, four (33.3%) were alive at 3 months, and three(25%) were alive at 1 year. Kaplan–Meier survival curvesof patients who experienced an ME or CPA are shown inthe Figure. The differences in postevent mortality at 1 yeardid not reach statistical significance (P ¼ .074).

Procedure Type and ComorbiditiesThe frequency of ME/CPA by procedure type is reported inTable 2. The frequency of an ME or CPA was highest (38%;n ¼ 21) during hemodialysis access cases (dialysis catheterplacement/exchange, n ¼ 11; fistulography, n ¼ 5; anddeclotting, n ¼ 5). Of the 12 CPAs, nine (75%) occurred inhemodialysis access cases (dialysis catheter placement/exchange, n ¼ 6; fistulography, n ¼ 1; and declotting, n¼ 2). The RR of an ME or CPA occurring duringhemodialysis access case compared with all other caseswas 5.2 (95% confidence interval ¼ 3.02–8.95; P o .0001).Comorbidities were present in 92.5% of all patients

(n ¼ 49; 37 MEs, 12 CPAs). The frequencies of various

Figure. Kaplan–Meier survival curves following CPAs and MEs. (Ava

Table 2 . Frequency of Event by Procedure Type

Procedure Total Encounte

Arterial procedures 3,066

Venous procedures except dialysis catheters 15,168

Dialysis catheter placement/exchange 2,088

Fistulography 1,258

Declotting 393

All other interventions 16,561

All nondialysis access interventions 34,795

All dialysis access interventions 4,132

All interventions 38,927

Values in parentheses are percentages.

CPA ¼ cardiopulmonary arrest, ME ¼ medical emergency.

comorbidities are summarized in Table 3. The meanimmediate preprocedural ASA status classifications forME, CPA, and all events were 2.88, 3.33, and 2.98,respectively (P ¼ .12). Patients who were 70 years of ageor older (n ¼ 16) at the time of the ME/CPA were lesslikely to be alive 12 months after the event (19%; n ¼ 3)than those who were younger than 70 years of age at thetime of the event (51%; n ¼ 19; P ¼ .04).

Procedure Outcome and ComplicationsSeven events (13%; six MEs, one CPA) occurred beforethe start of the procedure, six in the periproceduralholding area and one in the procedure room before thebeginning of the intervention. The events that occurredbefore the start of the procedure were in two outpatientsand five inpatients. Overall, 35 procedures (64%) were com-pleted successfully. Procedural technical complications werereported in 42% of cases (n¼ 23), and all were attributed asthe primary cause of the ME or CPA. These procedural

ilable in color online at www.jvir.org.)

rs ME (n ¼ 43) CPA (n ¼ 12) All (N ¼ 55)

4 (0.13) 1 (0.03) 5 (0.16)

11 (0.07) 2 (0.01) 13 (0.09)

5 (0.24) 6 (0.29) 11 (0.53)

4 (0.32) 1 (0.08) 5 (0.4)

3 (0.76) 2 (0.51) 5 (1.3)

16 (0.1) 0 16 (0.1)

31 (0.09) 3 (0.009) 34 (0.1)

12 (0.29) 9 (0.22) 21 (0.51)

43 (0.11) 12 (0.03) 55 (0.14)

Table 3 . Comorbidities

Comorbidity ME (n ¼ 41) CPA (n ¼ 12) All (N ¼ 53)

Liver disease 7 (17) 1 (8) 8 (15)

Renal disease* 16 (39) 11 (92) 27 (51)

ESRD or acute renal failure† 13 (32) 10 (83) 23 (43)

CAD (or equivalent) 10 (24) 7 (58) 17 (32)

Congestive heart failure 6 (15) 2 (17) 8 (15)

Malignancy 16 (39) 1 (8) 17 (32)

Chronic pulmonary disease 5 (12) 1 (8) 6 (11)

Diabetes mellitus 12 (29) 3 (25) 15 (28)

History of VTE 3 (7) 3 (25) 6 (11)

Percentages do not add up to 100% because some patients had multiple comorbidities. Values in parentheses are percentages.

CAD ¼ coronary artery disease, CPA ¼ cardiopulmonary arrest, ESRD ¼ end-stage renal disease, ME ¼ medical emergency, VTE ¼venous thromboembolic disease.

*Defined by a glomerular filtration rate o 90 mL/min.†Hemodialysis-dependent.

Table 4 . Procedural Complications Directly Attributed as Primary Cause of Adverse Event

Complication ME (n ¼ 41) CPA (n ¼ 12) All (N ¼ 53)

Total 22 1 23

Oversedation 9 (21) 0 9 (16)

Arrhythmia related to intravascular device 4 (9) 0 4 (7)

Vascular injury 3 (7) 1 (8) 4 (7)

Contrast agent reaction 3 (7) 0 3 (6)

Air embolism 1 (2) 0 1 (2)

Retroperitoneal hematoma 1 (2) 0 1 (2)

Fall from IR table 1 (2) 0 1 (2)

Values in parentheses are percentages.

CPA ¼ cardiopulmonary arrest, IR ¼ interventional radiology, ME ¼ medical emergency.

Volume 24 ’ Number 12 ’ December ’ 2013 1783

complications, rather than sequelae of patient comorbid-ities, were attributed as the primary cause of 22 MEs (51%)and one CPA (8%, P ¼ .018). Table 4 summarizes thefrequencies of the respective complications.The most frequently reported complication was

oversedation (16%; n ¼ 9; all MEs). Moderate seda-tion with intravenous fentanyl and midazolam wasused for 66% of cases (n ¼ 36; 29 MEs, seven CPAs).Fentanyl only was used in an additional three cases.The median dose of fentanyl in cases of oversedation(n ¼ 9) was 200 mg (interquartile range, 150–200 mg),compared with 150 mg (interquartile range, 50–250 mg)in all other cases (n ¼ 27; P ¼ .58). The median doseof midazolam in cases of oversedation (n ¼ 9) was 3mg (interquartile range, 2–5 mg), compared with 2.5mg (interquartile range, 0.5–4 mg) in all other cases (n¼ 27; P ¼ .58).

DISCUSSION

IR is distinct from other surgical, radiologic, andmedical subspecialties. It requires expertise in image-guided minimally invasive procedures applied to multi-ple diseases and organs (10). Use of the least invasive

techniques currently available minimizes risk to thepatient and improve health outcomes. As such, it is ofparamount importance for interventional radiologists tobe aware of the patient population most at risk for MEor CPA during IR procedures and to identify practicesto minimize the risks of significant adverse events.In the present study, an ME or CPA occurred in

0.14% of approximately 39,000 IR cases overall. Onlyone patient (2% of events) could not be successfullyresuscitated and died in the IR unit. Despite theinfrequent nature of these events the and high initialsuccess rate of treating ME/CPA in IR, the posteventmortality rate was relatively high. Following the event,almost one fourth of patients died before discharge, andfewer than half of patients were alive at 3 months.Higher mortality rates were observed following CPAcompared with ME, but the difference was not statisti-cally significant (1 y survival rates, 25% for CPA vs46.5% for ME; P ¼ .074). Although the mortalityobserved after CPA and ME was relatively high,previous studies have found post-CPA mortality ratesto be even higher (11,12). Among 517 patients in the ICU,Abella et al (11) found 59.6% were able to be resuscitated,30.4% survived to discharge from the ICU, 26.9% survivedto discharge from the hospital, and 24.3% survived to 1

Nadolski et al ’ JVIR1784 ’ Medical Emergencies and Cardiopulmonary Arrests in IR

year. Older studies (12) have reported even lower rates ofsurvival (11.3%) at 1 year after in-hospital cardiac arrest.The observed differences in survival between the presentstudy and earlier investigations of post-CPA survival canbe at least partially attributed to differences in patientpopulations and subgroups.These differences highlight the importance of being aware

not just which patients are likely to have an ME/CPA butalso which patients are more likely to recover. For example,the improved survival of patients who experience a CPA inthe present study compared with previous studies may besecondary to an overall low incidence of PEA and asystole(n ¼ 9; 16.4%) as the cardiac rhythm identified at the timeof the event. These cardiac arrhythmias have been shown tobe associated with poorer post-event survival (13,14). In 143patients with witnessed events, Brindley et al (14) found thatthe odds ratio of survival was significantly lower after PEAor asystole (adjusted odds ratio, 21.0; 95% confidenceinterval, 6.2–71.7) compared with primary respiratoryarrest. Previous studies (15) also support a relativelyimproved postevent survival in patients undergoing IRprocedures. In a study of 441 patients surviving CPA todischarge, Skrifvars et al (15) found an odds ratio of 11.3 ofsurviving to 1 year in patients whose event occurred in IRcompared with patients whose event occurred on a generalhospital ward.Regardless of the relatively high initial survival rates of

these events in IR, they negatively impact the care of thesepatients. Only 64% of procedures (n ¼ 35) could becompleted successfully. Additionally, 71% of patients (n ¼39) had an increase in level of care, and 72% of outpatients(n ¼ 13) had to be admitted after the event. Although thecost of the hospital stay and added procedures were notrecorded in the present study, these adverse events havefinancial implications for hospitals and patients.In the present study, patients undergoing procedures

related to hemodialysis access had a significantly higherRR of experiencing an ME or CPA compared with thoseundergoing any other intervention (RR ¼ 5.2; 95%confidence interval ¼ 3.02–8.95; P o .0001). Previousinvestigations (16) have found a higher RR of CPA inpatients undergoing fistulography. In the present study,the increased frequency of events was not limited tofistulography but also present in patients who underwentcatheter insertion or exchange, suggesting that the risk ismore a function of renal failure in this population asopposed to the procedure. Patients with end-stage renaldisease are known to be at increased risk of cardiacmortality compared with their age-, sex-, and race-controlled cohorts without renal disease. The risk ofsudden cardiac death in the dialysis population may beas high as 24% (17). Given the RR of ME or CPA in thedialysis population, increased vigilance during their carefor early warning signs of instability is warranted. Inaddition, patients with end-stage renal disease receivecare at freestanding dialysis access centers, which maynot be equipped with rapid response or code teams to

provide additional support during an ME or CPA. Thisscenario underscores the need for interventional radiol-ogists and their staff to be familiar with signs ofimpending ME or CPA and their treatment algorithms.The present study highlights the need for continued

vigilance for signs of early instability in the holding area. Inthe present series, the holding area was not infrequently thesite of ME/CPA (33% of all events). Continued frequentmonitoring of vital signs and mental status has been shownto be helpful in predicting clinical deterioration (1). Inparticular, a patient’s pulmonary function, which histo-rically has been demonstrated to be a less accuratelymonitored vital sign, appears to be an important earlymarker of deterioration (1). Hypoxia or respiratory failurewas the etiology in 16% of all events. Of the 18 events thatoccurred in the holding area, 33% (n ¼ 6) were for hypoxiaor respiratory failure. The frequency of respiratory eventsoverall and particularly in the periprocedural holding areaprovides an opportunity for increased surveillance toreduce the occurrence of these events.In all, MEs were significantly more likely to be directly

attributable to procedural complication than CPAs. Themost frequent complication attributed as the cause of anevent was oversedation (n ¼ 9). All patients who experi-enced oversedation survived the initial event. Interestingly,the dosages of fentanyl and midazolam administeredbetween patients who had oversedation and those whodid not was not statistically different (P ¼ .58). Addition-ally, none of the patients who experienced oversedationreported a previous complication or sensitivity to anestheticagents. In the present series, only three patients who had anME/CPA indicated a previous complication from anesthe-sia or medications administered during a procedure (historyof contrast agent reaction, n ¼ 2; previous CPA fromanesthesia, n ¼ 1). Reported incidences of sedation-relatedcomplications from administration of fentanyl and mid-azolam in IR range between 2% and 4.7%, and a smallsubset of these patients will experience clinical deteriorationleading to ME or CPA (18,19). The relative frequency ofoversedation as a complication in IR and its potential tocause a reversible and survivable ME highlights the need forall IR personnel to have intimate knowledge of recognizingand treating sedation-related complications (18,20).The present study is not without limitations. The study

is retrospective, with only 1 year of follow-up and withouta population of individuals with similar comorbidities withwhom to compare survival. Additionally, there were threeevents for which complete records of the ME/CPA werenot available. Etiology of serious adverse events can bemultifactorial, and it can be difficult to accurately deter-mine the main causative factor. In the present study, thecause of the event and its relationship to the procedurewere defined by the treating interventionalist at the time ofthe procedure. Therefore, differences in interpretation ofsimilar events between treating physicians cannot beaccounted for in this retrospective study. Additionally,whether the identified MEs/CPAs were the first such event

Volume 24 ’ Number 12 ’ December ’ 2013 1785

for each individual patient in his or her lifetime could notbe determined with absolute certainty. Previous MEs/CPAs or comorbidities may confer higher risk for futureevents or mortality following such an event, which cannotbe accounted for in the present study. In addition,algorithms for activating emergency responses can differbetween institutions, and incidences and outcomes mayvary between institutions.In conclusion, MEs/CPAs occur infrequently in patients

undergoing IR procedures. However, it remains importantfor all IR personnel to be aware of the incidence of theseevents, of the relevant emergency alarm mechanisms, andof treatment algorithms in their institutions. Although theincidence of MEs/CPAs in patients undergoing IR proce-dures is low, 1-year mortality following these events isrelatively high. The RR of MEs/CPAs is significantlyhigher in hemodialysis access interventions compared withother procedures, demonstrating the need for increasedvigilance when caring for these patients, even during lessinvasive procedures such as catheter placement/exchange.Oversedation was the most frequently encountered com-plication related to an ME/CPA, highlighting the need forinterventionalists to be well trained in recognizing signs ofoversedation and managing its clinical consequences.

REFERENCES

1. Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl JMed 2011; 365:139–146.

2. Sandroni C, Nolan J, Cavallaro F, Antonelli M. In-hospital cardiac arrest:incidence, prognosis and possible measures to improve survival. IntensCare Med 2007; 33:237–245.

3. Larkin GL, Copes WS, Nathanson BH, Kaye W. Pre-resuscitation factorsassociated with mortality in 49,130 cases of in-hospital cardiac arrest: areport from the National Registry for Cardiopulmonary Resuscitation.Resuscitation 2010; 81:302–311.

4. McCurd M, Wood S. Rapid response systems: identification and manage-ment of the “prearrest state.” Emerg Med Clin North Am 2012; 30:141–152.

& SIR, 2013

J Vasc Interv Radiol 2013; 24:1785–1787

http://dx.doi.org/10.1016/j.jvir.2013.09.008

The author has not identified a conflict of interest.

From the Department of Radiology, Duke University Medical Center, Box 3808,Room 1502, Durham, NC 27710. Final revision received September 17, 2013;accepted September 23, 2013. Address correspondence to T.P.S.; E-mail:tony.smith@duke.edu

5. Jones D, Duke G, Green J, et al. Medical emergency team syndromesand an approach to their management. Crit Care 2006; 10:R30.

6. Buxton AE, Calkins H, Callans DJ, et al. ACC/AHA/HRS 2006 key dataelements and definitions for electrophysiological studies and procedures:a report of the American College of Cardiology/American Heart Associ-ation Task Force on Clinical Data Standards (ACC/AHA/HRS WritingCommittee to Develop Data Standards on Electrophysiology). Circulation2006; 114:2534–2570.

7. Schein RM, Hazday N, Pena M, Ruben BH, Sprung CL. Clinical ante-cedents to in-hospital cardiopulmonary arrest. Chest 1990; 98:1388–1392.

8. Buist M, Bernard S, Nguyen TV, Moore G, Anderson J. Associationbetween clinically abnormal observations and subsequent in-hospitalmortality: a prospective study. Resuscitation 2004; 62:137–141.

9. Kause J, Smith G, Prytherch D, et al. A comparison of antecedents tocardiac arrests, deaths and emergency intensive care admissions inAustralia and New Zealand, and the United Kingdom—the ACADEMIAstudy. Resuscitation 2004; 62:275–282.

10. Kaufman JA, Reekers JA, Burnes JP, et al. Global statement defininginterventional radiology. J Vasc Interv Radiol 2010; 21:1147–1149.

11. Abella BS. Not all cardiac arrests are the same. CMAJ 2011; 183:1572–1573.

12. Cooper S, Janghorbani M, Cooper G. A decade of in-hospital resuscita-tion: outcomes and prediction of survival? Resuscitation 2006; 68:231–237.

13. Herlitz J, Bång A, Gunnarsson J, et al. Factors associated with survivalto hospital discharge among patients hospitalised alive after out ofhospital cardiac arrest: change in outcome over 20 years in thecommunity of Göteborg, Sweden. Heart 2003; 89:25–30.

14. Brindley PG, Markland DM, Mayers I, Kutsogiannis DJ. Predictors ofsurvival following in-hospital adult cardiopulmonary resuscitation. CMAJ2002; 167:343–348.

15. Skrifvars MB, Castren M, Nurmi J, Thoren AB, Aune S, Herlitz J. Dopatient characteristics or factors at resuscitation influence long-termoutcome in patients surviving to be discharged following in-hospitalcardiac arrest? J Intern Med 2007; 262:488–495.

16. Rueb GR, Brady WJ, Gilliland CA, et al. Characterizing cardiopulmonaryarrest during interventional radiology procedures. J Vasc Interv Radiol2013; 24:1774–1778.

17. Wang AYM, Lam CWK, Chan IHS, Wang M, Lui SF, Sanderson JE.Sudden cardiac death in end-stage renal disease patients: a 5-yearprospective analysis. Hypertension 2010; 56:210–216.

18. Patatas K, Koukkoulli A. The use of sedation in the radiology depart-ment. Clin Radiol 2009; 64:655–663.

19. Arepally A, Oechsle D, Kirkwood S, Savader SJ. Safety of conscioussedation in interventional radiology. Cardiovasc Intervent Radiol 2001; 24:185–190.

20. Johnson S. Sedation and analgesia in the performance of interventionalprocedures. Semin Intervent Radiol 2010; 27:368–373.

INVITED COMMENTARY

Beware the Simple Case

Tony P. Smith, MD

ABBREVIATIONS

CPA = cardiopulmonary arrest, ME = medical emergency, RRT = rapid response team

This issue of the Journal of Vascular and InterventionalRadiology (JVIR) contains two studies on cardiopulmo-nary arrests (CPAs) in interventional radiology. Bothreport patients undergoing procedures at two largeacademic centers, the Hospital of the University ofPennsylvania (1) and the University of Virginia Schoolof Medicine (2).