Traumatic Deaths in the Emergency Room a Retrospective Analysis

8/3/2019 Traumatic Deaths in the Emergency Room a Retrospective Analysis ...

http://slidepdf.com/reader/full/traumatic-deaths-in-the-emergency-room-a-retrospective-analysis- 1/8

European Journal of Trauma and Emergency Surgery Focus on Diagnostic and Prognosis of Severely Traumatized Patients

Traumatic Deaths in the Emergency Room: ARetrospective Analysis of 115 Consecutive CasesTim Söderlund

1, Ilkka Tulikoura

1, Mika Niemelä

2, Lauri Handolin

1

AbstractObjective: The aim of the present study was to char-acterise traumatic deaths occurring in the emergencyroom (ER) and to assess retrospectively the quality ofgiven emergency care by evaluating whether any of thedeaths could be identified as potentially preventable.Methods: All consecutive deaths of trauma patientsbetween 1 January 1998 and 31 December 2006 in the ER

of the Töölö Hospital Trauma Centre were retrospec-tively reviewed. The inclusion criterion was death of atrauma patient occurring in the ER. Both the pre- and in-hospital medical charts and the autopsy reports of thepatients were reviewed.Results: A total of 115 patients fulfilled the inclusioncriteria, and the autopsy reports were obtained for all ofthese cases (100%). The patients were mainly males(n = 84; 73%), and the median age of the patients was51 years (range 1-93 years). The average injury severityscore (ISS) was 34.6. Blunt trauma was the most com-mon type of injury in the study population. A total of 115injuries in 50 patients were missed in both the clinicaland radiological surveys in the ER, i.e., a missed injurywas identified in 43% of the cases. Of these patients,15.7% had a clinically significant missed injury (AIS ‡ 4).

Based on our review of all available material, we con-sider that 11 deaths (9.6%) were potentially preventable.Conclusions: Missed injuries did not play a major role inthe preventable deaths. Seven potentially preventabledeaths were considered to be failures in the surgicaldecision-making process, resulting in futile non-opera-tive treatment or a delay in surgical bleeding control.

Key Words

Preventability of death Æ Trauma benchmarking ÆTraumatic death Æ Trauma patient outcome

1 Department of Orthopedics and Traumatology, Töölö Hospital,Helsinki University Central Hospital, Helsinki, Finland,

2 Department of Neurosurgery, Töölö Hospital, Helsinki UniversityCentral Hospital, Helsinki, Finland.

Received: September 13, 2008; revision accepted: February 10, 2009;Published Online: May 12, 2009

Eur J Trauma Emerg Surg 2009 Æ No. 5 URBAN & VOGEL

Eur J Trauma Emerg Surg 2009;35:455-62

DOI 10.1007/s00068-009-8179-0

IntroductionTrauma is one of the leading causes of death, accounting

approximately for every tenth death worldwide [1, 2]. It

is also the leading cause of death in the population under

40 years of age in Western societies [3]. In Finland, therewere 4,125 trauma-related deaths in 2003, and trauma

was the third most common cause of death on a whole-

population basis [4]. Approximately 50% of the trauma

deaths occur in the early phase - within 24 h of the injury

[5, 6]. In Canada, 82% of the trauma patients dying in the

emergency room (ER) of a trauma centre had a clinically

missed intra-abdominal injury that was observed in au-

topsy [6]. The incidence of autopsies carried out after

traumatic death varies from 6 to 97% [6-8], which may

lead to an underestimation of injuries and cause a bias in

the estimation of the probability of survival [6, 9].

The Toolo Hospital Trauma Centre (Helsinki Uni-versity Central Hospital) provides acute trauma care for

patients from Helsinki and its surroundings, resulting in

a catchment area of about 1.5 million people (approx.

25% of the Finnish population). Helsinki University

Central Hospital has three individual units that provide

emergency trauma care, with Toolo Hospital providing

acute care for all major traumas, with the exception of

patients < 16 years without a potential brain injury.

Thus, the number of adolescent trauma patients at Toolo

Hospital is low and consists only of those with a sus-

pected brain injury. The patients with penetrating torso

injuries are mostly admitted to another hospital, which

results in a low number of these patients as well at TooloHospital. There are approximately 20,000 admissions to

Toolo Hospital annually, including approximately 500

patients with an injury severity score (ISS) > 15.

455



Söderlund T, et al. Traumatic Deaths in the Emergency Room

An ambulance doctor service is available in the

city of Helsinki, and a helicopter-doctor service is

available in the surrounding cities. The coverage of

pre-hospital emergency care doctors is considered to

be good in the Helsinki University Central Hospitalcatchment area, and only seldom is a major trauma

patient taken to hospital without an accompanying

pre-hospital doctor.

The aim of the study reported here was to char-

acterise the traumatic deaths occurring in the ER of Toolo Hospital and to assess retrospectively the quality

of given emergency care by evaluating whether any of

the deaths could be identified as potentially prevent-

able in the light of the medical reports and post-mor-

tem autopsy reports. Finnish law requires an autopsy to

be performed on all victims dying in an accident, dying

in obscure circumstances and dying or suspected to

have died due to violence (both self-inflected and non-

self-inflected). Here, we present a unique data set of

115 consecutive deaths with an autopsy rate of 100%.

Materials and MethodsAll consecutive deaths of trauma patients between 1

January 1998 and 31 December 2006 in the ER of the

Toolo Hospital Trauma Centre were retrospectively

reviewed. An inclusion criterion was the death of a

trauma patient occurring in the ER. Non-traumatic

deaths due to a medical condition (such as non-trau-

matic intracranial bleeding), patients declared dead onarrival (DOA, defined as no active movement, no

palpable pulse, no breathing and no electrical cardiac

activity) and deaths of patients > 60 years of age with a

low-energy proximal femur fracture were excluded

from the study.

We reviewed the pre- and in-hospital medical

charts and the autopsy reports of all patients. The

collected data consisted of the mechanism of the in-

jury, recorded vital signs at the scene and in the ER,

clinically and radiologically diagnosed injuries, injuries

diagnosed in autopsy, results of the laboratory tests,

given care and timing in transportation and given care.The given care consisted of diagnostic and therapeutic

manoeuvres, including fluid resuscitation and blood

transfusions. Vital signs were defined as systolic blood

pressure (SBP), heart rate (HR), respiratory rate

(RR), Glasgow Coma Scale (GCS) [10] and peripheral

oxygen saturation.

The injuries were classified using the Abbreviated

Injury Scale (AIS) [11] for obtaining ISS [12] and new

injury severity score (NISS) [13]. The revised trauma

score (RTS) [14] was assessed, and the trauma-injury

456

severity score (TRISS) methodology [15] was used to

calculate the probability of survival, Ps, at the time of

admission to the ER. According to the AIS philosophy,

the AIS scores were downgraded in cases without

further specification of injuries. The AIS score wasdetermined by a trained and certified author (LH). The

hospital admission data were used for the calculating

the RTS and Ps for non-intubated patients. For intu-

bated patients, the pre-hospital data prior to intubation

were used to replace the missing values of GCS and

RR in order to minimize the bias caused by missing

values among the most severely injured patients [16] -

i.e. those intubated at the scene.

The pre- and in-hospital medical charts and the

autopsy reports for each patient were formally peer-

reviewed by all authors. Based on the available data,

each individual death was assessed as potentially pre-

ventable or non-preventable by two authors experi-

enced in general trauma surgery and one experienced

in neurotrauma and neurosurgery. All of the assess-

ments were carried out by the authors working in

consensus. Deaths were rated potentially preventable

if a treatable life-threatening condition could be ret-

rospectively identified based on the clinical examina-

tions, radiological studies or autopsy report, but was

not identified and/or treated. Also, if immediate or

earlier procedures to control the bleeding could

potentially have changed the outcome, the death was

classified as potentially preventable. A potentially

preventable death was also considered in cases where atreatable injury, such as tensionpneumothorax or car-

diac tamponade, was missed and therefore left un-

treated, thereby contributing clearly to the death.

All deaths were classified by the cause of the death

as brain injury (CNS), exsanguination or other reason.

A death due to CNS was defined as a brain injury and/

or a spinal cord injury incompatible with life regardless

of other injuries. Exsanguination was determined as

the cause of death if major bleeding was present,

regardless of possible attempts to control it, and when

CNS was not applicable. Causes of death other than

CNS or exsanguination were defined as another criticalinjury or condition incompatible with life, such as com-

promised airway, burn/inhalation injury, air embolism

or blunt cardiac injury.

ResultsA total of 115 patients fulfilled the inclusion criteria,

and the autopsy reports were obtained for all of these

patients (100%). The patients were mainly male

(n = 84, 73%), and the median age of the patients was





Table 1. Demographics of the patients.

Mechanisms of the injury/clinical

characteristics of the patients (n = 115)

Dominant injury, n (%)Blunt 94 (81.7)

Penetrating 17 (14.8)

Burn 4 (3.5)

Injury type, n (%)

Blunt assault 1 (0.9)

Burn 4 (3.5)

Crushing 1 (0.9)

Explosion 1 (0.9)

GSW 16 (13.9)

Major fall 21 (18.2)

Minor fall 20 (17.4)

Traffic 51 (44.3)

Clinical assessment scores/parameters (average ± SD)

ISS 35 ± 17

NISS 42 ± 15

RTS 3.75 ± 1.79

RR 12 ± 7

SBP 97 ± 49

GCS 4.6 ± 3.4

Hb (g/l) 98 ± 32

BE (mmol/l) -8.47 ± 6.5

TT (%) 60 ± 28

Minor fall: < 4 m; major fall: ‡ 4 m; GSW: gunshot wound; ISS: injury

severity score; NISS: new injury severity score; RTS: revised trauma score;

RR: respiratory rate; SBP: systolic blood pressure; GCS: Glasgow Coma

Scale; Hb, hemoglobin; BE: base excess; TT: thromboplastin time

51 years (range 1-93 years). Blunt trauma was the

most common type of injury. The mechanisms of

injuries are shown in Table 1. Eleven patients (9.6%)

were referred to from other hospitals, and 104 patients

were brought from the scene. Suicide was the reason

for injury in 32 cases. Most of the penetrating injuries

(13/17) were self-inflicted gunshot wounds in the head.

Two patients had a combined cause of injury: one

sustaining a major fall combined with hanging and the

other with a combination of a minor fall and drowning.

Ethanol intoxication (blood ethanol level > 1.00&)was determined post-mortem in 30 patients (26%), and

amphetamine use was determined in three patients

(2.6%). One patient sustaining a major fall was sus-

pected to have suffered from an acute serotonin syn-

drome due to an anti-depressive intoxication.

Pre-hospitalThe majority (n = 107, 93%) of the patients were

transferred by ambulance, and eight patients (7%) were

transferred by helicopter. Eight patients were trapped at


the scene and required extraction. The median time from

the accident to the arrival of the pre-hospital unit to the

scene was 16 min (IQ25-75 11-22 min), and the median

scene time was 27 min (IQ25-75 20-38 min). Ninety pa-

tients (78%) were intubated before arrival at the ER.Eleven blunt trauma patients (9.6%) received cardio-

pulmonary resuscitation (CPR) at the scene, four of

them dying of exsanguination at hospital. Pre-hospital

pleural decompression was performed on eight patients

(7%), but none of these patients had a cardiac arrest

necessitating CPR. Surgical airway was established on

one patient at the scene. The median volume of intra-

venous fluids administered before arrival at the ER was

1,500 ml (IQ25-75 1,000-2,250 ml).

Emergency RoomEmergency interventions carried out in the ER in-

cluded intubation (12 patients), chest tube insertion (28

patients), CPR (27 patients), thoraco-laparotomy (one

patient), escarotomies (two patients) and external

pelvic stabilization (two patients). Intra-cranial pres-

sure monitoring was started in the ER for three pa-

tients. The patients were treated in the ER for a

median of 98 min (IQ25-75 41-162 min) before death.

Nine patients (7.8%) died within 15 min after arrival at

the ER, and the majority of the patients (n = 103,

89.6%) died within 4 h after arrival. The median times

from ER admission to FAST (focused assessment of

sonography for trauma) was 10 min (IQ25-75 6-

15 min), to supine chest X-ray, 15 min (IQ25-75 10-20 min), and to computed tomography (CT) scan,

30 min (IQ25-75 20-40 min). The CT scans were per-

formed on 75 patients, and injuries causing the death

could be diagnosed from the CT scans in all cases.

The median volume of total fluids the patients re-

ceived in the ER was 3,000 ml (IQ25-75 1,000-

7,700 ml). Blood products were given to 53 patients in

the ER. For these patients the average number of the

units of red blood cells (RBC), fresh frozen plasma

(FFP) and platelets were 8.6, 1.1, and 0.4 units,

respectively. Patients in the exsanguination group re-

ceived 11.8, 1.1 and 0.5 units of RBC, FFP and plate-lets, respectively.

Trauma ScoringThe trauma-scoring characteristics, vital signs and re-

sults of the laboratory tests are presented in Table 1. In

the CNS group (n = 71), 41 patients were observed to

have injuries in other body parts as well, with the most

common being the thorax (n = 31) and the bony pelvis

(n = 26). The major bleeding sites at the patients dying

of exsanguination (n = 30) were the abdomen (n = 6),

457




Table 2. Cases of preventable deaths (n = 11).

Patients whose death Psb Reasons of preventabilityc

were considered to be

preventablea

77m 0.61 Injuries were underestimated; haemorrhagic shock was not identified; trauma resuscitation

was inadequate

37f 0.13 Intra-abdominal bleeding was diagnosed, but laparotomy was not performed after pelvic

angioembolisation in haemodynamically unstable patient

88m 0.05 No interventions made to control bleeding from diagnosed pelvic fracture

65f 0.85 Pre-hospital oesophageal intubation was not identified in ER

25m 0.36 Bleeding from pelvic fracture and ruptured spleen were not controlled, and patient received

only crystalloids and RBCs (18 units) during 3 h of treatment in the ER. Patient developed a

non-reversible coagulopathy

52f 0.97 Patient refused blood transfusions due to religious reasons and died due to bleeding from

several closed fractures

17f 0.41 Proximal control of bleeding was not done in traumatic pelvic amputation

26m 0.02 Blunt aortic injury was identified, but during the waiting of transfer to department of

cardiothoracic surgery the contained mediastinal bleeding turned non-contained bleeding to

pleural cavity

62f 0.64 Pelvic fracture was missed from CT scan and trauma resuscitation was inadequate

24m 0.22 Liver and spleen ruptures with intra-abdominal bleeding were diagnosed and treated

conservatively despite haemodynamic instability. Patient developed non-reversible coagu-

lopathy

74m 0.67 Patient had profuse bleeding from bilateral open femoral fractures and bleeding from open

humeral fractures. Bleeding was ineffectively controlled

RBC: Red blood cell; ER: Emergency Room; CT: computed tomography;aNumber followed by lowercase letter indicates age (years) and sex (m: male; f:

female) of the patient;b

Ps: probability of survival, calculated using TRISS;

records of the patients

the thorax (n = 6), the bony pelvis (n = 3) and acombination of several body regions (n = 15). The

average age of the patients in the CNS group was

51 years, in the exsanguination group, 43 years, and in the

other-reason group, 60 years. The average ISS of the

patients <50 years and >50 years was 39.7 and 29.5,

respectively.

The ISS ranged from 4 to 75, and the patients dying of

exsanguination had a higher average ISS (45) than those

of the CNS and other-reason groups (32 and 29,

respectively). The patients in the CNS group had a

higher SBP (105 vs. 82 mmHg). The average NISS

score was 42.4, and the NISS scores of 65 patients

(56.5%) were higher than the ISS score.

Preventable DeathsEleven deaths (9.6%) were considered to be potentially

preventable. These cases are presented in Table 2 with

the preventable reason that was considered to contrib-

ute to the death. One death of these 11 preventable

deaths was considered to be due to a missed injury in

the ER (missed pelvic fracture with retroperitoneal

bleeding, which could be seen in the CT scan). The P s

458

cPreventability of the death was evaluated after peer-review of all medical

was compared between the preventable and non-pre-ventable deaths (Table 3). In the non-preventable

group, 21% of the deaths were probably preventable

(Ps 0.51-0.75) or frankly preventable (Ps 0.76-1)

— compared to 45% in the preventable group. The

average Ps of the preventable death group was 0.45

(IQ25-75 0.05-0.67) compared to 0.31 (IQ25-75 0.09-

0.45) in the non-preventable group. The preventable

death group had a lower ISS (34 vs. 42) and a higher

RTS (5.40 vs. 3.58) than the non-preventable group.

The patients in the preventable death group received

Table 3. Comparison of the probability of survival between pre-

ventable and non-preventable deaths.

Ps Preventable Non-preventable

(n = 11) (n = 104)

0-0.25 4 (36) 52 (50)

0.26-0.5 2 (18) 30 (29)

0.51-0.75 3 (27) 14 (13)

0.76-1.00 2 (18) 8 (8)

Values are presented as the number of cases, with the percentage in

parenthesis





Table 4. Clinically significant (AIS

patients.

> 3) missed injuries (n = 18) in 18 fractures (3/18,

11%).

17%) and abdominal injuries (2/18,

Thorax

Rupture of descending aortaRupture of descending aorta

Rupture of descending aorta

Air embolism of right ventricle

Lung contusion

Heart contusion

Heart contusion

Heart contusion

Unilateral (tension) pneumothorax

Head and neck

Skull fracture

Spinal cord contusion

Skull fractureCIV fracture with spinal cord contusion

AIS Abdomen AIS

6 Liver rupture 56 Liver rupture 4

5

5

4

4

4

4

4

AIS Pelvis AIS

5 Pelvic fracture 5

4 Pelvic fracture 4

4 Pelvic fracture 44

The patients in the preventable death group

(n = 11) had 18 missed injuries, which included only

two injuries with AIS ‡ 4. There were 97 missedinjuries in the non-preventable death group (n = 104),

including 16 injuries with AIS ‡ 4.

DiscussionThe rate of potentially preventable deaths in the ER

was determined to be 9.6% in this study. Reported

rates of potentially preventable trauma deaths vary

from 1% in level 1 trauma centres up to 33% in smaller

volume units [17-21]. Most of our patients (89.6%)

died within 4 h after arrival to the ER. The majority of

clinically significant missed injuries are found in the

early-phase deaths [6], and preventable deaths at this

phase are mainly due to a failure to stop bleeding or

AIS: Abbreviated Injury Scale. Values of each injury are shown

an average of 12.4, 1.5 and 0.5 units of RBC, FFP and

platelets; in comparison, the patients in the non-pre-

ventable group received 7.2, 0.9 and 0.4 units, respec-

tively. The patients in the preventable group received

considerably higher volumes of intravenous fluids than

those in the non-preventable group (average 11.5 vs.

4.2 l).

The deaths were categorized into three groups

according to the time of arrival at the ER: during ‘‘nor -mal’’ office hours (0800-1600 hours); in the evening hours

(1600-2400 hours); during the night (0000-0800 hours).

The total number of deaths and the number of potentially

preventable deaths were 44/2 (4.5% preventable), 48/4

(8.3% preventable) and 23/5 (21.7% preventable) for

office hours, evening and night-time, respectively. In

eight of the 11 potentially preventable deaths, the sur-

geon-in-charge was the in-house resident initially without an

attending senior trauma surgeon.

Missed Injuries

The post-mortem examinations revealed a total of 115injuries in 50 patients that had been missed both in the

clinical and radiological surveys in the ER; this indi-

cates that a missed injury occurred in 43% of the cases.

The severity of 97 of these missed injuries ranged from

1 to 3 on the AIS; the severity of the remaining 18

injuries (18 patients) ranged from 4 to 6 on the AIS(Table 4). Thus, 15.7% of the patients had a clinically

significant missed injury (AIS ‡ 4). These AIS ‡ 4

grade injuries were mainly in the chest (9/18, 50%),

followed by head and spine injuries (4/18, 22%), pelvic


prevent hypoxia or to a delay in surgical treatment [18,

20, 22], as was also seen in this study.

In revealed by our analysis, one clearly preventable

death occurred due to ethical reasons as the patient

refused blood transfusions on religious grounds, and it

may be argued whether this death was acceptable or

not. One burn victim died of prolonged hypoxia due to

unrecognized placement of the endotracheal tube in

the oesophagus. Oesophageal placement of the endo-

tracheal tube was also identified in the autopsy of an-

other patient who had suffered a severe head injury.The oesophageal placement of the endotracheal tube

was not considered to contribute to the death because

the initial traumatic brain injury itself was incompati-

ble with life. In both cases, the intubation was done at

the scene by a pre-hospital unit, but the misplacement

was not recognized in the primary or secondary survey

in the ER. In the literature, the incidence of oesoph-

ageal insertion of the endotracheal tube by pre-hospi-

tal units has been reported to range from less than 0.5

up to 6% [23], but it is more than reasonable to argue

that such a condition should be found in the primary

survey (e.g. clinical examination, chest X-ray, andexpiratory CO2) upon the patient’s arrival at the ER.

Seven potentially preventable deaths were consid-

ered to be failures in surgical decision-making, result-

ing in futile non-operative treatment or a delay in the

surgical control of bleeding that subsequently led to

the development of uncontrollable coagulopathy. In

one of the seven cases there were, in addition to severe

pelvic bleeding, also missed injuries (liver rupture, AIS

3, and bleeding from mesenterial artery, AIS 2). The

development of coagulopathy was not sufficiently

459




prevented and/or treated in several cases, as the pa-tients received large volumes of crystalloids and RBC,

but relatively low amounts of FFP and platelets. In two

cases (77-year-old male and 62-year-old female), the

injuries were underestimated and partially missed, anda haemorrhagic shock was not recognized. The initial

resuscitation was suboptimal, leading to the develop-

ment of a non-reversible haemorrhagic shock and the

patients not responding to any further resuscitation.

The early recognition of a haemorrhagic shock is acommon pitfall in the initial resuscitation [24]; it can be

especially demanding to recognize this state in elderly

patients having arterial hypertension and using beta-

blockers, as the blood pressure and heart rate can be

misleading.

A higher fraction of patients in the potentially

preventable death group than in non-preventable

group had Ps > 0.50 (45 vs. 21%, respectively), and the

average Ps was higher in the potentially preventable

group (0.45 vs. 0.31). A P s > 0.50 has been used to

identify patients with unexpected death and thus wor-

thy of peer-review [25]. Five patients in the potentially

preventable death group had Ps > 0.50. If we had car-

ried out a peer-review only of deaths with P s > 0.50,

our rate of potentially preventable deaths would have

been 4.3%. For the Ps calculations, we used the TRISS

method with coefficients derived from the Major

Trauma Outcome Study (MTOS [26]). However, these

coefficients may not be appropriate for our study

population for several reasons. In MTOS. the patientsintubated before admission to ER were excluded (78%

of our patients were intubated at the scene), and most

of the excluded patients in MTOS were blunt trauma

patients (81.7% of our patients had blunt trauma) [26].

The mortality of the excluded patients in MTOS was

16.8% compared to 8.0% of the patients included in the

analyses [26]. Consequently, we suggest that in our study

population the decision for peer-review should not be

based on the TRISS method Ps > 0.50 as a cut-off.

At Toolo Hospital a senior trauma surgeon is on-

call at home during the night-time on most days, and a

resident surgeon is in charge of trauma resuscitations.The lack of an in-house senior trauma surgeon during

the night-time may partially explain the potentiallypreventable deaths in the light of delayed surgical

decision-making, as four of the five potentially pre-ventable deaths that occurred at night-time were con-

sidered to be due to a lack of experience. Twopreventable deaths which occurred during the office

hours were also considered to result from delayed

surgical decision-making, even though a senior trauma

surgeon was involved in the trauma resuscitation. Also,

460

one preventable death during the night-time was

treated initially by a senior trauma surgeon. A senior

trauma surgeon was present in the initial resuscitation

of the preventable deaths (n = 11) in three cases

(27%). In the non-preventable deaths (n = 104), theinitial treatment was performed by a resident surgeon

alone in 84 cases (81%). Thus, the senior trauma sur-

geon was initially present more often in the prevent-

able deaths than in the non-preventable ones. The

effect of the presence of a 24-h in-house trauma sur-

geon in terms of reducing the mortality of critically

injured trauma patients is a controversial question, as

the 24-h presence of such a specialist has been reported

to both decrease mortality [27] and have no significant

effect [28-30]. However, the volume of patients han-

dled by the trauma centre has been shown to have a

clear impact on the survival of the patients being

treated [29, 31, 32].

The Toolo Hospital trauma protocol was estab-

lished in 2002 to promote decision-making tools and

education for surgeons leading trauma resuscitations.

The trauma protocol is also aimed at providing a

standard quality of care around the clock. Systematic

trauma team simulation training was initiated in 2003

to encourage trauma teams to implement new proto-

cols and to improve overall team performance. The

massive transfusion protocol was established at the

beginning of 2005 and subsequently incorporated into

the trauma protocol. However, the ratios of RBC, FFP

and platelets before (9.1:1.2:0.5) and after (7:0.7:0.3)the implementation of the massive transfusion protocol

were not improved in patients dying in the ER. The

availability of CT scanning was good throughout the

study period, and FAST, performed by radiologists,

has also become a common routine in the trauma pa-

tient survey since 2003. At the beginning of our study

period, the technique for pelvic stabilization in the ER

was anterior external fixation, which was performed on

haemodynamically unstable patients with an unstable

fracture of the pelvic ring. Since 2005, we have used the

commercially available stabilization belt (T-POD;

Pyng Medical, Richmond, BC, Canada) for the initialstabilization of pelvic fractures in the ER. We did not

find any potentially preventable deaths during the last

2 years (2005-2006) in our cohort study. However, due

to overall low annual numbers of traumatic deaths in

the ER (23 deaths during 2005-2006), we are unable to

draw any definitive conclusions on the possible effects

of the evolved trauma care on the potentially pre-

ventable deaths. A longer time survey is needed toidentify the possible effects of the changes in our

trauma care protocol.





Missed injuries did not play a major role in our

study. They contributed to three deaths, but only in

one case was a missed injury the main reason for pre-

ventability. In other studies, unexpected injuries (i.e.

missed injuries without CT) of the thorax and abdomenwere seen in 42-74% of the patients [33, 34]. The

number of missed injuries in the CNS group (21 pa-

tients with 41 missed injuries) was mainly due to the

fact that in many cases the treatment of the patient was

stopped after the head CT scan revealed brain injuries

incompatible with life. The post-mortem autopsies re-

vealed 56 missed injuries in 24 patients in the exsan-

guination group. This result is partly explained by the

high proportion of blunt polytraumas in this group (28/

29 patients) combined with the severely compromised

clinical condition of the patients, which prevented CT

scanning (body CT scan performed only on 10/29 pa-

tients), and partly by the fact that the patients died

before possible surgical exposures could have revealed

these injuries. Computed tomography scanning is not

feasible in critically injured patients, and the decisions

for surgical interventions in such cases have to be

based on the clinical examination. In treating these

patients, the experience of the trauma surgeon can be

assumed to have a major impact.

Computed tomography scanning is considered to

be the golden standard of blunt trauma patient survey,

and it is especially important in the early diagnostics of

sedated and intubated blunt trauma patients [33, 34].

In our study cohort, the majority of the patients wereintubated (89%), but 51 patients were haemodynami-

cally unstable (SBP £ 90). Thus, the majority of the

patients should have been scanned with CT, but the

haemodynamic instability prevented it. Computed

tomography scanning is also associated with a risk of

delaying some critical emergency procedures if the

patient in extremis is taken to relatively time-con-

suming scanning instead of making a prompt decision

to operate on the patient. However, there is a possi-

bility of incorporating CT scanning in a routine trauma

patient survey in a time-effective way and to gain the

benefits of full-body CT scanning without losing timeto start possible key-emergency operations [35]. The

ER and CT of Toolo hospital have recently been ren-

ovated, which has greatly facilitated the performance

of ‘‘fast track’’ scanning as part of the trauma patient

survey. The possible effects of this change in our

trauma protocol on timing and overall performance

will be seen after a few years.

We conclude that our rate of preventable deaths is

relatively high compared to those reported earlier in

Level 1 trauma centres. On the other hand, it can al -


ways be argued that the reported figures may not be

comparable as such due to the broad variety of defi-

nitions used for the parameters studied. All deaths

were peer-reviewed regardless of the estimated prob-

ability of survival. More accurate definitions of parameters as well as high-quality trauma registries are

needed for reliable benchmarking. The role of the

experienced senior trauma surgeon is of vital impor-

tance throughout the acute trauma care, and efforts

have to be made for obtaining a sufficient number of

well-educated and experienced surgeons to cover this

task 24/7. A continuous control of the quality of the

given care has to be part of the normal routine in a

major trauma centre; equally, performance improve-

ment should be an ongoing process in these centres.

AcknowledgmentsThe authors wish to thank Jouni Backman MD/PhD (Helsinki Uni-

versity Central Hospital, Department of Clinical Pharmacology), for

providing insight into the serotonin syndrome.

Conflict of interest statementThe authors declare that there is no actual or potential conflict of

interest in relation to this article.

References1. Murray CL, Lopez AD. Mortality by cause for eight regions of the

world: global burden of disease study. Lancet 1997;349:1269-76.2. Peden M, McGee K, Sharma G. The injury chart book: a graphical

overview of the global burden of injuries. World Health Orga-nization, Geneva, 2002.

3. Nijs S, Broos P. Trauma care systems in Belgium. Inj Int J Care Inj

2003;34:652-7.

4. Statistics Finland. Deaths, cause of death. Available at: http://

www.tilastokeskus.fi. 2003.

5. Søreide K, Krüger AJ, Vårdal AL, Ellingsen CL, Søreide E, Lossius

HM. Epidemiology and contemporary patterns of trauma

deaths: changing place, similar pace, older face. World J Surg2007; doi: 10.1007/s00268-007-9226-9.

6. Hodgson NF, Stewart TC, Girotti MJ. Autopsies and death cer-

tification in deaths due to blunt trauma: what are we missing?

Can J Surg 2000;43:130-6.

7. Jin PHFK, Klaver JF, Maes A, Ponsen KJ, Das C, Goslings JC.

Autopsies following death due to traumatic injuries in Neth-erlands: an evaluation of current practice. Inj Int J Care Inj

2008;39:83-9.

8. Ong AW, Cohn SM, Cohn KA, Jarmillo DH, Parbhu R, McKenney

MG, Barquist ES, Bell MD. Unexpected findings in trauma pa-tients dying in the intensive care unit: results of 153 consecutive

autopsies. J Am Coll Surg 2002;194:401-6.

9. Harviel JD, Landsman I, Greenberg A, Copes WS, Flanagan ME,

Champion HR. The effect of autopsy on injury severity andsurvival probability calculations. J Trauma 1989;29:766-73.

10. Teasdale G, Jennett B. Assessment of coma and impaired con-

sciousness: a practical scale. Lancet 1974;304:81-4.

461

http://www.tilastokeskus.fi./

http://www.tilastokeskus.fi./




11. Association for the Advancement of Automotive Medicine

(AAAM). The abbreviated injury scale 2005. AAAM, Barrington,

2005. Available at: http://www.aaam.org.

12. Baker SP, O’Neill B, Haddon W, Long WB. The injury severity

score: a method for describing patients with multiple injuries

and evaluating emergency care. J Trauma 1974;14:187-96.13. Osler T, Baker SP, Long W. A modification of the injury severity

score that both improves accuracy and simplifies scoring. JTrauma 1997;43:922-5.

14. Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA,

Flanagan ME. A revision of the trauma score. J Trauma

1989;29:623-9.

15. Boyd CR, Tolson MA, Copes WS. Evaluating trauma care: the

TRISS method. J Trauma 1987;27:370-8.

16. Skaga NO, Eken T, Steen PA. Assessing quality of care in a

trauma referral center: benchmarking performance by TRISS-based statistics or by analysis of stratified ISS data? J Trauma

2006;60:538-47.

17. Acosta JA, Yang JC, Winchell RJ, Simons RK, Fortlage DA,

Hollingsworth-Fridlund P, Hoyt DB. Lethal injuries and time to

deaths in a level I trauma center. J Am Coll Surg 1988;186:528-33.

18. Anderson ID, Woodford M, de Dombal FT, Irving M. Retrospec-

tive study of 1,000 deaths from injury in England and Wales. BrMed J 1988;296:1305-8.

19. Gilroy D. Deaths from blunt trauma, after arrival at hospital:

plus ça change, plus c’est la même chose. Inj Int J Care Inj

2005;36:47-50.

20. Gruen RL, Jurkovich GJ, McIntyre LK, Foy HM, Maier RV. Patterns

of errors contributing to trauma mortality: lessons learned from

2,594 deaths. Ann Surg 2006;3:371-80.

21. Shackford SR, MacKersie RC, Davis JW, Wolf PL, Hoyt DB. Epi-

demiology and pathology of traumatic deaths occurring at a

level I trauma center in a regionalized system: the importance of

secondary brain injury. J Trauma 1989;29:1392-7.

22. Tien HC, Spencer F, Tremblay LN, Rizoli SB, Brenneman FD.Preventable deaths from hemorrhage at a level I Canadiantrauma center. J Trauma 2007;62:142-6.

23. Jemmett ME, Kendal KM, Fourre MW, Burton JH. Unrecognized

misplacement of endotracheal tube in a mixed urban to rural

emergency medical service setting. Acad Emerg Med

2003;10:961-5.

24. MacKersie RC, Dicker RA. Pitfalls in the evaluation and man-

agement of the trauma patient. Curr Probl Surg 2007;44:

778-833.

25. Champion HR, Sacco WJ, Hunt TK. Trauma severity scoring to

predict mortality. World J Surg 1983;7:4-11.

26. Champion HR, Copes WS, Sacco WJ, Lawnick MM, Keast SL, Bain

LW, Flanagan ME, Frey CF. The Major trauma outcome study:

establishing national norms for trauma care. J Trauma

1990;30:1356-65.

462

27. Rogers FB, Simons R, Hoyt DB, Shackford SR, Holbrook T, Fort-

lage DA. In-house board-certified surgeons improve outcome

for severely injured patients: a comparison of two university

centers. J Trauma 1993;34:871-7.

28. Luchette F, Kelly B, Davis K, Johanningman J, Heink N, James L,

Ottaway M. Impact of the in-house trauma surgeon on initialpatient care, outcome, and cost. J Trauma 1997;42:490-7.

29. Pasquale MD, Peitzman AB, Bednarski J, Wasser TE. Outcome

analysis of Pennsylvania trauma centers: factors predictive of

nonsurvival in seriously injured patients. J Trauma 2001;50:465-74.

30. Thompson CT, Bickell WH, Siemens RA, Sacra JC. Community

hospital level II trauma center outcome. J Trauma 1992;32:336-43.

31. Smith RF, Frateschi L, Sloan EP, Campbell L, Krieg R, Edwards LC,

Barret JA. The impact of volume on outcome in seriously injured

trauma patients: two years’ experience of the Chicago trauma

system. J Trauma 1990;30:1066-76.

32. Konvolinka CW, Copes WS, Sacco WJ. Institution and per-sur-

geon volume versus survival outcome in Pennsylvania’s trauma

centers. Am J Surg 1995;170:333-40.

33. Deunk J, Dekker HM, Brink M, van Vugt R, Edwards MJ, van VugtAB. The value of indicated computed tomography scan of thechest and abdomen in addition to the conventional radiologic

work-up for blunt trauma. J Trauma 2007;63:757-63.

34. Traub M, Stevenson M, McEvoy S, Briggs G, Lo SK, Leibman S,

Joseph T. The use of chest computed tomography versus chest

x-ray in patients with major blunt trauma. Inj Int J Care Inj

2007;38:43-7.

35. Weninger P, Mauritz W, Fridrich P, Spitaler R, Figl M, Kern B,

Hertz H. Emergency room management of patients with blunt

major trauma: evaluation of the multislice computed tomog-

raphy protocol exemplified by an urban trauma center.

J Trauma 2007;62:584-91.

Address for CorrespondenceTim Söderlund

Department of Orthopedics and TraumatologyTöölö Hospital

Helsinki University Central HospitalTopeliuksenkatu 5

P.O. Box 26600029 HelsinkiFinland

e-mail: [email protected]


http://www.aaam.org./

http://www.aaam.org./

Documents

Traumatic Deaths in the Emergency Room a Retrospective Analysis