A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury

G Model JINJ 7268 No. of Pages 10

Injury, Int. J. Care Injured xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Injury journal homepage: www.elsevier.com/locate/injury

A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury Swathikan Chidambaram* , En Lin Goh, Mansoor A. Khan Department of Surgery and Trauma, Faculty of Medicine, Imperial College London, St Mary’s Hospital, London, United Kingdom

A R T I C L E I N F O

Keywords: Whole-body computed tomography Trauma Selective imaging WBCT

A B S T R A C T

Background: Traumatic injury is the third leading cause of death overall. To optimize the outcomes in these patients, hospitals employ whole-body computed tomography (WBCT) imaging due to the high diagnostic yield and potential to identify missed injuries. However, this delays time-critical interventions. Currently, there is an absence of any high-level evidence to support or refute either view. We present a meta-analysis of the available literature to elucidate the efficacy of WBCT in improving the outcomes of trauma, specifically the mortality rate. Methods: A systematic review of studies comparing WBCT and selective CT imaging in secondary survey was conducted, using MEDLINE, EMBASE, the Cochrane Review and Scopus databases. The articles were evaluated for intervention using WBCT to reduce mortality rate, followed by subgroup analysis for other secondary measures, using Review Manager 5.3 software. Results: Eleven studies of 32,207 patients were included. There were lower overall (OR = 0.79; 95% CI 0.74,0.83, p < 0.05) and 24 h mortality rates (OR = 0.72, 95% CI 0.66,0.79, p < 0.05) in the WBCT cohort. Additionally, patients in the WBCT arm spent less time in the emergency room (MD = 14.81; 95% CI 17.02, 12.60, p < 0.00001) and needing ventilation (MD = 2.01; 95% CI 2.41, 1.62, p < 0.05) despite a higher baseline injury severity score. Conclusion: The analysis shows that WBCT is associated with better outcomes, including a lower overall and 24 h mortality rate, however the included studies are mostly observational and show considerable heterogeneity. Further work is required to make definitive clinical recommendations for a tailored algorithm in managing trauma patients. Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved.

Advance(s) in knowledge 1. In blunt trauma, the use of whole-body computed tomography (WBCT) was associated with lower odds of overall mortality and 24 h mortality. 2. WBCT-use was associated with less time spent in the emergency room, hospital and intensive care unit and fewer ventilation days.

2. WBCT allows for more rapid diagnosis, which translates into earlier treatment and reduced sequelae.

Summary statement In patients presenting with blunt trauma, the use of WBCT leads to a lower mortality and morbidity in the immediate period. Introduction

Implication(s) for patient care 1. The use of WBCT following blunt trauma is associated with improved immediate and long-term outcomes.

* Corresponding author at: Faculty of Medicine, Imperial College London, South Kensington Campus, London SW7 2AZ, United Kingdom. E-mail address: [email protected] (S. Chidambaram).

Traumatic injury is the leading cause of mortality, morbidity and permanent disability in patients under the age of 45 years, and the third leading cause of death overall [1,2]. In the United States, costs associated with trauma care are estimated to exceed $500,000 per patient and $514 billion in total. In the emergency setting, injured patients require immediate life-saving treatment and hospitals employ a sophisticated sequence of protocols to

http://dx.doi.org/10.1016/j.injury.2017.06.003 0020-1383/Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

2

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

recognize and manage fatal injuries before other systemic investigations are performed. The role of computed tomography (CT) in trauma management has rapidly expanded due to the capability to image the whole body or selected regions with high specificity and sensitivity [3–7]. Compared to conventional selective imaging (SI), whole body CT (WBCT) scanning can identify less apparent injuries that would otherwise remain undiagnosed, which translates into earlier treatment, better outcomes, shorter time spent in the hospital and fewer subsequent scans done in the long-term [8–13]. These benefits have led to more trauma centers employing routine WBCT in their assessment protocols [12–15]. The use of WBCT in the immediate time period has potential to delay time-critical interventions [16–19], increase exposure to unnecessary radiation and costs. In particular, it is estimated that 2% of cancers may be attributed to radiation from CT scans [20], with a 16% increase in cancer risk with each successive CT scan performed [21]. However, this is related to patient age and is more significant in the first few years of life. Since the typical trauma population comprises of young adult males, it is likely that the risk of radiation quoted may not be representative in these individuals. Thus, this underlines the importance optimizing the risk/benefit ratio to stratify patients on the likely outcomes based on their pre-existing characteristics and expert opinion to determine the most appropriate mode of imaging. Currently, there is no high-level evidence to support or refute either view. Numerous primary studies and systematic reviews have investigated the usefulness of WBCT [22–25]. Given the low level of evidence in the primary studies, these reviews could not determine a definitive conclusion of WBCT utility. Since 2014, multiple studies and trials have been published, including the recent REACT-2 trial, which was a multi-center, prospective, randomized controlled trial (RCT) [26]. We present a systematic review and meta-analysis of all the currently available literature, with the aim to elucidate the efficacy of WBCT in improving the outcomes following trauma. Methods A systematic review of studies comparing WBCT and SI during the initial assessment of patients who have undergone trauma was conducted. This systematic review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Search strategy and selection of studies Scientific publications investigating the use of WBCT in trauma were identified using MEDLINE, EMBASE, the Cochrane Review and Scopus databases. The search was performed on September 15, 2016. Search terms included (“whole body computed tomography” OR “wbct” OR “full body computed tomography” OR “fbct” OR “total body computed tomography” OR “tbct” OR “panscan” OR “whole body imaging” OR “selective imaging”). This was then combined using the AND modifier with a second search (“trauma” OR “injury” OR “wounds and injuries” OR “emergency”). The search results were limited to include only original studies. The reference lists of eligible articles were also hand-searched for additional relevant publications.

between WBCT and SI protocols, for any trauma (blunt, penetrating, or blast). Comparative cohort studies, non-randomized prospective studies and RCTs were included. However, case series, case reports, narrative reviews, editorials, and conference abstracts; studies without comparison groups; studies with pediatric patients or less than five patients; and publications in a non-English language were excluded. Data extraction and quality assessment Both qualitative and quantitative data were extracted by two independent reviewers (S.C. and E.L.G.). Data extracted included the year of publication, study design, sample size, country of study, type of patients, patient characteristics, outcome measures, and conclusions. All primary studies were appraised for rigorousness using the Newcastle-Ottawa Scale and Wright’s Levels of Evidence, both of which, assess potential sources of bias and variation in the selection, comparability and generalizability of non-randomized studies. Data synthesis Data for the overall mortality rate were extracted to calculate the odds ratio (OR) and 95% confidence interval (CI). Secondary outcome measures analyzed included the 24 h mortality rate; time spent in the emergency room (ER); time spent in the intensive care unit (ICU); time spent in the hospital; number of ventilation days; Injury Severity Score (ISS); and proportion of patients with Abbreviated Injury Scale (AIS) score 3 for the head, thorax, abdomen and extremities. ORs were calculated for the dichotomous variables, while mean outcomes were used for continuous variables. Statistical methods Review Manager 5.3 (Cochrane Collaboration, Oxford, United Kingdom) was used for statistical analysis of the data. Two types of modelling were used to assess the heterogeneity of the data: fixedeffects and random-effects. The random-effects model was chosen for the overall mortality rate to reflect the many confounding effects that cannot be accounted for. Meanwhile, the fixed-effects model was used for the remaining analyses, since the number of studies reporting these parameters was small. In all cases, statistical heterogeneity was assessed by using I2 statistic and was categorized as low, moderate and high for an I2 statistic of above 25%, 50% and 75%, respectively. Results above 60% were considered as substantial heterogeneity. Results The database search yielded 620 studies, and six studies were identified through searching other sources. Of these, 195 duplicates were removed. Titles and abstracts of the remaining 461 studies were assessed for eligibility, and a further 425 studies were removed. A further 25 studies were excluded after full-text review due to incompatible outcome measures; specifically, the lack of a mortality rate (Fig. 1). A total of 11 studies, representing 32,207 patients met the criteria for inclusion in the meta-analysis. Study characteristics

Selection of studies: inclusion and exclusion criteria Titles and abstracts were scanned, and irrelevant studies were excluded. Full text articles of the remaining studies were then retrieved and evaluated for inclusion. The inclusion criteria required articles to have compared the overall mortality rate

The characteristics of the primary studies are summarized in Table 1a and b [26–36]. There were five prospective cohort studies, five retrospective studies and one RCT. All studies compared the effect of WBCT or SI on the mortality rate of the patients as the primary outcome. Five studies also reported on the 24-h mortality

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

3

Fig. 1. PRISMA diagram of the meta-analysis.

of the patients [26–28,33,34]. Three studies compared WBCT with a comparison group that consisted of at least a chest x-ray, pelvic xray, FAST, and organ-focused CT imaging [29,31,32]. Five studies evaluated the baseline ISS for both cohorts [27–31]. Four studies had reported on the proportion of patients with an AIS  3 for the head, thorax, abdomen and extremities [27–30]. Lastly, studies had also described the number of days in the ER, ICU [27–29,31], in the hospital [27,28,31] or requiring ventilation [12,15,33]. The definition of WBCT was consistent across all studies, but some differences were noted in the scanning protocols identified (Table 1b). Risk of bias The non-randomized studies included were evaluated for sources of bias using the Wright’s Level of Evidence and the Newcastle-Ottawa Scale. All studies achieved an excellent score of 7/8 on the Newcastle-Ottawa Scale. They attained maximum points for the “selection” category, but lost points for

comparability of the study cohorts and for lack of adequate follow-up, both of which potentiate possible bias. Using a funnel plot (Fig. 2), the risk of publication bias was examined. Asymmetry was noted in the plot and caused by the studies by Wada et al. and Hutter et al. [29,33]. Both studies favor WBCT, but have a higher standard error of the effect estimate. The asymmetry suggests the possible existence of small, unpublished studies with negative results as well as the inherent clinical heterogeneity. Patient characteristics The studies predominantly involved blunt trauma patients. Two studies [35,36] had included trauma of other types, but this was a non-significant proportion of less than 3% of the cohort that was eventually excluded from final analysis. Patient factors that may have confounding effects on the mortality rate were not statistically significant at baseline, including the age, the gender proportion and mechanism of injury. However, significant differences were present in other physiological cardiorespiratory

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

4

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

Table 1a Characteristics of studies. Study

Year

Huber2013 Wagner

Definition of WBCT

Scanning Protocol

WBCT is defined as unenhanced CT of the head followed by contrastenhanced CT of the chest, abdomen, and pelvis, including the complete spine. It can be conducted as single-pass or segmented WBCT. By contrast, no CT or only dedicated CT of one or combined body regions was defined as non-WBCT.

The participating hospitals were free to choose their own diagnostic algorithms depending on the type of CT scanners and local emergency department protocols. Neither information about the location of the CT scanner (in or near the trauma room or in the department of radiology) nor information about the specific local CT protocols is recorded in the trauma registry. Due to the high number of 216 participating hospitals in this study, the recorded data can be interpreted as representative for the currently practised standard of trauma care.

Huber2009 Wagner 2013 WBCT was defined as the imaging of the head, thorax, abdomen and The trauma CT scanner is a GE Lightspeed VCT 64-slice helical scanner (GE Hsiao extremities, as noted in the scanning protocol. Healthcare, Waukesha, WI, USA), which is located in the Radiology Department adjacent to the ED. The standard whole-body CT protocol includes a noncontrast scan of the brain and neck, and a contrast-enhanced scan of the chest, abdomen and pelvis. Additional phases might include a delayed phase through relevant body regions, usually for suspected active bleeding sites, and/or precontrast and early contrast-enhanced phases of the chest or abdomen, usually for suspected large vessel injury. 2011 WBCT was defined as the imaging of the head, thorax, abdomen and The following procedures and scanning parameters applied to the trauma panHutter extremities, as noted in the scanning protocol. scan: tube voltage 120 kV, native scan of the head and the cervical spine with the upper extremities positioned adjacent to the body at axial slice thickness of 0.62 mm, followed by multi-planar reconstruction with a slice thickness of 2 mm. After repositioning the arms over the head, 100 ml of iopromide (Ultravist1 300, Bayer-Schering, Germany, containing 300 mg/mL of iodine) were injected at a flow rate of 4 ml/ s, and the diagnostic procedure was completed by contrast-enhanced imaging in the arterial phase of the thorax, abdomen, pelvis, vertebral column and the upper part of the legs. Additionally, a venous phase of the intra-abdominal organs was performed. The slice thickness was 0.62 mm for acquisition of raw data, 2.5 mm for axial, 3.0 mm for coronal, and 5.0 mm for sagittal reconstruction of the trunk. For the vertebral column, a slice thick- ness of 2.0 mm was employed in either plane to guarantee high resolution in this sensible area. Scanning methods and indications for WBCT depended on the protocol of each 2013 WBCT was defined as a CT scan including the head, neck, chest, Kimura abdomen, and pelvis during initial trauma management at emergency center, about which no information was available emergency centers. Non-WBCT was defined as CT scanning without including one or more of the previously mentioned body regions. 2016 It was defined as CT of the chest, abdomen, and pelvis, including the The protocol for the intervention (total-body CT) group consisted of a two-step Sierink complete spine, as per the imaging protocol. acquisition (from vertex to pubic symphysis) without gantry angulations, starting with a non-enhanced CT of the head and neck with arms alongside the trunk. The second scan covered the chest, abdomen, and pelvis. The preferred technique for the second scan was split-bolus intravenous contrast imaging immediately after raising the arms alongside the head.26 CT scanners at the participating sites were all 64-slice multidetector row CT scanners. The standard radiological trauma work-up was done according to ATLS guidelines. The multidetector CT scanner was located in the trauma room or in a room adjacent to the emergency department. 2013 It was defined as CT of the chest, abdomen, and pelvis, including the TBCT scanning consisted of a two-step whole-body acquisition (from vertex to Sierink complete spine, as per the imaging protocol. pubic symphysis), starting with head and neck non-enhanced CT (NECT) with arms alongside the body. For the second complementary scan a split-bolus intravenous contrast protocol was used directly after repositioning the arms alongside the head. This scan covered thorax, abdomen, and pelvis. Wada 2013 It was defined as CT of the chest, abdomen, and pelvis, including the The locations of the CT scanners in both institutions are on the same floor as the complete spine, as per the imaging protocol. trauma rooms. CT was performed at the discretion of the attending physician based on individual patient condition and not according to a pre-defined protocol. Weninger 2007 It was defined as CT of the chest, abdomen, and pelvis, including the The first phase was the examination of head and upper cervical spine (to C4/C5) complete spine, as per the imaging protocol. without any contrast medium; cervical collars were left in place. If possible, the patient’s arms were then positioned above the head, and the torso was examined after intravenous administration of nonionic contrast medium (Scanlux 370 mg/mL, Sanochemia, Vienna). Arterial contrasted phase for examination of thoracic, abdominal, and pelvic organs (C4 to lesser trochanters) started with a delay of 25 s (thoracic scan with “bolus tracking” of the descending aorta at 120 HU). After a 35-s delay the venous perfusion phase was started to examine abdomen and pelvis from the diaphragm to the lesser trochanters (abdominal scan during portal venous phase). 2010 It was defined as CT of the chest, abdomen, and pelvis, including the The whole-body MSCT is performed according to a predefined standardized Wurmb complete spine, as per the imaging protocol. protocol using a multislice CT with a 16-row-scanner, as shown in Table 2 of the paper. Yeguiyan 2012 WBCT is defined as unenhanced CT of the head followed by contrast- No details given on the scanning protocol. enhanced CT of the chest, abdomen, and pelvis, including the complete spine. It can be conducted as single-pass or segmented WBCT. By contrast, no CT or only dedicated CT of one or combined body regions was defined as non-WBCT.

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

5

Table 1b Radiological criteria and scanning protocols used in the studies. Study

Year

Country

Design

No. of Wright’s Patients Level

Newcastle- Inclusion Criteria Ottawa Scale

Exclusion Criteria

Germany

Multi-center, retrospective, cohort

16,719

III

7/8

Penetrating trauma

Huber2009 Germany Wagner

Multi-center, retrospective, cohort

4621

II

7/8

Hsiao

2013

Australia

Single-center, prospective, cohort

660

II

7/8

Hutter

2011

Germany

Single-center, prospective, cohort

1144

II

7/8

Kimura

2013

Japan

Multi-center, retrospective, cohort

5208

III

7/8

Blunt trauma patient with initial SBP > 75 mmHg and GCS of 3–12

Sierink

2016

The Multi-center Netherlands randomized controlled trial

304

I

N/A

Patients aged 18 years or older with trauma with compromised vital parameters, clinical suspicion of lifethreatening injuries, or severe injury

Sierink

2013

The Single-center, Netherlands prospective, cohort

1083

II

7/8

Wada

2013

Japan

Multi-center, retrospective, cohort

152

III

7/8

Weninger 2007 Austria

Single-center, retrospective, cohort

370

III

7/8

Wurmb

2010

Germany

Multi-center, prospective, cohort

318

II

7/8

Yeguiyan

2012

France

Single-center, prospective, cohort

1950

II

7/8

Trauma patients with the presence of one of the following vital parameters Respiratory rate [29/min or \10/min Pulse[120/min Systolic blood pressure \100 mmHg Estimated exterior blood loss Abnormal pupillary reaction on site OR patients with one of the following clinically suspicious diagnoses Fractures of at least two long bones Flail chest, open chest, or multiple rib fractures Pelvic fracture Unstable vertebral fractures Spinal cord injury Patients with blunt trauma who required Patients who were transferred from emergency bleeding control (surgery or other hospitals; patients with traumatic transcatheter arterial embolization). cardiopulmonary arrest on arrival; and patients with severe brain injury that was the direct cause of death. patients with blunt trauma admitted Patients with minor injuries; deaths between February 1, 2003, and before admission; deaths in the ER December 31, 2004 with an Injury Severity Score (ISS) /17 and at least one life threatening injury of head, thorax, or abdomen with an abbreviated injury score (AIS) /4, who survived at least until admission to the ICU Group I: patients with multiple trauma Exclusion criteria: patients who required who required emergency surgery and immediate emergency laparotomy or who were diagnosed with the thoracotomy directly in the resuscitation conventional trauma protocol from 2001 area without getting a complete to 2003. diagnostic work-up were excluded from Group II: patients with suspected the analysis. multiple trauma who required emergency surgery and who were initially diagnosed with the MSCT trauma protocol from 2004 to 2006. at least 18 years old and had a severe (a) penetrating traumas and (b) deaths blunt trauma defined as trauma occurring before the implementation of requiring admission into an ICU within any advanced life-sustaining treatment

Huber2013 Wagner

Blunt trauma, injury-severity score of at least 16, and available information about whole-body CT during trauma-room treatment. Only those patients who were admitted directly from the incident scene, and not transferred from other hospitals blunt trauma patients (age 0.16 years), ISS $16, and available information about the Revised Injury Severity Classification (RISC) score, WBCT during trauma room treatment and the systolic blood pressure on hospital admission. Only those patients were included who were admitted directly from the incident scene and not transferred from other hospitals. All adult (age > 15 years) presentations that triggered trauma team activation and were CT scanned during the initial ED assessment were included All patients who had been exposed to a blunt high-velocity incident with a high pre-test probability of major trauma

Patients who died or received emergency surgery within the first 30 min after arrival at the hospital were excluded due to a possible “immortal time bias”

Trauma patients who were transferred from other facilities were excluded.

Pregnant women were excluded from undergoing a pan-scan. Super-obese patients with a body weight of > 200 kg were excluded Patients’ records without any one of the predictors of the Trauma and Injury Severity Score (TRISS) or outcomes were excluded Patients referred from another hospital; low-energy blunt trauma; stab wound in 1 body region; too unstable to undergo CT and required either CPR or immediate operation Known age \18 years Known pregnancy Referred from another hospital Any patient judged to be too unstable to undergo a CT scan and requires (cardiopulmonary) resuscitation or immediate operation

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

6

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

Table 1b (Continued) Study

Year

Country

Design

No. of Wright’s Patients Level

Newcastle- Inclusion Criteria Ottawa Scale

Exclusion Criteria

72 h after injury or, in the case of early death before ICU admission, trauma managed by a mobile ICU (MICU)

Fig. 2. Funnel Plot for the meta-analysis.

Table 2 Comparison of baseline ISS.1 Study

Mean ISS in WBCT group (SD)

Mean ISS in SI group (SD)

p-value for severity comparison

Weninger et al. [2007] Huber-Wagner et al. [2009] Hutter et al. [2011] Kimura et al. [2013] Huber-Wagner et al. [2013]

26.6 32.4 28.3 26.0 29.7

27.6 (11.5) 28.4 (12.4) 26.4 (12.2) 23.0 (29.5) 27.7 (11.9)

Not significant <0.001 <0.001 <0.0001 <0.001

(10.3) (13.6) (11.8) (21.9) (12.2)

measurements, which were inevitable given the nature of trauma care. Baseline ISS was documented by five studies for the WBCT (n = 13,378) and SI (n = 14,461) arms. Four studies, with notably larger sample sizes, reported a significantly higher ISS (p < 0.001) in the WBCT cohort (Table 2). Four studies reported on the proportion of patients with an AIS  3. To account for the different sites of trauma, the AIS scores were reported separately for the head, thorax, abdomen and extremities. The number of patients in both arms differed for each subgroup analysis (Table 3). In total, there were 48,498 and 49,186 data points for the WBCT and SI arms, respectively. Huber-Wagner et al. reported a statistically higher proportion of patients with AIS  3 for injuries to the thorax, abdomen and extremities in the WBCT arm. The proportion of patients with head injuries with AIS  3 was comparable between the WBCT and SI arms. Typically, there was a slightly higher percentage of injuries related to the thorax, abdomen and extremities in the WBCT arm, while the SI arm had more patients

1 Abbreviations used in Tables 1–3: ISS- injury severity score; AIS Abbreviated Injury Scale; SD standard deviation; WBCT whole body computed tomography; SI Selective Imaging

with injuries of the head (Table 2). The studies did not report on any organ-specific injuries for either arm. Primary outcome measure: overall mortality rate The eleven studies included measured overall mortality rate as the primary outcome measure. The total sample size was 32,207, with similar numbers in the WBCT (n = 16,106) and SI (16,101) arms. There were 2857 deaths and 3526 deaths in the WBCT and SI arms, respectively. There was evidence of heterogeneity between studies (I2 = 83%, p < 0.00001), and consequently, random-effects analysis demonstrated a significantly lower overall mortality rate in the WBCT arm (pooled OR = 0.79; 95% CI 0.74,0.83, p < 0.05) (Fig. 3). Twenty-four hour mortality rate The 24 h mortality rate was appropriate in assessing the usefulness of WBCT in the immediate timeframe. The total sample size was 20,206, with more patients in the WBCT arm (n = 11,752) than the SI arm (n = 8454). There were 997 deaths and 976 deaths in the WBCT and SI arms, respectively. There was evidence of significant heterogeneity between studies (I2 = 85%; p < 0.0001).

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

7

Table 3 Comparison of proportion of patients with baseline AIS score 3. Study

Head AIS WBCT

Huber-Wagner et al. [2009] Hutter et al. [2011] Kimura et al. [2013] Huber-Wagner et al. [2013]

Thorax AIS SI

884 (59.1%) 1882 (60.2%) 330 210 (67.1%) (54.3%) 1431 2665 (92.0%) (93.6%) 5229 4686 (56.6%) (62.6%)

pvalue

WBCT

Abdomen AIS SI

pvalue

WBCT

SI

Extremities AIS pvalue

WBCT

SI

1035 (69.3%) <0.001 407 (66.9%) 0.044 720 (90.1%)

1589 <0.001 390 (26.1%) 660 (50.8%) (21.1%) 163 (52.1%) <0.001 124 (20.4%) 61 (19.5%)

<0.001 614 (41.1%) 1073 (34.3%) <0.001 – –

843 (85.7%

<0.001 5873 (63.6%)

3659 (48.9%)

416 (36.9%) <0.001 3760 (40.7%)

0.51

0.046

138 (48.3%) <0.001 2034 (22.0%)

176 (51.8%) 0.38 1379 (18.4%)

545 (33.5%) 2453 (32.8%)

p-value <0.001 – N.S. <0.001

Fig. 3. Overall Mortality Rate.

Fig. 4. 24-hour mortality rate.

The fixed-effects model demonstrated a lower 24 h mortality rate of statistical significance in the WBCT arm (OR = 0.72, 95% CI 0.66,0.79, p < 0.05) (Fig. 4).

(I2 = 91%; p < 0.00001). The fixed-effects model showed patients in the SI arm spent a 1.97 days less in ICU (pooled mean difference = 1.97; 95% CI 1.59, 2.34, p < 0.05) (Fig. 6).

Time spent in the ER

Time spent in the hospital

Three studies compared the time spent in the ER between the WBCT (n = 2287) and SI (n = 3625) arms (Fig. 5). Patients spent 14.81 min less in the ER for the WBCT arm (pooled mean difference = 14.81; 95% CI 17.02, 12.60, p < 0.00001) despite a higher ISS.

Three studies had compared the length of stay in the hospital, with 10,912 patients in the WBCT and 10,798 patients in the SI arms. There was evidence of moderate heterogeneity between studies (I2 = 72%; p = 0.03). The fixed-effects model demonstrated a shorter stay for the SI arm by 1.03 days (pooled mean difference = 1.03; 95% CI 0.25,1.81, p = 0.03) (Fig. 7).

Time spent in the ICU Number of ventilation days Four studies had compared the length of stay in the ICU, with 11,520 patients in the WBCT and 11,111 patients in the SI arms. There was evidence of high heterogeneity between studies

Three studies had compared the number of days requiring ventilation. There was evidence of heterogeneity between studies

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

8

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

Fig. 5. Time in Emergency Room.

Fig. 6. Time in Intensive Care Unit.

Fig. 7. Time in Hospital .

Fig. 8. Ventilation Days.

(I2=; p < 0.00001). The fixed-effects model demonstrated a higher number of ventilation days for the SI cohort (pooled mean difference = 2.01; 95% CI 2.41, 1.62, p < 0.05) (Fig. 8). Discussion Using the available evidence, this paper concludes a statistically significant reduction in the overall and 24-h mortality rate with routine use of WBCT compared to SI (Figs. 3 and 4), using both random- and fixed-effects analysis. This contends previous analyses, both primary studies and reviews, and supports the role of WBCT in the acute and sub-acute setting [23–25]. Notably, a statistical difference in the 24-h mortality was established using the fixed-effects analysis, but not the random-effects modelling. This discrepancy may be attributed to other uncontrollable factors such as specialization of trauma center and expertise of medical personnel that have confounding effects on the results. Secondary analyses comparing the time spent in the ER and days requiring ventilation showed that patients in the WBCT arm fared better in both categories. However, patients in the SI arm had a shorter duration of stay in the ICU and correspondingly, the

hospital. These differences may be indicative of fewer and less severe complications experienced by the SI cohort due to the lower baseline ISS. As only a few studies reported these data, fixed-effects analysis was used to demonstrate the significant differences, and the ISS and proportion of patients with AIS  3 at baseline was assessed to reduce any heterogeneity engendered. These findings illustrate that WBCT imaging is associated with a lower morbidity of trauma in the acute phase, in addition to reduced mortality. A critical determinant of outcome from trauma management is the baseline injury severity. Patients with more severe injuries fare worse, regardless of the imaging protocol. In ten of the studies, the ISS was higher in the WBCT group. A similar effect was also seen for the proportion of patients with an AIS  3 in each group. AIS  3 is indicative of severe trauma, and associated with a mortality rate greater than 10% [37]. Moreover, AIS scores can be stratified according to injury site, which enables more specific analysis between WBCT and SI groups, given that injuries to different locations will have contrasting sequelae of complications. On the whole, the proportion of patients with AIS  3 was greater in the WBCT arm, suggesting an increased risk of mortality in this group of patients. However, the present analysis reveals a lower mortality

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

rate following WBCT, which is indicative of the benefits of this approach. Interestingly, the use of SI was preferred to WBCT in patients with head injuries, which may reflect the tailored use of the different imaging protocols. In patients with injuries of the thorax, abdomen and extremities, the use of WBCT allowed for a more thorough evaluation of the injuries and hence resulted in a lower overall and 24 h mortality. Conversely, head injuries require time-critical interventions that can only be provided after more targeted initial management using SI and the practice will vary depending on individual patient factors, including the mechanism of injury. Further information will be required to establish the risk of mortality in these patients to guide the decision-making of clinicians in stratifying these patients. Nevertheless, while WBCT may pick up undiagnosed injuries, the severity of these injuries and subsequently its influence on course of management may be minimal, so any benefits derived will be off-set by the delay in intervention. However, a missed injury is only recorded, when it is subsequently identified and correlated to prior diagnostic efforts. This renders much uncertainty about current data on ‘missed injuries’. Unfortunately, no further evaluation can be made in this aspect, since none of the included studies accounted for the ‘missed injuries’ in either the WBCT or SI arms. As noted earlier, the routine use of WBCT imaging for trauma patients may increase the risk of malignancy. Gordic et al. reported the use of WBCT in the initial assessment of trauma patients to be associated with a significantly higher radiation dose of 29.5 mSv compared to 15.9 mSv with the use of SI [38,39]. However, the introduction of iterative reconstructive techniques into CT imaging has led to a decrease in WBCT dose from 15 to 20 mSv to 5–10 mSv [27]. Moreover, WBCT reduces the trauma-related imaging time and need for further imaging. This is especially evident in trauma patients with a higher baseline severity. In their RCT, Sierink et al. observed that trauma patients with an ISS  16 received a higher radiation dose in the immediate setting but a comparable total radiation dose throughout the admission [26]. Thus, it is apparent that patients with the greatest baseline injury severity are most likely to benefit from WBCT but identifying these patients remains a significant issue. Whilst triage criteria are typically based on the mechanism of injury and clinical features, these tend to underestimate the actual severity in 30% of cases [32]. There are currently no large-scale studies that exclusively investigate if there was a significant difference between WBCT and SI protocols, and if this difference has a tangible impact on patient outcomes. Although a cancer-related mortality of 1 in 1250 has been attributed to CT scans, whether the morbidity and mortality due to imaging outweighs that of ‘missed injuries’ is, again, another note of much controversy. This is especially given that the claims are supported by qualitative expert opinions and quantitative predictions that are not always based on the most realistic of situations. The authors acknowledge that this analysis has several limitations. Firstly, only one RCT has been conducted in this field of trauma to date, which was included in the analysis. Additionally, the remaining studies display considerable heterogeneity, with numerous bias and confounding factors. Hence, the improved mortality associated with WBCT is derived from low level evidence and more robust prospective trials are required to confirm this. Further, the points of contention with regards to the dose of radiation received by patient and number of missed injuries are not reported by most primary studies. The effects of these measures are more long-term, so it is not always possible to account for these factors, without which, a definitive conclusion on the benefits and harms of WBCT imaging cannot be made. Another limitation was that the definition of trauma in the studies were based on an initial subjective assessment made by the emergency department team

9

although patients were subsequently categorized according to ISS. Additionally, there were subtle differences in the imaging protocols between the studies included, such as the location of the CT scanner, and the exact imaging parameters. Some studies such as Wada et al. stated that no preformed imaging protocol was used and was let to the discretion of the attending physician [33]. Studies have reported an incidental finding rate of approximately 45% in WBCT [36], with 36% located in the abdomen and thorax [40]. In another study, Munk and colleagues described a higher incidental finding rate of 30% than overall traumatic abnormality [41]. However, there is a paucity of studies with follow-up to appraise the clinical benefit of identifying these findings. Lastly, the extent of heterogeneity, as noted by the Cochrane Q test value, between the studies was significant but inevitable given the nature of the primary studies for the afore-mentioned reasons. In conclusion, the present review demonstrates low level evidence suggesting that WBCT is associated with a lower mortality rate. However, further work is still required in the form of RCTs and economic cost analyses to make recommendations for a more tailored algorithm in managing trauma patients. A shorter stay will translate to lesser costs per patient, which is beneficial if it does not compromise the care provided, as estimated by the National Audit Office as well as the Medical Expenditure Panel Survey [42,43]. As mentioned before, patients will be also subjected to lower dose of radiation and hence lesser risk of malignancy, as noted by the 7th BEIR report [44]. This will need further studies to correlate the results of the primary survey, WBCT results and final outcomes, including mortality rate. Conflicts of interest statement There are no financial and personal relationships with other people or organisations that could inappropriately influence (bias) the work, including employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/registrations, and grants or other funding. References [1] Holcomb JB, Tilley BC, Baraniuk S. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the proppr randomized clinical trial. JAMA 2015;313(5):471–82. [2] Rhee P, Joseph B, Pandit V, Aziz H, Vercruysse G, Kulvatunyou N, et al. Increasing trauma deaths in the United States. Ann Surg 2014;260(1):13–21. [3] Klingenbeck-Regn K, Schaller S, Flohr T, Ohnesorge B, Kopp AF, Baum U. Subsecond multi-slice computed tomography: basics and applications. Eur J Radiol 1999;31(2):110–24. [4] Leidner B, Beckman OM. Standardized whole-body computed tomography as a screening tool in blunt multitrauma patients. Emerg Radiol 2001;8(1):20–8. [5] Deunk J, Brink M, Dekker HM, Kool DR, Blickman JG, van Vugt AB, et al. Routine versus selective multidetector-row computed tomography (MDCT) in blunt trauma patients: level of agreement on the influence of additional findings on management. J Trauma 2009;67(5):1080–6. [6] Tillou A, Gupta M, Baraff LJ, Schriger DL, Hoffman JR, Hiatt JR, et al. Is the use of pan-computed tomography for blunt trauma justified? A prospective evaluation. J Trauma 2009;67(4):779–87. [7] Deunk J, Dekker HM, Brink M, van Vugt R, Edwards MJ, van Vugt AB. The value of indicated computed tomography scan of the chest and abdomen in addition to the conventional radiologic work-up for blunt trauma patients. J Trauma 2007;63(4):757–63. [8] Kanz K-G, Paul AO, Lefering R, Kay MV, Kreimeier U, Linsenmaier U, et al. Trauma management incorporating focused assessment with computed tomography in trauma (FACTT) potential effect on survival. J Trauma Manage Outcomes 2010;4(1):1–10. [9] Rieger M, Czermak B, El Attal R, Sumann G, Jaschke W, Freund M. Initial clinical experience with a 64-MDCT whole-body scanner in an emergency department: better time management and diagnostic quality? J Trauma 2009;66(3):648–57. [10] Ahvenjarvi L, Mattila L, Ojala R, Tervonen O. Value of multidetector computed tomography in assessing blunt multitrauma patients. Acta Radiol 2005;46 (2):177–83. [11] Smith CM, Woolrich-Burt L, Wellings R, Costa ML. Major trauma CT scanning: the experience of a regional trauma centre in the UK. Emerg Med J 2011;28 (5):378–82.

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003

G Model JINJ 7268 No. of Pages 10

10

S. Chidambaram et al. / Injury, Int. J. Care Injured xxx (2017) xxx–xxx

[12] Salim A, Sangthong B, Martin M, Brown C, Plurad D, Demetriades D. Whole body imaging in blunt multisystem trauma patients without obvious signs of injury: results of a prospective study. Arch Surg 2006;141(5)468–73 discussion 73-5.. [13] Sampson MA, Colquhoun KB, Hennessy NL. Computed tomography whole body imaging in multi-trauma: 7 years experience. Clin Radiol 2006;61 (4):365–9. [14] Schoeneberg C, Schilling M, Burggraf M, Fochtmann U, Lendemans S. Reduction in mortality in severely injured patients following the introduction of the Treatment of patients with severe and multiple injuries guideline of the German society of trauma surgery?a retrospective analysis of a level 1 trauma center (2010–2012). Injury 2014;45(3):635–8. [15] Gupta M, Schriger DL, Hiatt JR, Cryer HG, Tillou A, Hoffman JR, et al. Selective use of computed tomography compared with routine whole body imaging in patients with blunt trauma. Ann Emerg Med 2011;58(5)407–16 e15. [16] Ruchholtz S, Waydhas C, Schroeder T, Piepenbrink K, Kuhl H, Nast-Kolb D. The value of computed tomography in the early treatment of seriously injured patients. Chirurg 2002;73(10):1005–12. [17] Wedegartner U, Lorenzen M, Nagel HD, Weber C, Adam G. Diagnostic imaging in polytrauma: comparison of radiation exposure from whole-body MSCT and conventional radiography with organ-specific CT. Rofo 2004;176(7):1039–44. [18] Kalra MK, Maher MM, Toth TL, Hamberg LM, Blake MA, Shepard JA, et al. Strategies for CT radiation dose optimization. Radiology 2004;230(3):619–28. [19] Laack TA, Thompson KM, Kofler JM, Bellolio MF, Sawyer MD, Laack NN. Comparison of trauma mortality and estimated cancer mortality from computed tomography during initial evaluation of intermediate-risk trauma patients. J Trauma 2011;70(6):1362–5. [20] Brenner DJ, Hall EJ. Computed tomography — an increasing source of radiation exposure. New Engl J Med 2007;357(22):2277–84. [21] Mathews JD, Forsythe AV, Brady Z, Butler MW, Goergen SK, Byrnes GB, et al. Cancer risk in 680,000 people exposed to computed tomography scans in childhood or adolescence: data linkage study of 11 million Australians. BMJ 2013;346:f2360. [22] Surendran A, Mori A, Varma DK, Gruen RL. Systematic review of the benefits and harms of whole-body computed tomography in the early management of multitrauma patients: are we getting the whole picture? J Trauma Acute Care Surg 2014;76(4):1122–30. [23] Caputo ND, Stahmer C, Lim G, Shah K. Whole-body computed tomographic scanning leads to better survival as opposed to selective scanning in trauma patients: a systematic review and meta-analysis. J Trauma Acute Care Surg 2014;77(4):534–9. [24] Sierink JC, Saltzherr TP, Reitsma JB, Van Delden OM, Luitse JS, Goslings JC. Systematic review and meta-analysis of immediate total-body computed tomography compared with selective radiological imaging of injured patients. Br J Surg 2012;99:52–8 Suppl 1. [25] Healy DA, Hegarty A, Feeley I, Clarke-Moloney M, Grace PA, Walsh SR. Systematic review and meta-analysis of routine total body CT compared with selective CT in trauma patients. Emerg Med J 2014;31(2):101–8. [26] Sierink JC, Treskes K, Edwards MJR, Beuker BJA, den Hartog D, Hohmann J, et al. Immediate total-body CT scanning versus conventional imaging and selective CT scanning in patients with severe trauma (REACT 2): a randomised controlled trial. Lancet 2017. [27] Huber-Wagner S, Biberthaler P, Häberle S, Wierer M, Dobritz M, Rummeny E, et al. Whole-body CT in haemodynamically unstable severely injured patients a retrospective, multicentre study. PLoS One 2013;8(7):e68880.

[28] Huber-Wagner S, Lefering R, Qvick LM, Korner M, Kay MV, Pfeifer KJ, et al. Effect of whole-body CT during trauma resuscitation on survival: a retrospective, multicentre study. Lancet 2009;373(9673):1455–61. [29] Hutter M, Woltmann A, Hierholzer C, Gartner C, Buhren V, Stengel D. Association between a single-pass whole-body computed tomography policy and survival after blunt major trauma: a retrospective cohort study. Scand J Trauma Resusc Emerg Med 2011;19:73. [30] Kimura A, Tanaka N. Whole-body computed tomography is associated with decreased mortality in blunt trauma patients with moderate-to-severe consciousness disturbance: a multicenter, retrospective study. J Trauma Acute Care Surg 2013;75(2):202–6. [31] Weninger P, Mauritz W, Fridrich P, Spitaler R, Figl M, Kern B, et al. Emergency room management of patients with blunt major trauma: evaluation of the multislice computed tomography protocol exemplified by an urban trauma center. J Trauma 2007;62(3):584–91. [32] Wurmb TE, Quaisser C, Balling H, Kredel M, Muellenbach R, Kenn W, et al. Whole-body multislice computed tomography (MSCT) improves trauma care in patients requiring surgery after multiple trauma. Emerg Med J 2011;28 (4):300–4. [33] Wada D, Nakamori Y, Yamakawa K, Yoshikawa Y, Kiguchi T, Tasaki O, et al. Impact on survival of whole-body computed tomography before emergency bleeding control in patients with severe blunt trauma. Crit Care 2013;17(4):1– 7. [34] Yeguiayan JM, Yap A, Freysz M, Garrigue D, Jacquot C, Martin C, et al. Impact of whole-body computed tomography on mortality and surgical management of severe blunt trauma. Crit Care 2012;2012. [35] Hsiao KH, Dinh MM, McNamara KP, Bein KJ, Roncal S, Saade C, et al. Wholebody computed tomography in the initial assessment of trauma patients: is there optimal criteria for patient selection? Emerg Med Aust 2013;25(2):182– 91. [36] Sierink JC, Saltzherr TP, Beenen LF, Russchen MJ, Luitse JS, Dijkgraaf MG, et al. A case-matched series of immediate total-body CT scanning versus the standard radiological work-up in trauma patients. World J Surg 2014;38(4):795–802. [37] Thomas A, Gennarelli EW. The Abbreviated injuryscale 2005. IL: Des Plaines; 2008 Update. [38] Gordic S, Alkadhi H, Hodel S, Simmen HP, Brueesch M, Frauenfelder T, et al. Whole-body CT-based imaging algorithm for multiple trauma patients: radiation dose and time to diagnosis. Br J Radiol 2015;88(1047):20140616. [39] Asha S, Curtis KA, Grant N, Taylor C, Lo S, Smart R, et al. Comparison of radiation exposure of trauma patients from diagnostic radiology procedures before and after the introduction of a panscan protocol. Emerg Med Aust 2012;24(1):43– 51. [40] van Vugt R, Dekker HM, Deunk J, van der Vijver RJ, van Vugt AB, Kool DR, et al. Incidental findings on routine thoracoabdominal computed tomography in blunt trauma patients. J Trauma Acute Care Surg 2012;72(2):416–21. [41] Munk MD, Peitzman AB, Hostler DP, Wolfson AB. Frequency and follow-up of incidental findings on trauma computed tomography scans: experience at a level one trauma center. J Emerg Med 2010;38(3):346–50. [42] General RbtCaA. Managing high value capital equipment in the NHS in England. [cited 2016 18 September]. [43] Quality AfHRa, Medical Expenditure Panel Survey [cited 2016 18 September]. [44] Davidson ST. Radiation: any dose is too high. Environ Health Perspect 2005;113 (11):A735.

Please cite this article in press as: S. Chidambaram, et al., A meta-analysis of the efficacy of whole-body computed tomography imaging in the management of trauma and injury, Injury (2017), http://dx.doi.org/10.1016/j.injury.2017.06.003