Answers are grounded in this paper and cite their sources.
Dear Editor, Physical examination is a fundamental component of the primary survey of trauma patients, designed to promptly identify life-threatening injuries1. Missed diagnoses may delay treatment of time-critical injuries, whereas over-diagnosis may lead to unnecessary triage or treatment2. Accurate diagnosis is therefore key to ensuring appropriate and timely definitive management, which has been shown to reduce morbidity and mortality after trauma2. However, the ability of physical examination to identify life-threatening torso injuries remains unclear. While several factors can influence the accuracy of physical examination, such as the setting (pre-hospital versus in-hospital), mechanism of injury (blunt versus penetrating trauma), severity of injury, presence of distracting injuries and impaired consciousness, there is no comprehensive review of the diagnostic accuracy of torso examination in trauma patients3. Existing reviews have focused on blunt abdominal and pelvic trauma, produced conflicting results, and do not include the most recent original data on this subject3,4. To address these knowledge gaps, we conducted a systematic review and meta-analysis of the diagnostic accuracy of physical examination to identify potentially life-threatening torso injuries in adults and to evaluate factors that may affect its accuracy. This study adheres to the Preferred Reporting Items for Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies (PRISMA-DTA). The protocol was prospectively registered on PROSPERO (CRD42021297840). The full methodology is available in the Supplementary Methods. Data from 35 studies, encompassing 54 243 injured patients, were included in this meta-analysis. The overall quality of included studies was assessed as moderate using the QUADAS-2 framework. Thirteen studies, describing 2465 patients, reported the diagnostic accuracy of chest examination. Among these patients, 1122 (45.5 per cent) had potentially life-threatening chest injuries. Clinical examination exhibited a moderate ability to identify these injuries (pooled sensitivity 76 per cent, specificity 86 per cent, Fig. 1). For abdominal examination, eight studies, describing 3280 patients, reported diagnostic accuracy. Among these patients, 494 (15.1 per cent) had potentially life-threatening abdominal injuries. Clinical examination again demonstrated a moderate ability to identify these injuries (pooled sensitivity 64 per cent, specificity 75 per cent, Fig. 1). Seventeen studies, describing 48 309 patients, reported the accuracy of pelvic examination. Among these patients, 8276 (17.1 per cent) had potentially life-threatening pelvic injuries. Once more, clinical examination demonstrated a moderate ability to identify these injuries (pooled sensitivity 78 per cent, specificity 90 per cent, Fig. 1). We performed pre-specified subgroup analyses for clinical factors that can influence the diagnostic accuracy of torso examination. Two factors—setting and level of consciousness—significantly affected diagnostic accuracy. Clinical examination performed in the pre-hospital setting was less likely to identify life-threatening injuries than when performed in-hospital (pooled sensitivity 46 per cent versus 76 per cent, P < 0.0001, Fig. 1). Similarly, examination of patients with a reduced level of consciousness significantly reduced the ability to identify life-threatening torso injuries compared to examination of awake patients (pooled sensitivity 63 per cent versus 90 per cent, P < 0.0001, Fig. 1). No significant effect was observed for injury mechanism or severity. Full reporting is available in the Supplementary Results and Appendices (Supplementary Figures 1–12, Tables 1–5). This study presents a comprehensive meta-analysis of the diagnostic accuracy of physical examination in trauma patients, highlighting its limitations in correctly identifying life-threatening torso injuries. The key implication is that physical examination alone cannot definitively rule in or rule out such injuries, which remain a significant cause of preventable trauma deaths. Even with normal examination findings, there remains a considerable chance of missed injuries, particularly in the pre-hospital setting or among unconscious patients. Consequently, our findings emphasize the place of immediate total-body CT during the initial hospital assessment of patients with suspected injuries5. In the pre-hospital setting, we advocate relying on an overall risk assessment alongside diagnosis for triage and therapeutic decisions, augmented by prognostic and diagnostic adjuncts (for example, ultrasound and validated decision support systems) where possible. No funding was required for this work. J.M.W. and Z.P. have received research funding from a precision trauma care research award from the Combat Casualty Care Research Program of the U.S. Army Medical Research and Materiel Command (DM180044). J.M.W. received funding from the Royal College of Surgeons of England and the Rosetrees Trust for salary support during the conduct of the study. Thomas Durrands (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review & editing), Mark Murphy (Data curation, Formal analysis, Investigation, Writing—review & editing), Jared M. Wohlgemut (Data curation, Formal analysis, Investigation, Writing—review & editing), Henry De'Ath (Writing—review & editing), and Zane Perkins (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing—review & editing). The other authors declare that they have no competing interests. Supplementary material is available at BJS online. The data that support the findings of this study are available from the corresponding author (T.D.), upon reasonable request. The paper is not based on previous communications to a society or meeting. Systematic Reviews & Meta-analyses; Level III (predominance of level III or lower studies).
No takes yet. Share an insight, caveat, or question.
Durrands et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: