Anterior-posterior view by full-body digital X-ray to rule out severe spinal injuries in Polytraumatized patients

Abstract Background Spinal injuries are present in 16–31% of polytraumatized patients. Rapid identification of spinal injuries requiring immobilization or operative treatment is essential. The Lodox-Statscan (LS) has evolved into a promising time-saving diagnostic tool to diagnose life-threatening injuries with an anterior-posterior (AP)-full-body digital X-ray. Methods We aimed to analyze the diagnostic accuracy and the interrater reliability of AP-LS to detect spinal injuries in polytraumatized patients. Therefore, within 3 years, AP-LS of polytraumatized patients (ISS ≥ 16) were retrospectively analyzed by three independent observers. The sensitivity and specificity of correct diagnosis with AP-LS compared to CT scan were calculated. The diagnostic accuracy was evaluated by using the area under the ROC (receiver operating characteristic curve) for sensitivity and specificity. Interrater reliability between the three observers was calculated using Fleiss’ Kappa. The sensitivity of AP-LS was further analyzed by the severity of spinal injuries. Results The study group included 320 patients (48.5 years ±19.5, 89 women). On CT scan, 207 patients presented with a spinal injury (65%, total of 332 injuries). AP-LS had a low sensitivity of 9% (31 of 332, range 0–24%) and high specificity of 99% (range 98–100%). The sensitivity was highest for thoracic spinal injuries (14%). The interrater reliability was slight (κ = 0.02; 95% CI: 0.00, 0.03). Potentially unstable spinal injuries were more likely to be detected than stable injuries (sensitivity 18 and 6%, respectively). Conclusion This study demonstrated high specificity with low sensitivity of AP-LS in detecting spinal injuries compared to CT scan. In polytraumatized patients, AP-LS, implemented in the Advanced Trauma Life Support-algorithm, is a helpful tool to diagnose life-threatening injuries. However, if spinal injuries are suspected, performing a full-body CT scan is necessary for correct diagnosis.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.