Characteristics of the included studies.

This study aimed to assess the predictive ability of the shock index (SI) and the shock index, pediatric age-adjusted (SIPA) for mortality among pediatric patients with trauma (aged ≤ 18 years). A systematic search used PubMed, Embase, and Cochrane Library databases to identify pertinent articles published from their inception to 13 February 2023. For each SI and SIPA, the pooled sensitivity, specificity, diagnostic odds ratio (DOR), and area under the summary receiver operating characteristic curve (AUC) with the corresponding 95% confidence intervals were calculated. We planned a priori meta-regression analyses to explore heterogeneity using the following covariates: country, clinical setting, type of center, data source, and cutoff value. Twelve studies were included based on the inclusion criteria. Among them, nine studies with 195,469 patients were included for the SIPA at the hospital, four studies with 4,970 patients were included for the pre-hospital SIPA, and seven studies with 606,445 patients were included to assess the ability of the SI in predicting mortality. The pooled sensitivity and specificity with 95% confidence interval for predicting mortality were as follows: 0.58 (0.44–0.70) and 0.72 (0.60–0.82), respectively, for the SIPA at the hospital; 0.61 (0.47–0.74) and 0.67 (0.61–0.73), respectively, for the pre-hospital SIPA; and 0.71 (0.59–0.81) and 0.45 (0.31–0.59), respectively for the SI. The DOR were 3.80, 3.28, and 2.06 for the SIPA at the hospital, pre-hospital SIPA, and SI, respectively. The AUC were 0.693, 0.689, and 0.618 for the SIPA at the hospital, pre-hospital SIPA, and SI, respectively. The SI and SIPA are simple predictive tools with sufficient accuracy that can be readily applied to pediatric patients with trauma, but SIPA and SI should be utilized cautiously due to their limited sensitivity and specificity, respectively.