Pelvic angiography is effective for emergency pediatric patients with pelvic fractures: a propensity-score-matching study with a nationwide trauma registry in Japan

Abstract

Purpose: The aim of this study was to evaluate the association between the implementation of pelvic angiography (PA) and outcome in emergency pediatric patients with pelvic fracture. Methods: We extracted data on pelvic fracture patients aged ≤ 19 years between 2004 and 2015 from a nationwide trauma registry in Japan. The main outcome was hospital mortality. We assessed the relationship between implementation of PA and hospital mortality using one-to-one propensity-score-matching analysis to reduce potential confounding effects in comparing the PA group with the non-PA group. Results: In total, 1351 patients were eligible for our analysis, with 221 patients (16.4%) included in the PA group and 1130 patients (83.6%) included in the non-PA group. For all patients, the proportion of hospital mortality was higher in the PA group than in the non-PA group [13.6% (30/221) vs 7.1% (80/1130), crude odds ratio (OR) 2.062 (95% confidence interval (CI), 1.318–3.224); p = 0.002]. In the propensity-score-matched patients, the proportion of hospital mortality was lower in the PA group than in the non-PA group [10.5% (22/200) vs 18.2% (38/200), p = 0.027]. This finding was confirmed in both the multivariable logistic regression model [adjusted OR 0.392 (95% CI, 0.171–0.896); p = 0.026] and the conditional logistic regression model [conditional OR 0.484 (95% CI, 0.261–0.896); p = 0.021]. Conclusion: The implementation of PA was significantly associated with lower hospital mortality among emergency pediatric patients with pelvic fractures compared with the non-implementation of PA.