Pulmonary Complications and Sepsis Following Severe Acute Subdural Hematoma in Patients Who Underwent Craniotomy versus Craniectomy: A Propensity Score Matched Analysis

Abstract

Background The purpose of the study was to evaluate the impact of craniotomy (CO) and decompressive craniectomy (DC) for evacuation of acute subdural hematoma (SDH) on pulmonary complications and sepsis. Methods Study data were obtained from the National Trauma Data Bank (2007–2010). Only patients who met all of the following criteria were included in this analysis: sustained blunt injuries, presented with severe traumatic brain injury, sustained an associated SDH, presented with an initial Glasgow Coma Scale (GCS) score ≤ 8 and an Abbreviated Injury Scale score of head ≥ 3, and underwent a CO or DC within 4 hours of hospital arrival. Patient characteristics and outcomes were compared between CO and DC, the two procedural groups. The data were first compared between the two unmatched groups; then propensity score matching and a matched pairs analysis were performed. Results From the total population of 2,370 patients, 1,852 (78%) of them underwent CO, and the remaining 518 (22%) underwent DC. Some differences were found between the CO and DC groups regarding age (mean [standard deviation (SD)]: 47.9 years [22.8] versus 39.6 years [20.1]; p < 0.001), sex (male: 70.1% versus 74.7%; p = 0.05), race (white: 77.4% versus 73.4%; p = 0.06), the injury mechanism (fall: 50.7% versus 33.2%; p < 0.001), Injury Severity Score (mean [SD]: 28.0 [9.3] versus 30.5 [10.0]; p < 0.001), and GCS score (median [interquartile range] 3 [3–5] versus 3 [3–4])). After the propensity score matching, no significant differences were found between the groups on the variables just listed (all p > 0.05). No significant differences were seen between the CO and DC groups in the incidences of these conditions: acute respiratory distress syndrome (ARDS) (12.0% versus 8.1%; p = 0.20), pneumonia (34.9% versus 37.6%; p = 0.60), pulmonary embolism (PE) (3.5% versus 1.6%; p = 0.30), and systemic sepsis (6.2% versus 8.1%; p = 0.5). Conclusion Although most of the patients underwent CO for acute SDH, no significant differences were observed in the incidence of ARDS, pneumonia, PE, or systemic sepsis when compared with patients who underwent DC.