Kinoshita, Takahiro, Yoshiya, Kazuhisa, Fujimoto, Yasunori, Kajikawa, Ryuichiro, Kiguchi, Takeyuki, Hara, Masahiko, Wakayama, Akatsuki, and Yoshimine, Toshiki
Background When it comes to evacuating intracranial hemorrhagic lesions in patients with traumatic brain injury (TBI), neurosurgeons perform either a craniotomy or a decompressive craniectomy (DC). The aim of the present study was to estimate the impact of DC on outcomes in elderly patients. Methods This retrospective cohort study, conducted in a neurosurgical institute in Japan from April 2009 to June 2014, included 91 consecutive patients with TBI (aged 60 years or older) who underwent evacuation of intracranial hemorrhagic lesions. Patients were divided into 2 groups: craniotomy only or DC. We set the primary endpoint as an unfavorable outcome (death or vegetative state), as evaluated on the Glasgow Outcome Scale at 6 months after injury. The secondary endpoints included existence of delayed hemorrhage and occurrence of hydrocephalus requiring shunt placement. The inverse probability of treatment weighting method was used to develop a propensity model to adjust for baseline imbalances between groups. Results The DC group exhibited greater severity both in clinical and computed tomography findings according to baseline characteristics. After we adjusted for these differences by inverse probability of treatment weighting using the propensity score, DC was significantly associated with unfavorable outcomes (adjusted odds ratio, 8.00; 95% confidential interval, 2.30–27.84; P = 0.002) and delayed hemorrhage (adjusted odds ratio, 13.42; 95% confidential interval, 1.52–118.89; P = 0.022). There was no significant difference in the occurrence of hydrocephalus requiring shunt placement. Conclusions DC in conjunction with evacuation of intracranial hemorrhagic lesions was associated with worse functional outcome in elderly patients with TBI. [ABSTRACT FROM AUTHOR]