The role of early surgical intervention in civilian gunshot wounds to the head.

Surgical management of gunshot wounds of the head has remained a controversial issue in the care of civilian patients. In an attempt to determine who might benefit from aggressive surgical intervention, we examined 89 patients over a 3-year period who had suffered cranial gunshot wounds and had at least one computed tomographic scan of the head after admission. Patients were divided into those receiving early (less than 24 hours) surgical intervention (ES, n = 27), late (greater than 24 hours) surgical intervention (LS, n = 6) or no surgical intervention (NS, n = 56). Overall mortality was 63%. Ten of 27 patients (37%) in the ES group died compared with 46 of 56 patients (82%) in the NS group (p less than 0.0001). Glasgow Coma Scale (GCS) scores in the ES group averaged 7.86 +/- 4.72 and in the NS group 5.59 +/- 4.42 (p less than 0.05). The GCS scores in the LS group (all of whom survived) were significantly higher than those of the other two groups, 12.17 +/- 4.10. The number of patients with GCS scores of 3 or 4 on admission was significantly less in the ES (41%) than in the NS group (66%, p = 0.035) and survival was better with surgery (36%) than without (3%, p = 0.007). Patients with mass lesions (clot, ventricular blood) were more often found in the ES group (17/27) than in the NS group (18/56) (p = 0.008). Patients with bihemispheric injuries fared better with surgery (7 of 14 survivors) than without (2 of 33 survivors, p = 0.0003). Only one infectious complication (brain abscess) was encountered in the LS group. No delayed intracranial complications in survivors in the NS group were seen.(ABSTRACT TRUNCATED AT 250 WORDS)