Cumulative incidence and predictors of neurosurgical interventions following nonsevere traumatic brain injury with mildly abnormal head imaging findings.

BACKGROUND

Incidence and predictors of neurosurgical interventions following nonsevere traumatic brain injury (TBI) with mildly abnormal head computed tomographic (CT) findings are poorly defined. Despite this, neurosurgical consultation is routinely requested in this patient population. Our objective was to determine incidence of neurosurgical intervention in this patient population and identify clinical and radiographic features predicting the subsequent need for these interventions.

METHODS

We identified all consecutive adult patients with nonsevere TBI admitted from January 1, 2001, through December 31, 2010. The definitions of "mildly abnormal initial head CT findings" and "neurosurgical interventions" were determined a priori by author consensus. Cumulative incidence of neurosurgical interventions was determined, and multivariate logistic regression was used to identify independent predictors of neurosurgical intervention.

RESULTS

Of 677 patients, 51 underwent neurosurgical intervention for a cumulative incidence of 7.5%. Only 1.6% required an intracranial procedure. In adjusted analysis, presence of coagulopathy (odds ratio [OR], 2.21; 95% confidence interval [CI], 1.13-4.3; p = 0.02), suspected cerebrospinal fluid leak (OR, 11.36; 95% CI, 2.83-45.58; p = 0.001), any basal cistern or sylvian fissure subarachnoid hemorrhage (OR, 2.94; 95% CI, 1.56-5.57; p = 0.001), depressed skull fracture (OR, 2.84; 95% CI, 1.29-6.28; p = 0.01), or unstable repeated head CT findings (OR, 2.81; 95% CI, 1.52-5.2; p = 0.001) remained an independent predictor of the need for subsequent neurosurgical intervention.

CONCLUSION

Among patients with nonsevere TBI and mildly abnormal head imaging findings in which routine neurosurgical consultation is obtained, there is a low incidence of neurosurgical interventions. Our findings suggest that routine early neurosurgical consultation in this patient population may not be necessary; however, this should be tested in a prospective, comparative study.

LEVEL OF EVIDENCE: Prognostic study, level III; therapeutic study, level IV.