An Evidence-Based Approach to the Efficient Use of Computed Tomography Imaging in the Neurosurgical Patient

Abstract

Computed tomography (CT) is the current standard for rapidly diagnosing some of the more common structural pathologies that affect the neurosurgical patient perioperatively. With this convenience comes the potential for its overuse. To investigate the utility of head CT scans ordered for various clinical indications. All head CT studies ordered by the UCLA Neurosurgery Department from August 15, 2011 through December 15, 2011, were prospectively studied. Variables collected included demographic information, diagnosis, surgical procedures, indication for CT, CT findings, and whether the study led to a documentable change in management. There were 801 head CT studies ordered for the 462 patients who were admitted to the neurosurgical service. The authors identified 14 indications for ordering a head CT with the following probabilities of a positive finding: examination change (17/56, 30.3%), follow-up (4-6 hours after intracerebral hemorrhage; 16/126, 12.7%), CT angiography (11/30, 36.7%), routine postoperative imaging (6/126, 4.7%), postventriculostomy placement (4/62, 6.5%), immediately before (4/31, 12.9%) or after removal of (2/42, 4.8%) a ventriculostomy, surveillance (>24 hours after intracerebral hemorrhage or external ventricular drain placement) (3/66, 4.5%), headaches (2/8, 25%), ground level fall (1/8, 12.5%), intracranial pressure spikes (2/6, 33.3%), and delayed (6-24 hours after intracerebral hemorrhage; 1/25, 4%). The probability of discovering a clinically significant finding varies widely for each of the listed study indications. This prospective analysis of all CT scans ordered at a single institution suggests that imaging studies obtained without a change in neurological status were unlikely to produce a positive finding, and even when there was a positive finding, it was extremely unlikely to result in any intervention.