Intrathecal Urokinase Infusion Through a Microcatheter into the Cisterna Magna to Prevent Cerebral Vasospasm: Experimental Study in Dogs

Abstract

BACKGROUND AND PURPOSE: Our preliminary report on intrathecal urokinase (UK) infusion into the cisterna magna (CM) with a microcatheter showed good results in terms of vasospasm prevention in humans. In this study, we evaluated the relationship between different urokinase (UK) infusion sites and their effect on vasospasm prevention by using our canine subarachnoid hemorrhage (SAH) model.

METHODS: At 24 hours after SAH induction, we injected 1000 IU/kg UK into the cisterna magna (CM) or lumbar sac (LS) of dogs by using a microcatheter inserted at the lumbar region. We then obtained serial angiograms and chronologically examined the changes in the mean diameter of the basilar artery (BA) during a 14-day period to determine the effect of the different injection sites on vasospasm prevention. At 24 hours after UK injection, one dog from each group was killed for gross inspection of the subarachnoid clot. To measure its concentration in the CM and sylvian fissure, UK (1000 IU/kg) was injected into the CM or LS of dogs without SAH; measurements were taken at 15-minute intervals until 4 hours after injection.

RESULTS: At 24 hours after UK injection, subarachnoid clot in front of the brain stem persisted strongly in the LS group; it had almost disappeared in the CM group. In the LS group, there was severe and persistent BA constriction during the 14-day observation period. In the CM group, the BA was constricted on day 3; however, gradual dilatation occurred over time. The mean diameter of the BA on days 7, 10, and 14 was 48.2%, 53.9%, and 58.9% in the LS group and 62.6%, 70.5%, and 82.3% in the CM group. The difference between the two groups was significant on days 7, 10, and 14 (P < .05). In dogs without SAH, the average maximum UK concentration in the CM and the sylvian fissure was 2.5 and 6.7 times higher, respectively, in the CM group than in the LS group.

CONCLUSIONS: In our canine SAH model, the administration of UK into the CM was significantly more effective in preventing cerebral vasospasms than was administration into the LS.