Do Patient Characteristics Explain the Differences in Outcome Between Medically Treated Patients in SAMMPRIS and WASID?

BACKGROUND AND PURPOSE

The Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) medical group had a much lower primary end point rate than predicted from the preceding Warfarin Aspirin Symptomatic Intracranial Disease (WASID) trial. This result has been attributed to the aggressive medical therapy used in SAMMPRIS, but an alternative hypothesis is that SAMMPRIS patients were at lower risk. We undertook analyses to evaluate these competing hypotheses.

METHODS

Using proportional hazards regression, we compared the SAMMPRIS primary end point between SAMMPRIS medical patients and WASID patients meeting the same qualifying criteria adjusted for confounding baseline characteristics.

RESULTS

The unadjusted comparison of the SAMMPRIS primary end point showed a significantly higher risk for WASID patients (P=0.009, logrank test) with 12 month Kaplan-Meier estimates of 21.9% in WASID and 12.6% in SAMMPRIS and hazard ratio 1.9 (95% confidence interval =1.2-3.0). The analyses identified the following confounding factors that varied between the studies and that conferred a higher risk: lack of statin use at enrollment (hazard ratio =1.8, 95% confidence interval =1.1-2.9, P=0.027) that was more prevalent among WASID patients (39% versus 14%, P<0.0001) and prior infarcts in the territory of the symptomatic vessel (hazard ratio =1.8, 95% confidence interval =1.1-2.9, P=0.023) that was more prevalent among SAMMPRIS patients (34% versus 22%, P=0.015).The hazard ratio for WASID versus SAMMPRIS adjusted for these 2 characteristics was 1.9 (95% confidence interval =1.1-3.2).

CONCLUSIONS

After adjustment for confounding baseline characteristics, WASID patients had an almost 2-fold higher risk of the SAMMPRIS primary end point, which supports the hypothesis that the lower rate of the primary end point in the medical arm of SAMMPRIS compared with WASID patients was as a result of the aggressive medical management used in SAMMPRIS.

CLINICAL TRIAL REGISTRATION

URL: http://www.clinicaltrials.gov. Unique identifier: NCT00576693.