312 DOES EXCLUDING PATIENTS WITH ORTHOSTATIC HYPOTENSION AFFECT SAFETY ESTIMATES IN HYPERTENSION TRIALS? A SYSTEMATIC REVIEW AND COMPARATIVE META-ANALYSIS

Abstract

Abstract Background In order to apply the findings of antihypertensive randomised controlled trials to older adults, it is important to understand the populations included and consider the generalizability of the results to cohorts that may have been excluded. The purpose of this study was to determine whether participants with orthostatic hypotension were included in randomised controlled trial of antihypertensive therapy and whether adverse event rates of particular importance to this cohort; namely falls or syncope differed based on their exclusion. Methods We performed a systematic review and meta-analysis of randomised controlled trials comparing antihypertensives to placebo, combination of antihypertensive agents compared to fewer antihypertensive or higher compared to lower blood pressure targets that reports falls or syncope outcomes. A random-effects meta-analysis was used to estimate a pooled treatment-effect overall in subgroups of trials which excluded patients with orthostatic hypotension and trials which did not exclude patients with orthostatic hypotension. Difference in treatment effect was assessed by testing P for interaction. The primary outcome measure was falls events. Results Forty-Seven trials were included, including 18 trials which excluded those with orthostatic hypotension and 29 trials which did not exclude those with orthostatic hypotension. Thirteen trials (n=94,222) reported falls. The baseline incidence of falls in the control group was 4.8% in trials which excluded orthostatic hypotension compared to 8.8% in trials which did not exclude participants with orthostatic hypotension. The association of antihypertensive treatment and falls was similar for trials which excluded those with orthostatic hypotension (OR 1.00; 95%CI, 0.89-1.13) and trials which did not exclude those with orthostatic hypotension (OR, 1.02; 95%CI, 0.88 –1.18). Conclusion The exclusion of patients with orthostatic hypotension may under-estimate the event rate of adverse events such as falls but does not appear to affect relative risk estimates associated with antihypertensives.