• Diabetes Mellitus
• Diabetes & Endocrinology Clinical Trials Clinical Trials

Two analyses from the RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes) study [NCT00379769; Home PD et al. Lancet 2009] of cardiovascular (CV) outcomes in patients who were treated with rosiglitazone were presented at the European Society of Cardiology 2009 Annual Congress. The first evaluated rates of CV death or CV hospitalization, and the second evaluated rates of coronary events.

The first analysis showed no difference in the primary endpoint of CV hospitalization or CV death but did demonstrate increased rates of heart failure (HF) that led to hospitalization or death in subjects who were randomized to rosiglitazone plus metformin or a sulfonylurea (RSG) compared with a control group that was treated with a combination of metformin and a sulfonylurea.

Professor Michel Komajda, MD, Université Pierre et Marie Curie, France, Paris, presented the results of the post hoc analysis, showing that over the 5.5 years of follow-up in the RECORD study, subjects in the RSG group experienced similar rates of the primary endpoint (HR, 0.99; 95% CI, 0.85 to 1.16; p=0.93) compared with those on control, meeting the criteria for noninferiority. There were, however, significantly more fatal/nonfatal HF events in the RSG group (61 events in the RSG group vs 29 in the control group; HR, 2.10; 95% CI, 1.35 to 3.27; p = 0.001). The HF event rates for the two groups began to diverge early and continued to diverge throughout the trial.

Five baseline variables were shown to be significant predictors of HF risk: treatment with rosiglitazone (HR, 2.25; 95% CI, 1.42 to 3.58), age ≥60 years (HR, 3.81; 95% CI, 2.34 to 6.20), waist circumference ≥104 cm (HR, 3.52; 95% CI, 2.08 to 5.98), the presence of microalbuminuria/proteinuria (HR, 3.35; 95% CI, 2.18 to 5.14; all p<0.001), and baseline beta-blocker therapy (HR, 1.86; 95% CI, 1.20 to 2.90; p=0.006).

“These findings support the current recommendation that rosiglitazone should not be used in patients with symptomatic HF and should not continue to be used in the presence of HF,” said Prof. Komajda. Professor Kenneth Dickstein, MD, Stavanger University Hospital, Stavanger, Norway, the discussant for the presentation, noted that subjects with a history of treatment for HF were excluded from RECORD and that use of RSG in such patients also should be avoided.

A second analysis by Home et al. (Lancet 2009) also evaluated the first occurrence of MI as an endpoint and showed no difference between the RSG group and the control (HR, 1.14; 95% CI, 0.80 to 1.63) from the RECORD trial. Professor John McMurray, University of Glasgow, Glasgow, Scotland, presented results of a post hoc analysis that was focused on more broadly defined coronary events, including an analysis of time to the first of three expanded coronary endpoints in the overall trial cohort, total (including recurrent) events, and events that occurred among those subjects who experienced a “first MI” during the trial. The three additional composite outcomes were: any acute coronary syndrome ([ACS], defined as fatal MI, sudden death, or hospitalization for cardiac arrest, acute MI, or unstable angina [UA] pectoris); any ACS or hospitalization with “other” angina (defined as ACS plus “other” CV hospitalization attributed to angina pectoris); and any ACS, “other angina,” or coronary revascularization.

Similar to the results of the MI analysis in the RECORD study, no difference was observed between the RSG group and the control group for any of the newly analyzed composite outcomes (Table 1).

There were a total of 15 deaths in the RSG group (7 acute MI; 8 sudden cardiac deaths) versus 22 in the control group (10 acute MI; 12 sudden cardiac deaths). Total coronary events (cardiac arrest, acute MI, UA, other angina, or revascularization) also were similar between groups, wherein 221 events were experienced by 127 subjects in the RSG group versus 230 events in 128 subjects in the control group. Overall numbers of recurrent events in subjects who had a first MI also did not differ between treatment groups. Among the 60 survivors of a first MI in the RSG group, there were 7 recurrent MIs, 3 cases of UA, and 11 deaths (7 CV deaths). Among the 52 survivors of a first MI in the control group, there were 11 recurrent MIs, 2 cases of UA, and 12 deaths (10 CV deaths).

“In the RECORD trial, contrary to the meta analysis published by Nissen and Wolski [Nissen SE and Wolski K. N Engl J Med 2007],” said Prof. McMurray, “we did not see statistically significant increase in coronary outcomes, an excess of recurrent coronary events, or an excess of total or cardiovascular mortality in subjects treated with rosiglitazone compared with those receiving conventional therapy.”

• © 2009 MD Conference Express