Shihchen Kuo1, Wen Ye2, Justin Duong3, William H Herman4. 1. Department of Internal Medicine, University of Michigan, Ann Arbor, MI, United States. 2. Department of Biostatistics, University of Michigan, Ann Arbor, MI, United States. 3. College of Literature, Science and the Arts, University of Michigan, Ann Arbor, MI, United States. 4. Department of Internal Medicine, University of Michigan, Ann Arbor, MI, United States; Department of Epidemiology, University of Michigan, Ann Arbor, MI, United States. Electronic address: wherman@umich.edu.
Abstract
AIMS: It is unclear whether the favorable impact of empagliflozin on cardiovascular outcomes (CVOs) is due to its effect on multiple cardiometabolic risk factors (CRFs). METHODS: We used the Michigan Model for Diabetes, a validated computer simulation model, and published data from the EMPA-REG OUTCOME trial to estimate three-year CVOs in the placebo and pooled empagliflozin treatment groups to assess whether the observed benefits might be attributable to differences in CRFs. RESULTS: When we programmed the model to match the baseline characteristics of the trial population and the reported trajectories of five CRFs (weight, HbA1c, systolic blood pressure, low- and high-density lipoprotein cholesterol), the simulated hazard ratio (HR) for the primary composite CVO did not differ from the reported result. The simulated HRs for fatal/nonfatal myocardial infarction and coronary revascularization procedure fell within the reported 95% confidence intervals (CIs), but those for fatal/nonfatal stroke, hospitalization for heart failure, cardiovascular death, and all-cause mortality fell outside the reported 95% CIs. The effects of empagliflozin on CRFs accounted for approximately half of the observed benefit for the primary composite CVO, but explained smaller proportions of risk reductions for hospitalization for heart failure, cardiovascular death, and all-cause mortality. CONCLUSIONS: The effects of empagliflozin on multiple CRFs account for some but not all of reduced risks of CVOs in the EMPA-REG OUTCOME trial. More comparable control of established CRFs in type 2 diabetes CVO trials of antidiabetic agents with pleiotrophic effects would facilitate the interpretation of the observed outcomes.
RCT Entities:
AIMS: It is unclear whether the favorable impact of empagliflozin on cardiovascular outcomes (CVOs) is due to its effect on multiple cardiometabolic risk factors (CRFs). METHODS: We used the Michigan Model for Diabetes, a validated computer simulation model, and published data from the EMPA-REG OUTCOME trial to estimate three-year CVOs in the placebo and pooled empagliflozin treatment groups to assess whether the observed benefits might be attributable to differences in CRFs. RESULTS: When we programmed the model to match the baseline characteristics of the trial population and the reported trajectories of five CRFs (weight, HbA1c, systolic blood pressure, low- and high-density lipoprotein cholesterol), the simulated hazard ratio (HR) for the primary composite CVO did not differ from the reported result. The simulated HRs for fatal/nonfatal myocardial infarction and coronary revascularization procedure fell within the reported 95% confidence intervals (CIs), but those for fatal/nonfatal stroke, hospitalization for heart failure, cardiovascular death, and all-cause mortality fell outside the reported 95% CIs. The effects of empagliflozin on CRFs accounted for approximately half of the observed benefit for the primary composite CVO, but explained smaller proportions of risk reductions for hospitalization for heart failure, cardiovascular death, and all-cause mortality. CONCLUSIONS: The effects of empagliflozin on multiple CRFs account for some but not all of reduced risks of CVOs in the EMPA-REG OUTCOME trial. More comparable control of established CRFs in type 2 diabetes CVO trials of antidiabetic agents with pleiotrophic effects would facilitate the interpretation of the observed outcomes.
Authors: Michael Willis; Christian Asseburg; April Slee; Andreas Nilsson; Cheryl Neslusan Journal: Pharmacoeconomics Date: 2021-02-13 Impact factor: 4.981