TY - JOUR

T1 - The effect of sodium-glucose transport protein 2 inhibitors on mortality and heart failure in randomized trials versus observational studies

AU - Krogh, Jesper

AU - Hjorthoj, Carsten

AU - Kristensen, Soren L.

AU - Selmer, Christian

AU - Haugaard, Steen B.

PY - 2021

Y1 - 2021

N2 - Aim Randomized clinical trials (RCTs) allocating type 2 diabetes patients to treatment with sodium-glucose transport protein 2 (SGLT-2) inhibitors or placebo have found significant effects on the risk of heart failure and modest effects on mortality. In the wake of the first trials, a number of observational studies have been conducted, some of these reporting a mortality reduction of 50% compared to active comparators. In this review, we systematically assess and compare the results on all-cause mortality, cardiovascular mortality and heart failure hospitalization observed in RCTs with the results obtained in observational studies.Method We performed a systematic bibliographical search including cardiovascular outcome trials and observational studies assessing the effect of SGLT-2 inhibitors on mortality and heart failure.Results Seven RCTs and 23 observational studies were included in the current review. The observed heterogeneity between study results for all-cause mortality (p-interaction < 0.001) and cardiovascular mortality (p-interaction < 0.001) was explained by study type, whereas this was not the case for heart failure (p-interaction = 0.18).Conclusion Methodological considerations such as the omission of important confounders, immortal-time bias and residual confounding such as unmeasured social economic inequality may be the cause of the inflated results observed in observational studies and that calls for caution when observational studies are used to guide treatment of patients with type 2 diabetes.

AB - Aim Randomized clinical trials (RCTs) allocating type 2 diabetes patients to treatment with sodium-glucose transport protein 2 (SGLT-2) inhibitors or placebo have found significant effects on the risk of heart failure and modest effects on mortality. In the wake of the first trials, a number of observational studies have been conducted, some of these reporting a mortality reduction of 50% compared to active comparators. In this review, we systematically assess and compare the results on all-cause mortality, cardiovascular mortality and heart failure hospitalization observed in RCTs with the results obtained in observational studies.Method We performed a systematic bibliographical search including cardiovascular outcome trials and observational studies assessing the effect of SGLT-2 inhibitors on mortality and heart failure.Results Seven RCTs and 23 observational studies were included in the current review. The observed heterogeneity between study results for all-cause mortality (p-interaction < 0.001) and cardiovascular mortality (p-interaction < 0.001) was explained by study type, whereas this was not the case for heart failure (p-interaction = 0.18).Conclusion Methodological considerations such as the omission of important confounders, immortal-time bias and residual confounding such as unmeasured social economic inequality may be the cause of the inflated results observed in observational studies and that calls for caution when observational studies are used to guide treatment of patients with type 2 diabetes.

KW - SGLT&#8208

KW - 2 inhibitors

KW - real&#8208

KW - world studies

KW - propensity score

KW - heart failure

KW - mortality

KW - diabetes

KW - TYPE-2 DIABETES-MELLITUS

KW - CVOT SUMMIT

KW - CARDIOVASCULAR OUTCOMES

KW - CVD-REAL

KW - SGLT2 INHIBITORS

KW - LOWERING DRUGS

KW - LOWER RISK

KW - DEATH

KW - DAPAGLIFLOZIN

KW - ASSOCIATION

U2 - 10.1111/dme.14600

DO - 10.1111/dme.14600

M3 - Review

C2 - 33991127

VL - 38

JO - Diabetic Medicine Online

JF - Diabetic Medicine Online

SN - 1464-5491

IS - 9

M1 - 14600

ER -