Thiazolidinedione Therapy Associated with Improved Clinical Outcomes in Heart Failure Patients with Diabetes

Abstract

Background: The role of thiazolidinediones (TZDs) in the treatment of diabetic patients (pts) with Heart Failure (HF) is controversial. Though their favorable endovascular effects may contribute to delayed HF progression, their use has been associated with fluid retention and clinical decompensation. Concern regarding fluid retention, particularly when used in combination with Insulin (I), has led to several guideline recommendations against the use of TZDs in HF. To date, there have been no trials assessing the impact of TZD therapy on HF outcomes. We analyzed our experience with TZDs to better define the role of these agents in HF. Methods: We retrospectively analyzed 97 consecutive diabetic pts referred to our Heart Failure Treatment Program with systolic dysfunction (LVEF ≤ 35%) and NYHA class II-IV HF. 37% were treated with TZDs and 15% with TZD+I. The TZD+ and TZD− groups were well-matched with regard to baseline LVEF (TZD+ 23.7%, TZD− 25.9%, p=NS), glycosylated hemoglobin (TZD+ 7.1, TZD− 6.9, p=NS), and NYHA class (TZD+ 2.56, TZD− 2.79, p=NS). Baseline renal function was slightly worse in the TZD− group (1.15 vs 1.33 mg/dl, p=0.03). All pts were treated with ACE-inhibitors or ARBs and 97% with beta blockers. We assessed clinical outcomes at one year. Results: All 97 pts were alive at one year. Clinical outcomes were analyzed by TZD use both alone and in combination with I:Clinical Outcomes by Thiazolidinedione UseTZD+TZD−ΔNYHA class−0.21−0.12p=NSAll-cause hospitalizations (per patient)0.190.71p=0.01HF hospitalizations (per patient)0.030.16p=0.04Total hospital days0.672.72p=0.02Clinical Outcomes by Combined TZD and I UseTZD,I+TZD,I−ΔNYHA Class−0.13−0.16p=NSAll-cause hospitalizations (per patient)0.200.59p=0.11HF hospitalizations (per patient)0.200.43p=NSTotal Hospital Days (per patient)0.072.30p=0.0005There was a trend toward increased diuretic requirements in the TZD,I+ group (69.3mg vs 45.5 mg furosemide/day, p=0.08). This was not seen in pts treated with TZDs alone. Conclusions: We saw no adverse effect of TZDs on outcomes. In fact, TZD use was associated with reductions in both hospitalizations and hospital days. The hospital day reduction was independent of I use. These findings challenge recommendations against TZD use in HF. Clinial trials will determine the degree to which these findings can be extrapolated to broader HF populations.