Novel clinical risk score identifies T2D patients at greatest risk of heart failure
A newly developed clinical risk score can identify patients with type 2 diabetes (T2D) at greatest risk for hospitalization for heart failure (HHF) who would derive the most benefit from treatment with dapagliflozin, according to a study presented at ESC 2019.
“The TIMI Risk Score for Heart Failure in Diabetes (TRS-HFDM) … identified a gradient of increasing absolute reduction in the risk of HHF in patients treated with an SGLT-2* inhibitor,” said Dr David Berg from the Brigham and Women’s Hospital in Boston, Massachusetts, US.
“The results of this analysis … offer a practical tool to assist clinicians with risk stratification, counselling, and therapeutic decision making in patients with T2D,” said Berg and co-authors.
The association between 25 risk indicators** and HHF was first assessed in 8,212 T2D patients (with either established or multiple risk factors for atherosclerotic CVD) who received placebo during the SAVOR-TIMI 53*** study (derivation cohort; median age 65 years, 33 percent female), 228 of whom experienced ≥1 HHF events over a median 2.1-year follow-up.
The TRS-HFDM clinical risk score was then developed based on five independent clinical risk indicators of HHF (prior heart failure, prior atrial fibrillation [AF], coronary artery disease [CAD], estimated glomerular filtration rate [eGFR], and urine albumin-to-creatinine ratio [UACR]) which were selected due to significant associations with risk of HHF (p<0.001). “Each … indicator was assigned an integer weight proportional to the regression coefficient”, with increasing scores signifying higher risk (eg, 0 points = low risk, ≥3 points = very high risk). Prior heart failure and UACR >300 mg/g were accorded 2 points each, with the others each accorded 1 point.
The score was then assessed in 8,578 T2D patients who received placebo in the DECLARE-TIMI 58# trial (external validation cohort; median age 65 years, 38 percent female), 286 of whom experienced ≥1 HHF events over a median 4.2-year follow-up period.
The integer-based score of 0–7 points revealed a >20-fold gradient in the risk of HHF in the derivation and validation cohorts (score of ≥4 vs 0 points; ptrend<0.001), with a Harrell C index of 0.81 and 0.78, respectively. [ESC 2019, abstract 410]
The score was effective in assessing HHF risk in patients with established CVD and in those with multiple risk factors (Harrell C index, 0.80 and 0.77, respectively), as well as in those with and without history of heart failure (Harrell C index, 0.70 and 0.77, respectively).
The relative reduction in HHF risk with dapagliflozin was comparable among patients regardless of risk score (25–34 percent). However, absolute risk reduction (ARR) was greater among those with a higher risk of HHF at baseline (ptrend=0.04), with ARRs of 0.3 (hazard ratio [HR], 0.66 vs placebo), 0.6 (HR, 0.74), 1.5 (HR, 0.67), and 2.7 percent (HR, 0.75) in those with low, intermediate, high, and very high HHF risk, respectively. This was the equivalent of needing to treat 303, 172, 65, and 36 patients, respectively, to prevent one HHF event at 4 years.
According to the researchers, the mechanisms behind the benefits of different glucose-lowering drug classes on CVD risk is undetermined. “[T]he distinct profile of cardiovascular benefit for each of these drug classes highlights the need for improved models of risk stratification to optimally treat patients with T2D,” they said.
“[B]y using a simple, validated clinical risk score for HHF, clinicians can better characterize the risk profile of their patients and identify those patients who have the greatest absolute reduction in risk of HHF from treatment with SGLT2 inhibitors [as well as with future therapies],” they said, highlighting that the score should also be validated in non-clinical trial populations of patients with T2D.