Screening and Treatment Using a Risk-Based Approach with NT-proBNP Guidance in Diabetes Mellitus: The STRONG-DM Pilot Study.

BACKGROUND

Adults with type 2 diabetes mellitus (T2DM) have an elevated risk for developing heart failure (HF), but prescription rates of preventive therapies remain low.

OBJECTIVES

The purpose of this study was to evaluate the feasibility and effectiveness of a virtual cardio-kidney-metabolic (CKM) prevention strategy, including risk assessment and consultation for individuals with high risk of developing HF, to guide the prescription of preventive therapies.

METHODS

The study included patients with T2DM and no history of HF who were seen by a participating primary care physician (PCP) and matched control patients. The intervention included PCP education, electronic health record-based recommendations for N-terminal pro-B-type natriuretic peptide (NT-proBNP) screening, and virtual CKM consultations among patients with high risk of developing HF based on the WATCH-DM score (≥12) or NT-proBNP (≥125 pg/mL). Key outcomes included prescription rate of CKM therapies (sodium-glucose cotransporter 2 inhibitors, glucagon-like-peptide-1 receptor agonists, or finerenone) during the 6-month follow-up period that were compared between the intervention and matched (age, sex, race, CKM risk factors, CKM therapies) control group identified retrospectively from the cohort of patients seen during the same study period by nonparticipating PCPs within the same practices.

RESULTS

Among 545 patients (62.4% women; 33.6% Black), PCPs ordered NT-proBNP screening in 93.4% of patients, and 36.9% were identified as high risk for developing HF. NT-proBNP screening identified 16.7% additional patients as high-risk for HF who were deemed low risk based on WATCH-DM. Over the 180-day median follow-up, the intervention (vs matched control) group had significantly greater new prescriptions of sodium-glucose cotransporter 2 inhibitors (20.3% vs 5.8%; P < 0.001) and finerenone (4.1% vs 0.2%; P < 0.001), with no significant difference in glucagon-like-peptide-1 receptor agonists initiation. Overall, the intervention group demonstrated a 3-fold higher rate of initiating any CKM therapy (4.33 vs 1.43 new prescriptions/100-person-months; incidence rate ratios [IRRs]: 3.03; 95% CI: 2.18-4.31; P < 0.001). The greater uptake of CKM therapies in the intervention compared with control group was driven by greater prescriptions among high-risk (IRR: 5.03; 95% CI: 3.06-8.77) vs low-risk patients (IRR: 1.82; 95% CI: 1.15-2.96; P interaction = 0.005).

CONCLUSIONS

Among adults with T2DM and high risk for developing HF, a virtual CKM prevention strategy was feasible and increased prescription rates of preventive therapies. This virtual CKM prevention strategy should be evaluated further in a randomized clinical trial.