Association Between In-Hospital Exposure to Pharmacological Classes Recommended for Heart Failure and Mortality in a Heterogeneous Cardiovascular Critical-Care Population: A Multicenter Cohort Study

Background and Objectives: Pharmacological classes recommended for heart failure (HF) are also prescribed for indications other than HF. Their use in heterogeneous cardiovascular critical-care populations therefore does not necessarily represent HF-directed guideline-directed medical therapy. We examined the association between in-hospital exposure to these medication classes and in-hospital mortality, without assuming HF-specific treatment indications or causal treatment effects. Materials and Methods: This retrospective multicenter cohort included 8622 adults in a cardiovascular critical-care dataset with admission dates from June 2014 to October 2021. Exposure was defined as administration during at least 50% of the index hospitalization of each of four medication classes: a renin–angiotensin system inhibitor or angiotensin receptor–neprilysin inhibitor (counted as one class), a β-blocker, a mineralocorticoid receptor antagonist, and a sodium–glucose cotransporter 2 inhibitor. Patients were categorized by exposure to 0–1, 2, or ≥3 classes. The primary outcome was in-hospital mortality. Multivariable logistic regression adjusted for age, sex, hypertension, diabetes mellitus, documented HF, chronic kidney disease, and admission systolic blood pressure >100 mmHg. Results: Among 8622 patients, only 8.2% had documented HF; 6985 were exposed to 0–1 medication classes, 1388 to 2 classes, and 249 to ≥3 classes. Crude in-hospital mortality was 9.46%, 2.09%, and 3.21%, respectively (overall p < 0.001). Compared with the 0–1-class group, the adjusted odds ratios (ORs) for mortality were 0.19 (95% confidence interval [CI], 0.13–0.28) for 2 classes and 0.25 (95% CI, 0.12–0.53) for ≥3 classes. Documented HF prevalence and N-terminal pro–B-type natriuretic peptide (NT-proBNP) levels increased across exposure categories, indicating differences in clinical phenotype and likely treatment selection. Conclusions: In a heterogeneous cardiovascular critical-care cohort, greater in-hospital exposure to pharmacological classes commonly recommended for HF was associated with lower observed in-hospital mortality relative to exposure to 0–1 classes. However, medication indications were unavailable, the association was not strictly monotonic, and treatment eligibility, clinical stability, and survivor-treatment or immortal-time bias may substantially explain the findings. These descriptive, hypothesis-generating results neither establish treatment benefit nor justify HF-specific pharmacotherapy in patients without an appropriate indication.