Highlight
– In a Danish nationwide active‑comparator new‑user cohort (July 2020–2023), initiating SGLT2 inhibitors as the third foundational therapy after RASI and beta‑blocker was associated with a 30% lower risk of all‑cause mortality versus initiating MRAs (weighted HR 0.70, 95% CI 0.57–0.86).
– Cardiovascular mortality was also lower with SGLT2 initiation (wHR 0.65, 95% CI 0.49–0.87); the effect on heart failure hospitalization was directionally favorable but non‑significant (wHR 0.89, 95% CI 0.74–1.07).
Background and clinical context
Heart failure with reduced ejection fraction (HFrEF) remains a leading cause of morbidity and mortality worldwide. Contemporary guideline frameworks recommend early initiation and up‑titration of four “foundational” pharmacotherapies: renin‑angiotensin system inhibitors (RASI, including angiotensin‑converting enzyme inhibitors or angiotensin receptor blockers and, increasingly, angiotensin receptor‑neprilysin inhibitor [ARNI]), beta‑blockers, mineralocorticoid receptor antagonists (MRAs), and sodium‑glucose co‑transporter‑2 (SGLT2) inhibitors. Landmark randomized trials have established substantial benefits for MRAs (e.g., RALES, EMPHASIS‑HF) and SGLT2 inhibitors (e.g., DAPA‑HF, EMPEROR‑Reduced) on mortality and heart failure outcomes.
In practice, these therapies are often introduced sequentially rather than simultaneously, and clinicians lack definitive randomized data to guide optimal sequencing. The Danish nationwide cohort study by Svanström et al. (Lancet Regional Health Europe, 2025) addresses a pragmatic question: among patients with HFrEF already receiving RASI and beta‑blocker, which agent should be introduced third — an MRA or an SGLT2 inhibitor — to best reduce mortality and cardiac events?
Study design and methods
This was a nationwide, observational pharmacoepidemiologic study in Denmark from July 2020 through 2023. Inclusion criteria focused on adults aged ≥45 years with documented left ventricular ejection fraction ≤40% who were on background RASI and beta‑blocker therapy. The investigators used an active‑comparator new‑user design, comparing patients newly started on an MRA (spironolactone 63% or eplerenone 37%) versus those newly started on an SGLT2 inhibitor (dapagliflozin 74% or empagliflozin 26%), specifically as the third foundational therapy. This design reduces bias related to treatment initiation dynamics and attempts to emulate a pragmatic trial.
Baseline covariates were balanced using inverse‑probability of treatment weighting (IPTW) derived from propensity scores, aiming to account for measured confounding (demographics, comorbidities, prior hospitalizations, concurrent medications, and healthcare‑utilization measures). The primary outcome was all‑cause mortality. Secondary outcomes included cardiovascular death, heart failure hospitalization, and the composite of death or HF hospitalization. Weighted hazard ratios (wHRs) were estimated using Cox proportional hazards models.
Key findings
The cohort comprised 4,185 new MRA users and 2,565 new SGLT2 inhibitor users. Unweighted crude event rates showed 423 deaths in the MRA group (unweighted rate 6.3 per 100 person‑years) and 155 deaths in the SGLT2 group (5.8 per 100 person‑years).
After IPTW adjustment to balance measured baseline characteristics, initiation of an SGLT2 inhibitor versus an MRA was associated with:
- All‑cause mortality: wHR 0.70 (95% CI 0.57–0.86), corresponding to an absolute risk difference of −2.1 per 100 person‑years (95% CI −0.9 to −3.2).
- Cardiovascular death: wHR 0.65 (95% CI 0.49–0.87).
- Heart failure hospitalization: wHR 0.89 (95% CI 0.74–1.07) — a non‑statistically significant trend favoring SGLT2 inhibitors.
- Composite of cardiovascular death or HF hospitalization: wHR 0.83 (95% CI 0.71–0.97).
These results indicate a consistent mortality benefit associated with choosing an SGLT2 inhibitor over an MRA when used as the third agent after RASI and beta‑blocker in this population.
Interpretation and clinical plausibility
The observed mortality advantage with SGLT2 inhibitors is biologically plausible and consistent with randomized evidence. DAPA‑HF (dapagliflozin) and EMPEROR‑Reduced (empagliflozin) demonstrated reductions in the composite of cardiovascular death or heart failure hospitalization and favorable effects on disease progression across background therapy, including in patients already treated with MRAs. SGLT2 inhibitors exert multiple cardioprotective mechanisms beyond glycemic control — natriuresis, reduction in intravascular volume, improved myocardial energetics, anti‑inflammatory and antifibrotic effects, and renal protection — which may contribute to mortality reduction.
MRAs (spironolactone, eplerenone) have long‑standing mortality benefits, particularly in patients with more severe symptoms and low EF, mediated through blockade of aldosterone‑driven fibrosis, sodium retention, and neurohormonal activation. However, real‑world use can be constrained by hyperkalemia risk and declining renal function, potentially limiting persistence and dosing.
Thus, in a contemporary cohort, preferential initiation of SGLT2 inhibitors may confer broader tolerability and survival benefits, while MRAs remain a critical component of guideline‑directed medical therapy for many patients.
Strengths of the study
- Nationwide coverage in Denmark with linkage of prescription, hospitalization, and vital status registries provides near‑complete ascertainment of exposures and hard outcomes.
- The active‑comparator new‑user design reduces immortal time bias and confounding related to treatment sequencing.
- IPTW based on an extensive set of measured covariates attempts to emulate randomization and balance groups for observed confounders.
- Outcomes were clinically relevant and robust (all‑cause and CV mortality).
Limitations and cautions
- Residual confounding by unmeasured factors remains possible in any observational study. Important clinical variables that may not be fully captured in administrative data include New York Heart Association (NYHA) functional class, natriuretic peptide levels (BNP/NT‑proBNP), precise EF values beyond the ≤40% inclusion, blood pressure, serum potassium and creatinine at baseline and during follow‑up, and frailty indices.
- Indication bias may exist: clinicians may preferentially start MRAs in patients perceived as sicker (e.g., more symptomatic, lower EF) or with specific phenotype, which could inflate mortality in the MRA group despite weighting. Conversely, SGLT2 inhibitors may have been preferentially prescribed for patients with diabetes or better renal profiles.
- Data on concomitant ARNI use, device therapy (ICD/CRT), or doses of RASI and beta‑blockers were not detailed; cross‑trial comparisons and interpretability depend on these background therapies.
- Follow‑up duration may be limited for some outcomes; long‑term comparative effectiveness and safety (e.g., hyperkalemia rates, kidney outcomes) require further evaluation.
- Generalizability to other healthcare systems, ethnic groups, or regions with different prescribing patterns may be limited.
Clinical implications
The study supports prioritizing SGLT2 inhibitor initiation as the third foundational therapy after RASI and beta‑blocker in many patients with HFrEF, given the associated reductions in all‑cause and cardiovascular mortality in a large real‑world cohort. This does not imply MRAs are dispensable. MRAs retain a proven role — particularly in patients with persistent symptoms, advanced remodeling, or low EF — and should be considered when not contraindicated (e.g., acceptable potassium and renal function).
Practical application should be individualized: consider SGLT2 initiation early in patients with HFrEF unless contraindications exist (e.g., type 1 diabetes, severe renal dysfunction per local labeling), and monitor renal function and electrolytes. MRAs should be added promptly if residual indications remain, with careful potassium and creatinine surveillance. Integration with ARNI therapy, device use, and patient preferences remains essential.
Research and guideline implications
These observational findings provide important real‑world evidence that complements randomized trials. However, definitive sequencing recommendations would ideally come from randomized comparisons or pragmatic trials testing sequencing strategies. Future work should focus on head‑to‑head randomized or high‑quality emulation studies across diverse populations, with granular clinical data (NYHA class, biomarkers, device therapy) to better characterize effect heterogeneity and safety profiles.
Guideline committees may incorporate these real‑world data into consensus statements emphasizing earlier SGLT2 initiation, but recommendations should continue to emphasize individualized, comprehensive implementation of all four foundational therapies where feasible.
Conclusion
In this large Danish nationwide cohort, initiation of an SGLT2 inhibitor as the third foundational therapy after RASI and beta‑blocker was associated with a clinically meaningful reduction in all‑cause and cardiovascular mortality compared with initiating an MRA. While the findings support prioritizing SGLT2 inhibitors in treatment sequencing for many patients with HFrEF, clinicians should weigh individual patient characteristics, potential contraindications, and the continued role of MRAs. Randomized sequence trials or enriched observational emulations with more granular clinical data would strengthen causal inference and inform definitive sequencing guidance.
Funding and data
The study was supported by the Novo Nordisk Foundation and Karolinska Institutet. The analysis used national registry data from Denmark (July 2020–2023) as reported in Svanström et al., Lancet Regional Health Europe 2025.
Selected references and reading
– Svanström H, Mkoma GF, Hviid A, Pasternak B. Initiation of SGLT2 inhibitors versus mineralocorticoid receptor antagonists as third‑line therapy in heart failure with reduced ejection fraction: a nationwide cohort study. Lancet Reg Health Eur. 2025 Oct 27;60:101510.
– McMurray JJV, et al. DAPA‑HF Investigators. Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction. N Engl J Med. 2019.
– Packer M, et al. EMPEROR‑Reduced Investigators. Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med. 2020.
– Pitt B, et al. RALES Investigators. The effect of spironolactone on morbidity and mortality in severe heart failure. N Engl J Med. 1999.
– Zannad F, et al. EMPHASIS‑HF Investigators. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med. 2011.
– 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure.
Note: Readers should consult the full article by Svanström et al. for detailed methodology, subgroup analyses, and supplementary data.