Risk Factors Associated with Cardiac Dysfunction in Men with Erectile Dysfunction

Dyslipidemia more than doubled the odds of diastolic heart dysfunction in 398 men with ED; BMI was the only factor linking to both types

Journal: The Journal of Sexual Medicine | Published: 2026-03-09 | Type: Cross-sectional study | PMID: 41843808 Authors: Durukan Emil et al. — Depts. of Urology and Cardiology, Copenhagen University Hospital-Herlev & Gentofte Funding/COI: Not listed

Summary

This Danish cross-sectional study used echocardiography to look for signs of cardiac dysfunction in 398 men presenting with ED. It found that BMI was independently associated with both left ventricular (LV) systolic and diastolic dysfunction, while elevated HbA1c and diastolic blood pressure were linked to systolic dysfunction, and dyslipidemia to diastolic dysfunction. The practical takeaway is that cardiometabolic profiling in ED patients may reveal subclinical cardiac disease — but the design can't say which came first.

Claims

• LV systolic dysfunction — in multivariable analysis:
• BMI: OR 1.08 (95% CI 1.01–1.15) per unit
• Diastolic blood pressure: OR 1.04 (95% CI 1.01–1.07) per mmHg
• HbA1c: OR 1.06 (95% CI 1.01–1.12)
• LV diastolic dysfunction — in multivariable analysis:
• BMI: OR 1.09 (95% CI 1.00–1.19) — barely clears significance
• Dyslipidemia: OR 2.54 (95% CI 1.11–5.72) — the largest effect in the paper
• High-sensitivity CRP and lipid profile were significant in univariable analyses for systolic dysfunction but dropped out in multivariable models
• Cohort sourced from both urology outpatient clinic and randomly selected men >40 years from the general population

Study Quality

The 398-person cohort is a legitimate strength — echocardiographic phenotyping at this scale in a urology population is not common. The multivariable models are appropriate for a cross-sectional design, and the prospective ClinicalTrials.gov registration (NCT05285280) adds credibility. Recruitment from two distinct channels (clinic referrals and population-based random invitations) was a sensible attempt to reduce selection bias.

That said, the study cannot establish directionality. The associations between cardiometabolic markers and echocardiographic findings are plausible and biologically coherent, but this design fundamentally cannot answer whether metabolic disease predicts cardiac dysfunction in ED men, or whether men with worse baseline cardiac status are more likely to present with ED. Effect sizes, while statistically significant, are modest for most predictors — OR of 1.04–1.09 per unit of BMI or blood pressure translates to small absolute differences.

Red Flags

• No causal inference possible — the cross-sectional design is stated plainly but should temper enthusiasm for clinical implications
• Funding and conflicts of interest not reported — for a paper co-authored by cardiologists at an academic center this is more procedural gap than likely bias, but it's still absent
• Dyslipidemia CI is wide (OR 2.54, 95% CI 1.11–5.72) — the lower bound barely clears 1.0, and the upper bound is 5.72; this estimate is imprecise
• BMI's diastolic dysfunction association also barely clears significance (CI 1.00–1.19) — rounding artifact territory
• Definition of "cardiac dysfunction" thresholds and echo protocols are not described in the abstract; reader has to trust they were applied consistently

Strengths

• Prospectively registered
• Large cohort for this type of echocardiographic assessment in a urology setting
• Well-characterized population with multiple clinical variables enabling genuine multivariable adjustment
• Dual recruitment strategy reduces pure clinic-selection bias
• Collaboration between urology and cardiology departments — appropriate expertise for the outcome

Verdict

This is a solid, well-designed cross-sectional study doing exactly what it claims to do. It documents that cardiometabolic risk factors are not randomly distributed among men with ED who show echocardiographic signs of cardiac dysfunction — BMI, HbA1c, diastolic BP, and dyslipidemia cluster with measurable LV impairment. That's useful epidemiology. The paper earns no more than it claims, though: it cannot and does not establish causation, the dyslipidemia finding has a wide confidence interval, and the absence of reported funding/COI is a minor gap. Worth reading if you're in the ED-cardiovascular interface space; not the kind of finding that reshapes clinical practice on its own.