Cardiac rehabilitation referral in heart failure units: clinical profile and outcomes from the Spanish National SEC-EXCELENTE-IC Registry

Introduction and aim

Cardiac rehabilitation (CR) holds a Class I recommendation for heart failure (HF) patients to improve functional capacity and reduce hospitalizations. However, its implementation in real-world clinical practice remains suboptimal. This study aimed to characterize the clinical profile and comorbidities of HF patients referred to CR and evaluate its independent impact on 1-year clinical outcomes within a nationwide quality-certification program (SEC-EXCELENTE).

Methods

We conducted a prospective analysis of 2,245 patients with 1-year follow-up from the SEC-EXCELENTE registry, involving 68 accredited HF units in Spain between 2019 and 2023. Patients were stratified by CR referral. Comprehensive baseline characteristics, comorbidities, and guideline-directed medical therapy were assessed. The primary endpoint was a 1-year composite of all-cause mortality or HF hospitalization. Multivariable Cox regression models, including clinical complexity and treatment variables, were used to identify independent predictors.

Results

Out of 2,245 patients, only 219 (9.79%) were referred to CR. Referred patients were significantly younger (65.2 ± 11.7 vs. 71.7 ± 12.2 years; p<0.001), predominantly male (73.5% vs. 63.2%; p=0.003), and more likely to have an ischemic etiology (50.5% vs. 28.8%; p<0.001). They presented with lower LVEF (34.4 ± 12.8 % vs. 41.2 ± 15.5%; p<0.001), better renal function (eGFR 67.8 ± vs. 59.4 ml/min; p<0.001), and a higher prevalence of sinus rhythm (79.7% vs. 60.8%; p<0.001) (Table 1).

Treatment optimization was markedly superior in the CR group, with higher use of Sacubitril/Valsartan (52.1% vs. 40.5%; p=0.001), MRA (75.8% vs. 58.1%; p<0.001), and iSGLT2i (68.0% vs. 51.9%; p<0.001). Notably, quadruple therapy was achieved in 52.3% of CR patients compared to 31.2% in the non-referred group (p<0.001). Although CR referral was associated with a lower crude incidence of the primary endpoint (26.6 vs. 35.3 per 100 person-years; p=0.058), this association was attenuated after multivariable adjustment for age, comorbidities, and GDMT (Adjusted HR 0.98; 95% CI 0.72–1.35; p=0.921) (Figure 1).

Conclusions

In this contemporary national registry, CR referral remains remarkably low (<10%) and is heavily biased toward younger, male, ischemic patients with lower clinical complexity. These patients also receive significantly more intensive pharmacological treatment. The lack of independent prognostic impact suggests that the apparent clinical benefit of CR referral in real-world settings is largely driven by a "healthy-user" selection bias and a more aggressive optimization of the four pillars of HF therapy.Table 1.Clinical Profile and 1-Year OutcomesFigure 1.Survival by CR Referral Status