PULIDO - prediction usefulness of a 5-level score in determining outcomes after cardiac rehabilitation

Abstract Background Cardiopulmonary exercise testing (CPET) provides multiple physiologic markers with established prognostic relevance. However, simple, practical CPET-based tools for risk stratification in patients entering cardiac rehabilitation (CR) are lacking. Aim To identify CPET parameters associated with all-cause mortality and develop a cumulative, clinically applicable CPET-based risk score. Methods We retrospectively analyzed patients enrolled in a CR program who underwent CPET prior to CR initiation. Clinical, laboratory, echocardiographic, and CPET data were collected from patients’ records. CPET variables associated with lower survival were identified using receiver operating characteristic (ROC) analyses to determine optimal thresholds. A cumulative score (0–5) was constructed by assigning one point per abnormal CPET parameter. Survival was assessed with Kaplan–Meier curves and multivariate Cox proportional hazards models. Results Among 476 patients (mean age 61 years; 80% male; 84% ischemic heart disease), 67% had hypertension, 27% diabetes, 70% dyslipidemia, and 61% a history of smoking. Median serum creatinine and NT-proBNP were 0.93 mg/dL and 460 pg/mL, respectively; mean left ventricular ejection fraction was 50%. Over a mean follow-up of 2.5 years, 20 deaths occurred. Five CPET parameters were associated with lower survival: peak VO2 11.7 mL·kg−¹·min−¹, percent-predicted peak VO2 57%, PETCO2 31 mmHg, VE/VCO2 slope 32, and ventilatory power 5.7 (HR 2.5). The cumulative CPET score demonstrated good discrimination for mortality (AUC 0.709); a score ≥3 identified patients with significantly reduced survival (log-rank p0.001). In multivariate Cox regression, the score remained an independent predictor of all-cause mortality after adjusting for LVEF, age, NT-proBNP, eGFR, and cardiovascular risk factors (HR 1.551, 95% CI 1.160–2.075, p=0.003). Conclusions A simple additive score derived from five CPET parameters measured before CR initiation provides strong prognostic discrimination for all-cause mortality and may serve as a practical tool for risk stratification in cardiac rehabilitation candidates.For image description, please refer to the figure legend and surrounding text.