Abstract Aims To assess and compare long-term mortality and predictors thereof in de novo cardiac resynchronization therapy defibrillators (CRT-D) vs. upgrade from an implantable cardioverter-defibrillator (ICD) to CRT-D. Methods and results Study population consisted of 595 consecutive patients with CRT-D implanted between 2002 and 2015 in a tertiary care, university hospital, in a densely inhabited, urban region of Poland [480 subjects (84.3%) with CRT-D de novo implantation; 115 patients (15.7%) upgraded from ICD to CRT-D]. In a median observation of 1692 days (range 457–3067), all-cause mortality for de novo CRT-D vs. CRT-D upgrade was 35.5% vs. 43.5%, respectively (P = 0.045). On multivariable regression analysis including all CRT recipients, the previously implanted ICD was an independent predictor for death [hazard ratio (HR) 1.58, 95% confidence interval (CI) 1.10–2.29, P = 0.02]. For those, who were upgraded from ICD to CRT-D, the independent predictors for all-cause death were as follows: creatinine level (HR 1.01, 95% CI 1.00–1.02, P = 0.01), left ventricular end-systolic diameter (HR 1.07, 95% CI 1.02–1.11, P = 0.002), New York Heart Association (NYHA) IV class at baseline (HR 2.36, 95% CI 1.00–5.53, P = 0.049) and cardiac device-related infective endocarditis during follow-up (HR 2.42, 95% CI 1.02–5.75, P = 0.046). A new CRT scale (Creatinine ≥150 μmol/L; Remodelling, left ventricular end-systolic ≥59 mm; Threshold for NYHA, NYHA = IV) showed high prediction for mortality in CRT-D upgrades (AUC 0.70, 95% CI 0.59–0.80, P = 0.0007). Conclusion All-cause mortality in patients upgraded from ICD is significantly higher compared with de novo CRT-D implantations and reaches almost 45% within 4.5 years. A new CRT scale (Creatinine; Remodelling; Threshold for NYHA) has been proposed to help survival prediction following CRT upgrade.