Nationwide analysis of mortality and causes of death in heart failure: updated insights from the Thai Heart Failure Registry (THFR).
Heart failure (HF) remains associated with high mortality worldwide despite advances in guideline-directed therapy.
To characterise the patterns and causes of death among Thai patients with chronic HF enrolled in a nationwide registry and to identify predictors of mortality.
We performed a prospective cohort analysis of 2879 HF patients enrolled in the Thai Heart Failure Registry (THFR) which has enrolled patients since 2020. The present analysis included those enrolled between January 2021 and December 2023. Causes of death were ascertained via electronic health records and adjudicated as cardiovascular (CV), non-CV or undetermined. Mortality incidence rates were calculated per 100 person-years. Multivariable Cox proportional hazards models identified independent predictors of all-cause mortality.
Over a median follow-up of 23 months, 301 patients (10.5%) died (incidence rate 6.8 per 100 person-years). Of these, 149 (49.5%) were CV deaths, 116 (38.5%) were non-CV deaths, 36 deaths (12.0%) were undetermined. In HF with reduced ejection fraction/HF with mildly reduced ejection fraction patients, CV deaths predominated compared with non-CV mortality (5.3% vs 3.9%), whereas HF with preserved ejection fraction patients experienced higher non-CV mortality compared with CV mortality (5.5% vs 4.1%), particularly sepsis. Independent predictors of mortality included ischaemic aetiology (HR 1.34; 95% CI 1.06 to 1.69), atrial fibrillation (HR 1.57; 1.24-2.00), peripheral oedema at presentation (HR 2.11; 1.57-2.85), hyponatraemia (serum Na<135 mmol/L; HR 1.83; 1.39-2.39), hypokalaemia (serum K<3.5 mmol/L; HR 1.44; 1.06-1.95), elevated blood urea nitrogen (>25 mg/dL; HR 1.56; 1.23-1.98) and prior loop diuretic use (HR 1.61; 1.21-2.14). Good guideline-directed medical therapy adherence is associated with a lower mortality despite the adjustment of confounders.
Among Thai HF patients, mortality remains high, with distinct patterns of CV versus non-CV deaths across left ventricular ejection fraction phenotypes. Targeted strategies addressing ischaemic heart disease, rhythm management, fluid status and electrolyte disturbances may reduce mortality.
To characterise the patterns and causes of death among Thai patients with chronic HF enrolled in a nationwide registry and to identify predictors of mortality.
We performed a prospective cohort analysis of 2879 HF patients enrolled in the Thai Heart Failure Registry (THFR) which has enrolled patients since 2020. The present analysis included those enrolled between January 2021 and December 2023. Causes of death were ascertained via electronic health records and adjudicated as cardiovascular (CV), non-CV or undetermined. Mortality incidence rates were calculated per 100 person-years. Multivariable Cox proportional hazards models identified independent predictors of all-cause mortality.
Over a median follow-up of 23 months, 301 patients (10.5%) died (incidence rate 6.8 per 100 person-years). Of these, 149 (49.5%) were CV deaths, 116 (38.5%) were non-CV deaths, 36 deaths (12.0%) were undetermined. In HF with reduced ejection fraction/HF with mildly reduced ejection fraction patients, CV deaths predominated compared with non-CV mortality (5.3% vs 3.9%), whereas HF with preserved ejection fraction patients experienced higher non-CV mortality compared with CV mortality (5.5% vs 4.1%), particularly sepsis. Independent predictors of mortality included ischaemic aetiology (HR 1.34; 95% CI 1.06 to 1.69), atrial fibrillation (HR 1.57; 1.24-2.00), peripheral oedema at presentation (HR 2.11; 1.57-2.85), hyponatraemia (serum Na<135 mmol/L; HR 1.83; 1.39-2.39), hypokalaemia (serum K<3.5 mmol/L; HR 1.44; 1.06-1.95), elevated blood urea nitrogen (>25 mg/dL; HR 1.56; 1.23-1.98) and prior loop diuretic use (HR 1.61; 1.21-2.14). Good guideline-directed medical therapy adherence is associated with a lower mortality despite the adjustment of confounders.
Among Thai HF patients, mortality remains high, with distinct patterns of CV versus non-CV deaths across left ventricular ejection fraction phenotypes. Targeted strategies addressing ischaemic heart disease, rhythm management, fluid status and electrolyte disturbances may reduce mortality.
Authors
Yingchoncharoen Yingchoncharoen, Phrommintikul Phrommintikul, Buakhamsri Buakhamsri, Senthong Senthong, Chotenimitkhun Chotenimitkhun, Boonyapiphat Boonyapiphat, Charoenyos Charoenyos, Puwanant Puwanant, Krittayaphong Krittayaphong
View on Pubmed