Letter to the editor Tendencies in cause of death in patients with chronic heart failure and depressed systolic function. Response

To the Editor,

We congratulate and thank the authors for their letter, which reaffirms and complements our findings, highlighting the relevance of noncardiovascular mortality in the long-term in patients with heart failure (HF).¹ All the patients in our study¹ had a left ventricular ejection fraction < 40%; it has been demonstrated that, at this value, treatment affects disease prognosis, compared with 50% in the study by Moliner et al.² This feature, along with the more advanced functional class in our population, would have predisposed to a greater risk of death directly related to HF in the short- and mid-term. In our study, all patients were followed up for 4 years, allowing us to draw clear conclusions on mortality during this period. In the study by Moliner et al., the median follow-up was 4.2 years, which appeared similar but was actually highly variable, with quartiles of 1.9 years and 7.8 years, because the authors included patients throughout the study period (2002-2018), the same period in which the causes of death were studied. Therefore, the patients included in the earlier years and with a longer follow-up time (> 4 years) can be considered HF survivors, who are more vulnerable to death from other causes in the long-term. This would explain the sharp increase in noncardiovascular mortality observed in the last 3 years: up to two thirds of deaths in 2018.² Therefore, there is a bias due to the longer follow-up of HF survivors. However, both studies concur in finding a lower sudden death rate attributable to improvements in treatment, which reaffirms the importance of treatment adherence. They are also complementary, as the study by Moliner et al.² allows us to see what would happen to patients in our study who survived beyond this 4-year period. Current treatments have mainly reduced the risk of sudden death and delayed death due to HF, which, if avoided, means that other forms of noncardiovascular death predominate at long-term follow-up. Whether it is simply a question of time or whether there is a correlation between diseases such as HF and cancer remains an open question.