This study investigated the clinical characteristics, medications, and outcomes of HF patients with HF from a very elderly population. While previous clinical studies of patients with ADHF have reported a mean age range of 71–78 years, the median age of the patients in this study was 86 years, and many of these patients had a life expectancy longer than the average. Many patients were not included in the previous study, and studies are needed to collect and organize data on these patients in the future.
Characteristics of HF patients in the super-aging society
The proportion of patients with HFpEF in this study was high (57.2%). Even within this age range, the patients were older and had a higher proportion of women than those with HFrEF. Previous clinical studies have also shown that patients with HFpEF tends to be older and more likely to be female, suggesting that the proportion of patients with HFpEF is increasing in older populations [7–9]. In contrast, in previous studies, patients with HFrEF tended to have more coronary artery disease complications than those with HFpEF, and patients with HFpEF tended to have more atrial fibrillation complications than those with HFrEF; however, no significant differences in complication rates were found in this study [7–9]. BNP and NT-proBNP levels were similar to those reported in previous studies and tended to be higher in patients with lower LVEF.
Pharmacotherapy of HF patients in a super-aging society
Although the JCS guidelines recommend ACEI/ARB, BB, and MRA to improve the prognosis of HFrEF, the induction rates of ACEI/ARB, BB, and MRA tended to be low for HFrEF in this study (62.9%, 73.4%, and 29.8%, respectively). This was not sufficient compared with previous clinical studies, and age and renal function could have affected the results [10–12]. In patients with HFrEF, the incidence of composite events did not differ between those treated with and without ACEI/ARB, BB, or MRA. Previous reports have shown that guideline-based medical treatment (using both ACEI/ARB and BB: GBMT) for patients with HFrEF aged < 80 years of age reduced the incidence of composite events; the present study results were similar, as there was no significant difference in the incidence of composite events between patients with HFrEF with and without GBMT for patients with HFrEF aged > 80 years of age [2]. In contrast, in patients with HFpEF, the incidence of composite events was lower in those taking ACEI/ARB and higher in those taking BB and MRA in univariate analysis. Previous studies examining outcomes in patients with HFpEF with or without ACEI/ARB administration [13], found no significant differences in CD and HF readmission but tended to have better outcomes. Another study reported no significant differences in cardiac death or HF readmission with or without BB and MRA administration [14, 15]. However, in the multivariate analysis, significant differences in the presence or absence of these drugs disappeared, suggesting that age and the severity of the patient’s background may have had an effect. In terms of the incidence of all-cause death, the presence or absence of ACEI/ARB administration was significantly different in patients with HFmrEF in the multivariate analysis; however, the hazard ratio was lower in patients with CKD or chronic lung disease, which is different from the results of previous Japanese studies, suggesting that the very small number of patients (n = 62) involved in the study was the reason for the extreme results [16].
Outcomes of HF patients in a super-aging society
The in-hospital mortality rate in previous studies was 6.4-8% in Japanese cohort [5, 17] and 2.1–3.4% in a European cohort [18], and 3.1% in an American cohort [9], whereas in this study, it was much higher at 14.1%. This was expected because the study included very elderly patients. The 1-year overall rate of composite events was 37.7%, which was higher than the 23.7% reported previously [19] and was significantly higher for HFrEF than for HFpEF. Differences in LVEF varied, with some reports showing that HFpEF was as poor as HFrEF, whereas others showed significantly worse results for HFrEF, as observed in the present study. The 1-year incidence of all-cause death varied from 7.3–21.8% in previous Japanese reports [5, 20], but tended to be higher in this study (19.7% overall). The trend toward worse outcomes (composite events and all-cause death) in the present study was likely due, in large part, to age. In contrast, the event rate for composite events was higher in the HFpEF group than in the HFrEF group, despite the older age of the patients. This may be related to the lower rate of introduction of cardioprotective agents such as ACEI/ARB, BB, and MRA. The efficacy of these agents in very elderly patients with HF, who have not been included in large clinical studies, needs to be prospectively investigated.
Limitations
Our study has several limitations. First, we collected the LVEF data during index hospitalization; however, there was no information on the exact timing of echocardiography during hospitalization, which may have been temporarily affected by co-existing conditions. The presented analysis was based on a single LVEF assessment during hospitalization and was not repeated during follow-up; therefore, we did not observe a change in LVEF. Second, treatment during hospitalization was at the discretion of the attending physician, and the prescriptions recorded were those at the time of discharge and may have been originally taken or subsequently discontinued, increased, or added. Third, during the period covered in this study, the new HF drugs angiotensin receptor neprilysin inhibitors (ARNI) and SGLT2I had not yet been indicated for the treatment of HF; therefore, the new HF drugs were not introduced, and their introduction rates were very low. Fourth, while some studies have reported worse prognoses for frail patients about HF treatment and frailty, others have suggested that patients with frailty may benefit more from HF medications [21]. Although this study included a very elderly population, which would be expected to have more frail patients, we were unable to collect sufficient measures of frailty and thus could not address this relationship. Fifth, HF with supernormal EF has been reported to have a poor prognosis, and care should be taken in interpreting the results of this study, since the prognosis may vary depending on whether the patient is classified as supernormal or not, even among patients classified as HFpEF [22]. Sixth, nursing home use may influence the event rate of HF readmission, and, information on social factors, such as information on caregiving, is needed [23]. The possibility that potential confounding factors may influence prognosis cannot be ruled out, so caution must be exercised in drawing the conclusion that HFpEF has a good prognosis based on the results of this study. Seventh, to examine the characteristics of HF patients in this region, we included patients with acute coronary syndrome if they met the criteria, but these patients should have been excluded to more accurately assess the association between EF and outcome. Finally, the number of patients analyzed was too small to draw a definitive conclusion, and this was an observational study conducted at a single institution.