Frailty has been found to be a potential predictor of increased adverse outcomes in older cardiac surgical populations [15]. In this study, we analyzed the effects of frailty on older patients who underwent concomitant VR and BRFA. There were four main findings. First, a significant proportion of the study patients had a combination of frailty. Second, frail patients tended to have poorer cardiac functions. Third, during hospitalization, frailty affected the duration of ICU stay and hospital stay for VR with concomitant BRFA procedures. Moreover, frail patients had a higher risk of pulmonary complications. Finally, in terms of late clinical outcomes, frail patients had higher risk-adjusted all-cause mortality and all-cause hospitalization rates.

Frailty is considered a general indicator of patient vulnerability and is highly associated with adverse health outcomes in the field of geriatrics [16], and it is gaining traction in the field of cardiac surgery [17]. With the development and promotion of AF surgical radiofrequency ablation technology in recent years, valve surgery concurrently with AF surgical radiofrequency ablation technology has obvious advantages, and successful conversion to sinus rhythm eliminates the burden of symptoms of AF after surgery for valvular heart disease. The overall sinus rhythm rate at the latest follow-up was 69.4%, which was comparable to that calculated in previous studies. Nakamura et al. reported that the rate of freedom from AF in 143 patients (age 65 years or older) with concomitant surgical ablation was 74.9% [18]. The question of whether to perform BRFA in older patients with valvular disease combined with AF simultaneously has been previously explored in several studies [7,8,9,10], including one of our own [14], and some well-known risk factors, such as age, body mass index, and left atrial size, have been shown to adversely affect patients undergoing heart valve surgery combined with AF ablation [19,20,21]. However, to our knowledge, comparative clinical data on risk stratification using a frailty index in older patients undergoing surgical radiofrequency ablation techniques for concurrent AF are lacking. Surgeons often refuse to perform concomitant BRFA because of the patient's frailty.

Previous studies have shown that age, AF, cardiovascular disease, and other factors are associated with the degree of frailty [3, 22,23,24]. Yang et al. used the CHS frailty phenotype and the Edmonton Frailty Scale to detect frailty in approximately 8.3% of older patients (> 65 years of age) with AF [4], whereas our retrospective observational study showed that frailty was higher than in their study, with 36.2% of older patients hospitalized with AF. This may be because our study patients all had valvular diseases resulting in an increased proportion of frail patients. Frail patients reported worse cardiac function, suggesting that there may be an association between the degree of cardiac function and frailty status. One possible explanation is that cardiac insufficiency may lead to decreased mobility, fatigue, poor appetite, and weight loss, which subsequently leads to weakness in frail patients. In our multivariate logistic regression and Cox proportional survival analyses, we included many potential confounding variables, such as cardiac function class, as covariates. This essentially eliminated the effect of these potential confounders and further validated the effect of frailty status on surgical outcomes in older patients with valve disease combined with AF. We found that frail patients had prolonged hospital stay and postoperative time in the ICU. This is similar to the reports of previous studies that used various frailty criteria [25, 26]. This may be partly attributed to the fact that in our study, postoperative pulmonary complications occurred significantly more in frail patients than in non-frail patients, which significantly prolonged the length of hospital stay and time in the ICU [27].

In older patients, we need to balance the risk of intervention with the improvement of QoL [28]. Previous studies have shown that frailty is an independent predictor of increased intensity of arrhythmia symptoms and worsening QoL [29]. In our study, we used a disease-specific method of QoL assessment, the AFEQT questionnaire, to assess patients with AF. After adjusting for important patient characteristics, adjusted baseline QoL was also found to be significantly lower in frail patients than in non-frail patients. This may be due to the fact that the current assessment of QoL in patients with AF is highly dependent on the patient's symptom status, and frailty may make patients more sensitive to AF symptoms and perceive disease symptoms differently. In addition, analysis of the component domains of AFEQT showed that at baseline, frail patients had more severe symptoms, were more limited in their daily activities, and had more concerns about their treatment. Therefore, when analyzing the impact of radiofrequency ablation on QoL in older patients, clinicians should be aware of these differences and ensure that symptom burden and QoL at baseline are carefully assessed before selecting a treatment strategy. When follow-up QoL was investigated, patients of the frail group continued to have lower overall AFEQT scores with no significant improvement in follow-up compared to those of the non-frail group. Direct comparison of preoperative and postoperative QoL in older patients is limited by the presence of censored data, which requires further study.

Predicting early and late clinical outcomes and risks can help inform treatment decisions. However, current cardiac surgery risk models, including the EuroSCORE II and STS risk models, play an important role in predicting early cardiac surgery outcomes including hospital mortality, but they often fail to directly predict long-term outcomes [30]. These risk models often do not take into account a comprehensive assessment of frailty, which may be an important determinant of the outcome, particularly in older patients. An important implication of correctly identifying frailty is how to provide the appropriate surgical treatment modality for older patients with valvular disease combined with AF. Although our previous study suggested that this technique could provide clinical benefit to older patients, in practice, most surgeons do not consider older patients for simultaneous BRFA procedures.

In our study, frailty was assessed based on the patient's status prior to hospitalization, and we found an increased risk of early postoperative pulmonary complications and in-hospital death in the frail population. In the late clinical outcome of our study, survival was significantly lower in patients in the frail group who were discharged after concomitant BRFA with VR than in those in the non-frail group. Frailty was also identified as a risk factor for all-cause mortality according to univariate and multivariate analyses. Of note, we found it interesting that although the adjusted risk of all-cause mortality and all-cause hospitalization risk for frail patients were 3.06 and 2.36 times higher than those for non-frail patients, respectively, adjusted stroke incidence, cardiovascular-related mortality, and rehospitalization for cardiovascular causes were not significantly different between frail patients and non-frail patients. Therefore, the characteristics of readmission burden, stroke incidence, and cardiovascular-related mortality in frail patients mentioned in this study need to be considered in future evaluations of frail patients undergoing concomitant BRFA with VR in older patients. Our findings suggest that frailty assessment tools may provide valuable information for preoperative communication with patients about their condition, preoperative stratification of risk for late mortality, prediction of major in-hospital adverse events, QoL, and late clinical outcomes.

However, it is also important to highlight that many frail patients underwent surgery without experiencing poor outcomes. Future research should focus on reducing the impact of frailty on clinical outcomes, and teams involved in performing VR and BRFA procedures need to be skilled in managing frail patients. Surgeons should be aware that a significant proportion of older patients who undergo VR and BRFA may be frail and establish with geriatricians the practice of preoperative comprehensive geriatric assessment, postoperative management of frail patients, and improvement in postoperative QoL. With a closer working relationship, specific care models should be developed in the future for this population to reduce the incidence of frailty and improve the clinical outcomes in this patient population. Our finding also suggests that an accurate preoperative frailty assessment may provide a better risk–benefit assessment for each patient, which may have important implications for Medicare and Medicaid costs. Furthermore, there is a need to focus on how treatments and interventions specifically affect all-cause outcomes in frail patients and predict the discharge of frail patients in order to improve discharge planning and coordinate health care costs and resource use. These should be confirmed in further studies.

Our study has limitations that are inherent to retrospective studies. We did not include vulnerable patients who were considered to be "inoperable" by their surgeons and whose frailty may have been quite severe. We cannot extrapolate what the clinical outcome of these patients would have been if they had undergone simultaneous BRFA procedures. Therefore, clinicians should consider factors such as further grading of frailty before interventions in older patients with AF considered too debilitated to review the indications for surgery. Additionally, in clinical practice, there is currently a lack of consensus regarding the best tool to assess frailty. Although the CHS frailty assessment method is easy to administer and perform and is a fairly simple qualitative assessment tool, its inability to indicate the degree of frailty through scoring makes it possible that our findings do not fully reflect the impact of frailty. More patients and a longer follow-up period with testing of other possible definitions of frailty are needed to further confirm the observed effects. In addition, the benefit in patients who did not recover sinus rhythm may not be as significant as in patients who recovered sinus rhythm. In our hospital, older patients over the age of 65 years have relatively rarely chosen mechanical valves. Since thromboembolic events are common after mechanical valve replacement, we excluded patients with mechanical valve replacement to more accurately compare the occurrences of postoperative strokes in the two groups. However, further large-scale studies are needed to address this issue. Nevertheless, our current findings may provide the premise for hypothesis generation and may serve as hypothetical information for future prospective studies to clarify the impact of preoperative frailty status on VR combined with BRFA.