Impact of Frailty Syndrome on Early and Midterm Adverse Events in Patients With Acute Coronary Syndrome

Introduction

With an population aging, the proportion of older adults in the group of patients with acute coronary syndromes (ACS) is on the rise, and estimates show that 1 in every 3 patients with ACS is aged 75 years or over [1]. Older age is associated with the growing burden of comorbidities, disabilities, and frailty. Several definitions of frailty have been proposed, and, according to the most popular definitions, it is a syndrome of decreased capability to cope with stressors and increased vulnerability to adverse events, resulting from a decline of the biological function of an individual [2–4]. Frailty is common in patients with cardiovascular disease: in patients over 60 years of age, frailty was found in 30% with coronary artery disease, 74% with aortic stenosis, and 80% with heart failure [5]. ACS still constitutes one of the major causes of cardiovascular morbidity and mortality, and the rate of in-hospital and after-discharge adverse events related to ACS remains high [6]. While the proportion of patients with ACS who are frail is increasing, the extent of the direct impact of frailty on outcomes in these patients has not been fully elucidated. The 2023 European Society of Cardiology guidelines on ACS management in general recommend application of the same diagnostic and therapeutic tools in older adult patients as in younger patients [7]. However, patients with frailty are believed to be at increased risk of major adverse cardiovascular and cerebrovascular events (MACCEs) and to be more susceptible to complications of invasive procedures. To address existing evidence gaps on the association between frailty and ACS outcomes beyond chronological age, we conducted this prospective, observational study investigating the impact of frailty on early and midterm adverse events in elderly patients with ACS.

Multiple tools have been developed for frailty assessment, some based on the “phenotype” concept, others on the “cumulative deficit” concept [2,8,9]. The novel aspect of our study was the use of the Tilburg Frailty Indicator (TFI), which is a valuable tool for frailty assessment in patients with ACS. First, it is a multidimensional instrument that offers holistic evaluation of frailty status of an individual, integrating the assessment of 3 domains: physical, psychological, and social. This type of holistic approach is recommended by the World Health Organization [10]. Second, the TFI has a Polish validation and is a questionnaire-based score, which is of great importance in acute settings of ACS, when patients cannot perform physical activity [11–13].

Material and Methods

STATISTICAL METHODS:

Statistical analysis of the results was performed using STATISTICA v. 13.3 (TIBCO Software Inc, Palo Alto, CA, USA) and R for Windows version 4.4.3 (R Core Team, Vienna, Austria).

The Shapiro-Wilk test was used to assess the conformity of empirical distributions of continuous quantitative variables, such as age and length of hospitalization, and discrete variables, such as TFI, with theoretical normal distributions. For quantitative variables, depending on their distribution, mean (M) and standard deviation (SD) values were calculated, or medians (Me) and lower (Q1) and upper (Q3) quartiles.

For qualitative variables, counts (n) and percentages (%) were calculated and presented in contingency tables. Hypotheses regarding the independence of qualitative features in the 2 groups were verified using the Fisher exact test. The significance of differences in mean values between the 2 groups for variables with a near-normal distribution and homogeneous variances was checked using the t test. The homogeneity of variances was checked using the Levene test.

Cut-off values for continuous variables delineating 2 states, such as the presence or absence of frailty syndrome, were determined based on receiver operating characteristic (ROC) curve analysis and the Youden index. For each prognostic parameter, the area under the ROC curve was estimated, and for the proposed cut-off value, the sensitivity, specificity, and accuracy of the test were calculated.

Logistic regression was used to assess the influence of quantitative predictors (continuous and discrete) and qualitative predictors (nominal and ordinal) on a binary dependent variable. Model coefficients were estimated using the maximum likelihood method. Preliminary selection of qualitative variables for the model was performed based on the results of contingency table analysis. The significance of the association between variables and the dependent variable was verified using the chi-square test (Pearson test and maximum likelihood). To assess the strength and direction of the association, appropriate odds ratios along with confidence intervals were calculated. Variables with a P value ≤0.25 in univariable analyses were included in the multivariable logistic regression model. For continuous predictors, the assumption of a linear relationship between each predictor and the logit of the outcome was assessed. The necessity of a Box-Tidwell transformation for continuous variables included in the model was assessed using the boxTidwell function implemented in the CAR package of R. The goodness of fit of the model to the data was assessed using the Hosmer-Lemeshow test. In all statistical tests used, a significance level of α=0.05 was adopted. A test result was considered significant when the 2-sided probability was P<0.05.

Results

IN-HOSPITAL COMPLICATIONS:

Table 2 presents the occurrence of in-hospital complications, which were defined as follows: major bleeding according to the TIMI scale, ventricular arrhythmias (sustained and nonsustained ventricular tachycardia), atrioventricular conduction disturbances requiring pacing, sudden cardiac arrest, early stent thrombosis, acute heart failure (Killip Kimball III or IV), stroke, prolonged hospitalization, and in-hospital death. The overall in-hospital complication rate was higher in the frailty group than in the non-frailty group (61.3% vs 25.4%; P<0.001). Of all analyzed in-hospital complications, PLOS was observed more frequently in the frailty group than in the non-frailty group (54% vs 16.9%; P<0.001), while no significant differences were found in the occurrence of other in-hospital complications. Patients with frailty had almost 5 times the odds of in-hospital complications and 6 times the odds of PLOS, compared with patients without frailty. Univariate and multivariate regression analysis demonstrated left ventricular ejection fraction (LVEF) less than 38% and TFI score of 5 points or higher as independent predictors of in-hospital adverse events (Table 3).

MACCES AT 6 MONTHS:

MACCEs analyzed at 6 months following the index hospitalization for ACS included myocardial infarction, urgent re-PCI, stroke, and all-cause death. There was a trend toward increased MACCEs at 6 months in the frailty group vs non-frailty group, although it did not reach statistical significance (15.3% vs 5.1%; P=0.056) (Table 2). However, the analysis regarding the severity of the frailty syndrome revealed a significantly increased rate of MACCEs at 6 months in patients with moderate/severe frailty than in patients without frailty (22.2% vs 5.1%; P=0.019) (Table 4). What is noteworthy is that the risk of all-cause death at 6 months was increased in the frailty group, with frailty associated with 7 times the odds of all-cause death (10.9% vs 1.7%; OR 7.12; P=0.03).

Univariate analysis demonstrated that age 75 years and older, LVEF 35% or lower, and TFI of 8 points or more were associated with MACCEs at 6 months; however, in the multivariate analysis only, LVEF of 35% or lower and TFI of 8 points or more were independent predictors of MACCEs (Table 3).

Discussion

STRENGTHS AND LIMITATIONS:

The strengths of this study include that it was a prospective, observational study. Frailty scale was precisely assessed in each individual patient by trained medical professionals with the use of a validated score. TFI is a score based on a questionnaire, with no physical activity of an examined individual required, which is important in the acute settings of ACS hospitalization.

Limitations include that the study was a single-center study with a limited number of participants. The maximal follow-up time was 6 months. It should be noted that several predictors, including TFI of 8 or more and LVEF of 35% or lower, exhibited relatively wide 95% confidence intervals. This indicates a lack of precision in the point estimates and a degree of instability in the results, likely due to small sample sizes or a limited number of events in certain subgroups (eg, in the mortality analyses). Therefore, these findings should be interpreted with caution.

Conclusions

Overall frailty is associated with worse in-hospital outcomes and moderate/severe frailty, with an increased rate of MACCEs 6 months following the index ACS. The impact of frailty on in-hospital and midterm outcomes in patients with ACS is independent of chronological age. The assessment of frailty in older adult patients hospitalized with ACS provides incremental information for risk stratification. Routine evaluation of frailty using validated instruments may improve risk stratification and should be considered in ACS care pathways.