Multimorbidity, polypharmacy, and mortality in older patients with pacemakers

Abstract Background The prevalence of multimorbidity and polypharmacy and its association with all‐cause mortality in older patients with pacemakers are largely unknown. We aimed to clarify the prevalence of multimorbidity and polypharmacy, and its association with all‐cause mortality in patients ≥75 years of age with pacemakers. Methods We retrospectively investigated 256 patients aged ≥75 years (mean age 84.0 ± 5.3 years; 45.7% male) with newly implanted pacemakers. The study endpoint was all‐cause mortality (“with events”). Multimorbidity was defined as a Charlson Comorbidity Index ≥3. Polypharmacy was defined as the use of ≥5 medications. Results During the follow‐up period (median, 3.1 years), 60 all‐cause deaths were reported. The Charlson Comorbidity Index (2.9 ± 1.9 vs. 1.7 ± 1.7, p < .001) and prevalence of multimorbidity (56.7% vs. 26.0%, p < .001) were significantly higher in deceased patients than in survivors. The number of drugs (6.9 ± 3.0 vs. 5.9 ± 3.3, p = .03) and the prevalence of polypharmacy (78.3% vs. 63.8%, p = .04) were significantly higher in patients with events than in those without events. The event‐free survival rate was significantly higher among patients without multimorbidity than in those with multimorbidity (log‐rank, p < .001), and was also significantly higher among patients without polypharmacy than in those with polypharmacy (log‐rank, p < .001). Multimorbidity (hazard ratio [HR]: 3.21; 95% confidence interval [CI]: 1.85–5.58; p < .001) and polypharmacy (HR: 1.97; 95% CI: 1.03–3.77; p = .04) were independent predictors of all‐cause mortality. Conclusions Multimorbidity and its associated polypharmacy, which are common in the older population, are prevalent in patients with pacemakers and are independent predictors of poor prognosis.


| INTRODUC TI ON
With the increased number of older individuals in the Japanese population, patients with multimorbidity, defined as individuals living with two or more chronic health conditions, are often encountered. 1 Multimorbidity is associated with decreased quality of life, impaired functional status, reduced physical and mental health, and increased mortality. 2 Furthermore, multimorbidity is associated with a high treatment burden, polypharmacy, and considerably greater health service usage, including emergency hospital admission and a corresponding increase in medical expenditures. 3 Simultaneously, patients who require a pacemaker (PM) implant are more often older than previously observed. 4,5 Therefore, multimorbidity is a major challenge for older patients with PMs. However, data on multimorbidity in older patients with PMs are scarce. Moreover, polypharmacy likely leads to worse outcomes. 6 Nevertheless, data on polypharmacy in such patients are limited. In addition, aging leads to a decline in activities of daily living (ADL). Accordingly, this study aimed to clarify the prevalence of multimorbidity, polypharmacy, and dependent ADL, and to evaluate its association with all-cause mortality in patients with PMs aged ≥75 years.

| Study population
This was a single-center, retrospective, observational study. A total of 374 consecutive patients with newly implanted PMs for standard pacing indications were investigated at Nagoya City University Hospital between January 2010 and June 2020. We collected baseline characteristics, comprehensive echocardiographic indices, laboratory data, PM parameters (including pacing mode: physiological or ventricular pacing), underlying diseases, and the number of drugs taken. All data were extracted from the patients' medical records.
Blood samples, echocardiographic indices, underlying disease, data regarding the number of drugs taken, and Barthel index (BI) 7 were collected at discharge. The individual percentages of right ventricular pacing were also obtained from the medical records as the mean value for each individual during the 3-month period prior to the end of the study. The study endpoint was all-cause mortality ("with events"). Patients aged ≥75 years were eligible for inclusion if they had been followed up for at least 6 months. Patients aged <75 years (n = 109) and those who had been followed up for less than 6 months (n = 9) were excluded. The study protocol was approved by the Nagoya City University Graduate School of Medical Sciences and Nagoya City University Hospital Institutional Review Board (reference no. 60-20-0210) and was carried out in accordance with the principles of the Declaration of Helsinki. The requirement for written informed consent was waived by the Nagoya City University Graduate School of Medicine and the University Hospital Institutional Review Board due to the retrospective nature of the study.

| Multimorbidity, polypharmacy, and BI
The Charlson Comorbidity Index (CCI), 8 which predicts mortality by classifying or weighting comorbidities, was measured, with the severity of comorbidities categorized as follows: mild, CCI scores of 1-2; moderate, scores of 3-4; and severe, scores ≥5. In this study, multimorbidity was defined as a CCI score of ≥3. The underlying disease definitions were derived from the CCI criteria. All regular daily medications at discharge were counted. Polypharmacy was defined as the daily use of ≥5 regular medications. The BI consists of 10 items: feeding, bathing, grooming, dressing, bowel control, bladder control, toileting, chair transfer, ambulation, and stair climbing. This index is widely used as an indicator of ADL in routine clinical practice in geriatric medicine (BI range, 0-100; BI ≥85 indicates independent ADL). 7

| Statistical analysis
The SPSS statistical software (version 23.0, SPSS Inc.) was used for all statistical analyses. Continuous variables are presented as means ± standard deviations for normally distributed variables, and as medians and interquartile ranges (IQR) for non-normally distributed variables. Categorical variables are summarized as frequencies (%). For the comparison of two groups, continuous variables were compared using unpaired Student's t-tests for normally distributed variables and Mann-Whitney U tests for non-normally distributed variables. Differences in prevalence between the groups were compared using the Chi-square test. For event-free survival analysis, Kaplan-Meier curves were obtained and compared using log-rank tests. We calculated hazard ratios (HRs) derived from the multivariate Cox regression analysis to identify predictors of all-cause mortality. We used significant variables that were identified in univariate Cox regression models with a p-value of <.05; these included the following: age, atrioventricular block (AVB) and sick sinus syndrome (SSS) as indications for PM implantation, multimorbidity, polypharmacy, and dependent ADL (BI <85). As there were too many individual factors related to underlying diseases and medications, and also because this study focused on the association between multimorbidity and the number of drugs and prognosis, the CCI and polypharmacy were used as representative variables. A p-value <.05 was considered statistically significant. Cardiovascular deaths, including heart failure (n = 8), sudden death (n = 5), and cerebral infarction (n = 2), were observed in 15 patients. 12.8%, p = .046) ( Table 1). Figure 1 shows the distribution of the    F I G U R E 1 Distribution of (A) the Charlson Comorbidity Index (CCI), (B) the number of drugs, and a Barthel index value of 85 as a cutoff between patients who died of any cause ("with events") and those who survived. The proportion of patients who died was higher among patients with multimorbidity, with polypharmacy, and a Barthel index of <85

| Baseline characteristics
proportion of patients with events was >35% of those with two or more of the three factors (multimorbidity, polypharmacy, and dependent ADL) and was 4.8% of those with none of the three. Among those who died, the proportion of patients with dependent ADL alone (33.3%) was higher than that for those with multimorbidity alone (22.2%) or polypharmacy alone (17.6%). Patients with multimorbidity were more likely to take multiple medications (83.5%; 71 out of 85 patients). Furthermore, even among those without multimorbidity, more than half of the patients received polypharmacy (57.9%; 99 of 171 patients).

| Clinical outcome
The endpoint-free survival rate was significantly higher in patients without multimorbidity (log-rank, p < .001, Figure 3A), without polypharmacy (log-rank, p = .004, Figure 3B), and with independent ADL (log-rank, p < .001, Figure 3C) than in those with multimorbidity, polypharmacy, and dependent ADL.  Among patients with multimorbidity, polypharmacy, and dependent ADL, the proportion of patients with events was more than 35% among those with two or more of the three, and 4.8% among those with none of the three. An unexpectedly high number of patients with polypharmacy were present, including in the group without multimorbidity or Alzheimer's disease, rheumatic or connective tissue disease, and heart failure) overlapped with CCI components. Five diseases (diabetes, hypertension, malignancy, stroke, and dementia) were the most frequent underlying diseases in this study. Therefore, we believe that the use of the CCI is appropriate. The prevalence of multimorbidity, defined as a CCI ≥3, was 33.2% in our study. A previous study reported that approximately 20% of patients over 75 years of age had a CCI ≥3. 13 The differences observed are a result of the targeted cohorts; our study focused on patients with PMs, whereas the previous study focused on the general population. Since multimorbidity reflects an aging society, an extension of healthy life expectancy by improved lifestyles and/or social environment may help palliate this problem.
A previous study reported that cancer and vascular deaths accounted for approximately 80% of deaths in patients with multimorbidity. 14 In the present study, approximately 50% of all deaths were caused by these two disease groups, while approximately 30% of deaths were due to senility. This difference results from a difference in cohorts; our study targeted much older patients than the aforementioned review. Therefore, the results of our study are consistent with the finding that cancer and cardiovascular deaths are the most common causes of death in patients with multimorbidity. Regarding the underlying diseases, the prevalence of prior heart failure was significantly higher in deceased patients than in survivors, and heart failure was one of the major causes of death in this study. However, although patients frequently died of malignant diseases, the prevalence of underlying malignant disease did not differ between the groups. Therefore, these underlying diseases did not directly lead to death. The so-called "competing causes of death" were also observed in older patients with PMs.
This study showed that polypharmacy was common among older patients with PMs, with or without multimorbidity. One of the reasons for polypharmacy in patients without multimorbidity was that a single patient had multiple prescriptions for a single disease.
Multiple medications were required for a single patient with diabetes, depression, and urological diseases in this study. Furthermore, multiple prescriptions were given to a single patient for diseases, such as constipation and vertigo. It has been suggested that polypharmacy has several adverse effects, including increased incidence of falls, 15 frailty, 16 fractures, 17 renal dysfunction, 18 and hospitalization. 19 Polypharmacy likely leads to worse outcomes, given the higher likelihood of drug-drug interactions and reduced treatment adherence. 6 In particular, polypharmacy in patients with AF aged >75 years was associated with adverse outcomes. 20 However, the exact definition of polypharmacy has not yet been established. A systematic review reported that the most common definition of polypharmacy was the concurrent use of five or more drugs, 21 which is consistent with that used in our study. A previous study using this definition reported that the prevalence of polypharmacy was >15% F I G U R E 3 Kaplan-Meier curves for all-cause mortality in patients with or without multimorbidity (A), with or without polypharmacy (B), and with dependent or independent ADL. The endpoint-free survival rate was significantly higher in patients without multimorbidity and polypharmacy and with independent ADL than in those with multimorbidity, polypharmacy, and dependent ADL in the general population of the United States. 22  This study has some limitations. First, this was a single-center retrospective observational study that included a limited number of patients. Second, the numbers of diseases and medications were evaluated only at discharge. Since the number of diseases and drugs could change during the follow-up period, our study could not clarify the impact of these changes on patients' prognosis. Despite these limitations, we believe that the results of our study are meaningful. Finally, this study focused on whether three factors, multimorbidity, polypharmacy, and dependent ADL, affect the prognosis in patients with a PM. Therefore, no conclusions could be drawn regarding the impact of improvement of ADL by PM implantation on the prognosis. A further study is therefore needed to examine this issue.

| CON CLUS IONS
Multimorbidity and its associated polypharmacy, which are common in older populations, are also prevalent in patients with PMs.
Polypharmacy is commonly observed regardless of disease prevalence. Additionally, multimorbidity, polypharmacy, and dependent ADL were found to be independent predictors of poor prognosis, including among older patients with PMs.

ACK N OWLED G M ENTS
We acknowledge the contribution of Satoshi Osaga, Clinical Research Management Center, Nagoya City University Hospital, in the statistical analysis of this study.

CO N FLI C T S O F I NTE R E S T
Authors declare no conflict of interest for this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
All the relevant data are within the manuscript.