Frailty is a geriatric syndrome characterized by functional impairments and is associated with poor outcomes; however, the prevalence of frailty and its association with health status in patients treated with percutaneous coronary intervention (PCI) are unknown. To assess the prevalence of frailty and its association with health status in PCI-treated patients, we studied 629 patients ≥65 years old undergoing PCI from October 2005 through September 2008. Frailty was characterized using the Fried criteria: weight loss >10 lbs. in previous 1 year, exhaustion, low physical activity, poor gait speed, and poor grip strength (3 features = frail; 1 feature to 2 features = intermediate frailty; 0 feature = not frail). Health status was assessed using the Short-Form 36 and the Seattle Angina Questionnaire (SAQ). Multivariable linear regression models were used to estimate the independent association between frailty and health status. Complete data on 545 patients demonstrated that 19% (n = 117) were frail, 47% (n = 298) had intermediate frailty, and 21% (n = 130) were not frail. Frail patients had more co-morbidities and more frequent left main coronary artery or multivessel disease after adjusting for age and gender (p <0.05 across groups). Multivariable linear regression demonstrated poorer health status in frail patients compared to nonfrail patients as evidenced by lower Short-Form 36 scores, lower SAQ scores for physical limitation, and lower SAQ scores for quality of life (p <0.001 for each health status domain). In conclusion, 1/5 of older patients are frail at the time of PCI and have greater comorbid burden, angiographic disease severity, and poorer health status than nonfrail adults.
Although optimizing patients’ health status (their symptoms, function, and quality of life) are important therapeutic goals for all patients, they are particularly relevant to older adults. Patients’ health status is not only an important outcome in its own right but also has been shown to be associated with clinical events in patients with coronary artery disease (CAD) and quality of life after percutaneous coronary intervention (PCI). The association between health status and age-associated impairments such as frailty are unknown but are needed to better identify opportunities to improve care for the aging population of patients with CAD. Frailty is an emerging concept that is characterized by decreased resilience to stressors and increased physiologic vulnerability. Fried et al defined frailty as a clinical syndrome with progressive decreases in reserves and physical function, operationalized its assessment, and reported its association with adverse outcomes. Although traditional cardiovascular co-morbidities in patients with CAD have been studied, prevalence and correlates of geriatric syndromes such as frailty are less understood. With the availability of standardized definitions and valid instruments, the aims of this study were to assess the prevalence of frailty in PCI-treated older patients and to describe baseline health status measurements of frail patients compared to those without frailty. In addition, we sought to understand the incremental association of frailty to baseline quality of life and disease-specific quality of life after accounting for demographics, co-morbidities, and angiographic severity of patients’ CAD.
Methods
Patients ≥65 years of age undergoing PCI at the Mayo Clinic in Rochester, Minnesota and Franciscan Skemp Hospital in LaCrosse, Wisconsin from October 2005 to September 2008 and who survived to hospital discharge were prospectively enrolled in a study assessing frailty and health status. A cross-sectional study design was used to administer standardized health status questionnaires and to perform functional assessments of frailty and abstraction of clinical co-morbidities from medical records. Patients with residual neurologic deficits after a stroke, severe Parkinson disease, or dementia were excluded from the study because of the concern regarding the reliability and validity of functional assessments in these patients. The study was approved by the institutional review board of the Mayo Clinic and participants provided written informed consent.
We used the definition of frailty by Fried et al and Walston and Fried in which musculoskeletal, neuroendocrine, and nutritional defects are considered in the determination of frailty. Five criteria were measured in each study participant including unintended weight loss (>10 lbs. in preceding year), exhaustion, physical activity, gait speed (walking time >15 feet), and grip strength. Exhaustion was measured by the Center of Epidemiologic Studies depression subscale. The 2 items in this scale are how often in the previous week the patient believed that (1) “I felt that everything I did was an effort” and (2) “I could not get going.” Subjects who answered “a moderate amount of time (3 to 4 days)” or “most of the time” to either statement were categorized as meeting the exhaustion criteria for frailty. Physical activity was measured by the short version of the Minnesota Leisure Time Activity questionnaire. Gait speed was assessed as walking time across a 15-foot distance in an unobstructed well-lit hallway. Participants walking with an assist device were permitted to do so for the test. Handgrip strength was tested in the subject’s dominant hand using a Jamar handgrip dynamometer (Jamar, Bolingbrook, Illinois). Participants were told to squeeze as hard as possible and results were scored as kilograms. The details for the thresholds in the 5 frailty criteria used are provided ( Supplementary Appendix ). The assessment for frailty was performed in patients after their PCI procedure. Patients were defined as frail if they were found to have deficits in ≥3 core elements and as intermediately frail if they had deficits in 1 to 2 core elements. Patients without any deficits in the core frailty measurements were classified as nonfrail (i.e., normal).
Co-morbidity was assessed using 2 separate instruments. The Charlson index, a well-validated measurement of co-morbidity burden, is based on 12 long-term conditions and its corresponding weights were originally determined based on their association with 1-year mortality. We used the same procedure to derive a score for the Charlson index in our study participants. The CAD-specific index has been compared to the Charlson index and has been shown to perform similarly; however, the CAD-specific index incorporates CAD severity and left ventricular ejection fraction in addition to baseline clinical co-morbidities. Higher values of these indexes indicate a greater burden of co-morbidities.
Health status assessments were performed after the PCI procedure through administration of the Short-Form 36 (SF-36) for physical component and mental component scores and the disease-specific Seattle Angina Questionnaire (SAQ), instruments that have been shown to be valid and reproducible self-report tools for ascertaining health-related quality of life. The SF-36 is a generic assessment of patients’ health status using an 8-scale profile of functional health and well-being that form 2 distinct domains: physical component score and mental component score. A score of 50 reflects the United States population mean and each 10 points represents 1 SD from that mean. Higher scores indicate a better health status. The SAQ quantifies 5 clinically relevant dimensions of CAD: physical limitation, angina stability, angina frequency, treatment satisfaction, and quality of life. Scales range from 0 to 100, where higher scores indicate better functioning, fewer anginal symptoms, and better quality of life. A 5-point difference in mean SAQ scores is considered clinically significant.
Univariate continuous distributions are summarized as median and interquartile range unless otherwise specified. Comparisons between groups were tested using linear models with a linear contrast of group means, thus testing for a mean trend from the no-frailty to the frail group. Age and gender were added to the linear model to obtain age- and gender-adjusted p values. Discrete data including 30-day outcomes of major cardiovascular events are summarized as frequency (percentage) and were tested across frailty groups with the Armitage trend test. Age- and gender-adjusted p values for these comparisons were obtained by modeling the binomial variable as a function of frailty (coded as consecutive integers for the 3 groups) and age and gender. Partial Spearman correlation coefficients adjusted for age and gender were computed to measure the association among frailty, quality of life, and co-morbidity measurements.
Multivariable linear regression was used to assess the association among patient demographics, baseline characteristics, frailty, and quality of life measurements. Five multiple imputation datasets were created using the aregImpute function in S-Plus (available from the Hmisc library by Frank Harrell) to allow all observations to be included in the analysis. Regression models were fit within each imputed dataset; the 5 sets of estimates were then combined according to the rule of Rubin. Box–Cox transformations were created for non-normally distributed variables. However, the results from models with transformed data provided similar conclusions as those without transformed data; for ease of interpretation, results from untransformed data are presented. Partial residual plots were used to assess assumptions of linear associations for continuous covariates; no evidence of nonlinearity was seen in these plots. Interactions among site, CAD-specific index, and frailty measurements were tested.
Results
We screened 1,885 patients of whom 629 (33.4%) consented to participate. Slightly more men (69% vs 63%) and younger patients (74.3 ± 6.4 vs 75.8 ± 6.9 years) consented compared to nonconsenting patients. Mean age of participants was 74.8 ± 6.4 years. Of participants, 117 (18.6%) were frail, 298 (47.4%) had intermediate frailty, and 130 (20.6%) were not frail. Frailty status could not be classified in 84 patients (13.3%) because of incomplete or incorrectly completed forms. The most common frailty measurement was slow gait speed (41%) and the least common was unintended weight loss of >10 lbs. in the preceding year (7.5%).
Characteristics of the study population are listed in Table 1 . Patients with features of frailty tended to be older, had a higher body mass index, were more likely to be women, and were more likely to have more co-morbidities including hypertension, diabetes mellitus, chronic kidney disease, peripheral arterial disease, congestive heart failure, atrial fibrillation, stroke or transient ischemic attack, and myocardial infarction (adjusted p <0.05 for comparison across groups). Table 2 presents procedural and angiographic characteristics of the study cohort. There was a higher frequency of multivessel or left main coronary artery disease in frail patients (74%) compared to intermediately frail (68%) or nonfrail (60%) patients (p = 0.019) and frailty was found to be associated with the presence of multivessel or left main coronary artery disease after adjusting for age and gender (p = 0.005). Table 3 presents baseline health-related quality of life stratified by frailty status in patients undergoing PCI. Mean SAQ physical limitation score was significantly lower in frail patients compared to intermediately frail patients and nonfrail patients, indicating worse disease-specific physical limitations in patients with frailty. Similarly, mean SAQ quality-of-life score was significantly lower in frail patients compared to intermediately frail patients and nonfrail patients, suggesting that patients with the frailty phenotype had greater associated quality-of-life impairments from angina. SAQ scores for angina frequency and treatment satisfaction were similar across frailty strata. SF-36 scores for the physical component and mental component score domains were lowest in frail patients compared to intermediately frail and nonfrail patients, suggesting that older adults with frailty undergoing PCI have lower baseline health-related quality of life compared to nonfrail patients. Table 4 lists 30-day outcomes for major cardiovascular events (death, myocardial infarction, and revascularization) by frailty status. Event rates for short-term clinical outcomes were generally low for the cohort overall, with approximately 5% developing myocardial infarction 30 days after PCI within each frailty stratum. After adjusting for age and gender, there were no statistically significant differences in any of the clinical events by frailty category.
| Characteristics | Not Frail | Intermediate Frailty | Frail | p Value | Adjusted p Value ⁎ |
|---|---|---|---|---|---|
| (n = 130) | (n = 298) | (n = 117) | |||
| Age (years), mean ± SD | 72.6 ± 5.8 | 74.6 ± 6.0 | 77.4 ± 6.8 | <0.001 | |
| Men | 106 (82%) | 205 (69%) | 65 (56%) | <0.001 | |
| Body mass index (kg/m 2 ) | 29.0 ± 4.2 | 30.1 ± 5.4 | 30.7 ± 6.6 | 0.014 | <0.001 |
| Hypertension | 91 (72%) | 244 (82%) | 103 (88%) | 0.001 | 0.008 |
| Current smoker | 4 (3%) | 18 (6%) | 7 (6%) | 0.30 | 0.047 |
| Diabetes mellitus | 20 (16%) | 93 (31%) | 47 (40%) | <0.001 | <0.001 |
| Chronic kidney disease | 5 (4%) | 39 (13%) | 28 (25%) | <0.001 | <0.001 |
| Peripheral arterial disease | 7 (6%) | 29 (10%) | 25 (22%) | <0.001 | <0.001 |
| Congestive heart failure | 14 (11%) | 47 (16%) | 32 (28%) | <0.001 | 0.013 |
| Atrial fibrillation | 9 (7%) | 41 (14%) | 30 (26%) | <0.001 | <0.001 |
| Cerebrovascular accident or transient ischemic attack | 11 (9%) | 41 (14%) | 23 (20%) | 0.011 | 0.004 |
| Vascular surgery | 11 (9%) | 42 (14%) | 25 (22%) | 0.003 | 0.005 |
| Myocardial infarction | 29 (23%) | 94 (32%) | 41 (38%) | 0.013 | 0.043 |
| Percutaneous coronary intervention | 38 (29%) | 117 (39%) | 41 (35%) | 0.31 | 0.28 |
| Coronary artery bypass grafting | 21 (16%) | 79 (27%) | 39 (33%) | 0.002 | 0.012 |
| Chronic lung disease or obstructive pulmonary disease | 9 (7%) | 35 (12%) | 31 (26%) | <0.001 | <0.001 |
| Rheumatologic disease | 45 (35%) | 135 (46%) | 68 (58%) | <0.001 | 0.005 |
| Any neoplasm | 32 (25%) | 64 (22%) | 25 (22%) | 0.53 | 0.44 |
| Metastatic solid neoplasm | 13 (10%) | 39 (13%) | 14 (12%) | 0.68 | 0.89 |
| Lymphoma | 0 (0%) | 2 (1%) | 2 (2%) | 0.12 | 0.22 |
| Unexpected fall within 6 months | 5 (4%) | 46 (15%) | 27 (23%) | <0.001 | <0.001 |
| Fracture | 42 (33%) | 105 (36%) | 36 (31%) | 0.76 | 0.84 |
| Sleep apnea | 16 (13%) | 50 (17%) | 27 (24%) | 0.022 | 0.002 |
| Charlson index, median (quartile 1, quartile3) | 2.0 (1.0, 3.0) | 2.0 (1.0, 4.0) | 3.0 (2.0, 5.0) | <0.001 | <0.001 |
| Coronary artery disease–specific index, median (quartile 1, quartile 3) | 1.0 (1.0, 3.0) | 3.0 (1.0, 4.0) | 3.0 (2.0, 6.0) | <0.001 | <0.001 |
| Characteristics | Not Frail | Intermediate Frailty | Frail | p Value | Adjusted p Value ⁎ |
|---|---|---|---|---|---|
| (n = 130) | (n = 298) | (n = 117) | |||
| Urgency of percutaneous coronary intervention | 0.26 | 0.41 | |||
| Elective | 53 (41%) | 99 (33%) | 35 (30%) | ||
| Urgent | 54 (42%) | 137 (46%) | 62 (53%) | ||
| Emergency | 23 (18%) | 60 (20%) | 20 (17%) | ||
| Myocardial infarction before percutaneous coronary intervention | 0.65 | 0.99 | |||
| No myocardial infarction within 24 hours | 93 (73%) | 193 (66%) | 80 (69%) | ||
| Non–ST-segment elevation myocardial infarction | 18 (14%) | 59 (20%) | 20 (17%) | ||
| ST-segment elevation myocardial infarction | 17 (13%) | 42 (14%) | 16 (14%) | ||
| Multivessel or left main disease | 74 (60%) | 187 (68%) | 82 (74%) | 0.019 | 0.005 |
| Total number of stents placed | 1.4 ± 0.8 | 1.4 ± 0.9 | 1.6 ± 1.0 | 0.27 | 0.13 |
| Number of segments treated | 1.4 ± 0.7 | 1.4 ± 0.7 | 1.6 ± 0.8 | 0.034 | 0.011 |
| Number of coronary arteries treated | 0.24 | 0.17 | |||
| 1 | 110 (85%) | 252 (85%) | 93 (79%) | ||
| 2 | 19 (15%) | 42 (14%) | 22 (19%) | ||
| 3 | 1 (1%) | 3 (1%) | 2 (2%) | ||
| Percutaneous coronary intervention in native left anterior descending coronary artery | 69 (53%) | 121 (41%) | 58 (50%) | 0.52 | 0.46 |
| Percutaneous coronary intervention in native right coronary artery | 39 (30%) | 108 (36%) | 38 (32%) | 0.65 | 0.40 |
| Percutaneous coronary intervention in native left main coronary artery | 2 (2%) | 9 (3%) | 11 (9%) | 0.002 | 0.003 |
| Percutaneous coronary intervention in native left circumflex coronary artery | 36 (28%) | 68 (23%) | 31 (26%) | 0.79 | 0.81 |
| Vein graft intervention | 6 (5%) | 35 (12%) | 9 (8%) | 0.36 | 0.97 |
| Use of drug-eluting stent | 111 (85%) | 242 (81%) | 94 (80%) | 0.29 | 0.23 |
| Glycoprotein IIb/IIIa use | 78 (60%) | 182 (61%) | 64 (55%) | 0.42 | 0.99 |
| Blood loss requiring transfusion | 1 (1%) | 8 (3%) | 6 (5%) | 0.037 | 0.088 |
| Pseudoaneurysm | 0 (0%) | 0 (0%) | 1 (1%) | 0.13 | 0.94 |
| Hypotension | 8 (6%) | 24 (8%) | 11 (9%) | 0.34 | 0.35 |
| Femoral bleed | 0 (0%) | 3 (1%) | 1 (1%) | 0.41 | 0.47 |
| Hematoma | 2 (2%) | 4 (1%) | 6 (5%) | 0.063 | 0.32 |
| Gastrointestinal bleed | 1 (1%) | 1 (0%) | 2 (2%) | 0.41 | 0.64 |
| Retroperitoneal bleed | 0 (0%) | 0 (0%) | 1 (1%) | 0.13 | 0.94 |
| Central nervous system bleed | 0 (0%) | 0 (0%) | 0 (0%) | — | — |
| Characteristics | Not Frail | Intermediate Frailty | Frail | p Value for Trend | Adjusted p Value ⁎ |
|---|---|---|---|---|---|
| (n = 130) | (n = 298) | (n = 117) | |||
| Seattle Angina Questionnaire | |||||
| Physical limitation | 77.6 ± 17.8 | 70.2 ± 20.0 | 59.5 ± 24.5 | <0.001 | <0.001 |
| Anginal frequency | 71.3 ± 24.6 | 73.4 ± 24.6 | 65.9 ± 28.1 | 0.10 | 0.080 |
| Treatment satisfaction | 85.6 ± 17.3 | 84.4 ± 17.9 | 82.5 ± 17.2 | 0.16 | 0.17 |
| Quality-of-life scale | 63.0 ± 23.8 | 59.7 ± 23.9 | 55.1 ± 26.3 | 0.013 | 0.013 |
| Center for Epidemiologic Studies depression score | 5.8 ± 5.8 | 10.5 ± 9.1 | 15.4 ± 10.4 | <0.001 | <0.001 |
| Short-Form 36 | |||||
| Physical component score | 42.9 ± 9.6 | 37.0 ± 9.6 | 29.0 ± 8.3 | <0.001 | <0.001 |
| Mental component score | 55.6 ± 7.4 | 52.3 ± 9.7 | 48.8 ± 11.2 | <0.001 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
