Abstract
Introduction:
Frailty and decreased physical performance are associated with poor outcomes after kidney transplant. Less is known about their relationship with pretransplant outcomes. The aim of this study was to characterize associations between frailty and physical performance with death on the kidney transplant waiting list.
Design:
Since December 2014, high-risk kidney transplant candidates at our center (age > 59, diabetic and/or history of >3 years dialysis) have undergone frailty and physical performance testing using Fried Criteria and the Short Physical Performance Battery.
Results:
Between December 2014 and November 2016, 272 high-risk candidates underwent testing and were approved for transplant. Both frailty and physical performance score were significantly associated with death on the waiting list (hazard ratio [HR]: 6.7, confidence interval [CI]: 1.5-30.1; P = .01; HR: 0.8 per 1-point increase, CI: 0.7-1.0; P = .02, respectively). The relationship between frailty, physical performance score, and death on the waiting list appeared to be independent of age, diabetes, or duration of dialysis.
Discussion:
Frailty and decreased physical performance appear to be independently associated with increased mortality on the kidney transplant waiting list. Further studies are needed to determine whether improving frailty and physical performance prior to transplant can decrease waiting list mortality.
Introduction
The transplant community’s ability to identify kidney transplant candidates at high risk of adverse outcomes is poor. In addition, transplant candidates are becoming older and sicker. 1,2 This combination of insufficient predictive tools with increasingly complex candidates can have dangerous consequences—candidates who would benefit most from transplant can be turned down, while candidates who would have poor outcomes can be approved. Given the scarcity of donor organs for transplantation, objective methods of identifying candidates at high risk of adverse outcomes are needed.
Frailty and physical performance testing are measures that could improve risk stratification in kidney transplant candidates. Frailty and decreased physical performance are promising risk factors to study because they are potentially reversible, unlike many other risk factors that are irreversible. 3 –6 Both frailty and decreased physical performance are associated with significant morbidity and mortality after kidney transplant. For example, frailty and decreased physical performance have been associated with posttransplant delirium, longer hospital length of stay, rehospitalizations, and death. 7 –11 Despite their role in predicting posttransplant outcomes, little is known about the relationship between frailty and decreased physical performance with pretransplant outcomes. 12
The goals of this study were to characterize frailty and physical performance in high-risk kidney transplant candidates at our center and examine the association between frailty and physical performance with death on the kidney transplant waiting list. High-risk candidates had one or more of the following characteristics: age >59, diabetes, and history of >3 years dialysis. Physical performance was measured using the Short Physical Performance Battery (SPPB). The SPPB combines results of balance, chair stand, and gait speed testing and has been shown to predict disability, hospitalizations, and death in nontransplant elderly. 13 –15
Methods
Study Design
Since December 2014, high-risk kidney transplant candidates at Mayo Clinic Rochester have undergone frailty and physical performance testing during their outpatient transplant evaluation as outlined below. High-risk candidates are defined as having one or more of the following characteristics: age >59, diabetic, and history of >3 years dialysis. We retrospectively analyzed candidates evaluated between December 2014 and November 2016 who underwent frailty and physical performance testing and were ultimately approved for kidney transplant at our center. Clinical information was abstracted from electronic databases. Our study was approved by the institutional review board of Mayo Clinic.
Frailty and Physical Performance Testing
Frailty and physical performance testing were conducted during the patient’s annual evaluation by selected physical therapists (B.B., E.G., and S.H.) using infrared photocells and an electronic timing system to achieve a highly precise measurement. 16 Frailty was measured according to Fried Criteria. Patients who had 3 or more of the following characteristics were defined as frail: shrinking (unintentional weight loss), weakness (low grip strength), self-reported exhaustion, low physical activity, and slow walking speed. 17 Patients who did not have 3 or more of these characteristics were defined as not frail. Physical performance was measured using the SPPB that provides a composite measure of balance (tandem, semi-tandem, and side-by-side stands), gait speed (4-meter walk), and chair stand time (time required to perform 5 chair stands). 18 The SPPB takes approximately 10 minutes to complete. Patients receive a score ranging from 0 (unable to perform) to 4 (no difficulty performing) for each of the 3 components. Component scores are then summed to provide a total SPPB score ranging from 0 to 12. 18
Data Analysis
Data were expressed as means and standard deviation or median and range as appropriate. Continuous data were compared by Student t test and nonparametric tests (Kruskall-Wallis) for normally distributed and skewed data, respectively. Proportions were compared with the χ2 test. Survival analysis was performed using Kaplan-Meier and Cox regression. Patient follow-up was censored at the time of kidney transplant. For these analyses, patients were divided according to a SPPB score of <10 or ≥10 based on published literature showing scores <10 to be associated with all-cause mortality across different patient populations. 19 Cardiovascular disease was defined as history of myocardial infarction, coronary artery bypass surgery, or coronary artery stenting.
Results
Study Population
During the study time period, 468 high-risk candidates were evaluated and approved for kidney transplant at our center, of which 274 patients met with the physical therapist. Two of those patients refused to complete the frailty and SPPB testing and were therefore excluded from our analysis. The remaining 272 (58.1%) patients underwent frailty and physical performance testing. The baseline demographics of the 272 patients who completed testing are outlined in Table 1. Compared to candidates who did not complete testing, candidates who were tested were older (61.8 [9.3] vs 57.9 [13.3] years; P = .02), more likely to have diabetes (59.6% vs 40.3%; P < .0001), more likely to have a history of cardiovascular disease (26.8% vs 14.3%; P = .0009), had a higher body mass index (30.2 [5.7] vs 28.9 [5.7] kg/m2; P = .02), and had a shorter history of dialysis (22.4 [27.7] vs 45.1 [54.8] months; P < .0001). Ultimately, 88 (32.4%) of 272 patients received kidney transplants. Mean follow-up time was 12.4 (6.6) months.
Characteristics of Kidney Transplant Candidates.
Frailty and Physical Performance
Among the patients who completed testing, 39 (14.3%) were frail. Median SPPB score was 12 (interquartile range [IQR]: 11-12). Overall, 35 (12.9%) patients had a SPPB score <10 (poor performance). The relationship between frailty and SPPB score <10 is illustrated in Figure 1. Median SPPB score was lower in frail patients compared to nonfrail patients (10 [IQR: 6-12] vs 12 [IQR: 11-12], P < .0001]. Only 18 (6.6%) patients were both frail and had a SPPB score <10. On multivariate analysis, female gender was significantly associated with being frail or having a SPPB score <10 (see Table 2).
Distribution of the Short Physical Performance Battery scores in relationship to frailty.
Variables Associated With Frailty and a Short Physical Performance Battery (SPPB) Score <10.a
Abbreviations: CI, confidence interval; HR, hazard ratio.
aUnivariate analysis was performed for both frailty and SPPB score <10 using the following clinically meaningful variables: age, female gender, dialysis, history of cardiovascular disease, Caucasian race, body mass index. Variables significantly associated with either frailty or SPPB score <10 (P < .05) were included in the multivariate analysis.
Frailty, Physical Performance, and Mortality
After wait listing, 7 (2.6%) of 272 patients died. Estimated overall mortality was 1.5% and 7.1% by 12 and 24 months after wait listing. Frailty was significantly associated with death on the waiting list (hazard ratio [HR]: 6.7, confidence interval [CI]: 1.5-30.1; P = .01). Frailty remained associated with death on the waiting list after individually adjusting for age (HR: 7.1, CI: 1.6-32.4; P = .01), female gender (HR: 11.5, CI: 2.5-53.6; P = .002), diabetes (HR: 6.6, CI: 1.5-29.4; P = .01), history of cardiovascular disease (HR: 6.2, CI: 1.4-28.3; P = .02), preemptive status (HR: 6.7, CI: 1.5-30.7; P = .01), and cardiac troponin T (HR: 5.7, CI: 1.2-27.4; P = .03). Of the frail patients, 4 (10.3%) died on the waiting list. Survival rates of frail patients are described in Figure 2.
Association between frailty and survival on the kidney transplant waiting list.
The SPPB score was also significantly associated with death on the waiting list (HR: 0.8 per 1-point increase, CI: 0.7-1.0; P = .02). The association between SPPB score and death remained significant after individually adjusting for age (HR: 0.73 per 1-point increase, CI: 0.58-0.91; P = .006), female gender (HR: 0.72 per 1-point increase, CI: 0.58-0.89; P = .002), diabetes (HR: 0.82 per 1-point increase, CI: 0.68-0.98; P = .03), history of cardiovascular disease (HR: 0.79 per 1-point increase, CI: 0.65-0.95; P = .01), preemptive status (HR: 0.79 per 1-point increase, CI: 0.65-0.96; P = .02), and cardiac troponin T (HR: 0.86 per 1-point increase, CI: 0.71-1.05; P = .14). Of the 35 patients with a SPPB score <10, 5 (14.3%) died on the waiting list. Survival rates of patients with a SPPB score <10 are displayed in Figure 3.
Association between Short Physical Performance Battery (SPPB) score <10 and survival on the kidney transplant waiting list.
Conclusions
Our study was the first to examine the relationships with both frailty and decreased physical performance and death on the kidney transplant waiting list. In our cohort of 272 high-risk kidney transplant candidates, the prevalence of frailty was 14.3%, the median SPPB score was 12 (IQR: 11-12), and 12.9% had a SPPB score <10. Female gender was significantly associated with both frailty and SPPB <10. Both frailty and SPPB score were significantly associated with death on the waiting list (HR: 6.7, CI: 1.5-30.1; P = .01; HR: 0.8 per 1-point increase, CI: 0.7-1.0; P = .02, respectively). These relationships remained significant after individually adjusting for variables including age, diabetes, history of cardiovascular disease, preemptive status, and cardiac troponin T, a variable previously shown to be associated with death on the kidney transplant waiting list. 20
Frailty has recently been recognized as a risk factor for death on the kidney transplant waiting list. 12 Our study confirms this finding and also describes a relationship between decreased SPPB scores and death on the kidney transplant waiting list. SPPB scores, especially scores <10, are a strong predictor of mortality in multiple different populations, including community-dwellers, inpatients, outpatients, and different age groups. Like frailty, the SPPB is a promising risk predictor because scores likely reflect the combination of multiple factors such as aging, medical comorbidities, and cognition. 19 The SPPB may offer several advantages when compared to frailty, however. Unlike frailty, the SPPB does not involve self-report, a process subject to bias that may be especially pronounced in candidates eager to be approved for transplant. In addition, the SPPB provides a more robust measure of lower extremity function than single tests such as gait speed, which is the only lower extremity measure included in frailty testing. Thus, the SPPB may discriminate patients with poor physical performance and serve as a benchmark for exercise interventions. 19
The strengths of our study include the highly precise measurement of both frailty and physical performance by a small group of dedicated physical therapists who used infrared photocells and an electronic timing system. Limitations include the single-center design of the study and the small number of deaths on the waiting list that limited multivariable analysis. In addition, not all high-risk candidates underwent frailty and SPPB testing. Reasons for high-risk candidates not undergoing testing included limited physical therapy appointment availability and missed referrals. However, the goal of our study was to examine frailty and physical performance in high-risk candidates, and the patients who were tested were indeed older and had more comorbidities than candidates who were not tested. The fact that patients who were tested had a history of less time on dialysis may reflect the high mortality of diabetics on dialysis.
In conclusion, frailty and decreased physical performance appear to be independently associated with increased mortality on the kidney transplant waiting list. These measures may improve risk stratification of high-risk kidney transplant candidates. Which is the better predictor remains unclear and will need to be addressed in future multicenter studies with longer follow-up. The SPPB is an attractive alternative to frailty measurement, given that it is an objective measure that does not rely on self-report. In addition, it is a robust assessment of lower extremity function and can serve as a benchmark for interventional studies. Further studies are needed to determine whether improving frailty and physical performance prior to transplant can improve survival on the waiting list.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Mayo Clinic Department of Internal Medicine, the Division of Nephrology and Hypertension, and the Robert D. and Patricia E. Kern Center for the Science of Healthcare Delivery.