Abstract
Background
Frailty is common and associated with poor outcomes among kidney transplant (KT) recipients. While frailty improves in the first 3 months post-KT with restored kidney function, longer-term trajectories are likely to plateau/decline due to aging and other stressors (eg, immunosuppression). We evaluated longer-term post-KT trajectories of the physical frailty phenotype (PFP) and its components in adult patients at 2 centers.
Methods
PFP components were measured at admission, 1, 3, 6 months, 1 year, and annually thereafter post-KT. We used adjusted mixed-effects models to describe repeated measures of continuous components (weight, gait speed, grip strength, activity) and generalized estimating equations to quantify longitudinal, binomial response patterns (PFP; exhaustion).
Results
Among 1 336 recipients (mean age = 53) followed for a median of 1.9 years (interquartile range [IQR] = 0.1–3.2), likelihood of frailty declined in the first 2.5 years post-KT (adjusted odds ratio [aOR] = 0.96, 95% confidence interval [CI]: 0.95, 0.98), but increased after 2.5 years post-KT (aOR = 1.03, 95% CI: 1.00, 1.05). In the first 2.5 years post-KT, recipients demonstrated increases in weight (0.4 lbs/month, 95% CI: 0.3, 0.5), grip strength (0.2 kg/month, 95% CI: 0.1, 0.2), and activity (23.9 kcal/month, 95% CI: 17.5, 30.2); gait speed remained stable (−0.01 s/month, 95% CI: 0.01, 0.003). Additionally, likelihood of becoming exhausted declined post-KT (OR = 0.99, 95% CI: 0.98, 1.00). After 2.5 years post-KT, recipients demonstrated decreased grip strength (−0.07 kg/month, 95% CI: −0.12, −0.01) and activity (−20 kcal/month, 95% CI: −32.3, −8.2); they had stable weight (−0.003 lbs/month, 95% CI: −0.17, 0.16), gait speed (−0.003 s/month, 95% CI: −0.02, 0.01), and likelihood of becoming exhausted (OR = 1.01, 95% CI: 0.99, 1.02).
Conclusion
Despite frailty improvements in the first 2.5 years, recipients’ frailty worsened after 2.5 years post-KT. Specifically, they experienced gains in strength, activity, and exhaustion in the first 2.5 years post-KT, but declined in strength and activity after 2.5 years post-KT while experiencing persistent slowness. Clinicians should consider monitoring recipients for worsening frailty after 2.5 years despite shorter-term improvements.
Keywords: Frailty, Outcomes, Renal, Transplant
Kidney transplantation (KT) restores kidney function and is the preferred treatment for end-stage kidney disease (ESKD). It provides numerous health benefits for recipients of all ages, doubling remaining life expectancy and decreasing risk of mortality (1,2); however, after restoration of kidney function, KT recipients continue aging and are at risk of new or worsening geriatric conditions, such as frailty. Frailty is a syndrome characterized by decreased physiological reserve and manifests as a distinct vulnerability to acute or chronic stressors (3), such as KT or consumption of immunosuppressive medications (4–6). Frailty is prevalent in 15%–20% of adult KT recipients at the time of transplant and is associated with poor post-KT outcomes including delirium, longer length of stay, early hospitalization readmission, immunosuppression intolerance, poor health-related quality of life, cognitive decline, and mortality (4,7–12). While pre-KT frailty is associated with inferior post-KT outcomes, frailty itself has been shown to improve by 3 months post-KT (13). However, longer-term frailty trajectories are unknown, and it is unclear whether frailty trajectories continue to improve with restored kidney function despite increasing age in this population post-KT.
Though frailty is measured in many ways, the most commonly used metric is the physical frailty phenotype (PFP) (14). The PFP is operationalized based on the co-occurrence of any 3 of 5 manifested components of multiple compromised physiological mechanisms, measured by grip strength, gait speed, physical activity, exhaustion, and unintentional weight loss (3). However, little is known about the long-term trajectories of each of these frailty components post-KT, given the combination of stressors (eg, aging) with benefits (ie, restoration of kidney function). Additionally, past studies in the general population have hypothesized that certain components of frailty are more likely to manifest later than others, suggesting that they may be markers of more severe frailty (15–17). The differential impact of renal failure and post-KT renal recovery on frailty components is yet to be clarified but may reflect unique aspects of post-KT recovery or provide further insights into frailty severity in this population.
We sought to characterize the trajectories of the PFP and its components over a 5-year period post-KT by leveraging a 2-center prospective cohort study of KT recipients. Additionally, because frailty is prevalent among both younger and older adult KT recipients, we evaluated whether post-KT frailty trajectories differed by age.
Method
Study Design
We leveraged an ongoing, 2-center cohort study of frailty in KT recipients (December 2008–May 2019) from the Johns Hopkins Hospital in Baltimore, Maryland and the Michigan Medical Center in Ann Arbor, Michigan. A total of 1 336 eligible KT recipients (English-speaking and 18 years or older) were enrolled in the study at KT admission and included in this analysis. Recipients were assessed for physical frailty (3), as described below, then followed over time during routine clinical visits at approximately 1, 3, 6 months, 1 year, and annually thereafter post-KT. Recipient, donor, and transplant factors at the time of KT admission were self-reported or abstracted from medical records, including age, sex, race, education, diabetes status, Charlson Comorbidity Index (CCI) adapted for ESKD (18,19), cause of ESKD, years on dialysis, and donor type. Additional assessments including impairment in lower extremity function (Short Physical Performance Battery [SPPB]) (20), functional dependence in activities of daily living (ADLs) (21), functional dependence in instrumental activities of daily living (IADLs) (22), health-related quality of life, and Centers for Epidemiologic Studies—Depression (CES-D) scale were conducted at the time of KT admission.
All clinical and research activities being reported are consistent with the Declaration of Helsinki and the Declaration of Istanbul. The Institutional Review Boards of Johns Hopkins Hospital and the University of Michigan approved this study, and all enrolled participants provided written informed consent.
PFP and Its Components
The PFP, a validated tool to measure frailty in ESKD and KT populations of all ages (4,8,11,23–25), was ascertained at KT admission. Its 5 criteria were based on the original guidelines derived by Fried et al. in the Cardiovascular Health Study (3) and include (a) shrinking (self-report of unintentional weight loss of >10 lbs [measured dry weight] in the past year); (b) weakness (grip strength below an established gender- and BMI-based cutoff using a handheld dynamometer); (c) exhaustion (self-report based on 2 questions from the CES-D) (26); (d) low activity (kcals/week below an established cutoff based on the Minnesota Leisure Time Physical Activity questionnaire); and (e) slowness (gait speed for 15 feet below an established gender- and height-based cutoff) (3). Each of the criteria was scored as 0 or 1 representing absence or presence of that component and were summed to create a total score ranging from 0 to 5. Scores of 3–5 were defined as frail, as previously determined for ESKD and KT populations (8–13,23,24,27–29). Additionally, 4 components of the PFP, including weight (pounds), grip strength (kilograms), activity (kcal/week), and gait speed (seconds), were considered continuously.
Descriptive Analyses
Percentages, means with standard deviations (SDs), and medians with interquartile ranges (IQRs) were generated for participant characteristics and frailty components, and differences by frailty status were tested using Fisher’s exact tests, t-tests, and Kruskal–Wallis tests for categorical, normally, and nonnormally distributed continuous variables, respectively. Additionally, we generated unadjusted quadratic prediction and log odds plots to visualize post-KT trajectories for continuous and categorical PFP components, respectively.
Post-KT Trajectories of Frailty and Its Components
We used mixed-effects models with fixed and random effects for person (intercept) and time (slope) to describe repeated measures of continuous physical frailty components (weight, gait speed, grip strength, and activity), controlling for recipient, donor, and transplant factors. A knot was added at 2.5 years post-KT to account for the curve–linear relationship based on exploratory data analyses using lowess plots and unadjusted quadratic prediction plots, which revealed curve–linear trajectories with inflections at approximately 2.5 years post-KT. An unstructured correlation structure was selected for the random effects to generate the best possible model fit, allowing the model to directly calculate each variance and covariance value reflected by the data. Additionally, we conducted mixed-effects logistic regression models using adjusted generalized estimating equations to quantify longitudinal, binomial response patterns of exhaustion and PFP (frail vs nonfrail) and generate odds of exhaustion and frailty over time. To assess whether any of the trajectories differed by age (≥65 vs 18–64 years), we tested for interaction by age in separate models using the Wald test for interaction.
Sensitivity Analyses
Because activity is not normally distributed (Supplementary Figure 1), we conducted sensitivity analyses to address the skewed distribution using a random-effects Tobit model for the abovementioned trajectory analyses, a method conceived to estimate linear relationships when there is left- or right-censoring in the measured outcome (30). Additionally, given that weight trajectories do not account for intentionality, a key element of vulnerability for the PFP among KT recipients, we assessed the longitudinal, binomial response patterns of unintentional weight loss to account for intentionality. In order to determine whether overall estimates were robust to potential bias due to differential drop-out or attrition, we assumed an immortal cohort and incorporated inverse probability weighting in conjunction with GEE (WGEE) trajectories of frailty components measured on a continuous scale (ie, weight, grip strength, gait speed, and activity level) (31–36). Finally, to assess whether results were robust to potential differential rates of death by age, we conducted a joint analysis incorporating the random-effects model and a survival model accounting for mortality.
Statistical Analyses
All analyses were performed using Stata version 15 (StataCorp, College Station, TX). Two-sided p values <.05 were considered statistically significant.
Results
Study Population
Of the 1 336 KT recipients who met criteria for inclusion and were followed for a median of 1.9 years (IQR = 0.1–3.2), the mean age was 53 years (SD = 14.0), 39.7% were female, 53.8% were White, 38.4% were Black, and 67.1% had education beyond a high school degree (Table 1). Diabetes was the most common comorbidity (30.9%). Prior to KT, recipients had been on dialysis for a median of 2.8 years (IQR = 1.3–5.0).
Table 1.
Demographic and Health Characteristics at Time of Admission for Kidney Transplant (KT) by Frailty Status (n = 1 336)
| Factor | Frailty Status | |||
|---|---|---|---|---|
| Overall (n = 1 336) | Frail (n = 208) | Nonfrail (n = 1 128) | p | |
| Age, mean (SD) | 52.7 (14.0) | 55.4 (14.0) | 52.3 (13.4) | .004 |
| Female, n (%) | 530 (39.7) | 85 (40.9) | 445 (39.5) | .70 |
| Race, n (%) | .40 | |||
| White | 719 (53.8) | 101 (48.6) | 618 (54.8) | |
| Black | 513 (38.4) | 90 (43.3) | 423 (37.5) | |
| Asian | 50 (3.7) | 8 (3.9) | 42 (3.7) | |
| Other | 54 (4.0) | 9 (4.3) | 45 (4.0) | |
| Education, n (%) | .01 | |||
| High school degree or less | 401 (32.9) | 78 (40.8) | 323 (31.4) | |
| >High school degree | 819 (67.1) | 113 (59.2) | 706 (68.6) | |
| Deceased donor, n (%) | 495 (39.5) | 68 (34.7) | 427 (40.4) | .15 |
| Cognitive impairment (3MS <80) | 94 (7.0) | 22 (10.6) | 72 (6.4) | .04 |
| SPPB impaired, n (%) | 500 (52.3) | 100 (71.9) | 400 (49.0) | <.001 |
| ADL dependent, n (%) | 8 (1.6) | 4 (4.3) | 4 (1.0) | .04 |
| IADL dependent, n (%) | 58 (11.2) | 21 (22.3) | 37 (8.8) | <.001 |
| CCI, median (IQR) | 1 (0–3) | 2 (0–3) | 0 (0–2) | <.001 |
| Comorbidities, n (%) | ||||
| Myocardial infarction | 74 (5.6) | 16 (7.7) | 58 (5.2) | .14 |
| Peripheral vascular disease | 80 (6.1) | 17 (8.2) | 63 (5.7) | .16 |
| Cerebral vascular disease | 48 (3.6) | 14 (6.8) | 34 (3.1) | .01 |
| Chronic lung disease | 62 (4.7) | 13 (6.3) | 49 (4.4) | .28 |
| Rheumatological disease | 179 (13.6) | 39 (18.8) | 140 (12.6) | .02 |
| Peptic ulcer disease | 52 (4.0) | 11 (5.3) | 41 (3.7) | .25 |
| Diabetes | 409 (30.9) | 83 (40.1) | 326 (29.2) | .002 |
| Metastatic cancer | 5 (0.4) | 0 (0.0) | 5 (0.5) | >.99 |
| Leukemia | 3 (0.2) | 0 (0.0) | 3 (0.3) | >.99 |
| Lymphoma | 2 (0.2) | 1 (0.5) | 1 (0.1) | .29 |
| HIV | 29 (2.2) | 4 (1.9) | 25 (2.3) | >.99 |
| Congestive heart failure | 70 (6.0) | 18 (9.6) | 52 (5.3) | .03 |
| Cause of ESKD, n (%) | .32 | |||
| Glomerular diseases | 271 (27.7) | 49 (29.2) | 222 (27.3) | |
| Diabetes | 183 (18.7) | 35 (20.8) | 148 (18.2) | |
| Hypertension | 305 (31.1) | 55 (32.7) | 250 (30.8) | |
| Other | 221 (22.6) | 29 (17.3) | 192 (23.7) | |
| Years on dialysis, median (IQR) | 2.8 (1.3–5.0) | 3.1 (1.4–5.5) | 2.7 (1.3–4.9) | .45 |
Notes: 3MS = modified Mini-Mental State Examination total score (range = 0–100); ADL = activities of daily living; IADL = instrumental activities of daily living; CCI = Charlson Comorbidity Index; SPPB = Short Physical Performance Battery; HIV = human immunodeficiency virus; ESKD = end-stage kidney disease.
Unadjusted Prevalence of Frailty and Its Components at KT Admission
Of the KT recipients, 208 (15.6%) were frail, 46.6% had weakness, 46.3% low activity, 28.9% exhaustion, 16.2% slowness, and 14.2% unintentional weight loss at the time of KT admission. Frail recipients were more likely to be older (55.4 vs 52.3 years, p = .004), have cognitive impairment (10.6% vs 6.4%, p = .04), have a higher CCI (2 vs 0, p < .001), diabetes (40.1% vs 29.2%, p = .002), cerebral vascular disease (6.8% vs 3.1%, p = .01), congestive heart failure (9.6% vs 5.3%, p = .03), SPPB impairment (71.9% vs 49.0%, p < .001), and dependence in ADLs (4.3% vs 1.0, p = .04) and IADLs (4.3% vs 1.0, p = .04).
Adjusted Levels of Frailty and Its Components at KT Admission
After adjustment, the odds of overall frailty was 0.13 (95% CI: 0.09, 0.18), and older recipients were 1.73-fold (adjusted odds ratio [aOR] = 1.73, 95% CI: 1.24, 2.42) more likely to have frailty at the time of KT admission than younger recipients. The adjusted mean weight at KT admission was 194.3 lbs, grip strength was 31.7 kg, gait speed was 4.9 seconds, and activity was 867.2 kcal/week (Table 2). Additionally, the odds of meeting the exhaustion criterion was 0.34 (95% CI: 0.27, 0.44). At admission for KT, there were no significant differences by age in weight, grip strength, gait speed, activity, or exhaustion (Table 2).
Table 2.
Adjusted Estimates of Frailty Component Trajectories Post-Kidney Transplantation (KT), Overall and by Age (n = 1 336)
| Subgroup | Level (95% CI) | Monthly Change (95% CI) | |||||
|---|---|---|---|---|---|---|---|
| At KT | 2.5 Years Post-KT | 5 Years Post-KT | 0–2.5 Years Post-KT | p | 2.5–5 Years Post-KT | p | |
| Frailty (odds) | 0.13 (0.09, 0.18) | 0.04 (0.02, 0.07) | 0.09 (0.05, 0.17) | 0.96 (0.95, 0.98) | <.001 | 1.03 (1.00, 1.05) | .03 |
| Age (years) | |||||||
| ≥65 | 0.18 (0.12, 0.26) | 0.12 (0.06, 0.23) | 0.11 (0.03, 0.44) | 0.99 (0.97, 1.01) | .26 | 1.00 (0.95, 1.05) | .96 |
| 18–64 | 0.10 (0.07, 0.15) | 0.04 (0.02, 0.07) | 0.07 (0.03, 0.14) | 0.97 (0.96, 0.99) | <.001 | 1.01 (0.99, 1.05) | .30 |
| Odds ratio | 1.73 (1.24, 2.42) | 2.86 (1.40, 5.82) | 1.74 (0.40, 7.6) | 1.02 (0.99, 1.04) | .22 | 0.98 (0.93, 1.04) | .59 |
| Weight (pounds) | 194.3 (187.6, 201.0) | 206.1 (199.0, 213.2) | 206.0 (197.7, 214.2) | 0.39 (0.30, 0.48) | <.001 | −0.003 (−0.17, 0.16) | .97 |
| Age (years) | |||||||
| ≥65 | 194.4 (187.0, 201.9) | 197.2 (187.9, 206.5) | 200.5 (184.8, 216.1) | 0.09 (−0.12, 0.31) | .40 | 0.11 (−0.38, 0.59) | .66 |
| 18–64 | 197.5 (190.2, 204.7) | 211.1 (203.4, 218.8) | 210.3 (201.5, 219.2) | 0.45 (0.36, 0.55) | <.001 | −0.03 (−0.20, 0.15) | .78 |
| Difference | −3.0 (−10.0, 4.0) | −13.9 (−23.2, −4.6) | −9.9 (−26.2, 6.4) | −0.36 (−0.60, −0.12) | .003 | 0.13 (−0.38, 0.65) | .61 |
| Grip strength (kg) | 31.7 (30.2, 33.2) | 36.5 (34.9, 38.2) | 34.4 (32.4, 363.5) | 0.16 (0.13, 0.19) | <.001 | −0.07 (−0.12, −0.01) | .01 |
| Age (years) | |||||||
| ≥65 | 27.5 (25.8, 29.3) | 30.4 (27.9, 32.9) | 24.9 (20.3, 29.4) | 0.10 (0.02, 0.17) | .01 | −0.18 (−0.35, −0.02) | .03 |
| 18–64 | 32.8 (31.1, 34.4) | 38.0 (36.2, 39.8) | 36.2 (34.0, 38.4) | 0.17 (0.14, 0.20) | <.001 | −0.06 (−0.12, −0.005) | .03 |
| Difference | −5.2 (−6.9, −3.6) | −7.6 (−10.1, −5.0) | −11.3 (−16.1, −6.5) | −0.08 (−0.16, 0.001) | .05 | −0.12 (−0.30, 0.05) | .16 |
| Gait speed (seconds) | 4.9 (4.6, 5.2) | 4.8 (4.4, 5.1) | 4.7 (4.2, 5.2) | −0.005 (−0.01, 0.003) | .20 | −0.003 (−0.02, 0.01) | .69 |
| Age (years) | |||||||
| ≥65 | 5.6 (5.3, 5.9) | 6.0 (5.3, 6.6) | 50.2 (4.0, 6.3) | 0.01 (−0.01, 0.03) | .22 | −0.03 (−0.07, 0.02) | .23 |
| 18–64 | 4.7 (4.4, 5.1) | 4.5 (4.1, 4.9) | 4.5 (4.0, 5.0) | −0.01 (−0.02, 0.0002) | .06 | 0.0005 (−0.01, 0.01) | .95 |
| Difference | 0.9 (0.6, 1.2) | 1.5 (8.3, 2.1) | 0.7 (−0.5, 1.9) | 0.02 (−0.001, 0.04) | .06 | −0.03 (−0.07, 0.02) | .24 |
| Activity (kcal/week) | 867.2 (661.7, 1 072.7) | 1 583.19 (1 321.8, 1 844.6) | 975.1 (608.2, 1 341.9) | 23.9 (17.5, 30.2) | <.001 | −20.3 (−32.3, −8.2) | .001 |
| Age (years) | |||||||
| ≥65 | 715.6 (481.2, 949.9) | 1 688.5 (1 229.4, 2 147.5) | 868.1 (−44.1, 1 780.2) | 32.4 (16.7, 48.1) | <.001 | −27.3 (−61.3, 6.6) | .11 |
| 18–64 | 924.1 (699.9, 1 148.3) | 1 588.7 (1 302.5, 1 874.9) | 1 021.0 (627.6, 1 414.5) | 22.2 (15.2, 29.1) | <.001 | −18.9 (−31.8, −6.0) | .004 |
| Difference | −208.5 (−430.2, 13.2) | 99.8 (−388.7, 588.2) | −153.0 (−1 118.4, 812.5) | 10.3 (−6.9, 27.5) | .24 | −8.4 (−44.7, 27.9) | .65 |
| Exhaustion (odds) | 0.34 (0.27, 0.44) | 0.23 (0.17, 0.32) | 0.28 (0.18, 0.44) | 0.99 (0.98, 1.00) | .002 | 1.01 (0.99, 1.02) | .38 |
| Age (years) | |||||||
| ≥65 | 0.24 (0.17, 0.33) | 0.42 (0.24, 0.73) | 0.46 (0.16, 1.37) | 1.02 (1.00, 1.04) | .11 | 1.00 (0.96, 1.05) | .87 |
| 18–64 | 0.32 (0.25, 0.42) | 0.30 (0.21, 0.42) | 0.36 (0.22, 0.59) | 1.00 (0.99, 1.01) | .61 | 1.01 (0.99, 1.03) | .49 |
| Odds ratio | 0.74 (0.54, 1.02) | 1.40 (0.76, 2.58) | 1.28 (0.40, 4.07) | 1.02 (1.00, 1.05) | .09 | 1.00 (0.95, 1.05) | .91 |
Notes: ESKD = end-stage kidney disease. Mixed-effects models for continuous components (weight, grip strength, gait speed, activity) and generalized estimating equations for dichotomous components (exhaustion) were used to estimate level and change per month of each frailty component in separate models. All models were adjusted for age, sex, race, cause of ESKD, years on dialysis, diabetes status, and donor type. Intergroup differences and monthly changes that are statistically significant at p < .05 are bolded.
Adjusted Trajectories of Frailty and Its Components
Trajectories in frailty and its components in the first 2.5 years post-KT
In the first 2.5 years post-KT, recipients experienced a lower likelihood of frailty over time compared to time of KT (aOR = 0.96, 95% CI: 0.95, 0.98). Individual components of PFP demonstrated varying trajectories post-KT (Figures 1 and 2). In the first 2.5 years post-KT, recipients experienced changes in nearly all components of the frailty phenotype after adjustment (Table 2). Specifically, recipients demonstrated per-month increases in weight (slope = 0.39 lbs/month, 95% CI: 0.30, 0.48), grip strength (slope = 0.16 kg/month, 95% CI: 0.13, 0.19), and activity (slope = 23.9 kcal/week/month, 95% CI: 17.5, 30.2). Recipients also demonstrated a lower likelihood of reporting exhaustion (aOR = 0.99, 95% CI: 0.98, 1.00; p = .002) in the first 2.5 years post-KT compared to time of KT. The only component in which recipients did not demonstrate any change in the first 2.5 years post-KT was gait speed (slope = −0.005 s/month, 95% CI: −0.01, 0.003).
Figure 1.
Unadjusted quadratic prediction plots of continuous physical frailty phenotype components (gait speed, grip strength, weight, and activity) among kidney transplant (KT) recipients (n = 1 336).
Figure 2.
Unadjusted predicted probabilities of the frailty (physical frailty phenotype) and exhaustion among kidney transplant (KT) recipients post-KT (n = 1 336).
Trajectories did not differ significantly by recipient age for overall frailty or for any of its components except for weight (p for interaction = .003); specifically, younger recipients experienced steep increases in weight (slope = 0.45 lbs/month, 95% CI: 0.36, 0.55), while older recipients remained stable (slope = 0.09 lbs/month, 95% CI: −0.12, 0.31) in the first 2.5 years post-KT (Table 2).
Trajectories in frailty and its components after 2.5 years post-KT
Between 2.5 and 5 years post-KT, recipients experienced a greater likelihood of frailty over time compared to time of KT (aOR = 1.03, 95% CI: 1.00, 1.05; p = .03). Regarding its components, recipients had stable weight (slope = −0.003 lbs/month, 95% CI: −0.17, 0.16) and gait speed (slope = −0.003 s/month, 95% CI: −0.02, 0.01), as well as no change in the likelihood of reporting exhaustion during this period compared to time of KT (aOR = 1.01, 95% CI: 0.99, 1.02). However, recipients demonstrated a decrease in grip strength (slope = −0.07 kg/month, 95% CI: −0.12, −0.01) and activity (slope = −20.3 kcal/week/month, 95% CI: −32.3, −8.2). These trajectories of overall frailty and its components did not differ significantly by recipient age (all p for interactions >.05; Table 2).
Sensitivity Analyses
After accounting for nonnormal distribution of activity, direction and magnitude of associations remained relatively robust across analyses among KT recipients. Specifically, in the first 2.5 years post-KT, activity increased (slope = 20.68 kcal/week/month, 95% CI: 15.2, 26.2; p < .001) among KT recipients, but declined after 2.5 years post-KT (slope = −8.6 kcal/week/month 95% CI: −13.9, −3.3; p = 0.001). These trajectories did not differ by age before or after 2.5 years post-KT (p > .05).
Additionally, assessments of binomial response patterns of unintentional weight loss generated similar results to continuous findings. Specifically, in the first 2.5 years post-KT, there was a lower likelihood of unintentional weight loss (ie, weight stability and/or gain) (aOR = 0.99, 95% CI: 0.98, 1.00; p = .02), but a greater likelihood of unintentional weight loss after 2.5 years (aOR = 1.04, 95% CI: 1.02, 1.06; p < .001). However, unlike the analyses of continuous weight, these trajectories did not differ by age before or after 2.5 years post-KT (p > .05).
For all continuous frailty components, inferences regarding overall trajectories remained relatively robust after accounting for attrition using WGEE (Supplementary Table 1). The sole exceptions were that grip strength improved (slope = 0.11 kg/month, 95% CI: 0.07, 0.15) instead of declined after 2.5 years post-KT, and that gait speed declined (slope = −0.01 s/month, 95% CI: −0.02, −0.002) rather than remained stable after 2.5 years post-KT. Regarding differences in trajectories by age after accounting for differential rates of mortality using joint models, all inferences remained robust (Supplementary Table 2).
Discussion
In this 2-center cohort study of 1 336 adult KT recipients, we found that recipients experienced a lower likelihood of frailty over time in the first 2.5 years post-KT (aOR = 0.96, p < .001), but a higher likelihood of frailty in the long term between 2.5 years and 5 years post-KT (aOR = 1.03, p = .03). Nearly all components of the PFP demonstrated significant changes in the first 2.5 years after KT, including increases in weight (slope = 0.39 lbs/month), grip strength (slope = 0.16 kg/month), and activity (slope = 23.9 kcal/week/month), as well as a lower likelihood of reporting exhaustion (OR 0.99, p = .002). Between 2.5 and 5 years post-KT, frailty components stabilized (weight: slope = −0.003 lbs/month, exhaustion: aOR = 1.01) or demonstrated modest decreases (grip strength: slope = −0.07 kg/month, activity: slope = −20.3 kcal/week/month). Gait speed was the sole component that demonstrated no changes post-KT in the short- (slope = −0.005 s/month) or long term (slope = −0.003 s/month). Additionally, older and younger KT recipients experienced similar trajectories of frailty and its components except for weight change.
The decreased likelihood of frailty and improved frailty components in the first 2.5 years post-KT are encouraging. These findings corroborate the dynamic nature of frailty observed in prior studies pre- and post-KT (13,37) and underscore the overall benefit of KT and restoration of kidney function despite the insult of surgery in patients with poor reserve (13). However, we observed a greater likelihood of frailty after 2.5 years post-KT, manifested through the stabilization (weight, gait speed, and exhaustion) or modest worsening (grip strength and activity) of frailty components. Whether these long-term changes represent the limits of KT-specific benefit on frailty; the impact of chronic consumption of immunosuppressive medications (4–6); the contributions of chronic comorbidity, such as lifelong diabetes; a lapse in more intensive follow-up, encouragement, and intervention from providers; or simply the impact of aging, require further investigation. A better understanding of the reasons why frailty components plateau or worsen at this point could guide additional interventions to optimize recovery of resilience in this population.
Though frailty is often considered a disease primarily of older adults (38), as shown in prior studies, we found that younger KT recipients were frail (39) and experienced similar post-KT frailty component trajectories to older KT recipients. These findings suggest that both primary and secondary manifestations of frailty are likely present in this population, reflecting the impact of age- and disease-specific factors on functional reserve, respectively (40). All KT recipients are at risk for secondary frailty due to their ESKD and the stress of KT, as acute and chronic diseases exhaust the reserve function of the body’s organ systems (41). As recipients recover from their transplant surgery, they fully, or partially, recover from the cause of their secondary frailty, which is consistent with our findings of frailty improvement in the first 2.5 years post-KT.
Weight trajectories post-KT were one of the only areas in which younger and older KT recipients differed, with younger KT recipients experiencing more dramatic increases in weight compared to their older counterparts in the first 2.5 years post-KT. This difference might reflect differences in postoperative management, such as changes in prescribed immunosuppressive medication, given that older recipients are more likely to be frail, frail patients have a higher risk of immunosuppression intolerance (42), and certain immunosuppressives may induce glucose intolerance, hyperlipidemia, and weight gain (43). However, evaluating weight loss as a component of frailty is challenging given that 42% of the US population is obese (44), including a substantial proportion of the frail population (45–47). Indeed, characterizing trajectories of unintentional weight loss as a binomial component of the PFP demonstrated similar overall trajectories, but no differences in those trajectories by age, reinforcing “intentionality” as a key component of vulnerability (48) that captures physiological changes beyond those presented by chronological age.
Studies of the natural history of frailty have identified PFP components that serve as the earliest markers of frailty (weakness and exhaustion) versus those that are manifested closer to the onset of frailty (weight loss) when frailty is severe enough to be identified by the PFP among community-dwelling older adults (15–17). Such studies have suggested that the hierarchical ordering of different frailty components could reflect frailty severity (49). While our data cannot confirm these hypotheses, we found that weakness and exhaustion, along with low activity, were among the most common frailty criteria. Additionally, our observations of patients recovering from KT—and their pre-KT early-stage renal disease—presented a unique picture of frailty reversal; the observed increases in weight, grip strength, activity, and exhaustion suggest that these components recover before gait speed in the context of KT, which may show later improvements beyond the 5 years of our study. Further investigation is needed to establish whether the reversal of frailty components follows similarly reported sequential patterns, and whether there are clinical implications for a “frailty component hierarchy” within the context of KT.
This study is not without limitations. Though this study involved a 2-center design, with a diverse KT recipient population, results may not be fully generalizable to the national recipient population. Additionally, given the longer follow-up time, bias due to attrition is another potential limitation, though this is unlikely to affect inferences significantly based on our sensitivity analyses. Nonetheless, strengths of this study include the prospective, validated measurement of frailty pre- and post-KT, as well as long-term follow-up of study participants with repeated assessments. Additionally, due to the large sample size, this study had sufficient power to explore differences in trajectories by age, providing the unique opportunity to glean how KT can reverse a leading cause of frailty (ie, kidney disease) among those where aging is not the major contributor of their frailty status. Finally, the lack of a nontransplanted comparison group limits inferences about the effects of aging versus other factors on long-term frailty trajectory.
Conclusion
In conclusion, despite the lower likelihood of frailty in the first 2.5 years post-KT, the likelihood of frailty increased after 2.5 years post-KT. In the first 2.5 years, increases in strength, activity, and weight, as well as decreases in exhaustion, contributed to the lower likelihood of frailty; however, KT recipients experienced persistent slowness during this period. After 2.5 years post-KT, recipients demonstrated declines in grip strength and activity while other components remained unchanged, contributing to the observed increase in likelihood of frailty in the long term. Clinicians should consider monitoring KT recipients for worsening frailty status given declines in strength and activity after 2.5 years post-KT, as well as persistent slowness in the first 2.5 years post-KT. The long-term stagnation or worsening of frailty components highlights an opportunity to further improve post-KT frailty and potentially improve outcomes post-KT.
Supplementary Material
Contributor Information
Nadia M Chu, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA.
Jessica Ruck, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
Xiaomeng Chen, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
Qian-Li Xue, Division of Geriatric Medicine and Gerontology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA.
Silas P Norman, Department of Internal Medicine, Division of Nephrology, University of Michigan School of Medicine, Ann Arbor, Michigan, USA.
Dorry L Segev, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA.
Mara A McAdams-DeMarco, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA.
Funding
This study was funded by the National Institute on Aging (grant number R01AG055781). Study investigators were funded by the National Institute of Diabetes and Digestive and Kidney Disease, the National Institute of Allergy and Infectious Diseases, and the National Institute on Aging: grant numbers K01AG064040 (PI: N.M.C.), P30AG021334 (N.M.C.), F32AG067642 (PI: J.R.), P30AG021334 (Q.L.X.), K24AI144954 (PI: D.L.S.), R01AG055781 (PI: M.A.M.-D.), and R01DK114074 (PI: M.A.M.-D.).
Conflict of Interest
None declared.
Author Contributions
N.M.C.: participated in concept design, data analysis, interpretation, drafting, critical revision, and approval of the article. J.R.: participated in interpretation, drafting, critical revision, and approval of the article. X.C.: participated in critical revision and approval of the article. Q.-L.X.: participated in critical revision and approval of the article. S.P.N.: participated in critical revision and approval of the article. D.L.S.: participated in concept design, critical revision, and approval of the article. M.A.M.-D.: participated in concept design, interpretation, critical revision, and approval of the article.
References
- 1. Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med. 1999;341(23): 1725–1730. doi: 10.1056/NEJM199912023412303 [DOI] [PubMed] [Google Scholar]
- 2. Knoll GA. Is kidney transplantation for everyone? The example of the older dialysis patient. Clin J Am Soc Nephrol. 2009;4(12):2040–2044. doi: 10.2215/CJN.04210609 [DOI] [PubMed] [Google Scholar]
- 3. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146–M157. doi: 10.1093/gerona/56.3.m146 [DOI] [PubMed] [Google Scholar]
- 4. McAdams-DeMarco MA, Law A, Tan J, et al. Frailty, mycophenolate reduction, and graft loss in kidney transplant recipients. Transplantation. 2015;99(4):805–810. doi: 10.1097/TP.0000000000000444 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Fozouni L, Lai JC. Frailty as an immune-mediated condition that leads to an increased risk of acute cellular rejection in liver transplant recipients. Clin Liver Dis (Hoboken). 2020;15(6):243–245. doi: 10.1002/cld.960 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Harhay MN, Rao MK, Woodside KJ, et al. An overview of frailty in kidney transplantation: measurement, management and future considerations. Nephrol Dial Transplant. 2020;35(7):1099–1112. doi: 10.1093/ndt/gfaa016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chu NM, Gross AL, Shaffer AA, et al. Frailty and changes in cognitive function after kidney transplantation. J Am Soc Nephrol. 2019;30(2): 336–345. doi: 10.1681/ASN.2018070726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. McAdams-DeMarco MA, Olorundare IO, Ying H, et al. Frailty and postkidney transplant health-related quality of life. Transplantation. 2018;102(2):291–299. doi: 10.1097/TP.0000000000001943 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Haugen CE, Mountford A, Warsame F, et al. Incidence, risk factors, and sequelae of post-kidney transplant delirium. J Am Soc Nephrol. 2018;29(6):1752–1759. doi: 10.1681/ASN.2018010064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. McAdams-DeMarco MA, King EA, Luo X, et al. Frailty, length of stay, and mortality in kidney transplant recipients: a national registry and prospective cohort study. Ann Surg. 2017;266(6):1084–1090. doi: 10.1097/SLA.0000000000002025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. McAdams-DeMarco MA, Law A, Salter ML, et al. Frailty as a novel predictor of mortality and hospitalization in individuals of all ages undergoing hemodialysis. J Am Geriatr Soc. 2013;61(6):896–901. doi: 10.1111/jgs.12266 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. McAdams-DeMarco MA, Law A, King E, et al. Frailty and mortality in kidney transplant recipients. Am J Transplant. 2015;15(1):149–154. doi: 10.1111/ajt.12992 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. McAdams-DeMarco MA, Isaacs K, Darko L, et al. Changes in frailty after kidney transplantation. J Am Geriatr Soc. 2015;63(10):2152–2157. doi: 10.1111/jgs.13657 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Buta BJ, Walston JD, Godino JG, et al. Frailty assessment instruments: systematic characterization of the uses and contexts of highly-cited instruments. Ageing Res Rev. 2016;26:53–61. doi: 10.1016/j.arr.2015.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Stenholm S, Ferrucci L, Vahtera J, et al. Natural course of frailty components in people who develop frailty syndrome: evidence from two cohort studies. J Gerontol A Biol Sci Med Sci. 2019;74(5):667–674. doi: 10.1093/gerona/gly132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Xue QL. The frailty syndrome: definition and natural history. Clin Geriatr Med. 2011;27(1):1–15. doi: 10.1016/j.cger.2010.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Xue QL, Bandeen-Roche K, Varadhan R, Zhou J, Fried LP. Initial manifestations of frailty criteria and the development of frailty phenotype in the Women’s Health and Aging Study II. J Gerontol A Biol Sci Med Sci. 2008;63(9):984–990. doi: 10.1093/gerona/63.9.984 [DOI] [PubMed] [Google Scholar]
- 18. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245–1251. doi: 10.1016/0895-4356(94)90129-5 [DOI] [PubMed] [Google Scholar]
- 19. Hemmelgarn BR, Manns BJ, Quan H, Ghali WA. Adapting the Charlson Comorbidity Index for use in patients with ESRD. Am J Kidney Dis. 2003;42(1):125–132. doi: 10.1016/s0272-6386(03)00415-3 [DOI] [PubMed] [Google Scholar]
- 20. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85–M94. doi: 10.1093/geronj/49.2.m85 [DOI] [PubMed] [Google Scholar]
- 21. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA. 1963;185:914–919. doi: 10.1001/jama.1963.03060120024016 [DOI] [PubMed] [Google Scholar]
- 22. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9(3):179–186. doi: 10.1111/j.1532-5415.1983.tb03391.x [DOI] [PubMed] [Google Scholar]
- 23. Garonzik-Wang JM, Govindan P, Grinnan JW, et al. Frailty and delayed graft function in kidney transplant recipients. Arch Surg. 2012;147(2): 190–193. doi: 10.1001/archsurg.2011.1229 [DOI] [PubMed] [Google Scholar]
- 24. McAdams‐DeMarco MA, Law A, Salter ML, et al. Frailty and early hospital readmission after kidney transplantation. Am J Transplant. 2013;13:2091–2095. doi: 10.1111/ajt.12300 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Nastasi AJ, McAdams-DeMarco MA, Schrack J, et al. Pre-kidney transplant lower extremity impairment and post-kidney transplant mortality. Am J Transplant. 2018;18(1):189–196. doi: 10.1111/ajt.14430 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. doi: 10.1177/014662167700100306 [DOI] [Google Scholar]
- 27. McAdams-DeMarco MA, Ying H, Olorundare I, et al. Individual frailty components and mortality in kidney transplant recipients. Transplantation. 2017;101(9):2126–2132. doi: 10.1097/TP.0000000000001546 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. McAdams-DeMarco MA, Tan J, Salter ML, et al. Frailty and cognitive function in incident hemodialysis patients. Clin J Am Soc Nephrol. 2015;10(12):2181–2189. doi: 10.2215/CJN.01960215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. McAdams-DeMarco MA, Suresh S, Law A, et al. Frailty and falls among adult patients undergoing chronic hemodialysis: a prospective cohort study. BMC Nephrol. 2013;14:224. doi: 10.1186/1471-2369-14-224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Tobin J. Estimation of relationships for limited dependent variables. Econometrica. 1958;26:24–36. doi: 10.2307/1907382 [DOI] [Google Scholar]
- 31. Rubin DB. Multiple Imputation for Nonresponse in Surveys. John Wiley & Sons; 2004. [Google Scholar]
- 32. Robins JM, Rotnitzky A, Zhao LP. Analysis of semiparametric regression models for repeated outcomes in the presence of missing data. J Am Stat Assoc. 1995;90:106–121. doi: 10.2307/2291134 [DOI] [Google Scholar]
- 33. Schafer JL. Multiple imputation in multivariate problems when the imputation and analysis models differ. Statistica Neerlandica. 2003;57:19–35. doi: 10.1111/1467-9574.00218 [DOI] [Google Scholar]
- 34. Aloisio KM, Swanson SA, Micali N, Field A, Horton NJ. Analysis of partially observed clustered data using generalized estimating equations and multiple imputation. Stata J. 2014;14(4):863–883. [PMC free article] [PubMed] [Google Scholar]
- 35. Birhanu T, Molenberghs G, Sotto C, Kenward MG. Doubly robust and multiple-imputation-based generalized estimating equations. J Biopharm Stat. 2011;21(2):202–225. doi: 10.1080/10543406.2011.550096 [DOI] [PubMed] [Google Scholar]
- 36. Beunckens C, Sotto C, Molenberghs G. A simulation study comparing weighted estimating equations with multiple imputation based estimating equations for longitudinal binary data. Comput Stat Data Anal. 2008;52:1533–1548. doi: 10.1016/j.csda.2007.04.020 [DOI] [Google Scholar]
- 37. Chu NM, Deng A, Ying H, et al. Dynamic frailty before kidney transplantation: time of measurement matters. Transplantation. 2019;103(8):1700–1704. doi: 10.1097/TP.0000000000002563 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. McAdams-DeMarco MA, Ying H, Olorundare I, et al. Individual frailty components and mortality in kidney transplant recipients. Transplantation. 2017;101(9):2126–2132. doi: 10.1097/TP.0000000000001546 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Chu NM, Chen X, Norman SP, et al. Frailty prevalence in younger end-stage kidney disease patients undergoing dialysis and transplantation. Am J Nephrol. 2020;51(7):501–510. doi: 10.1159/000508576 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Fried L, Walston J. Frailty and failure to thrive. In: Hazzard W, Blass J, Halter J, eds. Principles of Geriatric Medicine and Gerontology. McGraw Hill; 2003:1487–1502. [Google Scholar]
- 41. Bergman H, Ferrucci L, Guralnik J, et al. Frailty: an emerging research and clinical paradigm—issues and controversies. J Gerontol A Biol Sci Med Sci. 2007;62(7):731–737. doi: 10.1093/gerona/62.7.731 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. McAdams-DeMarco MA, Chu NM, Segev DL. Frailty and long-term post-kidney transplant outcomes. Curr Transplant Rep. 2019;6(1):45–51. doi: 10.1007/s40472-019-0231-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Lo A. Immunosuppression and metabolic syndrome in renal transplant recipients. Metab Syndr Relat Disord. 2004;2(4):263–273. doi: 10.1089/met.2004.2.263 [DOI] [PubMed] [Google Scholar]
- 44. Hales C, Carroll M, Fryar C, Ogden C.. Prevalence of obesity and severe obesity among adults: United States, 2017–2018. NCHS Data Brief, no 360. Statistics NCfH; 2020. [PubMed] [Google Scholar]
- 45. Woods NF, LaCroix AZ, Gray SL, et al. Frailty: emergence and consequences in women aged 65 and older in the Women’s Health Initiative Observational Study. J Am Geriatr Soc. 2005;53(8):1321–1330. doi: 10.1111/j.1532-5415.2005.53405.x [DOI] [PubMed] [Google Scholar]
- 46. Blaum CS, Xue QL, Michelon E, Semba RD, Fried LP. The association between obesity and the frailty syndrome in older women: the Women’s Health and Aging Studies. J Am Geriatr Soc. 2005;53(6):927–934. doi: 10.1111/j.1532-5415.2005.53300.x [DOI] [PubMed] [Google Scholar]
- 47. Puts MT, Visser M, Twisk JW, Deeg DJ, Lips P. Endocrine and inflammatory markers as predictors of frailty. Clin Endocrinol (Oxf). 2005;63(4): 403–411. doi: 10.1111/j.1365-2265.2005.02355.x [DOI] [PubMed] [Google Scholar]
- 48. Harhay MN, Chen X, Chu NM, Norman SP, Segev DL, McAdams-DeMarco M. Pre-kidney transplant unintentional weight loss leads to worse post-kidney transplant outcomes. Nephrol Dial Transplant. 2021;36(10):1927–1936. doi: 10.1093/ndt/gfab164 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Chu NM, Bandeen-Roche K, Xue QL, Carlson MC, Sharrett AR, Gross AL. Physical frailty phenotype criteria and their synergistic association on cognitive functioning. J Gerontol A Biol Sci Med Sci. 2021;76(9):1633–1642. doi: 10.1093/gerona/glaa267 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.