Abstract
Background and Aims
Frailty is a complex health state of increased vulnerability to stressors and is common in those with chronic kidney disease (CKD). Frailty is strongly associated with progressive renal impairment and independently linked with adverse outcomes in all stages of CKD such as disability, falls, hospitalisation, and mortality. How frailty varies across the disease trajectory is not well-defined, particularly in those with early stages of disease. Identifying those with CKD who may be at risk of becoming frail could inform the early intervention and understanding how frailty changes across the disease spectrum may aid better future management of these patients.
Method
This is a secondary analysis of data of n=4736 CKD patients (42% female, mean age 59.7 (SD: 16.3) years) from a prospective observational study of physical activity behaviours. A modified assessment of frailty (based on the Fried phenotype model) was conducted across each of the CKD stages ((n=262 CKD stage 1/2, n=678 CKD stage 3, n=610 CKD stage 4/5, n=1094 haemodialysis (HD), n=177 peritoneal dialysis (PD), n=1915 transplant (TX)). Markers of frailty were defined as (1) poor endurance (defined as a VO2 peak of <20ml/kg/min estimated from the Duke Activity Index Scale); (2) low physical activity level (classified as ‘insufficiently active’ using General Practice Physical Activity Questionnaire); and (3) slowness (self-reported slow walking pace (less than 3mph)). Participants were classified as ‘frail’ if ≥2 markers were present. Frequency analysis and chi-squared tests were conducted to identify and compare patients with ≥2 markers of frailty across the disease trajectory. Binominal logistical regression was performed to determine which factors were associated with being frail.
Results
Across the total cohort, the majority (56%) of patients exhibited ≥2 markers of frailty. The presence of frailty increased concurrently with disease decline (39% CKD stage 1/2, 58% CKD stage 3, 74% CKD stage 4/5, 76% HD, 66% PD, 39% TX). The prevalence of frailty was significantly higher in patients in CKD stage 3, CKD stage 4/5, HD and PD patients when compared to CKD stage 1/2 and TX patients (p<.001). Frailty was significantly higher in patients in CKD stage 4/5 and on HD compared to those in CKD stage 3 (p<.001). Patients who were female (OR = 1.42 [1.23 to 1.65] p<.001), older (OR= 1.05 [1.05 to 1.06] p <.001), and of non-white ethnicity (OR = 2.41 [2.00 to 2.90] p<.001) had an increased likelihood of being frail. Increased number of comorbidities (OR = 1.26 [1.18 to 1.34] p<.001) were associated with an increased likelihood of being frail (Figure 1).
Conclusion
The proportion of patients with ≥2 markers of frailty increased with disease progression until it reached a peak in HD with almost 8 out of 10 HD patients exhibiting frailty. Frailty prevalence was reduced in those with a renal TX. Over half of those in CKD stage 3 and three-quarters with CKD stage 4/5 had ≥2 markers of frailty present. Worryingly, markers of frailty were present early in the disease process with over 1/3 of patients with CKD stage 1/2 classified as frail by this modified phenotype model. Age, sex, ethnicity, and number of comorbidities were associated with the likelihood of being frail. It is important that healthcare providers actively attempt to identify patients at risk of frailty to ensure appropriate interventions addressing risk factors that may exacerbate the progression of frailty can be implemented.