Prevalence and associated factors of frailty in patients with chronic kidney disease: a cross-sectional analysis of PEAKING study

This is a cross-sectional analysis of adult patients from the Physical Evaluation and Adverse outcomes for patients with chronic Kidney disease IN Guangdong, China (PEAKING) study. The results of this study are reported according to STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines [18].

The PEAKING study is an open cohort study established in 2017 and aims to investigate the level of physical activity and adverse outcomes in patients with CKD. Patients registered in the chronic disease management clinic of Guangdong Provincial Hospital of Chinese Medicine (A tertiary hospital of the region, located in Guangzhou city, the capital city of Guangdong province, China) were invited if they met the following inclusion criteria: (1) age over 18 years; (2) diagnosed as CKD with estimated glomerular filtration rate (eGFR) less than 60 ml/min/1.73 m² or abnormal kidney biomarkers (such as proteinuria, hematuria, etc.) for more than 90 days [19]. Patients were not invited if they met the following exclusion criteria: (1) has received or expected to receive kidney replacement therapy within 1 year; (2) pregnant or lactating women or those planning pregnancy within 1 year; (3) acute myocardial infarction or acute cerebrovascular event, acute obstructive nephropathy requiring surgery within 3 months; (4) severe arrhythmia or heart failure (New York Heart Association class grade III or above) which could not be controlled by medication; (5) active malignant tumor, decompensated cirrhosis or diseases of the hematopoietic system; (6) serious mental illness, or unable to cooperate with the investigation and treatment due to other reasons; (7) physically disability to perform physical activities;

In our study, participants were included in the analysis if they met the following: (1) complete baseline demographics and laboratory test data; (2) valid accelerometer data. The study was approved by the Ethics Committee of Guangdong Provincial Hospital of Chinese Medicine (B2015-152-02), and all participants gave informed consent.

FRAIL scale (FS) was used to evaluate the degree of frailty at baseline as recommended by the International Conference of Frailty and Sarcopenia Research (ICFSR) [20]. FS has been used in patients with all stages of CKD and demonstrated good reliability and validity [14, 21]. FS is a self-report tool and derives its name from the five domains: fatigue, resistance, ambulation, illness, and loss of weight [22]. FS score ranges from zero to 5 points. One domain represents 1 point. Those with a score more than 2 points are deemed to be with frail, 1 to 2 points are considered as prefrail stage, and those with zero points are classified as robust (Supplement Table 1).

Upon enrollment, a case report form (CRF) was used to collect baseline data of participants, including demographic data (sex, age, marital status, educational level, health insurance, working status, smoking and alcohol drinking habit). Comorbidities such as hypertension, diabetes, cardiovascular disease, stroke, chronic obstructive pulmonary disease, bronchial asthma, arthritis, malignant tumor, and osteoporosis were recorded on the CRF allowing calculation of Charlson comorbidity index (CCI) [23]. Information about the kidney disease was collected as well, including primary disease of CKD, disease duration, and kidney biopsy report if available.

Participants underwent physical examinations at baseline for anthropometric parameters including body weight and height, and waist circumference. Body mass index (BMI) was calculated accordingly and categorized as severely underweight (< 16.5 kg/m²), underweight (< 18.5 kg/m²), normal (18.5–22.9 kg/m²), and overweight (> 23 kg/m²) [24]. In addition, their physical performance and physical activity were measured using established approaches: handgrip strength was measured as a proxy for upper limb performance using a digital handgrip dynamometer (EH101, CAMRY Sensun Weighing Apparatus Group Ltd, Guangdong, China), calculated from the largest readings of three measurements using participants’ dominant hand; the ActiGraph GT3X + accelerometers (ActiGraph, LLC Pensacola, FL, USA) were employed to evaluate daily step counts as a reliable and objective measure of physical activity. It incorporates a triaxial accelerometer that detects and records acceleration forces in different directions. These forces are then converted into activity counts, which reflect the intensity and duration of movement. This sophisticated device has gained widespread adoption by over 1500 colleges and institutions in more than 65 countries or regions worldwide [25]. They serve as precise instruments for capturing and assessing multiple facets of physical activity, encompassing not only daily step counts but also capturing the nuances of intensity, duration, and frequency of physical activity, sedentary behavior, and energy expenditure. Subjects were required to wear an accelerometer on the right hip in daytime for 9 consecutive days [26]. Data were considered valid if a participant had at least 3 days (including one non-working day) of at least 10 hours per day recorded [27]. We selected an epoch length of 60 seconds [28]. A non-wear-time was defined as an interval of at least 60 min of zero activity counts [28]. At the end of the measurement period, the accelerometer data were retrieved through the ActiLife software. Daily step counts were defined as the average of daily step counts from valid wearing days.

Finally, participants were required to have laboratory tests at baseline including the number of white blood cell, percentage of neutrophils, hemoglobin, estimated glomerular filtration rate (eGFR) estimated by CKD-EPI Creatinine Eq. (2021) [29], spot urine protein–creatinine ratio, serum potassium, total serum calcium, serum albumin, serum uric acid, serum triglyceride, and serum total cholesterol.

Data were described as percentages (%) for categorical variables and mean and standard deviation (M ± SD) or median and interquartile range [M(IQR)] for continuous variables. To test the statistical difference among groups (robust, prefrail and frail), univariate analyses were conducted using Chi-squared test for categorical variables and analysis of variance (with post hoc Tukey analysis) or Kruskal–Wallis test (with Mann–Whitney U test for post hoc analysis) for continuous variables as appropriate. Variables that showed statistical significance of P < 0.1 in the univariate analyses were included in the multivariate logistic regression analysis, which estimated the odds ratio (OR) and 95% confidence interval (CI) with the robust group as reference group. We further treated FS score as a continuous dependent variable and examined associated factors of FS score in multiple linear regression analysis with stepwise backward variable selection, incorporating the same panels of variables in the aforementioned logistic regression. A two-tailed examination with P value less than 0.05 was used as an indication of statistical significance. All analyses were conducted using the Stata 15.0 and the free statistics analysis platform.

Higher BMI (OR = 1.25, 95% CI 1.10–1.42, P = 0.001) was associated with a higher likelihood of being prefrail; an increase of 5 units eGFR (OR = 0.92, 95% CI 0.87–0.99, P = 0.008) was inversely associated with being prefrail (Fig. 2).

An increase of 100 steps per day (OR = 0.95, 95% CI 0.91–0.99, P = 0.01) and an increase of 5 units eGFR (OR = 0.82, 95% CI 0.68–0.99, P = 0.045) were inversely associated with being frail, and higher BMI was associated with a higher likelihood of being frail (OR = 1.52, 95% CI 1.11–2.06, P = 0.008) (Fig. 3).

In the multiple linear regression analysis, with FS scores as the dependent variable, daily step counts (β = − 0.008, P < 0.001) and eGFR (β = − 0.005, P < 0.013) were associated with a lower severity of frailty, BMI (β = 0.074, P = 0.001) was associated with a higher severity of frailty among patients with ND-CKD.