Background: Sarcopenia is common in chronic kidney disease (CKD) and linked to higher mortality, but identifying those at risk remains challenging. Indices combining serum creatinine and cystatin C (eGFRratio and eGFRdifference) have been studied, but have tended to perform worse in those with CKD. This study aimed to examine the relationship of Creatinine Muscle Index (CMI), an estimate of glomerular filtration of creatinine, with sarcopenia and mortality in a non-dialysis CKD population.
Methods and findings: NURTuRE-CKD is a prospective, multicentre cohort study of people with non-dialysis CKD in the UK. Two thousand nine hundred ninety-six individuals were enrolled between July 2017 and September 2019. Cystatin C measurements were available in 2,930 adults. CMI (mg/day) was calculated as eGFR cystatin C × serum creatinine concentration. The relationships between CMI and: (1) probable sarcopenia (defined as the best hand grip strength of <27 kg for males and <16 kg for females); (2) individual muscle function measures including hand grip strength (kg) and timed get-up-and-go (TUG) (seconds); (3) all-cause mortality were assessed using Spearman's correlation, logistic regression, and Cox proportional hazards models, stratified by sex and adjusted for age, ethnicity, body mass index, smoking status, Charlson Comorbidity Index, urine albumin-to-creatinine ratio, and C-reactive protein. TUG test is the time taken to stand from a chair, walk 3 m, turn, return, and sit down. Among 1,723 males and 1,207 females, the median (IQR) age was 66 (53-74) years, and the median eGFRcreatinine was 34 (24-47) ml/min/1.73m2. A total of 806 participants (27.5%) had probable sarcopenia, and over a median follow-up period of 50 (41 to 56) months, 527 (18%) died. The adjusted OR for probable sarcopenia per 100 mg/day increase in CMI was OR 0.72 (95% CI 0.67, 0.78 p value <0.001) in males and OR 0.81 (95% CI 0.73, 0.89 p value < 0.001) in females. CMI correlated positively with grip strength (ρ = 0.47 [0.43, 0.50] and 0.45 [0.40, 0.49]) and negatively with TUG (ρ = -0.37 [-0.41, -0.32] and -0.44 [-0.49, -0.40]) in males and females, respectively. In adjusted models, the HR for mortality per 100 mg/day increase in CMI was HR 0.85 (95% CI 0.78, 0.90 p value < 0.001) in males and HR 0.77 (95% CI 0.67, 0.87 p value < 0.001) in females. In males and females, respectively, the C-index of CMI for probable sarcopenia (0.73 and 0.71) and mortality (0.70 and 0.76) was higher than that of the eGFR ratio (probable sarcopenia: 0.64 and 0.61; mortality: 0.60 and 0.65; all p < 0.001) and the eGFR difference (probable sarcopenia: 0.59 and 0.57; mortality: 0.56 and 0.59; all p < 0.001) Limitations include the observational design, the assessment of muscle function without direct measurement of muscle mass, and limited generalisability to CKD populations not followed in secondary care.
Conclusions: In persons with CKD, CMI-a biomarker reflecting creatinine generation-was independently associated with muscle function and mortality, supporting its utility in populations with reduced kidney function.
