Association of chronic kidney disease with muscle deficits in children

Abstract

The effect of chronic kidney disease (CKD) on muscle mass in children, independent of poor growth and delayed maturation, is not well understood. We sought to characterize whole body and regional lean mass (LM) and fat mass (FM) in children and adolescents with CKD and to identify correlates of LM deficits in CKD. We estimated LM and FM from dual energy x-ray absorptiometry scans in 143 children with CKD and 958 controls at two pediatric centers. We expressed whole body, trunk, and leg values of LM and FM as Z-scores relative to height, sitting height, and leg length, respectively, using the controls as the reference. We used multivariable regression models to compare Z-scores in CKD and controls, adjusted for age and maturation, and to identify correlates of LM Z-scores in CKD. Greater CKD severity associated with greater leg LM deficits. Compared with controls, leg LM Z-scores were similar in CKD stages 2 to 3 (difference: 0.02 [95% CI: -0.20, 0.24]; P = 0.8), but were lower in CKD stages 4 to 5 (-0.41 [-0.66, -0.15]; P = 0.002) and dialysis (-1.03 [-1.33, -0.74]; P < 0.0001). Among CKD participants, growth hormone therapy associated with greater leg LM Z-score (0.58 [0.03, 1.13]; P = 0.04), adjusted for CKD severity. Serum albumin, bicarbonate, and markers of inflammation did not associate with LM Z-scores. CKD associated with greater trunk LM and FM, variable whole body LM, and normal leg FM, compared with controls. In conclusion, advanced CKD associates with significant deficits in leg lean mass, indicating skeletal muscle wasting. These data call for prospective studies of interventions to improve muscle mass among children with CKD.

Figure 1.

Whole body lean mass deficits are evident only in dialysis participants because of elevated trunk lean mass in CKD; muscle deficits, represented by leg lean mass deficits proportional to the severity of CKD, are evident in CKD 4 to 5 and dialysis. The figure indicates differences in whole body, trunk, and leg lean mass-for-height Z-scores between CKD participants and controls. Differences in whole body, trunk, and leg lean mass Z-scores adjusted for sex, race, age, height (sitting height or leg length), pubertal status, fat mass Z-score, and study site are presented.

Figure 2.

Whole body and leg fat mass do not differ significantly between CKD participants and controls, but trunk fat mass is significantly higher among all CKD participants. The figure shows differences in whole body, trunk, and leg fat mass-for-height Z-scores between CKD participants and controls. Differences in whole body, trunk, and leg fat mass Z-scores adjusted for sex, race, age, height (sitting height or leg length), pubertal status, and study site are presented. Central distribution of adiposity was evident in all CKD groups.

Figure 3.

An example of one of the body composition centile curves constructed to allow generation of z-scores in controls and CKD participants (DXA leg LM for leg length centile curves for non–African-American girls). The leg LM-for-leg length centile curves constructed using data from healthy non-African-American girls are shown, with 3rd, 10th, 25th, 50th, 75th, 90th, and 97th percentile lines. In addition, leg LM is plotted relative to leg length for each individual non-African-American female CKD participant. CKD 2 to 3 participants are represented with circles, CKD 4 to 5 participants with squares, and dialysis participants with triangles. Similar centile curves were created for each sex/race category for each of the considered body composition measures. It should be noted that this graph represents an unadjusted comparison between CKD participants and controls, and that the severity of deficits is likely to be underestimated from the appearance of the graph. On average, CKD participants will be older than controls with the same leg length.