In the present study, we used CT images to assess body composition at the L3 level in a cohort of patients with CKD stages 3–5 to explore factors associated with two CT-derived markers of myosteatosis, %IMAT and average muscle attenuation. This approach allowed us to show that myosteatosis in this cohort of carefully phenotyped patients was associated with several nutritional and metabolic parameters including older age, increased adiposity, and obesity-related metabolic alterations. Furthermore, higher %IMAT and lower attenuation were associated with a higher mortality risk.

Similar findings have been reported in the general population. In a large well-controlled cohort of healthy subjects, Delmonico et al. [24] were the first to demonstrate the increase in myosteatosis with age, a phenomenon that occurred regardless of body weight and muscle mass gain or loss. In the context of CKD, it is probable that factors beyond age contribute to the development of myosteatosis. Evidence shows that patients with advanced CKD on dialysis have increased myosteatosis, as assessed by MRI, compared to age-matched controls [9]. Additionally, Kim et al. [10] identified myosteatosis, assessed by muscle density, as an independent predictor of CKD progression in 149 patients, with a median follow-up of 7.5 years. In our study, we did not find any association between creatinine clearance and myosteatosis. However, we did not evaluate changes in renal function over time, which limits our ability to confirm previous findings. Nevertheless, individuals with CKD often have several conditions already associated with myosteatosis. Available evidence suggests that aging, poor nutritional status, inflammation, oxidative stress, mitochondrial dysfunction, and insulin resistance might act synergistically in the development of myosteatosis [25]. Obesity also seems to be a factor associated with myosteatosis in patients with CKD. In fact, our group has recently shown the association between myosteatosis and other markers of body composition in hemodialysis patients [17]. In the aforementioned work, among four body composition phenotypes (normal, sarcopenia only, obesity only, and sarcopenic obesity), both groups with obesity had the highest prevalence of myosteatosis in comparison to the group with normal body composition [17]. Research involving other patient populations, as well as healthy individuals, further supports a link between obesity and its associated metabolic disturbances and the occurrence of myosteatosis [5, 6, 26]. One possible explanation for this association is that with increased adiposity, adipocytes may exceed their fat-storage capacity, resulting in the accumulation of ectopic fat in lean tissues, including skeletal muscle, liver, and pancreas [3].

Myosteatosis by CT is currently evaluated by assessing IMAT or muscle attenuation. While %IMAT reflects only the intermuscular adipose tissue (i.e., adipose tissue in between muscle fibers and muscle groups), the average muscle attenuation is a measure of muscle density and reflects adipose tissue within skeletal muscle fibers and muscle cells. Lower values of muscle attenuation reflect a greater amount of intramuscular adipose tissue, which will consequently influence muscle density. In fact, a study comparing CT and MR, has shown that, intramyocellular lipid stores rather than extramyocellular lipid stores, better reflected CT-assessed muscle attenuation [27]. Despite limitations in assessing IMAT through a single cross-sectional CT area, the differentiation between adipose tissue outside and inside the fibers may be important because they might have different effects on muscle and metabolic health [3]. A study using MR suggested that intramyocellular lipids rather than extramyocellular lipids influences insulin resistance [28]. However, in studies using CT, both parameters have been shown to be related to inflammatory markers [12, 25], and insulin resistance [29]. In our study, CRP and HOMA index (borderline significance) were higher in the lower tertile of muscle attenuation and, in the case of HOMA index, it was higher in the third tertile of %IMAT. At multivariable linear regression, CRP remained independently associated with muscle attenuation. More in general, CCI was associated with both parameters at univariate analysis, and with %IMAT in multivariable linear regression, while metabolic syndrome remained an independent predictor of both markers of myosteatosis in multivariable linear regression analysis. Similar observations were found in studies involving subjects with type-2 diabetes [5, 6]. Particularly, a recent systematic review showed increased presence of myosteatosis (defined both by IMAT and muscle density) in subjects with diabetes, and its association with insulin resistance [5]. Low muscle density has also been associated with low-grade inflammation [26]. In fact, ectopic adipose tissue in muscles secretes cytokines, leading to localized inflammation [2]. However, it is still unclear whether myosteatosis is solely a marker of metabolic derangements, or if it plays a role in the development of insulin resistance and inflammation.

The association between myosteatosis and health outcomes has also been investigated in relation to increased cardiovascular risk and mortality. Obesity, diabetes, inflammation, and dyslipidemia are known cardiovascular risk factors. Myosteatosis, too, is linked with such risks; studies show it is associated with increased cardiovascular mortality in older men [30], and with higher CAC-scores [31]. Additionally, myosteatosis predicts cardiovascular events and mortality in hemodialysis patients [32], unlike low skeletal muscle mass [33]. In the present study, CAC-score (a well-established marker of increased cardiovascular risk) was associated with both investigated parameters of myosteatosis. At adjusted analysis, all-cause mortality risk was also increased for patients with higher %IMAT and lower muscle attenuation, both when parameters were considered as continuous and categorical variables, but only %IMAT was associated with cardiovascular mortality. Considering the low number of cardiovascular events in non-dialysis CKD cohorts, a larger cohort may be needed to confirm the association between cardiovascular mortality and myosteatosis parameters.

Our study has limitations. Firstly, in order to determine the severity of myosteatosis, we divided the cohort into tertiles, which was arbitrary but allowed the identification of metabolic differences between the worst tertiles and the best tertiles. Secondly, because this is a secondary analysis of a larger study and includes a reduced number of patients from the original cohort, coupled with the intrinsic characteristics of patients with CKD, there might be insufficient statistical power to discern certain differences in hard outcomes. Nevertheless, we were able to show important associations between myosteatosis parameters, metabolic derangements and all-cause mortality in a group of patients with non-dialysis CKD. Finally, our study used CT to evaluate myosteatosis, which, despite being considered a gold standard, is not without limitations. CT-assessed muscle density could be affected by fluid retention. Fluid retention is usually a complication in advanced stages of CKD, however, patients with non-dialysis CKD seldom have fluid overload in an extent that could influence the abdominal muscle density.

In conclusion, in patients with CKD the extent of CT-assessed abdominal myosteatosis was associated with higher age, abdominal adiposity, and markers of metabolic dysfunction. Moreover, in the adjusted analysis, higher %IMAT and lower attenuation were associated with a higher mortality risk.