Non-obese non-alcoholic fatty liver disease and the risk of chronic kidney disease: a systematic review and meta-analysis

Abstract

Background: Data on risk of developing chronic kidney disease (CKD) between non-obese and obese non-alcoholic fatty liver disease (NAFLD) patients are limited. We aimed to reveal the risk difference of incident CKD between non-obese and obese NAFLD patients.

Methods: We searched PubMed, Embase, and Web of Science databases for studies which reported the incidence of CKD in non-obese and obese NAFLD from inception to 10 March 2024. The primary and secondary outcomes were pooled. Subgroup analysis was used to examine the heterogeneity.

Results: A total of 15 studies were incorporated. The incidence of CKD in non-obese and obese NAFLD were 1,450/38,720 (3.74%) and 3,067/84,154 (3.64%), respectively. Non-obese NAFLD patients had a comparable risk of CKD as obese NAFLD (odds ratio [OR] 0.92, 95% confidence interval [95% CI] [0.72-1.19], I² = 88%). No differences in estimated glomerular filtration rate and serum creatinine between non-obese and obese NAFLD were found. The mean differences (MD) and 95% CI were 0.01 [-0.02 to 0.04] and 0.50 [-0.90 to 1.90], respectively. In subgroup analyses, non-obese NAFLD had higher eGFR when diagnosed with ultrasound (MD 1.45, 95% CI [0.11-2.79], I² = 21%). Non-obese NAFLD had higher creatinine in non-Asian (MD 0.06, 95% CI [0.01-0.11], I² = 55%) and when taking BMI > 30 as the criterion for obesity (MD 0.06, 95% CI [0.00-0.12], I² = 76%). The occurrence of CKD did not differ when non-obese NAFLD were categorized into overweight and normal-weight types.

Conclusions: Non-obese NAFLD patients experienced the same risk of CKD compared to obese NAFLD.

Keywords: CKD; Creatinine; NAFLD; Non-obese; eGFR.

Figure 1. Workflow diagram of study selection.

Figure 2. Forest plot and pooled estimates of the effect of non-obese NAFLD on the risk of incident CKD compared to obese NAFLD (Akahane et al., 2020; Chon et al., 2020; Hu et al., 2022a, 2022b; Kwon et al., 2023; Liu et al., 2021; Mikolasevic et al., 2020; Nabi et al., 2023).

Figure 3. Forest plot and pooled estimates of the effect of non-obese NAFLD on the risk of incident CKD compared to obese NAFLD stratified by region (A), the methodology used for the diagnosis of NAFLD (B), type of study (C), the cut-off value of non-obese NAFLD (D), quality of studies (E) (Akahane et al., 2020; Chon et al., 2020; Hu et al., 2022a, 2022b; Kwon et al., 2023; Nabi et al., 2023; Iwaki et al., 2022; Kim et al., 2022; Yang et al., 2018; Leung et al., 2017; Ampuero et al., 2018; Mikolasevic et al., 2020; Ragab et al., 2021).

Figure 4. Forest plot and pooled estimates of the effect of non-obese NAFLD on the level of eGFR compared to obese NAFLD.

eGFR, estimated glomerular filtration rate (Akahane et al., 2020; Hu et al., 2022b; Kim et al., 2022; Chon et al., 2020).

Figure 5. Forest plot and pooled estimates of the effect of non-obese NAFLD on the level of eGFR compared to obese NAFLD stratified by the methodology used for the diagnosis of NAFLD (Akahane et al., 2020; Kim et al., 2022; Chon et al., 2020).

Figure 6. Forest plot and pooled estimates of the effect of non-obese NAFLD on the level of serum creatine compared to obese NAFLD (Iwaki et al., 2022; Kim et al., 2022; Yang et al., 2018; Leung et al., 2017; Ampuero et al., 2018; Mikolasevic et al., 2020; Ragab et al., 2021).

Figure 7. Forest plot and pooled estimates of the effect of non-obese NAFLD on the level of serum creatine compared to obese NAFLD stratified by the cut-off value of non-obese NAFLD (A), region (B), the methodology used for the diagnosis of NAFLD (C) (Iwaki et al., 2022; Kim et al., 2022; Yang et al., 2018; Leung et al., 2017; Ampuero et al., 2018; Mikolasevic et al., 2020; Ragab et al., 2021).