Association of remnant cholesterol with kidney outcomes in adults with type 2 diabetes: a systematic review and meta-analysis

Figures

• 

  Figure 1. PRISMA 2020 flow diagram of study selection.

• 

  Figure 2. Exploratory overall association between remnant cholesterol and kidney outcomes in adults with type 2 diabetes. Squares and horizontal lines indicate study-specific estimates and 95% confidence intervals, respectively; the diamond indicates the pooled estimate from the random-effects model. The dashed vertical line indicates the null value of 1. CI, confidence interval; DKD, diabetic kidney disease; ESKD, end-stage kidney disease; HR, hazard ratio; OR, odds ratio.

• 

  Figure 3. Longitudinal association between remnant cholesterol and kidney outcomes in adults with type 2 diabetes. The left panel presents the random-effects model based on 12 independent cohort estimates. The right panel presents leave-one-out sensitivity analyses in which each corresponding estimate was omitted in turn and the model was refitted using the same analytical method. Squares and horizontal lines represent individual estimates and 95% confidence intervals; diamonds represent pooled estimates. The dashed vertical line indicates the null value of 1. CI, confidence interval; DKD, diabetic kidney disease; ESKD, end-stage kidney disease; HR, hazard ratio; RC, remnant cholesterol.

• 

  Figure 4. Outcome-specific longitudinal associations between remnant cholesterol and kidney outcomes in adults with type 2 diabetes. Separate random-effects analyses are presented for incident diabetic kidney disease, diabetic kidney disease progression, kidney failure or end-stage kidney disease, and incident albuminuria. For incident diabetic kidney disease, the yellow diamond represents the analysis of all eligible estimates and the blue diamond represents the sensitivity analysis restricted to studies scoring 8–9 stars on the Newcastle–Ottawa Scale. Incident albuminuria was reported by one eligible study and was therefore not pooled. Squares and horizontal lines indicate study-specific hazard ratios and 95% confidence intervals; diamonds indicate pooled estimates. CI, confidence interval; DKD, diabetic kidney disease; ESKD, end-stage kidney disease; HR, hazard ratio; NOS, Newcastle–Ottawa Scale; RC, remnant cholesterol.

• 

  Figure 5. Kidney outcomes according to baseline kidney status. (A) Incident DKD among participants without baseline DKD. (B) Outcomes among participants with established DKD: DKD progression (B1) and kidney failure/ESKD (B2). (C) Kidney outcomes in cohorts with mixed baseline kidney status. *For Mori et al., the estimate in panel A was restricted to participants with an eGFR ≥60 mL/min/1.73 m² and an ACR <30 mg/g at baseline; incident DKD was operationalized as a sustained ≥40% decline in eGFR for ≥90 days. In panel B1, three mutually exclusive baseline DKD strata were combined using inverse-variance weighting before inclusion. Squares and horizontal lines indicate study-specific hazard ratios and 95% confidence intervals; diamonds indicate pooled estimates. CI, confidence interval; DKD, diabetic kidney disease; ESKD, end-stage kidney disease; HR, hazard ratio; RC, remnant cholesterol.

• 

  Figure 6. Continuous and exploratory dose–response associations between remnant cholesterol and incident diabetic kidney disease. (A) Forest plot of direct continuous estimates from two articles comprising three independent cohorts, standardized to a 0.1 mmol/L increase in remnant cholesterol. (B) Exploratory dose–response analysis based on five cohorts from four articles. The left plot shows study-specific categorical estimates, with representative doses derived from reported means, medians, or category boundaries. The right plot shows the pooled three-knot restricted cubic spline and its 95% confidence interval, using an RC concentration of 0.40 mmol/L as the reference. CI, confidence interval; DKD, diabetic kidney disease; HR, hazard ratio; RC, remnant cholesterol.

• 

  Figure 7. Cross-sectional association between remnant cholesterol and prevalent or early diabetic kidney disease. The left panel presents the random-effects model based on six cross-sectional studies. The right panel presents leave-one-out sensitivity analyses in which each study was omitted in turn and the model was refitted using the same analytical method. Squares and horizontal lines represent study-specific odds ratios and 95% confidence intervals; diamonds represent pooled estimates. The dashed vertical line indicates the null value of 1. CI, confidence interval; DKD, diabetic kidney disease; OR, odds ratio; RC, remnant cholesterol.

• 

  Figure 8. Cardiometabolic adjustment models for the association between remnant cholesterol and kidney outcomes. (A) Analyses restricted to estimates adjusted for blood pressure or hypertension, glycemia, and adiposity, presented for the overall study, longitudinal cohorts, and cross-sectional studies. (B) Analyses restricted to estimates adjusted for blood pressure or hypertension, glycemia, adiposity and diabetes duration, presented for the overall study, longitudinal cohorts, and cross-sectional studies. Squares and horizontal lines indicate study-specific estimates and 95% confidence intervals; diamonds indicate pooled estimates. CI, confidence interval; DKD, diabetic kidney disease; ESKD, end-stage kidney disease; HR, hazard ratio; OR, odds ratio; RC, remnant cholesterol.

• 

  Figure 9. Baseline kidney markers according to remnant cholesterol category. (A) Mean differences in baseline estimated glomerular filtration rate between the highest and lowest RC categories. (B) Ratios of baseline albumin-to-creatinine ratio medians between the highest and lowest RC categories. Squares and horizontal lines indicate study-specific estimates and 95% confidence intervals; diamonds indicate pooled estimates. ACR, albumin-to-creatinine ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; MD, mean difference; RC, remnant cholesterol.