Kidney Week

Abstract: TH-PO1098

Revisiting the Lipid Paradox in Nondialysis-Dependent CKD and Mediation Effect of BMI on the Association of LDL Cholesterol and All-Cause Mortality

Session Information

• CKD: Therapies, Innovations, and Insights
  November 06, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2302 CKD (Non-Dialysis): Clinical, Outcomes, and Trials

Authors

• Takahashi, Rina, The Lundquist Institute, Torrance, California, United States

Rina Takahashi, MD, PhD

• Tran, Diana, The Lundquist Institute, Torrance, California, United States

Diana Tran, MPH

• Cohen-Hagai, Keren, Meir Medical Center, Kefar Sava, Center District, Israel

Keren Cohen-Hagai, MD

• Rhee, Connie, VA Greater Los Angeles Healthcare System, Los Angeles, California, United States

Connie Rhee, MD, MS

• Sumida, Keiichi, The University of Tennessee Health Science Center, Memphis, Tennessee, United States

Keiichi Sumida, MD, PhD, MPH, FASN

• Kovesdy, Csaba P., The University of Tennessee Health Science Center, Memphis, Tennessee, United States

Csaba P. Kovesdy, MD, FASN

• Kalantar-Zadeh, Kamyar, The Lundquist Institute, Torrance, California, United States

Kamyar Kalantar-Zadeh, MD, DrMed, PhD, MPH, FASN

Background

There are mixed reports on the lipid paradox in CKD, including whether a reverse association truly exists between LDL-C and mortality. The mortality risk associated with LDL-C may be confounded or mediated by factors such as BMI. We hypothesized that the observed lipid paradox in CKD might be mediated by BMI, given the obesity paradox.

Methods

In a national TRI-CKD cohort of 30,559 U.S. veterans with incident CKD (2004-2018), defined as two outpatient eGFR <60 mL/min/1.73m², ≥90 days apart, both ≥25% below baseline. Patients were categorized into five LDL-C groups: <70, 70–<100, 100–<130, 130–<160, and ≥160 mg/dL. Cox proportional hazards models examined the association between LDL-C and mortality with progressive adjustment for demographics, comorbidities, labs including eGFR, and medications. Mediation analysis assessed the role of BMI.

Results

Patients with higher LDL-C tended to be younger (71.5 to 64.3), while mean eGFR remained stable (52.5 to 55.1 mL/min/1.73m²). LDL-C was weakly but significantly associated with BMI (β = –0.11, p = 0.001), but the effect size was negligible (R² = 0.0003). LDL-C showed an inverse linear association with mortality in unadjusted models, consistent with the lipid paradox. With progressive adjustment, the association became U-shaped in the fully adjusted model. Mediation analysis showed no significant mediation effect of BMI (β_indirect < 0.01).

Conclusion

In this large national cohort of U.S. veterans with incident CKD, LDL-C exhibited a U-shaped association with all-cause mortality after comprehensive adjustment for potential confounders. Although LDL-C was modestly correlated with BMI, mediation analysis indicated no significant mediation effect of BMI on the LDL–mortality relationship. These findings suggest that the lipid paradox in CKD is not explained by the obesity paradox, and that LDL-C may instead be independently associated with mortality, specifically in non-dialysis CKD.

Funding

• Veterans Affairs Support

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.20257nebrb82