Kidney Week

Abstract: TH-PO0605

Target Trial Emulation Evaluating Effects of Niaoduqing Granules on Mortality in Stage 3-4 CKD: A Real-World Cohort Study of 18,315 Patients

Session Information

• CKD: From Global Burden to Precision Care - Epidemiology, Demographics, Geographies, and Risk Factors
  October 22, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2202 CKD (Non-Dialysis): Clinical, Outcomes, and Trials

Authors

• Li, Xinyan, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Xinyan Li, MD

• Guo, Wen-Cong, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Wen-Cong Guo, MD

• Wang, Bin, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Bin Wang, MD, PhD

• Wang, Fengmei, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Fengmei Wang, PhD

• Wen, Yi, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Yi Wen, PhD

• Liu, Bi-Cheng, Southeast University Zhongda Hospital, Nanjing, Jiangsu, China

Bi-Cheng Liu, MD, PhD

Background

Niaoduqing (NDQ) granules, a Chinese herbal formula containing rhubarb, has been widely used in China for chronic kidney disease (CKD) for three decades, with prior studies showing benefits in improving renal function and slowing progression. However, robust evidence on its long-term efficacy, especially on mortality, remains lacking. This study evaluated NDQ's association with all-cause and cardiovascular (CVD) mortality in stage 3-4 CKD via target trial emulation.

Methods

Using the China Renal Data System (CRDS) from 32 tertiary hospitals (2004-2022), we identified new NDQ users (6-month washout) and non-users with stage 3-4 CKD. Primary outcomes were all-cause and CVD mortality, linked to the Chinese Center for Disease Control and Prevention (CDC) death registry. Overlap weighting (OW) balanced baseline covariates to emulate randomization, and weighted Cox models estimated HRs.

Results

Of 18,315 patients (4,917 with NDQ, 13,398 without NDQ), baseline characteristics were well-balanced after OW (all SMDs <0.01). NDQ reduced all-cause mortality (HR = 0.627; p < 0.001) and CVD mortality (HR = 0.561; p < 0.001). The protective effects were consistent across almost all subgroups, especially in patients aged ≤60 years, CKD stage 3a, and those on RAASi/SGLT2i. Sensitivity analyses, including restricting to long-term follow-up, excluding early deaths, and avoiding potential COVID-19 impact, confirmed robustness.

Conclusion

In this target trial emulation using a large real-world cohort with OW, initiation of NDQ significantly reduced all-cause and cardiovascular mortality in stage 3-4 CKD, supporting its role as a beneficial adjunctive therapy.

Figure 1. Study design and weighted cumulative incidence of mortality outcomes.

Figure 2. Subgroup analyses for all-cause and cardiovascular mortality.

Funding

• Government Support – Non-U.S.