Identifying subgroups of renal function trajectories.

Area of Science:

• Nephrology
• Biostatistics
• Chronic Kidney Disease Research

Background:

• Chronic kidney disease (CKD) patient renal function can exhibit diverse progression patterns.
• Understanding these trajectories is crucial for personalized patient management and treatment strategies.

Purpose of the Study:

• To evaluate the efficacy of latent class linear mixed modeling (LCMM) in identifying distinct renal function trajectory subgroups within a CKD cohort.
• To illustrate the clinical relevance of identified subgroups in a multicenter hospital-based CKD patient population.

Main Methods:

• Analysis of the NephroTest cohort data, including 1967 CKD patients with baseline glomerular filtration rate (GFR) measurements.
• Utilized LCMM to identify subgroups of GFR trajectories, comparing baseline patient characteristics across identified subgroups.
• Employed both measured GFR (mGFR) via 51Cr-EDTA renal clearance and estimated GFR (eGFR) using the CKD-EPI equation.

Main Results:

• Five distinct mGFR trajectory classes were identified.
• Three classes showed slow linear mGFR decline; two classes exhibited high baseline mGFR with either nonlinear decline (11 patients) or nonlinear improvement (94 patients).
• Elevated proteinuria and blood pressure at baseline were associated with severely decreased or rapidly declining mGFR trajectories. Similar patterns were observed when using eGFR.

Conclusions:

• LCMM successfully identified five clinically relevant subgroups of renal function trajectories in the CKD cohort.
• This methodology can be applied to other CKD cohorts for enhanced characterization of disease progression profiles.
• LCMM facilitates the investigation of specific risk factors linked to distinct CKD progression patterns.