Optimizing diabetic kidney disease animal models: Insights from a meta-analytic approach

Diabetic kidney disease (DKD) is a prevalent complication of diabetes, often leading to end-stage renal disease. Animal models have been widely used to study the pathogenesis of DKD and evaluate potential therapies. However, current animal models often fail to fully capture the pathological characteristics of renal injury observed in clinical patients with DKD. Additionally, modeling DKD is often a time-consuming, costly, and labor-intensive process. The current review aims to summarize modeling strategies in the establishment of DKD animal models by utilizing meta-analysis related methods and to aid in the optimization of these models for future research. A total of 1215 articles were retrieved with the keywords of "diabetic kidney disease" and "animal experiment" in the past 10 years. Following screening, 84 articles were selected for inclusion in the meta-analysis. Review manager 5.4.1 was employed to analyze the changes in blood glucose, glycosylated hemoglobin, total cholesterol, triglyceride, serum creatinine, blood urea nitrogen, and urinary albumin excretion rate in each model. Renal lesions shown in different models that were not suitable to be included in the meta-analysis were also extensively discussed. The above analysis suggested that combining various stimuli or introducing additional renal injuries to current models would be a promising avenue to overcome existing challenges and limitations. In conclusion, our review article provides an in-depth analysis of the limitations in current DKD animal models and proposes strategies for improving the accuracy and reliability of these models that will inspire future research efforts in the DKD research field.