Utility of serum 2-microglobulin for prediction of kidney outcome among patients with biopsy-proven diabetic nephropathy

Aim: To examine whether serum beta 2-microglobulin (beta 2-MG) could improve the prediction performance for kidney failure with replacement therapy (KFRT) among patients with diabetic nephropathy (DN).Methods: Patients with biopsy-proven DN at Nara Medical University Hospital were included. The exposure of interest was log-transformed serum beta 2-MG levels measured at kidney biopsy. The outcome variable was KFRT. Multivariable Cox regression models and competing-risk regression models, with all-cause mortality as a competing event, were performed. Model fit by adding serum beta 2-MG levels was calculated using the Akaike information criterion (AIC). The net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indexes were used to evaluate the improvement of predictive performance for 5-year cumulative incidence of KFRT by serum beta 2-MG levels.Results: Among 408 patients, 99 developed KFRT during a median follow-up period of 6.7 years. A higher serum beta 2-MG level (1-unit increase in log-transformed serum beta 2-MG level) was associated with a higher incidence of KFRT, even after adjustments for previously known clinical and histological risk factors (hazard ratio [95% confidence interval {CI}]: 3.30 [1.57-6.94] and subdistribution hazard ratio [95% CI]: 3.07 [1.55-6.06]). The addition of log-transformed serum beta 2-MG level reduced AIC and improved the prediction of KFRT (NRI and IDI: 0.32 [0.09-0.54] and 0.03 [0.01-0.56], respectively).Conclusions: Among patients with biopsy-proven DN, serum beta 2-MG was an independent predictor of KFRT and improved prediction performance. In addition to serum creatinine, serum beta 2-MG should probably be measured for DN.