#4141 serum bicarbonate and ckd progression: a longitudinal analysis by the group-based trajectory model

Abstract Background and Aims Follow-up studies in CKD patients showed an inverse relationship between serum bicarbonate and CKD progression, but only sparse longitudinal observations testing this relationship exist. Bicarbonate supplementation slowed the progression rate of CKD in 5 trials, but no such effect was registered in six trials. All these trials had limited power and were at high risk of bias. In the lack of well-designed clinical trials with sufficient power, longitudinal studies provide circumstantial new evidence for the hypothesis that bicarbonate supplements can mitigate the risk of adverse kidney outcomes. Methods We performed a longitudinal study in a cohort of 528 patients with at least 3 longitudinal measurements of bicarbonate over 32 months (inter-quartile range 30 – 36 months). The association between bicarbonate trajectories over time and the incidence rate of renal events (>30% eGFR reduction, dialysis or transplantation) were investigated by a two stages analysis: 1) in the first stage, we identified distinctive group-based bicarbonate trajectories analysis (GBTM, Annu Rev Clin Psychol. 2010;6:109-38); 2) in the second stage we assessed the association between bicarbonate trajectories groups and the risk of renal events by Cox proportional hazard model. Results One hundred and twenty-six patients had renal events on longitudinal observation. Four distinct trajectories of bicarbonate (based on intercept) were identified and labelled as low trajectory [including 11% of patients, mean (± SD) bicarbonate: 19.07± 2.5 mEq/L], moderate [35% of patients, bicarbonate: 21.2± 2.3 mEq/L], moderate-high [42% of patients, bicarbonate: 25.1± 2.2 mEq/L] and high [including 12% of patients, bicarbonate: 27.9± 2.8 mEq/L]. In analyses adjusting for age, gender, smoking, diabetes, total cholesterol, systolic BP, background CV comorbidities, anti-hypertensive drugs, haemoglobin, albumin, phosphate, hs-CRP, 24h proteinuria, and eGFR, the hazard rate (HR) of renal events decreased in a dose-dependent fashion from the lowest bicarbonate trajectory (reference category, HR 1) to the moderate (HR:0.70, 95%CI 0.43-1.14), high moderate (HR:0.54, 95%CI 0.31-0.93), and high bicarbonate category (HR:0.31, 95%CI 0.12-0.80) (P for trend=0.005). Thus, for patients in the high bicarbonate category (27.9± 2.8 mEq/L), there is an 69% risk reduction for a renal composite endpoint. Conclusion In a longitudinal analysis of a cohort of CKD patients, high bicarbonate levels are associated with a substantially reduced risk of renal events. These findings represent a strong call for well-designed, adequately powered randomised trials testing the effect of bicarbonate supplements or pharmacologic interventions that increase serum bicarbonate on renal outcomes. Given the substantial risk reduction in patients in the highest bicarbonate trajectory, these trials are an absolute clinical research priority.