Trimethylamine-N-oxide (TMAO) and clinical outcomes in patients with end-stage kidney disease receiving peritoneal dialysis.

BACKGROUND Trimethylamine-N-oxide (TMAO) is a gut bacteria-derived metabolite of l-carnitine and choline. A high concentration of TMAO has been proven to relate to cardiovascular disease (CVD), all-cause mortality and chronic kidney disease progression. We aimed to investigate the relation between the value of serum TMAO and outcomes for peritoneal dialysis (PD) patients. METHODS This is a prospective cohort study with data retrospectively analysed. All incident PD patients were enrolled and followed up. Log-rank test, competing risk survival analysis and COX regression were performed to test the effect of serum TMAO on developing first-episode peritonitis, all-cause and CVD mortality. RESULTS A wide distribution of serum TMAO concentration was observed in 513 PD patients, with a median level of 72.3 (43.7, 124.7) µmol/L. Patients with lower TMAO concentration were more likely to be without diabetes and hypertension. Patients with lower TMAO concentration showed better residual kidney function and solute clearance at baseline. Participants in the higher three TMAO quartiles showed an increased risk for first-episode peritonitis (p = 0.039). By competing risk survival analysis, after adjusting for age, sex, diabetes mellitus, CVD, body mass index, albumin, high-sensitive C-reactive protein, potassium, phosphorus, residual kidney function, normalised protein equivalent of total nitrogen appearance and calendar year of catheter implantation, patients in the higher three TMAO quartiles had a statistically or marginally higher risk for first-episode peritonitis compared with patients in the lowest quartile, with hazard ratio (HR) 1.65 (1.05, 2.58), 1.46 (0.92, 2.31) and 1.66 (1.05, 2.61), respectively. In the COX model, patients in the third quartile TMAO group had significantly higher CVD mortality risk compared with the lowest quartile group, as HR 2.27 (1.02, 5.05) after adjusting for various factors. As for all-cause mortality, TMAO did not show any associated effects. CONCLUSIONS Serum TMAO concentration is associated with the risk of first-episode peritonitis and CVD mortality in PD patients. No obvious association between serum TMAO and all-cause mortality was observed.