UC Irvine

Background and objectives

The optimal target for glycemic control has not been established for diabetic peritoneal dialysis (PD) patients.

Design, setting, participants, & measurements

We examined mortality-predictability of hemoglobin A1c random serum glucose in a contemporary cohort of diabetic PD patients treated in DaVita dialysis clinics July 2001 through June 2006 with follow-up through June 2007.

Results

We identified 2798 diabetic PD patients with A1c data. Serum glucose correlated with A1C (r=0.51). Adjusted all-cause death hazard ratio and 95% confidence interval for baseline A1c increments of 7.0 to 7.9%, 8.0 to 8.9%, 9.0 to 9.9%, and ≥10%, compared with 6.0 to 6.9% (reference), were 1.13 (0.97 to 1.32), 1.05 (0.88 to 1.27), 1.06 (0.84 to 1.34), and 1.48 (1.18 to 1.86); and for time-averaged A1c values were 1.10 (0.96 to 1.27), 1.28 (1.07 to 1.53), 1.34 (1.05 to 1.70), and 1.81 (1.33 to 2.46), respectively. The A1c-mortality association was modified by hemoglobin level such that higher all-cause mortality was evident only in nonanemic patients. Similar but non-significant trends in cardiovascular death risk was found across A1c increments. Adjusted all-cause death HR for time-averaged blood glucose 150 to 199, 200 to 249, 250 to 299, and ≥300 mg/dl, compared with 60 to 99 mg/dl (reference), were 1.02 (0.70 to 1.47), 1.12 (0.77 to 1.63), 1.45 (0.97 to 2.18), and 2.10 (1.37 to 3.20), respectively.

Conclusions

Poor glycemic control appears associated incrementally with higher mortality in PD patients. Moderate to severe hyperglycemia is associated with higher death risk especially in certain subgroups.