An Amino Acid-Based Peritoneal Dialysis Fluid Buffered with Bicarbonate versus Glucose/Bicarbonate and Glucose/Lactate Solutions: An Intraindividual Randomized Study

Abstract

Objective

In order to study acute metabolic changes and peritoneal transport, amino acids as osmotic agent and bicarbonate as buffer were tested as new agents in peritoneal dialysis (PD) solutions.

Design

In a prospective, cross-over, randomized, intraindividual study, we investigated the acute metabolic changes following the application of three different PD fluids: (1) a 1% amino acid-based PD solution buffered with bicarbonate (34 mmol/L) (Amino/Bic); (2) a 1.5% glucose anhydrous-containing bicarbonate-buffered solution (34 mmol/L) (Glu/Bic); and (3) a conventional 1.5% glucose anhydrous-based dialysis solution with lactate (35 mmol/L) (Glu/Lac).

Setting

University medical center.

Patients

Ten nondiabetic patients stable on continuous ambulatory peritoneal dialysis (time on dialysis, 42.5 ± 21.5 months) were treated and monitored with the test solutions over a 6-hour dwell. Three different study days followed in a randomized order for each patient (interval of 1 – 3 weeks). Blood and dialysate samples were taken at 0.25, 0.5, 1, 2, 4, and 6 hours. Immediately after the 1-hr dwell (and after sampling), the patients received a standardized breakfast, thereby simulating usual food intake.

Results

Following the application of Amino/Bic a significant increase in plasma amino acids occurred, with peak levels (maximum 250% increase) after either the 1-hr or the 2-hr dwell. Before taking the standard meal (0.5 hr, 1 hr), the mean serum glucose level with Amino/Bic was 8% ± 13% lower than with Glu/Bic (p = 0.06) and 14% ± 8% lower than with Glu/Lac (p < 0.01). This difference was still significant after the standard breakfast and also for the whole dwell (average serum glucose 0.5 – 6 hr: Amino/Bic, 91 ± 6 mg/dL; Glu/Bic, 100 ± 8 mg/dL; Glu/Lac, 102 ± 7 mg/dL; p < 0.01 MANOVA). The serum insulin profiles did not differ between the fluids. A transperitoneal protein- and amino acid-related nitrogen loss of 0.49 ± 0.18 g and 0.48 ± 0.12 g per dwell was measured using Glu/Bic and Glu/Lac, while a positive balance of 1.80 ± 0.43 g was achieved with Amino/ Bic. The parameters of acid-base status (pH, HCO₃, pCO₂) remained nearly unchanged in the blood, irrespective of the solution used, while dialysate values differed markedly. No significant differences with respect to ultrafiltration (Amino/Bic, –68 ± 199 mL/6hr; Glu/Bic, –51 ± 89 mL/6hr; Glu/ Lac, –2 ± 134 mL/6hr) and peritoneal creatinine clearance (Amino/Bic, 4.9 ± 0.6 mL/min; Glu/Bic, 5.1 ± 0.6 mL/min; Glu/ Lac, 4.8 ± 0.5 mL/min) were measured.

Conclusions

Our results demonstrate that ultrafiltration and small solute clearance over a 6-hour dwell with a 1% Amino/Bic solution were comparable to those of 1.5% Glu/Bic and 1.5% Glu/Lac. Reduced serum glucose concentrations were found with Amino/Bic and this fluid compensated the transperitoneal protein-nitrogen loss of about three glucose dwells. Bicarbonate buffering (34 mmol/L) did not change blood acid-base status combined with either glucose or amino acids.