OBJECTIVE : The hemodialysis (HD) session per se is a catabolic event contributing to protein-energy wasting via several mechanisms including nutrient losses. Amino acids (AAs) losses in the dialysate are estimated from 6 to 10 g per treatment. The HD patient plasma AA concentration is usually lower than in normal subjects. This is even more marked in patients with long dialysis vintage or malnutrition. METHODS : The aim of the study was to evaluate the effect on mass balance of a branched-chain AA (BCAA)-enriched (valine, isoleucine, leucine) dialysis fluid in a group of 6 stable HD anuric patients, fasting since 12 hours. The specific choice of BCAA relied on their key role on protein and muscle anabolism and their usual decreased plasma concentration in HD patients. Each patient was prescribed in a cross-over design and random order, either receiving a standard high-flux HD or an HD treatment using a BCAA-enriched acid concentrate designed to achieve a physiological plasma concentration of BCAAs. HD prescription remained unchanged during the 2 phases of study. Dialysate electrolytes prescription was kept constant for each individual patient, as well as dialysate glucose concentration (5.5 mmol/L). Pre- and post-dialysis BCAAs concentrations were measured by Ion-Exchange Liquid Chromatography. Postdialysis concentrations were corrected for hemoconcentration, and net mass transfer was calculated. RESULTS : Six stable prevalent end-stage kidney disease patients were studied. They consisted of 5 men and 1 woman, aged 69.9 years, with body mass index of 25.2 kg/m2. Treatment schedule consisted of treatment time 4 hours, high-flux polysulfone membrane (1.8 m2), blood flow 350 mL/minute, and dialysate/blood flow ratio at 1.5. The average BCAAs concentration in dialysate was targeted to physiological levels and assessed in 6 different samples, respectively for plasma valine, isoleucine, and leucine at 271, 78, and 145 μmol/L. With standard dialysate, plasma valine decreased from 204.5 to 130.8 (P = .0014). Plasma isoleucine and leucine changes were not significant, respectively from 65.7 to 59.3 μmol/L and 110.3 to 113.4 μmol/L. When using the BCAA-enriched dialysis fluid, plasma valine increased from 197.2 to 269.2 μmol/L (P = .0001), plasma isoleucine and leucine respectively from 63.2 to 84.7 (P = .0022) and from 107.2 to 161.6 μmoles/L (P = .0002). Dialysis dose estimated from KT/V did not differ between the sessions. The mass transfer with BCAA-enriched dialysate was +115, +16, and + 83 μmol per session for leucine, isoleucine, and valine, respectively. CONCLUSIONS : In conclusion, the addition of BCAAs at physiological concentration in the dialysis fluid contributes to restore physiological plasma concentrations for valine, isoleucine, and leucine at the end of a dialysis session. As BCAAs are essential to muscle balance, this could help to limit losses of BCAAs, restore physiological BCAAs concentrations, and decrease muscle catabolism observed during the HD treatment. Further outcome-based studies are needed to confirm this hypothesis on a larger scale and longer treatment time.