Sevelamer hydrochloride, as a phosphate binder that contains neither aluminum nor calcium, is expected to improve the prognosis of dialysis patients. However, sevelamer hydrochloride has been reported to lower the serum bicarbonate level. In the present study, we performed a retrospective study on the potential influences of sevelamer hydrochloride on metabolic acidosis in hemodialysis patients. The subjects were 72 patients who underwent hemodialysis at our hospital. Thirty-six patients taking sevelamer hydrochloride and 36 patients matched for sex, diabetes mellitus, age and duration of dialysis who were not taking sevelamer hydrochloride were studied. We assigned the 36 patients who had been taking sevelamer hydrochloride to the 'sevelamer group', and the 36 patients not taking sevelamer hydrochloride were the control group. Statistical significance was evaluated by a t-test and Pearson's correlation coefficient. In the sevelamer group, the mean levels of bicarbonate, base excess and pH decreased significantly after administration, compared with the values before administration, but in the control group, aggravation of acidosis was not seen. The levels of bicarbonate, base excess and pH after the medication of sevelamer hydrochloride were found to be significantly and negatively correlated with the daily dose of sevelamer hydrochloride. The levels were also found to be significantly and negatively correlated with the cumulative dose of sevelamer hydrochloride; however, the value of the mean levels of chlorine and the anion gap did not increase with sevelamer hydrochloride. Sevelamer hydrochloride caused metabolic acidosis in a dose-dependent manner in hemodialysis patients without hyperchloremia.