We previously hypothesized that PCR is dependent on Kt/V (urea). To test the hypothesis, 30 patients from two hemodialysis (HD) units with PCR values consistently less than 1 regardless of the Kt/V (urea) without an obvious cause (such as malignancy, peptic ulceration, etc.) were identified. These patients were randomized into an experimental group where an increase in the Kt/V (urea) was made or into a control group where a Kt/V (urea) remained unchanged. In the experimental group the minimal increment in Kt/V (urea) was 0.3 plus a minimum requirement of a Kt/V (urea) greater than or equal to 1. The control group were to maintain the existing Kt/V (urea) within +/- 0.2. Urea kinetic modeling was monitored monthly and the patients followed for three months. Of 30 patients, 18 fulfilled the criteria (9 experimental; 9 controls). In the experimental group, the initial Kt/V (urea) was 0.82 +/- 0.19 (means +/- SD); the final Kt/V (urea) was 1.32 +/- 0.21 (P = 0.000). These patients had an initial PCR of 0.81 +/- 0.08. This increased to 1.02 +/- 0.15 (P = 0.005). The control group had an initial Kt/V (urea) of 0.82 +/- 0.16. The final Kt/V (urea) was 0.87 +/- 0.15. Their initial PCR was 0.87 +/- 0.14 and the final PCR was 0.86 +/- 0.087 (P = NS). In the experimental group, the mean increments in Kt/V (urea) and PCR were 0.49 +/- 0.17 and 0.21 +/- 0.16 (P less than 0.005).(ABSTRACT TRUNCATED AT 250 WORDS)