Search

Urinary total protein (UTP) determinations are notoriously inaccurate, poorly reproducible, and difficult to interpret in early renal disease, causing many investigators to measure urinary albumin instead. In this study, we compare a new nonimmunologic fluorescent dye (AB-dye) for measuring albumin with the more expensive and cumbersome radioimmunoassay. We tested 207 urine specimens from patients with variable protein concentrations and divided the results into five arbitrary ranges (0 to 20, 21 to 50, 51 to 100, 101 to 200, and 201 to 400) for chi-square analysis. There was a high degree of correlation between the two methods (chi-square = 260.8 with 16 degrees of freedom; P < 0.001). The correlation was also high when analyzed by linear regression (R = 0.86; F < 0.01). Based on our comparison of total protein and albumin concentration in the same urine samples, we hypothesized that patients with mild proteinuria may not necessarily have microalbuminuria. Urine samples with UTP between 150 and 400 μg/mL were tested for albumin by the AB-dye. Of 41 samples in this range, 18 (44%) had normal albumin levels. We conclude that measuring urinary albumin with the AB-dye is comparable in performance to radioimmunoassay and could replace UTP determinations, especially for patients with borderline elevations of UTP, many of whom do not have microalbuminuria.

We have previously found that carbamylated hemoglobin (carHb) levels are increased in chronic renal failure and correlate positively with blood urea nitrogen (BUN) levels and with the duration of exposure to urea. In a fashion analogous to glycosylated hemoglobin in diabetic patients, it is possible that carHb may better reflect BUN levels before hemodialysis (preBUN) and also between hemodialysis sessions. We therefore tested the hypothesis that carHb could be a better index of adequacy of hemodialysis than the urea reduction ratio (URR). Fifty hemodialysis patients had carHb measured every 2 months for 14 months; the carHb level was compared with URR and preBUN levels, as assessed by changes in absolute numbers and trends of the BUN levels between hemodialyses. Mean URR was above 61% throughout the 14 months. Mean carHb levels did not change significantly during the study and were only weakly correlated with URR. However, there was a much better correlation between predialysis BUN and carHb, suggesting that carHb levels reflect more accurately the changes in BUN between hemodialysis sessions. To further test this hypothesis, we subdivided the patients arbitrarily, depending on the change in preBUN between two consecutive carHb measurements. We found significantly lower carHb levels when BUN decreased or remained stable than when it increased or was persistently high. In patients with decreasing or stable BUN, carHb was significantly lower than in patients with persistently high or increasing BUN (carHb 81.5 +/- 3.6 microg valine hydantoin [VH]/g Hb v 123.7 +/- 11.7 microg VH/g Hb, respectively; P0.001). URR was not different between groups. In addition to changes in BUN levels, carHb was correlated by multiple regression analysis with the presence of diabetes, weight, and plasma HCO3. The relationship between diabetic patients and carHb levels was complex because such patients tend to have higher preBUN levels, higher protein catabolic rate, and lower HCO3 levels. These results demonstrate that carHb reflects the changes between dialysis BUN and may serve as a more accurate index of uremia control. Clinically, it appears that well-dialyzed patients have carHb levels lower than 100 microg VH/g Hb.