A comparison of urinary albumin-total protein ratio to phase-contrast microscopic examination of urine sediment for differentiating glomerular and nonglomerular bleeding.

BACKGROUND

Hematuria can be classified as either glomerular or nonglomerular, depending on the bleeding source. We recently reported that urinary albumin-total protein ratio is potentially useful for identifying the source of hematuria.

STUDY DESIGN

Diagnostic test study.

SETTING & PARTICIPANTS: 579 fresh urine specimens with microhematuria (> or =5 red blood cells/high-power field) collected from patients with the source of the hematuria confirmed on histopathologic and/or imaging studies and clinical criteria assessed.

INDEX TEST

Each urine specimen was evaluated morphologically by using phase-contrast microscopy and biochemically by using urinary albumin-total protein ratio, albumin-creatinine ratio, and total protein-creatinine ratio.

REFERENCE TEST

Each patient had a definitive clinical diagnosis established by means of biopsy (64.4%), imaging studies (21.2%), and routine optimal microscopic examination of urine sediment (14.3%).

RESULTS

Of 579 specimens, 329 were obtained from patients with glomerular disease and 250 were obtained from patients with nonglomerular disease. Mean urinary albumin-total protein, albumin-creatinine, and total protein-creatinine ratios for those with glomerular versus nonglomerular diseases were 0.73 +/- 0.11 versus 0.41 +/- 0.14 mg/mg (P < 0.001), 1,110 +/- 1,850 versus 220 +/- 560 mg/g (P < 0.001), and 1,600 +/- 3,010 versus 480 +/- 1,160 mg/g (P < 0.001), respectively. The percentage of patients with greater than 3% glomerular red cells was 83.3% versus 24.8% (P < 0.001). Receiver operating characteristic curve analysis showed that areas under the curve for albumin-total protein ratio, albumin-creatinine ratio, and total protein-creatinine ratio were 0.992, 0.781, and 0.688, respectively (P < 0.001, albumin-total protein versus albumin-creatinine; P < 0.001, albumin-total protein versus total protein-creatinine). At cutoff values of 0.59 mg/mg, 71 mg/g, and 265 mg/g, albumin-total protein ratio, albumin-creatinine ratio, and total protein-creatinine ratio had sensitivities and specificities of 97.3% and 100%, 78.9% and 61.1%, and 68.8% and 62.0% for detecting glomerular disease, respectively. Phase-contrast microscopy had sensitivity of 83.3% and specificity of 75.2% for detecting glomerular disease.

LIMITATIONS

Albumin-total protein ratio cannot be used in patients with urinary total protein less than 5 mg/dL (<0.05 g/L). Use of only 1 sample from 1 patient may not be sufficient to obtain definitive results.

CONCLUSIONS

Urinary albumin-total protein ratio is much more useful than phase-contrast microscopy for differentiating between glomerular and nonglomerular disease in patients with microscopic hematuria.