Measurement of occult upper gastrointestinal tract blood loss: A direct comparison of radiochromium and haem‐porphyrin assay techniques

Faecal haem-porphyrin assay by the HemoQuant method has many practical advantages over the well-validated, radiochromium (51Cr-tagged red cell) method for measuring gastrointestinal blood loss. Because haem may be absorbed but the chronic ion is not, the two measures were directly compared in low-grade bleeding from the proximal gastrointestinal tract. Blood loss was measured by both methods simultaneously in 40 patients with osteoarthritis before and during medication with aspirin preparations. Mean (geometric) daily blood loss before aspirin usage measured 0.60 mL radiochromium (range 0.13-1.62) and 0.47 mL (0.14-1.40) by HemoQuant (P = 0.042). On aspirin, bleeding rose to 1.57 mL/day (0.43-4.85) by radiochromium and to 0.72 mL/day (0.23-3.0) by HemoQuant (P < 0.0001). The two measures correlated well, r = 0.847 (P << 0.0001), but the regression coefficient was 0.417, reflecting the lower estimates of bleeding by HemoQuant. In four normal subjects who ingested 51Cr-labelled red cells (26-41 mL) over 3 days, recovery of 51Cr was complete (103 +/- 2%, +/- s.e.), but recovery of haem-porphyrins was only 63 +/- 13% (P = 0.01), presumably because of absorption of haem. Although faecal haem-porphyrin assay is of considerable clinical utility, it is a quantitative index rather than an absolute measure when low amounts of bleeding originate from the proximal gastrointestinal tract.