Detection of occult upper gastrointestinal tract bleeding: performance differences in fecal occult blood tests

To compare rates of detection of occult upper gastrointestinal (GI) tract bleeding by guaiac (Hemoccult II [HO]), immunochemical (HemeSelect [HS]), and heme-porphyrin (HemoQuant [HQT]) fecal occult blood tests. PATIENTS, SUBJECTS, AND METHODS: In a cross-sectional study to detect native occult upper GI tract bleeding, single stools were collected from 56 patients with iron deficiency and a proven hemorrhagic GI tract lesion. In a longitudinal study to detect simulated occult upper GI tract bleeding, 3 stool samples were serially collected from 10 clinically normal subjects after ingestion of 5 and 15 mL of autologous blood. All stool samples were subjected to blinded fecal occult blood determinations with use of the 3 tests.

In the cross-sectional study, the HQT test detected 88% (37/42) of hemorrhagic upper GI tract lesions compared with 26% (11/42) detected by the HO test (P<.001) and 2% (1/42) by the HS test (P<.001). In the longitudinal study, all preingestion fecal occult blood test results were negative. After ingestion of 5 mL of blood, the HQT result became positive in 60% (6/10), and the HO and HS results remained negative (P=.03). After ingestion of 15 mL of blood, the HQT result became positive in all 10 cases, the HO result was positive in 6 (P=.12 vs HQT), and the HS result was positive in none (P=.002 vs HQT); all 3 stool samples collected after the 15-mL ingestion were positive in each of the 10 subjects by the HQT test but in only 1 subject by the HO test (P=.003).

The HQT test detects occult upper GI tract blood loss significantly more frequently than the HO or HS test.