ABSTRACT
UNLABELLED: Gastrointestinal hemorrhage is a frequent complication in digestive diseases, requiring follow-up by specialized units such as the digestive bleeding unit. Hemoglobin is determined daily in the management of these patients, and consequently a rapid and reliable method for hemoglobin determination is required. OBJECTIVE: To study the reliability of HemoCue in patients with gastrointestinal hemorrhage and to determine whether there are any differences in hemoglobin measurement between HemoCue and the reference laboratory. MATERIALS AND METHODS: We performed a cross-sectional, observational paired sample study of 54 patients treated in the digestive bleeding unit. Hemoglobin was measured simultaneously by HemoCue (Angelholm, Switzerland) and by laboratory determination in patients with gastrointestinal bleeding, both in the acute phase (first 24h of bleeding) and in the stable phase. The results are expressed as means. To compare means, Student's t-test was used for paired data, Pearson's r was used for correlations and Bland and Altman analysis was used to quantify differences. RESULTS: Fifty-four patients (34 males) were included. The cause of the hemorrhage was duodenal/gastric ulcer in 26 patients, variceal bleeding in eight, esophagitis in five, Mallory Weiss syndrome in four, portal hypertensive gastropathy in nine and other causes in one. In the acute phase, mean arterial pressure (MAP) was 83.8, mean hemoglobin in capillary blood was 8.07 g/dl by HemoCue in 0.81 min, and mean hemoglobin in venous blood was 8.17 by laboratory determination in 44.87 min. Pearson's correlation coefficient was 0.979. In the stable phase, MAP was 100.7, and hemoglobin with HemoCue was 9.32 in 0.8 min and 9.4 by laboratory determination in 47.83 min. Pearson's correlation coefficient in the stable phase was 0.966. CONCLUSIONS: There is a good correlation between HemoCue and laboratory determinations. Consequently, HemoCue is a quick and reliable method both in the acute and stable phases of gastrointestinal bleeding.
