Introduction: Monitoring of hemoglobin is necessary in patients with gastrointestinal haemorrhage. Take blood sample and analysis at laboratory with automatum is the gold standard. A fast determination of hemoglobin at the bedside is possible with a portable haemoglobinometer Hemocue.
Aims and methods: To assess correlation of capillary hemoglobin by Hemocue at the bedside, with venous hemoglobin by Coulter STKS in laboratory, for the monitoring of patients with a gastrointestinal haemorrhage hospitalized at november 2001 to july 2002. Statistical analysis used t test student apparied, and the determination of correlation coefficient r.
Results: Fourteen males and 8 females 61+/-17 aged had taking 204 venous blood samples and 204 capillaries blood simultaneously. Gastrointestinal haemorrhage was related with variceal bleeding (n=11), a duodenal ulcer (n =6), an esophagitis peptic ulcer (n =1), a diverticulous colitis (n =2), a gastric cancer (n =1), and an ischemic colitis (n =1). Initial hemoglobin was at 7.6+/-2,4 g/dL. Transfusion of 3+/-3 blood unit were realised in 17 patients. Means of hemoglobin were 9.15+/-1,62 g/dL on venous blood analyzed in laboratory, and 9.43 +/- 1,72 g/dL on blood capillary by Hemocue at the bedside. Correlation coefficient r was 0.87 (p <0.001). Variceal or others bleeding, fluid infusion for hypovolaemia, red cell transfusion and widespread of low level hemoglobin do not disturbed the validity of results.
Conclusion: In acute period of a gastrointestinal hemorrhage, monitoring of hemoglobin by Hemocue with capillary blood at the bedside is a reliable method.