Objective: To assess the volume of blood samples withdrawn for laboratory testing in intensive care unit (ICU) patients and to determine the influence of the resulting blood loss on transfusion requirements in patients staying in the ICU for more than seven days.
Study design: Prospective clinical open study.
Patients: Fifty patients treated in the ICU over the 3-month study period, neither admitted for a systematic postoperative monitoring, nor experiencing bleeding or haemolysis.
Methods: For each patient the following data were obtained: duration of ICU stay, volume of the daily withdrawn blood, the concentration of haemoglobin (Hb) at the time of ICU admission, ICU discharge and before each transfusion, volume of transfused blood.
Results: A mean volume of 62 +/- 29 mL.d-1 of blood was taken. It decreased from 85 +/- 6 mL on admission day, to 66 +/- 6 mL after seven days and 60 +/- 8 mL after 14 days. About 27% of the withdrawn blood was rejected (initial blood reflowing through cannula and connection tube). Twenty-one patients (42%) had a length of stay greater than 7 days. In this population, a first group (13/21) was given transfusions of packed red cells during their hospitalisation and a second group (8/21) was not transfused. The mean volume of blood taken per day (67 +/- 21 mL.d-1 vs 55 = 15 mL.d-1) and the total volume (1.204 +/- 810 mL vs 810 +/- 389 mL) were not significantly higher in the transfused group. Conversely, the mean haemoglobin concentration on ICU admission (97 +/- 22 g.L-1 vs 136 +/- 26 g.L-1) was significantly lower (P = 0.001) in the transfused patients.
Conclusion: Blood losses from blood withdrawal for laboratory tests are important and in agreement with the results of other reports. It is generally accepted that iatrogenic blood loss of this magnitude can cause anaemia if repeated over a prolonged period. Conversely, our data suggest that blood sampling does not contribute significantly to anaemia and transfusion requirements in patients with a prolonged ICU stay.