Reducing risk of iatrogenic anemia by revising phlebotomy practice

A simple change in phlebotomy practice – switching from conventional to low-volume sample blood tubes – results in a 42 % reduction in patient blood loss, and consequent reduced risk of iatrogenic anemia. That is the positive headline finding of this recently published study conducted at the Royal Adelaide Hospital in South Australia. 

For many hospital patients who require frequent blood testing, the accumulated associated blood loss can be of sufficient volume to materially increase their risk of anemia or increase severity of pre-existing anemia, with consequent need for red-cell transfusion. Since red-cell transfusion is associated with its own range of risks for the recipient patient, it is to be avoided if possible. Conservation of the patients own blood (i.e. minimizing blood loss) is a central strategy in patient blood management (PBM) guidelines aimed at reducing the necessity for transfusion. PBM guidelines promote several strategies for minimizing phlebotomy-associated (iatrogenic) blood loss, including the use of small-volume phlebotomy tubes. 

As the authors of this study observe, common hematological and biochemical assays using modern instrumentation only require 10-20 µL of blood and commonly collected volumes per test exceed this requirement 20-40 fold. They designed their study to assess the absolute reduction in iatrogenic blood loss that can be achieved by switching to small-volume tubes. 

Prior to study implementation, the authors first sought to validate the use of small-volume tubes. Paired samples (one collected into conventional EDTA tube and the other collected into small-volume EDTA tube) were collected simultaneously from 20 patients. All samples were submitted for 5-parameter hematology testing. The same approach was used to test agreement between paired lithium-heparin samples for 21-parameter biochemistry testing. Statistical analysis of paired results revealed acceptable collinearity for all parameters (coefficient of determination, R2 typically in the range of 0.990-1.0); the small-volume tubes to be used in the study had thus been validated for both hematology and biochemistry testing. 

The study population comprised two cohorts of hematology inpatients; the first (control) cohort comprised 68 patients who were admitted between 2009 and 2010; blood samples from this cohort were collected into conventional tubes with the following fixed volume fills: 4 mL for EDTA (hematology) tubes; 4 mL for sodium-citrate (coagulation-study) tubes; 8 mL for lithium-heparin (biochemistry) tubes; 8 mL for clot-activator tubes; and 9 mL for EDTA (transfusion-specific) tubes.

The second (intervention) cohort comprised 102 patients admitted between 2012 and 2013; blood samples from this cohort were collected into small-volume tubes with the following fixed volume fills: 2 mL for EDTA and sodium-citrate tubes; 2.5 mL for lithium-heparin tubes; 5 mL for clot-activator tubes; and 6mL for EDTA (transfusion-specific) tubes.

In terms of age, gender split and median length of stay the two cohorts were not significantly different. Of note, length of stay was on average 17 days for the two cohorts, highlighting the fact that hematological inpatients generally have a relatively long hospital stay and therefore presumably, a consequent high risk of accumulative iatrogenic blood loss from regular blood testing. 

There was no statistical difference between the two cohorts in terms of the mean number of blood tubes collected per patient, but the mean total volume of blood collected per day was 20.0 mL for the control cohort and 11.5 mL for the intervention cohort; thus, use of small-volume tubes was associated with reduction in blood loss of 8.5 mL/day (i.e. 42 % reduction). During a 3-week admission, which is clearly not unusual for hematology inpatients, the mean reduction in iatrogenic blood loss achieved by the use of small-volume tubes thus amounts to 180 mL, certainly enough to materially preserve hemoglobin concentration (loss of 1 mL blood results in a mean decrease in hemoglobin of 0.07 g/L). 

EDTA sample error rate was recorded for a total of 7,810 EDTA samples collected from the control cohort and 10,570 EDTA samples collected from the intervention cohort. Defined sample errors were: sample clotted; platelet clumping; fibrin present; hemodilution and insufficient sample. Overall, error rates were low in both control and intervention cohorts, although there was a statistically significant higher proportion of samples with fibrin present among the intervention group (0.5 % intervention cohort versus 0.2 % control cohort). For all other errors assessed, there was no significant difference between the two cohorts. 

In discussion of their study the authors report that since completion of their study in 2013, small-volume blood tubes have been used for all venesections on hematology inpatients at their hospital. The change in phlebotomy practice has met with ”broad clinical acceptance and minimal implementations issue.” There are plans to implement the policy across the entire hospital.