The effect of blood-drawing techniques and equipment on the emolysis of ED laboratory blood samples
Dear Editor:
I would like to take this opportunity to comment on a recent letter to the editor published in the October 2003 issue regarding the research study, “The Effect of Blood Drawing Techniques and Equipment on the Hemolysis of ED Laboratory Blood Samples.” I was disappointed in the letter and commend both the research study authors and the Editor for their responses to the letter.
My comments and perspectives are based on my research background and having recently completed a very similar project that was published recently.1 Rather than attack the negativity of the letter, I would like to start by commending the authors of the research project. To encourage nursing research, we need to provide thoughtful comments and encourage nurses to become active in research, not provide unhelpful, uninformed critiques. I also attended the oral poster session by Ms Grant on this project at the National Scientific Assembly ENA meeting in September this year in Philadelphia and was able to speak with the investigator. Ms Grant started this project as a staff nurse who was frustrated about a clinical problem that had a very negative impact on patient care. She wanted to do something about it other than complain. She was not a researcher, nor did she have research training. Yet she pursued, consulted, and completed a great project. Her sampling technique and methods were reasonable, unbiased, and did allow for the ability to make conclusions. Ms Grant did not make any conclusions not supported by data, and she clearly called for further research on the topic. In particular, she admitted she could not explain why their results did not find a difference in hemolysis rates between catheter size as found in previous research. She did not recommend global practice changes but reported the results of their practice changes.
I would like to comment on a few findings. The overall hemolysis rate for the project was extremely high (32%). We recently completed a similar project with a larger sample size and a more detailed analysis. Our overall rate of hemolysis was much less (7%).1 We also compared hemolysis rates between intravenous catheters and steel needles (vacutainer draws) and examined the following additional variables: catheter size, location of draw, and patient age. We used a different brand of catheter than did Grant. I do wonder if there is something else going on at the investigator's institution; perhaps the other brand of catheters have an even higher rate of hemolysis. We also used a single and even more strict definition of hemolysis. Our blood bank technicians used any discoloration of pink or darker as the definition of hemolysis. Theoretically, we should have had an even higher rate of hemolysis than did Grant.
Our recent data confirm Grant's findings that specimens drawn through an intravenous catheter are more likely to hemolyze. We used multiple regression to estimate the effect of multiple variables on the ability to predict hemolysis. Specimens were 6.73 (RR, P = 0.001) times more likely to hemolyze when drawn through a catheter compared with a steel needle (10% vs 1.5%). Translated, specimens are almost 7 times more likely to hemolyze when drawn off a catheter than through a steel needle. This is not a finding we were hoping for, however, you cannot argue with these data.
Grant also commented on their inability to explain the lack of effect of catheter size. Previous data have demonstrated higher rates of hemolysis with smaller gauge catheters.2., 3., 4. Our data confirmed these findings. In our sample of 605 specimens, smaller gauge catheters (20- to 24-gauge) were 7.42 times (RR, P = 0.005) more likely to hemolyze when compared with 14- or 16-gauge catheters, and 18-gauge catheters were 3.6 times (RR, P = .067, NS) more likely to hemolyze. Again, translated, the worst thing you can do is draw your sample off of a 20- to 24-gauge catheter. The risk is less with an 18-gauge catheter and almost none when drawn through a 14- or 16-gauge catheter. I would encourage the researchers to reanalyze the data with use of a multiple regression technique, which may or may not help answer the question of the lack of effect from catheter size.
Finally, we also examined the effect of location of draw and found an increased risk of hemolysis for samples drawn from sites other than the antecubital fossa (RR = 2.61, P = 0.003). Specimens were about 2.6 times more likely to hemolyze when drawn from the hand, wrist, etc.
Hemolysis of specimens is a serious problem many emergency departments are facing on a daily basis. Causes of hemolysis are complicated and multifactorial. We do not believe the answer is a single practice change. We have educated the staff about our findings and have not chosen to “write a policy.” Patients' veins come in all shapes and sizes. Frequently, it is very difficult to obtain 2 sticks. We have suggested the staff obtain 2 sticks when possible or draw their specimens off of a larger (14- to 18-gauge) catheter from the antecubital fossa. Based on each patient's anatomy and veins, staff should realize the worst thing to do is draw a specimen off of a 20-gauge intravenous catheter from the hand. Clearly, drawing from any intravenous catheter will significantly increase your likelihood of hemolysis. This is not a popular finding; however, maybe it is time we need to really try to start basing our practice on research whenever possible.
Finally, we also learned that laboratories often “reject” specimens for many reasons; hemolysis is only one reason. When attempting to look into this problem at an individual institution, make sure hemolysis rates represent only that and not mislabeling of specimens, etc.
Again, I would like to congratulate Ms Grant and her team for being so bold as to conduct this research project, publish their findings, present at a national conference, and respond so professionally to a letter to the editor.
FULL TEXT