Hot Topic

Phlebotomy Top Gun


Receive notification when new Hot Topics are published:

Click CC to turn on closed captioning.

Published: August 2013

Print Record of Viewing

Dr. Karon presents technical issues involved in the collection of blood specimens during intravenous starts. He will discuss what we know about this practice and speak to the quality of intravenous start blood collection.

Presenter: Brad Karon, MD, PhD

Questions and Feedback

Contact us:


Download the PDF

For those of you who have not attended our phlebotomy conference here in Rochester, the title of my hot topic today, “Phlebotomy Top Gun,” may seem a little odd, but I hope I’ll be able to explain to you what is “Phlebotomy Top Gun” and what we try to do at our annual phlebotomy conference.

I have no disclosures relevant to today’s presentation.

So every year at our phlebotomy conference, I do present a talk that’s always called “Phlebotomy Top Gun.” The format of “Phlebotomy Top Gun” is a case-based presentation, and I solicit from you, the attendees, cases, issues, or scenarios you like to year about. I present these scenarios, and using an audience response voting mechanism, the audience and the attendees of the conference vote on the answer they feel is appropriate to the case. I then present the material, evidence, data, which I’m going to do for this question and issue; and at the end of the case, the attendees vote again, and I get to see if I’ve changed anybody’s mind. So this is an actual case from a “Phlebotomy Top Gun” presentation at a previous phlebotomy conference in Rochester, and the question was—“Regarding the practice of drawing blood during IV starts, consider the following or which of the following statements do you feel is true.”

  • The first answer or choice would be: It’s no more problematic than a venous-line draw.
  • The second choice would be: It’s no more problematic than an arterial-line draw.
  • The third choice would be: IV-start draws have hemolysis rates that are several-fold higher than any other type of blood draw.
  • And the fourth choice would be: With proper training, IV-start draws can be done without any problems.

And at this point, during the live conference, attendees using an audience response system would vote for the answer they thought was appropriate, and we would get to see in real time the distribution of results that our audience felt was correct.

So, again, this is a case that I presented at a previous phlebotomy conference; and what I do next then with each case in the “Phlebotomy Top Gun” presentation is go through any data, evidence, internal or external to Mayo Clinic that exist and then re-poll at the end to see if anybody’s mind has been changed; and I give you what I feel is the correct answer. So regarding IV blood draws, there have been a fair amount of formal studies done, and I’ll be presenting 2 published studies today. The first from the Journal of Emergency Nursing in 2005 was an observational study of 100 patients’ blood draws that were drawn through an IV start. The study was fairly well performed in the sense that they did identify which blood-drawing device (a BD Autoguard catheter), that they identified whether they did or didn’t use extension tubing (in this case, they did use extension tubing), and the type of syringe used for the blood draw. The number of draws was reasonable (100 blood draws) and 382 separate blood samples collected. And you can see that almost 13% of specimens had some degree of hemolysis that was visibly detected. But more significantly, about 4% of samples were rejected by the laboratory for moderate hemolysis, which is generally in most labs around 200 mg/dL for hemoglobin. And they looked at the variables associated with specimens that were rejected for hemolysis, and they found that a 22-gauge IV (a small IV) or a difficult IV placement (not surprising) was associated with a greater risk for sample hemolysis.

The next study I’ll talk about was published in the Journal of Emergency Nursing in 2008. This was a prospective study, meaning they looked at current practice, and fairly large of 853 different nurse-collected samples. They had the same group of nurses either collecting through IV starts or performing venipuncture; and therefore, they compared the rate of hemolysis in those samples collected by nurses through direct venipuncture versus those collected during an IV start. They used a Vacutainer collection system for both types of collection, and they found a very significant difference in the rate of hemolysis—about 5% rate of hemolysis when samples were collected during IV starts versus about 0.3% for direct venipuncture. This study did define hemolysis as any degree of free hemoglobin that could be visibly detected in the separated serum or plasma sample, which might be one drawback. We would probably rather prefer an expression of the amount of samples or percent of samples that were rejected by the laboratory because they had too much hemoglobin. But it does serve to show that certainly there’s a higher percentage of samples collected through IV starts that have visible hemoglobin in serum or plasma as opposed to those collected by direct venipuncture.

Finally, this is some internal quality improvement data we did at Mayo Clinic here. This was part of a quality improvement initiative looking at reducing hemolyzed specimens in our core testing laboratory. We did a 1-month snapshot of all hemolyzed blood samples in the core clinical laboratory at Mayo Clinic; and as you can see, about a little over 20% of all hemolyzed samples in this 1-month period were collected during IV starts and the remainder collected during Vacutainers or with butterflies or with other type of syringes. But certainly, only a small percentage of our overall samples are collected during IV starts, and the fact that they made up over 20% of all hemolyzed samples suggests they’re much more problematic than any other type of blood draw.

So we went on after we obtained this quality improvement data and we actually looked at—Is there something we can do to improve the integrity of blood specimens collected during IV starts? The first thing we did is we looked at 2 different areas, and we only have 2 different areas in the practice, where we routinely do IV-start blood collections. In one case in the Emergency Department, it’s nursing doing the collections; and the other (and this is a preop area), it was laboratory personnel doing the collections. So we looked at about 300 specimens collected from both areas, and we looked at—Is there a big difference in the hemolysis rate between laboratory-collected and nursing-collected? And you can see 6.7% versus 8.5%—the numbers are a little different; but basically, on the smaller data sets, really both high rates of hemolysis and not a lot of difference in terms of whether laboratory or nursing personnel are doing the collections.

We then went on to say—Is there something we can do to make the process better? So we did a second trial, and this is focusing in both places as well, but smaller numbers, about 200 specimens that we did; and instead of doing our --what was our previous process of just drawing blood during the IV start, we used a Luer-lock device and added a 2-cc discard volume to try to see if that would reduce the amount of hemolyzed samples. And we went, as you saw in the last slide, from our baseline of about 7% to 8% hemolysis rates down to a little less than 6% of samples being hemolyzed when we added in the Luer-lock device with a 2-cc discard volume. So how does this compare to rates of rejection or hemolysis in other sample types?  Well, across the institution, we know our overall redraw rate or unacceptable specimen rate is <1%, and about a third of those are due to hemolysis or hemolyzed samples, so overall hemolysis rates for all samples, certainly <0.25% for samples mostly collected by venipuncture compared to over 5% by IV start.

So the conclusions that we drew from this are that IV starts are inherently problematic; but it’s not to say that they can’t or shouldn’t be done, but you should assume that you’re going to have higher hemolysis rates in samples drawn through IV starts than you are other sample types. We saw some ability of the type of device used, and technique has some ability to improve the hemolysis rates; and certainly larger IVs are better, possibly extension tubing or discard volume may help; but even under the sort of optimal conditions that we could achieve at Mayo Clinic, the hemolysis rates varied from anywhere to 2 to 10 times higher than any other type of blood draw. And when it comes to IV-start blood draws, hemolysis rates of around 3% to 5% are probably about the best you can do; and again, the literature would suggest you can probably do a lot worse with IV-start draws.

So at this point during the conference, we would go back and re-poll the audience to see what they felt -- if they had changed their mind on the correct answer. So we present the same question regarding the practice of drawing blood during IV starts. Which of these 4 answers would they feel is correct and the answer, in my opinion, would be correct would be that IV blood draws have a hemolysis rate several-fold higher than any other type of blood draw and that, again, while you can do something perhaps with technique and devices to improve that, it’s always going to be an inherently problematic sample type.

So that’s the end of my case presentation. If you enjoyed this idea of getting phlebotomy knowledge and being challenged with scenarios and cases, I would encourage you to attend our upcoming phlebotomy conference, “Phlebotomy 2013: Advancing the Profession and Practice of Phlebotomy.” The conference will be October 3-4 here in Rochester, MN. For those of you who choose to come and choose to register online; at the point you do go to register online, you’ll see a little question box, and that is the point where you enter your question for me in terms of a question and a case, a scenario you would like me to use to design the upcoming “Phlebotomy Top Gun” presentation I’ll be doing in October in our phlebotomy conference. Also, when you fill out your survey, there will be an opportunity to also, at that point, make comments or ask questions; and if you have anything you like to hear about related to phlebotomy or any cases or scenarios in that venue you’d like to suggest, we’d be happy to hear from you then as well. Thank you for listening today and have a wonderful day.