Expires: February 12, 2023
Brad Karon, M.D., Ph.D., is Co-Director of Laboratory Services for the Department of Laboratory Medicine and Pathology at Mayo Clinic in Rochester, Minnesota. He is also a Professor of Laboratory Medicine and Pathology.
Contact us: firstname.lastname@example.org.
Thank you, Dr. Pritt. To those of you who have not attended a phlebotomy conference in the past, the title of today’s session, “Phlebotomy Top Gun,” may seem a little odd. I hope I’ll be able to explain to you what “Phlebotomy Top Gun” is all about and what we try to do at our annual phlebotomy conference.
I have no disclosures relevant to today’s presentation.
Every year at our phlebotomy conference, I present a talk that’s always called “Phlebotomy Top Gun.” The format for “Phlebotomy Top Gun” is a case-based presentation. I solicit from attendees of the conference the questions, issues, and areas of concern that they would like to hear about. Using the questions submitted by the audience, I present these as case-based scenarios. Then, using an audience-response voting mechanism, the attendees of the conference vote on what they believe the correct answer to the case or question to be. I then present the collective experience of the Mayo practice, as well as any published data and evidence I can find related to the topic.
At the end of the case presentation, the audience votes again. For each case, I can see whether I’ve been able to change anyone’s mind on the question at issue by presenting the data, evidence, and information relevant to this particular topic. This is an actual case from last year’s “Phlebotomy Top Gun” presentation. In last year’s presentation, we were addressing the issue of order of draw and whether the order of draw is still relevant in today’s modern phlebotomy practice. The question I posed, based on a question submitted by an attendee last year, was the following:
Which of the following statements describes the importance of drawing a serum tube before the K EDTA tube?
At this point in the phlebotomy conference, the attendees will vote on what they believe to be the right answer, and at the end of the case we will re-poll. I’ll reveal what I believe to be the right answer, and I’ll get to see if I changed anyone’s mind.
A great number of cases, but not all, start with a review of the CLSI guidelines, which have a great deal of information about recommended procedures for blood collection. The relevant guideline in this case, which is often the case, is GP41, which addresses collection of diagnostic venous blood specimens. The guidelines cover order-of-draw recommendations, which continue to be present in the guidelines as blood culture first, followed by citrated tubes, serum tubes, heparin, then EDTA, with special emphasis on not collecting the EDTA tube before the serum tube. The guidelines do clarify that although these recommendations are valid for both liquid anticoagulants (glass tubes), which are really not used other than for a few special coagulation-type tests in our practice, and the more commonly used plastic tubes that have lyophilized additive in them. The guidelines go on to say, for additives not listed, that you consider the type of additive and where it fits into the order of draw. The guidelines suggest not collecting serum clot activator tubes before a citrate tube for coagulation study to avoid carryover of the clot activator into the coagulation tube. They specifically mention in the guidelines that sodium (Na) carryover from Na citrate or heparin tubes into a serum tube is not really significant. For trace metals, follow the manufacturer’s instructions for various tubes and collection devices. That’s a review of the CLSI guidelines, which is where I start with many of the cases or questions presented at the phlebotomy conference.
In the case of order of draw, this is one of the rare situations where we have a CLSI guideline to consider, but we also have the European Federation for Clinical Chemistry Working Group on preanalytical phases of testing, which did an extensive review (referenced at the end of this presentation for those of you who would like to read it). This European group did an extensive review and had a consensus conference on a number of preanalytical phases or preanalytical questions about blood collection. This document goes on to list various problems that have been linked to incorrect order of blood draw, including 1) hypernatremia due to Na citrate or Na EDTA, which if you recall a slide ago, CLSI said was not an issue. The European Federation suggested that it could be an issue. 2) Hyperkalemia due to K EDTA contamination, which is really what we are going to focus on—K contamination from K EDTA tubes into serum. 3) Hypocalcemia, low magnesium (MG), zinc, alkaline phosphatase, other enzymes—again, these are due to EDTA contamination of serum tubes. 4) Poor coagulation due to the transfer of anticoagulants, clot activator for example, into citrated tubes. 5) Clot activator interference with coagulation testing, as I mentioned, and 6) dilution effects if you pour one tube into another.
The European Federation guidelines go beyond what CLSI has done, which provides you guidelines for what you should do in a few references. Again, I encourage people to read the European Federation guidelines, which is a very nice document that goes into why this is important, what the mechanisms of K EDTA contamination are, and how much evidence exists that they do or don’t actually happen. The European Federation went over three potential mechanisms for anticoagulant contamination for K EDTA contamination of serum tubes when the order of draw is not followed. 1) Direct transfer or pour off. Again, obviously, if you need serum and you collect K EDTA and you pour that into a serum tube, then you’re going to have a contaminated sample. This is obviously a very bad practice and we don’t expect to see that where we have trained phlebotomists. 2) Backflow during closed loop (vacutainer) collection. If you snap the K EDTA tube onto the vacutainer adaptor, there is potential for some blood that has K EDTA on it to stick to the vacutainer adaptor, and that could get transferred into the serum tube. We’re going to show you in a minute that the studies really don’t suggest that actually happens with lyophilized K EDTA. This is what may have changed going to plastic tubes with a lyophilized reagent—like lyophilized anticoagulant, we don’t see that there is carryover from backflow on a vacutainer adaptor. 3) Contamination by syringe. If you use a syringe to dose the tubes in the incorrect order (dosing the K EDTA first, then the serum), there is the potential to carry over some of that K EDTA from the purple top K EDTA tube to the serum tube. Data on this, whether it really happens and how often it happens, is somewhat spotty or controversial, but really that is what we are going to talk about. If this doesn’t happen anymore, then maybe order of draw is no longer important.
However, in general what the European Federation for Clinical Chemistry Working Group for Preanalytical Phase concluded was: order of draw was not important for routine venipuncture using a vacutainer adaptor, but contamination still did occur during syringe transfer. Because it did, rather than training one way for routine vacutainer and another for syringe, it was easiest to follow order of draw as defined in CLSI guidelines and by this guideline group all the time.
These studies are really those cited in the European Federation Working Group document. One study was of 11 healthy volunteers, drawn by an experienced phlebotomist. In this experiment they drew a serum tube, a K EDTA tube, then a second serum tube to try to measure in a healthy outpatient if there was any measurable EDTA carried over from the middle tube into the second serum tube. Well, the answer was no. EDTA was not detectable in any of the serum samples when looking at the measurements that would be affected by K EDTA contamination: K obviously, then the metals (calcium, magnesium, and zinc), the enzymes that need calcium, and metals like alkaline phosphates. There was no difference between the serum tube before the EDTA and the serum tube drawn after. So while it was a small study of 11 healthy volunteers, it really didn’t find that contamination occurred when you drew an EDTA tube before the serum tube.
A larger study cited 57 outpatients attending an anticoagulation clinic and 58 healthy volunteers—now we’re getting to 120 or so patients involved—and multiple experienced phlebotomists. They drew a serum, then a citrate, then a second serum, and the other set of volunteers used a serum, K EDTA, and then a serum. And again similar results: there was no difference between the mean K, Na, calcium, MG, or phosphorus in the serum collected first and the serum collected after an EDTA or Na citrate tube. This is consistent with the smaller study I just showed, which shows that order of draw probably doesn’t matter in routine outpatient phlebotomy using a vacutainer adaptor. We are not seeing carryover between tubes.
In contrast, here’s a study done on hospital inpatients in one United Kingdom hospital. For one month they simply said, well, let’s look at all inpatient blood serum K, where the K is over 6 mmol/L, and let’s just see if we can detect measurable EDTA in those. So they had 117 samples with serum K over 6 mmol, and 25% of those had detectable EDTA, which was defined as > 0.1 mmol/L in this study. So that would suggest that carryover is fairly frequent. Just selecting serum samples with high K, we’re seeing that a lot of these samples—almost a quarter of them—have some amount of EDTA that presumably had to come from a K EDTA tube. There is not another logical explanation for where EDTA in a serum sample would come from. Again, most of these were inpatients and almost all of these patients, when they had their K measured a second time that started over 6, returned normal and had no EDTA. Again, this is further evidence that this was likely K EDTA contamination in these patients with high K serum values. EDTA concentrations all were measurable, but the more EDTA, the higher the K and the lower the calcium, zinc, MG, and alkaline phosphate. This was more evidence that what they were seeing was a variable amount of K EDTA getting into serum samples somehow. The study didn’t say how it got there, it just showed it got there and that it caused high K, low calcium, MG, zinc, and alkaline phosphate.
The same group of authors went on to do a multi-center study to say, well, this was one UK hospital, now let’s look at five hospitals. These hospitals were all in the UK as well. They looked at these five different hospitals over a month, looking at all serum K over 6 and seeing how many of those had measurable EDTA to see if the first study was a fluke. They looked at different types of tubes: Sarstedt versus Greiner tubes, and one site had BD. Each site had between 100 and 300 of these serum samples with K over 6. You can just look at the last column to see the percentage that had EDTA contamination varied from 1% to almost 7%. That very first study showed almost 25% of these serum samples had K over 6 contaminated with EDTA, and here we’re seeing it still happens but at lower numbers, where 1 to 7% of these serum samples with high K were contaminated by K EDTA.
What’s our experience been like here at the Mayo Clinic? We also do a similar thing. We have a number of rules built into our middleware system in the lab to detect possible contamination with either K EDTA or IV fluid, but specifically there is a flag that goes up if the K is over 6 mmol on an inpatient. Then it will look at other values that are requested with that K and alert for potential contaminations. When this flag goes up for potential contaminations, a caregiver is contacted, and if this value is not expected for that patient, we’ll redraw it. Most of the time we find that the K comes back relatively normal if there is a contamination suspected. Most of the time that K contamination is from K given intravenously to the patient and not suspected to be K from EDTA. Unusual cases that we review go to a resident or fellow on call, and we review those weekly in our call rounds. It seems like once a quarter or once a month we have a sample with a very high K that doesn’t look real, and often a resident or fellow will ask us to measure it, if they didn’t order a calcium or MG, and it will come back with similar results to the studies that I showed: a very high K and a very low calcium or MG, then it will be redrawn with a normal K and it will look like K EDTA contamination. We don’t go back to determine if it was drawn with a syringe or a vacutainer, but we suspect what we’re seeing here is what’s been shown in other studies: that some low rate, between 1% and 5%, of these very high serum K samples have somehow been contaminated with K EDTA. We suspect in happened during syringe draws that were done with improper order of draw, where the K EDTA tube was dosed first and then the serum tube.
Conclusions for order of draw are that evidence suggests that K EDTA contamination does not occur when you have a closed draw that is a vacutainer needle and vacutainer adaptors with plastic tubes and lyophilized K EDTA. It doesn’t really look like there’s backflow still happening where K EDTA gets stuck to the vacutainer adaptor and goes into the serum tube if it’s drawn after an EDTA tube for a vacutainer collection. The greatest data on syringe draws comes from these five UK hospitals, the study I showed, where K EDTA contamination is seen in somewhere between 2 to 15% of all samples that had very unusual K, calcium, MG, or zinc concentrations. I want to make clear that 2 to 15% of all serum samples are contaminated with K EDTA, but when you select for something like K over 6 or very low calcium or MG values, then you see that 2 to 15% of those samples do appear to be contaminated K EDTA. So it seems that order of draw is still important, and with syringe draws particularly you can carry over K EDTA from the purple top to the serum tube if you do not dose them in the correct order or essentially follow the recommended order of draw. At Mayo Clinic we are still rarely seeing samples that look like they have K EDTA contamination. In our practice we were able to figure out that a vacutainer or syringe collection could be happening more often than we’re detecting with our middleware rules and systems. The underlying assumption I’m left with is that order of draw is still important: you should draw the serum before the K EDTA purple top tube. The underlying assumption, from the data and evidence out there, is that contamination happens primarily with syringe draws, and it doesn’t seem that backflow contamination with a routine vacutainer collection happens. But I would agree with the European Federation Working Group guidelines. If you’re going to do it for syringe draws, it’s probably just easiest to follow the recommended order of draw for all blood draws.
During the phlebotomy conference we would then go back and re-poll the audience, after we had given our experience and reviewed the published data and guidelines that are out there. We would go back and re-poll our audience with the same question: Which of the following statements describes the importance of drawing a serum tube before the K EDTA tube? Again, same answers:
The first three of these are plausible, and the fourth I just threw in for fun, because our lab equipment can’t tell the difference—K is K. We would allow the audience to re-poll, and I would present what I believe to be the answer, which is number three, that order of draw is still important because during syringe collections contaminations with K EDTA may rarely occur if the order of draw is not followed.
So that’s the end of my presentation, and, again, this is case one of four or five that I presented at last year’s “Phlebotomy Top Gun” presentation at our phlebotomy conference.
These are references for those who would like to read more. I appreciate your time today.