Feature Story

Partial drawback: iffy APTTs lead to tube’s exit

What laboratories should do

Karen Sandrick

When the hospital laboratories of Hahnemann University and the Medical College of Pennsylvania in Philadelphia were getting ready to merge in 1997, Jamie Siegel, MD, then director of both coagulation laboratories, had to synchronize her procedures. One change involved switching from standard full-draw (4.5 mL) to partial-draw (2.7 mL) sodium citrate tubes.

But Dr. Siegel worried that the move to the partial-draw tubes might affect the accuracy of activated partial thromboplastin times when monitoring patients on unfractionated heparin therapy. The preanalytical variables that influence APTT were already known-how much time passes before a blood sample gets to the laboratory, how long the laboratory takes to process the sample, which reagent is used in testing. So when her colleagues Kathleen Sazama, MD, JD, Vanlila Swami, MD, and David Bernard, MD, PhD, agreed the partial-draw tube might be another preanalytical variable, Dr. Siegel set out to compare APTTs from heparinized patients using both types of tubes.

Dr. Siegel evaluated a total of 74 blood samples drawn in the two institutions and found "huge differences" in some patients with prolonged APTTs, she told CAP TODAY. While there was no apparent difference in the results of coagulation testing among normal individuals, the mean abnormal range in APTT for one hospital was 64.6 seconds on samples obtained with the full-draw tube and 53.6 seconds with the partial-draw tube. At the other institution, the mean abnormal range was 62.7 seconds for the full-draw and 50.1 seconds for the partial-draw tubes (Am J Clin Pathol. 1998;110:184-187). "If you look at individual numbers, some samples had up to 30-plus seconds of a difference, which can be significant when you’re dealing with heparin," she says.

Getting an APTT that’s 20 or 30 seconds shorter than the gold standard value obtained with full-draw tubes translates into higher heparin doses than necessary, says Dr. Siegel, who now is associate professor of medicine in the division of hematology at Robert Wood Johnson University Hospital, New Brunswick, NJ. "We didn’t know if that, in fact, would increase bleeding risk for patients, but we knew it theoretically could," she points out. And because of that possibility, the Medical College of Pennsylvania did not make the switch.

Partial-draw sodium citrate tubes have been around for many years. The tubes were developed to reduce the amount of blood drawn from patients who must undergo laboratory testing regularly. The partial-draw tubes have the same configuration as their full-draw counterparts-13 ¥ 75 mm; they collect less blood because they have less interior vacuum.

Until recently, no one suspected partial-draw tubes introduced preanalytical variability. Mark Hosler, MD, medical director of the laboratory at Southeast Missouri Hospital, Cape Girardeau, has been using the tubes for five or six years and hasn’t heard about adverse clinical bleeds.

But during the last 18 months, studies conducted in Dr. Hosler’s laboratory as well as others across the country have revealed two major problems: APTTs computed from samples drawn with partial-draw tubes were falsely shortened in up to 15 percent of patients on unfractionated heparin, and platelet counts tended to be artificially low.

According to Becton Dickinson, of Franklin Lakes, NJ, the largest supplier of partial-draw sodium citrate tubes worldwide and sponsor of the clinical studies, these problems affect only small segments of the patient population. Decreased APTT occurs with patients on unfractionated heparin; samples from normal blood donors or patients on warfarin therapy do not exhibit artificial change in APTT with partial-draw tubes. And the falsely low platelet count pertains to the one percent to two percent of patients who have EDTA-associated platelet clumping, explains Jeffry B. Lawrence, MD, vice president for medical affairs.

Becton Dickinson has not received complaints from clinician customers about bleeding as a result of the shortened APTT. "There is enough noise in the APTT system that clinicians may not notice a bleeding problem. And if clinicians end up giving 10 percent more heparin than needed, that may not be enough of an increase for anyone to pick up an increased rate of bleeding," explains John Brandt, MD, senior clinical research pathologist, Lilly Research Laboratories, Indianapolis, and a member of the CAP Coagulation Resource Committee.

Becton Dickinson nevertheless has launched a massive educational campaign about the drawbacks of the partial-draw tubes, and it will withdraw the tubes from the market on Sept. 30.

"The potential for leading to inappropriate monitoring of unfractionated heparin therapy is real," Dr. Brandt says. "If you’re basing your heparin therapy on samples drawn with these tubes, you may end up giving more heparin than needed, and the more heparin you give, the higher the risk of bleeding."

BD first noticed discrepancies in APTT while running a series of followup tests after correcting a problem with tube stoppers. The company had discovered that certain lots of stoppers used in its citrate tubes had begun to release divalent cations into the citrate solution and, in the process, caused shortening of both prothrombin time and APTT. After changing to new stoppers, BD found that the variability in prothrombin time disappeared, as did the negative bias in APTT with some APTT reagents. However, in initial studies other reagents still showed some APTT shortening with the partial-draw tubes. Therefore, the company decided to ask independent hospital laboratories to study the performance of coagulation tubes across the entire spectrum of today’s highly sensitive coagulation reagents.

The studies were all conducted in the same way, Dr. Lawrence points out. Phlebotomists in each institution obtained blood samples from normal blood donors, individuals with various hematological disorders, and patients receiving unfractionated heparin or warfarin, using the full-draw 4.5-mL tube and two types of partial-draw tubes-the 2.7 mL and 1.8 mL.

The studies measured clotting time for normal donors and for warfarin and heparin patients using the PT reagents Thromboplastin C Plus (Dade), Thromboplastin HS (Sigma), RecombiPlastin (Hemoliance), and Innovin (Dade), and the APTT reagents ACTIN, ACTIN FS, and ACTIN FSL (all Dade), FSL (Sigma), Thrombosil I (Hemoliance), SynthASil (Hemoliance), APTT-C (IL Test), and Platelin (Organon Teknika). The studies also counted platelets in K3EDTA, sodium citrate, and CTAD (citrate, theophylline, adenosine, and dipyridamole cocktail) tubes and tabulated PT and APTT values for each of these tube categories.

The most critical part of these studies, Dr. Lawrence says, determined the risk each of the three types of tubes posed for misclassifying heparinized patients. The studies calculated a heparin therapeutic range using an antifactor Xa chromogenic substrate assay, and for comparison evaluated the standard therapeutic range of 1.5 to 2.5 times the midpoint of the reference range. The studies then classified patients as overanticoagulated if their APTT was above the therapeutic range or underanticoagulated if it fell below.

"We generated for each one of these reagents a number that we call the misclassification rate, which indicates how many patients would have been classified one way with the full-draw tube and another way with the partial-draw tube," Dr. Lawrence explains.

The misclassification risk for the 1.8-mL partial-draw tube ranged from 6.7 percent with the SynthASil reagent to 14.8 percent with the ACTIN FSL reagent. The risk for the 2.7-mL tube was 9.3 percent with ACTIN FSL and 10 percent with the other two reagents.

BD has since learned that the shortened APTT is due to two aspects of the configuration of the partial-draw tube: a greater amount of headspace and less vacuum. "There is more space between the top of the column of blood and the bottom of the stopper in partial-draw tubes than in standard draw tubes. This increased headspace contributes to platelet activation, which releases platelet factor 4. Platelet factor 4 neutralizes heparin in the blood sample, and that leads to the artificially shortened APTT result," Dr. Lawrence says.

Because there is less vacuum in the partial-draw tubes, it takes longer for the tubes to fill with blood. "Platelets, as they are being drawn from the patient, are exposed to a certain shear stress because of the difference between the pressure in the tube and the pressure in the patient’s veins," Dr. Lawrence adds. "With the lower evacuation of the tube, the platelets are exposed to shear stress for a longer period of time in the partial-draw tubes. So the platelets become activated, which leads to platelet clumping; the clumped platelets are not counted by the automated analyzer, and you can get artificially low platelet counts."

Prothrombin time is not affected, probably because it involves fewer biochemical reactions than APTT. "With prothrombin time you’re basically looking only at tissue factor pathway activation, whereas with APTT, there are many more biochemical steps that can be influenced by platelet activation," Dr. Lawrence notes.

Dr. Hosler’s laboratory, which serves seven small hospitals and 120 physicians as well as 288-bed Southeast Missouri Hospital, participated in four studies for BD last year that compared clotting times in plastic tubes, full-draw tubes, and its in-house partial-draw tubes and found a wide range of APTT values. In one study, according to Julie Wengert, head of the blood bank and hematology, "APTTs with the full-draw tube were in the therapeutic range, but then as you went down to the 2.7- and 1.8-mL draws, the APTTs dropped down sometimes 20 to 25 seconds. Some patients ran 60 seconds in the full-draw tubes and 35 or 40 seconds in the 2.7- and 1.8-mL draws."

In another study, the laboratory noticed that CTAD tubes were consistent across the three draw volumes for heparinized patients. "The 2.7-mL and 1.8-mL tubes were equivalent to the 4.5-mL draw," Dr. Hosler says. "But the clotting times decreased significantly when the sodium citrate tubes were used," he adds.

Now that BD has similar data from studies on more than 400 patients, it is taking a number of steps to get the word out. It has sent a letter to all its customers worldwide that explains the variability in platelet counts and in APTT monitoring of patients on unfractionated heparin.

"Many people have never thought of the coagulation tube as the source of variability that could be clinically significant," Dr. Lawrence says. "We want to inform the hematology and clinical communities that for certain patient populations, the coagulation tube can be a variable."

To be sure this information reaches those who use the tubes, BD is changing the labeling so it warns against using partial-draw tubes for platelet counting or with samples from heparinized patients. The company established a Web site (www.bd.com/coag) to provide another outlet for the information, and it will present data from its studies at professional society meetings around the world and in peer-reviewed journals. And at the end of September, it will discontinue the product.

BD is recommending that laboratories use one of its alternative products in lieu of the partial-draw tubes, such as its pediatric or CTAD tubes (see "What laboratories should do," page 50). The pediatric tubes are smaller than standard 4.5-mL full-draw sodium citrate tubes and, consequently, will draw lower volumes of blood. Although the CTAD tube has the same configuration as the partial-draw tube, its cocktail of theophylline, adenosine, and dipyridamole platelet inhibitors prevents platelet activation.

The company also hopes to increase awareness of the effects of variability in the preanalytical phase. "One thing we’ve learned over the last 20 years is that, as instrumentation has gotten better and more reliable and more reproducible, and as coagulation reagents have gotten more sensitive, the preanalytical phase has become more important as a potential source of error," Dr. Lawrence says.

"Customers have a right and, frankly, an obligation to ask the vendor of whatever coagulation tube they’re using, ’Does this have documented clinical performance in the patient population that I’m studying?’ Laboratories have an obligation to determine that the specimen containers they’re using for coagulation are free from bias that potentially could be clinically significant," Dr. Lawrence asserts.

Early in his career, Dr. Lawrence, like other laboratory directors, concentrated on getting the best possible data, achieving the highest sensitivity and specificity, greatest precision, and greatest accuracy. "That is what we considered to be our work product," he says.

In hospitals and other health care settings today, laboratories are redefining their work product to include patient outcome. "Of course, we want to have improved performance of the APTT and platelet count, but ultimately that’s not an end in itself. The end we’re really focusing on is improved anticoagulation therapy and patient management. That’s where laboratory physicians and other laboratory professionals can make the biggest contribution to medicine," he adds.

And that is where concepts like the risk of misclassifying heparinized patients come in. The difference between an effective and a toxic dose of heparin is narrow. Standard clinical practice for patients on IV heparin is to calculate the initial dose on the basis of body weight, then readjust six hours later depending on heparin’s effect on clotting. "Whether the doctor increases, decreases, or maintains the current dose of heparin is based on where the APTT of the patient fits in relation to the therapeutic range. Therefore, the accuracy of the APTT result is critical for patient management," Dr. Lawrence notes.

Misclassification was defined in terms of the gold standard BD fulldraw glass tube because published results obtained with that tube formed the basis of clinical recommendations for treating heparinized patients. For a coagulation tube to be free of misclassification, it should yield the same result as the full-draw tube when identifying patients as falling inside or outside the heparin therapeutic range.

"Based on the clinical studies we’ve done with the products that will remain on the market-the full-draw tubes, pediatric tubes, and CTAD tubes-all will be associated with a zero misclassification rate," Dr. Lawrence says. "So laboratories will be able to use them with confidence that the results they give to the doctor will be accurate."

Karen Sandrick is a freelance writer in Chicago.