College of American Pathologists
Printable Version

  Feature Story


cap today

In two settings, plusses pile up for POC coag testing

September 2003
Karen Lusky

In a perfect clinical world, surgeons could obtain a patient’s coagulation test results as fast as they do their hemostats to control bleeding. And patients in Coumadin clinics would get their anticoagulation adjusted the instant their INR slipped out of therapeutic range.

In the real world, point-of-care coagulation testing and treatment algorithms are as close as it gets to that imaginary ideal, though everyone agrees the POC devices are far from perfect. Yet a growing number of studies show that the point-of-care approach appears to produce impressive clinical and cost outcomes in cardiac surgery and outpatient anticoagulation clinics.

“Rapid turnaround time for coagulation tests guides surgeons in controlling surgical bleeding and in avoiding unnecessary blood transfusion,” says Haifeng Mark Wu, MD, assistant professor of pathology and director of the clinical coagulation laboratory at Ohio State University, Columbus. Dr. Wu’s laboratory is considering using POC coagulation testing for cardiac surgery and organ transplantation in its new Ross Heart Hospital. “Without POC values, over-transfusion tends to occur in surgery,” he adds. “And everyone understands the importance of saving blood. Not only are blood components invaluable products that cost our hospital alone $10 million a year, it’s best for patients to avoid unnecessary transfusions.”

POC coagulation testing helps reduce blood transfusion because it gives surgeons the information they need to determine the etiology of bleeding during the procedure and in the immediate post-op period. “Microvascular bleeding is a clinical, visual diagnosis,” says David Jobes, MD, professor of anesthesia at Children’s Hospital of Philadelphia and the Hospital of the University of Pennsylvania, also in Philadelphia. “You look at a wound surface and make a judgment: Is that a discreet vessel bleeding? If not, your eyes cannot tell you anything else except it’s abnormal bleeding.”

So what did surgeons do before POC coagulation tests? “In a sense, clinicians used open-heart patients to test heparin dosing until the early 1970s, when they adopted activated clotting times as the first point-of-service test to monitor heparinization during surgery,” Dr. Jobes says.

“ACT has become a standard POC test during cardiac bypass surgery,” he adds. “People don’t always think about it in those terms, but that’s what it is. And the reason it’s done POC is that no laboratory test could help us manage the patient in a timely manner. With the huge dose of heparin administered, all of the lab tests would be blown out of the water, and the turnaround makes using the lab impossible.

“That’s the point of point-of-care testing—you get the test results in proximity to the problem, when you don’t have the option of treating a patient without just guessing what to do,” Dr. Jobes says.

The data are in
Outcomes data are beginning to quantify just how well activated clotting times and other POC coagulation tests curtail transfusion and the need for re-operation.

A study performed by Dr. Jobes and his colleagues at the University of Pennsylvania Medical Center found that POC coagulation testing performed during cardiac bypass virtually eliminated return to the OR for wound exploration due to nonsurgical coagulopathic bleeding. “And where bleeding was treated with blood products, the need for cryoprecipitate was markedly reduced,” Dr. Jobes tells CAP TODAY. The study’s findings were presented at the National Academy of Clinical Biochemistry’s second point-of-care symposium last year.

The researchers employed a combination of platelet counts (using a slightly smaller version of a standard cell counter, the Abbott Cell-Dyn 1400), prothrombin time test, and activated partial thromboplastin time test. “All of the POC testing,” says Dr. Jobes, “was done in the OR suite in conjunction with a patient-specific heparin dosing system and verification of the reversal of heparin with a thrombin time and a heparin-neutralized thrombin time” — all International Technidyne Corp. devices.

Once Dr. Jobes was certain the cardiac bypass patients had no residual heparin in their blood, he would order a PT, APTT, and platelet count. Depending on the test results, the surgical team treated patients before they left the operating room. Once patients left the OR, they received routine ICU testing: a lab-based PT, APTT, fibrinogen, and platelets.

But before patients went to the ICU, all the testing was done outside the OR complex by a POC lab 20 feet away, “if you can call what we had a lab,” Dr. Jobes says.

“We created it on a small cart with four wheels. The instruments that use whole blood are quite small—only one is big, and that’s the platelet counter, which is now the size of a cardboard box, 2.5 by 2.5 feet.” (Many operating rooms use Plateletworks by Helena Laboratories at the point of service to assess platelet count and function.)

The findings of the University of Pennsylvania Medical Center study, which looked at patients receiving point-of-care testing from June 1994 to May 1996, are in sync with those of other randomized trials: POC coagulation testing pays off for patients. For example, in 2001 Mayo Clinic conducted a pilot using a POC testing algorithm for open-heart surgery that resulted in a 60 percent reduction in the number of patients with excessive microvascular bleeding who received platelets, fresh frozen plasma, and cryoprecipitate, alone or in combination. The study is reported in Anesthesia (Nuttall GA, et al. 2001; 94: 773–781) and CAP TODAY (Titus K. 2001; 15[12]: 22–26).

Mayo Clinic now does activated clotting times during the bypass to monitor heparin levels. Its post-bypass algorithm includes platelets, PT, APTT, a thrombelastograph (TEG hemostasis analyzer by Haemoscope Corp.), and fibrinogen. “TEG and the PT and APTT are done POC,” says Paula Santrach, MD, co-director of clinical laboratories at Mayo Clinic and a member of the CAP Point-of-Care Testing Committee. “We have to use fresh whole blood, so the testing has to be close.” Platelet counts and fibrinogen are done stat in the lab removed from the OR, but the turnaround time is still about five to 20 minutes.

The TEG monitors the “thrombodynamic properties” of blood by getting it to clot in an environment that mimics sluggish venous flow, according to the manufacturer.

“During surgery, we do the PTs and APTTs [using a POC device], which takes about five minutes,” Dr. Santrach says. “The TEG system takes 15 to 20 minutes to get a partial tracing, and 60 minutes for a complete tracing.”

Making the case for savings
The clinical benefits of POC coagulation testing during surgery are replicable and dramatic. But you have to look at the big picture to see the cost savings. “That’s because if you just compare the cost per test, the POC is more expensive than doing the tests in the lab,” Dr. Santrach says.

OSU’s Dr. Wu concurs. “The POC reagent is expensive, and generally more so than in the lab because the volume of POC testing is low.” And QC, he notes, must meet CAP accreditation requirements. “Also, one needs to have a computerized interface between the POC instrumentation and the hospital laboratory information system,” he says. (The latter is not a CAP Laboratory Accreditation Program requirement.)

Yet the savings become obvious when you compare the cost of POC coagulation testing to the cost of blood products and return visits to the OR. For example, Dr. Santrach says, the cost of the tests Mayo Clinic does in its algorithm for open-heart surgery patients is lower than that of a unit of red cells—and much lower than the cost of six units of platelets or two units of fresh frozen plasma.

POC coagulation testing in the OR is cost-effective, and more so when accompanied by a treatment algorithm, Dr. Jobes says. “The problem is ... you can get into the ‘no, not my budget’ infighting when dealing with the cost center to which you want to attribute the expense. So you can end up going round and round until you find someone to agree to adjust budgets on both sides.”

Grappling with lack of standardization
On the clinical side, variability in test results among various devices remains a challenge. This lack of standardization stands in the way of developing a universal or standardized algorithm for managing patients during surgery.

“Since the devices aren’t standardized, how could one standardize the transfusion triggers?” Dr. Santrach asks. “They tried to [standardize] PT by using the INR, but there’s experience that shows that even that doesn’t correct for all the differences. In addition, there’s no way to standardize the APTT.”

Charleston Area Medical Center, West Virginia, found a tremendous difference in the amount of heparin the surgical team administered during cardiac bypass when it changed to the Bayer Rapidpoint Coag analyzer. Using the Bayer device to monitor heparin therapy in cardiovascular surgery, the team administers, on average, 10,000 fewer units of heparin per procedure than it did with its previous machine. “We have also seen less bleeding and less need for blood products with the new machine,” says Jason Newsome, lab technical specialist in cardiovascular perfusion at the medical center.

What accounts for such a difference in ACT results? “The various types of ACT are subject to variability due to dilution and temperature of the blood,” Dr. Jobes says. “And smaller volumes used in the test systems produce less variability secondary to those influences.”

With POCcoagulation testing, normal range varies markedly from one reagent vendor to another, agrees Bayer spokesperson Jan Price. In fact, Bayer typically discusses with its customers up front the need for the hospital to do studies that correlate the APTT point-of-care test with the lab APTT test. If the normal ranges are different, the hospital should change the reporting sheet for normal ranges.

“Every vendor has a different range that may [vary] from what’s used in the lab,” Price cautions. “They correlate, but they don’t match. That’s one of the biggest educational hurdles to using POC coagulation tests. Most vendors will help the hospital determine the correlation and normal ranges.”

Some hospitals, of course, would prefer to have a test that matches what they’re doing. But just as no two labs match each other on the same tests, no POC testing for coagulation matches directly to the laboratory’s testing for PT and APTT. And there is no similar lab device for ACT.

With ACT, you have to compare one vendor to the next, and in most cases they don’t match. Even multiple instruments provided by a single vendor won’t provide matching test results, according to Price. “The results are different enough that you want to set up ranges,” she says. “And the user just has to understand that the test results correlate but don’t match, so you have to set up normal ranges and different target times from device to device. You can’t just come out and say, ‘Shoot for a 500 [seconds] every time for ACTs,’ because each vendor may be different.”

Mayo Clinic confronted the practical reality of variability among devices when its POC instrument for PT and APTT testing was temporarily recalled by the manufacturer. Since the open-heart surgery team uses the PT and APTT for transfusion triggers, switching to a different piece of POC equipment would have required Dr. Santrach to revalidate the triggers and probably change them to match the previous ones. She decided against it. “We would have had to run things in parallel with fresh samples from real patients, which meant a good six-month project, and we were told we would probably get the POC equipment back by then,” she says. In the meantime, the laboratory is running the PTs and APTTs for the cardiac surgery team.

“You can’t just say you will [take action] at the upper limit of normal because that’s going to vary from one device to another—that’s really the underlying issue,” Dr. Santrach says. “And once you start switching instruments and reagents, you can get quite different results. And the upper limit of normal may not be a good transfusion trigger, as patients may have abnormal results but no bleeding.”

Whether or not the patient is bleeding is a trigger in and of itself, agrees Charleston Area Medical Center’s Newsome. A patient may have a functional platelet count of 30,000, with the key word being functional, he says, but if he or she isn’t bleeding, the physician probably won’t treat something that’s not causing a direct problem.

Overcoming other challenges
There are other challenges to POC coagulation testing in the OR. For one, the devices primarily use fresh whole blood from a fingerstick, venous sample, or heparinized arterial line. “So the preanalytical variables and the integrity of the sample are very important,” Dr. Santrach says. “If you don’t do that right, you can end up with the wrong answer.”

Furthermore, patients given the same dose of anticoagulant may have different results. One patient may respond to heparin with a marked prolongation of the activated clotting time—another patient may have a different response curve.

Some subsets of patients are more difficult to manage during the intraoperative period because they were already on low-molecular-weight heparin before surgery, or they may have received clot busting therapy in the emergency room or platelet aggregation inhibitors during cardiac catheterization in an emergent situation. But the POC coagulation test identifies the mechanism that needs treatment: protamine sulfate to reverse too much heparinization, vitamin K for warfarin overdose, or platelets for the platelet inhibitors.

Despite the potential stumbling blocks, POC coagulation testing and treatment algorithms in the OR are expected to move into the mainstream. “The advance of technology coupled with new and evolving treatment strategies will progressively introduce POC testing,” Dr. Jobes predicts. “Time-dependent procedures will require it, and the cost-revenue management strategies [hospitals use] will change,” paving the way for the testing.

The testing may also expand from the OR into the post-recovery period. Newsome reports that his medical center is “having conversations” about using more POC coagulation testing after cardiac surgery, at least during the first 12 hours of recovery.

Monitoring patients’ INR
Point-of-care INR testing in physicians’ offices and clinics is less dramatic than in the operating theater. Yet use of the method to manage chronic warfarin patients appears to improve clinical outcomes and reduce the cost of treating serious adverse events. It also generates revenue that adds up in a high-volume setting.

At OSU, which started point-of-service INR management in January, clinic staff asks patients a few standardized questions and obtains a fingerstick blood sample that it puts in the CoaguChek machine (Roche Diagnostics). Within about 90 seconds, the device produces a reading, which it feeds directly into the computerized system via a cable.

“The computerized software support program—called CoagClinic—then calculates the warfarin dosage using a computerized algorithm,” says Mark Wurster, MD, a hematologist and internist and director of anticoagulation at Ohio State University. Dr. Wurster is also co-founder and medical director of Standing Stone, a disease state management company that holds the patent on CoagClinic.

If the patient’s INR is too high or low, a message pops up on the computer screen telling the practitioner what the American College of Physicians says to do.

The average length of time for the visit is 12 to 15 minutes, and the clinic is able to bill separately for the laboratory test and the face-to-face encounter. By contrast, Medicare will not pay for telephone management of patients’ Coumadin therapy. The uncompensated service adds up for doctors’ offices and clinics: “The American Medical Association estimates that it costs about $7.50 every time someone picks up the phone, which includes overhead,” Dr. Wurster says.

Using the fingerstick POC test and face-to-face encounter, the clinic or physician’s office bills Medicare $5.42 or thereabouts for the test, which is essentially a break-even proposition. Medicare pays about $13 for a CPT 99211, the minimum evaluation and management code for incident-to-service provided by a nurse or pharmacist, Dr. Wurster says. “That may be different for private insurers, and Medicaid may pay a little less, but that’s the ballpark figure,” he adds.

If the hospital provides the services in the outpatient setting, it can bill under the outpatient prospective payment system using an APC code. “The APC for the service pays about four times the incident-to-billing and INR testing,” Dr. Wurster says.

The revenue produced by the point-of-care testing approach isn’t huge on a per patient basis. But Dr. Wurster notes that his former medical group at St. Luke’s Hospital, Kansas City, did 12,000 such visits a year, which would contribute a bit to the bottom line.

The face-to-face encounter also gives nurses and other health care professionals an opportunity to assess and teach patients. For example, when patients visit the OSU anticoagulation clinic, medical staff queries them about symptoms they might be having and determines if they are taking new medications or have changed their diet.

“It’s also essential to look at herbal formulations a patient may be taking, as these can interfere with clotting,” Dr. Wurster says. “For example, green tea is loaded with vitamin K and can reverse a person’s INR very quickly.” The clinic staff also knows to warn certain patient groups against overindulging in vitamin-K-rich collard greens at harvest time.

The beauty of the point-of-service testing method, in Dr. Wurster’s view, is that you catch an abnormal INR before the patient leaves the health care setting. “The software program gives you guidelines to consider holding one to three doses and calculates a dose reduction using a computerized algorithm,” he says. “It will also try to get to the new dose using the same pill size, which prevents patients from getting confused.”

A few bruises vs. inpatient hospitalization
To assess the impact of the POC approach, Dr. Wurster compared outcomes at his former practice at the multispecialty continuity clinic at St. Luke’s Hospital, which used POC INR testing, with the traditional approach at OSU, which was in place before January. In comparing the two groups of patients, Dr. Wurster examined three endpoints:

INR compliance (defined for the purpose of the study as the number of visits with INR results in the therapeutic range/total visits). The compliance rate was about 55 percent on average in the St. Luke’s test group using POC INR testing. That compared to 38 percent using the traditional approach at OSU (centralized lab testing and telephone management of anticoagulant dosing).

Treatment-related complications (defined as hospitalization and emergency room/outpatient encounters for a diagnosis related to anticoagulation when the INR was outside the therapeutic range). The experimental group had four complications during the study period from 2002 to 2003, all of them bruises on the lower extremities evaluated on an outpatient basis. The OSU patients had 13 inpatient admissions.

Costs. The experimental group’s complication treatment costs added up to $250, compared with $335,000 for the 13 inpatient admissions at OSU. For the test group, this represented a relative risk reduction of 89 percent.

In a followup national benchmarking study involving 30 outpatient settings (seven primary care, 16 subspecialty, and seven hospital-based clinics) using POC anticoagulation therapy and the CoagClinic software in 2002, Dr. Wurster and his cohorts found the average INR compliance rate to be 52 percent. That compared with the 38 percent compliance rate at OSU using the traditional telephone management approach.

The benchmarking study also looked at two other recognized studies involving POC testing in subspecialty practices—one with an INR compliance rate in the 30 percent range and another with an INR compliance rate of 45 percent. “The [latter two] studies were actually looking at a subspecialty approach to anticoagulation and were considered to represent a ‘gold standard’ approach,” Dr. Wurster says.

Dr. Wurster’s benchmark study found no statistically significant differences between INR compliance rates achieved in primary care sites and hospital-based clinics and subspecialty groups using the new method. In his view, that means the technology can be used in primary care offices just as well as in cardiology and other subspecialty settings.

Dr. Wurster presented the results of the benchmarking study at the Anticoagulation Forum in San Francisco in May. The study will also be published in an upcoming issue of the Journal of Thrombosis and Thrombolysis.

With the results of the benchmarking study now in, Dr. Wurster says researchers can hone in on questions about the gold standard for treatment based on the INR compliance rate.

POC INR testing isn’t all roses. As OSU’s Dr. Wu points out, the testing is cost-effective in the clinic setting in terms of managing patients but not in terms of the cost per test. Labs can, in fact, do the testing more cost-effectively because of the higher volume. “But in terms of patient outcomes, patient satisfaction and management, the POC coagulation testing is likely to be cost-effective,” Dr. Wu agrees. “To run the tests in the lab takes between 40 to 70 minutes overall.”

Another concern is that the POC devices use whole blood while the lab uses plasma to determine the INR. “But if the POC instrument vendor uses appropriate calibration/ validation to account for that, then accuracy is less of a concern,” Dr. Wu adds.

Even so, when the INR climbs too high, POC testing may not be as accurate. In that regard, “the POC instrument is good for a screening test, as most patients will have INR within the therapeutic range,” Dr. Wu says. But when patients have a Coumadin overdose, and the INR is very high on the POC instrument, the reading may not be accurate.

For that reason, the OSU coagulation clinic asks the hospital laboratory to repeat POC tests that produce an INR of 4.5 or higher. The patient waits at the clinic for the backup lab result. “Every time we have done that,” Dr. Wurster says, “the backup INR comes up in a range close enough that treatment recommendations are not changed significantly.”

Patients’ perception of the convenience of point-of-care testing may also affect its success in some settings. Dr. Santrach reports that Mayo Clinic’s experience in primary care shows that only about half of patients choose the point-of-service approach.

“Some people don’t necessarily have time for face-to-face encounters and prefer to just get their blood drawn and deal with it by telephone later,” she says. “It depends on how quickly the patients can get their blood drawn [by the lab] versus coming in for a visit to see the physician or nurse.”
Karen Lusky is a writer in Brentwood, Tenn.