Poor testing, dosing dog fetomaternal bleeds

Karen Titus

September 2013—If Mark Brecher, MD, were compiling a greatest hits list of medical successes of the 20th century, there’s one advance he’d be sure to include: the introduction of Rh immune globulin in the late 1960s to prevent the Rh sensitization of Rh-negative mothers who deliver an Rh-positive baby.

The number of patients affected may be small, but it’s not insignificant, says Dr. Brecher, chief medical officer, LabCorp, and adjunct professor, University of North Carolina, Chapel Hill. If the mother’s immune system becomes sensitized to the Rh-D antigen, the result can adversely affect a current pregnancy or the mother’s ability to carry a future pregnancy safely to term. In the United States, some 13 percent of mothers are Rh-negative, while the chance that they’ll deliver an Rh-positive baby is about 60 percent, says Dr. Brecher, a member of the CAP Transfusion Medicine Resource Committee. With Rh immunoprophylaxis, the risk of the mother being alloimmunized by a D-positive fetus dropped from roughly 13 percent to less than 0.1 percent.

A typical dose of RhIg, 300 μg, covers 30 mL of whole blood or 15 mL of red cells spilled by the baby. When a larger fetomaternal hemorrhage occurs, however—as it does in about 0.3 percent of cases—physicians can’t rely on the standard dose. In these cases, laboratories need to determine the volume of fetomaternal hemorrhage and calculate how many vials of RhIg need to be given.

If only they could get it right.

Arriving at the correct dosage would, in Disney parlance, “be a dream come true.” But a dream it remains. Laboratories continue to have difficulty with fetomaternal hemorrhage, or FMH, primarily because they make calculation errors and use a test—usually the acid-elution Kleihauer-Betke—that does a poor job of quantifying fetal red cells in maternal circulation.

Not that clinicians are rising up in protest. “The average OB-GYN has no clue,” says Kenneth Moise Jr., MD, professor of OB-GYN and of pediatric surgery, University of Texas Medical School at Houston. “The average obstetrician, when asked about fetomaternal hemorrhage, says, ‘That’s a lab problem. I expect them to educate me on what to do.’”

Unfortunately, plenty of data have demonstrated ongoing failure to manage FMH more precisely.

In 2009, the Transfusion Medicine Resource Committee analyzed data on FMH testing from the 2006 and 2007 CAP Surveys. The results did not inspire confidence. The main finding, as noted in a Transfusion commentary by three committee members (published online Feb. 27 and scheduled for print publication in September), was that even though nearly 67 percent of participating labs used the standard method, found in the AABB Technical Manual, for calculating RhIg dose, almost 21 percent recommended an incorrect dose (11.5 percent too much, 9.2 percent too little).

“We kept getting these crazy responses. Either people were dosing too high or too low,” says Dr. Brecher, who at the time was the chief editor of the manual. “When we tried to focus in on what the problem was, we would tell people to calculate it using the [AABB recommendations]. We found people were still making mistakes.”

The committee then tried to take test error out of the picture. Even when labs had the equation set up for them—they were given figures for the percent of red blood cells detected and for the maternal blood volume—“we still had about 16 to 17 percent of people who could not do the calculation right,” says Dr. Brecher.

“Which was in some ways frightening,” he adds.

The Transfusion Medicine Resource Committee, in an effort Dr. Brecher led, responded by creating an online RhIg Dose Calculator. It’s a Microsoft Excel program that requires only that users plug in the numbers—the dose is then calculated automatically. The calculator is posted on the CAP Web site (www.cap.org, under Committees & Leadership, Transfusion Medicine Resource Committee, and then Transfusion Medicine Topic Center), and it does not require validation as LIS software, says S. Gerald Sandler, MD, a TMRC member and lead author of the Transfusion commentary. Most importantly, he says, the calculator works.

If you use it.

TMRC analysis, also in 2009, of subsequent Surveys (which are done twice a year) found that nearly 45 percent of laboratories ignored the new tool and continued to calculate RhIg dosing manually; of these, almost 17 percent recommended an erroneous dose. Only 1.6 percent of labs that used the calculator, in contrast, submitted a dosing error.

The calculator appears to be growing in popularity, Dr. Brecher says, noting that it shows up regularly in literature citations. In addition, “We know that it’s been downloaded thousands of times from the CAP Web site”—in the last 12 months, almost 1,500 times. Dr. Sandler, however, says that “a significant” number of laboratories in the Survey group don’t use it. “I can’t explain why,” says Dr. Sandler, who is also professor, pathology and medicine, and medical director of the blood bank, Department of Laboratory Medicine, MedStar Georgetown University Hospital, Washington, DC.

Even when the calculator is used, it’s only as good as the numbers that are plugged into it. Unfortunately, those numbers are overwhelmingly the product of the Kleihauer-Betke test.
The committee’s initial 2009 analysis alluded to this issue, noting that flow cytometry was more precise than the acid-elution assay for quantifying FMH. At that time, 3.8 percent of labs used flow, while almost 74 percent used acid-elution methods. (The remaining 11.5 percent said they used “other” methods.)

The 2012 analysis showed the numbers haven’t budged. A whopping 96.1 percent of participants used an acid-elution method, while only 3.9 percent used flow cytometry. Of the laboratories using an acid-elution assay, more than 46 percent recommended an inadequate dose of RhIg, and just over 29 percent recommended an overdose.

Kleihauer-Betke is not a good test, to put it bluntly. “It’s not specific enough,” says Meghan Delaney, DO, MPH, assistant medical director, Puget Sound Blood Center, and assistant professor, University of Washington, Seattle. “People could have widely variant responses, and therefore don’t get the right answer.”

Most laboratories get very little practice in running the test. So when they “pull out the kit, they’re going to be reading the instructions again to figure out how to do the stain,” says Dr. Moise. If it’s not done correctly, it’s difficult to count the adult versus the fetal cells. “It’s just not going to be a very accurate test,” he says.

The answer would seem to be obvious: Switch from the KB test to flow cytometry. Says Dr. Sandler: “It’s not as though we have a virus and no test to detect it.”
(What about improving the Kleihauer-Betke test? Says Dr. Brecher: “In theory, anything can be improved, but I haven’t seen anything suggesting that this test can be improved.”)

A certain torpor has set in, however, which has kept that solution visible but out of reach, like an object in an art museum.