Grand Rounds

Rh disease: It's still a threat

By Kenneth J. Moise, Jr., MD

Once a common cause of perinatal death, Rhesus disease is now quite rare in pregnant women, thanks in large part to advances in ultrasound and DNA technology. But the fact that roughly 7 out of every 1,000 liveborn infants are delivered by Rh-sensitized women emphasizes the need for more vigorous preventive efforts and up-to-date management skills.

What's in a name? Sometimes it can signal important advances in the management of a life-threatening disease. The term hemolytic disease of the fetus/newborn, for instance, has replaced hemolytic disease of the newborn because modern diagnostic techniques now allow us to detect the disorder much earlier. Similarly, most experts use the term Rhesus alloimmunization rather than the older expression isoimmunization to describe the formation of maternal antibodies to the RhD red cell antigen-reflecting deeper insights into the pathophysiology of the disorder.

To prevent the disease, routine postpartum use of Rhesus immune globulin (RhIG) in Rh-negative patients was introduced in the United States over 40 years ago. A subsequent recommendation for routine antenatal use at 28 weeks' gestation was introduced 20 years later. Despite these efforts, a recent review of the 2001 birth certificates in the US by the Centers for Disease Control and Prevention indicates that Rh sensitization still affects 6.7 out of every 1,000 live births.¹ We are failing to prevent all cases of Rh alloimmunization for several reasons: inadvertent antenatal and postpartum omission, failure to use the drug for other antenatal indications, and insufficient dosing at delivery in the case of a large fetomaternal hemorrhage.²

Screening patients for Rh antibodies

All pregnant patients should undergo an antibody screen at the first prenatal visit. Those who are RhD-negative but weak Rh positive (formerly called Du-positive) are not at risk for Rhesus alloimmunization and therefore don't need RhIG. If an RhD-negative patient's initial antibody screen is negative, it's not necessary for physicians to do further diagnostic testing until 28 weeks' gestation. Unless the patient's partner is RhD-negative, administer a 300-µg dose of RhIG at this point in gestation. The American Association of Blood Banks (AABB) recommends a repeat antibody screen at 28 weeks, but the American College of Obstetricians and Gynecologists has left this to the discretion of the clinician.³ Although it has not yet been proven cost-effective, a repeat screen will detect the rare patient who becomes RhD sensitized early in pregnancy. If a repeat screen is obtained, the intramuscular RhIG injection can be administered before the patient is sent for venipuncture. The results of the antibody screen should not be affected by the administration of RhIG if the interval between procedure is less than 12 hours.

Since the half life of RhIG is approximately 16 days, about 1 in 5 patients will have a very low anti-D titer (usually 1:2 or 1:4) at the time of admission for labor at term.^4,5 On occasion, the cord direct Coombs at birth can also be positive as a result of the passively acquired maternal antibody; these cases don't cause symptoms in the newborn. At delivery, a cord blood sample should be tested for RhD typing. If the neonate is RhD-positive, administer 300 µg within 72 hours of delivery.

In about 0.1% of deliveries, fetomaternal hemorrhage in excess of 30 mL occurs; more than the standard dose of RhIG will be required in these cases. Routine screening of all women for excessive fetomaternal bleeding at the time of delivery is now recommended by AABB.⁶ Typically this initially involves a sheep rosette test that is read qualitatively as positive or negative. If negative, one vial of RhIG (300 µg) is given. If positive, the bleed is quantitated with a Kleihauer-Betke stain or fetal cell stain by flow cytometry. One should consult the blood bank pathologist to determine the number of doses of RhIG to administer.

The mechanism by which RhIG prevents sensitization is not well understood. Biochemical studies have revealed that the standard dose is insufficient to block all of the antigenic sites on the fetal red cells in the maternal circulation.⁷ Therefore, if RhIG is inadvertently omitted after delivery, the drug can offer some protection if administered within 13 days. It should not be withheld as late as 28 days after delivery if the need arises.⁸

Additional indications for RhIG are listed in Table 1. Three of these deserve special mention. Controversy surrounds the use of RhIG for threatened abortion. It's probably not indicated when only spotty vaginal bleeding occurs but it should be used in patients with significant clinical bleeding; doses can be repeated at 12-week intervals as necessary. Although a 50-µg dose can be used up to 13 weeks' gestation, most hospitals no longer stock this preparation and it costs about the same as the standard 300-µg dose. A second indication for RhIG that is often overlooked is blunt trauma to the maternal abdomen, particularly at the time of a motor vehicle accident. Finally, if 300 µg of RhIG are given late in gestation for external cephalic version or third- trimester amniocentesis for fetal lung maturity, a repeat dose is unnecessary if delivery occurs within 3 weeks, assuming there's no fetomaternal hemorrhage by maternal testing.

TABLE 1
Other indications for RhIG administration

Spontaneous abortion

Threatened abortion

Elective abortion

Ectopic pregnancy

Hydatidiform mole

Amniocentesis

Chorionic villus biopsy

Placenta previa with bleeding

Suspected abruption

Intrauterine fetal demise

Blunt trauma to the abdomen (including motor vehicle accidents)

External cephalic version

Tools of the trade

If a woman does become sensitized, RhIG is no longer effective. Fortunately, clinicians have a battery of tools to follow these pregnancies. But with all the rapid advances in this field, it's wise to consult with a maternal-fetal medicine specialist when faced with such patients.

Antibody titer. Perhaps the oldest available test is the maternal antibody titer. Once the maternal antibody screen returns positive for anti-D, a titer should be ordered. The titer is considered critical if it has been linked to an increased risk of fetal hydrops for a particular institution. An anti-D titer of 1:32 in the first affected pregnancy is often used. However, one should be cautious in interpreting antibody titers; they're only crude estimates of the amount of circulating antibody.

Titers today are performed much the same as they have been for decades. Preserved human red cells are used as the indicator to measure a biologic endpoint. These cells have a shelf life of 4 weeks, which means it's likely that subsequent titers will be performed using a different lot of indicator cells. In addition, large differences in titer in the same patient can be seen between laboratories because many commercial facilities use techniques like enzymatic treatment of the indicator red cells to overestimate the actual value of the titer. Additionally, newer gel technology will often produce titer results two or more dilutions higher than expected with older tube technology. Most labs no longer perform titers in parallel. With this method, stored frozen serum is thawed and subjected to the same indicator red cells and laboratory technician interpretation as the current maternal sample. A change of more than one dilution (i.e., 1:4–1:16) represents a true increase in maternal titer.

Ultrasound. Ultrasound has become the cornerstone of fetal therapy for hemolytic disease of the fetus and newborn. An early study should be obtained for dating because many of the parameters used to gauge fetal disease-including

OD₄₅₀ (the change in optical density), peak middle cerebral artery (MCA) Doppler, and fetal hematocrit-change with gestational age.

One of the most significant breakthroughs in recent years has been research that validates the peak systolic MCA Doppler velocity as a reliable screening tool to detect fetal anemia (Figure 1). The vessel can be easily visualized with color flow Doppler. Pulsed Doppler is then used to measure the peak systolic velocity of the MCA just distal to its bifurcation from the internal carotid artery. Enhanced fetal cardiac output and a decrease in blood viscosity contribute to an increased blood flow velocity in fetal anemia. Since the general trend is for the MCA velocity to increase with advancing gestational age, results are reported in multiples of the median (MOMs) much like serum alpha fetoprotein.

Initial retrospective studies suggested that a value of greater than 1.5 MOMs had a positive predictive rate for moderate-to-severe fetal anemia of 74% with a 10% false-positive rate.⁹ More recent prospective studies have revealed a positive predictive value of 53% and a negative predictive value of 98%, results comparable to the use of

OD₄₅₀.¹⁰ MCA Dopplers can be started as early as 18 weeks' gestation but are not reliable after 35 weeks. The advantage of serial MCA measurements is that they reduce the need for invasive diagnostic procedures like amniocentesis and cordocentesis by more than 70%.¹⁰

Fetal blood typing through DNA analysis. Paternal testing should begin early in the evaluation process of a sensitized patient. If you're certain who the father is and he's RhD-negative, there's no further need for testing. If on the other hand, he is RhD-positive, serologic testing for the other Rh antigens (C, c, E, e) and the use of race-specific population tables can determine with some degree of accuracy whether he is homozygous or heterozygous for the RhD antigen. This can be performed at most blood banks.

If the father is heterozygous for the RhD antigen, amniocentesis can be undertaken as early as 15 weeks to perform a fetal genotype for RhD. Studies have now shown that the RhD gene is complex, with many reported mutations and rearrangements in its structure. In these cases, the fetus can inherit these alterations from either parent. Both paternal and maternal blood samples should be sent to the reference laboratory with the amniotic fluid aliquot in an effort to enhance the accuracy of the fetal DNA result. If the patient's partner is not available or if there is a question regarding paternity, paired maternal titers can be tested 8 to 10 weeks apart. If there's a fourfold increase in titer or greater (i.e., from 1:4 to 1:16 or more), an RhD-negative fetal genotype by previous amniocentesis may be erroneous. A more recent development in fetal RhD typing involves the isolation of free fetal DNA in maternal serum.¹¹ In the United Kingdom, this technique has virtually replaced amniocentesis for fetal RhD determination in the case of a heterozygous paternal phenotype. It should be available for clinical use in the US in the near future.

Amniocentesis for

OD_450. Many centers have yet to adopt serial MCA Dopplers. As an alternative, some employ serial amniocenteses for

OD₄₅₀ in conjunction with serial MCA Dopplers until they are comfortable with the latter, as there is a well-established learning curve for the newer procedure. Although the tried and proven Liley curve was published in 1961, the recent modification by Queenan and coworkers (Figure 2) is more useful due to its accuracy at gestational ages of less than 27 weeks.¹² A rise or plateauing trend into the Rh-positive (affected) zone warrants more invasive testing through cordocentesis.

Amniocentesis for fetal lung maturity. Because the fluorescence depolarization technique (TDx-FLM, Abbott Laboratories, Abbott Park, Ill.) is so easy to perform, it has become widely accepted as the primary test for fetal lung maturity. The amount of polarized light that passes through the sample is inversely proportional to the final result for fetal lung maturity. High levels of amniotic fluid bilirubin due to fetal hemolysis can therefore cause a false elevation. Other tests for fetal lung maturity such as phosphatidylglycerol quantitation, lamellar body count, or the lecithin-sphingomyelin ratio, should be employed in cases of rhesus disease as these assays are not affected by excess bilirubin.

Cordocentesis. Introduced in the mid-1980s, direct access to the umbilical cord vessels by U/S-guided needle puncture allows clinicians to measure fetal hematocrit, reticulocyte count, bilirubin level, and direct Coombs. Initial enthusiasm for cordocentesis as a primary surveillance tool has waned due to the 1% incidence of fetal loss and a chance for enhanced maternal sensitization. Today, cordocentesis is reserved as a second-line diagnostic tool once amniocentesis or MCA Doppler suggests fetal anemia.

Clinical management

If you're managing a first sensitized pregnancy, observe these steps (Figure 3):

• Follow maternal titers every 4 weeks up to 24 weeks' gestation; repeat every 2 weeks thereafter.

• Once a critical value (usually 1:32) is reached, begin serial MCA Dopplers at about 24 weeks' gestation. Alternatively, initiate amniocenteses every 10 days to 2 weeks for

OD₄₅₀. In cases of a heterozygous paternal phenotype, perform amniocentesis at around 24 weeks to determine the fetal RhD status. Send maternal and paternal blood samples (usually in an EDTA "purple top" tube) with the amniotic fluid. If an RhD-negative fetus is found, no further testing is warranted.

• If the MCA Doppler is >1.5 MOMs or the

OD₄₅₀ value enters the Rh-positive (affected) zone of the Queenan curve, perform cordocentesis with blood readied for intrauterine transfusion for a fetal hematocrit of <30%.

• Initiate antenatal testing with non-stress testing or biophysical profiles at 32 weeks' gestation.

• If repeat MCA velocities remain <1.5 MOMs or

OD₄₅₀ values remain below the Rh-positive (affected) zone, perform amniocentesis at 35 weeks' gestation for

OD₄₅₀ and fetal lung maturity.

*If mature lungs are found and the

OD₄₅₀ value has not reached the Rh-positive (affected) zone, induce at 37 weeks' gestation to allow for hepatic maturity in an effort to prevent hyperbilirubinemia.

*If immature lungs are found and

OD₄₅₀ has reached the Rh-positive (affected) zone, treat with 30 mg of oral phenobarbital tid and induce labor in 1 week. This will accelerate fetal hepatic maturity and allow for more efficient neonatal conjugation of bilirubin. In these cases, a unit of packed RBC cross-matched to the pregnant patient should be prepared prior to delivery so it's available if the pediatrician must do an emergency neonatal transfusion.

*If immature lungs are found and

OD₄₅₀ is not in the Rh-positive (affected) zone, repeat the amniocentesis at 37 weeks.

If you are faced with a patient who had a previously affected fetus or infant that has had a transfusion (Figure 4):

• Maternal titers are not helpful in predicting the onset of fetal anemia after the first affected gestation.

• In cases of a heterozygous paternal phenotype, perform amniocentesis at 15 weeks' gestation to determine the fetal RhD status. If an RhD-negative fetus is found and paternity is certain, no further testing is warranted.

• Begin MCA Doppler assessments or serial amniocenteses for

OD₄₅₀ (using the Queenan curve) at 18 weeks' gestation. Repeat at 1 to 2 week intervals.

• If a rising value for MCA Doppler >1.5 MOMs or rising

OD₄₅₀ value into the Rh-positive (affected) zone of the Queenan curve is noted, perform cordocentesis with blood readied for intrauterine transfusion for fetal hematocrit of <30%.

What role should intrauterine transfusion play?

First introduced in 1963 by Sir William Liley, intrauterine transfusion (IUT) has withstood the test of time as the most successful fetal therapy.¹³ Initially, clinicians used the peritoneal cavity to do these transfusions, but then they discovered that hydropic fetuses absorbed transfused red cells poorly. As experience with cordocentesis accumulated, the direct intravascular transfusion (IVT) of donor red cells into the fetal umbilical vein at its placental insertion became the most common method of IUT in the US. Some US centers combine IVT with the original intraperitoneal method in an effort to prolong the intervals between procedures.¹⁴ In England, the intrahepatic portion of the umbilical vein is preferred as the site for IVT.

Once a physician decides to use IUT, special preparation of the donor red cells is required. Typically, a fresh type O, RhD-negative unit is screened to be sure it doesn't contain cytomegalovirus antibodies. The unit is crossmatched to the pregnant patient and then packed to a final hematocrit of 75% to 85%. This step allows a minimal blood volume to be administered to the fetus during the transfusion. The blood is then leukoreduced with a special filter and irradiated with 25 Gy to prevent graft-vs-host reaction.

The IUT procedure is usually performed in an inpatient setting, with the help of conscious sedation and local anesthetic. Prophylactic antibiotics are given but tocolytics are rarely required. We use continuous U/S guidance to find the umbilical cord insertion. After initial puncture, a sample of fetal blood is sent for hematocrit and other values. A small dose of a paralytic agent is administered into the umbilical vein to stop fetal movement. We then connect sterile tubing to the donor unit, which is attached to the procedure needle; blood is then infused by syringe through a closed system (Figure 5). The amount to transfuse is based on the estimated fetal weight using U/S and the initial hematocrit.¹⁵ A final sample is obtained to measure the fetal hematocrit when the procedure is finished. After the procedure, the patient undergoes continuous fetal monitoring until fetal movements resume.

Limited visualization of the umbilical cord insertion precludes successful IVT prior to 18 weeks' gestation, and most centers will not perform IUT's after 35 weeks. After the first IVT, the second procedure is usually planned 7 to 10 days later with an expected decrement in the fetal hematocrit of approximately 1%/day. Subsequent procedures are repeated at 2- to 3-week intervals based on fetal response and suppression of fetal erythropoiesis.

After the last procedure, the physician will usually schedule labor induction at 38 to 39 weeks' gestation to allow for fetal liver and lung maturity. The addition of maternal phenobarbital (30 mg tid for 7–10 days) may further enhance the ability of the fetal liver to conjugate bilirubin, thereby preventing the need for neonatal exchange transfusions.¹⁶ Many of these neonates don't require phototherapy and are discharged at the end of a routine postpartum stay. Breastfeeding is not contraindicated.

In centers that have a lot of experience in this area, the overall perinatal survival rate with IUT is 85% to 90%.¹⁷ Hydropic fetuses fare more poorly with a 15% decrease in survival, when compared to their non-hydropic counterparts. The success of serial IVT has resulted in a new phenomenon in the neonate. Suppression of fetal erythropoiesis in conjunction with the lack of the need for neonatal exchange transfusions results in the development of a profound anemia in the first month of life. With that in mind, be sure that weekly hematocrit and reticulocyte counts are performed until there is evidence of renewed production of red cells. "Top-up" red cell transfusions may be required in as many as 50% of cases, particularly if the neonate becomes symptomatic from its anemia.¹⁸

Data on the neurodevelopment of neonates transfused by IVT are limited. Most studies point to more than a 90% chance of intact survival.¹⁹ Hydrops fetalis does not seem to affect this outcome. Sensineural hearing loss may be slightly increased due to prolonged exposure of the fetus to high levels of bilirubin. A hearing screen should be performed during the early neonatal course and repeated by two years of life.

REFERENCES

1. Martin JA, Hamilton BE, Ventura SJ, et al. Births: final data for 2001. Natl Vital Stat Rep. 2002;51:1-104.

2. Hughes RG, Craig JI, Murphy WG, et al. Causes and clinical consequences of Rhesus (D) haemolytic disease of the newborn: a study of a Scottish population, 1985-1990. Br J Obstet Gynaecol. 1994;101:297-300.

3. Synder EL, Shoos Lipton K. Prevention of hemolytic disease of the newborn due to anti-D. Prenatal/perinatal testing and Rh immune globulin administration. American Association of Blood Banks Bulletin No. 98-2. Bethesda, Md. February 1998, pp 1-6.

4. Goodrick J, Kumpel B, Pamphilon D, et al. Plasma half-lives and bioavailability of human monoclonal Rh D antibodies BRAD-3 and BRAD-5 following intramuscular injection into Rh D-negative volunteers. Clin Exp Immunol. 1994;98:17-20.

5. ACOG. Prevention of RhD alloimmunization. ACOG Practice Bulletin No.4. Washington, DC; 1999.

6. Breecher ME Technical Manual of American Association of Blood Banks. Bethesda, Md: American Association of Blood Banks, 2002.

7. Kumpel BM. On the mechanism of tolerance to the Rh D antigen mediated by passive anti-D (Rh D prophylaxis). Immunol Lett. 2002;82:67-73.

8. Bowman JM. Controversies in Rh prophylaxis. Who needs Rh immune globulin and when should it be given? Am J Obstet Gynecol. 1985;151:289-294.

9. Mari G, Deter RL, Carpenter RL, et al. Noninvasive diagnosis by Doppler ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med. 2000;342:9-14.

10. Zimmerman R, Carpenter RJ Jr, Durig P, et al. Longitudinal measurement of peak systolic velocity in the fetal middle cerebral artery for monitoring pregnancies complicated by red cell alloimmunisation: a prospective multicentre trial with intention-to- treat. BJOG. 2002;109:746-752.

11. Randen I, Hauge R, Kjeldsen-Kragh J, et al. Prenatal genotyping of RHD and SRY using maternal blood. Vox Sang. 2003;85:300-306.

12. Queenan JT, Tomai TP, Ural SH, et al. Deviation in amniotic fluid optical density at a wavelength of 450 nm in Rh-immunized pregnancies from 14 to 40 weeks' gestation: a proposal for clinical management. Am J Obstet Gynecol. 1993;168:1370-1376.

13. Liley AW. Intrauterine transfusion of foetus in haemolytic disease. Br Med J. 1963;2:1107-1109.

14. Moise KJ Jr, Carpenter RJ Jr, Kirshon B, et al. Comparison of four types of intrauterine transfusion: effect on fetal hematocrit. Fetal Ther. 1989;4:126-137.

15. Mandelbrot L, Daffos F, Forestier F, et al. Assessment of fetal blood volume for computer-assisted management of in utero transfusion. Fetal Ther. 1988;3:60-66.

16. Trevett T, Dorman K, Lamvu G, et al. Does antenatal maternal administration of phenobarbital prevent exchange transfusion in neonates with alloimmune hemolytic disease? Am J Obstet Gynecol. 2003;189:S214.

17. Schumacher B, Moise KJ Jr. Fetal transfusion for red blood cell alloimmunization in pregnancy. Obstet Gynecol. 1996;88:137-150.

18. Saade GR, Moise KJ, Belfort MA, et al. Fetal and neonatal hematologic parameters in red cell alloimmunization: predicting the need for late neonatal transfusions. Fetal Diagn Ther. 1993;8:161-164.

19. Hudon L, Moise KJ Jr, Hegemier SE, et al. Long-term neurodevelopmental outcome after intrauterine transfusion for the treatment of fetal hemolytic disease. Am J Obstet Gynecol. 1998;179:858-863.

DR. MOISE is Upjohn Distinguished Professor of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, University of North Carolina School of Medicine, Chapel Hill, N.C. He is also Director of the UNC Treatment Center for Fetal Hemolytic Disease. A pioneer in management of the alloimmunized pregnancy, he has performed approximately 300 intrauterine transfusions in the last 13 years.

Ken Moise. Grand Rounds: Rh disease: It's still a threat.

Contemporary Ob/Gyn

May 1, 2004;49:34-48.