Abstract

All infants undergoing an exchange transfusion at King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia, during a 7½-year period, were studied retrospectively. One hundred thirteen exchange transfusions were performed in 57 infants. Of 45 infants with hemolytic disease, three were stillborn. There were four early neonatal deaths and two infant deaths caused by hemolytic disease of the newborn. The 42 liveborn infants required 93 exchange transfusion for hemolytic disease of the newborn. In 15 infants requiring 20 exchange transfusions, the indication was hyperbilirubinemia of prematurity. Rh-hemolytic disease of the newborn was the most common etiology of severe anemia or hyperbilirubinemia requiring an exchange transfusion. Isoimmunization was due to anti-D in 34 patients, anti-c in three, anti-E and anti-c in one, anti-A and anti-B in four and anti-Jk^b in one patient. One of the cases of hemolytic disease of the newborn was due to G6PD deficiency, and in one case the cause was unknown. The incidence of moderate to severe Rh-hemolytic disease of the newborn was 7.0/1000 births contributing significantly to perinatal mortality (6.3%) and to early neonatal mortality (7.0%) in the hospital. Rh-hemolytic disease of the newborn is a problem in Saudi Arabia that could be reduced with improved antenatal and obstetric care and ante- and postnatal anti-D-globulin prophylaxis.

KEYWORDS: Erythroblastosis, fetal–Saudi Arabia, Exchange transfusion, whole blood

Introduction

A number of conditions may cause hemolytic disease of the newborn (HDN), but blood group incompatibility still remains the most important.1 Significant disease is primarily associated with the D antigen of the Rh-group (Rh-HDN) and with incompatibility of ABO factors (ABO-HDN), but more than 60 different red cell antigens are capable of eliciting an antibody response in a suitable recipient that could result in HDN. Regionalized programs, initially for postnatal and more recently antenatal prophylaxis of Rh immunization by intramuscular injection of human anti-D globulin, have significantly decreased the incidence of Rh immunization in many countries.2–6 A decrease in the average family size and improved ante- and perinatal care for the already immunized mother and her offspring have also contributed to decreasing mortality and morbidity from HDN.2 HDN remains, however, a major problem in countries without well developed anti-D-globulin programs and a large average family size. In Saudi Arabia, 30% of the deliveries in urban7 and 90% in rural areas8 take place at home. Furthermore, the fertility rate is high at seven births per woman.9 In a recent study of perinatal mortality in our institution, Rh-related HDN was found to cause 9% of early neonatal deaths.10 The purpose of this retrospective study is to describe the incidence, etiology, and outcome of moderate to severe HDN, and its impact on perinatal mortality in a referral center in Saudi Arabia.

Methods

Moderate to severe HDN in this study was defined as that degree of severity causing stillbirth or resulting in a liveborn infant's requiring exchange transfusion. All charts of neonates requiring exchange transfusions for hyperbilirubinemia admitted to the neonatal intensive care unit at King Faisal Specialist Hospital (KFSH) during the period, 1 January 1977 to 30 June 1984, were reviewed. All stillbirths due to erythroblastosis fetalis were included. Information regarding maternal reproductive history was collected from the mother's chart. Antenatal care included maternal antibody titers, ultrasonography, nonstress and stress tests, obtaining amniotic fluid for spectrophotometry, and estimation of fetal lung maturity. Intrauterine fetal blood transfusions were not performed at this hospital or elsewhere in the Kingdom. The indications for exchange transfusions were positive direct Coombs' test, hemoglobin at birth < 10 g/dl or cord bilirubin ≥ 5 mg/dl, or a rise of indirect bilirubin ≥ 0.5 mg/dl/hr.¹¹ The indirect bilirubin was at no point allowed to reach 20 mg/dl in term and 16 to 18 mg/dl in preterm infants. All exchange transfusions were performed via the umbilical vein. Active resuscitation and full supportive neonatal intensive care were given to all neonates as clinically indicated.

Results

A total of 113 exchange transfusions were performed during the study period. Fifteen neonates underwent 20 exchange transfusions for hyperbilirubinemia of prematurity, but none had evidence of hemolytic disease and all were excluded from further study. The 42 neonates with HDN that underwent 93 exchange transfusions remained in the study group. Three more stillbirths with erythroblastosis fetalis brought the total to 45 infants suffering from moderate to severe HDN. These 45 infants were born to 38 mothers. Three mothers gave birth twice and two mothers three times during the study period. In this study, Rh-HDN followed by jaundice of prematurity were the most frequent causes of severe anemia or hyperbilirubinemia requiring exchange transfusion (Table 1).

Twelve mothers had more than one Rh antibody detected in their sera. Evidence of isoimmunization against more than one antigen was found in two neonates (Table 2). Maternal antibody screen titers varied from 1/16 to 1/8192 and did not correlate with the severity of HDN. Four early neonatal deaths and two infant deaths were related to Rh-HDN. One neonate with HDN, previously reported,¹² died of triploidy 69,XXY and was excluded from the mortality figures related to Rh-HDN. The etiology of HDN in the population is shown in Table 2. The majority (38) of patients had Rh-HDN (84%); only five patients had other forms of blood group incompatibility. Four patients had ABO-HDN (9%) and one infant's mother was Jk^b-sensitized, probably following blood transfusion at the time of surgery for choriocarcinoma of the vagina four years earlier. One male neonate developed HDN secondary to G6PD deficiency. One female neonate had hemolytic disease of unknown origin that also had affected several older siblings, suggesting an inherited red cell enzyme deficiency.

Table 1. Etiology of severe hyperbilirubinemia or HDN requiring exchange
transfusions in infants in Riyadh (1977–1984)

Table 2. Etiology of moderate and severe HDN at King Faisal Specialist Hospital,
Riyadh, Saudi Arabia, 1977–1984

Thirty-eight neonates were born to 31 Rh-sensitized women, whose mean age was 31 years. The mean gestational age was 34 weeks. The previous reproductive history of these mothers, excluding the present pregnancy, is summarized in Table 3. The total number of births in the hospital during the study period was 5421, and the overall perinatal mortality at the hospital during the study period was 20.6/1000 births. The perinatal mortality in the study population was 184/1000 births. The incidence of Rh-related HDN that resulted in fetal death or in the need for exchange transfusion among infants born in the hospital was 7.0/1000 briths. Rh-HDN was responsible for 6.3% of the perinatal deaths and 7.0% of the early neonatal deaths in the hospital. The Apgar scores, need for supportive therapy, and outcome related to hemoglobin levels at birth in the neonates are presented in Table 4.

Table 3. Maternal reproductive history of 38 neonates with Rh-HDN

Discussion

So far no nationwide survey has been conducted in Saudi Arabia regarding HDN, and this report has to be interpreted carefully, as it concerns the experience in one institution on a selected population. As expected, Rh-HDN was the most common etiology for severe anemia or hyperbilirubinemia in infants requiring an exchange transfusion. A true incidence of HDN in Saudi Arabia cannot be derived from this study on a selected population, but the incidence of 7.0/1000 of moderate or severe Rh-HDN seen in this study is similar to a past incidence of serologic evidence of Rh-HDN in 4.3 to 9.1/1000 births in North America at a time when prophylaxis programs had just begun.^2,3 Mesleh on the other hand, in a study of perinatal mortality at the Armed Forces Hospital in Riyadh in 1982, could not attribute any of 63 perinatal deaths to Rh incompatability.¹³ Dramatic improvements were seen after introduction of Rh immune globulin prophylaxis in western countries.^2–6 In the province of Manitoba, Canada, the incidence of Rh immunization fell from 9.1/1000 total births in 1965–1969, before prophylaxis was commenced, to 2.2/1000 in 1975–1981, when both ante- and postnatal prophylaxis was available.² It is expected that by 1990 in Manitoba, of 16,500 births only five Rh-positive infants affected with erythroblastosis will be seen yearly.² Similar progress has been made in Sweden. In 1981 only four perinatal deaths among 93,678 total births could be related to maternal immunization, and only 206 exchange transfusions were performed.⁵ Similar results could be expected in Saudi Arabia if regionalized ante- and perinatal care including early detection of Rh-negative and isoimmunized mothers and Rh-immunoglobulin prophylaxis were made available to all women at risk.

The distribution of irregular antibodies among our infants was compared to that in a study from the United States before prophylaxis was introduced¹⁴ and to another study from Denmark,⁴ which has a well developed program for prophylaxis. Our findings fall midway between these studies with a relatively high proportion of Rh-anti-bodies other than anti-D or DC as a cause of HDN. The incidence of Rh-negative individuals among Saudi blood donors at King Faisal Specialist Hospital is 6.7%, similar to the 7% incidence among pregnant women in the southwestern part of Saudi Arabia and the 8% incidence among mothers that delivered at the Armed Forces Hospital in 1982.^13,15 The very high number of pregnancies experienced by each mother in the population predisposes to an isoimmunization to Rhantigens weaker than anti-D. It explains the relative importance of those Rh-antigens as a cause of HDN. The increased risk of immunization is also reflected in that 12 of the mothers had more than one Rh-antibody detected in their serum. In our experience, maternal antibody titers did not correlate with severity of HDN, as has been found before among multiparous women.¹⁶

Table 4. Symptoms, signs, invasive procedures, and outcome related to HGB levels at birth in 35 neonates with Rh-HDN

* Died at 40 days of age from multiple organ failure.

† Died from intractable diarrhea of infancy possibly related to HDN at 8 months of age.

§ Slightly delayed development at 15 months of age.

As seen in Table 3, the maternal population that gave birth to infants with Rh-HDN was one with a very high obstetric risk. These mothers were relatively old and had on the average been pregnant more than seven times; this pregnancy rate corresponds to the fertility rate of seven per women expected in Saudi Arabia.^7,9

The perinatal mortality rate (PMR) (227/1000 births) and the rate for reproductive losses (337/1000 pregnancies) as calculated from the maternal histories prior to the present pregnancy were extremely high. With the care offered in the hospital, the PMR for the study population dropped to 184/1000 births but was still very high compared to the overall PMR of 20.6/1000 births at the hospital during the study period. Rh-HDN contributed significantly to perinatal mortality (6.3%) and to early neonatal mortality (7.0%). All stillbirths were hydropic. The early neonatal deaths occurred among those severely affected neonates who had a hemoglobin of < 5.0 g/dl, were hydropic, had severe birth asphyxia, and required assisted ventilation (Table 4). A more active approach to these pregnancies using antenatal intrauterine blood transfusions could probably have saved some of the hydropic fetuses and neonates.^17–19 Recently, midtrimester intrauterine exchange transfusions of the fetus at 15 to 23 weeks' gestation have been described and could be of potential value for the most severely affected fetuses.²⁰ A survival of 75% has been reported with active antenatal management of hydropic fetuses.^17,18

This study shows that HDN is an important cause of perinatal mortality and morbidity at King Faisal Specialist Hospital and is likely to contribute significantly to perinatal mortality and morbidity in Saudi Arabia. HDN in the study population was mainly due to Rh-antigens, but hemolysis due to rare antigens and red cell enzyme defects was also seen. Seventy-three (73%) percent of HDN was anti-D related and could have been prevented by proper antenatal and postnatal prophylaxis with human anti-D globulin injections.

If anti-D prophylaxis becomes more widely implemented in the population, the number of infants with HDN due to other Rh antigens will remain the same, but increase in relative importance. A reduction in average family size would decrease the incidence of Rh-HDN caused by anti-c̄, anti-C, and anti-E. Rh-HDN will probably remain a problem for many years in Saudi Arabia, as the antenatal care in general is still poor, and 30% to 90% of the deliveries take place at home.^7,8 Every effort should be made to standardize perinatal care for all women, to detect the mother at risk, and to prevent her from becoming immunized. Presently, the best program for prevention of Rh-HDN isoimmunization of a whole population can be found in Finland, where every Rh-negative woman who becomes pregnant is registered through the antenatal care program into a central computer.²¹ This type of organization could serve as a model in Saudi Arabia. Regionalized perinatal care is needed to take care of the already immunized mother so she can receive antenatal care and be delivered at an institution appropriately equipped and staffed to handle her possibly critically ill offspring. Presently no facilities for treating the hydropic fetus antenatally are available in the country; and postnatal care, however aggressive, is of little value in these cases. A small multidisciplinary team able to perform intrauterine transfusions should be established at one regional center to which mothers with severely affected fetuses could be transferred. Plasmapheresis, for which facilities are available, could be used in selected cases in combination with intrauterine transfusions. A regionalized system for the care of the Rh-immunized mother in Saudi Arabia could be extended to include mothers with pregnancies at risk due to other conditions and thus improve the presently high perinatal mortality rate in the country.²² A nationwide survey to determine the incidence and treatment of HDN in Saudi Arabia is needed to be able to evaluate fully the contribution of HDN to perinatal morbidity, mortality, and handicap.

References

1. Naiman JL: Erythroblastosis fetalis. In: Oski FA, Naiman JL, eds. Hematologic Problems in the Newborn, 3rd ed. Philadelphia: W.B. Saunders Co., 1982;283–346.

2. Bowman JM, Pollock J.: Rh immunization in Manitoba: progress in prevention and management. Can Med Assoc J 1983;129: 343–5.

3. Hensleigh PA: Preventing rhesus isoimmunization. Antepartum Rh immune globulin prophylaxis versus a sensitive test for risk identification. Am J Obstet Gynecol 1983;146(7): 749–55.

4. Mickley H.: Hyppigheden af haemolytisk sygdom hos nyfödte (HSN) forårsaget af irregulaere antistoffer. Ugeskr Laeger 1983;145(23): 1763–5.

5. Medical birth registration in 1981. Statistics of the National Board of Health and Welfare. In: Statistiska medelanden HS, Stockholm, Sweden: Statistics Sweden; 1984

6. Palmer DC, Drew JH, Jaundice: a 10 year review of 41,000 live born infants. Aust Paediatr J 1983;19(2):86–9.

7. Chattopadhyay SK, Sengupta BS, Chattopadhyay C, Zaidi Z, Showail H: Maternal mortality in Riyadh, Saudi Arabia. Br J Obstet Gynaecol 1983;90(9): 809–14.

8. Bhatty MA, Al-Sibai H, Marwah SM: A survey of mother and child care in the Saudi community in Rabaiyah, Tarut Island. Saudi Medical Journal 1983;4: 37–43.

9. United Nations Economic Commission for Western Asia. In: The Population Situation in the ECWA region – Saudi Arabia. Beirut: Population Division ECWA 1979;1–27.

10. Ohlsson A, Serenius F: Perinatal mortality at King Faisal Specialist Hospital and Research Centre 1976–1981. In: 8th Saudi Medical Conference Abstracts. Riyadh, Saudi Arabia: Saudi Arabian National Guard, 1983;57.

11. Behrman RE, Kliegman RM: Hemolytic disease of the newborn. In: Behrman RE, Vaughan VC, eds. Nelson Textbook of Pediatrics. 12th ed. Philadelphia: W.B. Saunders Co. 1983;383–8.

12. Ohlsson A, Sheth K, Lewall D, Robinson C: Hematologic and ultrasonographic findings in triploid syndrome. The King Faisal Specialist Hospital Medical Journal 1984;4(3):281–4.

13. Mesleh R: Perinatal mortality of the Riyadh Armed Forces Hospital, Riyadh. Saudi Medical Journal 1985;6(2):135–44.

14. Giblett ER: Blood group antibodies causing hemolytic disease of the newborn. Clin Obstet Gynec 1964;7:1044–55.

15. Hartley DRW: One thousand obstetric deliveries in the Asir province, Kingdom of Saudi Arabia: A review. Saudi Medical Journal 1980;1:187–96.

16. Jacobs WM: Fetal survival rates in Rh-immunized gravidas: correlation of antibody titers and past obstetric history. Obstet Gynecol 1962;19: 806–7.

17. Bowman JM, Manning FA: Intrauterine fetal transfusions: Winnipeg 1982. Obstet Gynecol 1983;61(2):203–9.

18. Harman CR, Manning FA, Bowman JM, Lange IR: Severe Rh-disease – poor outcome is not inevitable. Am J Obstet Gynecol 1983;145(7):823–9.

19. Queenan JT: Current management of the Rh-sensitized patient. Clin Obstet Gynecol 1982;25(2):293–301.

20. MacKenzie IZ, MacLean DA, Fry A, Evans S: Midtrimester intrauterine exchange transfusion of the fetus. Am J Obstet Gynecol 1982;143(5):555–9.

21. Schneider J: Stand der Rhesusprophylaxe im Jahre 1982. Arch Gynecol 1983;235(1–4):411–18.

22. Ohlsson A: Better perinatal care in Saudi Arabia. Annals of Saudi Medicine 1985;5(3):169–78.