Identifying ABO incompatibility in newborns: selective vs automatic testing.
Two alternative approaches to identifying ABO incompatibility have been suggested. One approach is to look for ABO incompatibility only in those infants who become sufficiently jaundiced to warrant diagnostic studies to determine the cause of jaundice. For term babies, this includes any infant with clinical jaundice in the first day of life or a bilirubin level greater than 171 [mu]mol/L (10 mg/dL) anytime thereafter. The trend toward early discharge is a potential problem for this selective approach to testing.
An alternative approach is to screen every infant born to a mother with type O blood to determine blood type and DAT status. All infants who are DAT positive can then be followed closely for hyperbilirubinemia. This routine testing will identify a larger number of infants, most of whom will have no clinical disease, and could lead to unnecessary laboratory tests and phototherapy.
Two studies that looked specifically at the automatic testing approach concluded that there are no appropriate screening tests capable of selecting those ABO incompatible infants who will have hemolytic disease severe enough to require treatment.[4,5] It was concluded that screening for hemolytic disease due to ABO incompatibility was not cost-effective, and the importance of clinical observation of the newborn was emphasized. However, neither study compared the outcomes of infants tested automatically with those tested selectively.
The purpose of this study was to compare the effectiveness of automatic and selective testing in identifying ABO incompatibility in infants of mothers with group O blood types. The following specific questions were asked: (1) Does the selective approach lead to higher peak bilirubin levels in infants and more readmissions for phototherapy? (2) Does the automatic approach lead to more laboratory tests and phototherapy?
The study took place at the St Paul--Ramsey Medical Center in St Paul, Minnesota. Infants in the normal newborn nursery who were cared for by the family practice service were the study population. Infants in the intensive care nursery were excluded from the study. The study infants received care from family physician residents and faculty. A phototherapy nomogram by Cockington is prominently displayed in the nursery and used as a guide for management of hyperbilirubinemia.
In 1989 the Ramsey pediatric service began automatically testing the cord blood of infants of mothers with type O, Rh-positive blood to determine type, Rh-factor, and DAT status. (The cord blood of all infants of Rh-negative mothers are routinely typed so that the mothers may be treated with anti-Rh immunoglobulin.) Although the family practice service did not initiate routine testing, many infants on the service were mistakenly tested.
Labor and delivery records for 1990 and 1991 were reviewed to identify infants of women with type O, Rh-positive blood. Charts of the infants who received care from the family practice service were reviewed retrospectively for demographic information, laboratory testing, and jaundice recognition and management. The management of infants with automatic and selective cord blood testing were compared. Statistical analysis was done with chi-square and Student's t tests.
Three hundred one infants of mothers with type O, Rh-positive blood were identified. One hundred thirteen (38%) infants had their cord blood type, Rh-factor, and DAT status determined automatically without a physician order. Of the 188 infants whose cord blood was not tested automatically, 34 (18%) had their cord blood tested selectively by physician order.
Infants whose cord blood was tested automatically are compared demographically with those who were tested selectively (Table 1). The two groups of infants did not differ significantly by sex, race, gestational age, weight, feeding method, or follow-up appointment time. The length of hospital stay was slightly longer for infants tested automatically. Twenty-nine (26%) of the infants tested automatically had blood type A or B, and 14 [TABULAR DATA OMITTED] (12%) were DAT positive. Eighteen of the infants tested selectively had blood type A or B, and 13 were DAT positive.
Jaundice management of the two groups of infants did not differ greatly (Table 2). Similar percentages of infants in both groups had jaundice noted by the nurse and physician. The number of infants having total bilirubin level were similar between the two groups.
Three infants had total bilirubin levels above 342 [mu]mol/L (20 mg/dL). All were DAT positive and had been tested selectively. One of these infants was first noted to be clinically jaundiced at 17 hours of age, and phototherapy was started at 23 hours of age. Despite appropriate treatment, the infant's bilirubin reached a Table 2. Jaundice Management of Infants Tested Automatically and Selectively
Tested Tested Automatically Selectively (n=113) (n=188) Jaundice noted by nurse (%) 45 35 Jaundice noted by physician (%) 25 22 Total bilirubin determined (%) 32 28 Hemoglobin determined (%) 7 6 Mean peak total bilirubin level, 168 (9.8) 174 (10.2)
Phototherapy used (%) 3.5 4.3 Readmission for phototherapy (%) 1.8 0.5
peak of 366 [mu]mol/L (21.4 mg/dL) at 68 hours of age. The other two infants were breast fed and noted to be clinically jaundiced at 25 and 30 hours of age. Bilirubin levels were checked on these infants at the time of the next routine blood drawing, a delay of 15 and 18 hours, respectively. At that time the levels were already greater than 342 [mu]mol/L (20 mg/dL).
The percentage of infants receiving phototherapy was similar. Of the 12 infants receiving phototherapy, 11 were blood group A or B, and 10 were DAT positive. Five of the infants receiving phototherapy were noted to be clinically jaundiced within the first 24 hours of life. All were clinically jaundiced within the first 48 hours of life. Two of the three infants readmitted for phototherapy had undergone routine testing. None of the infants received an exchange transfusion.
The selective and automatic approaches to identifying ABO incompatibility had similar outcomes in this population. The selective approach did not result in identifying infants with significantly higher peak total bilirubin levels than those of infants in the group automatically tested. Furthermore, readmissions for phototherapy occurred no more frequently in the selectively tested group. Although the three infants with total bilirubin levels above 342 [mu]mol/L (20 mg/dL) were tested selectively, the extreme elevation in one infant occurred despite early recognition of jaundice and optimal management. The other two cases could have potentially been avoided by checking bilirubin levels promptly when jaundice was first noted rather than waiting for the next routine blood drawing. Automatic cord blood testing did not result in increased use of laboratory testing or phototherapy. Therefore, either approach appears to be reasonable.
A major disadvantage of automatically testing infants of blood type O, Rh-positive women is the cost. The charge for determining cord type and Rh factor and performing the DAT at Ramsey is $33. Charges elsewhere are much higher. When the proportion of women with type O, Rh-positive blood (42%) and the annual number of births in the United States (3.7 million) are considered, substantial costs could be incurred by automatic testing of cord blood.
The findings in this study, however, may not apply to all clinical settings, particularly in facilities where earlier discharge of newborns is common. Despite national trends toward sending infants home sooner, most of the babies in this study were discharged on the third day of life. Only 6 (2%) infants were discharged on the day of birth and 72 (24%) on the second day of life. All of the infants requiring phototherapy were clinically jaundiced before 48 hours of age. In clinical settings where there are more frequent discharges before the third day of life, automatic cord blood testing may be preferable. Nevertheless, automatic testing cannot substitute for close clinical follow-up, particularly in babies discharged before 48 particularly in babies discharged before 48 hours of life.
An important limitation of this study was its small size. In particular, few infants were readmitted for phototherapy. A true difference in jaundice management or outcome may have been missed because of a type II error.
The Ramsey family practice service has continued its use of selective cord blood testing of infants of women with group O, Rh-positive blood. Physicians in similar clinical settings who automatically test these infants should consider adopting a selective approach.
[1.] Fanaroff AA, Martin RJ, eds. Neonatal-perinatal medicine: diseases of the fetus and infant. 5th ed. St Louis, Mo: Mosby-Year Book, 1992:242, 954-5.
[2.] Avery GB, ed. Neonatology: pathophysiology and management of the newborn. 3rd ed. Philadelphia: J. B. Lippincott Co, 1987: 653-4.
[3.] Rudolph AM, ed. Pediatrics. 18th ed. Norwalk, Conn: Appleton & Lange, 1987:1037.
[4.] Han P, Kiruba R, Ong R, et al. Haematolytic disease due to ABO incompatibility: incidence and value of screening in an Asian population. Aust Paediatr J 1988; 24:35-8.
[5.] Levine DH, Meyer HBP. Newborn screening for ABO hemolytic disease. Clin Pediatr 1985; 24:391-4.
[6.] Cockington RA. A guide to the use of phototherapy in the management of neonatal hyperbilirubinemia. J Pediatr 1979; 95:281-5.
[7.] Newman TB, Easterling MJ, Goldman ES, Stevenson DK. Laboratory evaluation of jaundice in newborns: frequency, cost and yield. Am J Dis Child 1990; 144:364-8.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||blood typing and testing|
|Author:||Madlon-Kay, Diane J.|
|Publication:||Journal of Family Practice|
|Date:||Sep 1, 1992|
|Previous Article:||Are Papanicolaou smears enough? Acetic acid washes of the cervix as adjunctive therapy: a HARNET study.|
|Next Article:||Primary care research funding sources.|