Objectives: To determine the incidence of clinically significant allo-antibodies in antenatal care (ANC) patients, and make recommendations on laboratory management of such cases in similar settings in Zimbabwe.
Design: A retrospective study.
Setting: Harare Central Hospital, a tertiary medical centre in Harare.
Subjects: Patients attending the ANC clinic at Harare Central Hospital.
Main outcome measures: Blood group tests, allo-antibody screen, development of haemolytic disease of the newborn.
Results: 3,000 patients were grouped and screened and 96.7% were found to be Rhesus positive, 0.5% were Rhesus Du positive and 2.8% were Rhesus negative. An overall antibody incidence of 1.7% (n = 50) was obtained, 1.0% (n = 30) of which were strongly positive and 0.7% (n = 20) were so weakly positive so that no antibodies could be identified. Antibodies identified from those patients with strongly positive antibody screen were anti-D 13.3% (n = 4), anti-E 6.7% (n = 2), anti-Jsb 3.2% (n = 1), anti-Lea 23.3% (n = 7) and anti-Leb 20% (n = 6). Antibodies of unknown specificity were detected from 20% (n = 6) of the patients. Four (13.3%) of the specimens were insufficient for antibody identification. Clinical records of those patients with a strongly positive antibody screen were examined and anti-D and anti-Jsb were observed to have caused severe to fatal Haemolytic Disease of the Newborn (HDN). The four anti-D positive cases resulted in two still births and two jaundiced babies. The single anti-Jsb positive antibody case resulted in an intra-uterine death. Antibodies that are generally considered of no clinical significance did not cause HDN in this study.
Conclusion: Anti-D remains the most important allo-antibody causing HDN, regardless of the availability of anti-D immunoglobulin for prophylaxis. Only Rhesus D negative women and those who have clinically significant antibodies need have repeat antibody screens during the rest of the pregnancy. In line with the current policy of screening all patients at booking, the policy on repeats is not clear and was not evident in this study.