Glucose-6-phosphate dehydrogenase deficiency neonatal screening: preliminary evidence that a high percentage of partially deficient female neonates are missed during routine screening.

OBJECTIVES: To provide preliminary evidence that the currently employed semiquantitative method of screening for glucose-6-phosphate dehydrogenase (G6PD) deficiency can only detect infants who are totally deficient for G6PD and misses all cases of partial G6PD deficiency.
SETTING: General population: 2150 randomly selected blood samples from the Blood Donation Department, Speliopouleion General Hospital, Athens, Greece. Neonate population: 2000 samples from neonates (50% male; 50% female) in maternity hospitals in the greater Athens area. High risk population: a total of 545 individuals from 133 families in the Athens area, the minimum acceptance criteria being the parents and any brother or sister.
METHOD: Blood specimens from neonates or adults were collected and either spotted and dried on special filter paper (Schleicher and Schull 2992, Darmstadt, Germany) or used in tubes after being heparinised. For the quantitative evaluation of G6PD enzyme activity, the Quantase G6PD screening kit (Quantase Limited, Perth, UK) was used. Quantase G6PD controls (Quantase Limited) were used at three levels of G6PD. These controls are rated at 24, 30, and 37 degrees C. Alternatively, we used the Sigma G6PDH controls (Sigma Chemical Company, St Louis, USA) which are rated at 30 and 37 degrees C. The assay was performed according to the instructions included in the kit with the modification for haemoglobin normalisation.
RESULTS: General population: 36 females who were classified as having normal enzymatic activity with the semiquantitative test, were classified as partially deficient with the quantitative test. Neonate population: using the quantitative test, the percentage of G6PD deficient neonates in this population was 5.5%, compared with 3.17% reported in routine screening using the semiquantitative method. High risk population: the quantitative method detected 28 cases of total or partial G6PD deficiency in sisters of males with known total deficiency. The semiquantitative
CONCLUSIONS: A considerable amount of partially G6PD deficient female neonates (heterozygotes) are undetected and classified as having normal enzymatic activity using the semiquantitative method, which uses a cut off of 2.1 U/g haemoglobin (Hb). The use of a fully quantitative G6PD screening kit is proposed, employing the automated haemoglobin normalisation and a cut off of 6.4 U/g Hb. Any neonate with an activity below this mark should be regarded as G6PD deficient, and all preventive measures should be taken.