Abstract
Abstract 4268
Glucose-6-phosphate dehydrogenase (G6PD) deficiency, affecting more than 500 million people worldwide, is one of the most common inherited disorders. There are 186 G6PD mutations published, with some that cluster within defined ethnic/racial groups, and in some populations more than one variant is encountered. These prevalent variants are generally considered to be benign in adults, but their morbidity, presenting as acute hemolytic anemia, in young children can be significant. However, comprehensive molecular characterization of ethnically prevalent G6PD mutants and their relative clinical significance is lacking.
In a region where the frequency of G6PD deficiency in the population is significant (>3%) and fava beans, green and dried, are a dietary staple, the risk of acute hemolytic crisis is significant. We studied 80 consecutively hospitalized unrelated Gaza Strip Palestinian children (ages 34–52 months, mean=39.5) for acute hemolytic anemia and found G6PD deficiency accounted for most cases (65 of 80 children; 81%) and almost all occurred after ingestion of green fava beans. Molecular analyses revealed 5 separate mutations, with 3 accounting for most cases. These were G6PD Mediterraneanc.563T, African G6PD A-c.202A/c.376G, and G6PD Cairoc.404C with relative allele frequencies of 0.33, 0.26, and 0.18 respectively. The detection of G6PD Cairo in 14 unrelated hospitalized children was surprising as this variant was encountered previously only in a single Egyptian individual. This unanticipated molecular heterogeneity of G6PD variants among Gaza Strip Palestinians provided a unique opportunity to test their relative clinical significance. We compared the prevalence of these variants to a control group consisting of 466 self-reported healthy, non-anemic, unrelated Palestinian adults (46 women and 420 men; 512 X-chromosomes). Within this cohort, the proportion with the G6PD deficient variants was 3.3%. The most common mutation was G6PD A+c.376A (allele frequency 0.016), comprising 47% of the variants detected while the G6PD variants Mediterranean and A- had allele frequencies of 0.0098 and 0.0078, respectively, comprising 29.4% and 23.5% of the detected variants. G6PD Cairo was not detected. The proportion of G6PD Mediterranean and G6PD A- variants in the general population are equivalent to their representation in the hospitalized children population. However, the number of G6PD Cairo variants in the hospitalized cohort was vastly larger than that observed in the general population indicating that these carriers are at a greater risk of hospitalization (Table 1). We found a dramatically increased risk of acute hemolytic anemia requiring hospitalization associated with all three variants. Fisher's Exact test was used to determine the association between the risk of hospital admission and presence of the variants and was significant for all at p<0.0001. Additionally, odds ratios for hospitalization are 41.1 (for G6PD A-) and 70.0 (for G6PD Mediterranean) compared to wild type [odds ratio was infinite for G6PD Cairo since it was not detected in the healthy population].
This is the first study comparing the relative clinical impact of three prevalent G6PD deficient variants in the same population and habitat to identify clinically significant effects due to the different genetic mutations.
. | G6PD . | |||||
---|---|---|---|---|---|---|
. | Mediterranean . | A- . | Cairo . | |||
. | + . | - . | + . | - . | + . | - . |
Hospitalized children | 26 | 54 | 21 | 59 | 14 | 66 |
Healthy population | 3 | 463 | 4 | 462 | 0 | 466 |
Odds ratio (95% CI) | 70.0 (20.1–367.4) | 41.1 (13.1–168.2) | ∞ | |||
Fisher's Exact p-value | <0.0001 | <0.0001 | <0.0001 |
. | G6PD . | |||||
---|---|---|---|---|---|---|
. | Mediterranean . | A- . | Cairo . | |||
. | + . | - . | + . | - . | + . | - . |
Hospitalized children | 26 | 54 | 21 | 59 | 14 | 66 |
Healthy population | 3 | 463 | 4 | 462 | 0 | 466 |
Odds ratio (95% CI) | 70.0 (20.1–367.4) | 41.1 (13.1–168.2) | ∞ | |||
Fisher's Exact p-value | <0.0001 | <0.0001 | <0.0001 |
Confidence interval (CI)
-NSR and MMS contributed equally to this work.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal