This is the first study performed to evaluate the prevalence of G6PDd in a large number of individuals through enzymatic activity accompanied by genotyping analysis in malaria endemic regions on the Pacific coast in Colombia. This approach allowed to preliminarily evaluate the frequency of allelic variants associated with G6PDd in municipalities from this region of Colombia.
This study found 6.56 % of analysed individuals showing G6PD deficient activity including both intermediate and severe deficiency. This overall frequency is lower than the 10 % prevalence reported previously for Colombia in a recent review which includes studies from Latin America [2, 21], the 12 % reported in Buenaventura for 242 individuals [21], 14.7 % in 508 individuals in Turbo [25] and the 12.7 % found in Cali [23]. However, the results found in this study were higher than previous reports including a G6PDd prevalence of 3.1 % in a mestizo population in Bogotá [22] and 2.2 % in 500 Amerindians and Afro-Amerindians individuals in Medellin [25]. The differences in G6PD activity found in various Colombian studies are probably due to composition of the different populations evaluated; a higher G6PDd prevalence in Afro-descendants is likely, particularly in malaria endemic regions.
The A? and A+ allelic variants are the most often reported globally associated with G6PDd, although the A? variant is the one most frequently evaluated in G6PDd studies in populations of African descent. In this study, 15.23 % of the individuals had G6PD A genotypes, mostly A? (11.26 % A? and 3.97 % A+), which is in agreement with a previous study showing that A? was the predominant G6PDd allelic variant in some countries in Latin America and the Caribbean [2]. Furthermore, the frequency of the G6PD A genotypes found in this study was similar to a recent study performed in Honduras that detected 11.81 % of individuals with A? genotype and 4.27 % for A+ [12]. These similarities are probably due to the population structure evaluated in these two studies, mainly Afro-descendants located in coastal regions. In contrast, the frequency of G6PD A? genotype found in a study performed in Venezuela was lower (2.56 %) [13].
Tumaco (26.66 %) and Buenaventura (24.39 %) showed the highest allelic G6PDd frequencies (Fig. 3). The population of these two municipalities is mainly Afro-descendants (~90 %). Interestingly, these municipalities are malaria endemic sites in Colombia with a predominance of P. vivax malaria infections. The lowest proportion of individuals with G6PDd variants was found in Quibdo.
When the enzymatic activity was evaluated, a greater proportion of G6PD deficient females compared to deficient males was found. However, from genotyping analysis, the proportion of hemizygous males with G6PDd variants was higher than females (heterozygous or homozygous). In Honduras, a greater proportion of females with G6PDd allelic variants were reported compared to hemizygous males [12].
In some individuals it was founded G6PDd probably associated with the presence of A + and A? genotypes. A significant association between deficient enzyme activity and the presence of G6PDd genetic variants was found (p = 0.04098). Eight individuals presented functional G6PDd and displayed the corresponding genetic variants, however, these variants were not detected in the remaining 20 individuals that showed reduced G6PD enzyme activity. A plausible explanation is that G6PDd in these individuals is probably associated with other variants not evaluated in this study, which have not been characterized for the Colombian population and could be identified in the future by DNA sequencing of all exons.
This is the first report regarding frequency of G6PDd allelic variants in regions with endemic P. vivax malaria on the Colombian Pacific coast. Additionally, this study shows an important association between genotype and phenotype in these regions. Previous studies did not report allelic variants in individuals with reduced G6PD activity [22]. Here the A? genotype was detected in most of the samples with severe G6PDd (Table 3). It is surprising that no reports on haemolysis associated with the use of PQ use are available from these regions despite its frequent use and apparent good treatment compliance.
Used for treatment of both P. falciparum (the majority of malaria cases in Africa) and P. vivax (70 % of malaria cases in Colombia) PQ is important for elimination of liver hypnozoites in P. vivax malaria. The PQ regimen for P. vivax malaria consists of a much higher dose than the currently recommended WHO single low dose (0.25 mg base/kg) to block P. falciparum malaria transmission which is associated with a considerably lower risk of haemolytic toxicity [3]. Acute haemolytic anaemia (AHA) is the most common complication triggered by PQ in G6PD deficient-individuals in Latin America and the Caribbean region [2]. A recent report showed that G6PD A? was the allelic variant present in autopsy tissue samples from a previously reported death in the Brazilian Amazon due to PQ treatment of P. vivax malaria [33, 34]. Therefore, in Latin American countries such as Colombia and Brazil where P. vivax is the main parasite causing malaria, it is recommended that PQ administration requires prior determination of G6PD status (normal or deficient), However, this recommendation is not followed in most endemic countries of the continent, and it is uncertain what proportion of malaria cases, mainly P. vivax that are treated with therapeutic protocols including PQ, develop haemolysis. Substantial reduction of malaria by implementation of mass drug administration (MDA) using PQ has been observed in regions with high prevalence of P. vivax and varying prevalence of G6PDd [35]. The particularly high prevalence of sub-microscopic infections in Colombia [28] must be considered by malaria control programmes. The implementation of MDA through mass PQ prophylactic treatment (MPPT) [36], including sub-microscopic carriers, is possibly an important approach to reduction of transmission in Colombia. Such implementation would benefit from development of a suitable method for diagnosis of G6PDd in these locations. Testing for G6PDd is currently limited by cost, infrastructure, and logistics.
This study offered estimates of G6PDd prevalence in malaria endemic areas with considerable populations of African descent in a relatively small sample, a limitation of our study. Importantly, genotyping for the first time in these regions showed a high proportion of the G6PD A? variant in these populations. However, more information is needed about G6PDd prevalence in malaria endemic areas in Colombia, and other endemic countries of the Latin American region, including additional variants not evaluated in this study that may be associated with G6PDd in 20 individuals with no identified G6PD alleles. Further mapping of G6PDd prevalence and variants will help design appropriate malaria elimination strategies including MDA with PQ. A prospective study with a larger sample size including a close follow up to detect potential development of haemolysis and haemolytic anaemia in P. vivax patients under PQ treatment was recently initiated in collaboration with the MoH of Colombia. The MoH is currently preparing pilot activities towards malaria elimination which include the introduction of PQ treatment for P. falciparum cases, and more intensive search and treatment of asymptomatic cases, which are mostly P. vivax cases demanding 2 week PQ regimens.