In the two localities that hosted the study, G1P[8] was the prevalent genotype, accounting for 38% in Abomey-Calavi and 30.0% in Cotonou or 32.0% across the two sites. According to a systematic review [16] describing the epidemiological situation in Africa, G1P[8] similarly was the predominant genotype with an overall prevalence of 22.64%. Taking into account the five regions of Africa (West, East, Central, North and South), the rate obtained varied from one region to another with a higher prevalence in North Africa (37.10%), while it was 14.35% in the West Africa [16].
At country level in West Africa, before vaccine introduction, Ghana reported G1P[8] (20%) as the predominant genotype [17], but others genotypes such as G6P[6] was found to be medically important in Burkina Faso [18], G12P[8] in Côte d’Ivoire [19] and G4P[8] and G12P[8] in Nigeria [20, 21]. During the post vaccine period, Ghana reported G1P[8] as the fourth common genotype detected with a rate of 8.0% [17]. Thus, there was a decrease in the prevalence rate and a replacement of strains after the introduction of the vaccine. Nearly similar observations are made in Zambia where G1P[8] initially reported as the predominant genotype (49%) in 2008 during the pre-vaccine period remained predominant the first year after vaccine introduction at 25% but changed in subsequent years [22]. On the other hand, according to the study of Seheri and colleagues, in Eastern and Southern African countries before and after vaccine introduction, data from six countries showed no difference in strains circulation during the pre- and post-vaccine introduction eras [23]. So, it is not conclusive that after vaccine introduction there is a switch of strains due to vaccine pressure or this is just an annual fluctuation of strains.
In systematic review by Ouermi and colleagues showed that G1P[8] was the most predominant genotype, followed by G2P[4], G9P[8] and G2P[6] [16]. Similarly, in our study, G2P[4] proved to be the second-most prevalent genotype circulating in the south of the country where the study was conducted. However, in Niger and the Democratic Republic of Congo, G2P[4] was the predominant genotype before vaccine introduction [24, 25]. However, G2P[4] was found circulating in the Central Africa Republic at a rate of 13%, in Cameroon at 5.9%, Burkina Faso at <1% and Gabon at <2% [26-29].
G12P[8] is the second emerging genotype, detected for the first time in Philippines in 1987, in children under 2 years of age [30]. Mainly limited to Asian countries, G12P[8] has been reported worldwide over the last 20 years [31].
Commonly detected at low rates in Africa, over the past decade, G12P[8] has been isolated as a predominant genotype in some countries such as Côte d'Ivoire and Nigeria in the pre-vaccination period [19, 21], and Ghana in the post-vaccination period [17]. Also, according to some studies, G12P[8] may come from a reassortment between human and pig rotavirus strains [31]. The presence of this genotype in our study without any pre-existing vaccination context, strongly suggests that G12P[8] appearance in Benin, is due to natural rotavirus genotype occurrence, not vaccine pressure.
On the other hand, G9P[8] recognized as being the fifth-most prevailing genotype [9, 11] was found in our study at a rate of 3%. This genotype seems to be more present in East and North Africa [32, 33]. Even at this relatively low level, it deserves to be monitored in Benin since it was one of the emerging genotypes of the last 20 years [34].
During the study 2% (4/186) of the samples were untypeable for G or P, which is less than the rate of untypeable samples previously reported from Sub-Saharan countries (8.6-14.6%) [23]. This low level of untypeable G/P genotypes could be explained by the technique used which incorporates more recently designed primers.
Unlike reported by the African Rotavirus Surveillance Network where mixed genotypes accounted for 12-14% [23], in our study, a single sample (1%) showed a mixed genotype. Indeed, close association of humans with domesticated animals in most countries of the developing region lead to gene reassortment events within commonly circulating animal rotavirus strains and thus give rise to a large genomic diversity and frequent occurrence of mixed infections [19]. However, studies from some Africans settings have shown no mixed genotypes [23] or a very low rate (1%) of mixed infections [20] similar to our findings. Also, as suggested by comments from the works of Boula and colleagues, the incorporation and use of new and updated one step multiplex genotyping assays in combination with sequencing in this study could strongly explain the low rate of mixed genotypes.
Atypical rotavirus G and P combinations were detected at low frequency (2%) similar to the rate described previously by Seheri and colleagues [23]. No correlation between age group and genotype was observed in this study, although it has been noted a higher variability of genotypes in children less than 18 months. This observation could probably due to the fact that more samples came from the youngest age groups.
It is important that further studies be conducted to determine the true intrinsic determinants of occurrence of rotavirus gastroenteritis in children. Studies on the genetic differences in histo-blood group antigens in our populations (secretor, non-secretor, Lewis antigens) could help to learn more about rotavirus epidemiology.