Mathematical models were constructed to understand the transmission dynamics of P. falciparum malaria in an area where access to health-care has significantly reduced malaria transmission. Active malariometric surveys suggest that a very high fraction of clinical episodes in the area, perhaps higher than 97% are promptly treated . Prompt and effective treatment may be a very cost-effective strategy for malaria control in low or unstable malaria transmission settings, because most individuals are likely to develop acute febrile illness after P. falciparum infection [39–41]. Significant reductions in mortality and morbidity of malaria after deploying a strategy of early detection and prompt treatment has been documented . However, model simulations find that, despite significant reductions in incidence, P. falciparum transmission is likely to continue. This suggests that the elimination of malaria may be difficult even in an area where malaria health systems are highly effective.
Given the fact that multiple surveys of large swaths of the population are unable to find individuals infected asymptomatically, gametocyte carriage in the absence of asexual parasites seems to be vital for maintaining transmission in this heavily controlled area [7, 24]. Findings of our model suggest that there is a strong seasonal fluctuation in the population of residual gametocyte carriers, which is consistent with previous observations in other low transmission areas [7, 9]. An interesting finding is that while the prevalence of gametocytes is relatively small during the dry season, it does not drop to zero. Presumably then these cases are responsible for the source of mosquito infection at the start of wet season [7, 9].
Gametocyte reduction is of great interest for malaria control, particularly in low-endemic malaria areas. One of the strategies for the malaria elimination programme, recommended by the WHO, is to identify and treat all malaria patients as well as to reduce onward transmission caused by gametocytaemia . The policy change from mefloquine treatment to artesunate-mefloquine has been shown to reduce transmission. Artemisinin combination therapies can reduce the asexual parasite burden 100 times faster than mefloquine, which can subsequently inhibit development of more mature gametocytes [31–33, 41]. The gametocyte model indicates that there is still an added value of the follow-up primaquine treatment even when the initial gametocyte density is small due to the switch from mefloquine to ACTs. In addition, in areas where artemisinin combination therapy has been used, gametocyte carriage is still common in the 7–21 days following treatment [13, 30, 41, 43–45]. In many countries, a single oral dose of primaquine is included in the standard anti-malarial drug regimen with the aim of further reducing gametocyte carriage, even when artemisinin-based therapy is used [6, 12].
However, while in these areas primaquine can be extremely effective at clearing gametocytes that persist after treatment with schizonticidal agents [13, 14, 43], the timing and duration of gametocyte carriage and subsequent infectiousness have not been considered carefully when primaquine is deployed as a transmission-blocking agent. Findings of the gametocyte model indicate that the effectiveness of primaquine in reducing the duration of infectiousness depends critically on timing. Primaquine is most beneficial when the administration is delayed, about eight days following initial treatment, to coincide with the release of a large cohort of mature gametocytes into the blood, which emerged from a large number of merozoites during an acute attack. The effect of primaquine is significantly reduced when the drug is given too early or too late. Although an immediate primaquine treatment can affect a small cohort of mature gametocytes that emerge from the first crop of merozoites that appear in circulation at the time of an acute attack, primaquine will be cleared from the system before the largest cohort of gametocytes mature. If primaquine is given too late, mature gametocytes will be able to circulate and infect mosquitoes until the drug is administered.
The benefits of optimally timed primaquine are greatest in those areas where the early treatment programme to cure asexual blood stage infections is very successful; a high fraction of clinical malaria episodes are expected to receive the standard treatment within one to three days after acute attack. In such areas, optimally timing primaquine administration shows a potential impact on overall malaria transmission at the population-level. The current results show that follow-up primaquine treatment can reduce the duration of infectiousness over the existing strategy of using artesunate-mefloquine alone, with a combined total net reduction in transmission of 98%, a 95-fold reduction in R
. An important observation is that the added value of optimally timed primaquine can have relatively large effects on reducing transmission only if a high fraction of patient infections are treated and cured with first-line anti-malarial drugs (i.e. when P is high), suggesting that the first emphasis should be on treating those with clinical malaria. Because primaquine effectively reduces transmission only in those patients who have cleared their asexual parasites, and because the average duration of an asymptomatic infection is approximately six months, the benefit of reducing the duration of gametocyte carriage is of little importance unless at least 90% of clinical malaria episodes are effectively treated. Treatment to clear asexual parasites and prevent asymptomatic infections can only reduce the duration of infectiousness insofar as the gametocytes are also cleared. In such situations, primaquine can be very effective at further reducing the duration of infectiousness, and the added value of good follow-up with primaquine treatment has nearly the same proportional effects on potential transmission as does the primary treatment.
In addition, findings from the model indicate that when all symptomatic P. falciparum patients receive the follow-up primaquine treatment at day eight, the P. falciparum malaria incidence can be reduced nearly to zero. However, the model suggests that a small number of undetected imported cases can pose a big threat for malaria elimination. To reach the elimination goal, vigilance to detect and cure imported asymptomatic cases may also be required. These findings support the WHO recommendations for a malaria elimination programme .
Mathematical modeling has been widely used to model transmission dynamics of malaria, and control interventions . However, findings from the models require a careful interpretation. This study intends to construct simple models that provide a valuable insight into the feasibility of malaria elimination in a low malaria transmission area. Results from the models should be considered as approximations that are likely to differ because of the natural variability under field conditions.
In addition, although the model assumed that the level of individual infectiousness follows the log-sigmoid relationship with the gametocyte density among non-immune adults, the duration of infectiousness in the analysis may be underestimated. Infectivity to mosquitoes is observed even when gametocyte densities fall below detection level by microscopy or by a molecular method [24, 47]. However, the trend of duration of infectiousness over different timings of primaquine administration does not change when different infectivity levels are applied. The model also does not take into account the additional gametocyte carriage from recrudescent infections. Though gametocytaemia is estimated to be greater in recrudescent infections than in primary infections, in an area where artesunate-mefloquine combination therapy is used, few recrudescent infections are expected [33, 48]. Lastly, the model assumes that most infected individuals develop clinical symptoms and are treated. In low and unstable transmission areas this assumption is generally correct; however pockets of higher transmission may exist, and the importance of asymptomatic asexual blood-stage infections in these areas in continuing transmission over the dry season can be significant. While surveys suggest that there are basically no asymptomatic carriers in the region in question, studies in other areas have found that sub-patent infections can persist for many months [9, 10, 49]. Thus, while the models suggest that optimally-timed primaquine administration can significantly impact the incidence of malaria in a low-transmission area well served by health centers, asymptomatic individuals in the area and not just imported carriers may also play a significant role in sustaining transmission and should be considered in any elimination plan.