The challenge of diagnosing Plasmodium ovale malaria in travellers: report of six clustered cases in french soldiers returning from West Africa

Malaria is a threat to travellers staying in tropical endemic areas, including soldiers under field conditions [1, 2]. Each year, 40,000 French soldiers stay in or go through malaria-infected areas, using mandatory mosquito-fighting measures and chemoprophylaxis (currently doxycycline). In 2006, the French Army experienced 557 cases of malaria, 60% of which were due to Plasmodium falciparum[3]. About half of these malarial attacks were diagnosed in soldiers after their return from endemic areas [3]. Continuous monitoring identified a recent increase in imported non-falciparum malaria, mainly Plasmodium vivax from French Guiana and Plasmodium ovale from West Africa [3, 4]. This is a report of six clustered cases of imported P. ovale malaria in French soldiers who were infected in the Ivory Coast, pointing out the difficulties in diagnosing this species.

All infected cases were men (mean age: 32 years, range: 30-36). The first symptoms developed on average 50 days after their return (range: 41-72) and an average of 20 days after the end of chemoprophylaxis (range: 11-42). The patients consulted on average two days after the onset of symptoms. The most frequent symptoms were fever (6/6), mostly tertian (4/6) with temperatures higher than 38.5°C in half the cases, aches (6/6) and shivers (5/6). Other symptoms included nausea (3/6), abdominal pain (2/6), diarrhoea (2/6), or cough (2/6). No patient had a splenomegaly. Biological changes (Table 1) included thrombocytopaenia lower than 100,000/mm³ (3/6), hepatic cytolysis (3/6), hyperbilirubinaemia (1/6), and elevated CRP (6/6). The haemoglobin count was normal for all patients.

The diagnosis was established using thick and thin blood smears, after at least three searches for each case (up to seven times for one case), and on average eight days after the onset of symptoms (range: 3-12). Parasitaemia was always lower than 0.5%. Rapid diagnostic tests were all negative both for the HRP2 antigen (specificity for P. falciparum) and for the pLDH antigen (specificity for Plasmodium spp, thus for P. ovale).

Three patients received chloroquine (25 milligrams/kilo within three days). Three patients received quinine (24 mg/kg/d for seven days) because of suspected co-infection with P. falciparum (two cases) or digestive intolerance (one case treated by intravenous quinine). All attacks were cured. One patient experienced a P. ovale relapse 30 days after his first attack. He received a further course of chloroquine. All patients also received a course of primaquine (15 mg/d for 14 days) to avoid malarial relapse, none having a glucose-6-phosphate-dehydrogenase (G6PD) deficiency. No new relapse was reported after a 24-month follow-up.

For travellers, P. falciparum is the main target of malarial prevention because this species is widely endemic, very common and clinically severe [1]. Although very rarely life-threatening, relapsing malaria species, P. vivax and P. ovale, are also of concern for travellers in endemic areas. In contrast to P. vivax, which is widespread in Asia and Central and Southern America, P. ovale is only endemic in some African countries [5]. Two subspecies of P. ovale have been identified in West Africa [6]. Plasmodium ovale is responsible for rare, travel-acquired infection [7]. However, the French Army experienced a significant increase in the incidence rate of P. ovale malaria attacks among French soldiers returning from the Ivory Coast between 2002 and 2007, but not among soldiers in the field, while the incidence of Pl. falciparum regularly decreased over the same period [3, 4]. These changes could reflect epidemiological changes in the local transmission of malaria and/or a lower effectiveness of doxycycline chemoprophylaxis on P. ovale than on P. falciparum. Indeed, P. ovale was very uncommon while using the association chloroquine-proguanil in the French Army before 2002. Thus, preventing P. ovale infection with doxycycline seems challenging because of the lack of action on hypnozoites. A similar failure has also been recently reported with atovaquone plus proguanil chemoprophylaxis [8].

Diagnosing P. ovale infection in infected patients is a challenge in the field or on return to non-endemic countries [9]. Symptoms can be due to a primary infection or, later to relapses that occur in one patient in ten because of delayed parasitaemia from intrahepatic dormant forms [1]. Shortening the time-lapse before effective treatment of this relapsing malaria is required to reduce morbidity, limit sick-leave and, diminish the risk of spleen rupture [10]. This experience illustrates a few traps that physicians should be aware of when attempting a diagnosis. First, all clinical features of P. ovale attacks are not specific. Nevertheless, a tertian fever-evocative although non-specific - was present in more than half the patients, possibly due to the time lapse to the delayed diagnosis, which had allowed synchronization of the replication cycle. Moreover, digestive or respiratory symptoms, which were present in half cases, can wrongly suggest a viral infection. However, any fever, associated or not, after return from an endemic area, should be considered to be due to malaria and be followed up by immediate thick and thin blood smears. As a result of health education before and during the mission, all the patients promptly consulted a general practitioner and underwent a search for malaria in a nearby laboratory. In contrast to Plasmodium falciparum and Plasmodium vivax malaria, the haemogram appeared not to be very helpful for Plasmodium ovale because anaemia is frequently missing and thrombocytopaenia mild or absent [7], especially within the first days of the attack. Biological markers for haemolysis were also missing in five cases out of six. Most of all, the sensitivity of routine microscopic searches for P. ovale in imported cases is very low [11], as shown in this series. Although more pyrogenic than P. falciparum[12], the level of parasitaemia for P. ovale is far lower at the same time after disease onset [7]. Unfortunately, all available rapid antigenic tests currently lack sensitivity to P. ovale, while they are highly sensitive to P. falciparum[13‐15]. They generally fail to detect P. ovale infection when thick films and blood smears are negative. This could be due to the very low level of circulating antigen or to inadequate antigens used for these tests. This could explain some acute unexplained fevers in the French Army in the field in Africa [3]. When the diagnosis of imported P. ovale malaria is suspected, routine microscopic searches with thick and thin blood smears should be repeated, up to three times and in an expert laboratory, if possible [9]. Considering the high sensitivity and specificity of molecular detection of P. ovale using PCR, this tool marks real progress in confirming the diagnosis, although it is still not routinely available [5, 7, 16]. It can be used as a second-line diagnosis tool to identify infra-microscopic parasitaemia, especially for unexplained relapsing fever in travellers.

The treatment for proven attacks is based on chloroquine (25 mg/kg for three days). The treatment against dormant stages in the liver consists of a radical cure with primaquine (0.5 mg/kg/d for 14 days) in patients without G6PD deficiency [17, 18]. Failures of primaquine are unusual, mostly due to poor observance or inadequate dosage [19].