Possible Autochthonous Malaria from Marseille to Minneapolis

We report 2 cases of Plasmodium falciparum malaria in southern France in a French woman and an American man of Togolese origin who reported no recent travel to malaria-endemic countries. Both infections occurred after a stay near Marseille, which raises the possibility of autochthonous transmission. Entomologic and genotypic investigations are described.


The Cases
On June 30, 2006, a 41-year-old woman was admitted to the North University Hospital in Marseille, France, with a 12-day history of fever, chills, and diarrhea. She had never traveled outside Europe and had no history of blood transfusion or injection drug use. She was born in Marseille and had lived there her entire life. Her home residence was >20 km from the nearest airport and 10 km from the seaport. The case-patient did not work outside the home and did not report any recreational activities near either location. Her neighborhood included families of Comorian descent. Laboratory testing demonstrated low leukocyte and erythrocyte counts and thrombocytopenia. Peripheral blood smear also demonstrated intra-erythrocytic forms consistent with P. falciparum infection with a parasitemia level of 0.1%. P. falciparum antigen was detected (Test NowICT; Fumouze, Levallois-Perret, France) and subsequently confi rmed by PCR. Treatment comprised a 3-day regimen of quinine-clindamycin (7) and the case-patient recovered.
A 34-year-old man with tactile fever, bilateral frontal headache, and diarrhea was examined at a clinic in Minneapolis, Minnesota, on May 1, 2006. This man had lived in Togo, his birth country, all of his life until he emigrated to Minnesota in April 2000. Before becoming ill, he had traveled to Paris and from Paris by train to Marseille on April 7, 2006. Also during this timeframe, he reported a day trip to Camargue (Sorgues, France) where he sustained multiple mosquito bites. On April 24, 2006, he returned to Paris by train and then traveled by air a few hours later to Minneapolis. The fi rst symptoms of malaria developed on April 27. His medical history included 4 episodes of malaria in childhood (none in the past decade) and no blood transfusions. On his last visit to Togo in December 2004, the case-patient had reported no chemoprophylaxis for malaria and had experienced no illness or intervening fevers since that time. Physical examination showed a moderately ill person with hematologic test results indicating thrombocytopenia and intra-erythrocytic ring forms consistent with P. falciparum malaria with a 3% parasitemia level. PCR testing confi rmed P. falciparum. He was given 4 tablets of atovoquone-proguanil once a day for 3 days and recovered completely.
We conducted entomologic surveillance on July 26, 2006, in and around the fi rst case-patient's residence in Marseille, 26 days after she became ill. No adult mosquitoes were found in the patient's home, on external staircases, in cellars located around the patient's home, or in surrounding buildings. Two pools of standing water were identifi ed: a single pool (1 m 2 , 2 cm deep) within a cellar in her building and another pool (10 m 2 , 5 cm deep) located 20 m outside the building. No other containers or places that could collect water were identifi ed. Several Culex theileri larvae were identifi ed in the outdoor pool, but there was no evidence of Anopheles larvae.
DNA was extracted from blood samples by using the ENZA blood DNA kit according to the manufacturer's recommendations (Biofi dal, Vaulx en Velin, France) and eluted in 100 μL of elution buffer per 250 μL of whole blood. Microsatellites loci (C4M79, Pf2689, TRAP, Pf2802, 7A11, and C4M69) were genotyped by fl uorescent end-labeled PCR. Primers sequence, PCR conditions, and methods for genotyping have been described elsewhere (8,9). Drug-resistant mutations were genotyped. Genotyping of the 2 isolates demonstrated different alleles for the 6 microsatellite loci. We also observed different alleles from codon 59 (dihydrofolate reductase), 437 (dihydropteroate  (Tables 1, 2).

Conclusions
The World Health Organization defi nes introduced autochthonous malaria as that acquired by mosquito transmission from an imported case in an area where malaria is not a regular occurrence (10) In France, it is rare to fi nd P. falciparum in a blood smear from patients who have not traveled to a malaria-endemic area in the previous 12 months. History of transfusion, organ transplantation, intravenous drug use, or mother-to-fetus transmission must also be excluded (10).
In Europe, some recent autochthonous malaria cases have been related to close proximity to airports and shipping ports receiving fl ights and water craft from malariaendemic areas (5,6,11). Inadvertent carriage of infective Anopheles mosquitoes by airplane, ship, baggage, or bilge water may be responsible for these occurrences. Also, large populations of migrants from areas highly endemic for malaria (12) may act as human reservoirs for potential gametocyte carriers. Marseille has a large population of persons of Comorian origin, and most of the patients with imported malaria cases diagnosed in Marseille contracted the disease during a trip to the Comoros Islands (13,14). Our fi rst casepatient's malaria may have been linked to her neighborhood, which included Comorian families who had recently traveled to Comoros. Despite the absence of Anopheles larvae near the fi rst patient's neighborhood, local transmission cannot be excluded during the late spring and summer. Summer temperatures in Marseille induce a short P. falciparum sporogonic cycle (≈11 days at 28°C), which is compatible with mosquitoes' longevity at that time.
The Minnesota case-patient likely represents a second case of autochthonous malaria. First, this patient had no illness to suggest an untreated, active infection since his last trip to Togo. Second, even if he had been persistently infected with P. falciparum and was semi-immune, this case would still represent an extremely long incubation period (16 months). Third, onset of symptoms was acute, hematologic testing showed infection with a 3% parasitemia level, and the case-patient was moderately ill, which suggests a more recent exposure. Moreover, the genotype was monoallelic for each locus, indicating a clonal infection, which would be expected in areas of relatively low transmission frequency. In contrast, in areas of frequent endemic transmission, such as Togo, multiple allelic polymorphisms would be expected (9). Finally, genotyping demonstrated different clones of P. falciparum; this fi nding suggests potential multiple introductions, temporally related, of the parasite into the environment The last malaria focus in continental France occurred near the end of World War II in the Camargue region, which was visited by our second case-patient in April 2006. However, large Anopheles populations, including potential vectors such as An. hyrcanus, An. melanoon, and An. atroparus, are still present from March to November (N. Ponçon and D. Fontenille, unpub.data) in southeastern France, generating an "anophelism without malaria" situation (15).