Because of violence and insecure conditions at illegal mining sites, French military health services conducted the mission. Twelve soldiers and military policemen accompanied the medical team to ensure their protection but also that of the miners’ community. A primary care clinic and laboratory were set up under tents at the Eau Claire gold mining camp from May 28, 2014, through June 6, 2014. Active malaria screening was offered to every person who sought care for any reason.

Diagnostic tests associating the malaria rapid diagnosis test (RDT) (SD Bioline Malaria Ag Pf/Pan; Standard Diagnostics, Inc., Giheung-gu, South Korea) and thin blood films were performed in the field. Patients who had positive results of an RDT, thin blood film, or both received treatment.

Data were collected by physicians concerning each person’s recent medical history, protection measures against mosquito bites, use of medications, and recent travel inside or outside of French Guiana. Patient anonymity was stringently respected; every patient was issued a unique identification number. Only verbal consent could be obtained to avoid references that might reveal the identity of undocumented persons engaged in illegal activities.

Dried blood spots were obtained on filter paper by fingerstick and packaged in individual plastic bags with a desiccant until processing. Plasmodium DNA was extracted subsequently and tested with a nested PCR targeting P. falciparum and P. vivax 18S rRNA genes, according to the method of Snounou et al. (6). The propeller domain of pfK13 gene was sequenced in P. falciparum isolates (7).

We defined Plasmodium infection as a positive RDT, thin blood film, or PCR result. Symptomatic Plasmodium infection was defined as a positive test result and fever (history of fever in preceding 24 hours and/or documented temperature >38°C during medical examination) and/or >2 of the following: nausea, vomiting, diarrhea, abdominal pain, anorexia, headache, or jaundice. Other Plasmodium infections were classified as asymptomatic.

Figure 2

Figure 2. Travels of gold miners (N = 205) living in Eau Claire, French Guiana, 2013–2014. Inset shows location of French Guiana in South America.

Overall, 205 persons freely sought medical care and accepted malaria screening. The sex ratio was 2.0 (137 [66.8%] men; 68 [33.2%] women). Median age was 39 years (interquartile range [IQR] 32–46 years; range 20–63 years). The workers had been gold panning for a median time of 7 years (IQR 3–14 years; range 1 month–44 years) and on illegal gold mining sites in French Guiana for a median time of 4 years (IQR 1–8 years; range 1 month–25 years). Most (97.6%, 200) patients came from Brazil, 4 (2.0%) came from Suriname, and 1 (0.5%) came from Guyana. Before their arrival at Eau Claire, patients had lived in Brazil (41.0%, 84), Suriname (24.9%, 51), or at another illegal gold mining site (19/69 [33.7%] different sites throughout inland French Guiana and 1 unknown site) (Figure 1). During the previous year, 66.0% (135/205) persons had traveled to >1 area outside Eau Claire: 54.1% (111/205) to Suriname (among those, 60.0% had traveled there >2 times); 22.4% (46/205) to Brazil; and 18.5% (38/205) to the main cities in the malaria-free area along the French Guiana coast (Figure 2).

Of the 205 patients, 156 (76.1%) reported >1 fever episode and 131 (63.9%) reported several (2–>4) episodes; 121 (59.0%) reported >1 malaria episode and 93 (45.4%) several episodes (Table 1). Self-medication with antimalarial drugs was reported by 120 (58.5%) patients, of whom 118 (98.3%) had reported malaria episodes in the past year. Artemisinin-based combination therapies (ACTs) were mainly used: dihydroartemisinin/piperaquine/trimethoprim (Artecom; Chongqing Tonghe Pharmaceutical Co. Mingshan Town, China) by 79 (63.7%) and artemether-lumefantrine (Coartem; Novartis Pharmaceuticals Corp. Basel, Switzerland) by 32 (26.7%). Chloroquine was also used alone or with ACTs by 13 (10.8%) patients. The medication schedules used were not clearly identified, but 53 (44.2%) patients took drugs for 1 or 2 days only. Nets were used by 37 (18.0%) and mosquito repellents by 41 (20.5%) of the 205 patients.

The overall prevalence of malaria infection was 48.3% (99/205). P. falciparum and P. vivax single infections accounted for 44.4% (44/99) and 29.3% (29/99) cases, respectively, and mixed infection with P. falciparum and P. vivax for 26.3% (26/99) (Table 2). RDTs, thin blood films, and PCR were positive for 40.4% (40/99), 32.3% (32/99), and 97.0% (96/99) of patients classified as positive for malaria, respectively. Asymptomatic infections accounted for 48.5% (48/99) of cases. Low parasitemia levels were systematically observed. Only 1 person had a parsitemia level >1%, and no differences in level were found between symptomatic and asymptomatic infections. The propeller region of the pfK13 gene was successfully sequenced in 26 P. falciparum isolates without any mutation detected.

During the 10 days of field work, almost all inhabitants of the gold mining site sought medical care. The high prevalence of malaria and asymptomatic Plasmodium infections observed confirms that malaria is hyperendemic there (8). Because of the mobility of gold miners within French Guiana, malaria also could be highly prevalent among persons at other illegal gold mining sites where competent vectors exist (3). This hypothesis is strengthened by the recurrent malaria outbreaks experienced by French forces involved in operations to control illegal gold mining, particularly in the center of the region (3).

Systematic self-medication by patients using ACTs without following a full course of treatment is a serious risk for emergence of resistance to artemisinin and associated drugs (9). The high price of ACTs in the field (2 g gold) was the primary reason patients gave for not completing treatments. The mobility of gold miners raises 2 issues: the reintroduction of the disease in malaria-free areas (10) and the spread of antimalarial drug resistance if the disease emerges (11). In 2013, parasite persistence on day 3 after treatment with Coartem was described in Suriname, and most participants in that study had worked in gold mines in French Guiana (12,13). In Guyana, pfK13 C580Y mutants obtained from samples in 2010 were recently reported (14). These results may be viewed as ominous. Collaboration between countries of the Guiana Shield to control malaria among mobile populations is urgently needed (15).