Effectiveness of Short Prophylactic Course of Atovaquone-Proguanil in Travelers to Sub-Saharan Africa


  • Eyal Leshem MD,

    1. The Center for Geographic Medicine, The Chaim Sheba Medical Center, Tel Hashomer, Israel
    Search for more papers by this author
  • Eyal Meltzer MD,

    1. The Center for Geographic Medicine, The Chaim Sheba Medical Center, Tel Hashomer, Israel
    2. Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
    Search for more papers by this author
  • Shmuel Stienlauf MD,

    1. The Center for Geographic Medicine, The Chaim Sheba Medical Center, Tel Hashomer, Israel
    2. Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
    Search for more papers by this author
  • Eran Kopel MD, MPH,

    1. Ministry of Health, Jerusalem, Israel
    Search for more papers by this author
  • Eli Schwartz MD

    Corresponding author
    1. The Center for Geographic Medicine, The Chaim Sheba Medical Center, Tel Hashomer, Israel
    2. Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
    • Corresponding Author: Professor Eli Schwartz, MD, Center for Geographic Medicine, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel. E-mail: elischwa@post.tau.ac.il

    Search for more papers by this author



Current guidelines recommend continuation of atovaquone-proguanil (AP) malaria prophylaxis for 7 days after leaving Plasmodium falciparum endemic areas. Evidence from clinical studies suggests that discontinuation of AP 1 day after exposure ends may be safe and effective. Our objective was to assess the effectiveness of short-course AP prophylaxis among travelers to sub-Saharan Africa.


To detect prophylactic failures associated with short-course AP prophylaxis discontinued 1 day after return, we conducted active surveillance during the years 2010 and 2011, by a retrospective telephone survey 1 to 6 months after travelers' return. Passive surveillance data were obtained from the Israel Ministry of Health (MOH) malaria registry.


Among 485 travelers to sub-Saharan Africa (cumulative exposure of 4,979 days), 421 (87%) discontinued AP 1 day after leaving the endemic region (cumulative exposure of 4,337 days). None of the 485 travelers reported malaria infection. The MOH malaria registry survey included 363 P. falciparum-infected patients during the years 2003 to 2011. The majority (n = 305; 84%) did not use any malaria prophylaxis. None of the patients had used AP (neither regular nor short course AP) for malaria prophylaxis.


We did not detect prophylaxis failures among a group of travelers who discontinued AP prophylaxis 1 day after leaving malaria-endemic areas. Passive surveillance in Israel did not detect any P. falciparum cases among AP users. We recommend further validation of our findings by clinical trials, prospective studies, and active surveillance in larger cohorts to assess the effectiveness of short-course AP prophylaxis in travelers.

In travelers visiting tropical regions, Plasmodium falciparum malaria is a major, potentially fatal cause of fever,[1] and malaria prophylaxis is recommended for travelers planning to visit endemic areas.[2] Atovaquone-proguanil (AP; 250 mg of atovaquone and 100 mg of proguanil hydrochloride; Malarone, GSK, Middlesex, UK) is a synergistic combination drug approved by the United States Food and Drug Administration for treatment and prophylaxis of P. falciparum malaria. The currently approved regimen of AP for malaria prevention is one tablet daily beginning 1 day before entry into a malaria-endemic area, continued daily during exposure, and ending 7 days after leaving the area.

Despite evidence from previous volunteer studies suggesting that use of atovaquone the day before exposure to P. falciparum-infected Anopheles mosquitoes may be sufficient to prevent the disease,[3] no studies have attempted to test the effectiveness of this shorter AP regimen. Moreover, older studies have shown causal prophylactic activity of proguanil given to volunteers 2 to 5 days after P. falciparum-infected Anopheles mosquitoes' exposure challenge.[4] Taken together with the long elimination half-life properties of atovaquone,[5] these studies suggest that ending the AP daily prophylaxis course 1 day after exposure ends is safe and effective in prevention of P. falciparum malaria. Therefore, since 2003 (the year in which AP was licensed in Israel), we have recommended that travelers may discontinue AP prophylaxis 1 day after leaving malaria-endemic areas (short-course AP prophylaxis). Our main purpose in suggesting short-course AP prophylaxis was to improve currently low rates of travelers' compliance with malaria prophylaxis.[6-8]

The travel clinic of the Center for Geographic Medicine at Sheba Medical Center in Israel consults approximately 7,000 travelers every year. Since 2003, we consulted over 6,000 travelers to sub-Saharan Africa (cumulative travel time over 60,000 days; equals over 164 travel years). During this time period we did not diagnose any cases of P. falciparum malaria in returned travelers following the use of short-course AP prophylaxis. The current study aimed to assess the effectiveness of short-course AP prophylaxis by active surveillance (a retrospective survey in travelers to sub-Saharan Africa) and by passive surveillance [case finding in the Israel Ministry of Health (MOH) malaria registry].


Study Population and Case Definition

To assess the effectiveness of short-course AP prophylaxis in a population at high risk of P. falciparum exposure we included only travelers who visited sub-Saharan Africa.[1, 9-13] Travelers planning to visit friends or relatives (VFR travelers) comprise a negligible minority among Israeli travelers to sub-Saharan Africa and were excluded from this study. We defined short-course AP prophylaxis as daily use of AP beginning 1 day before entering malaria-endemic area, continued daily, and ending 1 day after the traveler had left the endemic area.

Active Surveillance

Travelers who visited the pre-travel clinic at the Sheba Medical Center during January 2010 to December 2011, planned travel to sub-Saharan Africa, and were recommended the use of short-course AP prophylaxis were included. All travelers were informed about the manufacturer's recommended dosing schedule (continue AP 7 days after return), and about currently available evidence supporting the efficacy of a shorter course. All travelers had the option to receive a prescription for a long course (as recommended by the manufacturer) and were able to decide regarding the duration of the regimen when consulting their primary healthcare provider or pharmacist (from whom AP was purchased). The travelers were contacted 1 to 6 months after return from their trip, over telephone. The 6-month upper limit for inclusion was set to avoid recall bias. At least three attempts were made to contact each traveler. After giving their consent to participate in the survey, they answered a questionnaire. The questionnaire included demographic information, travel history, and information regarding prophylaxis regimen used, duration, and adherence. In addition, travelers were asked if malaria was diagnosed during travel or after return from travel. Travelers describing a febrile illness during or after travel were asked to describe the course of illness, presumed or final diagnosis, and treatment. The study was approved by the Sheba Medical Center Institutional Review Board.

Passive Surveillance

In Israel, malaria is a notifiable disease, and all cases of malaria are reported to the MOH. Each case is further investigated by an epidemic nurse from the health department. Demographics, malaria exposure, prophylaxis, and clinical details are documented. All P. falciparum malaria cases reported to the Israel MOH during 2003 to 2011 were reviewed. Information about prophylaxis use among confirmed malaria cases was retrieved. Information included the method used for malaria diagnosis, and if the patient used malaria prophylaxis, which prophylactic regimen was used (drug, dosage, and duration).


Active Surveillance

The telephone survey study included 500 travelers. Of those travelers, 485 (97%) were contacted and consented to participation. The study population traveled for a total of 4,979 days in malaria-endemic countries. Of the 485 travelers we contacted, 421 (87%) used short-course AP prophylaxis (cumulative period of 4,337 travel days); nine (2%) travelers (cumulative period of 95 travel days) continued taking AP for 2 to 7 days after leaving the endemic country; and 55 (11%) travelers (cumulative period of 547 travel days) did not take AP prophylaxis at all or discontinued AP prior to leaving the endemic country (Table 1).

Table 1. Travel destinations and duration of travelers who used short-course AP prophylaxis
 Number of travelersCumulative travel duration (days)
Tanzania and Zanzibar190(45)1,883(43)
Equatorial Guinea11(3)113(2)
South Sudan3(1)30(1)
Rwanda1 14 
Ghana1 9 
Gabon1 5 
Sierra Leone1 3 
Total421 4,337 

Among the 421 travelers who used short-course AP prophylaxis, 219 (52%) were male, mean age was 38.1 ± 20.3 years (median age 43 years, range 4–76) and mean travel duration was 10.1 ± 7.8 days (median 10 days, range 2–77). No cases of malaria were reported by any of the 485 study participants including the 421 participants who used short-course AP prophylaxis.

Passive Surveillance

During the years 2003 to 2011, 363 P. falciparum-infected patients were reported to the MOH and included in the study. The majority (305 patients; 84%) did not use any malaria prophylactic. Of 58 patients with malaria who used prophylaxis, none had used AP (neither regular nor short-course AP prophylaxis).


We did not detect failures of short-course AP prophylaxis ending 1 day after leaving a malaria-endemic area among a cohort of Israeli travelers to sub-Saharan Africa. Using both active and passive surveillance and questioning travelers less than 6 months after their return increased the reliability of reporting and enhanced our ability to detect prophylaxis failures had those occurred.

The current recommendation for ending AP malaria prophylaxis 7 days after leaving the endemic area is partially based on information from a study, in which 12 volunteers were given AP 1 day before P. falciparum-infected Anopheles mosquitoes' exposure challenge and continued AP for 7 days after exposure without acquiring malaria.[12] A more recent study by Deye and colleagues has shown that a single dose of AP given 1 day before P. falciparum-infected Anopheles mosquitoes' exposure challenge prevented malaria in six volunteers.[14] Previous studies have shown the efficacy of each of the AP components used separately in the prevention of P. falciparum malaria when used shortly before or after exposure: Shapiro and colleagues have reported that a single dose of 250 mg atovaquone given to volunteers (n = 6) 1 day prior to P. falciparum-infected Anopheles mosquitoes' exposure challenge conferred 100% protection.[3] Classic studies conducted by Fairley have shown that a single dose of proguanil (10–100 mg) given 2 to 5 days after P. falciparum-infected Anopheles mosquitoes' exposure challenge was 100% protective in nonimmune volunteers.[4] These studies support the hypothesis that a single dose AP given within a day of sporozoite inoculation prevents patency (ie, it kills the very early developmental stage in the liver), thus acting as causal prophylaxis. Moreover, AP exerts a long half-life elimination time of nearly 6 days.[5] Therefore, discontinuing AP 1 day after exposure ends, potentially allows safe P. falciparum prevention while avoiding unnecessary prolonged and expensive regimen. Nevertheless, a prospective randomized study comparing the efficacy of short and long AP regimen in travelers may not be conducted because of the extensive resources necessary for such an evaluation.

A recent survey of compliance with malaria prophylaxis was conducted at our post-travel clinic among ill-returned Israeli travelers returning from sub-Saharan Africa who received pre-travel consultation at other travel clinics in Israel. In this survey, only 105 of 247 (42%) travelers used malaria prophylaxis (E. Schwartz, unpublished data). This rate is similar to previously reported low rates of compliance with malaria prophylaxis advice among Swiss (32%) and French (18%) travelers to sub-Saharan Africa.[15, 16] In the current study, travelers who were advised by our clinic to use the short-course of prophylaxis reported substantially higher compliance rates when compared with the ill-returned travelers who visited other clinics for pre-travel consultation [430/485 (89%) vs 105/247 (42%); rate ratio = 2.1; 95% confidence interval 1.8–2.4; p < 0.001]. Although ill-returned travelers may differ from healthy travelers in health protection measures, this difference in malaria prophylaxis use may at least partially be attributed to better compliance with short-course AP prophylaxis. Recommending longer duration of malaria prophylaxis than necessary may be a factor contributing to low compliance rates with malaria prophylaxis, mostly because of concerns regarding cost, adverse events, and need for daily medication.

Several limitations should be considered when interpreting our results: recall bias may reduce reliability of reporting drug use. We studied only travelers who recently returned thereby reducing the risk of such bias and our questionnaire was used by trained professionals with an emphasis on open-ended questioning of duration of malaria regimen to reduce biased answering. The likelihood of not recalling an episode of P. falciparum malaria is very low. We are not able to confirm if any of the travelers was exposed to infected mosquito bites, because testing for anti-sporozoite antibodies was beyond the scope of this retrospective study. Additionally it is possible, although unlikely, that a case of malaria was missed by our active surveillance and was also not reported to the Israel MOH (passive surveillance). The cumulative exposure of our cohort may have been insufficient to detect a malaria case. Most travelers (88%) in our cohort traveled to East Africa where malaria risk is currently considered to be lower compared with Central or West Africa.[13] Evidence from two previous studies of malaria risk in travelers found that high risk was associated with travel to East Africa[9, 10]: an evaluation of malaria risk among Swedish travelers suggested an estimated rate of over 240 cases per 100,000 travelers to sub-Saharan Africa.[10] In this evaluation, the adjusted odds ratio of malaria acquisition was 341 in East Africa, 317 in Central Africa, and 277 in West Africa, as compared with travelers to Central America and the Caribbean.[10] In a previous report, the estimated annual rates of reported P. falciparum malaria per 100,000 travelers to West and East sub-Saharan Africa are 375 and 172, respectively.[9] In Israel, 18% of all imported malaria cases during 1999 to 2001 were acquired in Kenya, making it the leading source of malaria during that period.[11] Finally, AP metabolism may be variable among different subjects,[17] and diarrhea may reduce AP absorption.[18] Variability in absorption and metabolism may theoretically reduce the safety margin provided by short-course AP prophylaxis.

In summary, our study found no cases of prophylaxis failure in a cohort of travelers to sub-Saharan Africa who used short-course AP prophylaxis. On the basis of low compliance of travelers with current recommendations for AP use for malaria prophylaxis and based on current knowledge from clinical and pharmacokinetic studies, we suggest that travelers may discontinue AP prophylaxis 1 day after leaving endemic regions. Several evaluations of short-course AP prophylaxis should be considered to validate our findings. Clinical trials and larger, prospective cohort studies as well as active surveillance among travelers are necessary to further assess the effectiveness of short-course AP prophylaxis. The impact of recommending short-course AP prophylaxis on travelers' compliance with malaria chemoprophylaxis recommendations should be assessed.


We would like to thank Drs Emilia Anis and Daniel Goldman at the Israeli MOH, Jerusalem, Israel, for providing passive surveillance data for this report. No financial support was received for this study.

Declaration of Interests

The authors state that they have no conflicts of interest to declare.