Mefloquine vs. Doxycycline for Malaria Prophylaxis in Indonesian Soldiers


Synopsis: Consideration should be given by traveler clinics to prescribing a loading dose of mefloquine for persons entering a malarious area sooner than one week after initiation of mefloquine.

Source: Ohrt C, et al. Mefloquine compared with doxycycline for the prophylaxis of malaria in Indonesian soldiers. Ann Intern Med 1997;126:963-972.

Recommendations for malaria prophylaxis in chloroquine-resistant P. falciparum (CRPF) areas have always been a challenge for the traveler’s health physician. Few blinded or placebo-controlled studies have compared the efficacy and tolerability of the two leading drugs, mefloquine and doxycycline, used by U.S. physicians for chemoprophylaxis in CRPF areas.

Ohrt and colleagues, including a prestigious research working group (the U.S. Army Medical Component Armed Forces Research Institute, the U.S. Navy Medical Research Unit 2, the Indonesian Provincial Health Services, Indonesian Ministry of Health, the Indonesian Army, Hoffman-LaRoche Pharmaceuticals, and Walter Reed Army Institute of Research), conducted a superb randomized, double-blind, placebo-controlled field trial of chemoprophylaxis with these two drugs in Northeastern Irian Jaya, Indonesia.

The trial study site was conducted in a rural lowland area of Irian Jaya with intense malaria transmission. In the placebo group, 53 (77%) of the participants developed malaria over the three-month study with mostly P. falciparum (57%) or P. vivax (44%). The participants were 204 Indonesian soldiers who were given radical curative treatment before being randomized to either 100 mg daily doxycycline and weekly mefloquine placebo; 250 mg of mefloquine per week (preceded by a loading dose of 250 mg/d for 3 days) and daily doxycycline placebo; or placebos for both days. A study employee visited each post, identified the soldier, and watched the soldier swallow the pill or pills. Participant characteristics are described in the table. Follow-up weekly smears or smears when participants became ill continued for 13.5 weeks (range, 12.7-14.6 weeks), with seven participants followed for a mean of 9.4 weeks.

No malaria occurred in the 68 soldiers in the mefloquine group; thus, the protective efficacy was 100% (CI 96-100%). In the doxycycline group, P. falciparum occurred in one of the 67 soldiers yielding a protective efficacy of 99% (CI 94-100%). Both doxycycline and mefloquine were significantly better tolerated than placebo (P < 0.001 and P = 0.005, respectively), and doxycycline was better tolerated than mefloquine (P = 0.006). No symptom occurred more significantly in either the doxycycline or mefloquine group compared to the placebo group, except dizziness with mefloquine when compared to placebo (relative risk, 1.42; P > 0.2).


Participant Characteristics

Doxycycline Mefloquine Placebo Group Group Group Characteristic (n = 67) (n = 68) (n = 69)

Mean age ± SD (years) 26.0 ± 3.9 25.9 ± 4.2 25.7 ± 4.1

Mean weight ± SD (kg) 59.5 ± 6.6 57.9 ± 4.2 59.0 ± 4.8

Mean time between arrival in Irian Jaya, Indonesia, and study entry (days) 62.0 69.0 65.0

Splenomegaly 3% 0% 1%

Mean hematocrit ± SD 0.45 ± 0.03 0.45 ± 0.03 0.44 ± 0.03

History of malaria 59% 61% 62%

Median lifetime episodes of malaria (number) 1 1 1

Median time since most recent case of malaria (years) 5.0 4.8 3.8

* Before arriving in Irian Jaya, Indonesia

Adapted from Ohrt C, et al. Ann Intern Med 1997;126:963.


This study was an excellently conducted double-blind, placebo-controlled study which showed that both mefloquine and doxycycline were highly efficacious as prophylaxis for CRPF and P. vivax malaria in a largely non-immune population. Distinguishing features of this trial included: 1) intense exposure to malaria, 2) double-blind placebo control with double-dummy technique, and 3) assiduous monitoring of compliance and symptoms.

The most tantalizing question of this study is whether the findings are generalizable to a U.S. travel population. Although the authors made a cogent argument that the study participants were "non-immunes" (based on smears, infrequency of splenomegaly, normal hematocrits, and low transmission of malaria from areas from which the troops originated), 59-62% of the three groups (mefloquine, doxycycline, and placebo) had had a history of malaria. Results of prophylaxis studies in semi-immune persons may not be generalizable and, although a full semi-immune state is not reflected in these troops, it is rare for a U.S. traveler to have had malaria. A second question is whether a soldier’s compliance to daily drug administration with strict monitoring is equivalent to a traveler’s compliance to a daily drug such as doxycycline. Compliance issues might lead the health practitioner to favor mefloquine. Lastly, although one would hope that studies conducted in the army would encourage wide reporting of adverse symptoms to drugs, large cultural differences and varying thresholds for responding adverse effects could exist between the U.S. traveler and Indonesian soldiers.

A final take-home message is that few U.S. travel physicians use a loading dose of mefloquine prior to travel. Although this study did not compare loading doses with non-loading of mefloquine, with this sort of efficacy demonstrated and known pharmacokinetics of mefloquine, perhaps consideration should be given by traveler clinics to prescribing a loading dose for persons entering a malarious area sooner than one week after initiation of mefloquine.