BACKGROUND: In Zambia, malaria is one of the leading causes of morbidity and mortality, especially among under five children and pregnant women. For the latter, the World Health Organization recommends the use of artemisinin-based combination therapy (ACT) in the second and third trimester of pregnancy. In a context of limited information on ACT, the safety and efficacy of three combinations, namely artemether-lumefantrine (AL), mefloquine-artesunate (MQAS) and dihydroartemisinin-piperaquine (DHAPQ) were assessed in pregnant women with malaria. METHODS: The trial was carried out between July 2010 and August 2013 in Nchelenge district, Luapula Province, an area of high transmission, as part of a multi-centre trial. Women in the second or third trimester of pregnancy and with malaria were recruited and randomized to one of the three study arms. Women were actively followed up for 63 days, and then at delivery and 1 year post-delivery. RESULTS: Nine hundred pregnant women were included, 300 per arm. PCR-adjusted treatment failure was 4.7% (12/258) (95% CI 2.7-8.0) for AL, 1.3% (3/235) (95% CI 0.4-3.7) for MQAS and 0.8% (2/236) (95% CI 0.2-3.0) for DHAPQ, with significant risk difference between AL and DHAPQ (p = 0.01) and between AL and MQAS (p = 0.03) treatments. Re-infections during follow up were more frequent in the AL (HR: 4.71; 95% CI 3.10-7.2; p < 0.01) and MQAS (HR: 1.59; 95% CI 1.02-2.46; p = 0.04) arms compared to the DHAPQ arm. PCR-adjusted treatment failure was significantly associated with women under 20 years [Hazard Ratio (HR) 5.35 (95% CI 1.07-26.73; p = 0.04)] and higher malaria parasite density [3.23 (95% CI 1.03-10.10; p = 0.04)], and still women under 20 years [1.78, (95% CI 1.26-2.52; p < 0.01)] had a significantly higher risk of re-infection. The three treatments were generally well tolerated. Dizziness, nausea, vomiting, headache and asthenia as adverse events (AEs) were more common in MQAS than in AL or DHAPQ (p < 0.001). Birth outcomes were not significantly different between treatment arms. CONCLUSION: As new infections can be prevented by a long acting partner drug to the artemisinins, DHAPQ should be preferred in places as Nchelenge district where transmission is intense while in areas of low transmission intensity AL or MQAS may be used.