Chloroquine remains the standard schizontocide for Plasmodium vivax, but early recurrences and the practical difficulty of getting patients through 14 days of primaquine keep the question open. Curavivax was an open-label single-centre trial in Manaus, in the Brazilian Amazon, randomising 419 patients over 6 months of age with microscopically confirmed vivax malaria and G6PD activity above 30% into four arms: dihydroartemisinin-piperaquine or chloroquine, each with primaquine started on day 0 or delayed to day 42.
When primaquine was given from day 0, the two schizontocides were equivalent — day 42 recurrence 2.0% with chloroquine and 1.0% with dihydroartemisinin-piperaquine. When primaquine was delayed, they were not remotely equivalent: 27.3% recurrence with chloroquine alone (95% CI 18.3 to 37.8) against 2.4% with dihydroartemisinin-piperaquine alone (0.3 to 8.5), a hazard ratio of 0.08 (95% CI 0.00 to 0.22, p<0.0001). All regimens were well tolerated.
The mechanism explains the pattern. Piperaquine has a long terminal half-life and goes on suppressing blood-stage parasites for weeks after dosing, so it covers the gap that chloroquine leaves once its own levels fall. Primaquine, when actually taken, closes the same gap by clearing the liver stage. Give either one and recurrences are rare; give neither and a quarter of patients relapse or recrudesce within six weeks.
The practical reading for Indian practice, where vivax is the dominant species across much of the country and primaquine adherence over 14 days is a well-documented problem, is about which failure you are protecting against. If you are confident primaquine will be taken, chloroquine remains adequate. If you are not — an itinerant worker, no G6PD result yet, a patient you will not see again — a longer-acting schizontocide buys real protection during the period when radical cure is not happening. Two caveats: this was a single centre in the Amazon, where chloroquine resistance patterns are not India's, and G6PD activity below 30% was excluded, so it says nothing about how to manage deficient patients.
- Where primaquine will start immediately and adherence is assured, chloroquine remains an adequate schizontocide.
- Where radical cure will be delayed or adherence is doubtful, a long-acting schizontocide gives substantial protection against early recurrence.
- Check G6PD before primaquine; this trial excluded activity below 30% and offers no guidance there.
- A recurrence within 42 days is not automatically a new infection — consider relapse and treatment failure.
- Local chloroquine susceptibility differs from the Brazilian Amazon; apply the principle, not the protocol.
The statistics, in plain English
The striking comparison is between the two delayed-primaquine arms, and the gap there is large enough that it is not in doubt: 27.3% against 2.4%, with a hazard ratio of 0.08. The comparison between the two day-0 primaquine arms is the opposite — a hazard ratio of 0.48 with a confidence interval running from 0 to over 40,000. That absurd width is what happens when almost no events occur in either arm, and it means the comparison carries no information at all. Read that arm as 'both worked, we cannot rank them', not as a trend favouring one.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for infectious diseases, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free