Data from southeast Asia - where artemisinin-resistant malaria strains were first detected - broadly support WHO's 'working definition' for artemisinin resistance, but the currently used definitions require important refinements, according to a study by Lisa White and colleagues, from Mahidol University in Bangkok, Thailand, published in PLOS Medicine.
The drug artemisinin rapidly clears malaria parasites from the blood of infected patients - unless the parasites have developed resistance, in which case parasite clearance after artemisinin therapy (ACT) takes longer. The best measure of parasite clearance is the parasite half-life in the blood of a patient, and a common cut-off used to denote artemisinin resistance is 5 hours. The study shows that parasite half-life predicts the likelihood of an artemisinin-resistant infection for individual patients, but is influenced by how common resistance is in the particular area. The critical half-life varied between 3.5 hours (in areas where resistance is rare) and 5.5 hours (in areas where resistance is common). This means that there is no universal cut-off value in parasite half-life that can determine whether a particular infection is "sensitive" or "resistant".
Because measuring the parasite half-life requires frequent blood sampling which is difficult to do in resource-limited settings, WHO uses the following working definition for surveillance: artemisinin resistance in a population is suspected if more than 10% of patients are still carrying parasites three days after the start of ACT. Arguing that the cut-off used in the WHO working definition is based on limited data, the researchers examined how well the definition matches actual data from patients in areas with artemisinin-resistant parasites.
Applying a model specifically developed for this purpose, they found that the current WHO 'day-3' cut-off value of 10% is useful, but would be more informative if the parasite load at the start of ACT was taken into account. The authors also conclude that the WHO definition is in general a useful tool to identify areas with suspected artemisinin resistance, but "lacks accuracy in predicting the real proportion of artemisinin-resistant parasites, and should thus be followed by a more detailed assessment".