Artemisinin-based combination therapy for the treatment of uncomplicated Plasmodium vivax malaria and the prevention of relapses

  • Protocol
  • Intervention



This is the protocol for a review and there is no abstract. The objectives are as follows:

To compare ACTs with alternative antimalarial regimens for treating acute uncomplicated P.vivax malaria and the prevention of relapses.


Malaria is a disease of global public health importance whose social and economic burden is a major obstacle to human development in many of the worlds’ poorest countries (WHO 2007; WHO 2008). Transmission occurs person to person via the bite of mosquitoes infected with the protozoan parasite Plasmodium, of which five species are capable of causing disease in humans: P. falciparum, P. vivax, P. malariae, P. ovale, and (rarely) P. knowlesi (WHO 2006).

Although P. falciparum remains the dominant species worldwide, outside of Africa P. vivax is often equally prevalent and co-infections in a single clinical episode are very common (Mayxay 2004; WHO 2006; Price 2007). The relative importance of P. vivax has increased over recent years as control efforts against P. falciparum have shown signs of success, and as evidence accumulates that P. vivax malaria is not an entirely benign illness (Anstey 2009; Price 2009). P. vivax accounts for approximately 50% of all malaria episodes in Southern Asia and the Western Pacific, and up to 80% in South America and the Eastern Mediterranean Region (Mendis 2001; Price 2007). In Africa, P. vivax contributes less than 5% of malaria episodes but still results in 10-15 million cases annually. Estimates for the total burden of P. vivax range from 72 to 390 million clinical episodes per year (Price 2007).

Resistance of P. vivax to chloroquine, a cheap and simple treatment, is now widespread and has rendered this drug ineffective in parts of Indonesia and Papua New Guinea (Sumawinata 2003; Ratcliff 2007a; Karunajeewa 2008; Sutanto 2009). Low levels of resistance have also been reported from Burma, South Korea, Vietnam, India, Turkey, Ethiopia and parts of Southern Africa and South America (WHO 2006; Baird 2009; Price 2009). There is therefore an urgent need to find new and effective approaches to P. vivax therapy.

Description of the condition

Clinical syndromes

The clinical symptoms of malaria are caused by the direct and indirect effects of the blood stage parasite (the schizont) (Anstey 2009). Uncomplicated malaria is the milder form of the disease which presents as an acute febrile illness with headache, tiredness, muscle pains, abdominal pains, rigors, sweats, nausea and vomiting (WHO 2006).

In the absence of adequate treatment uncomplicated malaria can progress to severe or fatal forms of the disease characterized by severe anaemia, respiratory distress, renal failure, thrombocytopenia, and coma (Anstey 2009). Such severe disease with P. vivax has traditionally been considered to be extremely rare, however the development of more accurate diagnostic tests, particularly Polymerase Chain Reaction (PCR) techniques capable of reliably excluding P. falciparum as a cause, has revealed that severe disease due to P. vivax may cause significant morbidity (Barcus 2007; Tjitra 2008; Genton 2008; Kochar 2009; Baird 2009; Price 2009).

A clinical diagnosis of P. vivax malaria can be confirmed by detection and identification of the malaria parasite in the patient's blood using light microscopy, by rapid antigen testing kits, or by PCR techniques (rarely used in a clinical context) (WHO 2006).

The hypnozoite as the cause of spontaneous relapses

P. vivax differs from P. falciparum in having a liver stage, known as a hypnozoite, which can lie dormant in the liver following an acute infection. These hypnozoites are not susceptible to the antimalarial drugs used to treat the acute illness, and are the cause of spontaneous relapses which may occur weeks, months or even years after the initial episode (WHO 2006; Galappaththy 2007). The frequency and timing of these relapses varies geographically, reflecting variations in the behaviour of locally prevalent strains (Collins 1996; White 2002; Baird 2007). In the tropical regions of Southeast Asia and Oceania the risk of relapse is highest, with the first relapse occurring at around 3 to 5 weeks and often recurring repeatedly in the same patient (Collins 1996; Baird 1997; Baird 2007; Imwong 2007). Relapses in temperate regions are less common, often occur 6 to12 months later, and usually only once (Galappaththy 2007; Imwong 2007; Baird 2009).

Radical cure (complete removal of all stages of the malaria parasite from the infected patient) therefore requires additional therapy targeted at the hypnozoite. The only drug in common use for this purpose is primaquine, an 8-aminoquinolone. The current regimen recommended by the World Health Organization (WHO) is 15 mg/kg/day for 14 days, with higher doses necessary is some parts of Southeast Asia and Oceania (WHO 2006; Galappaththy 2007).

Assessing antimalarial efficacy in P.vivax

The hypnozoite causes considerable difficulty in the design and interpretation of P. vivax intervention trials. When P. vivax parasites reappear in the peripheral blood, which we will term a 'recurrence', there are no reliable methods to distinguish a recrudescence of the original infection (due to failure to adequately clear the schizont from the peripheral blood), from a spontaneous relapse (due to activation of the hypnozoite in the liver) or a re-infection (due to a new infection acquired from a subsequent mosquito bite) (WHO 2006; Baird 2009).

The most accurate test of a drugs ability to clear the blood stage would be to measure treatment failure prior to day 16, as post-treatment recurrent parasitaemia at this time-point will almost certainly be due to a recrudescence (WHO 2006; Baird 2009). This outcome however, may have little ability to discriminate between drugs, and is too early to exclude the possibility of subsequent recrudescence. Trials of artemisinin-based combination therapy (ACT) in the treatment of P. falciparum have recorded proven recrudescence well beyond 14 or even 28 days (Bloland 2003; Sinclair 2009).

Measuring recurrent parasitaemia at time points later than 16 days will include failures due to both recrudescences and relapses (Baird 2009). This distinction, though important in understanding the therapeutic value of a given treatment, may be of little interest to the patient as both may result in a further episode of clinical illness.

Demonstration of radical cure requires longer periods of follow-up, ideally up to one year, but is further complicated by the possibility that the observed recurrence is not due to recrudescence or relapse but to re-infections (White 2002; WHO 2006). The period of follow-up needed to make conclusions on radical cure, rather than delayed relapse, must exceed both the individual drug clearance time (the time taken to effectively reduce the drug levels to undetectable), and the expected relapse time of local strains (Bloland 2003; Baird 2009).

Drug resistance

Chloroquine resistance is a major problem, and there is further evidence of resistance developing among P. vivax strains to sulfadoxine-pyrimethamine (a common second-line treatment) and of decreased primaquine efficacy in parts of Southeast Asia (Baird 2009).

A 28-day test has been devised to assess chloroquine resistance in vivo. This defines resistance as the recurrence of P. vivax parasitaemia before day 28 in the presence of adequate chloroquine drug levels, regardless of the nature of the recurrence (recrudescence, relapse or re-infection). Prior to the development of resistance, recurrent P. vivax parasitaemia following chloroquine treatment was rarely seen before 28 days (Baird 1997; Baird 2009). It seems reasonable therefore to expect a replacement drug to also prevent recurrence (of any cause) for at least 28 days.

To combat the growing problem of resistance among P. falciparum strains, the WHO has now recommended that P. falciparum malaria is always treated using a combination of two drugs which act at different biochemical sites within the parasite. A parasite mutation producing resistance arising spontaneously during treatment, should then be killed by the partner drug, reducing or delaying the development of resistance and increasing the useful lifetime of each individual drug (White 1996; White 1999; WHO 2006). The availability of effective (and cheap) monotherapies for P. vivax, has until now negated the need for a similar approach with P. vivax. However, there are several reasons why ACT treatment for both species might now be desirable:

  • many episodes continue to be diagnosed clinically without identification of the species;

  • it would simplify national treatment protocols, and provide a rational approach to both proven, and undetected, P. falciparum and P. vivax co-infections;

  • it may decrease the market availability of monotherapies, and aid in the combat of resistance.

Description of the intervention

Artemisinin and its derivatives (artesunate, artemether and dihydroartemisinin) are short acting antimalarial drugs which have been shown to produce rapid relief from clinical symptoms and rapid clearance of the parasite from the peripheral blood (Pukrittayakamee 2000; WHO 2006). They are usually combined with a longer acting partner drug to produce ACTs. Until recently there had been no known resistance to the artemisinin derivatives but some resistance has now been reported in Southeast Asia and is being investigated (Dondorp 2009).

Artemisinin derivatives are generally reported as being well tolerated, and the safety profile of ACTs may be largely determined by the partner drug (Taylor 2004). Studies of artemisinin derivatives in animals have reported significant neurotoxicity (brain damage), but this has not been seen in human studies (Price 1999). Animal studies have also shown adverse effects on the early development of the foetus, but the artemisinin derivatives have not been fully evaluated during early pregnancy in humans. Other reported adverse events include gastrointestinal disturbance (stomach upset), dizziness, tinnitus (ringing in the ears), neutropenia (low levels of white blood cells), elevated liver enzymes (a marker for liver damage), and electrocardiographic (ECG) abnormalities (changes in cardiac conduction). Most studies, however, have found no evidence of ECG changes, and only non-significant changes in liver enzymes (WHO 2006; Nosten 2007). The incidence of type 1 hypersensitivity (allergic) reactions is reported to be approximately 1 in 3000 patients (Nosten 2007).

How the intervention might work

Therapy for P. vivax is divided into two components. First there is removal of the blood stage of the parasite by a drug with schizonticidal activity, traditionally chloroquine. Secondly there is treatment of the liver stage using a hypnozonticidal drug, usually primaquine, to achieve a radical cure. The artemisinin derivatives do not have a substantial effect on the liver stage of the parasite and therefore the primary aim of ACT treatment is eradication of the blood stage parasite and relief of clinical symptoms.(Pukrittayakamee 2000; WHO 2006).

In areas of high transmission, where the risk of new infections is high, the risks associated with primaquine, particularly the risk of haemolysis in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency, may outweigh the clinical benefits (WHO 2006). In this scenario, where primaquine is not routinely given, treatments which delay or prevent the initial relapses could have public health benefits (reduced clinical episodes, reduced loss of work, reduction in malaria associated anaemia).

A secondary aim of ACT treatment then, might be to delay clinical relapse, and there are now several trials which have investigated this (Hasugian 2007; Ratcliff 2007b). The ability of an ACT to delay the first relapse is, however, critically dependent on the half-life of the partner drug, with long half-life drugs still being present (in sufficient concentration) at the time of the first spontaneous relapse to clear the resultant blood stage parasites and prevent a recurrent clinical illness. This may be a desirable property in a drug combination from a patient perspective, but could contribute to further resistance developing as the partner drug is left unprotected by the rapidly eliminated artemisinin derivative.

Alternatively an ACT co-administered with the recommended 14 days of primaquine could demonstrate superiority to monotherapies plus primaquine if the ACT had fewer recrudescent failures, if it prevented re-infections (as a function of its long half-life), or if it delayed relapses due to primaquine failure.

In addition, the artemisinin derivatives have been shown to have an effect on developing Plasmodium gametocytes ie the stage of the parasite capable of infecting mosquitoes, and therefore an important stage in the transmission cycle (Nacher 2004; Price 1996). The interpretation of this indirect measure for reduced transmission is however unclear; there is evidence that even submicroscopic (undetectable) gametocytes are capable of transmission, and there is no consensus on what might constitute a clinically relevant reduction (Targett 2001). It is included as an outcome here for completeness.

Why it is important to do this review

Artemisinin combinations therapies are now the recommended therapy for P. falciparum malaria worldwide and despite the initial financial and programmatic hurdles they are now being rolled out in the majority of malaria endemic countries (WHO 2006).

As the effectiveness of chloroquine for P. vivax declines, alternative therapies are urgently needed, and ACTs are already being used to treat P. vivax, both intentionally and unintentionally, in many places. However, their exact role remains poorly defined, at least in part due to the methodological difficulties discussed above.

This review seeks to: assess the comparative effectiveness of ACTs in relation to the alternative non-ACTs; to elucidate any important differences between the ACTs currently available; and to discuss the merits of the alternative approaches.


To compare ACTs with alternative antimalarial regimens for treating acute uncomplicated P.vivax malaria and the prevention of relapses.


Criteria for considering studies for this review

Types of studies

Randomized controlled trials.

Types of participants

Adults and children (including pregnant or lactating women and infants) with symptomatic, microscopically confirmed, uncomplicated P. vivax malaria.

In studies recruiting patients with P. falciparum and P. vivax mono- or co-infections, only data from patients with proven P. vivax (+/- P. falciparum) at baseline will be included.

Types of interventions


A 3-day course of an ACT


A recognised non-ACT antimalarial therapy (chloroquine, sulfadoxine-pyrimethamine, amodiaquine, mefloquine, or combinations)


A 3-day course of an alternative ACT

(For the primary outcome 'Recurrence of P. vivax by day 14 or 28' we will subgroup trials according to those which also gave primaquine and those which did not. For the secondary outcome with prolonged follow-up 'Recurrence of P. vivax parasitaemia at 1 to12 months' we will only include trials which also give the WHO recommended dose of primaquine to both treatment arms)

Types of outcome measures

Primary outcomes

Recurrence of P. vivax parasitaemia by days 14 and 28.

(The term recurrence simply refers to reappearance of P. vivax parasites in the peripheral blood. It could include recrudescences, relapses and re-infections).

Secondary outcomes

Parasite clearance

Fever clearance

Recurrence of P. vivax parasitaemia at 1 to12 months (both regimens given with primaquine)

Haematological recovery as measured by changes in haemoglobin from baseline to last day of follow-up.

Measures of gametocyte carriage

Serious adverse events that require stopping treatment or admission to hospital

Search methods for identification of studies

We will attempt to identify all relevant studies regardless of language or publication status (published, unpublished, in press, ongoing).

Electronic searches

We will search the following databases: Cochrane Infectious Disease Group Specialized Register; Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library; MEDLINE; EMBASE; and LILACS, using the search terms detailed in Table 1. We will also search the metaRegister of Controlled Trials (mRCT) using  “vivax” and “arte* OR dihydroarte*”  as search terms.

Table 1. Detailed search strategy
  1. 1 Cochrane Infectious Disease Group Specialized Register; Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library; MEDLINE; EMBASE; and LILACS

Search set Search terms used for all databases1
42 or 3
51 and 4
6(search terms for RCTs)
75 and 6
8Limit 7  to Human

Searching other resources

We will check the reference lists of all trials identified by the methods described above. We will contact individual researchers and organizations working in the field, for unpublished and ongoing trials.

Data collection and analysis

Selection of studies

Felicity Brand (FB) and David Sinclair (DS) will independently review the results of the literature search and obtain full text copies of all trials thought to be relevant to this review. FB will scrutinize each trial report for evidence of multiple publications from the same data set.

FB and Nithya Gogtay (NG) will then independently assess the eligibility of each trial for inclusion in this review using an eligibility form based on the inclusion criteria stated above. We will report the excluded studies and the reasons for their exclusion. We will resolve any disagreements through discussion or, where necessary, by consultation with DS. If further clarification is necessary we will attempt to contact the authors for further information.

Data extraction and management

FB and NG will independently extract data using a pre-tested data extraction form. We will extract data on trial characteristics including: trial site, year, local malaria prevalence and transmission, trial methods, participants, interventions, and outcomes. We will resolve disagreements through discussion, review of the trial report, or, where necessary, by consultation with DS. Data on dose and drug ratio of the combinations will also be extracted.

We will extract the number randomized and the number analysed in each treatment group for each outcome. We will calculate and report the loss to follow up in each group.

For dichotomous outcomes, we will record the number of participants experiencing the event and the number of participants in each treatment group. For continuous outcomes, we will extract the arithmetic means and standard deviations for each treatment group together with the numbers of participants in each group. If the data have been reported using geometric means, we will record this information and extract standard deviations on the log scale. If medians have been extracted we will also extract ranges.

We will analyse data using Review Manager 5.

Assessment of risk of bias in included studies

FB and NG will independently assess the methodological quality of each trial, using the Cochrane Collaboration's tool for assessing the risk of bias (Review Manager 5), and discuss any differences of opinion with DS. We will follow the guidance to assess whether adequate steps were taken to reduce the risk of bias across six domains: sequence generation, allocation concealment, blinding (of participants, personnel and outcome assessors), incomplete outcome data, selective outcome reporting and other sources of bias. We will categorise these judgements as 'yes' (low risk of bias), 'no' (high risk of bias) or 'unclear'. Where our judgment is unclear we will attempt to contact the authors for clarification.

The risk of bias judgements will be displayed in a table and summarised in two diagrams: the 'risk of bias summary' and 'risk of bias graph'.

Measures of treatment effect

Dichotomous data will be presented and combined using risk ratios for both efficacy and safety. For continuous data summarised by arithmetic means and standard deviations, data will be combined using mean differences. Where continuous data have been summarised using geometric means they will be combined on the log scale using the generic inverse variance method and reported on the natural scale. When measures of effect, such as risk ratios or mean differences are presented, they will be accompanied by 95% confidence intervals. Medians and ranges will only be reported in tables.

Dealing with missing data

The primary analysis will be a complete-case analysis. Participants who were lost to follow-up or who developed P. falciparum parasitaemia during follow-up will be excluded. We will then conduct sensitivity analyses as described below to evaluate the robustness of this methodology. Participants who develop mixed parasitaemia during follow-up will be treated as failure in the same way as recurrence of P. vivax monoinfections.

If data from the trial reports are insufficient, unclear, or missing, we will attempt to contact the authors for additional information.

Assessment of heterogeneity

We will assess for heterogeneity amongst trials by inspecting the forest plots (to detect overlapping confidence intervals), reporting the I² statistic with a level of 50% to denote moderate levels of heterogeneity, and applying the Chi2 test with a P value of 0.10 to indicate statistical significance.

Assessment of reporting biases

We will consider publication bias using a funnel plot. We will keep it in mind that funnel plot asymmetry could be caused by publication bias, differences in methodological quality, or heterogeneity.

Data synthesis

The data will be analysed using RevMan 5 by FB with support from co-authors. Treatments will be compared directly using pair-wise meta-analyses. Meta-analyses will be stratified by geographical region where appropriate (to reflect any important differences in either transmission, timing of relapses, or resistance).

When no statistically significant heterogeneity is detected the fixed-effect meta-analysis model will be used. When moderate statistically significant heterogeneity is observed within groups that cannot be explained by sub-group or sensitivity analyses a random-effects meta-analysis model will be used to synthesize the data. When a pooled meta-analysis result is considered to be meaningless because of clinical or substantial statistical heterogeneity the results will be presented in a forest plot without a pooled estimate of effect.

Subgroup analysis and investigation of heterogeneity

We will investigate heterogeneity by conducting subgroup analysis to evaluate the contribution of differences in trial characteristics such as: risk of bias, age of participants, patterns of resistance, transmission intensity, differences in half-life of drugs and co-administration of primaquine.

Sensitivity analysis

We will conduct the following sensitivity analyses to restore the integrity of the randomization process and test the robustness of our results:

  • Sensitivity analysis 1: P. falciparum cases added in as failures.

  • Sensitivity analysis 2: All exclusions added in as failures.

  • Sensitivity analysis 3: P. falciparum cases added in as successes.

  • Sensitivity analysis 4: All exclusions added in as successes.


We would like to acknowledge the contribution of Drs R. Price and JK Baird whose useful comments and insights contributed greatly to the quality of this protocol. The editorial base for the Cochrane Infectious Disease Group is funded by the Department for International Development (DFID), UK, for the benefit of developing countries.


Protocol first published: Issue 4, 2010

Contributions of authors

All authors were involved in the conception and design of this protocol.

Declarations of interest

None known