Artemisinin‐based combination therapy for treating uncomplicated malaria

David Sinclair; Babalwa Zani; Sarah Donegan; Piero Olliaro; Paul Garner

doi:10.1002/14651858.CD007483.pub2

. 2009 Jul 8;2009(3):CD007483. doi: 10.1002/14651858.CD007483.pub2

Artemisinin‐based combination therapy for treating uncomplicated malaria

David Sinclair ^1,^✉, Babalwa Zani ², Sarah Donegan ¹, Piero Olliaro ³, Paul Garner ¹

Editor: Cochrane Infectious Diseases Group

PMCID: PMC6532584 PMID: 19588433

Abstract

Background

The World Health Organization recommends uncomplicated P. falciparum malaria is treated using Artemisinin‐based Combination Therapy (ACT). This review aims to assist the decision making of malaria control programmes by providing an overview of the relative benefits and harms of the available options.

Objectives

To compare the effects of ACTs with other available ACT and non‐ACT combinations for treating uncomplicated P. falciparum malaria.

Search methods

We searched the Cochrane Infectious Diseases Group Specialized Register; the Cochrane Central Register of Controlled Trials (CENTRAL); MEDLINE; EMBASE; LILACS, and the metaRegister of Controlled Trials (mRCT) to March 2009.

Selection criteria

Randomized head to head trials of ACTs in uncomplicated P. falciparum malaria.

This review is limited to: dihydroartemisinin‐piperaquine; artesunate plus mefloquine; artemether‐lumefantrine (six doses); artesunate plus amodiaquine; artesunate plus sulfadoxine‐pyrimethamine and amodiaquine plus sulfadoxine‐pyrimethamine.

Data collection and analysis

Two authors independently assessed trials for eligibility and risk of bias, and extracted data. We analysed primary outcomes in line with the WHO 'Protocol for assessing and monitoring antimalarial drug efficacy' and compared drugs using risk ratios (RR) and 95% confidence intervals (CI). Secondary outcomes were effects on P. vivax, gametocytes, haemoglobin, and adverse events.

Main results

Fifty studies met the inclusion criteria. All five ACTs achieved PCR adjusted failure rates of < 10%, in line with WHO recommendations, at most study sites.

Dihydroartemisinin‐piperaquine performed well compared to the ACTs in current use (PCR adjusted treatment failure versus artesunate plus mefloquine in Asia; RR 0.39, 95% CI 0.19 to 0.79; three trials, 1062 participants; versus artemether‐lumefantrine in Africa; RR 0.39, 95% CI 0.24 to 0.64; three trials, 1136 participants).

ACTs were superior to amodiaquine plus sulfadoxine‐pyrimethamine in East Africa (PCR adjusted treatment failure versus artemether‐lumefantrine; RR 0.12, 95% CI 0.06 to 0.24; two trials, 618 participants; versus AS+AQ; RR 0.44, 95% CI 0.22 to 0.89; three trials, 1515 participants).

Dihydroartemisinin‐piperaquine (RR 0.32, 95% CI 0.24 to 0.43; four trials, 1442 participants) and artesunate plus mefloquine (RR 0.30, 95% CI 0.21 to 0.41; four trials, 1003 participants) were more effective than artemether‐lumefantrine at reducing the incidence of P.vivax over 42 days follow up.

Authors' conclusions

Dihydroartemisinin‐piperaquine is another effective first‐line treatment for P. falciparum malaria.

The performance of the non‐ACT (amodiaquine plus sulfadoxine‐pyrimethamine) falls below WHO recommendations for first‐line therapy in parts of Africa.

In areas where primaquine is not being used for radical cure of P. vivax, ACTs with long half‐lives may provide some benefit.

23 April 2019

No update planned

Review superseded

Please refer to the Cochrane Special Collection: Sinclair 2014 https://doi.org/10.1002/14651858.SC000007/full

Plain language summary

Artemisinin‐based combination treatments for uncomplicated malaria

Malaria is a major cause of illness and death in many of the world's poorest countries. It is spread from person to person by the bite of mosquitoes infected with a microorganism called Plasmodium. The Plasmodium species P. falciparum is the most common cause of malaria worldwide and causes the majority of deaths. Uncomplicated malaria is the mild form of the disease which, if left untreated, can progress rapidly to become life threatening. The drugs traditionally used to treat uncomplicated malaria have become ineffective in many parts of the world due to the development of drug resistance.

The World Health Organization now recommends Artemisinin‐based Combination Therapy (ACTs) for treating uncomplicated malaria. The ACTs combine an artemisinin‐derivative (a relatively new group of drugs which are very effective) with another longer‐lasting drug to try and reduce the risk of further resistance developing.

This review summarizes the relative benefits and harms of the four ACTs in common use, one relatively new ACT (dihydroartemisinin plus piperaquine), and one combination which does not contain an artemisinin derivative but remains in use in some African countries (amodiaquine plus sulfadoxine‐pyrimethamine).

All five ACTs were shown to be highly effective at treating P. falciparum in most places where they have been studied. However, there were several trials where ACTs had high levels of treatment failure, which emphasises the need to continue to monitor their performance.

The new ACT, dihydroartemisinin plus piperaquine, was shown to be at least as effective as the ACTs currently in widespread use in Asia and Africa, and represents another option for malaria treatment.

ACTs were shown to be more effective than amodiaquine plus sulfadoxine‐pyrimethamine in countries from East Africa which probably represents high levels of resistance, to both drugs in this combination, in this region.

The second most common form of malaria, P. vivax, can also be treated with ACTs but requires additional treatment to cure the patient completely. This is because the P. vivax parasite can lie dormant in the liver for months or years before becoming active again. ACTs where the partner drug has a long duration of action may help to delay these relapses.

The ACTs seem to be relatively safe with few serious side effects. Minor side effects are more common but can be difficult to distinguish from the symptoms of malaria itself. Fifty trials were included in this review but did not include the most vulnerable populations; pregnant women and young infants (age < six months).

Background

Malaria is a disease of global public health importance. Its social and economic burden is a major obstacle to human development in many of the world's poorest countries. In heavily affected countries, malaria alone accounts for as much as 40% of public health expenditure, 30% to 50% of hospital admissions, and up to 60% of outpatient visits (WHO 2007). It has an annual incidence of approximately 250 million episodes and is the cause of more than a million deaths, most of them in infants, young children, and pregnant women (WHO 2008b).

Malaria is transmitted from person to person by the bite of mosquitoes infected with the protozoan parasite Plasmodium. Four Plasmodium species are capable of causing malaria in humans: P. falciparum, P. vivax, P. malariae, and P. ovale. Of these P. falciparum is responsible for over 90% of cases and almost all of the malaria deaths worldwide (WHO 2008b). P. vivax is also common and often presents as a co‐infection with P. falciparum in a single illness (Mayxay 2004). Uncomplicated malaria is the mild form of the disease which presents as a febrile illness with headache, tiredness, muscle pains, abdominal pains, rigors (severe shivering), and nausea and vomiting. If left untreated P. falciparum malaria can rapidly develop into severe malaria with anaemia (low haemoglobin in the blood), hypoglycaemia (low blood sugar), renal failure (kidney failure), pulmonary oedema (fluid in the lungs), convulsions (fitting), coma, and eventually death (WHO 2006). A clinical diagnosis of malaria can be confirmed by detection of the malaria parasite in the patient's blood. This has traditionally been done by light microscopy but increasingly rapid diagnostic tests are being used.

Resistance of P. falciparum to the traditional antimalarial drugs (such as chloroquine, sulfadoxine‐pyrimethamine, amodiaquine, and mefloquine) is a growing problem and is thought to have contributed to increased malaria mortality in recent years (WHO 2006). Chloroquine resistance has now been documented in all regions except Central America and the Caribbean. There is high‐level resistance to sulfadoxine‐pyrimethamine throughout South East Asia and increasingly in Africa. Mefloquine resistance is common in the border areas of Cambodia, Myanmar, and Thailand, but uncommon elsewhere. Resistance of P. vivax to sulfadoxine‐pyrimethamine is also increasing, and chloroquine resistance has been reported in some parts of Asia and Oceania (WHO 2006).

Artemisinin‐based antimalarials

Artemisinin and its derivatives (such as artesunate, artemether, and dihydroartemisinin) are antimalarial drugs with a unique structure and mode of action. The first published report of clinical trials appeared in the Chinese Medical Journal in 1979 (Qinghaosu 1979). Until recently there had been no reported resistance to the artemisinin derivatives; however the possibility of emerging resistance, on the Thai‐Cambodian border, is currently being investigated (WHO 2008a).

Artemisinin derivatives have been shown to produce faster relief of clinical symptoms and faster clearance of parasites from the blood than other antimalarial drugs (McIntosh 1999; Adjuik 2004; WHO 2006). When used as monotherapy, the short half‐life of the artemisinin derivatives (and rapid elimination from the blood) means that patients must take the drug for at least seven days (Meshnick 1996; Adjuik 2004). Failure to complete the course, due to the rapid improvement in clinical symptoms, can lead to high levels of treatment failure even in the absence of drug resistance. Artemisinin derivatives are therefore usually given with another longer‐acting drug, with a different mode of action, in a combination known as artemisinin‐based combination therapy or ACT. These combinations can then be taken for shorter durations than artemisinin alone (White 1999; WHO 2006).

The artemisinin derivatives also reduce the development of gametocytes (the sexual form of the malaria parasite that is capable of infecting mosquitoes) and consequently the carriage of gametocytes in the peripheral blood (Price 1996; Targett 2001). This reduction in infectivity has the potential to reduce the post‐treatment transmission of malaria (particularly in areas of low or seasonal transmission), which may have significant public health benefits (WHO 2006).

Artemisinin and its derivatives are generally reported as being safe and well tolerated, and the safety profile of ACTs may be largely determined by the partner drug (WHO 2006; Nosten 2007). 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 fetus, but the artemisinin derivatives have not been fully evaluated during early pregnancy in humans (Nosten 2007). Other reported adverse events include gastrointestinal (GI) 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).

Assessing antimalarial efficacy

The World Health Organization (WHO) recommends that first‐line antimalarials should have a treatment failure rate of less than 10%, and failure rates higher than this should trigger a change in treatment policy (WHO 2006). Treatment failure can be classified as:

Early treatment failure:

the development of danger signs or severe malaria on days one, two, three in the presence of parasitaemia;
parasitaemia on day two higher than on day 0;
parasitaemia and axillary temperature > 37.5 °C on day three;
parasitaemia on day three > 20% of count on day 0.

or late treatment failure:

development of danger signs, or severe malaria, after day three with parasitaemia;
presence of P. falciparum parasitaemia and axillary temperature > 37.5 °C on or after day four;
presence of P. falciparum parasitaemia after day seven.

The late reappearance of P. falciparum parasites in the blood can be due to failure of the drug to completely clear the original parasite (a recrudescence) or due to a new infection, which is especially common in areas of high transmission. A molecular genotyping technique called polymerase chain reaction (PCR) can be used in clinical trials to distinguish between recrudescence and new infection, giving a clearer picture of the efficacy of the drug and its post‐treatment prophylactic effect (White 2002; Cattamanchi 2003).

The WHO recommends a minimum follow‐up period of 28 days for antimalarial efficacy trials, but longer periods of follow up may be required for antimalarials with long elimination half‐lives (White 2002; WHO 2003). This is because treatment failure due to true recrudescence of malaria parasites may be delayed until the drug concentration falls below the minimum concentration required to inhibit parasite multiplication, which may be beyond 28 days. The WHO recommends 42 days follow up for trials involving lumefantrine and 63 days for trials of mefloquine (WHO 2003).

P. vivax malaria

P. vivax differs from P. falciparum in generally producing a milder illness and in having a liver stage known as a hypnozoite. These hypnozoites can lie dormant in the liver following an acute infection and cause spontaneous relapses at later dates.

As P. vivax often co‐exists with P. falciparum in a single illness, it is important to assess the effect of ACTs on the P. vivax parasite (Mayxay 2004; WHO 2006). ACTs have been shown to clear P. vivax from the peripheral blood, but they do not have a substantial effect on the liver stage of the parasite (Pukrittayakamee 2000). Although ACTs cannot provide a radical cure for P. vivax, their ability to delay the eventual relapse of P. vivax and provide a prolonged malaria free period may produce significant public health benefits.

It is important to note that when P. vivax parasitaemia occurs following initial treatment, PCR is unable to distinguish a recrudescence of the original infection (due to failure to clear the parasite from the peripheral blood) from a spontaneous relapse (due to failure to clear the liver stage) (WHO 2006).

Choice of combination treatment

The WHO now recommends that P. falciparum malaria is always treated using a combination of two drugs that act at different biochemical sites within the parasite (WHO 2006). If a parasite mutation producing resistance arises spontaneously during treatment, the parasite should then be killed by the partner drug, thereby reducing or delaying the development of resistance to the artemisinin derivatives, and increasing the useful lifetime of the individual drugs (White 1996; White 1999; WHO 2006). This policy emerged at the time when ACTs were primarily being considered, but other possibilities such as amodiaquine combined with sulfadoxine‐pyrimethamine (non‐ACTs) are also available.

The decision of which ACT to adopt into national malaria control programmes has been based on a combination of research and expert opinion. Systematic reviews can contribute to this decision by providing evidence on the:

relative effects on cure between combinations;
absolute cure levels achieved by a drug in a particular region;
safety and risk of adverse effects of the combination;
impact on gametocytes;
impact on haemoglobin levels; and
relative effects on P. vivax.

Other information that is also important to decision‐making include:

the appropriateness of the partner drug within a locality, based on informed judgements related to regional and national overviews of drug resistance and the intensity of malaria transmission;
the simplicity of the treatment regimen (co‐formulated products are generally preferred as they reduce the availability and use of monotherapy, which may in turn reduce the development of resistance);
the cost (since the ACT is likely to represent a large percentage of the annual health expenditure in highly endemic countries); and
other concerns such as fetal toxicity and teratogenicity.

To contribute to informed decision‐making, we have examined the comparative effects of ACTs for which co‐formulated products are currently available or shortly to be made available. We have included trials that have used co‐packaged or loose preparations of these same ACTs to provide information on relative effects of the different treatment options. While recent Cochrane Reviews have synthesized the evidence around individual ACT comparisons (Bukirwa 2005; Omari 2005; Bukirwa 2006; Omari 2006), this review broadens the inclusion criteria and pools the data into a single Cochrane Review. A comprehensive list of the available drugs and the treatment comparisons that have been assessed is shown in Appendix 1. The data are presented in answer to four questions:

How does dihydroartemisinin‐piperaquine (DHA‐P) perform?
How does artesunate‐mefloquine (AS+MQ) perform?
How does artemether‐lumefantrine (AL6) perform?
How does artesunate plus amodiaquine (AS+AQ) perform?

The comparison drugs were any of the above plus artesunate plus sulfadoxine‐pyrimethamine (AS+SP) and amodiaquine plus sulfadoxine‐pyrimethamine (AQ+SP).

Objectives

To compare the effects of ACTs with other available ACT and non‐ACT combinations for treating uncomplicated P. falciparum malaria.

A secondary objective was to explore the effects of the combinations on P. vivax infection.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials. Quasi‐randomized studies were excluded.

Types of participants

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

Trials that included participants with P. vivax co‐infection and mono‐infection were also eligible.

Types of interventions

Intervention

Three‐day course of an ACT (fixed dosed, co‐blistered, or individually packaged (loose)).

Control

Three‐day course of an alternative ACT or non‐artemisinin combination treatment (amodiaquine plus sulfadoxine‐pyrimethamine).

The specific ACTs included are: dihydroartemisinin‐piperaquine; artesunate plus mefloquine; artemether‐lumefantrine (six doses); artesunate plus amodiaquine and artesunate plus sulfadoxine‐pyrimethamine (Appendix 1).

Types of outcome measures

Primary outcomes

Total failure at days 28, 42, and 63; PCR‐adjusted and PCR‐unadjusted.

Secondary outcomes

P. vivax parasitaemia at day 28, 42, or 63 (all participants).
P. vivax parasitaemia at day 28, 42, or 63 (only participants with P. vivax at baseline).
Gametocyte carriage at day 7 or 14 (preference for day 14 in data analysis).
Gametocyte development (negative at baseline, and positive at follow up).
Change in haemoglobin from baseline (minimum 28 day follow up).

Adverse events

Deaths occurring during follow up.
Serious adverse events (life threatening, causing admission to hospital, or discontinuation of treatment).
Haematological and biochemical adverse effects (e.g. neutropenia, liver toxicity).
Early vomiting.
Other adverse events.

Search methods for identification of studies

Electronic searches

We searched the following databases using the search terms detailed in Appendix 2: Cochrane Infectious Diseases Group Specialized Register (March 2009); Cochrane Central Register of Controlled Trials (CENTRAL) published in The Cochrane Library (2009, issue 1); MEDLINE (1966 to March 2009); EMBASE (1974 to March 2009); and LILACS (1982 to March 2009). We also searched the metaRegister of Controlled Trials (mRCT) using 'malaria' and 'arte* OR dihydroarte*' as search terms (March 2009).

Searching other resources

We contacted individual researchers working in the field, organizations including the World Health Organization, and pharmaceutical companies (Atlantic, Guilin, Holleykin, HolleyPharm, Mepha, Novartis, Parke‐Davis, Pfizer, Sanofi‐Aventis, Roche) for information on unpublished trials (August 2008).

We also checked the reference lists of all trials identified by the methods described above.

Data collection and analysis

Selection of studies

David Sinclair (DS) and Babalwa Zani (BZ) reviewed the results of the literature search and obtained full‐text copies of all potentially relevant trials. DS scrutinized each trial report for evidence of multiple publications from the same data set. DS and BZ then independently assessed each trial for inclusion in this review using an eligibility form based on the inclusion criteria. We resolved any disagreements through discussion or, where necessary, by consultation with Paul Garner (PG). If clarification was necessary we attempted to contact the trial authors for further information. We have listed the trials that were deemed ineligible and the reasons for their exclusion in the 'Characteristics of excluded studies' table.

Data extraction and management

DS and BZ independently extracted data using a pre‐tested data extraction form. We extracted data on trial characteristics including methods, participants, interventions, and outcomes as well as data on dose and drug ratios of the combinations.

We extracted the number randomized and the number analysed in each treatment group for each outcome. We calculated and reported the loss to follow up in each group.

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

Primary outcome

The primary analysis drew on the WHO's protocol for assessing and monitoring antimalarial drug efficacy (WHO 2003). This protocol has been used to guide most efficacy trials since its publication in 2003, even though it was designed to assess the level of antimalarial resistance in the study area rather than for comparative trials. As a consequence a high number of randomized participants are excluded from the final efficacy outcome as losses to follow up or voluntary or involuntary withdrawals. For this reason we conducted a sensitivity analysis which aimed to restore the integrity of the randomization process (as is usual in trial analysis) and test the robustness of the results to this methodology. (For a summary of the methodology and sensitivity analysis see Appendix 3)

PCR‐unadjusted total failure

PCR‐unadjusted total failure (P. falciparum) was calculated as the sum of early treatment failures and late treatment failures (without PCR adjustment). The denominator excludes participants for whom an outcome was not available (e.g. those who were lost to follow up, withdrew consent, took other antimalarials, or failed to complete treatment) and those participants who were found not to fulfil the inclusion criteria after randomization.

PCR‐adjusted total failure

PCR‐adjusted total failure (P. falciparum) was calculated as the sum of early treatment failures, and late treatment failures due to PCR‐confirmed recrudescence. Participants with indeterminate PCR results, missing PCR results, or PCR‐confirmed new infections were treated as involuntary withdrawals and excluded from the calculation. Late treatment failures that occurred between days 4 and 14 were assumed to be recrudescences of the original parasite without the need for PCR genotyping (unless genotyped in the trial). The denominator excludes participants for whom an outcome was not available (e.g. those who were lost to follow up, withdrew consent, took other antimalarials, or failed to complete treatment) and those participants who were found not to fulfil the inclusion criteria after randomization.

These primary outcomes relate solely to failure due to P. falciparum. For both PCR‐unadjusted and PCR‐adjusted total failure, participants who experienced P. vivax during follow up were retained in the calculation if they were treated with chloroquine and continued in follow up. As long as they did not go on to develop P. falciparum parasitaemia they were classified as treatment successes. We excluded from the calculation those participants who experienced P. vivax and were removed from the trial's follow up at the time of P. vivax parasitaemia.

It was not always possible to guarantee that individual trials used the standard WHO definitions. We have accepted the trial authors' data unless we had specific reason to reclassify an individual participant or reject the data. Where this has been done we have stated clearly the reasons for doing so.

Secondary outcomes and adverse events

In a secondary analysis we examined the effects of ACTs on P. vivax. We have reported the incidence of P. vivax parasitaemia during follow up at days 28, 42, and 63. Where possible, we have stratified this analysis into participants who had P. vivax co‐infection at baseline and those negative for P. vivax at baseline.

Extracting data on gametocyte carriage was difficult due to the variety of ways that these data are presented in individual papers. In order to try to present useful data we contacted the lead author of all trials that reported on gametocytes for additional information which fitted our specified outcomes.

Haematological outcomes were also presented in a multitude of ways which prevented meta‐analysis. We have therefore presented these data as a narrative summary with forest plots where possible.

Other secondary outcomes have been presented using forest plots, tables, or narrative summaries as appropriate.

We extracted the number of serious adverse events and deaths and have presented these data in a forest plot. We have only included those trials that specifically report serious adverse events.

Data on early vomiting were extracted as a measure of tolerability of these combinations, and are presented as a forest plot. Other adverse events are presented in tables with a narrative summary.

Assessment of risk of bias in included studies

DS and BZ independently assessed the risk of bias for each trial using 'The Cochrane Collaboration's tool for assessing the risk of bias' (Higgins 2008). Differences of opinion were discussed with PG. We followed the guidance to assess whether adequate steps had been 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 have categorized these judgments as 'yes' (low risk of bias), 'no' (high risk of bias), or 'unclear'. Where our judgement is unclear we attempted to contact the trial authors for clarification.

This information was used to guide the interpretation of the data that are presented.

Measures of treatment effect

We analysed the data using Review Manager 5. Dichotomous data are presented and combined using risk ratios. For continuous data summarized by arithmetic means and standard deviations, data have been combined using mean differences. Risk ratios and mean differences are accompanied by 95% confidence intervals. Medians and ranges are only reported in tables.

Dealing with missing data

If data from the trial reports were insufficient, unclear, or missing, we attempted to contact the trial authors for additional information. If we judged the missing data to render the result uninterpretable we excluded the data from the meta‐analysis and clearly stated the reason. The potential effects of missing data have been explored through a series of sensitivity analyses (Appendix 3).

Assessment of heterogeneity

We assessed for heterogeneity amongst trials by inspecting the forest plots, applying the Chi² test with a 10% level of statistical significance, and also using the I² statistic with a value of 50% used to denote moderate levels of heterogeneity.

Data synthesis

The included trials have been given identity codes which include the first author, the year the study was conducted (not the year it was published) and the three‐letter international country code. Studies in forest plots are also listed in chronological order (by the final date of enrolment). We hope this will aid with interpretation of the review and forest plots.

Treatments have been compared directly using pair‐wise comparisons. For outcomes that are measured at different time points we have stratified the analysis by the time point. The primary outcome analysis is also stratified by geographical region as a crude marker for differences in transmission and resistance patterns.

Meta‐analysis has been performed within geographic regions where appropriate after assessment and investigation of heterogeneity. A random‐effects model was used where the Chi² test P value was less than 0.1 or the I² statistic was greater than 50%.

In addition, Olliaro‐Vaillant plots have been used to simultaneously display the absolute and relative benefits of individual ACTs at day 28.

Subgroup analysis and investigation of heterogeneity

We investigated potential sources of heterogeneity through the following subgroup analyses: geographical region, intensity of malaria transmission (low to moderate versus high malaria transmission), known parasite resistance, allocation concealment, participant age, and drug dose (comparing regimens where there are significant variations in drug dose).

Sensitivity analysis

We conducted a series of sensitivity analyses to investigate the robustness of the methodology used in the primary analysis. Our aim was to restore the integrity of the randomization process by adding excluded groups back into the analysis in a stepwise fashion (see Appendix 3 for details). Where these analyses altered the direction or significance of the measure of effect the revised results are presented and discussed.

Results

Description of studies

Results of the search

The search was conducted on 12 August 2008 and repeated on 26 March 2009. In total 517 trials were identified. Full text copies were obtained for 85 trials. Fifty trials are included in this review and 35 were excluded. A further four trials (Bousema 2004 KEN; Koram 2003 GHA; Martensson 2003 TZA; Van den Broek 2004 ZAR) were excluded from the primary analysis due to baseline differences between groups which had the potential to severely bias the result. These trials were retained for their data on adverse events.

Included studies

Forty‐six of the fifty included trials were conducted between 2003 and 2009.

Thirty‐one trials were conducted in Africa, 17 in Asia, one in South America (DHA‐P versus AS+MQ) and one in Oceania (DHA‐P versus AL6 versus AS+SP). There is obvious regional variability in which drugs are being studied. Trials from Asia mainly involve AS+MQ, AL6 and DHA‐P (plus one trial from Indonesia with AS+AQ). Only two studies from Africa have evaluated AS+MQ.

Pregnant and lactating women were excluded from all trials. The study population in Asian trials is older, with exclusion of children aged less than one year. African studies concentrated more on children and included those as young as six months.

Three trials (Hasugian 2005 IDN; Karunajeewa 2007 PNG; Ratcliff 2005 IDN) included participants with P. vivax mono‐infection at baseline. For our primary analysis we obtained data from the authors for only those participants who had P. falciparum or mixed infection (falciparum andvivax) at baseline.

One trial (Dorsey 2006 UGA) had an unusual study design where participants were followed up for more than one episode of malaria. For our primary analysis we obtained data from the authors for first episodes of malaria only.

The characteristics of the included studies are given in the 'Characteristics of included studies' table.

Excluded studies

The reasons for exclusion are given in the 'Characteristics of excluded studies' table.

The four additional studies excluded from the primary analysis had different inclusion criteria for different arms of the trial. Children aged less than one year were excluded from the AL6 treatment arm and reassigned to either AS+AQ or AS+SP. In these studies this led to significant baseline differences in age and weight, factors known to be associated with the outcomes. We explored the effects of including these trials in the largest meta‐analysis (AL6 versus AS+AQ, Analysis 9.9; Analysis 9.10). Inclusion of the trials with this bias shifted the results from no difference detected to favouring AL6. In the light of this we decided to exclude all trials that had systematically reallocated patients after randomization.

9.9 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 9 Sensitivity analysis: Total Failure Day 28 PCR unadjusted.

9.10 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 10 Sensitivity analysis: Total Failure Day 28 PCR adjusted.

Risk of bias in included studies

For a summary of the 'Risk of bias' assessments please see Figure 1 and Figure 2.

Methodological quality summary: review authors' judgements about each methodological quality item for each included study.

Methodological quality graph: review authors' judgements about each methodological quality item presented as percentages across all included studies.

Allocation

Generation of the randomized sequence was judged to be at low risk of bias for 33 trials, high risk of bias for 1 trial, and 16 trials were unclear regarding randomization methods.

Allocation concealment was judged to be at low risk of bias in 21 studies, high risk of bias in 19 studies and unclear in 10 studies. Descriptions which included the following details were accepted as adequate for concealment: opaque sealed envelopes; sealed sequentially numbered envelopes; or third party allocation. For primary outcomes we conducted a sensitivity analysis including only the trials with adequate allocation concealment.

Blinding

Of the included trials only 10 were judged to be at low risk of bias due to adequate blinding. Blinding or quality control of laboratory staff was conducted in 34 studies. Although this may be reassuring with regard to parasitological outcomes, secondary outcomes and particularly adverse event reporting will remain at high risk of bias.

Incomplete outcome data

We have reported the proportion of participants in each treatment arm for whom an outcome was not available and conducted sensitivity analyses to test the possible effect of these losses. Eight trials were judged to be at high risk of bias due to either moderate drop‐out (> 15%), differential drop‐out between groups that had the potential to alter the result, or participants missing from the primary analysis who could not be accounted for.

Selective reporting

Due to the varying half‐lives of drugs, the choice of which day to measure outcomes can influence the comparative effects of the drugs. If a drug with a long half‐life (DHA‐P or AS+MQ) is compared to a drug with a short half‐life (AS+AQ or AS+SP), day 28 outcomes may underestimate PCR adjusted failure with the long half‐life drug. At later time points (day 42 and 63) drugs with long half‐lives are likely to appear superior in preventing new infections (PCR unadjusted failure) which represents a prophylactic effect. We have kept this in mind when interpreting the data but did not judge the trials to be at high risk of bias.

Other potential sources of bias

Pharmaceutical companies provided financial support or study drugs in 15 trials. Further involvement of the pharmaceutical company in trial design or reporting is only described in one study (Djimde 2004 MLI).

Effects of interventions

In April 2009 we conducted the sensitivity analysis as described in Table 3 to test the robustness of our methodology. In general these analyses did not substantially change the direction, magnitude, or confidence intervals of the estimate of effect. Examples are shown in Analysis 1.12 and Analysis 1.13. Only sensitivity analyses of interest remain linked in this review.

1.12 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 12 Sensitivity analysis: Total Failure Day 63 PCR unadjusted.

1.13 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 13 Sensitivity analysis: Total Failure Day 63 PCR adjusted.

Question 1. How does dihydroartemisinin‐piperaquine (DHA‐P) perform?

Dosing concerns

Two dosing regimens have been commonly used in clinical trials of DHA‐P. These two regimens give the same total dose, but divided into three or four doses, given over three days. One trial (Ashley 2004 THA) directly compared the three‐dose regimen with the four‐dose regimen and found no difference at any time point (one trial, 318 participants, Analysis 14.1, Analysis 14.2).

14.1 — Comparison 14 Dihydroartemisinin‐piperaquine dose analysis: 3 dose vs 4 dose regimen, Outcome 1 Total Failure PCR unadjusted.

14.2 — Comparison 14 Dihydroartemisinin‐piperaquine dose analysis: 3 dose vs 4 dose regimen, Outcome 2 Total Failure PCR adjusted.

In comparisons comparing DHA‐P to AS+MQ, four trials used the three‐dose regimen, three trials used the four‐dose regimen and one trial used both. Stratifying the analysis by dosing regimen did not reveal any significant differences in efficacy between the two regimens (Analysis 15.1; Analysis 15.2; Analysis 15.3; Analysis 15.4; Analysis 15.5; Analysis 15.6).

15.1 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 1 Total Failure Day 63 PCR unadjusted.

15.2 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 2 Total Failure Day 63 PCR adjusted.

15.3 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 3 Total Failure Day 42 PCR unadjusted.

15.4 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 4 Total Failure Day 42 PCR adjusted.

15.5 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 5 Total Failure Day 28 PCR unadjusted.

15.6 — Comparison 15 Dihydroartemisinin‐piperaquine dose analysis (versus Artesunate plus mefloquine), Outcome 6 Total Failure Day 28 PCR adjusted.

Comparison 1. DHA‐P versus artesunate plus mefloquine

We found nine trials which assessed this comparison (eight in Asia and one in South America). Allocation concealment was assessed as 'low risk of bias' in five trials (Ashley 2003a THA; Ashley 2003b THA; Ashley 2004 THA; Grande 2005 PER; Mayxay 2004 LAO). Laboratory staff (outcome assessors) were blinded to treatment allocation in three trials (Ashley 2003a THA; Ashley 2003b THA; Ashley 2005 THA), and no other blinding is described.

Total failure

PCR adjusted treatment failure with DHA‐P was below 5% in all nine studies, and with AS+MQ in seven out of nine studies.

At day 63 comparative results were mixed. Trials from Asia favoured DHA‐P (Day 63, three trials, 1182 participants: PCR unadjusted RR 0.73, 95% CI 0.54 to 0.98, Analysis 1.1; PCR adjusted RR 0.39, 95% CI 0.19 to 0.79, Analysis 1.2) and the one trial from South America favoured AS+MQ (one trial, 445 participants: PCR unadjusted RR 6.19, 95% CI 1.40 to 27.35, Analysis 1.1; PCR adjusted no significant difference, Analysis 1.2). This difference may reflect the level of mefloquine resistance at the study sites. The performance of DHA‐P in the study in South America is similar to that in Asia, but the performance of AS+MQ was much improved with no PCR confirmed recrudescences.

1.1 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 1 Total Failure (P. *falciparum*) Day 63 PCR unadjusted.

1.2 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 2 Total Failure (P. *falciparum*) Day 63 PCR adjusted.

No significant differences were shown at other time points (Day 42, five trials, 1969 participants, Analysis 1.3, Analysis 1.4; Day 28, six trials, 2034 participants, Analysis 1.5, Analysis 1.6).

1.3 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 3 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

1.4 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 4 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

1.5 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 5 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

1.6 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 6 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

P. vivax

No significant difference was shown in the incidence of P. vivax parasitaemia at any time point (Day 63, four trials, 1661 participants; Day 42, three trials, 1251 participants; Day 28, one trial, 402 participants; Analysis 1.7). There were no significant differences in the incidence of P. vivax between groups with or without P. vivax at baseline.

1.7 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 7 P. *vivax* parasitaemia.

Gametocytes

The number of participants who developed detectable gametocytes (after being negative at baseline) was low in both groups, but significantly lower with AS+MQ (three trials, 1234 participants: RR 3.06, 95% CI 1.13 to 8.33, Analysis 1.8). AS+MQ may also clear gametocytes quicker than DHA‐P but the analysis is confounded by differences in gametocyte carriage at baseline (two trials, 1174 participants, Analysis 1.9).

1.8 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 8 Gametocyte development (in those negative at baseline).

1.9 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 9 Gametocytaemia carriage.

Anaemia

Five trials report on haematological changes. Individual studies did not show significant differences between groups (see Appendix 5). Two trials (Ashley 2003b THA; Ashley 2004 THA) report a decrease in haematocrit over the first seven days followed by recovery in both groups (figures not reported).

Adverse events

No difference has been shown in the frequency of serious adverse events (seven trials, 2374 participants, Analysis 1.10).

1.10 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 10 Serious adverse events (including deaths).

There is some evidence that DHA‐P is better tolerated than AS+MQ. Cental nervous system (CNS) related adverse events (at least one of sleep disturbance, dizziness, or anxiety) were reported as more common with AS+MQ in five out of the nine trials. Five trials also report significantly more nausea and vomiting with AS+MQ and two trials report more palpitations and dyspnoea. Abdominal pain and diarrhoea were reported as significantly more common with DHA‐P in one trial each. For a summary of adverse event findings see Appendix 4.

Early vomiting

Seven trials report some measure of early vomiting (vomiting related to drug administration) and no difference was shown in any trial (seven trials, 2473 participants, Analysis 1.11).

1.11 — Comparison 1 Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine, Outcome 11 Early vomiting.

Comparison 2. DHA‐P versus artemether‐lumefantrine (six doses)

We found six trials (four in Africa, one in Asia and one in Oceania) which assessed this comparison. Allocation concealment was assessed as low risk of bias in four trials (Kamya 2006 UGA; Ratcliff 2005 IDN; Yeka 2007 UGA; Zongo 2007 BFA). Laboratory staff were blinded to treatment allocation in five out of six trials.

Total failure

PCR adjusted treatment failure with DHA‐P was below 5% in four out of six studies and with AL6 in two out of six studies. Of note, one trial from Africa (Kamya 2006 UGA) found PCR adjusted failure to be > 10% with both combinations.

In trials from Africa DHA‐P performed significantly better than AL6 at day 42 (three trials, 1136 participants: PCR unadjusted Heterogeneity: Chi² P < 0.0001, I² = 91%, Analysis 2.1; PCR adjusted RR 0.39, 95% CI 0.24 to 0.64, Analysis 2.2). Although there is substantial heterogeneity among PCR unadjusted results the direction of effect is consistently in favour of DHA‐P.

2.1 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 1 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

2.2 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 2 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

In the one trial from Asia both drugs performed well with a non significant trend towards reduced re‐infections with DHA‐P (one trial, 356 participants, Analysis 2.1; Analysis 2.2; Analysis 2.3; Analysis 2.4).

2.3 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 3 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

2.4 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 4 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

In Oceania Karunajeewa 2007 PNG showed a reduction in PCR adjusted treatment failure at day 28 with AL6 but this effect was no longer significant at day 42 (one trial, 356 participants, Analysis 2.1; Analysis 2.2; Analysis 2.3; Analysis 2.4).

P. vivax

Participants treated with DHA‐P had significantly fewer episodes of P. vivax parasitaemia during 42 days follow up (four trials, 1442 participants: RR 0.32, 95% CI 0.24 to 0.43, Analysis 2.5). Of these four trials only one (Ratcliff 2005 IDN) included participants with P. vivax co‐infection at baseline.

2.5 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 5 P. *vivax* parasitaemia.

Gametocytes

Four trials reported the development of gametocytes in those negative at baseline and the results were highly heterogenous and could not be pooled (four trials, 1203 participants, heterogeneity: Chi² P = 0.006, I² = 76%, Analysis 2.6). This heterogeneity is consistent with the performance of the two drugs for total failure. In the two trials from Uganda (Kamya 2006 UGA and Yeka 2007 UGA) DHA‐P had significantly fewer treatment failures and was also significantly better at reducing gametocyte development. In trials with no difference for treatment failure (Zongo 2007 BFA and Mens 2007 KEN) there was also no difference in gametocyte development. Karunajeewa 2007 PNG and Ratcliff 2005 IDN report no differences in gametocyte carriage between groups but did not give figures.

2.6 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 6 Gametocyte development (in those negative at baseline).

Anaemia

Four trials report changes in haemoglobin from baseline to the last day of follow up (day 28 or 42). There is a non significant trend towards a benefit with DHA‐P but this is unlikely to be of clinical significance (four trials, 1356 participants, Analysis 2.7). In addition Karunajeewa 2007 PNG reports that haemoglobin remained similar in all groups (no figures given).

2.7 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 7 Anaemia.

Adverse events

No significant difference has been shown in the frequency of serious adverse events (five trials, 2110 participants, Analysis 2.8).

2.8 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 8 Serious adverse events (including deaths).

Kamya 2006 UGA and Karunajeewa 2007 PNG report no differences between groups (two trials, 671 participants). Ratcliff 2005 IDN reports more diarrhoea (P = 0.003) with DHA‐P (774 participants). Mens 2007 KEN reports more weakness (P = 0.035) with AL6 (146 participants). Yeka 2007 UGA reports more abdominal pain (P = 0.05) with AL6 (414 participants). Zongo 2007 BFA reports more abdominal pain (P < 0.05) and headache (P < 0.05) with AL6 (375 participants). For a summary of adverse event findings see Appendix 4.

Early vomiting

No difference has been shown in the frequency of drug related vomiting (two trials,1147 participants, Analysis 2.9).

2.9 — Comparison 2 Dihydroartemisinin‐piperaquine vs Artemether‐lumefantrine, Outcome 9 Early vomiting.

Comparison 3. DHA‐P versus artesunate plus amodiaquine

We found two trials (one in Africa and one in Asia) which assessed this comparison. Allocation concealment was assessed as low risk of bias in one trial (Hasugian 2005 IDN) and unclear in the other. In both trials laboratory staff were blinded to treatment allocation, but other staff and participants were unblinded.

Total failure

PCR adjusted treatment failure with DHA‐P was below 5% in both trials, and below 10% with AS+AQ.

DHA‐P performed significantly better than AS+AQ at day 28 (two trials, 679 participants: PCR unadjusted RR 0.53, 95% CI 0.35 to 0.81, Analysis 3.1; PCR adjusted RR 0.47, 95% CI 0.23 to 0.94, Analysis 3.2). The one trial that reports outcomes at day 42 (Hasugian 2005 IDN) had high losses to follow up (> 20%) at this time point (Analysis 3.3; Analysis 3.4).

3.1 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

3.2 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 2 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

3.3 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 3 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

3.4 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 4 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

P. vivax

Hasugian 2005 IDN reports significantly fewer episodes of P. vivax parasitaemia with DHA‐P by day 42 (one trial, 170 participants: RR 0.25, 95% CI 0.09 to 0.74, Analysis 3.5).

3.5 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 5 P. *vivax* parasitaemia.

Gametocytes

Both trials report no significant differences in gametocyte carriage during follow up (figures not reported).

Anaemia

Hasugian 2005 IDN found that the prevalence of anaemia at day seven (P = 0.02) and 28 (P = 0.006) was significantly higher with AS+AQ (authors own figures); in this trial recurrence of parasitaemia with both P. falciparum and P. vivax was higher in the AS+AQ group. Karema 2004 RWA found no significant difference in PCV between groups at days 0 or 14.

Adverse events

Hasugian 2005 IDN reports three serious adverse events with AS+AQ (two patients with recurrent vomiting on day three, one patient with bilateral cerebellar signs) (one trial, 334 participants, Analysis 3.6). Karema 2004 RWA does not comment on serious adverse events.

3.6 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 6 Serious adverse events (including deaths).

Hasugian 2005 IDN reports more nausea (P = 0.004), vomiting (P = 0.02), and anorexia (P = 0.007) with AS+AQ (334 participants). Karema 2004 RWA reports more vomiting (P = 0.007), anorexia (P = 0.005) and fatigue (P = 0.001) with AS+AQ (504 participants). For a summary of adverse event findings see Appendix 4.

Early vomiting

Hasugian 2005 IDN found no significant difference in the number of participants who vomited at least one dose of medication (one trial, 334 participants, Analysis 3.7).

3.7 — Comparison 3 Dihydroartemisinin‐piperaquine vs Artesunate plus amodiaquine, Outcome 7 Early vomiting.

Comparison 4. DHA‐P versus artesunate plus sulfadoxine‐pyrimethamine

We found one trial (from Oceania) which assessed this comparison. No attempt to conceal allocation was described. Laboratory staff were blinded to treatment allocation.

Total failure

At day 42 PCR adjusted treatment failure was > 10% in both groups.

There were no significant differences in treatment failure between the two arms (one trial, 215 participants, Analysis 4.1; Analysis 4.2; Analysis 4.3; Analysis 4.4)

4.1 — Comparison 4 Dihydroartemisinin‐piperaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

4.2 — Comparison 4 Dihydroartemisinin‐piperaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

4.3 — Comparison 4 Dihydroartemisinin‐piperaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 3 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

4.4 — Comparison 4 Dihydroartemisinin‐piperaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 4 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

P. vivax

Compared to AS+SP, DHA‐P significantly reduced the incidence of P. vivax parasitaemia by day 42 in participants treated for P. falciparum mono‐infection at baseline (one trial, 194 participants: RR 0.45, 95% CI 0.32 to 0.65, Analysis 4.5), or P. vivax ± P. falciparum at baseline (one trial, 75 participants: RR 0.46, 95% CI 0.27 to 0.79, Analysis 4.5).

4.5 — Comparison 4 Dihydroartemisinin‐piperaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 5 P. *vivax* parasitaemia by day 42.

Gametocytes

No significant differences in gametocyte carriage during follow up are reported (figures not reported).

Anaemia

Haemoglobin levels were reported to remain similar in both groups throughout follow up (figures not reported).

Adverse events

Monitoring for adverse events was undertaken but no differences between the groups were reported (see Appendix 4).

Early vomiting

Not reported.

Comparison 5. DHA‐P versus amodiaquine plus sulfadoxine‐pyrimethamine

We found two trials (both in Africa) which assessed this comparison. Allocation concealment was assessed as low risk of bias in one trial (Zongo 2007 BFA) and unclear in the other. Karema 2004 RWA blinded laboratory staff to treatment allocation. No other blinding is described.

Total failure

PCR adjusted treatment failure with DHA‐P was below 5% in both trials. In Rwanda, PCR adjusted treatment failure with AQ+SP was above 10%.

DHA‐P performed significantly better than AQ+SP at 28 days (two trials, 848 participants: PCR unadjusted RR 0.37, 95% CI 0.25 to 0.55, Analysis 5.1; PCR adjusted RR 0.30, 95% CI 0.17 to 0.54, Analysis 5.2). Zongo 2007 BFA did not show a difference at day 42 with both drugs performing well at this site (one trial, 341 participants, Analysis 5.3; Analysis 5.4).

5.1 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

5.2 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

5.3 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 3 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

5.4 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 4 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

P. vivax

Not reported.

Gametocytes

Zongo 2007 BFA found no difference in the development of gametocytaemia in participants who did not have detectable gametocytes at baseline (one trial, 367 participants, Analysis 5.5). Karema 2004 RWA reported no significant difference in gametocyte carriage during follow up but figures were not reported (one trial, 510 participants).

5.5 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 5 Gametocyte development.

Anaemia

Zongo 2007 BFA found no significant difference in haemoglobin at baseline or at day 42 (1 trial, 371 participants, Analysis 5.6). Karema 2004 RWA found that the packed cell volume (PCV) increased from baseline to day 14 in both groups, but at day 14 it was significantly lower with DHA‐P (one trial, 510 participants: MD ‐1.10, 95% CI ‐1.73 to ‐0.47, Analysis 5.6). This difference is unlikely to be of clinical significance.

5.6 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 6 Anaemia.

Adverse events

Zongo 2007 BFA reports that there were no serious adverse events (one trial, 371 participants). Karema 2004 RWA does not comment on serious adverse events.

Zongo 2007 BFA reports more abdominal pain (P < 0.05) and pruritis (P < 0.05) with AQ+SP (371 participants). Karema 2004 RWA reports more vomiting (P = 0.007), anorexia (P = 0.005), and fatigue (P = 0.001) with AQ+SP (510 participants). For a summary of adverse event findings see Appendix 4.

Early vomiting

Zongo 2007 BFA reports on vomiting medication on day 0 (as an exclusion criteria not an outcome) and there was no difference between groups (one trial, 383 participants, Analysis 5.7).

5.7 — Comparison 5 Dihydroartemisinin‐piperaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 7 Early vomiting.

Question 2. How does artesunate mefloquine (AS+MQ) perform?

Dosing concerns

AS+MQ has traditionally been administered using 15 mg/kg mefloquine on day one and 10 mg/kg on day two. A new fixed‐dose combination of AS+MQ is now available where mefloquine is given as a once daily dose of 8 mg/kg. One trial (Ashley 2005 THA) has directly compared these two regimens and found no significant difference (one trial, 423 participants, Analysis 16.1; Analysis 16.2). In addition five trials used loose tablets to deliver a once daily dose of mefloquine of 8 mg/kg in combination with artesunate. In all of these trials the proportion of treatment failures with the new regimen was below 10% and in three trials below 5% (Analysis 17.1; Analysis 17.2)

16.1 — Comparison 16 Artesunate Mefloquine dose analysis: FDC versus split dose regimen, Outcome 1 Total Failure Day 63 PCR unadjusted.

16.2 — Comparison 16 Artesunate Mefloquine dose analysis: FDC versus split dose regimen, Outcome 2 Total Failure Day 63 PCR adjusted.

17.1 — Comparison 17 Artesunate plus mefloquine dose analysis (versus Dihydroartemisinin‐piperaquine), Outcome 1 Total Failure Day 63 PCR adjusted.

17.2 — Comparison 17 Artesunate plus mefloquine dose analysis (versus Dihydroartemisinin‐piperaquine), Outcome 2 Total Failure Day 28 PCR adjusted.

Comparison 6. AS+MQ versus artemether‐lumefantrine (six doses)

We found eight trials (six in Asia and two in Africa) which assessed this comparison. Allocation concealment was assessed as low risk of bias in two trials (Mayxay 2003 LAO; Sagara 2005b MLI). Only one trial blinded microscopists to treatment allocation.

Total failure

In all eight trials both combinations performed well with PCR adjusted treatment failures below 5%.

In Asia, AS+MQ reduced overall treatment failure by day 42 compared to AL6 (four trials, 1000 participants: PCR unadjusted RR 0.53, 95% CI 0.29 to 0.94, Analysis 6.1). For PCR adjusted treatment failure there was substantial heterogeneity (four trials, 904 participants: heterogeneity Chi² P = 0.04, I² = 64%, Analysis 6.2), which related to one trial (Hutagalung 2002 THA). This trial was unusual in that P. vivax was very common during follow up and significantly more common following treatment with AL6. P. vivax was treated with chloroquine and participants continued in follow up. Therefore significantly more participants in the AL6 group received additional antimalarials which may have affected the result. Sensitivity analysis removing this trial shifts the result significantly in favour of AS+MQ.

6.1 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 1 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

6.2 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 2 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

There were no significant differences in PCR adjusted treatment failure at day 28 (five trials, 1479 participants, Analysis 6.4). One trial from Africa (Sagara 2005b MLI) did find a significant reduction in re‐infections with AS+MQ but this was not repeated elsewhere (Analysis 6.3).

6.4 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 4 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

6.3 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 3 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

P. vivax

AS+MQ performed significantly better than AL6 at reducing the incidence of P. vivax during 42 days of follow up (four trials, 1003 participants: RR 0.30, 95% CI 0.21 to 0.41, Analysis 6.5).

6.5 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 5 P. *vivax* parasitaemia.

Gametocytes

There is no evidence of an advantage with either drug at reducing gametocytaemia. There was no significant difference in gametocyte development in those negative at baseline (three trials, 883 participants, Analysis 6.6). Gametocyte carriage was generally low in the three trials which report it, with a statistically significant reduction in gametocyte carriage with AS+MQ on day seven, but not day three or 14 (three trials, 636 participants: Gametocyte carriage day seven RR 0.35, 95% CI 0.14 to 0.85, Analysis 6.7). Sagara 2005b MLI reports no differences between groups (no figures given).

6.6 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 6 Gametocyte development (in those negative at baseline).

6.7 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 7 Gametocyte carriage.

Anaemia

Six trials report some measure of haematological recovery. Hutagalung 2002 THA found a greater decrease in haematocrit at day seven with AS+MQ (9.3% AS+MQ versus 6.7% AL6, P = 0.02; authors own figures). None of the remaining five trials report a significant difference (see Appendix 5).

Adverse events

No difference has been shown in the frequency of serious adverse events (seven trials, 1773 participants, Analysis 6.8).

6.8 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 8 Serious adverse events (including deaths).

Three trials report significantly more CNS symptoms with AS+MQ (dizziness, headache, confusion, or sleep disturbance) and one reports more with AL6. Gastrointestinal (GI) symptoms (nausea, vomiting, abdominal pain, or anorexia) were significantly more common with AS+MQ in four trials. For a summary of adverse events see Appendix 4.

Early vomiting

No difference has been shown in the frequency of early vomiting (six trials, 1479 participants, Analysis 6.9).

6.9 — Comparison 6 Artesunate plus mefloquine vs Artemether‐lumefantrine, Outcome 9 Early vomiting.

Comparison 7. AS+MQ versus artesunate plus amodiaquine

We only found one trial in Africa (Faye 2003 SEN) which assessed this comparison. Allocation concealment and blinding were not described.

Total failure

In the 28 days of this trial, treatment failure was very low in both groups. It is therefore not possible to draw conclusions on the benefits of either drug. There were no significant differences in PCR unadjusted failure (one trial, 493 participants, Analysis 7.1) and no episodes of PCR confirmed recrudescence.

7.1 — Comparison 7 Artesunate plus mefloquine vs Artesunate plus amodiaquine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

P. vivax

Not reported.

Gametocytes

Gametocyte carriage was very low in both groups. Gametocytes were only detectable in three participants in the AS+MQ group on day three. At baseline, day seven and day 14 gametocytes were undetectable in all participants.

Anaemia

Twenty‐five percent of participants had haemoglobin measured on days 0 and 14 and no significant differences are reported.

Adverse events

In this trial there were no serious adverse events (one trial, 505 participants) and no differences between groups reported (see Appendix 4).

Early vomiting

Not reported.

Comparison n/a. AS+MQ versus artesunate plus sulfadoxine‐pyrimethamine

We did not find any trials which assessed this comparison.

Comparison 8. AS+MQ versus amodiaquine plus sulfadoxine‐pyrimethamine

We only found one trial in Africa (Faye 2003 SEN) which assessed this comparison. Allocation concealment and blinding were not described.

Total failure

In the 28 days of this trial, treatment failure was very low in both groups. It is therefore not possible to draw conclusions on the benefits of either drug. There were no differences in PCR unadjusted failure (one trial, 300 participants, Analysis 8.1) and there were no episodes of PCR confirmed recrudescence.

8.1 — Comparison 8 Artesunate plus mefloquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

P. vivax

Not reported.

Gametocytes

Detectable gametocytaemia was significantly less common with AS+MQ at days three and seven (Gametocyte carriage day three: RR 0.21, 95% CI 0.06 to 0.70; Gametocyte carriage day seven: RR 0.03, 95% CI 0.00 to 0.47, Analysis 8.3). At day 14 gametocytes were undetectable in all participants.

8.3 — Comparison 8 Artesunate plus mefloquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 3 Gametocyte carriage.

Anaemia

Twenty five percent of participants had haemoglobin measured on days 0 and 14 and no significant differences were reported.

Adverse events

In this trial there were no serious adverse events in either group (one trial, 306 participants) and no differences between groups reported (see Appendix 4).

Early vomiting

Not reported.

Question 3. How does artemether‐lumefantrine (6 doses) perform?

Dosing concerns

The six‐dose regimen of AL6 has been shown to be superior to the four‐dose regimen (Vugt 1999; Omari 2006). In this review we have only included the six‐dose regimen.

Comparison 9. AL6 versus artesunate plus amodiaquine

We found twelve trials (all in Africa) which assessed this comparison. Three of these trials were excluded after sensitivity analysis due to baseline differences which had the potential to bias the result in favour of AL6 (Analysis 9.9; Analysis 9.10). Of the remaining nine trials allocation concealment was assessed as low risk of bias in five trials (Adjei 2006 GHA; Bukirwa 2005 UGA; Dorsey 2006 UGA; Kobbe 2007 GHA; Mutabingwa 2004 TZA) and laboratory staff were blinded to treatment allocation in four trials.

Total failure

PCR adjusted treatment failure was below 5% for both AL6 and AS+AQ in six out of eight trials. In two more recent trials (both from Ghana), PCR adjusted treatment failure for both arms was above 5% and for AL6 above 10% (Analysis 9.2).

9.2 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 2 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

No difference has been shown in PCR adjusted total failure at day 28, either within individual trials or after pooling (eight trials, 1729 participants, Analysis 9.2). There is substantial heterogeneity in PCR unadjusted failure (nine trials, 3021 participants: heterogeneity Chi² P < 0.0001, I² = 76%, Analysis 9.1). Subgroup analysis seems to suggest regional differences, with studies from East Africa showing benefit with AL6 and recent studies from West Africa favouring AS+AQ (Analysis 9.1). However, substantial heterogeneity remains, and further subgroup analysis by trial characteristics and transmission intensity did not expand the interpretation of this heterogeneity.

9.1 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

P. vivax

One trial (Dorsey 2006 UGA) reported on P. vivax but there were too few patients to draw a conclusion (AL6: 8/202 at baseline and 3/202 during follow up, AS+AQ: No vivax at any time point).

Gametocytes

Bukirwa 2006 found that AL6 significantly reduced the development of gametocytaemia in patients who did not have detectable gametocytes at baseline (one trial, 305 participants: RR 0.34, 95% CI 0.15 to 0.74, Analysis 9.3). Three trials reporting gametocyte carriage over 14 days of follow up do not show a clear advantage with either combination (three trials, 1078 participants, Analysis 9.4).

9.3 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 3 Gametocyte development.

9.4 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 4 Gametocyte carriage.

Anaemia

Four studies reported some measure of haematological recovery from baseline to day 28 and did not show a difference between the two combinations (four trials, 2356 participants, Analysis 9.5). Guthmann 2004 AGO reported the proportion of participants who were anaemic (Hb < 11 g/dl) at day 0 and 28 and did not show a difference (one trial, 123 participants, Analysis 9.6). Three trials (Dorsey 2006 UGA; Faye 2003 SEN; Mutabingwa 2004 TZA) also reported measures of anaemia at day 14 and did not show a difference.

9.5 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 5 Anaemia.

9.6 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 6 Proportion anaemic (Haemoglobin < 11 g/dl).

Adverse events

No difference has been shown in the frequency of serious adverse events (six trials, 2749 participants, Analysis 9.7).

9.7 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 7 Serious adverse events (including deaths).

No important differences in adverse events were reported between groups. For a summary of adverse events see Appendix 4.

Early vomiting

No difference has been shown in the frequency of early vomiting (five trials, 1097 participants, Analysis 9.8).

9.8 — Comparison 9 Artemether‐lumefantrine vs Artesunate plus amodiaquine, Outcome 8 Early vomiting.

Comparison 10. AL6 versus artesunate plus sulfadoxine‐pyrimethamine

We found four trials (three from Africa and one from Oceania) which assessed this comparison. Two of these trials were excluded from the primary analysis due to baseline differences between the groups (Analysis 10.6; Analysis 10.7). Allocation concealment was judged to be at high risk of bias in the two remaining trials. Laboratory staff were blinded to treatment allocation in one trial.

10.6 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 6 Sensitivity analysis Total Failure Day 28 PCR unadjusted.

10.7 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 7 Sensitivity analysis: Total Failure Day 28 PCR adjusted.

Total failure

In Oceania, Karunajeewa 2007 PNG found no difference in PCR unadjusted failure (one trial, 217 participants, Analysis 10.1; Analysis 10.3), but did show a significant reduction in PCR adjusted treatment failure with AL6 at both day 28 and day 42 (one trial, 217 participants: Day 42 RR 0.33, 95% CI 0.13 to 0.86, Analysis 10.2; Day 28 RR 0.28, 95% CI 0.08 to 0.97, Analysis 10.4). PCR adjusted treatment failure with AS+SP was > 20% at day 42.

10.1 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 42 PCR unadjusted.

10.3 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 3 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

10.2 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum*) Day 42 PCR adjusted.

10.4 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 4 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

In Africa, Mukhtar 2005 SDN found no difference between the two groups (one trial, 157 participants, Analysis 10.3, Analysis 10.4).

P. vivax

Karunajeewa 2007 PNG found no differences in the incidence of P. vivax parasitaemia by day 42 in participants treated for P. falciparum mono‐infection at baseline (one trial, 196 participants), or those treated for P. vivax at baseline (one trial, 72 participants, Analysis 10.5)

10.5 — Comparison 10 Artemether‐lumefantrine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 5 P. *vivax* parasitaemia.

Gametocytes

Karunajeewa 2007 PNG reports no differences in gametocyte carriage between the two groups during follow up (figures not reported).

Anaemia

Karunajeewa 2007 PNG reports no differences in mean haemoglobin during follow up (figures not reported).

Adverse events

Two trials report on adverse events and no differences are noted between the two groups (Karunajeewa 2007 PNG; Van den Broek 2004 ZAR). For a summary of adverse events see Appendix 4.

Early vomiting

Not reported.

Comparison 11. AL6 versus amodiaquine plus sulfadoxine‐pyrimethamine

We found seven trials (all in Africa) which assessed this comparison. One trial was excluded from the primary analysis due to baseline differences between groups. Of the remaining trials allocation concealment was assessed as low risk of bias in two trials (Dorsey 2006 UGA; Zongo 2007 BFA) and laboratory staff were blinded to treatment allocation in four trials.

Total failure

PCR adjusted treatment failure with AL6 was below 5% in all six trials. The performance of AQ+SP was much more variable.

In East Africa, where treatment failure with AQ+SP was high, AL6 performed markedly better at day 28 (three trials, 1646 participants: PCR unadjusted RR 0.35, 95% CI 0.30 to 0.41, Analysis 11.1; PCR adjusted RR 0.12, 95% CI 0.06 to 0.24, Analysis 11.2).

11.1 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum)* Day 28 PCR unadjusted.

11.2 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum)* Day 28 PCR adjusted.

In contrast, in West Africa, where AQ+SP performed much better, there were fewer PCR unadjusted treatment failures with AQ+SP at both day 28 (three trials, 1130 participants: PCR unadjusted RR 2.88, 95% CI 1.86 to 4.47, Analysis 11.1) and day 42 (one trial, 345 participants: PCR unadjusted RR 2.64, 95% CI 1.66 to 4.21, Analysis 11.3). There were no significant differences between the two combinations after PCR adjustment (Analysis 11.2; Analysis 11.4).

11.3 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 3 Total Failure (P. *falciparum)* Day 42 PCR unadjusted.

11.4 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 4 Total Failure (P. *falciparum)* Day 42 PCR adjusted.

P. vivax

Only one trial (Dorsey 2006 UGA) reported on P. vivax and there were too few patients to draw a conclusion (AL6 8/202 at baseline and 3/202 during follow up, AQ+SP 4/253 at baseline and 0 during follow up).

Gametocytes

The prevalence of gametocyte carriage was significantly lower with AL6 at day three (three trials, 1331 participants: RR 0.43, 95% CI 0.25 to 0.75, Analysis 11.5) and day seven (four trials,1538 participants: RR 0.32, 95% CI 0.18 to 0.54, Analysis 11.5). Zongo 2007 BFA found no significant difference in the development of gametocytaemia in participants without detectable gametocytes at baseline (one trial, 371 participants, Analysis 11.6).

11.5 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 5 Gametocyte carriage.

11.6 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 6 Gametocyte development (in those negative at baseline).

Anaemia

Zongo 2005 BFA reports change in haemoglobin from baseline to day 28; Zongo 2007 BFA reports mean haemoglobin at baseline and day 42. Neither of these trials showed a clinically significant difference (two trials, 893 participants, Analysis 11.7). Four other trials assessed haematological recovery at shorter time points and did not detect a difference (Dorsey 2006 UGA; Fanello 2004 RWA; Faye 2003 SEN; Mutabingwa 2004 TZA).

11.7 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 7 Anaemia.

Adverse events

No difference has been shown in the frequency of serious adverse events (five trials, 2684 participants, Analysis 11.8).

11.8 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 8 Serious adverse events (including deaths).

Dorsey 2006 UGA reports more anorexia (P < 0.05) and weakness (P < 0.05) with AQ+SP (455 participants). Two trials report a significant increase in pruritis (P < 0.05, P < 0.0001) with AQ+SP. No further differences are noted. For a summary of adverse events see Appendix 4.

Early vomiting

Two trials report on the number of participants excluded for persistent vomiting on day 0. There were no differences between groups (two trials, 893 participants, Analysis 11.9).

11.9 — Comparison 11 Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 9 Early vomiting.

Question 4. How does artesunate plus amodiaquine perform?

Comparison 12. AS+AQ versus artesunate plus sulfadoxine‐pyrimethamine

We found seven trials (all in Africa) which assessed this comparison. Allocation concealment was judged as low risk of bias in only one trial (Bonnet 2004 GIN) and unclear in four. Laboratory staff were blinded to treatment allocation in two trials.

Total failure

PCR adjusted treatment failures with AS+AQ were < 10% in all seven trials, and with AS+SP in six out of seven trials.

Overall the number of PCR adjusted failures was low with no significant difference between groups (seven trials, 1419 participants, Analysis 12.2). There was substantial heterogeneity in PCR unadjusted failure rates between trials (seven trials, 1419 participants: heterogeneity: Chi² P < 0.00001, I² = 88%, Analysis 12.1). We attempted to investigate this heterogeneity with subgroup analysis on geographical region, allocation concealment, drug dose, stated resistance pattern, and age of participants, with no clear findings.

12.2 — Comparison 12 Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

12.1 — Comparison 12 Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

P. vivax

Not reported.

Gametocytes

We were able to combine the results of three trials reporting gametocyte carriage on days three, seven and 14 and no difference was shown at any time point (three trials, 532 participants, Analysis 12.3). The remaining four trials report that there were no differences in carriage between groups but do not give figures.

12.3 — Comparison 12 Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 3 Gametocyte carriage.

Anaemia

Five trials report that levels of anaemia improved following treatment in both groups. Three of these trials did not give figures (Djimde 2004 MLI; Swarthout 2004 ZAR; Van den Broek 2004 ZAR). Two trials report the proportion of patients with anaemia at baseline and day 28. The proportion improved in both groups with no significant differences between the two treatments (two trials, 452 participants, Analysis 12.4).

12.4 — Comparison 12 Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 4 Proportion of participants with anaemia.

Adverse events

No difference has been shown in the frequency of serious adverse events (four trials, 1108 participants, Analysis 12.5).

12.5 — Comparison 12 Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine, Outcome 5 Serious adverse events (including deaths).

Five trials reported on adverse events and no significant differences between treatments were noted. One trial (Djimde 2004 MLI) performed haematological and biochemical tests on days 7, 14, and 28 and no significant abnormalities were noted. For a summary of adverse events see Appendix 4.

Early vomiting

Not reported.

Comparison 13. AS+AQ versus amodiaquine plus sulfadoxine‐pyrimethamine

We found eight trials which assessed this comparison (all in Africa). Allocation concealment was assessed as low risk of bias in four trials (Dorsey 2006 UGA; Menard 2006 MDG; Mutabingwa 2004 TZA; Staedke 2003 UGA) and unclear in two. Laboratory staff were unaware of treatment allocation in seven trials.

Total failure

The efficacy of both drugs in these trials was highly variable.

A subgroup analysis demonstrates that it is in East Africa that AQ+SP is failing as a first‐line therapy. Heterogeneity is high, limiting meaningful pooling of data, but trials from East Africa tend to favour AS+AQ (five trials, 3317 participants, PCR unadjusted heterogeneity: Chi² P < 0.0001, I² = 91%, Analysis 13.1; three trials, 1515 participants, PCR adjusted heterogeneity: Chi² P = 0.03, I² = 73%, Analysis 13.2). AQ+SP performed well in Senegal in 2003, Mali in 2006 and Madagascar in 2006. We further investigated this heterogeneity with subgroup analysis on allocation concealment, drug dose, stated resistance pattern, and age of participants, with no clear findings.

13.1 — Comparison 13 Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 1 Total Failure (P. *falciparum*) Day 28 PCR unadjusted.

13.2 — Comparison 13 Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 2 Total Failure (P. *falciparum*) Day 28 PCR adjusted.

P. vivax

Not reported.

Gametocytes

AS+AQ significantly reduced the development of gametocytes in those negative at baseline (two trials, 1354 participants: RR 0.67, 95% CI 0.54 to 0.82, Analysis 13.3). Six trials measured gametocyte carriage during follow up. Three of these reported that there were no differences but did not give figures. Of the three trials which gave figures, only one (Faye 2003 SEN) found that AS+AQ significantly reduced carriage rates at days three and seven (Analysis 13.4).

13.3 — Comparison 13 Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 3 Gametocyte development.

13.4 — Comparison 13 Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 4 Gametocyte carriage.

Anaemia

All eight trials reported some measure of haematological recovery. No individual trial has reported a clinically important difference at day 14 or 28 (see Appendix 5).

Adverse events

No difference has been shown in the frequency of serious adverse events (seven trials, 4200 participants, Analysis 13.6).

13.6 — Comparison 13 Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine, Outcome 6 Serious adverse events (including deaths).

Dorsey 2006 UGA reports more anorexia (P < 0.05) and weakness (P < 0.05) with AQ+SP (485 participants). No differences are noted in any other trial. Four trials also undertook some biochemical monitoring and no important differences are noted. For a summary of adverse events see Appendix 4.

Early vomiting

Not reported.

Discussion

Summary of main results

Efficacy (as measured by total failure)

The WHO has set two standards for antimalarial drugs:

that a total failure rate (adjusted for new infections) of > 10% should trigger a change of first‐line drug policy; and
that a new drug being adopted as policy should have total failure rates (adjusted for new infections) of < 5%.

This review has demonstrated that:

In head to head trials the newest ACT, dihydroartemisinin‐piperaquine, achieved the standard of < 5% total failure in 15 out of the 17 studies it was involved in. DHA‐P appears to be at least as effective as AS+MQ in Asia (eight trials) providing a valuable alternative to current therapy. In clinical trials in Africa, DHA‐P may be more effective than the current widely used options AL6 (four trials) and AS+AQ (one trial), although these two drugs continue to perform well in many areas (Figure 3; Figure 4).
AS+MQ has performed well in trials from Asia and South America, with failure rates consistently low, but has been little studied in the African context (Figure 5; Figure 6).
AL6 and AS+AQ performed well in almost all studies they were involved in but Kamya 2006 UGA found failure rates in excess of 10% with AL6 and Yeka 2004 UGA reported > 10% failure with AS+AQ (Figure 7; Figure 8; Figure 9; Figure 10).
There is very little good quality evidence available comparing AS+SP to DHA‐P, AS+MQ or AL6 but it has performed well in head to head trials with AS+AQ.
The performance of the non‐ACT AQ+SP (which is only recommended as an interim measure by the WHO), was inadequate for first‐line use in several countries from East Africa. It was, however, still performing well in Senegal in 2003 (Faye 2003 SEN), Madagascar in 2006 (Menard 2006 MDG), and Burkina Faso in 2005 (Zongo 2005 BFA).

How does Dihydroartemisinin‐piperaquine perform? Summary of primary outcome: Effectiveness: Total Failure (*P. falciparum*) PCR adjusted.

**Olliaro‐Vaillant plot.** Day 28 PCR adjusted treatment failure data for trials of DHA‐P against all comparators are presented in this plot.

The horizontal red line represents the WHO standard of 10% treatment failure (PCR corrected). Plots below this line represent trials where DHA‐P performed to this standard.

The vertical blue line represents no difference between the two drugs. Plots to the right of this line represent trials where DHA‐P performed better than the comparator drug, and plots to the left represent trials where the comparator drug performed better than DHA‐P.

How does Artesunate plus mefloquine perform? Summary of primary outcome: Effectiveness: Total Failure (*P. falciparum*) PCR adjusted.

**Olliaro‐Vaillant plot.** Day 28 PCR adjusted treatment failure data for trials of AS+MQ against all comparators are presented in this plot.

The horizontal red line represents the WHO standard of 10% treatment failure (PCR corrected). Plots below this line represent trials where AS+MQ performed to this standard.

The vertical blue line represents no difference between the two drugs. Plots to the right of this line represent trials where AS+MQ performed better than the comparator drug, and plots to the left represent trials where the comparator drug performed better than AS+MQ.

How does Artemether‐lumefantrine perform? Summary of primary outcome: Effectiveness: Total Failure (P. *falciparum*) Day PCR adjusted.

**Olliaro‐Vaillant plot.** Day 28 PCR adjusted treatment failure data for trials of AL6 against all comparators are presented in this plot.

The horizontal red line represents the WHO standard of 10% treatment failure (PCR corrected). Plots below this line represent trials where AL6 performed to this standard.

The vertical blue line represents no difference between the two drugs. Plots to the right of this line represent trials where AL6 performed better than the comparator drug, and plots to the left represent trials where the comparator drug performed better than AL6.

How does Artesunate plus amodiaquine perform? Summary of primary outcome: Effectiveness: Total Failure (*P. falciparum*) PCR adjusted.

**Olliaro‐Vaillant plot.** Day 28 PCR adjusted treatment failure data for trials of AS+AQ against all comparators are presented in this plot.

The horizontal red line represents the WHO standard of 10% treatment failure (PCR corrected). Plots below this line represent trials where AS+AQ performed to this standard.

The vertical blue line represents no difference between the two drugs. Plots to the right of this line represent trials where AS+AQ performed better than the comparator drug, and plots to the left represent trials where the comparator drug performed better than AS+AQ.

Efficacy (P. vivax)

The two drugs with long half‐lives (DHA‐P and AS+MQ) have been shown to be superior to AL6 in reducing the incidence of P. vivax following treatment (for either P. falciparum or P. falciparum/P. vivax co‐infections). DHA‐P has also been shown to reduce the incidence of P. vivax compared to AS+AQ. Five trials have compared DHA‐P and AS+MQ and shown no difference.

There could be some public health benefits to using drugs with long half‐lives in this way, to prolong the malaria free period. One trial (Hasugian 2005 IDN) demonstrated a reduced risk of anaemia after treatment with DHA‐P. This is likely to be due to the lower incidence of both P. falciparum re‐infections and P. vivax in this group. As ACTs are ineffective at treating the liver stages of P. vivax, this effect may be lost as follow up continues as the majority of P. vivax will eventually relapse.

Prevention of transmission (as measured by gametocytes)

ACTs may be superior to AQ+SP (the only combination not containing an artemisinin derivative) in their effect on gametocytes. Gametocyte carriage at days three and seven was higher with AQ+SP compared to AS+MQ (one trial, 306 participants, Analysis 8.3) and AL6 (four trials, 1538 participants, Analysis 11.5). Gametocyte development in those negative at baseline was also higher with AQ+SP compared to AS+AQ (two trials, 1354 participants, Analysis 13.3). No difference was shown between AQ+SP and DHA‐P.

Artesunate plus mefloquine seems to be superior to DHA‐P in reducing the carriage of gametocytes and preventing gametocyte development. This effect may be a result of the relatively low artemisinin content of this combination. Pharmokinetic data suggest that dihydroartemisinin and artesunate are broadly bioequivalent (Newton 2002) but at current dosing the total dose of dihydroartemisinin over three days (6 mg/kg) is only half the total dose of artesunate (12 mg/kg).

DHA‐P did perform well against other combinations, and there is currently no evidence that it is inferior to AL6, AS+AQ or AQ+SP in its effect on gametocytes.

It should be noted that there is evidence that even submicroscopic levels of gametocytes (which are present in a significant number of patients after treatment) are capable of transmission (Bousema 2004 KEN).

Haematological recovery

Anaemia is a common complication of malaria. Following successful treatment of the parasite, the level of anaemia should improve gradually over time, provided there is no further re‐infection. This process can be hastened by supplementation with oral iron therapy.

In this review, where measures of haematological recovery were reported, there is no evidence of clinically important differences between the different ACTs.

Harms (as measured by adverse events)

The general lack of standardization in recording and reporting of adverse events unfortunately precludes the use of meta‐analysis to analyse safety data. In addition, very few of the included trials involved adequate blinding to prevent bias in adverse event reporting. Although serious adverse events seem to be uncommon, very few trials undertook the biochemical or haematological monitoring necessary to detect neutropenia or hepatotoxicity which have been previously reported.

DHA‐P seems to have a favourable profile in comparison to the other drugs. In the 17 trials involving DHA‐P, results are inconsistent, but individual trials have shown reduced incidence of vomiting, anorexia, abdominal pain, fatigue, and pruritis compared to AQ+SP, vomiting, anorexia, and fatigue compared to AS+AQ, abdominal pain and headache compared to AL6 and sleep disturbance, dizziness, anxiety, nausea and vomiting compared to AS+MQ.

AS+MQ seems to cause more sleep disturbance and dizziness than DHA‐P and AL6. Overall there are also probably more gastrointestinal symptoms with AS+MQ.

Combinations including amodiaquine do seem to cause more gastrointestinal upset when compared to DHA‐P but there is no convincing evidence of increased vomiting compared to AL6.

No clinically severe alterations in biochemical tests were noted in any of these trials.

AS+MQ tolerability in African children

There has been concern regarding the tolerability of AS+MQ in African children (WHO 2006). This concern was raised by Slutsker 1990 in a trial of mefloquine monotherapy in children aged three months to five years. They found vomiting rates of 16/56 (29%) with a single dose of 25 mg/kg and 26/65 (40%) with15 mg/kg; 13% and 8% were unable to tolerate a second dose respectively. Three important details from this trial should be noted: i) there was no comparison with an alternative therapy, ii) the one‐off dose was higher than in current regimens, and iii) the mean age of children was 13 months which is considerably younger than most trials of mefloquine in Asia.

In this review, we found two head to head trials of AS+MQ in Africa. Both of these studies excluded children aged < one year but vomiting was noted to be more common with AS+MQ in one of these trials (Sagara 2005b MLI). There are, in addition, several published single‐arm or excluded trials of AS+MQ use in Africa (Massougbodji 2002; Agomo 2008; Sagara 2008), but again these do not include the very young children as included in Slutsker 1990. It is therefore not possible with current evidence to say whether this poor tolerance is a consistent finding, whether it is substantially different from other available ACTs or whether the new regime of mefloquine 8 mg/kg/day is better tolerated.

Overall completeness and applicability of evidence

Due to the changing patterns of resistance, summary statistics should be interpreted with caution as the effectiveness of these combinations is likely to vary from place to place, and to change with time.

Evidence is generally lacking on the safety and efficacy of these combinations in very young children (< six months) and in pregnant and lactating women who were excluded from all of the included trials.

In addition to the ACTs presented here, two further combinations (dihydroartemisinin plus naphthoquine and artesunate plus sulfamethoxypyrazine‐pyrimethamine) are beginning to appear in the published literature and the market place, and these will be added to future updates of this review.

Quality of the evidence

The quality of the evidence has been assessed using the GRADE process (Guyatt 2008) and the results presented in the 'Summary of findings tables'. For these tables we asked the following questions:

1) Is dihydroartemsinin‐piperaquine a suitable alternative to the currently recommended ACTs?

There is high quality evidence that DHA‐P is at least as effective (at reducing PCR corrected treatment failure) as AS+MQ in Asia, and AL6 in Africa, and moderate quality evidence that DHA‐P is at least as effective as AS+AQ (Appendix 6).

2) Does amodiaquine plus sulfadoxine‐pyrimethamine remain a valid alternative to ACTs?

The performance of AQ+SP is highly variable and so it is difficult to make general statements on relative effects. There is moderate quality evidence that AQ+SP is inferior to DHA‐P and AL6 in East Africa and very low quality evidence that it is also inferior to AS+AQ (Appendix 6).

3) Does artesunate plus sulfadoxine‐pyrimethamine remain a valid alternative to other ACTs?

There is no good quality evidence comparing AS+SP to DHA‐P, AS+MQ or AL6. In trials comparing AS+SP to AS+AQ both drugs performed well and no clear difference was shown (Appendix 6).

4) Is artesunate plus mefloquine a valid alternative to the currently used ACTs in Africa?

AS+MQ generally performed well in trials in Asia against DHA‐P and AL6 (Appendix 6). The direct evidence from Africa versus AS+AQ and AQ+SP is of low quality (Summary of findings table 7; Summary of findings table 8). The high performance of AS+MQ is likely to be maintained in Africa where resistance to mefloquine is low.

For the comparison artemether‐lumefantrine versus artesunate plus amodiaquine see Appendix 6.

Potential biases in the review process

Data extraction was unblinded. All included trials are published; we were unable to obtain further unpublished data from pharmaceutical companies.

Authors' conclusions

Implications for practice.

All five ACTs performed adequately, to be used as first‐line therapies, in most sites where they were studied, however there are examples of failure rates above 10% with all combinations, emphasizing the need for continued monitoring and evaluation.

There is now a growing weight of evidence available to justify the use of dihydroartemisinin‐piperaquine as a first‐line treatment option for P. falciparum malaria.

There is evidence that the non‐artemisinin combination AQ+SP is failing in parts of East Africa where DHA‐P, AL6, and AS+AQ have been shown to be superior. There is also evidence that ACTs have a superior effect on gametocytes that may be of public health benefit particularly in low transmission settings.

The ACTs appear to be effective in treating the blood stage of P. vivax. There may also be some benefit in using drugs with long half‐lives to delay spontaneous relapses. This prophylactic effect needs to be balanced with the theoretical risk of promoting the development of drug resistance. Additionally, in areas where primaquine is being used to provide a radical cure this effect may not be be of clinical significance.

Evidence of the safety of artemisinins is accumulating. Serious adverse events with these drugs appear to be rare. However, these trials are not powered to detect rare but clinically important events and so it is imperative that active monitoring continues.

Implications for research.

There are several new ACT combinations in development which are likely to become commercially available in the next few years. Policy makers therefore have a greater range of potential products. In these circumstances, improved information on comparative efficacy, adverse events, and tolerability is invaluable for informed decision making.

Many trials are using relatively standardized primary outcomes. A move towards standardized approaches to measuring and reporting secondary outcomes, and adverse events, would greatly improve comparability between trials and meta‐analysis.

In the absence of mefloquine resistance, AS+MQ is likely to be highly effective in African countries but concerns regarding poor tolerability in young infants have restricted its use in this setting. There is in fact little evidence on the use of any of the ACTs in this age group, and head to head randomized trials are necessary to clarify or refute the specific concerns regarding AS+MQ and to provide more general guidance on the choice and use of ACTs in infants.

Further research is needed to clarify the role of specific ACTs in the treatment of P. vivax. It remains unclear as to whether a long acting ACT offers individual or public health benefits compared to standard treatments for radical cure.

The most vulnerable populations (pregnant women and very young infants) were excluded from all trials, and represent a critical gap in current knowledge.

What's new

Date	Event	Description
12 August 2009	Amended	Tables for treatment comparisons, search strategy, primary outcome measures, adverse events, anaemia, and summary of findings moved to appendices.

Question	Analysis	Comparisons
1. How does dihydroartemisinin‐piperaquine perform?	1	vs artesunate plus mefloquine
	2	vs artemether‐lumefantrine (6 doses)
	3	vs artesunate plus amodiaquine
	4	vs artesunate plus sulfadoxine‐pyrimethamine
	5	vs amodiaquine plus sulfadoxine‐pyrimethamine
2. How does artesunate plus mefloquine perform?	1	vs dihydroartemisinin‐piperaquine
	6	vs artemether‐lumefantrine (6 doses)
	7	vs artesunate plus amodiaquine
	‐	vs artesunate plus sulfadoxine‐pyrimethamine
	8	vs amodiaquine plus sulfadoxine‐pyrimethamine
3. How does artemether‐lumefantrine (6 doses) perform?	2	vs dihydroartemisinin‐piperaquine
	6	vs artesunate plus mefloquine
	9	vs artesunate plus amodiaquine
	10	vs artesunate plus sulfadoxine‐pyrimethamine
	11	vs amodiaquine plus sulfadoxine‐pyrimethamine
4. How does artesunate plus amodiaquine perform?	3	vs dihydroartemisinin‐piperaquine
	7	vs artesunate plus mefloquine
	9	vs artemether‐lumefantrine (6 doses)
	12	vs artesunate plus sulfadoxine‐pyrimethamine
	13	vs amodiaquine plus sulfadoxine‐pyrimethamine

Search set	CIDG SR^a	CENTRAL	MEDLINE^b	EMBASE^b	LILACS^b
1	malaria	malaria	malaria	malaria	malaria
2	arte*	arte*	arte*	arte*	arte*
3	dihydroarte*	dihydroarte*	dihydroarte*	dihydroarte*	dihydroarte*
4	amodiaq*	amodiaq*	amodiaq*	amodiaq$	amodiaq$
5	lumefantrine	lumefantrine	lumefantrine	lumefantrine	lumefantrine
6	Coartem*	Coartem*	Coartem*	Coartem$	Coartem$
7	mefloquine	mefloquine	mefloquine	mefloquine	mefloquine
8	2 or 3	2 or 3	2 or 3	2 or 3	2 or 3
9	4 or 5 or 6 or 7	4 or 5 or 6 or 7	4 or 5 or 6 or 7	4 or 5 or 6 or 7	4 or 5 or 6 or 7
10	1 and 8 and 9	1 and 8 and 9	1 and 8 and 9	1 and 8 and 9	1 and 8 and 9
11	—	—	Limit 10 to humans	Limit 10 to human	—

Analysis	Participants	PCR^b‐unadjusted		PCR‐adjusted
Analysis	Participants	Numerator	Denominator	Numerator	Denominator
Primary analysis	Exclusions after enrolment	Excluded^c	Excluded	Excluded	Excluded
	Missing or indeterminate PCR	Included as failures	Included	Excluded	Excluded
	New infections	Included as failures	Included	Excluded	Excluded
Sensitivity analysis 1^d	As 'Primary analysis' except: missing or indeterminate PCR	—	—	Included as failures	Included
Sensitivity analysis 2^e	As 'Sensitivity analysis 1' except: new infections	—	—	Included as successes	Included
Sensitivity analysis 3^f	As 'Sensitivity analysis 2' except: exclusions after enrolment	Included as failures	Included	Included as failures	Included
Sensitivity analysis 4^g	As 'Sensitivity analysis 2' except: exclusions after enrolment	Included as successes	Included	Included as successes	Included

Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine
Study ID	Adverse event monitoring	Blinding	Summary of adverse event findings
Ashley 2003a THA (134 participants)	Inpatient monitoring until day 28 FBC, U&E, LFT on days 0 and 7	Open label	SAE: No serious adverse events observed Biochemical: No evidence of toxicity observed Other: No differences between the groups reported
Ashley 2003b THA (356 participants)	Daily review until parasites cleared then weekly until day 63 A subset of patients in the DHA‐P group had FBC, U&E and LFT on days 0 and 7 and ECG monitoring before and after treatment	Open label	SAE: No serious adverse events observed GI: More abdominal pain reported with DHA‐P (P = 0.025) Nausea, vomiting, and diarrhoea not significantly different CNS: More sleep disturbance with AS+MQ (P = 0.008) Dizziness not significantly different Biochemical: Some minor fluctuations in LFTs CVS: No comment
Ashley 2004 THA (499 participants)	Clinical examination, symptom enquiry, and haematocrit daily until parasites cleared then weekly until day 63	Open label	SAE: 4 serious events with AS+MQ (death, severe anaemia, febrile convulsion, coagulopathy) and 11 with DHA‐P (2 deaths, bacterial sepsis, febrile convulsion, leptospirosis, haematemesis, nephritic syndrome, severe anaemia, respiratory infection, epigastric pain and vomiting). All except the one case of severe vomiting were judged to be unrelated or unlikely to be due to the study treatment GI: More diarrhoea with DHA‐P (P = 0.026); nausea, vomiting, and abdominal pain not significantly different CNS: No significant difference in dizziness or sleep disturbance Other: Urticaria occurred in 1 patient with DHA‐P but none with AS+MQ
Grande 2005 PER (522 participants)	Clinical assessment daily until day 3 then weekly until day 63 FBC, U&E, LFT, and PCV days 0 and 7, PCV days 14 and 63	Open label	SAE: 3 serious drug related events with AS+MQ requiring stopping treatment (encephalopathy, anxiety and arrhythmia, palpitations, and chest pain) GI: More nausea and vomiting with AS+MQ in adults (P = 0.02) but not significantly different in children. Abdominal pain and anorexia not significantly different CNS: More insomnia, dizziness and anxiety with ASMQ in adults (P = < 0.001) and more insomnia and anxiety with AS+MQ in children (P = < 0.001, 0.02). More somnolence with DHA‐P (P = 0.02) Biochemical: No clinically significant abnormal renal or liver test results
Janssens 2003 KHM (464 participants)	Clinical examination and symptom questionnaire days 0, 1, 2, 3. Only adverse events occurring in these 3 days are reported.	Open label	SAE: No serious adverse events observed GI: More nausea, vomiting, and anorexia with AS+MQ, only vomiting was significant (P = 0.03) CNS: More dizziness and sleep disturbance with AS+MQ (P = 0.002, 0.03) CVS: More palpitations with AS+MQ (P = 0.04)
Mayxay 2004 LAO (220 participants)	Daily review until parasites cleared then weekly until day 42	Open label	SAE: One neuropsychiatric reaction in AS+MQ group GI: More nausea and vomiting with AS+MQ (P = < 0.001, 0.02), abdominal pain and diarrhoea not significantly different CNS: More dizziness, sleep disturbance, nightmares, headache and weakness with AS+MQ (P = < 0.001, 0.02, 0.003, 0.001, 0.009) CVS/RS: More palpitations and dyspnoea with AS+MQ (P = 0.002, 0.04)
Smithuis 2004 MMR (652 participants)	Symptom questionnaire at days 0, 1, 2, 3 and 7. Only adverse events occurring in the first 7 days are reported.	Open label	SAE: No serious adverse events reported in the first 7 days GI: More nausea with AS+MQ but only significant in the group having supervised treatment (P = 0.05), diarrhoea, vomiting, and abdominal pain were not significantly different CNS: More dizziness with AS+MQ but only significant in the group having unsupervised treatment (P = 0.03), no other symptoms reported
Tangpukdee 2005 THA (180 participants)	Inpatient monitoring until day 28. Assessed using non‐suggestive questioning.	Open label	SAE: No serious adverse events observed Other: Reported as minor. No differences between groups reported
Tran 2002 VNM (243 participants)	Review at days 0, 2 and 7 LFTs on days 3, 7 and 28. Further follow‐up is unclear.	Open label	SAE: 12 events (10 vomiting, 2 dizziness) described as significant in AS+MQ group and none with DHA‐P (P = 0.002) Biochemical: No significant differences Other: All other adverse events described as minor with no differences between groups reported

Dihydroartemisinin‐piperaquine vs amodiaquine plus sulfadoxine‐pyrimethamine
Study ID	Adverse event monitoring	Blinding	Adverse events
Karema 2004 RWA (510 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, 21, and 28) An adverse event defined as any unfavourable and unintended sign associated temporally with the use of the drug administered Differential WBC count (and liver function tests at 1 site only) assessed at days 0 and 14	Open label	SAE: Not reported (1 seizure with AQ+SP) GI: More vomiting (P = 0.007) and anorexia (P = 0.005) with AQ+SP. No difference in abdominal pain, diarrhoea, nausea CNS: More fatigue with AQSP (P = 0.001). No difference in seizures, headache, dizziness, drowsiness CVS/RS: No difference in cough, angina, oedema Biochemical: No differences in mean PCV or mean WBC. No hepatotoxicity observed (one site only) Other: No difference in rash
Zongo 2007 BFA (371 participants)	A standardized history and examination on days 0, 1, 2, 3, 7, 14, 21, 28, 35, and 42 Adverse events defined as untoward medical occurrences Haemoglobin measured on days 0 and 42 or day of clinical failure	Open label	SAE: No serious adverse events were observed GI: Abdominal pain was more common with AQ+SP (P < 0.05). No difference in vomiting, diarrhoea, or anorexia. CNS: No difference in headache or weakness CVS/RS: No difference in cough Other: Pruritis more common with AQ+SP (P < 0.05)

Artesunate plus mefloquine vs Artemether‐lumefantrine
Study ID	Adverse event monitoring	Blinding	Adverse events
Faye 2003 SEN (294 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor, mainly gastralgia, dizziness, pruritis, asthenia, and vomiting Biochemical: No severe alterations in renal or hepatic function were observed
Hutagalung 2002 THA (490 participants)	Routine follow up daily until fever and parasites cleared then weekly to day 42 or any other day they became unwell At each visit a questionnaire on adverse events was completed An adverse event defined as symptoms or signs that were not present on admission and that developed after the start of treatment	Open label	SAE: None reported Overall comment: Both treatment regimens were well tolerated
Lefevre 1999 THA (219 participants)	Routine follow up at days 1, 2, 3, 7, 14, 21, and 28. Adverse events assessed at each visit. ECG monitoring and laboratory tests (including FBC liver and renal function tests) at baseline and each day of follow‐up.	Open label	SAE: No comment. GI: Abdominal pain, nausea, vomiting, diarrhoea, anorexia, constipation 18.3% AL vs 21.8% AS+MQ CNS: Headache, dizziness, and sleep disorder‐ 27.4% AL vs 16.4% AS+MQ CVS/RS: ECG 2% of each group showed QT prolongation of potential relevance with no cardiac complication Haematological: Slight worsening of anaemia after 3 days in both groups Biochemical: Liver function tests slightly abnormal at baseline. All baseline parameters normalized over the course of treatment. Renal function, electrolytes, glucose. Protein, urine tests showed no relevant changes after baseline in either group. Other: Skin reactions 8 AL vs 2 AS+MQ
Mayxay 2003 LAO (220 participants)	Routine follow up daily until fever and parasites cleared then weekly until day 42 or anytime they felt unwell Potential side effects were recorded at each visit	Open label	SAE: 3 serious neuropsychiatric events in AS+MQ group GI: Nausea and vomiting, abdominal pain, and diarrhoea more common with AS+MQ (P < 0.05) CNS: Weakness, dizziness, headache, confusion, and irritable/angry all more common with AS+MQ (P < 0.05). No difference in nightmares and tinnitus. CVS/RS: No difference in palpitations or dyspnoea Other: No difference in urticaria, herpes or blurred vision
Sagara 2005b MLI (270 participants)	Routine follow up on days 1, 2, 3, 7, 14, 21, and 28 Complete blood count, ALT and creatinine on 20% of participants on days 0 and 14 A serious adverse event was defined according to the International Conference on Harmonisation	Open label	SAE: Not mentioned GI: Vomiting more common with AS+MQ (P = 0.04). No significant difference in abdominal pain or diarrhoea. CNS: No significant difference in headache, weakness, dizziness (P = 0.06) or malaise Dermatological: No significant difference in pruritis or rash Biochemical: States 'both treatments were similar for laboratory adverse events'
Stohrer 2003 LAO (108 participants)	Treatment emergent symptoms and signs were recorded on days 0 to 3	Open label	SAE: 1 AL: severe diarrhoea, 1 ASMQ heavy sleep disorder and dizziness GI: None of the patients in either arm vomited within 1 hour of drug intake. No differences in abdominal pain, nausea, vomiting, diarrhoea, anorexia. CNS: Headache, dizziness, weakness, sleep disorder: 14 AL vs 22 ASMQ no significant difference
Van den Broek 2003a BGD (242 participants)	Routine follow up on days 0, 1, 2, 3, 7, 14, 21, 28, 35, and 42 and any other day when feeling ill Possible side effects assessed at each visit	Open label	SAE: None observed During the first 3 days headache, vomiting, nausea, and dizziness were significantly more common with AS+MQ (P < 0.05) Other complaints were: sleeplessness, pruritis/rash, epigastric pain, sweating with AS+MQ; blurred vision and anorexia with AL
Van Vugt 1998 THA (200 participants)	Routine follow up daily until fever and parasites cleared then weekly to day 28 A questionnaire for adverse effects was completed at each visit. Full neurological examination on days 0, 3, 7, and 28. Complete haematology and biochemistry (at one centre) on days 0, 3, 7, and 28.	Open label	SAE: 1 with AL: coma lasting 4 days 12 days after treatment, 1 with AS+MQ; generalized urticaria on day 1 Vomiting of medication: 4/150 AL vs 5/50 ASMQ (P = 0.045) GI: Anorexia, vomiting, nausea, abdominal pain, hepatomegaly less common with AL (12.7% AL vs 26% AS+MQ, P = 0.043) CVS: No electrocardiographic changes CNS: CNS symptoms (dizziness, sleep disorder, headache) less common with AL (6% AL vs 34% AS+MQ, P < 0.0001). One case of tremor and 2 cases of numbness with AL. Overall: Possible drug related adverse events less common with AL (33/150 AL vs 23/50 ASMQ, P = 0.002)

Artesunate plus mefloquine vs Artesunate plus amodiaquine
Study ID	Adverse event monitoring	Blinding	Adverse events
Faye 2003 SEN (505 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver, and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor; mainly gastralgia, dizziness, pruritis, asthenia, and vomiting Biochemical: No severe alterations in renal or hepatic function were observed

Artesunate plus mefloquine vs Amodiquine plus sulfadoxine‐pyrimethamine
Study ID	Adverse event monitoring	Blinding	Adverse events
Faye 2003 SEN (306 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver, and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor, mainly gastralgia, dizziness, pruritis, asthenia, and vomiting Biochemical: No severe alterations in renal or hepatic function were observed

Artemether‐lumefantrine vs Artesunate plus amodiaquine
Study ID	Adverse event monitoring	Blinding	Adverse events
Adjei 2006 GHA (227 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, and 28), including neurological assessment Audiological assessment on days 0, 3, 7, and 28 Total and differential WBC counts and liver enzymes on days 0, 3, 7, 14, and 28	Single blind (outcome assessors)	SAE: 1 patient treated with AS+AQ had severe anaemia on day 14 GI: No significant difference in nausea and vomiting between groups CNS: No significant difference in dizziness, fatigue, or excessive sleepiness between groups. Nystagmus was observed in 1 patient in each group, both cases had potential explanations from the past medical history. A positive Romberg's test was observed in 1 child treated with AL, again with a possible alternative diagnosis. Audiology: Hearing thresholds were significantly elevated in treated subjects as days 0, 3, 7, and 28 but no differences between participants and controls after 9 months Haematological: The mean neutrophil count was lower than baseline in both groups throughout follow up but there was no significant difference between groups. There was no significant difference in the incidence of neutropenia between groups (14/111 AL vs 13/116) Biochemical: No difference in liver enzymes were observed between groups. Liver enzymes were not observed to increase in response to treatment.
Bukirwa 2005 UGA (408 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, and 28), including neurological assessment An adverse event defined as any untoward medical occurrence	Single blind (outcome assessors)	SAE: One serious adverse event in each group (AL6 convulsion; AS+AQ pneumonia) both judged unlikely to be related to study meds CNS: No abnormalities in hearing or fine finger dexterity Overall comment: Adverse events of at least moderate severity: 125/202 AL vs 136/201 ASAQ (P = 0.25)
Dorsey 2006 UGA (434 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, and 28) An adverse event defined as any untoward medical occurrence Complete blood count and liver enzymes on days 0 and 14	Single blind (outcome assessors)	SAE: 29 serious adverse events (14/202 AL vs 15/232 ASAQ). Majority were seizures associated with fever. None considered probably or definitely related to study meds GI: Anorexia more common with ASAQ (P < 0.05). No significant difference in abdominal pain, vomiting or diarrhoea CVS/RS: No significant difference in cough CNS: No other significant differences in weakness Biochemical: Elevated liver enzymes occurred in 7 patients, all were attributed to other causes (6 viral hepatitis and 1 Salmonella bacteraemia) Other: No significant difference in pruritis
Falade 2005 NGA (132 participants)	Assessed at each visit (days 0 to 7, 14, 21, and 28) FBC, WBC and liver enzymes on days 0, 7 and 28 An adverse event defined as not present at enrolment but occurring during follow ‐up	Open label	SAE: There were no serious adverse events GI: No significant difference in abdominal pain or vomiting CVS/RS: No significant difference in cough or palpitations Haem: A significant transient decline in neutrophil counts between days 0 and 7 with AL which recovered by day 28 Biochemical: No statistically significant disturbance in blood chemistry. The study drugs did not adversely affect liver enzymes
Faye 2003 SEN (509 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor, mainly gastralgia, dizziness, pruritis, asthenia, and vomiting Biochemical: No severe alterations in renal or hepatic function were observed
Guthmann 2004 AGO (134 participants)	Adverse event monitoring not described	Unclear	AE not reported (2 patients excluded from AS+AQ group for vomiting and 1 from AL)
Kobbe 2007 GHA (237 participants)	'The comparative tolerability was assessed by the risk of occurrence of an adverse event' For each adverse event causality was assessed as recommended by the WHO	Open label	SAE: 2 SAE in each group, all classified as unlikely to be related to the treatment (asthma attack, febrile convulsion, enteritic bacterial infection, and severe anaemia) GI: No difference in GI symptoms including vomiting CVS/RS: No difference in respiratory symptoms Derm: No difference in dermatological symptoms
Koram 2003 GHA (105 participants)	Adverse event monitoring not described	Open label	AE not reported (3 patients with AS+AQ and 1 with AL were withdrawn for excessive vomiting)
Martensson 2003 TZA (407 participants)	Possible adverse events recorded at each visit (days 0, 1, 2, 3, 7, 14, 21, 28, 35, and 42) Differential white cell counts at days 0, 3, 7, 14, 21, and 28 An adverse event was defined as any undesirable medical occurrence regardless of whether it was related to the treatments	Unclear	SAE: 9 severe adverse events (2/200 AL vs 7/208 AS+AQ) all associated with clinically suspected severe malaria and not attributed to study drugs Haematological: No significant differences in mean WBC or neutrophil count between groups Overall comment: Both regimens generally well tolerated
Mutabingwa 2004 TZA (1034 participants)	Parents or guardians were asked to report on side effects, tolerability and usefulness of the treatment (days 0, 14, and 28)	Unclear	SAE: 1 death in the group treated with AL No other reporting of AE
Van den Broek 2004 ZAR (207 participants)	Possible side effects as passively reported to the examiner were recorded at each visit (days 0, 1, 2, 3, 7, 14, 21, and 28)	Open label	SAE: No severe adverse events judged to be related to the treatment given Overall comment: Common complaints were vomiting, diarrhoea, abdominal pain, and anorexia The frequency of potential adverse events was low (around 10%) and did not differ between groups. 1 case of urticaria occurred with AS+AQ
Owusu‐Agyei 2006 GHA (355 participants)	Field workers visited their homes to solicit adverse events on days 0, 2, 3, 7, 14, and 28	Open label	SAE: Not reported GI: No significant difference in diarrhoea, vomiting, nausea, anorexia, abdominal pain CNS: No significant difference in difficulty sleeping CVS/RS: No significant difference in cough, dyspnoea, palpitation Other: Body pain more common with AS+AQ. No difference in fever, runny nose, itching, joint pain, ulcers, yellow eyes

Artemether‐lumefantrine vs Amodiaquine plus sulfadoxine‐pyrimethamine
Study ID	Adverse event monitoring	Blinding	Adverse events
Dorsey 2006 UGA (455 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, and 28) An adverse event defined as any untoward medical occurrence Complete blood count and liver enzymes on days 0 and 14	Single blind (outcome assessors)	SAE: 30 serious adverse events (14/202 AL vs 16 AQ+SP). Majority were seizures associated with fever. None considered probably or definitely related to study meds. GI: Anorexia more common with AQ+SP (P < 0.05). No significant difference in abdominal pain, vomiting, or diarrhoea. CVS/RS: No significant difference in cough CNS: Weakness more common with AQ+SP (P < 0.05). No other significant differences. Biochemical: Elevated liver enzymes occurred in 7 patients, all were attributed to other causes (6 viral hepatitis and 1 Salmonella bacteraemia) Other: No significant difference in pruritis
Fanello 2004 RWA (500 participants)	All adverse events were recorded on the clinical record form (days 7, 14, 21, and 28) and a causality assessment was made PCV and WBC days 0 and 14	Open label	SAE: No comment on serious AE Overall comment: 251 patients reported one AE concomitant with administration of the drug with no differences between groups. AE possibly or probably related to the study drugs 22/251 AL, 35/249 AQ+SP P = 0.06 Haem: Mean WBC count at day 14 was similar in both groups (data not shown).
Faye 2003 SEN (310 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor, mainly gastralgia, dizziness, pruritis, asthenia, and vomiting Biochemical: No severe alterations in renal or hepatic function were observed
Mutabingwa 2004 TZA (1026 participants)	Parents or guardians were asked to report on side effects, tolerability, and usefulness of the treatment (days 0, 14, and 28)	Unclear	SAE: 1 death in each group No other reporting of AE
Zongo 2007 BFA (372 participants)	Assessed at each visit (days 0, 1, 2, 3, 7, 14, 21, 28, 35, and 42) Adverse events defined as any untoward medical occurrence	Open label	SAE: No serious adverse events GI: No significant difference in abdominal pain, vomiting, diarrhoea, or anorexia CVS/RS: No significant difference in cough CNS: No significant difference in headache or weakness. Other: Pruritis more common with AQ+SP (P < 0.05)
Zongo 2005 BFA (521 participants)	Assessed at each visit (days 0, 1, 2, 3, 7, 14, 21, and 28) Adverse events defined as any untoward medical occurrence	Double blind	SAE: 1 serious AE in each group (severe anaemia) GI: No significant difference in abdominal pain, vomiting, diarrhoea, or anorexia CVS/RS: No significant difference in cough or coryza CNS: No significant difference in headache or weakness Other: Pruritis more common with AQ+SP (P < 0.0001)

Is Dihydroartemisinin‐piperaquine as effective as Artesunate plus mefloquine for uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas worldwide  Intervention: Dihydroartemisinin‐piperaquine  Comparison: Artesunate plus mefloquine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artesunate plus mefloquine	Dihydroartemisinin‐piperaquine
*Efficacy: Total Failure (P. falciparum) Day 63 PCR adjusted ‐ Asia*	46 per 1000	18 per 1000   (9 to 36)	RR 0.39   (0.19 to 0.79)	1062  (3)	⊕⊕⊕⊕  high^1,2,3,4,5,6
*Efficacy: Total Failure (P. falciparum) Day 63 PCR unadjusted ‐ Asia*	151 per 1000	110 per 1000   (82 to 148)	RR 0.73   (0.54 to 0.98)	1182  (3)	⊕⊕⊕⊕  high^1,2,3,4,5,6
*Efficacy: Total Failure (P. falciparum) Day 63 PCR adjusted ‐ South America*	0 per 1000	Not estimable	RR 9.55   (0.52 to 176.35)	435  (1)	⊕  very low^7,8,9,10
*Efficacy: Total Failure (P. falciparum) Day 63 PCR unadjusted ‐ South America*	9 per 1000	56 per 1000   (13 to 246)	RR 6.19   (1.4 to 27.35)	445  (1)	⊕⊕⊕  moderate^7,8,9,11
*Vivax efficacy: P. vivax* parasitaemia by day 63**	180 per 1000	200 per 1000   (164 to 241)	RR 1.11   (0.91 to 1.34)	1661  (4)	⊕⊕⊕  moderate^{4,12,13,14,15}
Transmission potential: Gametocyte development (in those negative at baseline)	9 per 1000	28 per 1000   (10 to 79)	RR 3.06   (1.13 to 8.83)	1234  (3)	⊕⊕⊕⊕  high^4,11,13,16
Harms: Serious adverse events (including deaths)	7 per 1000	6 per 1000   (3 to 15)	RR 0.9   (0.38 to 2.15)	2617  (8)	⊕⊕  low^4,10,13,17
Harms: Early vomiting	88 per 1000	79 per 1000   (61 to 102)	RR 0.90   (0.69 to 1.16)	2473  (7)	⊕⊕  low^4,13,18,19
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Is Dihydroartemisinin‐piperaquine as effective as Artemether‐lumefantrine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas worldwide  Intervention: Dihydroartemisinin‐piperaquine  Comparison: Artemether‐lumefantrine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artemether‐ lumefantrine	Dihydroartemisinin‐ piperaquine
*Efficacy: Total Failure (P. falciparum) Day 42 PCR adjusted ‐ Africa*	117 per 1000	46 per 1000   (28 to 75)	RR 0.39   (0.24 to 0.64)	869  (3)	⊕⊕⊕⊕  high^{1,2,3,4,5,6,7}
*Efficacy: Total Failure (P. falciparum) Day 42 PCR unadjusted ‐ Africa*	380 per 1000	167 per 1000   (76 to 361)	RR 0.44   (0.20 to 0.95)	1136  (3)	⊕⊕⊕  moderate^2,3,4,6,8,9
*Efficacy: Total Failure (P. falciparum) Day 42 PCR adjusted ‐ Asia*	22 per 1000	17 per 1000   (4 to 83)	RR 0.77   (0.16 to 3.76)	317  (1)	⊕  very low^{1,10,11,12,13}
*Efficacy: Total Failure (P. falciparum) Day 42 PCR unadjusted ‐ Asia*	161 per 1000	97 per 1000   (56 to 169)	RR 0.60   (0.35 to 1.05)	356  (1)	⊕⊕  low^10,11,12,14
*Vivax efficacy: P. vivax* parasitaemia by D42**	197 per 1000	63 per 1000   (47 to 85)	RR 0.32   (0.24 to 0.43)	1442  (4)	⊕⊕⊕⊕  high^2,5,7,15,16
Transmission potential: Gametocyte development (in those negative at baseline)	—	—	—	1203  (4)	⊕  very low^17,18,19
Harms: Serious adverse events (including deaths)	6 per 1000	10 per 1000   (4 to 27)	RR 1.71   (0.66 to 4.46)	2110  (5)	⊕⊕  low^5,20,21
Harms: Early vomiting	23 per 1000	32 per 1000   (16 to 64)	RR 1.38   (0.68 to 2.78)	1147  (2)	⊕  very low^5,21,22
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Is Dihydroartemisinin‐piperaquine as effective as Artesunate plus amodiaquine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas worldwide  Intervention: Dihydroartemisinin‐piperaquine  Comparison: Artesunate plus amodiaquine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artesunate plus amodiaquine	Dihydroartemisinin‐piperaquine
*Efficacy: Total Failure (P. falciparum) Day 28 PCR adjusted*	73 per 1000	34 per 1000   (17 to 69)	RR 0.47   (0.23 to 0.94)	629  (2)	⊕⊕⊕  moderate^{1,2,3,4,5,6,7,8}
*Efficacy: Total Failure (P. falciparum) Day 28 PCR unadjusted*	161 per 1000	85 per 1000   (56 to 130)	RR 0.53   (0.35 to 0.81)	679  (2)	⊕⊕⊕  moderate^{2,3,4,5,6,7,8}
*Vivax efficacy: P. vivax* parasitaemia by day 42**	175 per 1000	44 per 1000   (16 to 130)	RR 0.25   (0.09 to 0.74)	170  (1)	⊕⊕⊕  moderate^9,10,11
Transmission potential: Gametocyte carriage	—	—	—	881 (2)	—¹²
Harms: Serious adverse events (including deaths)	18 per 1000	3 per 1000   (0 to 49)	RR 0.14   (0.01 to 2.71)	334  (1)	⊕  very low^9,10,13
Harms: Early vomiting	78 per 1000	41 per 1000   (17 to 101)	RR 0.53   (0.22 to 1.3)	334  (1)	⊕  very low^10,13,14
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Is Artesunate plus mefloquine superior to Artemether‐lumefantrine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas worldwide  Intervention: Artesunate plus mefloquine  Comparison: Artemether‐lumefantrine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artemether‐ lumefantrine	Artesunate plus mefloquine
*Efficacy: Total Failure (P. falciparum) Day 42 PCR adjusted*	28 per 1000	11 per 1000   (1 to 80)	RR 0.38   (0.05 to 2.84)	904  (4)	⊕  very low^1,2,3,4,5,6
*Efficacy: Total Failure (P. falciparum) Day 42 PCR unadjusted*	149 per 1000	79 per 1000   (43 to 140)	RR 0.53   (0.29 to 0.94)	1000  (4)	⊕⊕  low^1,2,3,4,5,7
*Vivax efficacy: P. vivax* parasitaemia by day 42**	246 per 1000	74 per 1000   (52 to 101)	RR 0.3   (0.21 to 0.41)	1003  (4)	⊕⊕⊕⊕  high^2,5,8,9
Transmission potential: Gametocyte carriage day 14	15 per 1000	6 per 1000   (1 to 31)	RR 0.41   (0.08 to 2.1)	536  (2)	⊕⊕  low^8,10,11
Harms: Serious adverse events (including deaths)	2 per 1000	6 per 1000   (1 to 28)	RR 2.96   (0.64 to 13.76)	1773  (7)	⊕⊕  low^8,11,12
Harms: Early vomiting	20 per 1000	21 per 1000   (11 to 42)	RR 1.07   (0.55 to 2.08)	1479  (6)	⊕  very low^8,11,12,13
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Artesunate plus amodiaquine vs Artesunate plus sulfadoxine‐pyrimethamine
Study ID	Adverse event monitoring	Blinding	Adverse events
Bonnet 2004 GIN (220 participants)	Adverse event monitoring not described	Open label	AE not reported
Djimde 2004 MLI ( participants)	Haemoglobin, glucose, complete blood count, liver enzymes, and creatinine were measured on days 0, 7, 14, and 28	Single blind (details not given)	SAE: One with AS+AQ. Overall comment: Adverse event distribution was unremarkable. Haematological: All treatment decreased the prevalence of abnormal values of leucocytes and platelets (figures not given) Biochemical: At day 14 the prevalence of grade 1 ALT toxicity was 9.7% AS+AQ vs 2.5% AS+SP (figures not given). These changes not thought to be clinically significant.
Guthmann 2003 AGO (187 participants)	Adverse event monitoring not described	Open label	AE not reported
Hamour 2003 SDN (161 participants)	Adverse event monitoring not described	Open label	SAE: No significant adverse events Overall comment: No significant adverse events were reported
Kayentao 2006 MLI (265 participants)	Adverse event monitoring not described	Single blind	One death occurred at day 7 after treatment with AS+SP. The parasitaemia was reported as cleared and cause of death unknown. Other AE not reported
Swarthout 2004 ZAR (180 participants)	Parents and guardians were asked about tolerability and potential side effects of the drugs (days 0, 1, 2, 3, 7, 14, 21, and 28)	Open label	SAE: None reported Overall comment: There were no adverse side effects reported by parents and both regimens were well tolerated
Van den Broek 2004 ZAR (192 participants)	Possible side effects as passively reported to the examiner were recorded at each visit (days 0, 1, 2, 3, 7, 14, 21, and 28)	Open label	SAE: No severe adverse events judged to be related to the treatment given Overall comment: Common complaints were vomiting, diarrhoea, abdominal pain and anorexia The frequency of potential adverse events was low (around 10%) and did not differ between groups. 1 case of urticaria occurred with AS+SP.

Artesunate plus amodiaquine vs Amodiaquine plus sulfadoxine‐pyrimethamine
Study ID	Adverse event monitoring	Blinding	Adverse events
Dorsey 2006 UGA (485 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, and 28) An adverse event defined as any untoward medical occurrence Complete blood count and liver enzymes on days 0 and 14	Single blind (outcome assessors)	SAE: 31 serious adverse events (15/232 AS+AQ vs 16/253 AQSP). Majority were seizures associated with fever. None considered probably or definitely related to study meds. GI: Anorexia more common with AQ+SP (P < 0.05). No significant difference in abdominal pain, vomiting or diarrhoea. CVS/RS: No significant difference in cough CNS: Weakness more common with AQ+SP (P < 0.05). No other significant differences Biochemical: Elevated liver enzymes occurred in 7 patients, all were attributed to other causes (6 viral hepatitis and 1 Salmonella bacteraemia) Other: No significant difference in pruritis
Faye 2003 SEN (521 participants)	All side effects were monitored actively (days 0, 1, 2, 7, 14, 21, and 28) and passively during the study 25% were randomly selected for blood counts, liver and renal function tests at days 0, 14, and 28	Open label	SAE: No serious adverse events Overall comment: The side effects observed with each treatment combination were minor, mainly gastralgia, dizziness, pruritis, asthenia and vomiting Biochemical: No severe alterations in renal or hepatic function were observed
Karema 2004 RWA (510 participants)	Assessed at each follow‐up visit (days 0, 1, 2, 3, 7, 14, 21, and 28) An adverse event defined as any unfavourable and unintended sign associated temporally with the use of the drug administered Differential WBC count (and liver function tests at one site only) assessed at days 0 and 14	Open label	SAE: Not reported (one seizure with AS+AQ, one with AQ+SP) GI: No differences in nausea, vomiting, diarrhoea, abdominal pain, or anorexia CVS/RS: No difference in cough, angina, oedema CNS: No difference in seizures, headache, dizziness, drowsiness, or fatigue Biochemical: No differences in mean PCV or mean WBC. No hepatotoxicity observed (1 site only) Other: No difference in rash
Kayentao 2006 MLI (265 participants)	Adverse event monitoring not described	Single blind	AE not reported
Menard 2006 MDG (166 participants)	Adverse event monitoring not described	Single blind (outcome assessors)	SAE: ‘No severe side effects attributable to the study medication’ No other reporting of AE
Mutabingwa 2004 TZA (1022 participants)	Parents or guardians were asked to report on side effects, tolerability, and usefulness of the treatment (days 0, 14, and 28)	Unclear	SAE: 1 death in the AQ+SP group died on the day of randomization No other reporting of AE
Staedke 2003 UGA (268 participants)	Assessed at each visit with a standardized history and examination. Neurological assessment on days 0, 7, 14, and 28. CBC, creatine and alanine transferase on days 0, 7, and 28.	Single blind (outcome assessors)	SAE: 16 serious adverse events (1/134 AS+AQ vs 6/134 AQ+SP CNS: ‘No important neurological events were seen’ Biochem: 1 severe anaemia with AS+AQ, 1 severe neutropenia with AQ+SP, 1 elevated alanine transaminase with AQ+SP No other comment on adverse events
Yeka 2004 UGA (1461 participants)	Adverse event monitoring not described	Single blind (outcome assessors)	SAE: 4/731 AS+AQ vs 10/730 AQ+SP. 2 additional patients died in the AQ+SP group No other reporting of AE

Dihydroartemisinin‐piperaquine vs Artesunate plus mefloquine
Study ID	Outcome measure and result	Significance test
Ashley 2003b THA	Median decrease in haematocrit by day 7: DHA‐P 6.3% (0% to 13.6%) vs AS+MQ 9.4% (2.6% to 14.3%) Mean haematocrit weekly from day 0 to 63: Presented graphically	P = 0.21
Ashley 2004 THA	Median change in haematocrit in each group, each week, from day 0 to 63: 'a decrease in haematocrit in both groups between days 0 and 7 followed by recovery in both groups'. Figures presented graphically.	Not reported
Janssens 2003 KHM	Mean haematocrit at day 63: DHA‐P 40.0% vs AS+MQ 40.2% (No differences at baseline)	Not reported
Mayxay 2004 LAO	Mean haematocrit following treatment (days 7 to 42): 'did not significantly differ between groups'. Figures not given.	Not significant P > 0.05
Smithuis 2004 MMR	Mean haemoglobin at day 28 (supervized treatment): DHA‐P 10.4g/dl vs AS+MQ 10.5g/dl Proportion anaemic (Hb < 10g/dl) on day 28 (superviZed treatment): DHA‐P 56/152 vs AS+MQ 59/156 (no differences at baseline)	P = 0.65      P = 0.85

Is Dihydroartemisinin‐piperaquine superior to Artesunate plus sulfadoxine‐pyrimethamine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas excluding Southeast Asia  Intervention: Dihydroartemisinin‐piperaquine  Comparison: Artesunate plus sulfadoxine‐pyrimethamine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artesunate plus sulfadoxine‐pyrimethamine	Dihydroartemisinin ‐piperaquine
Efficacy: Total Failure Day 42 PCR adjusted	202 per 1000	156 per 1000   (79 to 305)	RR 0.77   (0.39 to 1.51)	161  (1)	⊕  very low^1,2,3,4
Efficacy: Total Failure Day 42 PCR unadjusted	380 per 1000	391 per 1000   (281 to 551)	RR 1.03   (0.74 to 1.45)	215  (1)	⊕  very low^1,2,3,4
*Vivax efficacy: P. vivax* parasitaemia Day 42**	596 per 1000	268 per 1000   (191 to 387)	RR 0.45   (0.32 to 0.65)	194  (1)	⊕⊕  low^1,2,3,5
Transmission potential: Gametocyte carriage	—	—	—	215  (1)	—⁶
Harms: Serious adverse events (including deaths)	—	—	—	—	Not reported
Harms: Early vomiting	—	—	—	—	Not reported
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Is Artemether‐lumefantrine superior to Artesunate plus amodiaquine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Africa  Intervention: Artemether‐lumefantrine  Comparison: Artesunate plus amodiaquine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
Outcomes	Assumed risk	Corresponding risk
*Efficacy: Total Failure (P. falciparum) Day 28 PCR adjusted*	Artesunate plus amodiaquine	Artemether‐lumefantrine
*Efficacy: Total Failure (P. falciparum) Day 28 PCR adjusted*	19 per 1000	31 per 1000   (18 to 55)	RR 1.65   (0.95 to 2.87)	1729  (8)	⊕⊕⊕  moderate^1,2,3,4,5,6
*Efficacy: Total Failure (P. falciparum) Day 28 PCR unadjusted*	—	—	—	2617  (5)	⊕  very low^2,5,7,8,9
*Vivax efficacy: P. vivax* parasitaemia**	—	—	—	—	Not reported¹⁰
Transmission potential: Gametocyte carriage day 14	—	—	—	718  (2)	⊕  very low^11,12,13,14
Harms: Serious adverse events (including deaths)	13 per 1000	14 per 1000   (8 to 27)	RR 1.11   (0.59 to 2.08)	2617  (5)	⊕⊕  low^3,4,5,15
Harms: Early vomiting	83 per 1000	72 per 1000   (49 to 109)	RR 0.87   (0.59 to 1.31)	1097  (5)	⊕  very low^4,15,16
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).    CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Is Artesunate plus amodiaquine superior to Artesunate plus sulfadoxine‐pyrimethamine for treating uncomplicated malaria?
Patient or population: Patients with uncomplicated malaria  Settings: Endemic areas worldwide  Intervention: Artesunate plus amodiaquine  Comparison: Artesunate plus sulfadoxine‐pyrimethamine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of participants   (studies)	Quality of the evidence   (GRADE)
	Assumed risk	Corresponding risk
	Artesunate plus sulfadoxine‐pyrimethamine	Artesunate plus amodiaquine
*Efficacy: Total Failure (P. falciparum) Day 28 PCR adjusted*	44 per 1000	28 per 1000   (16 to 48)	RR 0.64   (0.37 to 1.08)	1419  (7)	⊕⊕  low^1,2,3,4,5
*Efficacy: Total Failure (P. falciparum) Day 28 PCR unadjusted*	—	—	—	1614  (7)	⊕  very low^1,2,4,6,7
*Vivax efficacy: P. vivax* parasitaemia**	—	—	—	—	Not reported
Transmission potential: Gametocyte carriage day 14	91 per 1000	81 per 1000   (46 to 140)	RR 0.89   (0.51 to 1.54)	520  (3)	⊕  very low^8,9,10,11
Harms: Serious adverse events (including deaths)	2 per 1000	2 per 1000   (0 to 14)	RR 0.99   (0.14 to 7.02)	1108  (4)	⊕  very low^9,10,11
Harms: Early vomiting	—	—	—	—	Not reported
The assumed risk* is the mean risk from the studies included in this review, calculated as the number of patients in the control groups with the event divided by the total number of patients in control groups. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure (P. falciparum) Day 63 PCR unadjusted	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.1 Asia	3	1182	Risk Ratio (M‐H, Fixed, 95% CI)	0.73 [0.54, 0.98]
1.2 South America	1	445	Risk Ratio (M‐H, Fixed, 95% CI)	6.19 [1.40, 27.35]
2 Total Failure (P. falciparum) Day 63 PCR adjusted	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 Asia	3	1062	Risk Ratio (M‐H, Fixed, 95% CI)	0.39 [0.19, 0.79]
2.2 South America	1	435	Risk Ratio (M‐H, Fixed, 95% CI)	9.55 [0.52, 176.35]
3 Total Failure (P. falciparum) Day 42 PCR unadjusted	5		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.1 Asia	5	1969	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.46, 1.69]
4 Total Failure (P. falciparum) Day 42 PCR adjusted	5		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
4.1 Asia	5	1898	Risk Ratio (M‐H, Fixed, 95% CI)	0.64 [0.30, 1.39]
5 Total Failure (P. falciparum) Day 28 PCR unadjusted	6		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
5.1 Asia	6	2034	Risk Ratio (M‐H, Random, 95% CI)	1.20 [0.22, 6.42]
6 Total Failure (P. falciparum) Day 28 PCR adjusted	6		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
6.1 Asia	6	2020	Risk Ratio (M‐H, Fixed, 95% CI)	0.69 [0.31, 1.56]
7 P. vivax parasitaemia	6		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
7.1 Mixed P. falciparum and vivax infection at baseline	5	2248	Risk Ratio (M‐H, Fixed, 95% CI)	0.84 [0.63, 1.12]
7.2 Total P. vivax parasitaemia by day 28	1	402	Risk Ratio (M‐H, Fixed, 95% CI)	7.43 [0.39, 142.89]
7.3 Total P. vivax parasitaemia by day 42	3	1251	Risk Ratio (M‐H, Fixed, 95% CI)	0.79 [0.57, 1.11]
7.4 Total P. vivax parasitaemia by day 63	4	1661	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.91, 1.34]
7.5 P. vivax parasitaemia by day 63 in those negative at baseline	3	1172	Risk Ratio (M‐H, Fixed, 95% CI)	1.22 [0.95, 1.56]
7.6 P. vivax parasitaemia by day 63 in those positive at baseline	2	79	Risk Ratio (M‐H, Fixed, 95% CI)	0.97 [0.57, 1.65]
8 Gametocyte development (in those negative at baseline)	3	1234	Risk Ratio (M‐H, Fixed, 95% CI)	3.06 [1.13, 8.33]
9 Gametocytaemia carriage	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.1 Gametocyte carriage day 0	2	1174	Risk Ratio (M‐H, Random, 95% CI)	1.07 [0.66, 1.73]
9.2 Gametocyte carriage day 7	2	1152	Risk Ratio (M‐H, Random, 95% CI)	2.00 [1.54, 2.58]
9.3 Gametocyte carriage day 14	2	1142	Risk Ratio (M‐H, Random, 95% CI)	5.14 [3.17, 8.33]
9.4 Gametocyte carriage day 21	2	1123	Risk Ratio (M‐H, Random, 95% CI)	7.23 [0.10, 519.79]
9.5 Gametocyte carriage day 28	2	1124	Risk Ratio (M‐H, Random, 95% CI)	9.68 [1.23, 75.98]
10 Serious adverse events (including deaths)	7	2374	Risk Ratio (M‐H, Fixed, 95% CI)	0.90 [0.38, 2.15]
11 Early vomiting	7	2473	Risk Ratio (M‐H, Fixed, 95% CI)	0.90 [0.69, 1.16]
12 Sensitivity analysis: Total Failure Day 63 PCR unadjusted	4		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
12.1 Total Failure (P. falciparum) Day 63 PCR unadjusted	4	1627	Risk Ratio (M‐H, Random, 95% CI)	0.94 [0.52, 1.70]
12.2 Total Failure Day 63 PCR unadjusted (losses to follow up included as failures)	4	1801	Risk Ratio (M‐H, Random, 95% CI)	0.95 [0.65, 1.38]
12.3 Total Failure Day 63 PCR unadjusted (losses to follow up included as successes)	4	1801	Risk Ratio (M‐H, Random, 95% CI)	0.94 [0.52, 1.68]
13 Sensitivity analysis: Total Failure Day 63 PCR adjusted	4		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
13.1 Total Failure (P. falciparum) Day 63 PCR adjusted	4	1497	Risk Ratio (M‐H, Random, 95% CI)	0.57 [0.17, 1.83]
13.2 Total Failure Day 63 PCR adjusted (indeterminate PCR included as failures)	4	1508	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.32, 1.39]
13.3 Total Failure Day 63 PCR adjusted (new infections included as successes)	4	1627	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.34, 1.35]
13.4 Total Failure Day 63 PCR adjusted (losses to follow up included as failures)	4	1801	Risk Ratio (M‐H, Random, 95% CI)	0.93 [0.67, 1.30]
13.5 Total Failure Day 63 PCR adjusted (losses to follow up included as successes)	4	1801	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.34, 1.33]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure (P. falciparum) Day 42 PCR unadjusted	5		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.1 Africa	3	1136	Risk Ratio (M‐H, Random, 95% CI)	0.44 [0.20, 0.95]
1.2 Asia	1	356	Risk Ratio (M‐H, Random, 95% CI)	0.60 [0.35, 1.05]
1.3 Oceania	1	216	Risk Ratio (M‐H, Random, 95% CI)	1.07 [0.76, 1.50]
2 Total Failure (P. falciparum) Day 42 PCR adjusted	5		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 Africa	3	869	Risk Ratio (M‐H, Fixed, 95% CI)	0.39 [0.24, 0.64]
2.2 Asia	1	317	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.16, 3.76]
2.3 Oceania	1	151	Risk Ratio (M‐H, Fixed, 95% CI)	2.31 [0.85, 6.23]
3 Total Failure (P. falciparum) Day 28 PCR unadjusted	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
3.1 Africa	2	484	Risk Ratio (M‐H, Fixed, 95% CI)	0.12 [0.05, 0.32]
3.2 Asia	1	451	Risk Ratio (M‐H, Fixed, 95% CI)	0.43 [0.17, 1.12]
3.3 Oceania	1	224	Risk Ratio (M‐H, Fixed, 95% CI)	1.27 [0.75, 2.15]
4 Total Failure (P. falciparum) Day 28 PCR adjusted	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
4.1 Africa	2	453	Risk Ratio (M‐H, Fixed, 95% CI)	0.57 [0.17, 1.99]
4.2 Asia	1	436	Risk Ratio (M‐H, Fixed, 95% CI)	0.90 [0.13, 6.36]
4.3 Oceania	1	193	Risk Ratio (M‐H, Fixed, 95% CI)	3.63 [1.04, 12.60]
5 P. vivax parasitaemia	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
5.1 Mixed P. falciparum and vivax infection at baseline	4	1608	Risk Ratio (M‐H, Fixed, 95% CI)	1.02 [0.73, 1.42]
5.2 P. vivax parasitaemia by D28	1	473	Risk Ratio (M‐H, Fixed, 95% CI)	0.05 [0.01, 0.36]
5.3 P. vivax parasitaemia by D42	4	1442	Risk Ratio (M‐H, Fixed, 95% CI)	0.32 [0.24, 0.43]
6 Gametocyte development (in those negative at baseline)	4	1203	Risk Ratio (M‐H, Random, 95% CI)	0.95 [0.35, 2.59]
7 Anaemia	4		Mean Difference (IV, Fixed, 95% CI)	Subtotals only
7.1 Mean haemoglobin (g/dl) at baseline	4	1356	Mean Difference (IV, Fixed, 95% CI)	‐0.07 [‐0.27, 0.13]
7.2 Mean haemoglobin (g/dl) at day 28	1	134	Mean Difference (IV, Fixed, 95% CI)	0.36 [‐0.03, 0.75]
7.3 Mean haemoglobin (g/dl) at day 42	1	375	Mean Difference (IV, Fixed, 95% CI)	0.30 [‐0.02, 0.62]
7.4 Mean change in haemoglobin (g/dl) from baseline to Day 42	2	835	Mean Difference (IV, Fixed, 95% CI)	0.26 [0.00, 0.51]
8 Serious adverse events (including deaths)	5	2110	Risk Ratio (M‐H, Fixed, 95% CI)	1.71 [0.66, 4.46]
9 Early vomiting	2	1147	Risk Ratio (M‐H, Fixed, 95% CI)	1.38 [0.68, 2.78]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure (P. falciparum) Day 28 PCR unadjusted	2	679	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.35, 0.81]
1.1 Africa	1	501	Risk Ratio (M‐H, Fixed, 95% CI)	0.54 [0.34, 0.85]
1.2 Asia	1	178	Risk Ratio (M‐H, Fixed, 95% CI)	0.50 [0.17, 1.42]
2 Total Failure (P. falciparum) Day 28 PCR adjusted	2	629	Risk Ratio (M‐H, Fixed, 95% CI)	0.47 [0.23, 0.94]
2.1 Africa	1	458	Risk Ratio (M‐H, Fixed, 95% CI)	0.59 [0.27, 1.27]
2.2 Asia	1	171	Risk Ratio (M‐H, Fixed, 95% CI)	0.15 [0.02, 1.22]
3 Total Failure (P. falciparum) Day 42 PCR unadjusted	1	152	Risk Ratio (M‐H, Fixed, 95% CI)	0.27 [0.10, 0.72]
3.1 Asia	1	152	Risk Ratio (M‐H, Fixed, 95% CI)	0.27 [0.10, 0.72]
4 Total Failure (P. falciparum) Day 42 PCR adjusted	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.10 [0.01, 0.81]
4.1 Asia	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.10 [0.01, 0.81]
5 P. vivax parasitaemia	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
5.1 Mixed P. falciparum and vivax infection at baseline	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	1.24 [0.67, 2.29]
5.2 P. vivax parasitaemia by day 28	1	181	Risk Ratio (M‐H, Fixed, 95% CI)	0.45 [0.04, 4.90]
5.3 P. vivax parasitaemia by day 42	1	170	Risk Ratio (M‐H, Fixed, 95% CI)	0.25 [0.09, 0.74]
6 Serious adverse events (including deaths)	1	334	Risk Ratio (M‐H, Fixed, 95% CI)	0.14 [0.01, 2.71]
7 Early vomiting	1	334	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.22, 1.30]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure (P. falciparum) Day 28 PCR unadjusted	9		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
1.1 East Africa	3		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.2 West Africa	5		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.3 South/Central Africa	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2 Total Failure (P. falciparum) Day 28 PCR adjusted	8	1729	Risk Ratio (M‐H, Fixed, 95% CI)	1.65 [0.95, 2.87]
2.1 East Africa	2	365	Risk Ratio (M‐H, Fixed, 95% CI)	0.83 [0.15, 4.59]
2.2 West Africa	5	1245	Risk Ratio (M‐H, Fixed, 95% CI)	1.81 [1.00, 3.26]
2.3 South/Central Africa	1	119	Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
3 Gametocyte development	1	305	Risk Ratio (M‐H, Fixed, 95% CI)	0.34 [0.15, 0.74]
4 Gametocyte carriage	3		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
4.1 Gametocyte carriage day 0	3		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4.2 Gametocyte carriage day 3	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4.3 Gametocyte carriage day 7	3		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4.4 Gametocyte carriage day 14	2		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
5 Anaemia	5		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
5.1 Mean haemoglobin (g/dl) at baseline	4		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.2 Mean haemoglobin (g/dl) at Day 28	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.3 Mean change in haemoglobin (g/dl) from baseline to Day 28	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.4 Mean haematocrit at baseline	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.5 Mean haematocrit at Day 28	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
6 Proportion anaemic (Haemoglobin < 11 g/dl)	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
6.1 At baseline	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
6.2 At day 28	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
7 Serious adverse events (including deaths)	6	2749	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.59, 2.08]
8 Early vomiting	5	1097	Risk Ratio (M‐H, Fixed, 95% CI)	0.87 [0.59, 1.31]
9 Sensitivity analysis: Total Failure Day 28 PCR unadjusted	12		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
9.1 Total Failure (P. falciparum) Day 28 PCR unadjusted	9	3021	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.60, 1.27]
9.2 Total Failure Day 28 PCR unadjusted (trials with baseline differences included)	12	3719	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.49, 0.97]
9.3 Total Failure Day 28 PCR unadjusted (losses to follow up included as failures)	9	3230	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.62, 1.06]
9.4 Total Failure Day 28 PCR unadjusted (losses to follow up included as successes)	9	3230	Risk Ratio (M‐H, Random, 95% CI)	0.89 [0.61, 1.30]
10 Sensitivity analysis: Total Failure Day 28 PCR adjusted	11		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
10.1 Total Failure (P. falciparum) Day 28 PCR adjusted	8	1729	Risk Ratio (M‐H, Fixed, 95% CI)	1.65 [0.95, 2.87]
10.2 Total Failure Day 28 PCR adjusted (trials with baseline differences included)	11	2311	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.69, 1.67]
10.3 Total Failure Day 28 PCR adjusted (indeterminate PCR included as failures)	8	1747	Risk Ratio (M‐H, Fixed, 95% CI)	1.72 [1.06, 2.78]
10.4 Total Failure Day 28 PCR adjusted (new infections included as successes)	8	2064	Risk Ratio (M‐H, Fixed, 95% CI)	1.70 [1.06, 2.75]
10.5 Total Failure Day 28 PCR adjusted (losses to follow up included as failures)	8	2196	Risk Ratio (M‐H, Fixed, 95% CI)	1.01 [0.78, 1.31]
10.6 Total Failure Day 28 PCR adjusted (losses to follow up included as successes)	8	2196	Risk Ratio (M‐H, Fixed, 95% CI)	1.75 [1.08, 2.83]

PERMALINK

Artemisinin‐based combination therapy for treating uncomplicated malaria

David Sinclair

Babalwa Zani

Sarah Donegan

Piero Olliaro

Paul Garner

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Background

Artemisinin‐based antimalarials

Assessing antimalarial efficacy

P. vivax malaria

Choice of combination treatment

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Intervention

Control

Types of outcome measures

Primary outcomes

Secondary outcomes

Adverse events

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Primary outcome

PCR‐unadjusted total failure

PCR‐adjusted total failure

Secondary outcomes and adverse events

Assessment of risk of bias in included studies

Measures of treatment effect

Dealing with missing data

Assessment of heterogeneity

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Results

Description of studies

Results of the search

Included studies

Excluded studies

9.9. Analysis.

9.10. Analysis.

Risk of bias in included studies

1.

2.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

1.12. Analysis.

1.13. Analysis.

Question 1. How does dihydroartemisinin‐piperaquine (DHA‐P) perform?

Dosing concerns

14.1. Analysis.

14.2. Analysis.

15.1. Analysis.

15.2. Analysis.

15.3. Analysis.

15.4. Analysis.

15.5. Analysis.

15.6. Analysis.

Comparison 1. DHA‐P versus artesunate plus mefloquine

Total failure

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure (P. falciparum) Day 28 PCR unadjusted	6		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.1 East Africa	3	1646	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.30, 0.41]
1.2 West Africa	3	1130	Risk Ratio (M‐H, Fixed, 95% CI)	2.88 [1.86, 4.47]
2 Total Failure (P. falciparum) Day 28 PCR adjusted	5		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1 East Africa	2	618	Risk Ratio (M‐H, Fixed, 95% CI)	0.12 [0.06, 0.24]
2.2 West Africa	3	1051	Risk Ratio (M‐H, Fixed, 95% CI)	1.39 [0.55, 3.47]
3 Total Failure (P. falciparum) Day 42 PCR unadjusted	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
3.1 West Africa	1	345	Risk Ratio (M‐H, Fixed, 95% CI)	2.64 [1.66, 4.21]
4 Total Failure (P. falciparum) Day 42 PCR adjusted	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
4.1 West Africa	1	284	Risk Ratio (M‐H, Fixed, 95% CI)	1.22 [0.44, 3.38]
5 Gametocyte carriage	4		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
5.1 Gametocyte carriage day 0	4	1545	Risk Ratio (M‐H, Fixed, 95% CI)	0.84 [0.51, 1.39]
5.2 Gametocyte carriage day 3	3	1331	Risk Ratio (M‐H, Fixed, 95% CI)	0.43 [0.25, 0.75]
5.3 Gametocyte carriage day 7	4	1538	Risk Ratio (M‐H, Fixed, 95% CI)	0.32 [0.18, 0.54]
5.4 Gametocyte carriage day 14	4	1536	Risk Ratio (M‐H, Fixed, 95% CI)	0.46 [0.21, 1.01]
6 Gametocyte development (in those negative at baseline)	1	371	Risk Ratio (M‐H, Fixed, 95% CI)	0.29 [0.08, 1.04]
7 Anaemia	2		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
7.1 Mean haemoglobin (g/dl) at baseline	2		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.2 Mean change in haemoglobin (g/dl) from baseline to Day 28	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.3 Mean haemoglobin (g/dl) at Day 42 or last day of follow up.	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
8 Serious adverse events (including deaths)	5	2684	Risk Ratio (M‐H, Fixed, 95% CI)	1.08 [0.56, 2.08]
9 Early vomiting	2	893	Risk Ratio (M‐H, Fixed, 95% CI)	1.42 [0.54, 3.68]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure PCR unadjusted	1	318	Risk Ratio (M‐H, Fixed, 95% CI)	1.72 [0.84, 3.53]
1.1 Day 63	1	318	Risk Ratio (M‐H, Fixed, 95% CI)	1.72 [0.84, 3.53]
2 Total Failure PCR adjusted	1	292	Risk Ratio (M‐H, Fixed, 95% CI)	0.56 [0.05, 6.09]
2.1 Day 63	1	292	Risk Ratio (M‐H, Fixed, 95% CI)	0.56 [0.05, 6.09]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure Day 63 PCR unadjusted	4	1784	Risk Ratio (M‐H, Random, 95% CI)	0.83 [0.49, 1.38]
1.1 DHA‐P 4 doses	3	1019	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.59, 1.10]
1.2 DHA‐P 3 doses	2	765	Risk Ratio (M‐H, Random, 95% CI)	1.44 [0.09, 22.81]
2 Total Failure Day 63 PCR adjusted	4	1634	Risk Ratio (M‐H, Random, 95% CI)	0.48 [0.18, 1.31]
2.1 DHA‐P 4 doses	3	908	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.17, 1.04]
2.2 DHA‐P 3 doses	2	726	Risk Ratio (M‐H, Random, 95% CI)	1.27 [0.03, 48.28]
3 Total Failure Day 42 PCR unadjusted	5	2126	Risk Ratio (M‐H, Random, 95% CI)	0.77 [0.43, 1.35]
3.1 DHA‐P 4 doses	3	957	Risk Ratio (M‐H, Random, 95% CI)	0.80 [0.50, 1.28]
3.2 DHA‐P 3 doses	3	1169	Risk Ratio (M‐H, Random, 95% CI)	0.88 [0.20, 3.81]
4 Total Failure Day 42 PCR adjusted	5	2043	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.20, 1.91]
4.1 DHA‐P 4 doses	3	903	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.14, 2.82]
4.2 DHA‐P 3 doses	3	1140	Risk Ratio (M‐H, Random, 95% CI)	0.70 [0.08, 5.87]
5 Total Failure Day 28 PCR unadjusted	6	2191	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.20, 2.65]
5.1 DHA‐P 4 doses	4	1075	Risk Ratio (M‐H, Random, 95% CI)	0.56 [0.10, 3.14]
5.2 DHA‐P 3 doses	3	1116	Risk Ratio (M‐H, Random, 95% CI)	1.29 [0.09, 18.93]
6 Total Failure Day 28 PCR adjusted	6	2171	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.19, 2.86]
6.1 DHA‐P 4 doses	4	1067	Risk Ratio (M‐H, Random, 95% CI)	0.79 [0.10, 6.11]
6.2 DHA‐P 3 doses	3	1104	Risk Ratio (M‐H, Random, 95% CI)	0.79 [0.08, 7.82]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure Day 63 PCR unadjusted	1	423	Risk Ratio (M‐H, Fixed, 95% CI)	0.93 [0.68, 1.27]
2 Total Failure Day 63 PCR adjusted	1	342	Risk Ratio (M‐H, Fixed, 95% CI)	0.66 [0.34, 1.28]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total Failure Day 63 PCR adjusted	3		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
2 Total Failure Day 28 PCR adjusted	2	279	Risk Ratio (M‐H, Fixed, 95% CI)	2.16 [0.23, 19.88]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Effectiveness: Total Failure (P. falciparum) PCR adjusted	11		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.1 Day 63: DHA‐P vs Artesunate plus mefloquine	4	1497	Risk Ratio (M‐H, Random, 95% CI)	0.57 [0.17, 1.83]
1.2 Day 42: DHA‐P vs Artemether‐lumefantrine	5	1337	Risk Ratio (M‐H, Random, 95% CI)	0.62 [0.29, 1.30]
1.3 Day 28: DHA‐P vs Artesunate plus amodiaquine	2	629	Risk Ratio (M‐H, Random, 95% CI)	0.42 [0.13, 1.35]
1.4 Day 42: DHA‐P vs Artesunate plus sulfadoxine‐pyrimethamine	1	161	Risk Ratio (M‐H, Random, 95% CI)	0.77 [0.39, 1.51]
1.5 Day 28: DHA‐P vs Amodiaquine plus sulfadoxine‐pyrimethamine	2	802	Risk Ratio (M‐H, Random, 95% CI)	0.32 [0.16, 0.64]

Methods	Trial design: A single blind randomized controlled trial Follow up: Clinical and laboratory assessment on days 0, 1, 2, 3, 7, 14, 28 and then monthly for 1 year Adverse event monitoring: Assessed at each visit up to 1 year using open questions about side effects, behavioural and developmental concerns. Neurological examination at each visit. Audiometry assessment on days 0, 3, 7, 28, and 1 year. WBC, aminotransferase and total bilirubin at days 0, 3, 7, 14, and 28.
Participants	Number: 227 randomized Inclusion criteria: Age 6 months to 14 yrs, axillary temp > 37.5 ºC, signs and symptoms of uncomplicated malaria, P. falciparum mono‐infection 2000 to 200,000/µl, willingness to comply with the follow up, informed consent Exclusion criteria: Signs or symptoms of severe malaria, chronic malnutrition or other severe disease, known intolerance or allergy to study meds, reported treatment with any of the study drugs during preceding month
Interventions	1. Artemether‐lumefantrine, fixed dose combination, 20 mg/120 mg tablets (Coartem: Novartis) 5 to 14 kg 1 tablet twice daily for 3 days 15 to 24 kg 2 tablets twice daily for 3 days 25 to 34 kg 3 tablets twice daily for 3 days > 35 kg 4 tablets twice daily for 3 days 2. Artesunate plus amodiaquine, loose combination (Plasmotrim: Mepha, Camoquine: Pfizer) AS 4 mg/kg once daily for 3 days AQ 10 mg/kg once daily for 3 days Only the first dose each day was supervized
Outcomes	ACPR at day 28, PCR adjusted and PCR unadjusted Adverse events including neurological, biochemical, and audiological events Not included in this review: Fever clearance Parasite clearance Further episodes of symptomatic malaria in 1 year
Notes	Country: Ghana Setting: Urban primary health facilities Transmission: Not described Resistance: AQ Dates: Oct 2004 to Dec 2006 Funding: Danish Council for Development Research, Global Fund for AIDS, TB and Malaria through the National Malaria Control Programme
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	'A computer generated randomisation scheme was prepared in advance'
Allocation concealment?	Low risk	'Allocated treatments were kept in sealed opaque envelopes'
Blinding?   All outcomes	Low risk	'All study personnel (except project nurses) were unaware of the assigned treatments'
Incomplete outcome data addressed?   All outcomes	Low risk	Low losses to follow up in both groups (7.2% AL6 vs 7.8% AS+AQ)
Free of selective reporting?	Low risk	All WHO outcomes reported. The WHO recommends 42 days follow up in studies of AL6. Day 28 outcomes may under estimate treatment failure with AL6.
Free of other bias?	Low risk	No other sources of bias identified

Methods	Trial design: A 3‐arm randomized controlled trial Follow up: All patients admitted to hospital for 28 days, oral temperature taken every 6 hours, parasite counts 12‐hourly until negative then daily for 28 days Adverse event monitoring: Adverse events defined as signs or symptoms that occurred or became more severe after treatment started. All patients had full blood counts, urea, electrolytes, creatinine, and liver function tests at days 0 and 7.
Participants	Number: 134 randomized into included treatment arms Inclusion criteria: Age > 14 yrs, weight > 40 kg, symptoms of malaria, P. falciparum parasitaemia, informed consent Exclusion criteria: Pregnancy or lactation, signs or symptoms of severe malaria, > 4% of red blood cells parasitized, contraindication to mefloquine, treatment with mefloquine in the previous 60 days, sulphonamides or 4‐aminoquinolones present in urine on admission
Interventions	1. Dihydroartemisinin‐piperaquine, fixed dose combination (Artekin: Holleykin) Total dose: 6 mg/kg DHA and 48 mg/kg P in 4 divided doses at 0, 8, 24 and 48 hours 2. Artesunate plus mefloquine, loose combination (Artesunate: Guilin, Mequin: Atlantic) AS 4 mg/kg once daily for 3 days MQ 8 mg/kg once daily for 3 days All doses supervized
Outcomes	Cure rate at day 28, all reappearances of parasites presumed to be recrudescences as patients hospitalized for duration Adverse events Not included in this review: Fever clearance time Parasite clearance time
Notes	Country: Thailand Setting: Bangkok Hospital for Tropical Diseases Transmission: Low transmission Resistance: Multiple‐drug resistance Dates: Jul 2002 to Apr 2003 Funding: Mahidol University, Tak Malaria Initiative Project, supported by Bill and Melinda Gates Foundation, Wellcome Trust of Great Britain
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	'The randomisation was computer generated (STATA; version 7; Statacorp)'. Randomized in blocks of 6
Allocation concealment?	Low risk	'The treatment allocation was concealed in sealed envelopes labelled with the study code'
Blinding?   All outcomes	High risk	'Laboratory staff reading the blood smears had no knowledge of the treatment received'. No other blinding described
Incomplete outcome data addressed?   All outcomes	Low risk	Similar loss to follow up in all groups (10.6% DHA‐P vs 11.9% AS+MQ)
Free of selective reporting?	Low risk	The WHO recommends 63 days follow up in studies of AS+MQ. Day 28 outcomes are likely to underestimate treatment failure with AS+MQ and DHA‐P.
Free of other bias?	Low risk	No other sources of bias identified

Methods	Trial design: A randomized controlled trial Follow up: Temperature and blood smears daily until clearance of fever and parasites, then weekly attendance until day 63 Adverse event monitoring: Adverse events defined as signs or symptoms that occurred or became more severe after treatment started. A subset of 55 patients in the DHA‐P group had full blood counts, urea, electrolyte, creatinine and liver function tests at days 0 and 7. 32 patients from the DHA‐P group also had ECG monitoring before and after treatment.
Participants	Number: 355 randomized into included treatment arms Inclusion criteria: Age 1 to 65 yrs, symptomatic P. falciparum parasitaemia, informed consent Exclusion criteria: Pregnancy or lactation, signs or symptoms of severe malaria, > 4% of red blood cells parasitized, contraindication to mefloquine, treatment with mefloquine in the previous 60 days
Interventions	1. Dihydroartemisinin‐piperaquine, fixed dose combination (Artekin: Holleykin) Total dose: 6 mg/kg DHA and 48 mg/kg P in 4 divided doses at 0, 8, 24, and 48 hours 2. Artesunate plus mefloquine, loose combination (Artesunate: Guilin, Mequin: Atlantic) AS 4 mg/kg once daily for 3 days MQ 8 mg/kg once daily for 3 days All doses supervized
Outcomes	Cure rate at day 63, PCR adjusted and unadjusted P. vivax during follow up, and mean time to reappearance Gametocyte development during follow up Mean haematocrit at days 0 and 7 Adverse events Not included in this review: Fever clearance time Parasite clearance time
Notes	Country: Thailand Setting: 4 clinics on the Thai‐Myanmar border Transmission: Unstable low and seasonal transmission Resistance: Multiple‐drug resistance Dates: Jul 2002 to Apr 2003 Funding: Wellcome Trust of Great Britain
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	'The randomisation was computer generated (STATA; version 7; Statacorp)'. Randomized in blocks of 9.
Allocation concealment?	Low risk	'The treatment allocation was concealed in sealed envelopes labelled with the study code'
Blinding?   All outcomes	High risk	'Laboratory staff reading the blood smears had no knowledge of the treatment received'. No other blinding described.
Incomplete outcome data addressed?   All outcomes	Low risk	Similar losses to follow up in all groups (12.8% DHA‐P vs 13.6% AS+MQ)
Free of selective reporting?	Low risk	All WHO outcomes reported
Free of other bias?	Low risk	No other sources of bias identified

Methods	Trial design: An open label randomized controlled trial Follow up: Temperature and blood smears daily until clearance of fever and parasites, then weekly attendance for clinical examination, symptom enquiry, malaria smear, and haematocrit until day 63 Adverse event monitoring: Adverse events were actively screened at each visit. Adverse events were defined as signs or symptoms that occurred or became more severe after treatment started.
Participants	Number: 500 randomized Inclusion criteria: Age 6 months to 65 yrs, weight > 5 kg, symptomatic P. falciparum mono‐infection or mixed infections, informed consent Exclusion criteria: Pregnancy or lactation, signs or symptoms of severe malaria, > 4% of red blood cells parasitized, treatment with mefloquine in the previous 60 days, contraindication to mefloquine
Interventions	1. Artesunate plus mefloquine, fixed‐dose combination, adult tablets 100 mg/220 mg, paediatric tablets 25 mg/55 mg (Far‐Manguinhos) 5 to 8 kg 1 paediatric tablet per day 9 to 17 kg 2 paediatric tablets per day 18 to 29 kg 1 adult tablet per day > 30 kg 2 adult tablets per day 2. Artesunate plus mefloquine, loose combination, (Arsumax: Sanofi‐Synthelabo, Lariam: Roche) AS 4 mg/kg once daily for 3 days MQ 15 mg/kg on day 1 and 10 mg/kg on day 2 All doses supervized
Outcomes	Cure rate at day 63, PCR adjusted and unadjusted P. vivax during follow up, and median time to reappearance Gametocyte development during follow up Mean haematocrit during follow up Adverse events Not included in this review: Fever clearance Parasite clearance
Notes	Country: Thailand Setting: 6 clinics on the Thai‐Myanmar border Transmission: Unstable low and seasonal transmission Resistance: Multiple‐drug resistance Dates: Nov 2004 to Jun 2005 Funding: DNDi, European Union International Co‐operation programme, Médecins sans Frontières, WHO/TDR, Wellcome Trust of Great Britain
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	'Randomised in blocks of 10 by a statistician using a computer‐generated randomisation'
Allocation concealment?	Low risk	'The treatment allocation was concealed in numbered, sealed envelopes...opened only after enrolment in the study'
Blinding?   All outcomes	High risk	An open label study. '50% of enrolment slides, 10% of follow up slides and all slides reported as showing recrudescence were subjected to a second blind reading'
Incomplete outcome data addressed?   All outcomes	High risk	Losses to follow‐up are moderate (15.5% FDC vs 15.3% loose). Reasons are not clearly stated and some losses may represent early treatment failures.
Free of selective reporting?	Low risk	All WHO outcomes reported
Free of other bias?	Low risk	No other sources of bias identified