SciELO - Scientific Electronic Library Online

 
vol.18 issue1The independent effect of central obesity on hypertension in adults living in Ribeirão Preto, SP, 2007. EPIDCV ProjectSynovitis and tenosynovitis in Brazil: analysis of sickness benefit claims author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Revista Brasileira de Epidemiologia

On-line version ISSN 1980-5497

Rev. bras. epidemiol. vol.18 no.1 São Paulo Jan./Mar. 2015

http://dx.doi.org/10.1590/1980-5497201500010014 

Original Articles

Primaquine and relapses of Plasmodium vivax. Meta analysis of controlled clinical trials

Jaime Carmona FonsecaI 

IGrupo Salud y Comunidad César Uribe Piedrahita, Universidad de Antioquia - Medellín (Antioquia), Colombia

ABSTRACT

BACKGROUND:

Primaquine (PQ) is used against relapses of vivax malaria (RVM) but several aspects about dosage are unknown, as the total effective dose (TD) in a number of days.

OBJECTIVE:

To compare PQ regimens against RVM in randomized or non randomized controlled clinical trials (CCTs).

METHODOLOGY:

Meta analysis. Information was sought until 31 December, 2012 in Lilacs, SciELO, PubMed (Medline), Cochrane Library, Cochrane Infectious Diseases Group, Embase. Experimental studies or CCT were used, always with concurrent control group. No matter whether or not the design was randomized, close label, supervised. It is not required that the study established difference between relapse and reinfection by molecular evidence. Inclusion and exclusion criteria for articles were applied and meet the inclusion criteria constituted adequate quality to be left in the meta analysis.

RESULTS:

23 ECC with or without random allocation of treatment met the selection criteria. Include four schemes of TD (TD mg number of days): 210 14 = 210 mg in 14 days; 210 7 = 210 mg in 7 days, 45 to 150 mg in 3 to 10 days, 0 (not PQ). If PQ is absent, recurrences occurr 34.48% versus 19.66% with PQ 210 14 (significant difference), 210 14 showed effectiveness equal to that of 210 7. Treatments 210 7 and 210 14 were statistically better than 45 to 150 effectiveness.

CONCLUSIONS:

The use of PQ is necessary to reduce recurrences and TD 210 mg given at 7 or 14 days is which is more effective but more studies are required to treatment 210 7.

Key words: Primaquine; Malaria; Plasmodium vivax; Recurrence; Effectiveness; Meta Analysis

INTRODUCTION

A unique feature of malaria by Plasmodium vivax and by Plasmodium ovale is the reappearance and recurrence of the parasitemia, with or without associated symptoms, weeks after having cured the disease with the total elimination of the blood parasites and the symptoms.

In cases of malaria by P. vivax, the malarial recurrence includes the resurgence and relapse and both of them are a result of the current plasmodial infection and they should be distinguish from reinfection (resulted from a new infection). The relapse is due to an asexual parasitemia resulted from liver hypnozoites of P. vivax and which occurs 4 to 5 weeks (usually for a limit of 4 weeks or 28 days) after the beginning of the treatment. Reinfection is an infection by another parasite clone that happens at any moment after the beginning of the treatment for the primary attack and it can only be proven through genetic analysis of the parasites, which is not yet available for the routine1.

Primaquine (PQ) is the only antimalarial available in the routine to be administered to those suffering from vivax or ovale malaria with the objective of avoiding recurrences1. Belonging to the group of the 8 aminoquinolines (8 AQ). There are many derivatives of 8 AQ with methoxy group in position 6 and substitutions in the side chain (radical) of position eight; the basic structure is the same with a change in the radical. It is the only 8 AQ in daily use to destroy the dormant liver stage parasites (hypnozoites) of P. vivax, against which it reaches maximum activity (+++). There is a slight effect (+) against the active liver stage parasites (trophozoites, schizonts, merozoites) of P. vivax and P. falciparum. It removes gametocytes and sporozoites of P. vivax and P. falciparum, with maximum power (+++). It has a very slight action on erythrocytic asexual forms of P. vivax (+, ) and null (0) effect on the P. falciparum. The curative action (elimination of dormant liver stage parasites DLSP: hypnozoites) depends on:

  • an amine group in position eight of the quinoline nucleus;

  • a methoxy group in position 6.

The alkylation of the amine group in position eight with a lateral chain of five carbons and a terminal amine confers it with great power1 6.

PQ seems more efficient when it occurs simultaneously with the blood schozonticides2 and, from all antimalarials, this is the one which has the wider activity spectrum3. The PQ has proven highly useful for simple primary and secondary chemoprophylaxis4 or against recurrences5 8.

The total dosis (TD) of 210 mg given in 14 days (210 14), for an adult weighting 60 kg (0.25 mg/kg/day for 14 days), is recommended by the WHO as the stardard in order to avoid recurrences7. Blocking the transmission of P. vivax through the use of drugs is achieved with the usual PQ dose, which produces, in 4 to 20 hours, the complete loss of transmission to the anopheline8.

The clinical and parasitary cure, which include avoiding recurrences of the infection by P. cynomolgi and P. vivax is the role of the TD of the 8 AQ9,10.

There are several recent reviews on the use of PQ to avoid the recurrences3,11,12, as well as meta analysis on the theme5,6,13 15. The 2 meta analysis of Cochrane6,13 assessed random and semi random controlled clinical trials (CCT) in humans who compared treatments with chloroquine only against chloroquine added with PQ, given as 210 14, and other PQ schemes. The meta analysis by John et al.15 included 87 studies published since 1950 under regimens of PQ anti reccurences and they were based on what was found in Medline and Cochrane Central Register of Controlled Trials. The meta analysis by Carmona Fonseca5 was based exclusively on descriptive studies. Then why to conduce this study? Because, regarding the review by Cochrane, we found 23 studies instead of 9; we analyzed 5 treatment schemes instead of 3 (0: zero mg of PQ; 75 5: 75 mg of PQ distributed into 5 days; 210 14: 210 mg of PQ distributed into 14 days); we worked with 6,350 patients instead of 3,423. In relation to the meta analysis made by John et al.15, it included segments of < 30 days (that measure the response of the acute attack); it included 7 studies without schizonticide, the TD categories used include one which contains the standard TD used in the standard procedure (14 days) or in 7 days, along with higher TD such as 315 14 (315 mg of PQ distributed into 14 days), 420 14 (420 mg of PQ distributed into 14 days), etc, and it does not discriminate the results among these values; because it includes Prospective and Retrospective studies; because it includes many studies without control group, it is to say Descriptive and solo CCT studies, without differentiating the recurrences in relation to the kind of study.

The objective of this review was to compare the effectiveness of PQ in avoiding the recurrences of malaria vivax, according to which it is administered with different dosing schemes (mg/kg-days of use) and according to which random and non random CCTs are shown.

MATERIALS AND METHODS

STUDY CLASS AND SPECIFIC QUESTIONS TO SOLVE WITH THE META ANALYSIS

A methodology was used for the meta analysis16. The questions to be answered were:

  • Is PQ effective in preventing the recurrences of vivax malaria?

  • Does the effectiveness of PQ in preventing the recurrences of vivax malaria depend on the applied TD when given a fixed time?

  • Does the effectiveness of PQ in preventing the recurrences of vivax malaria depend on the time (number of days) in which a given TD is effective?

  • Does the effectiveness of PQ in preventing recurrences of vivax malaria depend on the place where the infection occurs?

INFORMATION SOURCES, KEYWORDS, DATES, LANGUAGES AND PATIENTS

The information was searched for in Lilacs, SciELO, PubMed (Medline), Cochrane Library, Cochrane Infectious Diseases Group, Embase. We searched for information up to December 31, 2012, in every language. The bibliography of all written literature was used in order to search for articles.

For the search in PubMed, the following terms restricted to major topics were used (Restrict to MeSH Major Topic): "primaquine" and "malaria, vivax"; the filter "clinical trial" was used; the articles found were reviewed and the ones which were not CCT, comparing different PQ schemes in sick people, were discarded. Also searched in PubMed:

  • review articles with "primaquine, malaria vivax, recurrences, human";

  • meta analysis (primaquine, malaria vivax, recurrences, human).

The search in Embase used the words primaquine, malaria and vivax. In SciELO, SciELO Public Health and SciELO Brazil, Peru, Venezuela and Colombia the word "primaquina" was used. In Lilacs, the search was made with "primaquina", "vivax" and "recurrencia". In Cochrane Library and in Cochrane Infectious Diseases Group the search was with "primaquina".

In Google Académico the words "primaquina", "recurrencia", "vivax" were searched.

INCLUSION/EXCLUSIÓN CRITERIA OF THE ARTICLES; ASSESSMENT OF THE REPORTS

The design had to be experimental or CCT, always with a control group. It did not matter if the design was random, closed/blind, supervised. Every article had to inform with clarity and precision these data in order to be considered quality appropriate:

  • number of included patients, without demanding minimum sample size;

  • which treatment groups were used;

  • size of each group;

  • PQ doses administered and in how many days, having been used as the minimum of the schemes or different doses;

  • the minimum follow up time, after day 1 of the treatment of the acute malarial episode, was 60 days;

  • number, proportion or rate of recurrence in each group;

  • informing if the subjects were in malarial transmission areas right after receiving treatment;

  • establishing the diagnosis of malaria with thick smear (microscopy), with polymerase chain reaction or any laboratorial diagnostic test and informing that the species was P. vivax only, no matter the parasitemia found; i) receiving any schizontocide (usually chloroquine) in addition to PQ;

  • it was not required that the study would differ relapse from reinfection through molecular testing.

The extraction of data from each article was conducted by the author who used an Excel chart with the ten dates previously stated.

STATISTICAL ANALYSIS

The incidence of recurrences was calculated according to the protocol analysis17. The comparison of proportions was analyzed with the χ² test of Mantel Hanszel (χ2M H) of the Epitable module of the EpinInfo 6.0 software. The SPSS 10.0 software was used in order to compare measures of central tendencies between groups, using the nonparametric test of Kruskal Wallis (K W) for independent groups.

The meta analysis was made with the program of Joaquín Primo16, with the random effects model.

It was considered as statistically significant a probability of p < 0.05.

ETHICAL ASPECTS

The investigation project did not require ethical endorsement.

RESULTS

In PubMed, 33 writings were collected ("Primaquine"[Majr]) AND "Malaria, Vivax"[Majr] and "clinical trial" as filter), but only 14 of them were actually CCT on the comparison of PQ doses in treated patients with vivax malaria. In Embase, the search with the bibliographic list of other publications allowed us to find 12 other studies, but theses writings did not refer to the theme, or they would not fulfill the inclusion criteria or were new in relation to PubMed. In SciELO Public Health (Subject: Ciencias de la salud (List of journal by subjects) it was searched in "All indexes" and "Entire collection" and 5 articles were found, but none of them were CCT. In SciELO Brazil Peru, Venezuela and Colombia no references on CCT of PQ were found. In Lilacs, four references were found, though only one was CCT. In Cochrane Library there were six references and only two on recurrences of vivax malaria6,13. In Cochrane Infectious Diseases Group, we found the same as in Cochrane Library. In PubMed we found review articles and meta analysis ones; in both cases, many references were obtained5,6,13 15. The search in Scholar Google yielded 94 results; the titles in the writings were compared to the ones obtained in all other sources, but it resulted in no new reference which would meet the inclusion criteria.

In short, there were 23 CCTs with or without random treatment allocation, which met the selection criteria18 40: from PubMed 14 references were included, from Lilacs 1 and with the system's search procedures (search in physical archives of the Biblioteca Médica Universidad de Antioquia, n = 4, articles from the author's library and the ones obtained before with other objectives, n = 4) 8 were obtained. The studies were published in a lapse of 34 years, between 1978 and 2012: 23% between 1978 and 1999; 58% between 2000 and 2009; 19% between 2010 and 2012. The 23 studies assessed a total of 12 schemes (TD days in which they were applied) of PQ.

The PQ schemes may be analyzed under different perspectives:

  • according to the daily dose used (mg/kg/day) during a certain number of days (1 to 14 days);

  • according to the TD given in a certain number of days, i.e., the daily doses administered in "X" number of days;

  • a given amount of mg/kg/day once/week for a determined number of weeks.

The relations between daily doses (mg/kg/day) and TD in "X" days are:

  • 0.25 mg/kg/day for 14 days: it the standard procedure, here called 210 14, referring to what in an adult weighting 60 kg corresponds to 210 mg of TD in 14 days;

  • 0.50 mg/kg/day for 7 days: is the standard TD but administered in 7 days; scheme 210 7;

  • 1.17 mg/mg/day for 3 days: the standard TD given in 3 days; scheme 210 3;

  • 0.25 mg/kg/day for 3, 5, 7, 9 or 10 days: each one corresponds to the same daily doses of the standard scheme but with lower TD tan the standard one, supplying between 45 and 150 mg of TD in 3 to 10 days (including here a study which administered 45 mg in a single dose (meaning, in 1 day); scheme 45 to 150 (in 3 to 10), noting that here it is included a study which gave 150 in 5 days, i.e., 30mg per day;

  • 0.75 mg/kg/day for 1 day a week and for 8 weeks26 (a report): equivalent to 360 mg of TD, which is 71% higher than the standard TD (210 mg) but with very low daily doses [6,4 mg/day instead of 15 mg/day in an adult of 60 kg]); is included in the group 210 14;

  • 0.375 mg/kg/day for 14 days: 0.375 mg x 60 kg x 14 days = 315 mg of TD. Equivalent to 1.5 times (50% more) the standard TD;

  • 0.50 mg/kg/day for 14 days: 0.50 mg x 60 kg x 14 = 420 mg of TD. Corresponding to twice the standard TD;

  • 0.25 mg/kg/day for 3 to 10 days: 0.25 mg x 60 = 15 mg x 3 days = 45 mg of TD; if they are 4, 5, etc. days, corresponding to 60, 75, etc. mg of TD. Here we include a study which administered 150 mg of TD in 5 days, i.e., 30 mg/day instead of the standard 15 mg/day. They are all lower TDs than the standard ones. The TD 150 10, 150 5, 135 9, 105 7, 75 5 and 45 3 group in a category called 45 to 150 in 3 to 10 days;

  • did not use PQ: TD 0 mg.

We proceeded into grouping the studies according to their objectives (to compare the anti recurrence effectiveness of TDs of PQ given a fixed amount of time). The result were these 4 schemes of treatment in relation to the TD of PQ and the days in which this TD was used TD: 210 mg in 14 days; 210 mg in 7 days; 45 to 150 mg in 3 to 10 days; and 0 mg (not used). Among these 4 schemes, 5 comparisons are made:

  • 210 14 versus 210 7;

  • 210 14 versus 45 to 150 3 to 10;

  • 210 14 versus 0;

  • 210 7 versus 45 to 150 3 to 10; y

  • 45 to 150 3 to 10 versus 0.

The scheme 210 3 was observed in 2 studies22 23 and are included in the general analysis but not in the meta analysis because each one of the studies do not have a pair to compare to. The studies with TD of 315 (1 study) and 420 (2 studies) given in 14 days group up with the ones of 210 in 14 days.

The number of patients per PA scheme was as follows: with 210 14 there were 1,587 patients; with 210 7 there were 232; with 210 3 there were 101; with 45 to 150 in 3 to 10 days there were 2,380; with 0 (without PQ) there were 2,151 patients. In total, 6,451 patients were evaluated.

Children and adults were evaluated and dosed according to their body weight. The standard dose corresponds to 0.25 mg/day for a weight of 60 kg, implying in 15 mg/day. When it comes to children, the amount is lower, but in this work, they are similar to the ones of adults when using the same expected TD. Thus, expressions such as 210x14 refer to studies with children and adults who have always had the standard TD based in 0.25 mg/kg/day.

The total number of patients included in the analysis of the 23 articles is 6,451, divided into: 1587 in 210 14, 232 in 210 7, 101 in 210 3, 2380 with 45 to 150 and 2151 in PQ = 0.

The studies were carried out in 12 different countries, from 3 continents: America: 7 countries (Brazil, Colombia, Costa Rica, El Salvador, Mexico, Nicaragua, Peru), Africa: 1 country (Ethiopia), Asia: 4 countries (India, Pakistan, Thailand, China) (Table 1).

Table 1. Clinical controled studies included in the review sboute the effectiveness of primaquine regimens in order to prevent recurrences of malaria by P. vivax. 

Author, year of publication Country of the study Months of follow-up PQ scheme (TD‑days)*
210‑14 210‑7 210‑3 45 to 150-3 to 10 0
Da Silva, 2003 Brazil 6 91 97
Abdon, 2001 Brazil (Belem, Para) 6 40 39 40
Pinto, 1998 Brazil (Para) 3 39 105
Villalobos, 2000 Brazil (Rondonia) 2 31 30
Solari, 2000 Peru 2 28 28
Álvarez, 2006 Colombia 6 64 127
Carmona, 2009 Colombia 4 66 63**
Carmona, 2010 Colombia 4 35 38**
Bergonzolli, 2000 C. Rica/Nicaragua 6 32 95
Cedillos, 2012 El Salvador 7.5 147 144
Gomez, 1965 Mexico 9 363 389
Yeshiwondim, 2010 Ethiopia 5 132 108
Adak, 2001 India 12 220 224
Gogtay, 1999 India (Bombay) 6 63 62 60
Rajgor, 2003 India (Bombay) 6 103 101
Kim, 2012 India (Calculta) 15 42 59 50
Yadav, 2002 India (Orissa) 12 759 723
Leslie, 2008 Pakistan 11 55 71
Leslie, 2004 Pakistan (Afghan) 9 383 212
Rowland, 1999 Pakistan (Afghan) 12 100 250 350
Pukrittayakamee, 1994 Thailand (Bangkok) 3 18 21
Walsh, 2004 Thailand (Bangkok) 2 12 10
Li, 1999 China (Yunnan) 9 55 77
Total 6.8# 1587 232 101 2380 2151

*In Table 2 there are information on the expossed and the recurrences in each treatment, in each country;

**These were reviewed but they did not enter the meta-analysis' calculations because each one of them does not have a pair to compare to;

#Mean follow-up: 6.8 months (median: 6.0). Minimum: 2, maximum: 15 months.

The follow up time varied from 2 and 15 months, with an average of 6.7 months. The follow up with less than 4 months was observed in 24% of the evaluations; the 4 month one in 8% of them, the one of 5 months in 4%, 6 months in 28%, 8 months in 4%, 9 months in 12%, 11 months in 4%, 12 months in 12% and 15 months in 4%. It is possible that "long" length follow ups, when remaining in endemic areas, may have generated new infections, which did not differ from true relapses.

There is a significant difference in the follow up months in relation to the place (grouped in 5 places: East Africa, Asia, Central America Mexico, India Pakistan, South America): (p(Kruskal Wallis) = 0.007): the shortest follow up is in Asia (4.13 months) and the longer one in India Pakistan with 10.38 months.

For the 210 14 scheme, the proportion of recurrences is "high" with 4 months (19.00%) or with 9 months (17.11%) of follow up and it is "low" with 5 months (3.03%) or with 11 months (1.82%), with 2 months it is 12.86% and with 3 months it is 9.57%. There is no correlation between the percentage of recurrences and the months of follow up: rho = 0.066; p = 0.788). Something similar happens with TD 0 of PQ, but TD 45 to 150 shows evident stability in the proportion of recurrences in relation to time of follow up: the proportion varies between 16.19 and 28.16% for 2 to 15 months of follow up, but, also, if the value 16.19% is discarded, the range is between 21.68 and 28.16% for 6 to 15 months.

RECURRENCE ACCORDING TO THE TREATMENTS

The comparison by meta analysis (model of random effects) of the 5 treatment schemes of treatment with PQ (Table 2; Figure 1) shows the following:

Table 2. Five comparisons of four schemes of primaquine administration in controlled clinical trials. 

Country Treatment‑1 Treatment‑2 Weight (%) RR** 95%CI p-value
Recurrences Exposed Recurrences Exposed
210‑14a 5.88%* 210‑7b 5.97%*
Peru31 2 28 3 28 54.5 0.67 0.12 – 3.69 0.65692
Brazil18 2 40 1 39 65.5 1.95 0.18 – 20.64 0.58564
Total 4 68 4 67 RRdl 0.96 0.24 – 3.86 0.95976
210‑14a 11.58%* 45 to 150-3 to 10c 23.10%*
India25 1 63 16 62 13.0 0.06 0.01 – 0.45 0.00683
India39 16 42 16 59 26.3 1.40 0.80 – 2.48 0.32276
Brazil32 2 31 8 30 17.4 0.24 0.06 – 1.05 0.07290
Brazil18 2 40 8 40 17.2 0.25 0.06 – 1.11 0.07951
Colombia20 12 68 26 71 26.1 0.48 0.27 – 0.88 0.03598
C. Rica‑Nicaragua21 4 32 4 31 15.8 0.97 0.27 – 3.54 0.96422
Mexico 36 37 363 80 389 23.2 0.50 0.34 – 0.71 0.00040
Total 74 639 158 682 RRdl 0.50 0.28 – 0.90 0.01953
210‑14a 19.66%* 0d 34.48%*
India25 1 63 7 60 6.6 0.14 0.02 – 1.07 0.06215
India39 16 42 15 50 13.9 1.27 0.72 – 2.25 0.49985
India40 6 103 13 101 12.1 0.45 0.18 – 1.14 0.10834
Pakistan30 32 100 178 350 14.4 0.63 0.46 – 0.85 0.00721
Pakistan27 74 100 86 212 15.1 1.82 1.49 – 2.23 0.00002
Pakistan26 1 55 22 71 7.0 0.06 0.01 – 0.42 0.00546
Thailand33 3 25 8 10 11.1 0.15 0.05 – 0.45 0.00841
Thailand29 2 87 4 30 8.4 0.17 0.03 – 0.89 0.04541
Ethiopia38 4 132 9 108 10.9 0.36 0.12 – 1.15 0.09306
Total 139 707 342 992 RRdl 0.43 0.21 – 0.88 0.01995
210‑7b 6.19%* 45 to 150-3 to 10c 18.01%*
Brazil24 8 116 22 116 63.6 0.36 0.17 – 0.78 0.01428
Brazil18 1 39 8 40 9.1 0.13 0.02 – 0.98 0.05272
Brazil28 3 39 17 105 27.3 0.48 0.15 – 1.53 0.22490
Total 12 194 47 261 RRdl 0.36 0.19 – 0.66 0.00094
45 to 150-3 to 10c 20.20%* 0d 29.40%*
El Salvador35 35 147 60 144 17.9 0.57 0.40 – 0.81 0.00861
Pakistan30 128 250 178 350 20.3 1.01 0.86 – 1.18 0.95167
India25 16 62 7 60 10.7 2.21 0.98 – 499 0.08019
India19 59 220 134 224 19.4 0.45 0.35 – 0.57 0.00000
India34 49 759 62 723 17.7 0.75 0.52 – 1.08 0.13782
India39 16 59 15 50 13.9 0.90 0.50 – 1.64 0.78124
Total 303 1457 456 1551 RRdl 0.79 0.54 – 1.16 0.22791

a0.25 mg/kg/day for 14 days;

b0.50 mg/kg/day for 7 days;

c0.25 mg/kg/day for 3, 5, 10 days;

d0 mg/kg/day for 0 days;

*Percentage of recurrenre in each PQ scheme: (total Recurrences/total Exposed)x100;

**RR: relative risk = (%recurrences Treatment-1)/(%recurrences Treatment-2): RR < 1, Treatment-1 has less recurrences than Treatment-2; RR > 1, the contrary; RR = 1, both treatments have the same recurrences. RRdl: relative risk for DerSimonian-Laird; 95%CI: 95% confidence interval.

Figure 1. Forest’s Graphics (random effects model). 

  • the schemes 210 14 and 210 7 (2 studies) have equal anti recurrence effectiveness, with 5.88 and 5.97% of recurrences in each one of them, in this same order (p = 0.960 for RR of DerSimonian Laird);

  • the schemes 210 14 and 45 to 150 in 3 to 10 days (7 studies) allow very different recurrence proportions: 13.52 and 28.24% (reason: 2.1), respectively, and there is a significant statistical difference between them (p = 0.019);

  • the schemes 210 7 and 45 to 150 (3 studies) cause 6.19 and 18.01% of recurrence (reason: 2.91), with significant difference (p = 0.001);

  • the schemes 210 14 and 0 (9 studies) associate to recurrence of 19.66 and 34.48%, in this order (reason: 1.4), with no significant difference (p = 0.020);

  • the schemes 45 to 150 and 0 (5 studies) produce 20.20 and 29.40% recurrences (reason: 1.5), without significant difference (p = 0.228).

In short:

  • it is not completely strange that 210 14 and 210 7 are similar, since they have the same TD though with very different number of days;

  • the RR is statistically significant in the cases of schemes 210 14 versus 45 to 150 (p = 0.019), 210 0 (p = 0.020) and 210 7 versus 45 to 150 (p = 0.001). It should be noted that when controlling the heterogeneity of the comparison 210 14 versus 45 to 150 (see below), a significant difference between the 2 schemes is kept, always in favor of 210 14;

  • the absence of difference between 45 to 150 versus 0 exists for before and after controlling heterogeneity (see below).

It is observed, finally, that the standard TD 210 mg in 3 days23 lacked utility (53.85%) when compared to this same doses given in 7 or 14 days (14.71%), although the high daily doses were well tolerated.

The influence of the place (country) was evaluated as follows: within each PQ scheme all studies of each country are grouped and the proportions of the countries compared (Table 3). Within each one of the schemes using PQ (210 14, 201 7, 45 to 150) there is a significant difference due to the place (country) in the proportion of presented recurrence in 2 months of follow up. Also, if PQ is not used, there is a significant difference between the proportions of recurrence by country.

Table 3. Proportion of recurrence by place (country) under the same primaquine scheme. 

Position Country Exposed Recurrences (%) Reason:
Maximum/Country*
RR**
210‑14
1 Brazil 71 5.63 7.45 0.1343
2 Peru 28 7.14 5.87 0.1702
3 Colombia 124 17.74 2.37 0.4228
4 C. Rica/Nicaragua 32 12.5 3.36 0.2979
5 Mexico 363 10.19 4.12 0.2429
6 Ethiopia 132 3.03 13.85 0.0722
7 India 208 11.06 3.79 0.2635
8 Pakistan 255 41.96 1 1
9 Thailand 112 4.46 9.41 0.1063
p(χ²) = 0.000001 1325 15.70# 2.67 0.3741
210‑7
1 Brazil 222 6.76 3.38 0.2957
2 Colombia 35 22.86 1.00 1
p(χ²) = 0.00539 257 8.95# 2.55 0.3915
45 to 150-3 to 10
1 Brazil 291 18.9 2.7 0.3697
2 Colombia 71 36.62 1.4 0.7163
3 C. Rica/Nicaragua 31 12.9 3.96 0.2524
4 El Salvador 147 23.81 2.15 0.4658
5 Mexico 389 20.57 2.49 0.4023
6 India 1100 12.73 4.02 0.2490
7 Pakistan 250 51.12 1 1
p(χ²) = 0.000001 2279 20.54# 1.78 0.4017
PQ = 0
1 El Salvador 144 41.67 1.08 0.9222
2 Ethiopia 108 8.33 5.42 0.1844
3 India 1148 20.12 2.25 0.4454
4 Pakistan 633 45.18 1 1
5 Thailand 40 30.00 1.51 0.6640
p(χ²) = 0.000001 2073 28.85# 1.57 0.6385

*Indicates how many times the proportionof recurrences of a country is contained in the maximum value observed for the countries.

**RR: relative risk = the proportion of recurrences in a country with higher proportion (maximum), taken as not exposed to PQ, divided by the proportion of each country. It indicates the lowest risk there is in each country in relation to the maximum (always, the RR value of a specific country is lower than 1).

#The values of these lines are mean measured for this scheme.

Another important aspect related to the place (country) and the PQ schemes is that, in each scheme, the proportion of recurrences varies notoriously, even if PQ is not used. In 210 14 the lower proportion is observed in Ethiopia (3.03%) and Thailand (4.46%) and the higher one is in Pakistan (41.96%), which is a value at around 13 times the one in Ethiopia and Thailand. It should be noticed that in Brazil, Peru and Mexico there is a lower proportion than in India, a country usually pointed out as one of very few recurrences, while Colombia is the country in America with the highest recurrence proportion under the 210 14 scheme, which is standard. What happened in Pakistan is completely out of what was expected for any of the sites. In Thailand, the risk of recurrence is 3 times lower than in Colombia (17.74 versus 4.46%: 17.74/4.46 = 4.00), whereas it is 0.26 times lower than in Brazil and 1.5 times lower than in India.

The scheme 45 to 150 (including 75 5), the official one in India, presents the lowest values in this country (12.73%) and in Costa Rica and Nicaragua (12.90%), with 51.12% for Pakistan. The value 12,73% for India with 45 to 150 is very similar to 11.06% with 210 14. With this scheme, the recurrence in India is the same as in Costa Rica and Nicaragua, and it is 0.5 times lower than in Brazil.

If PQ is not used (PQ = 0), the lowest proportion of recurrences is found in Ethiopia (8.33%) and the highest, once again, in Pakistan (45.18%).

India has 5 studies and their analysis in relation to the 3 schemes used it the following: 210 14 has recurrences of 10.58% (22/208); 75 5 of 12.73% (140/1100) and 0 of 19.95% (231/1158) (p = 0.000). The comparison of these proportions by pairs indicates that 210 14 does not differ from 75 5 (p(χ²) = 0.388) but it does from PQ = 0 (p(χ²) = 0.003); also, 75 5 differs from 0 (p(χ²) = 0.000). In short, in India, the three schemes behave differently, being similar in anti recurrence effectiveness 210 14 and 75 0 and both of them are better than not using PQ.

HETEROGENEITY STUDY

• The evaluation of the heterogeneity of the studies indicates that the groups arranged for comparison have studies with little heterogeneity in the case of 210 14 versus 210 7 and 210 7 versus 45 to 150, though it is high in other cases: 210 14 versus 45 to 150; 210 14 versus 0; 45 to 150 versus 0. Thus, its conformation seems adequate in 2 comparisons and inadequate in 3 of those (Table 4). The problem of high heterogeneity in these comparisons is solved by eliminating one or 2 studies from each group, that is making a subgroup analysis which only combines studies that fulfill certain conditions or characteristics, in a way that those are more homogeneous, so: 210 14 versus 45 to 150: eliminates the study by Kim39, in India, the only one in 7 to demonstrate RR higher than 1.00. The RR becomes significant for the comparison of 6 studies (RR = 0.43; 95%CI 0.28 - 0.66; p = 0.000);

Table 4. Assessment on the heterogeneity of the studies (according to Primo16). 

Comparison
Q k p(Q) H I2 (%) 95%CI (I2) I2 H
√[Q/(k‑1)] 100%[Q ‑ (k‑1)]/Q (H2–1)/H2f
210‑14 versus 210‑7 0.52 2 0.46985 0.7211 91 just 1 gl 0.99 Low
210‑14 versus 45 to150-3 to 10 19.3 7 0.00365 17.944 69 32 – 86 0.99 High*
210‑14 versus 0 88.8 9 0.00000 33.185 91 85 – 95 0.99 High*
210‑7 versus 45 to150-3 to 10 1.22 3 0.54265 0.7810 -64 0 – 83 1.02 Low
45 to150-3 to 10 versus 0 39.9 6 0.00000 28.238 87 75 – 94 0.99 High*

Q of Cochran = ?² with k-1 dergees of freedom (gl); k: number of studies included in the analysis; p(Q): probability associated to Q; H of DerSimonian-Laird = v[Q/(k-1)] ; I2 = 100%[Q-(k-1)]/Q; Relation between H and I2 = (H2 - 1)/H2; H: Heterogeneity (H < 1.2: low; H = 1.2 to 1.5: moderated; H > 1.5: high);

*The heterogeneity is eliminated by excluding 1-2 studies in each group, so: 210-14 versus 45 to 150-3 to 10: eliminates the study by Kim39. Keeps signifficant (RR = 0.43; 95%CI 0.28 - 0.66; p = 0.00013) - now, H = 1.19; 210-14 versus 0: eliminates the studies by Kim39 and Leslie27. Becomes more signifficant (RR = 0.28; 95%CI 0.15 - 0.55; p = 0.00018) - now, H = 1.61; 45 to 150 versus 0: eliminates the studies by Gogtay25, Rowland30 and Adak19. Keeps signifficant (RR = 0.69; 95%CI 0.54-0.87; p = 0.00231) - Now, H = 1.04.

  • 210 14 versus 0: studies of Kim39 and Leslie27 are eliminated, the first one already commented and the second one carried out in Pakistan; the RR of those were 1.27 and 1.82, while in the other 7 studies the RR was < 0.640. The RR was kept significant for the comparison of 5 studies (RR = 0.24; 95%CI 0.13 - 0.42; p = 0.000);

  • 45 to 150 versus 0: the studies by Gogtay25, Rowland30, in Pakistan, and Adak19, in India, are removed, with extreme RR in the group of 6 studies: 2.21; 1.01 and 0.45; the 3 studies have RR between 0.57 and 0.90 and the RR is kept significant for the comparison of 5 studies (RR = 0.69; 95%CI 0.54 - 0.87; p = 0.002).

QUESTIONS AND ANSWERS ON THE EFFECTIVENESS OF PRIMAQUINE AGAINST RECURRENCES OF P. VIVAX

In a brief way and based in the findings described, the questions asked for this systematic review may be answered as follows:

  • Is PQ effective? Answer: yes. Indeed, the CCT in humans there is a clear benefit when comparing the schemes 210 14 and 201 7 to not using PQ or using it in lower doses than 210, which is the standard TD.

  • Does the effectiveness of PQ depend on the applied TD when given a fixed time? Answer: yes. The comparison of TD in CCT in humans (Table 2) showed that the standard TD of 210, given in 7 or in 14 days, is associated to recurrence proportions of 8.88% (gross average between 6.19 and 11.58%) against 20.54% to 45 to 150 (gross average between 18.01 and 23.07%) and to 29.01% for TD 0 of PQ (gross average between 27.91 and 30.11%). It is clear that the higher the TD, the lower the recurrences. Yes, outside the meta analysis procedure, we obtained the average of the proportion of recurrences in each treatment, the averages being 13.80 for 210 14 (219/1,587); 9.48% for 210 7 (22/232); 21.18% for 45 to 150 (204/2,380) and 30.68% for PQ = 0 (660/2,151). For 210 3 the average is 64.03% (89/139). The proportion of recurrences in general was 23.02% (1,494 between 6,489 patients), with 30.68% (660/2,151) in patients without PQ and 19.22% in patients with PQ (p(χ²) = 0.000). If we exclude the scheme 210 3 because of its exaggerated recurrences (higher to the group without PQ), it is found that the 4 remaining schemes have significant differences in the proportion of recurrences (p(χ²) = 0.000). The scheme 210 14 does no differ from 210 7 (p(χ²) = 0.070053). The scheme 45 to 150 differs significantly from not using PQ (p(χ²) = 0.000).

It is clear that the higher the TD, the lower the recurrences, with the new information that the time in which the TD is administered seems to influence it, for 210 mg in 3 days have very high recurrence, in 14 days it is 12.75% and the minimum is for 7 days with 7.17%.

  • Does the effectiveness of PQ depend on the number of days in which a given TD is effective? Answer: Yes. The standard TD (0.25 mg/kg/day) given in 7 to 14 days seems to have the same effectiveness, but given in 3 days it is of no use, according to the data informed22 23. It was not possible to clearly capture the influence of time in the follow up of the recurrence proportion.

  • Does the effectiveness of PQ depend on the place where the infection occurs? Answer: Yes. The data indicated a variation in recurrence according to the place: the higher proportion o f recurrences occurs in Pakistan (39.65%) and the lowest one in Ethiopia (5.68%) and it is lower in America (6 countries), with 17.83%, than in Asia (3 countries), with 25.22%.

DISCUSSION

The CCTs analyzed here indicate that the absence of PQ in the treatment led to the recurrence of 34.48 against 19.66% with PQ 210 14, with significant advantage for the use of PQ; also, 210 14 has effectiveness equal to 210 7. From this last schemes, there are very few evaluations with CCT, which need to be urgently increased, since 7 days of treatment against 14 of them is an enormous advantage.

It is very important to recognize that each scheme 210 7 and 210 14 is statistically better than 45 to 150 in 3 to 10 days (including 75 5), which has zero PQ anti recurrence capacity (PQ = 0). But 75 5 in India, and only there, could be useful, without being clear the reason why, because in this country the only significant difference is between 75 5 and 0, while it seems the same to give 210 rather than 75 or 210 rather than 0 mg of TD of PQ.

In the review by Cochrane6, with random CCTs, it was found that 75 5 lacks of preventive effects from the occurrences of vivax when compared with PQ = 0. This review also found that 210 14 produced significantly less recurrences than PQ = 0 and that the direct comparison of 210 14 and 75 5 showed that 210 14 was significantly lower.

The effectiveness of TD 210 given in 3, 5 or 7 days, was assessed, which involves giving out high daily doses of such. In 3 or in 5 days, it was of no use and the recurrences were very high (> 50%), in 4 or in 6 months22 23, but there were no significant toxic effects when the directions were strictly followed when using high daily doses41.

The influence of the place, according to this work, teaches that within each scheme using PQ (210 14, 201 7, 45 to 150) there are significant differences in the proportion of recurrences, and the same happens if PQ is not used. This seems to show that, among other points, the sensitivity of the hypnozoite clones of each country to PQ is very different and that the tendency of hypnozoites recurring also differs. Other variables which may act refer to, for example, the real intake of drugs and the existence of genetic and ethnic factors that are expressed in people with different capability to metabolize drugs.

The interest is in finding that with 210 14, Brazil, Peru and Mexico have lower proportions of recurrence than India, which supports the use of 210 mg in those countries.

The meta analysis on the effectiveness of PQ against vivax recurrences, according to the longitudinal descriptive field studies5, showed that the TD of 210 and 75 mg had the same effectiveness, but 75 mg had only been evaluated in India, where P. vivax seems to be more sensible to PQ than in other places. The meta analysis of the CCTs presented here show that 210 14 is better than 45 to 150 in 3 to 10 days (including TD 75) and that 45 to 150 is similar to using no PQ. The comparison of the questions which guided both analysis teaches the following about the effectiveness of PQ against recurrences of vivax malaria:

  • Is PQ effective? In both studies the answer is affirmative and both demonstrate that when PQ is not used, the risk of recurrence increases significantly.

  • Does the effectiveness of PQ depend on the applied TD when given a fixed time? The answer with descriptive studies showed that 75 mg had the same effectiveness of 210 in 7 or in 14 days, but with CCT the effectiveness of 210 in 7 or 14 days is frankly better than in the case of 45 to 150 mg in 3 to 10 days.

  • Does the effectiveness of PQ depend on the number of days in which a given TD is effective? The TD 210 in 7 or 14 days seems to have the same effectiveness, but given in 3 days it has no use, according to the informed data.

  • Does the effectiveness of PQ depend on the place where the infection occurs? The answer is yes in both studies.

It has already been noticed that the 2 meta analysis by Cochrane6,13 only included random and semi random CCT in humans who compared treatments with chloroquine against chloroquine plus PQ, given as 210 14, and other PQ schemes; which included 9 studies instead of the 23 CCT (randomized and non randomized) which we have added; in Cochrane 3 schemes were analyzed while here we analyzed 5; they worked with 3,423 patients and here there were 6,350. While in theory the CCT can have advantages on the non randomized CCTs, these advantages have to be shown in practice; besides that, many effects of the non randomized units to the treatment groups may be controlled in this analysis. Given the shortage of schemes used, going from 3 to 5 is an important advance of this report and the same happens when going from 3,423 to 6,350 patients. It is convenient to reinforce the similarities of the results between the review of Cochrane6 and the present study: that one found that 75 5, in relation to the use of PQ, lacked of preventive effects on the recurrences of vivax and this one reports that it was found that 210 14 produced significantly less recurrences than PQ = 0 and the comparison 210 14 versus 75 5 showed that 210 14 was significantly better.

A strength of the present study is the high level of patients involved, as well as the relatively high number of studies analyzed, whose quality, when meeting the inclusion criteria previously determined for the acceptance of the study, may be judged as good. However, it should be noted that the quantity of studies gathered in each comparison is low, varying between 2 (210 7 versus 210 14) and 9 (210 14 versus 0). Perhaps the main explanation for this low quantity of studies is due to the hegemony the standard scheme 210 14 has das, despite the questioning and the low interest of researchers to test other procedures.

CONCLUSION

It is concluded that:

  • the use of PQ is necessary to reduce the frequency of recurrences;

  • the standard TD is 0,25 mg/kg/day for 14 days is the best effectiveness in preventing the recurrences of vivax malaria and it should be maintained;

  • studies in different countries, regions (tropical versus subtropical) and continents must be carried out in order to assess the idea resulted from this work, also present in meta analysis of descriptive studies5, that this standard TD given in 7 days has the same effectiveness of the one given out in 14 days, but the studies on the earlier are few and are mostly restricted to Brazil.

REFERENCES

1. Vale N, Moreira R, Gomes P. Primaquine revisited six decades after its discovery. Eur J Med Chem 2009; 44(3): 937-53. [ Links ]

2. Edgcomb JH, Arnold J, Yount EJ, Alving AS, Eichelberger L, Jeffery G, et al. Primaquine, SN 13272, a new curative agent in vivax malaria; a preliminary report. J Natl Malar Soc 1950; 9(4): 285-92. [ Links ]

3. Baird JK. Effectiveness of antimalarial drugs. N Engl J Med 2005; 352(15): 1565-77. [ Links ]

4. Carmona-Fonseca J. La primaquina tiene alta eficacia en la quimioprofilaxis primaria simple antipalúdica. Metanálisis. Iatreia 2006; 19(3): 244-60. [ Links ]

5. Carmona-Fonseca J. Recurrencias de malaria por Plasmodium vivax según el uso de primaquina. Análisis de estudios descriptivos longitudinales. Rev Bras Epidemiol 2012; 15(3): 488-503. [ Links ]

6. Galappaththy GN, Omari AA, Tharyan P. Primaquine for preventing relapses in people with Plasmodium vivax malaria. Cochrane Database Syst Rev 2009; Issue 4 2009. [ Links ]

7. World Health Organization (WHO). Guidelines for the treatment of malaria. Documento WHO/HTM/MAL/2006.1108. Geneva: WHO; 2006. [ Links ]

8. Klein TA, Tada MS, Lima JB, Tang AT. Infection of Anopheles darlingi fed on patients infected with Plasmodium vivax before and during treatment with chloroquine plus primaquine in Costa Marques, Rondonia, Brazil. Mem Inst Oswaldo Cruz 1992; 87(2): 191-5. [ Links ]

9. Alving AS, Johnson CF, Tarlov AR, Brewer GJ, Kellermeyer RW, Carson PE. Mitigation of the haemolytic effect of primaquine and enhancement of its action against exoerythrocytic forms of the Chesson strain of Plasmodium vivax by intermittent regimens of drug administration. Bull World Health Organ 1960; 22: 621-31. [ Links ]

10. Schmidt LH. Chemotherapy of the drug-resistan malarias. Annu Rev Microbiol 1969; 23: 427-54. [ Links ]

11. Baird JK. Resistance to therapies for infection by Plasmodium vivax. Clin Microbiol Rev 2009; 22(3): 508-34. [ Links ]

12. Baird JK, Hoffman SL. Primaquine therapy for malaria. Clin Infect Dis 2004; 39(9): 1336-45. [ Links ]

13. Galappaththy GN, Omari AA, Tharyan P. Primaquine for preventing relapses in people with Plasmodium vivax malaria. Cochrane Database Syst Rev 2007; 24(1):CD004389. [ Links ]

14. Goller JL, Jolley D, Ringwald P, Biggs BA. Regional differences in the response of plasmodium vivax malaria to primaquine as anti-relapse therapy. Am J Trop Med Hyg 2007; 76(2): 203-7. [ Links ]

15. John GK, Douglas NM, von Seidlein L, Nosten F, Baird JK, White NJ, et al. Primaquine radical cure of Plasmodium vivax: a critical review of the literature. Malar J 2012; 11:280. [ Links ]

16. Primo J. Calculadora para metaanálisis. Versión actualizada en 10/7/2009; 1008,00 KB. CASP España. Unidad de Digestivo, Hospital de Sagunto. Disponible: http://www.redcaspe.org/herramientas. (Visitada en 1 de julio de 2009). [ Links ]

17. Lozano J, Dennis R. El experimento clínico. In: Ruiz AML, editor. Epidemiología clínica Investigación clínica aplicada. Bogotá: Editorial Panamericana; 2004. p. 233-64. [ Links ]

18. Abdon NP, Pinto AY, da Silva RS, de Souza JM. Assessment of the response to reduced treatment schemes for vivax malaria. Rev Soc Bras Med Trop 2001; 34(4): 343-8. [ Links ]

19. Adak T, Valecha N, Sharma VP. Plasmodium vivax polymorphism in a clinical drug trial. Clin Diagn Lab Immunol 2001; 8(5): 891-4. [ Links ]

20. Álvarez G, Piñeros JG, Tobón A, Ríos A, Maestre A, Blair S, et al. Efficacy of three chloroquine-primaquine regimens for treatment of Plasmodium vivax malaria in Colombia. Am J Trop Med Hyg 2006; 75(4): 605-9. [ Links ]

21. Bergonzoli G, Rivers Cuadra JC. Therapeutic efficacy of different antimalarial regimens in the Costa Rica-Nicaragua border region. Rev Panam Salud Publica 2000; 7(6): 366-70. [ Links ]

22. Carmona-Fonseca J. Malaria vivax en niños: recurrencias con dosis total estándar de primaquina administrada durante 3 frente a 7 días. Iatreia 2010; 23(1): 10-20. [ Links ]

23. Carmona-Fonseca J, Maestre A. Prevention of Plasmodium vivax malaria recurrence: Efficacy of the standard total dose of primaquine administered over 3 days. Acta Trop 2009; 112(2): 188-92. [ Links ]

24. da Silva RS, Pinto AY, Calvosa VS, de Souza JM. Short course schemes for vivax malaria treatment. Rev Soc Bras Med Trop 2003; 36(2): 235-9. [ Links ]

25. Gogtay NJ, Desai S, Kamtekar KD, Kadam VS, Dalvi SS, Kshirsagar NA. Efficacies of 5- and 14-day primaquine regimens in the prevention of relapses in Plasmodium vivax infections. Ann Trop Med Parasitol 1999; 93(8): 809-12. [ Links ]

26. Leslie T, Mayan I, Mohammed N, Erasmus P, Kolaczinski J, Whitty CJ, et al. A randomised trial of an eight-week, once weekly primaquine regimen to prevent relapse of Plasmodium vivax in Northwest frontier Province, Pakistan. PLoS One 2008; 3(8): e2861. [ Links ]

27. Leslie T, Rab MA, Ahmadzai H, Durrani N, Fayaz M, Kolaczinski J, et al. Compliance with 14-day primaquine therapy for radical cure of vivax malaria: a randomized placebo-controlled trial comparing unsupervised with supervised treatment. Trans R Soc Trop Med Hyg 2004; 98(3): 168-73. [ Links ]

28. Pinto AYN, Ventura AMRS, Calvosa VSP, Silva-Filho MG, Santos MA, Silva RSU, et al. Eficácia de quatro esquemas terapêuticos para malaria vivax em crianças. J Pediatr 1998; 74(3): 222-7. [ Links ]

29. Pukrittayakamee S, Vanijanonta S, Chantra A, Clemens R, White NJ. Blood stage antimalarial efficacy of primaquine in Plasmodium vivax malaria. J Infect Dis 1994; 169(4): 932-5. [ Links ]

30. Rowland M, Durrani N. Randomized controlled trials of 5- and 14-days primaquine therapy against relapses of vivax malaria in an Afghan refugee settlement in Pakistan. Trans R SocTrop Med Hyg 1999; 93(6): 641-3. [ Links ]

31. Soto LS, Tarazona ARS, Requena DM, Cuentas AEL. Ensayo clínico del tratamiento de la malaria vivax con esquema acortado de primaquina comparado con el esquema tradicional. Rev Soc Peru Med Interna 2002; 15(4): 196-9. [ Links ]

32. Villalobos-Salcedo JM, Tada MS, Kimura E, Menezes MJ, Pereira da Silva LH. In-vivo sensitivity of Plasmodium vivax isolates from Rondonia (western Amazon region, Brazil) to regimens including chloroquine and primaquine. Ann Trop Med Parasitol 2000; 94(8): 749-58. [ Links ]

33. Walsh DS, Wilairatana P, Tang DB, Heppner DG Jr, Brewer TG, Krudsood S, et al. Randomized trial of 3-dose regimens of tafenoquine (WR238605) versus low-dose primaquine for preventing Plasmodium vivax malaria relapse. Clin Infect Dis 2004; 39(8): 1095-103. [ Links ]

34. Yadav R, Ghosh SK. Radical curative efficacy of five-day regimen of primaquine for treatment of Plasmodium vivax malaria in India. J Parasitol 2002; 88(5): 1042-4. [ Links ]

35. Cedillos RA, Warren M, Jeffery GM. Field evaluation of primaquine in the control of Plasmodium vivax. Am J Trop Med Hyg 1978; 27(3): 466-72. [ Links ]

36. Mendoza IG. Comparative study of two regimens of radical treatment of vivax malaria in Mexico. WHO/Mal/527.65. 1965. Disponível: http://apps.who.int/iris/bitstream/10665/65295/1/WHO_Mal_527.65.pdf. (Visitada en 4 de junio de 2012). [ Links ]

37. Li X, Li C, Che L, Liu X, Li Z, Huang D, et al. [Observation on efficacy of artemether compound against vivax malaria]. Zhongguo Ji Sheng Chong Xue Yu Ji Sheng Chong Bing Za Zhi 1999; 17(3): 175 7. [Chinese] [ Links ]

38. Yeshiwondim AK, Tekle AH, Dengela DO, Yohannes AM, Teklehaimanot A. Therapeutic efficacy of chloroquine and chloroquine plus primaquine for the treatment of Plasmodium vivax in Ethiopia. Acta Trop 2010; 113(2): 105-13. [ Links ]

39. Kim JR, Nandy A, Maji AK, Addy M, Dondorp AM, Day NP, et al. Genotyping of Plasmodium vivax reveals both short and long latency relapse patterns in Kolkata. PLoS One 2012; 7(7): e39645. [ Links ]

40. Rajgor DD, Gogtay NJ, Kadam VS, Kamtekar KD, Dalvi SS, Chogle AR, et al. Efficacy of a 14-day primaquine regimen in preventing relapses in patients with Plasmodium vivax malaria in Mumbai, India. Trans R Soc Trop Med Hyg 2003; 97(4): 438-40. [ Links ]

41. Carmona-Fonseca J, Alvarez G, Maestre A. Methemoglobinemia and adverse events in Plasmodium vivax malaria patients associated with high doses of primaquine treatment. Am J Trop Med Hyg 2009; 80(2): 188-93. [ Links ]

Financing source: Colciencias (1115 5693 3361, Contrato n° 444 2013), Codi - Estrategia de Sostenibilidad 2013 2014 - Universidad de Antioquia.

Received: May 15, 2014; Revised: February 21, 2014; Accepted: July 31, 2014

Corresponding author: Jaime Carmona Fonseca. Salud y Comunidad César Uribe Piedrahita, Universidad de Antioquia, Carrera 51D, 62 29, piso 3, Medellín, Antioquia, Colombia. E mail: jaimecarmonaf@hotmail.com

Conflict of interests: nothing to declare

Creative Commons License This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.