Malaria is the leading cause of fever imported to France from the tropics, and Plasmodium falciparum
is the species most frequently involved (1
). Over the past two decades, the increasing rate of chloroquine resistance among P. falciparum
parasites throughout the world has led to the use of alternative treatments (1
). Although halofantrine is effective, accumulating evidence of its cardiotoxicity has limited its use (2
). In France, mefloquine is one of the first-line drugs recommended for use in the treatment of uncomplicated P. falciparum
malaria imported from areas of endemicity where P. falciparum
has a high rate of chloroquine resistance (1
). Quinine is an alternative. Quinine is used orally when mefloquine is contraindicated or had been used as chemoprophylaxis or is used intravenously when patients present with vomiting (1
). Although quinine is effective in most patients, it needs to be taken three times a day and reliably produces unpleasant side effects (nausea, vomiting, headache, tremor, tinnitus, transient loss of hearing, dysphoria, and hypoglycemia). Both treatment duration (7 days) and side effects affect patient compliance with quinine treatment. However, reduction of the duration of treatment exposes to the patient to the possibility of recrudescence. Although new treatments are required, their development is slow. Thus, the antimalarial properties of some antibiotics are of interest. In North America, quinine (or quinidine) is systematically combined with tetracycline for use in adults, whereas in France this combination is recommended only when patients have returned from southeastern Asia (1
). In recent years, Kremsner and colleagues (4
) have reported on comparative trials in which they have demonstrated the efficacies of short-term quinine-clindamycin regimens for the treatment of P. falciparum
malaria in Brazil or areas of endemicity in Africa. Although experts in the field provided guidelines including the possibility that clindamycin can be used instead of tetracycline (particularly in children) in association with quinine to treat imported cases of malaria (1
), this combination has never been evaluated to our knowledge in controlled trials among returned travelers (7
). The aim of the study described here was to compare the efficacy of a 7-day quinine regimen with that of a 3-day quinine-clindamycin regimen in the treatment of uncomplicated P. falciparum
malaria in travelers returning from the tropics to France.
The study was conducted in the Tropical Diseases Unit of the Hôpital Houphouët-Boigny in Marseilles, France, between June 1996 and December 1998. The trial was performed in accordance with the Declaration of Helsinki and its Hong Kong Amendment and according to the principles of good clinical practice, as defined by European Community guidelines and recommendations (III/3976/88-EN). The study protocol has been approved by the appropriate local ethics committees, and written informed consent was obtained from all patients. We included in the study patients presenting to our hospital with acute, uncomplicated P. falciparum
malaria if they fulfilled the following criteria: (i) body temperature greater than 37.5°C, (ii) age older than 15 years, (iii) nonpregnant state for women (a blood test for pregnancy was performed if needed), (iv) agreement to be an inpatient for 3 days, (v) agreement to comply with the study protocol during the 28 days monitoring, and (vi) provision of informed consent for noninclusion in the study. Criteria for noninclusion in the study were infection with species other than P. falciparum
, any sign or symptom of severe or complicated malaria (11
), colitis, intolerance of clindamycin, and concomitant antibiotic treatment. Patients were excluded if a review of the admission blood smears did not confirm the presence of malaria parasites or yielded a species other than P. falciparum
. The study was a 4-week, randomized, double-blind, placebo-controlled comparative trial. Eligible patients were randomly assigned to receive either a 7-day regimen of quinine (group Q) or a 3-day quinine-clindamycin regimen (group QC). Patients randomized into group Q received a 3-day regimen of intravenous infusions of quinine dihydrochloride (QUINIMAX [Sanofi Winthrop, Gentilly, France], in which 8 mg of quinine base per kg of body weight was infused over 4 h and then the same dose was continued every 8 h with a maximal dose of 1.5 g/day), followed by an oral 4-day quinine regimen administered in capsules (8 mg/kg every 8 h with each capsule containing 100 mg of quinine base, with a maximal dose of 1.5 g/day). Patients in group QC received intravenous perfusions of quinine as described above in combination with a 3-day regimen of intravenous clindamycin (DALACINE [Pharmacia & Upjohn SA, St Quentin-Yvelines, France], in which 5 mg/kg was infused over 1 h and then the same dose was continued every 8 h). To effect the double-blind trial design, an appropriate number of placebo treatments were used. Placebo capsules were prescribed for 4 days after the intravenous treatment to patients in group QC. Glucose solution (provided by Pharmacia & Upjohn SA) was used as a placebo and was given to the subjects in group QC as a substitute for the clindamycin given to the subjects in group Q. In order to protect the double-blind nature of the study, physicians, technicians, and all personnel involved in monitoring or conducting the trial were blinded to the trial drug codes until after completion of the trial. In case of an emergency, the pharmacist who prepared the drugs and placebo was able to unblind the treatment if required. On admission, the patient's history and clinical findings were recorded on standardized forms. The patients' characteristics at the baseline (age, sex, weight, country visited) were recorded. All subjects included in the trial were inpatients for at least the first 3 days of treatment. Body temperature was recorded every 6 h. Clinical signs and symptoms were recorded every 12 h. Thick blood smears were prepared before therapy (day 0) and every 24 h until two consecutive blood smears were free of asexual parasites for 48 h. Parasite counts were determined on May-Grümwald-Giemsa-stained thick blood smears in a blind fashion. A thick blood smear was classified negative when no parasite was detected in 200 high-power fields. Thereafter, smears were prepared on days 7, 14, and 28. Responses to treatment were classified as (i) early therapeutic failure (aggravation or fever and parasitemia on day 3), (ii) late therapeutic failure (fever and parasitemia between day 4 and day 28), and (iii) adequate clinical response (negative parasitemia and apyrexia at day 28). The primary endpoint was the time to parasite clearance. The secondary endpoint was time to resolution of fever, defined as the time needed for the body temperature to fall below 37.5°C for at least 48 h. Side effects evaluated as treatment-emergent symptoms were recorded every 12 h during hospitalization and thereafter on days 7, 14, and 28. In the case of diarrhea, stool tests were performed in order to identify Clostridium difficile
and its toxin. We decided to include a minimum of 50 patients in each group. The sample size was determined according to the results of previous studies conducted in areas of endemicity; those studies demonstrated that parasite and/or fever clearance times were significantly shorter among patients receiving the quinine-clindamycin regimen than among patients receiving quinine alone (4
). Data were analyzed with the SAS statistical software (SAS Institute Inc., Cary, N.C.). Data were described by the median, range, and interquartile range. The data from the two treatments arms were compared by using the Kruskal-Wallis H test. For categorical variables, percentages and corresponding exact binomial or normal approximations of the 95% confidence intervals (CIs) were calculated. The chi-square test or Fisher's exact test was used, as appropriate, for comparison of the two data from the treatment groups. The analysis of the 28-day cure rate was performed both for the intention-to-treat (ITT) and for the per-protocol (PP) patient populations. For the ITT analysis, loss to follow-up was considered a treatment failure. Parasite and fever clearance times were evaluated by survival analysis by using the log-rank test with Kaplan-Meier estimates of the survival function.
Of the 119 patients eligible for the study, 4 patients refused to provide consent and 115 were randomly assigned to receive a 7-day regimen of quinine or a 3-day treatment of quinine-clindamycin. Seven patients (three patients in group Q and four patients in group QC) were excluded on day 1 because a review of the admission blood smears did not confirm the presence of malaria parasites or yielded a species other than P. falciparum. These patients were not included in the analysis. A total of 108 patients (55 in group Q and 53 in group QC) were analyzed (Fig. ). The patients (age, sex, weight, country visited) and the baseline characteristics of the disease were similar for both groups. Five patients (three in group QC and two in group Q) were living in areas of endemicity in Africa and were visiting France when they presented with fever. The remaining 103 patients were French or new immigrants living in France for more than 5 years and were considered nonimmune persons with respect to malaria. All patients had returned from areas of endemicity where the chloroquine resistance rate is high or where multidrug resistance occurs in P. falciparum parasites, including 13 countries from sub-Saharan Africa (40 patients), the Comoro Islands (67 patients), and Asia (1 patients). Three patients (one in group Q and two in group QC) were lost to follow-up after they were discharged on day 4. These patients were considered treatment failures in the ITT analysis, while data for these patients were not included in the PP analysis. Among the 108 patients whose data were used for the ITT analysis (55 patients in group Q and 53 patients in group QC), the adequate clinical response was 94.6% (95% CI, 84.9 to 98.9) for group Q, whereas it was 96.2% (95% CI, 87 to 99.6) for group QC (P = 1). The difference in adequate clinical response between the two treatments in the ITT analysis was 1.7% (95% CI, −6.2 to 9.5). The median clearance times for both parasitemia (3 days in both groups; P = 0.5) and fever (43 h in group QC versus 52 h in group Q [P = 0.5]) were not statistically different between the two groups. Among the 105 patients whose data were used for the PP analysis (54 patients in group Q, 51 patients in group QC), the 28-day cure rate was 100% (95% CI, 93 to 100) for group QC, whereas it was 96.3% (95% CI, 87.3 to 99.6) for group Q (P = 0.5) (Table ). The difference in the adequate clinical response between the two treatment groups in the PP analysis was 3.7% (95% CI, −4.2 to 11.2). Side effects were reported among patients in both groups, with no significant difference (Table ). However, the treatment protocol had to be stopped because of severe side effects for two patients in group Q. Both were considered early failures. The first patient presented with a severe hemolytic anemia on day 2. She was hospitalized in an intensive care unit, and quinine treatment was stopped on day 3. She received an erythrocyte concentrate and remained well. The second patient presented with a severe toxic rash on day 5. Quinine was stopped, and the patient recovered. No severe adverse reaction was reported in any of the patients in group QC.
Controlled study of 3-day quinine-clindamycin (quinine-dalacine) treatment versus 7-day quinine treatment in patients returning to France from the tropics with uncomplicated falciparum malaria: trial profile.
Baseline characteristics of patients and outcome by treatment group in a controlled study of 3-day quinine-clindamycin treatment versus 7-day quinine treatment in patients returning to France from the tropics with uncomplicated falciparum malaria
Side effects by treatment group in a controlled study of 3-day quinine-clindamycin treatment versus 7-day quinine treatment in patients returning to France from the tropics with uncomplicated falciparum malariaa
In order to treat uncomplicated falciparum malaria, short-term quinine-clindamycin regimens have been tested in several clinical trials conducted in areas of endemicity (3
). In Africa, 88% of Gabonese children with uncomplicated falciparum malaria were successfully treated with a short-term quinine-clindamycin combination (5
). In another study conducted in Gabon, 92% of adults were cured with such a regimen (8
). More recently, Vaillant and colleagues (10
) reported on the efficacy of a 3-day quinine-clindamycin regimen in children from western Gabon, with a success rate greater than 97% by day 20 (10
). However, our study is the first to compare a 3-day quinine-clindamycin regimen with a 7-day quinine regimen for the treatment of imported uncomplicated P. falciparum
malaria in returned travelers. All patients in group QC were successfully cured. Our study lacked the power to detect a difference between the adequate clinical response rates between both regimens tested. However, assuming a type I error of 0.05 and a type II error of 0.20, it would have been necessary to include 2,562 patients in the study to detect the 1.6% difference that we observed between both groups in the ITT analysis. A huge number of patients need to be included in comparative trials of imported uncomplicated falciparum treatments to achieve a significant power. This is one of the reasons why recommendations for the treatment of returned travelers are more often based on guidelines provided by malaria advisory committees or experts in the field than on evidence-based data. In areas of endemicity, comparative studies usually include a group of patients who are treated with chloroquine. As chloroquine treatment frequently fails because of the existence of chloroquine resistance throughout the world, a difference in the response rates appears between the different treatment groups and the results are more acceptable in terms of statistical analysis.
In contrast to the results of Kremsner et al. (4
) in Gabon the difference in favor of the quinine-clindamycin regimen concerning the fever clearance time that we observed (43 versus 52 h) was not satistically significant. Similar side effects were reported in the two groups. However, we noted a trend toward a better tolerance of the quinine clindamycin regimen in group QC, because the protocol had to be stopped in two patients in group Q because of severe side effects. For the second patient, severe side effects occurred on day 5 and may have been avoided if the treatment had been stopped on day 3. The side effects on day 7 were not strictly recorded for each patient. Thus, we failed to demonstrate a shorter duration of hearing disturbance, which was expected among patients treated with the 3-day quinine-clindamycin regimen.
Although the use of oral drugs is usually recommended for the treatment of uncomplicated falciparum malaria, we used an intravenous regimen in this study to include patients who presented with vomiting and to ensure adherence. However, both quinine and clindamycin are well absorbed, and the combination may be as effective as an oral treatment (in the absence of vomiting).
In conclusion, a 3-day quinine-clindamycin regimen for the treatment of imported uncomplicated P. falciparum malaria in returned travelers was well tolerated and compared favorably with a 7-day quinine regimen, one of the standard regimens for the treatment of uncomplicated falciparum malaria in France. Benefits include the reduction of the duration of the treatment and a trend toward a diminution of side effects. Therefore, shortening of the duration of quinine treatment to 3 days may be possible with the addition of clindamycin. This short-term intravenous combination may be an alternative to oral mefloquine when patients present with vomiting. However, appropriate therapy of uncomplicated malaria should be oral, and further studies are warranted to demonstrate interest in an oral 3-day quinine-clindamycin regimen in returned travelers.