Clinical Effects of Nifuratel in Vulvovaginal Infections

Übersicht ArzneimForschDrugRes

Review

A meta-analysis of metronidazole-controlled trials Werner Mendlinga, Aldo Poli b, and Paola Magnani c

Klinik für Frauenheilkunde und Geburtsmedizin, VIVANTES-Klinika am Urban und im Friedrichshaina, Berlin (Germany), OPIS s.r.l.b, Desio (Milan, Italy), and Polichem SAc, Lugano (Switzerland)

Summary

Nifuratel (CAS 4936-47-4) displays a In the fixed and random effect analyses, strong antiprotozoarian and antibac- the confidence interval of Odds ratio in- terial activity and is provided with cer- cluded 1 and the p values for testing the tain fungicidal effect, but it is not active hypothesis of no difference between treat- against the physiologic flora. Its thera- ments were 0.656−1.266, p = 0.582 (fixed peutic effectiveness has been evaluated effects) and 0.643−1.290, p = 0.599 (ran- in more than 12,000 patients. The wide dom effects), respectively, indicating clinical experience with nifuratel con- equivalence. firms that the drug is safe and effective Furthermore, some controlled studies for the treatment of trichomoniasis, bac- and the wide clinical experience showed terial vaginosis, candidosis, and, particu- that the cure rate of nifuratel in patients larly, in patients suffering from mixed va- with mixed infections due to Tricho- ginal infection. monas vaginalis + Candida or Tricho- A meta-analysis of clinical trials com- monas vaginalis + bacteria or with bac- paring nifuratel and metronidazole (CAS terial vaginosis and mixed bacterial flora

443-48-1) in vulvovaginal infections was is higher than that of metronidazole, due Downloaded by: Harvard University Library. Copyrighted material. performed. All parallel-group metronida- to the wide spectrum of action of nifura- zole-controlled trials carried out in tel. patients with vulvovaginal infections have been included, complying with the following criteria: 1) cure assessed both as disappearance of symptoms and signs, and negative microbiological findings; Key words 2) microbiological tests performed with valid methods still used in current prac- ᭿ Bacterial vaginosis tice. ᭿ CAS 4936-47-4 Seven clinical trials have been se- ᭿ Inimur  lected, including overall 1767 patients, ᭿  832 out of whom were treated with nifur- Macmiror atel and 935 with metronidazole. The re- ᭿ Metronidazole sults of the meta-analysis confirmed the ᭿ Nifuratel, meta-analysis equivalence between nifuratel and me- ᭿ Trichomoniasis tronidazole: overall proportion of cured patients in the two groups were 88.5 % Arzneim.-Forsch./Drug Res. and 90.0 %, respectively, in the presence 52, No. 10, 725−730 (2002) of homogeneity among studies (p = 0.342).

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) Mendling et al. − Nifuratel 725 Übersicht

Zusammenfassung

Klinische Wirkungen von Nifuratel bei nen, die folgende Kriterien erfüllen, wur- 1,266, p = 0,582 (feste Effekte) bzw. vulvovaginalen Infektionen / Eine Meta- den berücksichtigt: 1) Die Infektion galt 0,643−1,290, p = 0,599 (zufällige Effekte), Analyse von Metronidazol-kontrollierten als geheilt, wenn die Symptome und was auf Äquivalenz hindeutet. Studien Krankheitszeichen verschwunden und Darüber hinaus haben einige kontrol- gleichzeitig die mikrobiologischen Ergeb- lierte Studien und die breite klinische Er- Nifuratel (CAS 4936-47-4) zeigt eine nisse negativ waren. 2) Die mikrobiologi- fahrung gezeigt, daß die Heilungsraten starke antiprotozoische und antibakte- schen Tests wurden mit validierten Me- von Nifuratel bei Patientinnen mit Misch- rielle Aktivität und besitzt eine gewisse thoden durchgeführt, die noch heute in infektionen durch Trichomonas vaginalis fungizide Wirkung, wirkt jedoch nicht ge- der Praxis angewandt werden. + Candida oder Trichomonas vaginalis + gen die physiologische Flora. Seine thera- Sieben klinische Studien wurden aus- Bakterien sowie bei Patientinnen mit bak- peutische Wirksamkeit wurde an mehr gewählt, an denen sich insgesamt 1767 terieller Vaginose und gemischte Bakte- als 12 000 Patientinnen getestet. Die Patientinnen beteiligten; 832 von ihnen rienflora höher sind als diejenigen von breite klinische Erfahrung mit Nifuratel wurden mit Nifuratel und 935 mit Metro- Metronidazol, was auf das breite Wir- bestätigt, daß das Präparat in der Thera- nidazol behandelt. Die Ergebnisse der kungsspektrum von Nifuratel zurückzu- pie von Trichomoniasis, bakterieller Vagi- Meta-Analyse bestätigten die Äquivalenz führen ist. nose, Candida-Mykosen und insbeson- von Nifuratel und Metronidazol: Der Ge- dere vaginalen Mischinfektionen sicher samtanteil der geheilten Patientinnen be- und wirksam ist. trug in den beiden Gruppen 88,5 % bzw. Eine Meta-Analyse von klinischen Stu- 90,0 % bei Homogenität der Studien (p = dien, in denen Nifuratel und Metronida- 0,342). In den Analysen der festen und zol (CAS 443-48-1) bei vulvovaginalen In- zufälligen Effekte lag das Vertrauensinter- fektionen verglichen wurden, wurde vall des Verhältnisfaktors (odds ratio) im durchgeführt. Alle Metronidazol-kontrol- Bereich von 1, und die p-Werte für die lierten Parallelgruppen-Studien mit Pa- Überprüfung der Hypothese der Äquiva- tientinnen mit vulvovaginalen Infektio- lenz beider Präparate betrugen 0,656−

1. Background also associated with complications, including abnormal Pap test, PID, and premature rupture of membranes in Most cases of discharge in women of childbearing age pregnancy and neonatal respiratory tract infection as are caused by bacteria; bacterial vaginosis represents well as neonatal genital infections [5−6]. a complex change in vaginal flora characterised by a Although the history and gynaecologic examination reduction in the prevalence and concentration of hy- are the source of important diagnostic clues, laboratory drogen peroxide producing lactobacilli, and an increase tests, such as vaginal pool wet mount exam, the amine in the prevalence and concentration of Gardnerella va- whiff test, vaginal pH, and the Q-tip test, should rou- ginalis, Mycoplasma hominis, anaerobic gram-negative Downloaded by: Harvard University Library. Copyrighted material. tinely be performed in all patients presenting a vaginal bacteria belonging to the genera Prevotella, Bacteroides, discharge to obtain a differential diagnosis, while va- and Peptostreptococcus species [1]. Bacterial vaginosis ginal cultures are required only for confirmation, ac- has been associated with several upper genital tract in- cording to the individual cases [7]. fections and obstetrical complications, such as pelvic Metronidazole (CAS 443-48-1) continues to be re- inflammatory disease (PID), post-caesarean endome- garded as the agent of choice in the management of tritis, post-hysterectomy pelvic infection, chorioam- both bacterial vaginosis and trichomoniasis. Although nionitis, premature rupture of membranes, and preterm metronidazole is highly active against anaerobic bac- − labour and delivery [2 4]. Candida vaginitis is the se- teria, it is less active against Gardnerella vaginalis and cond most common cause of vaginal infections and poor against Mobiluncus species. Despite this, a re- Candida albicans is responsible for 80 to 94 % of epis- peated 7-day course of metronidazole is often effective odes of Candida vulvovaginitis, while several investiga- treatment for a recurrence. A small number of women tors have reported an increased percentage of non-al- shows multiple recurrences. Prolonged oral use of me- bicans Candida species. Many factors increase suscept- tronidazole is associated with a high incidence of side ibility to vaginal Candida infections, such as pregnancy, effects located to the gastrointestinal tract and a risk of taking old high-dose oral contraceptives, diabetes, sys- peripheral neuropathies. Its use in pregnancy has been temic steroids, obesity and HIV infections. The third a matter of concern due to its theoretically supposed cause of infectious vaginitis is due to Trichomonas va- teratogenic effects [8]. ginalis; this organism has been found to coexist with a Several topical antifungals, such as clotrimazole, number of other infections, including gonococcal infec- miconazole, tioconazole, terconazole, nystatin, achieve tion and bacterial vaginosis. Trichomonas infection is high cure rates in excess of 80 % in case of uncompli-

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) 726 Mendling et al. − Nifuratel Review

cated vulvovaginal candidoses. The differences in for- assessed both as disappearance of symptoms and signs, and mulations are not considered to be clinically relevant. negative microbiological findings; 2) microbiological tests con- Oral azoles, such as ketoconazole, itraconazole and ducted with valid methods still used in current practice. fluconazole, have also been shown to achieve high cure Seven studies have been selected, four out of which were rates, but they are currently contraindicated in preg- performed in patients affected by Trichomonas vaginalis (T.v.) and three in mixed forms (see Table 1). A brief description of nancy [9]. these trials is reported as follows. The vulvo-vaginal infections have progressively Forty-two patients with recurrent trichomoniasis sub- changed during the last years, both quantitatively and sequent to treatment with nifuratel and 37 patients with recur- qualitatively, with the onset of mixed forms (where pro- rence after metronidazole treatment were assigned to receive a tozoa and mycetes are often associated with bacteria). second treatment with nifuratel (40 cases − 200 mg oral three 1) Nifuratel (CAS 4936-47-4) displays a strong antipro- times daily for 7 days and 250 mg/day vaginally for 10 days) or tozoarian and antibacterial activity and it is provided metronidazole (39 cases − 250 mg orally twice a day and 100 also with some fungicidal effect [10−11]. Unlike doxy- mg/day intravaginally for 6 days). The cure rate was 87.5 % cycline, nifuratel is not active against Lactobacillus spp. after nifuratel and 87.1 % after metronidazole [15]. No cross This fact is important in the vaginal infections where resistance between metronidazole and nifuratel could be the complete eradication of Lactobacillus spp. may pro- proved in these cases. duce change of vaginal microbial flora and vaginal pH. Patients with trichomonal vaginitis were assigned to receive Nifuratel has a very safe toxicological profile and it is metronidazole at the dosage of 200 mg orally three times a day for 7 days (120 women) or nifuratel at the dosage of 200 mg devoid of teratogenic effects; furthermore, the compar- orally three times a day for 7−10 days (90 women). The cure ison among past and recent clinical studies confirms rate was 73 % and 77 %, respectively [16]. that no resistance phenomena to the treatment with ni- An open, randomised, parallel group trial compared the ac- furatel are reported [11]. Its therapeutic effectiveness tivity of oral nifuratel and metronidazole in female patients has been evaluated in more than 12,000 patients, and it suffering from vaginitis due to Trichomonas vaginalis. The is on the market since 1965 in many European and ex- treatments (nifuratel 1200 mg/day for 10 days or metronidazole tra-European Countries. The wide clinical experience so 750 mg/day for 10 days) were randomly assigned to 184 far with nifuratel confirms that the drug is effective for patients (92 for each treatment group). The cure rate after the the treatment of trichomoniasis, bacterial vaginosis, 1st cycle of therapy was 73.9 % and 78.3 %, respectively, while candidoses, and, particularly, in patients suffering from the overall cure rate after 2 cycles was 80.4 % and 82.6 % [17]. mixed vaginal infection. A study compared nifuratel vaginal and oral treatment with Several trials comparing nifuratel with metronida- metronidazole, given at the following dosage: 500 mg/day va- zole and other agents currently used in the treatment ginally for 7 days + 600 mg/day orally for 10 days in trichomonal vaginitis. A second cycle of the two treatments was fore- of vulvovaginal infections have demonstrated the thera- seen in case of failure after the first one; concomitant oral peutic equivalence of the treatments [12−21]. Most of treatment for sexual partners was performed. The cure rates the studies have been performed before GCPs (good after the 1st cycle (80 % nifuratel − 79 % metronidazole) and clinical practice) came into force, and they reflect suit- the 2nd one (95 % nifuratel and 89 % metronidazole) were very able methods for that period. similar [18]. In order to assess the homogeneity of the results ob- The efficacy of nifuratel (oral 600 mg/day and vaginal 250 tained in these trials and to confirm the equivalence mg/day for 10 days) and that of metronidazole (oral 750 mg/ Downloaded by: Harvard University Library. Copyrighted material. between nifuratel and metronidazole in the treatment day and vaginal 500 mg/day for 10 days) were compared in of vulvovaginal infections, we applied the meta-analyt- 1050 cases. The cure rates were 79 % for nifuratel and 90 % for ical procedure, a retrospective quantitative technique metronidazole in patients with infections due to T.v. alone; that pools data from multiple trials. 58 % and 45 %, respectively, in T.v. and fungi infections, 75 % and 65 %, respectively, in T.v. and mixed bacterial flora infections. A second course in patients who were not cured after the first treatment showed a cure rate of 80 % and 96 %, respec- 2. Methods tively, in patients with T.v., of 90 % and 81 %, respectively, in An extensive search of all controlled clinical trials with nifuratel patients with T.v. and fungi and of 100 % and 95 % in patients in vulvovaginal infections was performed. We included in the with T.v. + mixed bacterial flora [19]. meta-analysis all parallel-group, metronidazole-controlled tri- One hundred and forty patients with vaginitis due to T.v., als carried out in patients with vulvovaginal infections, like Candida, bacteria or mixed infections were randomly assigned trichomoniasis, bacterial vaginosis, candidoses and mixed va- to receive oral nifuratel 600 mg/day for 7 days + vaginal nifura- ginal infections, that comply with the following criteria: 1) cure tel 250 mg/day for 10 days (70 cases) or oral metronidazole 600 mg/day for 7 days + vaginal nystatin 200000 IU/day for 7 days (70 cases). Cure rates were 75 % and 69.2 %, respectively [20]. Thirty six patients with vaginitis due to T.v., Candida albicans or “mixed non specific bacterial infections” [21] were 1) Available as Inimur  / Macmiror  (Taurus Pharma GmbH, treated with nifuratel (200 mg orally three times daily for 7 days Polichem S.A.) in form of sugar coated tablets (200 mg nifur- and 250 mg intravaginally for 10 days) or metronidazole (250 atel), pessaries (250 mg nifuratel) and 10 % ointment, re- mg 3 times daily orally and nystatin 1 vaginal pessary for 7 spectively. days). The cure rate was 54.5 % and 50 %, respectively.

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) Mendling et al. − Nifuratel 727 Übersicht

Table 1: Characteristics of the studies included in the meta-analysis.

No. of patients enrolled Study Route of administration / treatment (nifuratel / Diagnosis Dosage/day (reference) duration metronidazole)

Schmidt, H. 40 / 39 Trichomonas vaginalis N: oral, 7 days + vaginal, 10 days N: 600 mg + 250 mg [15] M: oral, 6 days + vaginal, 6 days M: 500 mg + 100 mg

Gjønnaess, H. 90 / 120 Trichomonas vaginalis N: oral, 7−10 days N: 600 mg [16] M: oral, 7 days M: 600 mg

Baron, A. 92 / 92 Trichomonas vaginalis N: oral, 10 days N: 1200 mg [17] M: oral, 10 days M: 750 mg

Block, E. 50 / 50 Trichomonas vaginalis N: oral, 7 days + vaginal 10 days N: 600 mg + 500 mg [18] M: oral, 7 days + vaginal 10 days M: 600 mg + 500 mg

Goisis, M. 482 / 568 Trichomoniasis (170/230) N: oral, 10 days + vaginal, 10 days N: 600 mg + 250 mg [19] Trichomonas vaginalis M: oral, 10 days + vaginal, 10 days M: 600 mg + 250 mg + fungi (50/38) Trichomonas vaginalis + mixed bacteria (262/300)

Pathak U.N. 56 / 52 Trichomonas vaginalis (13/14) N: oral, 7 days + vaginal, 10 days N: 600 mg + 250 mg [20] Moniliasis (28/23) M: oral, 7 days + vaginal, 7 days * M: 600 mg Mixed bacterial flora (15/15) + Nys 200,000 IU

Struthers J.O. 22 / 14 Trichomonas vaginalis (7/7) N: oral, 7 days + vaginal, 10 days N: 600 mg + 250 mg [21] Candidiasis (5/2) M: oral, 7 days + vaginal, 7 days * M: 750 mg Mixed bacterial flora (10/5) + Nys 1 pessary

N: nifurated, M: metronidazole, Nys: nystatin. a) Oral metronidazole combined with vaginal nystatin.

2.1. Measures of efficacy Table 2: Summary results. This meta-analysis focused on recovery from vulvovaginal in- No. of pa- Recovery Odds ratiotients fection (eradication of the pathogen) at the end of treatment Study (nifuratel / C n ie c (iuae / (Confidence(nifuratel as measure of efficacy. In all cases in which data were available, (reference) metro- metronida- Interval) nidazole) intent-to-treat analyses were performed; withdrawal was con- zole sidered as treatment failure. Schmidt H. 40/39 35 (88 %)/ 1.029 [15] 34 (87 %) (0.273−3.878) Gjønnaess H. 90/120 69 (77 %)/ 1.195 2.2. Statistical analyses [16] 88 (73 %) (0.634−2.253) Downloaded by: Harvard University Library. Copyrighted material. The significance (p < 0.05, two-sided) of the overall treatment Baron A. 92/92 74 (80 %)/ 0.866 − effect was evaluated by a chi-squared statistic (association [17] 76 (83 %) (0.411 1.824) test). Relative Risk (RR) and Odds Ratio (OR) with their 95 % Block E. 50/50 42 (84 %)/ 1.000 [18] 42 (84 %) (0.343−2.913) confidence intervals (CI) were calculated for individual trials Goisis M. 482/568 462 (96 %)/ 0.372 and for the summary results. For the analysis of dichotomous [19] 559 (98 %) (0.168−0.825) variables, a treatment effect model (fixed or random effects Pathak U.N. 56/52 42 (75 %)/ 1.333 model) was selected by testing for heterogeneity of the effect [20] 36 (69 %) (0.573−3.102) across the trials with a chi-square statistic, assuming a more Struthers J.O. 22/14 12 (55 %)/ 1.200 − liberal level of significance (p < 0.1). If the test of heterogeneity [21] 7 (50 %) (0.314 4.594) was significant, a random-effects analysis was carried out; in Total 832/935 736 (88 %)/ addition, for each outcome measure, we also used the fixed- 842 (90 %) effects and random-effects models to estimate summary treatment effects for all studies combined [22−23]. Further tests and analyses

Test of heterogeneity 6.780, 0.342 2.3. Sensitivity analyses (Q,p) To determine whether the results were unduly influenced by a Fixed effects analysis 0.912, (0.656−1.266), 0.582 single trial or a small number of trials, we repeated the meta- (Odds ratio, 95 % Confidence Interval, p) analysis, by using fixed and random effects models, including trials in chronological order to assess the robustness of treat- Random effects analysis 0.911, (0.643−1.290), 0.599 ment effects over a period of time and after successively with- (Odds ratio, 95 % Confidence Interval, p) drawing trials in decreasing order of their Odds ratio.

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) 728 Mendling et al. − Nifuratel Review

100% nifuratel metronidazole

90%

80%

70%

60%

50%

40%

30%

20%

10%

0% Schmidt H. Gjonnaes H. Baron A. Block E. Goisis M. Pathak U.N. Struthers J.O. Totals

Fig. 1: Recovery rate (%) with nifuratel or metronidazole in vulvovaginal infections.

3. Results Nifuratel is on the market since more than 30 years; clinical trials on the drug have been performed before Seven clinical trials with nifuratel in vulvovaginal infec- Good Clinical Practice came into force and they reflect tions met the inclusion criteria in the present analysis. suitable methods for that period. However, a lot of ex- In these trials a total of 1767 patients has been en- perience has been gained in clinical practice up today rolled, 832 out of whom were treated with nifuratel and and nifuratel has been confirmed to be effective in 935 with metronidazole (Table 1). The results of the patients suffering from trichomoniasis, bacterial vag- meta-analysis confirmed the equivalence between nifu- inosis, candidoses, and, particularly, in patients suffer- ratel and metronidazole (Table 2, Fig. 1): overall propor- ing from mixed vaginal infection. tion of cured patients in the two groups were 88.5 % The results of the present meta-analysis confirm that and 90.0 %, respectively, in the presence of heteroge- nifuratel is as effective as metronidazole, the gold neity among studies (p = 0.001). In the fixed and ran- standard therapy for trichomonal infections. The analy- dom effect analyses, the confidence interval of Odds ra- sis was not adjusted for study quality and, therefore, tio included 1 and the p values for testing the hypo- potentially at risk for this bias. In fact, it has been thesis of no difference between treatments were 0.656− shown that inclusion of studies of low methodological 1.266, p = 0.582 (fixed effects) and 0.643−1.290, p = 0.599 Downloaded by: Harvard University Library. Copyrighted material. quality tend to show an increased estimate of benefits. (random effects), respectively, indicating equivalence. However, the sensitivity analysis performed on the con- cerned data, did not detect any effect of exclusion of individual trials or a publication bias. 4. Discussion Furthermore, some controlled studies and the wide Nifuratel is a drug endowed with a wide spectrum of clinical experience showed that the cure rate of nifura- activity against micro-organisms responsible for vulvo- tel in patients with mixed infections due to Tricho- vaginal infections, such as Trichomonas vaginalis, aer- monas vaginalis + Candida or Trichomonas vaginalis + obic and anaerobic bacteria, Gardnerella vaginalis and bacteria or with bacterial vaginosis and mixed bacterial Candida spp. In trichomoniasis, in order to reach the flora is higher than that of metronidazole, due to the extravaginal localisations of the protozoa, it is essential wide spectrum of action of nifuratel. to combine the topical therapies (pessaries, ointment) with the oral one (sugar coated tablets). The partner should undergo the oral therapy since he is often the 5. References symptomless carrier of the parasite. In case of evident [1] Hill, G. B., Microbiology of bacterial vaginosis. Am. J. Obstet. local symptoms, also the local treatment of the male Gynaecol. 169, 450 (1993) partner with the ointment is recommended. Oral treat- [2] MacDermott, R. I. J., Bacterial vaginosis. Br. J. Obstet. ment is also indicated in vaginitis of uncertain origin. In Gynecol. 102, 92 (1995) trichomonal vaginitis, when it is neither desirable nor [3] Oleen-Burkey, M. A., Hillier, S. L., Pregnacy complica- practicable to carry out local therapy, good results will tions associated with bacterial vaginosis and their estimated be obtained with oral therapy alone, using a higher dose. costs. Infect. Dis. Obstet. Gynecol. 3, 149 (1995)

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) Mendling et al. − Nifuratel 729 Übersicht

[4] Hillier, S. L., Martius, J., Krohn, M. et al., A case-control [18] Block, E., Effekten av Nifuratel och metronidazol vid study of chorioamnionic infection and histologic chorioamni- behandling av Trichomoniasis. Läkartidningen 69, 5210 (1972) otis in prematurity. N. Engl, J, Med. 319, 971 (1988) [19] Goisis, M., Oppo, G. T., Treatment of female tricho- [5] Carr, P. I., Felsenstein, D., Friedman, R. H., Evaluation moniasis, in: F. Gasparri, G. Gargani, P. Periti (eds.), Diagnosis and management of vaginitis. J. Gen. Int. Med. 13, 335 (1998) and chemiotheraphy of urogenital infections, pp. 197−202. Ed- [6] Sobel, J. D., Vulvovaginitis in healthy women. Comp. izioni Mediche P. Periti, Firenze (1972) Ther. 25, 335 (1999) [20] Pathak, U. N., Sur, S. K., Farrand, R. J., Comparison of [7] Quan, M., Vaginitis: meeting the Clinical Challenge. Clin. metronidazole/nystatin and nifuratel in the treatment of vagin- Cornestone 3, 36 (2000) itis. Br. J. Clin. Pract. 29, 10 (1975) [8] SOGC (Society of Obstetricians and Gynaecologists of [21] Struthers, J. O., Treatment of patients with trichomonal, Canada), Clinical Practice Guidelines, Committee Opinion No. fungal and mixed bacterial vaginal infections at an out-patient 14, March 1997 gynaecological clinic, in: F. Gasparri, G. Gargani, P. Periti (eds.), [9] Reef, S., Levine, W. C., McNeil, M. M. et al., Treatment Diagnosis and Chemiotheraphy of urogenital infections, p. 157. options for vulvovaginal candidiasis. Background paper for de- Edizioni Mediche P. Periti, Firenze (1972) velopment of 1993 STD treatment recommendations. Clin. In- [22] Fleiss, J. L., The statistical basis of meta-analysis. Stat. fect. Dis. 20 (Suppl.), 580 (1995) Methods Med. Res. 2, 121 (1993) [10] Dubini, F., Furneri, P., Attivita` antimicrobica del Nifura- [23] DerSimonian, R., Laird, N., Meta-analysis in clinical tri- tel. G. Ital. Chemioter. 32, 545 (1985) als. Control. Clin. Trials 7, 177 (1986) [11] Mendling, W., Mailland, F., Microbiological and pharm- aco-toxicological profile of nifuratel and its favourable risk/be- nefit ratio for the treatment of vulvo-vaginal infections. Arz- neim.-Forsch./Drug Res. 52, 8 (2002) [12] Evans, B. A., Catterall, R. D., Nifuratel compared with metronidazole in the treatment of trichomonal vaginitis. Brit. Med. J., May 9, 1970, p. 335 [13] Heiss, H., Klinische Auswertung aktiver Arzneimittel gegen Trichomoniasis vaginalis im Doppel-Blindversuch. Wien. Med. Wochenschr. 121, 832 (1971) [14] Arnold, M., Vergleich von Nifuratel und Tinidazol bei Trichomonadenvaginitis. Ther. Umschau 31, 202 (1974) [15] Schmidt, H., Soost, H. J., Treatment of vaginal trichomoniasis and mycosis with nifuratel, in: F.Gasparri, G. Gargani, P. Periti (eds.), Diagnosis and chemiotheraphy of urogenital infections, pp. 191−195. Edizioni Mediche P. Periti, Firenze (1972) [16] Gjønnaess, H., Aure, J. C., Treatment of vaginal infections due to Trichomonas or candida, in: F. Gasparri, G. Gar- gani, P. Periti (eds.), Diagnosis and chemiotheraphy of urogenital infections, pp. 153−155. Edizioni Mediche P. Periti, Firenze (1972) Correspondence: [17] Baron, A., Roznicz miedzy leczienem ogolnym i skojar- Paola Magnani, Biol.Sc., zonym rzesistkowicy za pomoca metronidazolu i nifuratlu. Polichem S.A., via Caprino, 18, Wiad. Parazytol. 19 (2), 511 (1973) 6900 Lugano Paradiso (Switzerland) Downloaded by: Harvard University Library. Copyrighted material.

Arzneim.-Forsch./Drug Res. 52, No. 10, 725−730 (2002) 730 Mendling et al. − Nifuratel