Journal of Antimicrobial Chemotherapy (2004) 54, 211 216 DOI: 10.1093/jac/dkh278 Advance Access publication 26 May 2004 A double-blind randomized controlled trial of fusidic acid and metronidazole for treatment of an initial episode of Clostridium difficile-associated diarrhoea Marlene Wullt* and Inga Odenholt Department of Infectious Diseases, University Hospital, Malmö, Sweden JAC Received 17 November 2003; returned 16 January 2004; revised 2 April 2004; accepted 13 April 2004 Objectives: Few treatment options are currently available to treat patients suffering from an initial episode of Clostridium difficile-associated diarrhoea (CDAD). Patients and methods: A prospective, randomized controlled, double-blind trial was conducted to compare the efficacy of fusidic acid and metronidazole for treatment of patients experiencing a first episode of CDAD. The primary outcomes were clinical cure and clearance of determined on days 8 13, and secondary outcomes were clinical recurrence and reappearance of evaluated on days 35 40. Results: Of the patients in the fusidic acid group, 83% were clinically cured in comparison to 93% in the metronidazole group (P 5 0.116) at the first follow-up visit. Clearance of did not differ between the two groups at that time. Clinical recurrence and reappearance of were noted in 27% and in 13% of the patients receiving fusidic acid, respectively and in 29% and 10% of those given metronidazole at the second follow-up on days 35 40. Conclusion: Since three of the four primary and secondary outcomes were almost identical for the two groups, the results indicate that fusidic acid is as effective as metronidazole in curing an initial episode of CDAD and can therefore be considered as an adequate alternative for treatment of this disease. Keywords: antibiotics, alternative treatments, toxins, pseudomembranous colitis Introduction Clostridium difficile-associated diarrhoea (CDAD) constitutes an increasing problem in hospitals, and it primarily affects elderly patients with chronic diseases and surgical patients. 1 6 CDAD is associated with a variety of symptoms ranging from mild diarrhoea through moderately severe disease to pseudomembranous colitis. 7 Most patients experiencing a first episode of CDAD are cured by treatment with metronidazole or vancomycin. 8 However, the latter drug should no longer be considered as a first-line treatment due to the increasing incidence of vancomycin-resistant enterococci (VRE), which essentially leaves only metronidazole in the treatment of CDAD. Fusidic acid has been found to be effective against C. difficile in vitro. 9,10 Moreover an earlier open, non-blinded study had demonstrated that fusidic acid is adequately effective in the treatment of CDAD. 11 However, Wenisch et al. 12 later reported that fusidic acid offers inferior, but not statistically significant, clinical and microbiological efficacy for treatment of CDAD. Since the results of Wenisch et al. did not agree with our clinical experiences, this was the incentive to test fusidic acid in a larger investigator-initiated trial. To our knowledge, this is the first study to be focused solely on patients with an initial episode of CDAD and it is also hitherto the largest trial to examine the effectiveness of fusidic acid for the treatment of such patients. Materials and methods Study population and design The study was carried out at two university hospitals and seven general hospitals in southern Sweden from September 1999 to May 2002. Inpatients and outpatients in all departments of the hospitals were enrolled. The local Microbiology Laboratories daily informed the respective investigator teams of all positive... *Corresponding author. Tel: +46-40-331000; Fax: +46-40-336279; E-mail: marlene.wullt@skane.se... 211 JAC vol.54 no.1 q The British Society for Antimicrobial Chemotherapy 2004; all rights reserved.
M. Wullt and I. Odenholt assays. Enrolment was done by the investigator teams at the Departments of Infectious Diseases. Criteria for inclusion in the study were age >18 years, lack of hypersensitivity to fusidic acid or metronidazole, a positive assay from faeces within 6 days before enrolment (toxins A and B; ELISA, Premier EIA, Meridian Diagnostics, USA), and a history of ongoing diarrhoea (diarrhoea defined as three or more loose stools per day for at least 2 days). Exclusion criteria comprised the following: a prior clinically and microbiologically established episode of C. difficile diarrhoea within the previous 6 months, pregnancy, severe or concomitant medical conditions such as AIDS and metastatic cancer and the risk of non-compliance with the study protocol. Ongoing treatment with metronidazole, vancomycin, fusidic acid, or teicoplanin was also an exclusion criterion, and further treatment with these drugs was not allowed during the study period. Informed consent was obtained from all patients. The investigation was approved by the respective regional Medical Ethics Committees. Treatment protocol and interventions On enrolment day, a medical history was taken that included prior and ongoing symptoms, complete history of previous and ongoing antibiotics, earlier hospitalization, underlying diseases, interventions undergone, and other medications of importance (Table 1). Patients were randomized to fusidic acid 250 mg orally three times daily or to metronidazole 400 mg orally three times daily for 7 days. The first day of intake of the study medication was defined as day 1. Patients were asked to keep a diary covering the frequency and consistency of diarrhoea, presence or absence of other symptoms, side effects, and intake of the study medication. Follow-up visits were conducted on days 8 13 and 35 40. In cases of clinical failure or clinical recurrence, the patients were seen at follow-up visits within 2 days and then left the study without further follow-up. At follow-ups, medical history was taken, and stool samples were collected for later analysis of toxins A and B. Toxin assays were carried out at the Clinical Microbiology Laboratory at each centre. Randomization The study medication was delivered from Leo Pharma AB, Malmö, Sweden, in coded blister packs, each containing a daily dose of six capsules (metronidazole 200 mg 2 3 or placebo) or three tablets (fusidic acid 250 mg 3 or placebo). The placebo capsules and tablets did not differ in form or colour from the active counterparts. An independent statistician provided a computer-generated list of random set numbers. The investigator teams were unaware of the treatment allocation. Outcomes Primary outcomes were clinical cure and clearance of C. difficile toxin at the first follow-up (days 8 13). Clinical cure was defined Table 1. Baseline characteristics of the study population (n =131) and the 114 evaluable patients in both treatment groups Fusidic acid Metronidazole Intention-to-treat Characteristic (n=59) (n=55) (n=131) Mean (S.D.) age (years) 58.0(18.7) 58.4(18.2) 59.0(18.2) Gender (female/male) 41/18 33/22 80/51 Inpatient on enrolment 27(46) 28(51) 67(51.5) Outpatient on enrolment 32(54) 27(49) 64(49) Duration of diarrhoea before treatment mean (days) 22.3(35.1) 19.6(28.3) 22.4(36.4) median (days) quartiles 11(6.22) 9(5.18) 10(5.20) No. of bowel movements/24 h 3 5 19(32) 16(29.6) 45(35) 5 10 29(49) 24(43.6) 60(45.8) 10 15 6(10) 11(20) 17(13) >15 5(8) 4(7) 9(7) Fever (>37.98C) 14(24) 15(27) 34(26) Abdominal pain 40(69) 35(63) 82(63) Vomiting 12(20) 12(22) 28(21) Nausea 32(52) 23(42) 65(50) Underlying conditions cardiovascular/hypertensive 22(37) 15(27) 45(34) diabetes mellitus 5(8) 8(14) 14(11) acid-peptic disease 13(22) 9(16) 24(18) renal disease 6(10) 1(2) 7(5) hepatic disease 3(5) 2(4) 5(4) malignancy 8(14) 6(11) 17(13) chronic respiratory disease 2(3) 6(11) 12(9) inflammatory bowel disease 3(5) 2(4) 5(4) Surgery within last 4 weeks 8(14) 12(22) 22(17) Prior hospitalization 13(28) 13(34) 28(29) Prior antibiotic treatment 44(75) 42(76) 86(75) Ongoing antibiotic treatment 10(17) 9(16) 19(17) Values indicate numbers (percentages) unless stated otherwise. Note: There were no significant differences between the two treatment groups in terms of clinical features. 212
Treatment of CDAD as the cessation of diarrhoea within 5 8 days of initiating treatment, and clinical failure as persistence of diarrhoea on days 5 8. Persistence of was defined as a positive assay for on days 8 13. Secondary outcome was clinical recurrence, defined as the reappearance of diarrhoea on days 8 40 in clinically cured patients who had completed 7 days of treatment. The criterion for reappearance of was detection of on days 35 40, after a negative toxin assay on days 8 13. A positive assay at the first and second follow-ups was categorized as chronic persistence of C. difficile toxin, whereas a positive toxin assay on days 8 13 and a negative assay on days 35 40 was classified as late clearance of C. difficile toxin. Side effects were tertiary outcomes. Statistical analysis We analysed the results both by intention-to-treat and per protocol. We used Fisher s test, x 2 test, and Mann Whitney U-test to determine statistical significance. Odds ratios with 95% confidence intervals were calculated according to Mantel Haenszel. A P value less than 0.05 was considered significant. The primary objective of the study was to demonstrate the non-inferiority of fusidic acid compared to metronidazole for the clinical cure of CDAD and clearance of on days 8 13. With a postulated cure rate of 90%, the planned study population was 150 patients in each group in order to detect an absolute difference of 10% with a power of 80% and a 5% significance level. Figure 1. Design of randomized, double-blind, placebo-controlled trial to compare the efficacy of fusidic acid and metronidazole for treatment of patients experiencing a first episode of CDAD. 213
M. Wullt and I. Odenholt Results Altogether 131 patients could be enrolled and randomized according to the protocol. Of these, 114 patients were evaluable, 59 randomly assigned to treatment with fusidic acid and 55 to metronidazole (Figure 1). The two treatment groups were comparable with respect to baseline characteristics (Table 1). Table 2 illustrates clinical and microbiological outcomes for both treatment groups. The proportion of patients clinically cured on days 8 13 was 83% in the fusidic acid group and 93% in the metronidazole group (OR = 0.38; 95% CI 0.11 1.31; P = 0.116). The rate of persistence of was similar for the two groups (22% and 23%, respectively; P = 0.939). The results at the second follow-up were also similar for the two treatment groups. Clinical recurrence and reappearance of were noted in 27% and 13% of the patients receiving fusidic acid, respectively and in 29% and 10% of those given metronidazole. Five patients in the fusidic acid group and eight in the metronidazole group experienced side effects, primarily abdominal pain (three in each group) and nausea (one taking fusidic acid and three metronidazole). Other side effects were headache Table 2. Primary and secondary outcomes in prospective, randomized controlled, double-blind trial to compare the efficacy of fusidic acid and metronidazole for treatment of patients experiencing a first episode of CDAD Fusidic acid Metronidazole Follow-up n = 59 (%) n=55 (%) P value a Days 8 13 Clinical effect b 0.116 Cure 49(83) 51(93) Failure 10(17) 4(7) Bacteriological effect c 0.939 Clearance of 46(78) 41(77) Persistence of 13(22) 12(23) Days 35 40 Clinical effect d 0.971 No recurrence 33(73) 35(71) Recurrence 14(27) 15(29) Bacteriological effect e 0.691 Permanent clearance of 32(70) 34(68) Late clearance of 4(9) 3(6) Reappearance of 6(13) 5(10) Chronic persistence of 4(9) 8(16) Values represent numbers (percentages) of patients. a x 2 test. b Patients with a clinical failure, who left the study and were not further evaluated. c Faecal samples from two patients were missing for analysis in the metronidazole group. d Two patients in the fusidic acid group and one patient in the metronidazole group left the study before the second follow-up. e Faecal sample from one patient was missing for analysis in the fusidic acid group. (one receiving fusidic acid), exanthema (one taking metronidazole), and metallic taste (one given metronidazole). Discussion This study showed that there was no statistically significant difference in clinical and bacteriological cure rate between the two treatment groups. There was a tendency towards a higher frequency of clinical failure in the fusidic acid group compared with the metronidazole group, but this was not correlated with the rate of persistence of. A possible explanation for this is that the degree of illness indicated by current symptoms (such as duration of diarrhoea before enrolment, abdominal pain, nausea and underlying diseases) seemed to be more severe in patients taking fusidic acid than in those on metronidazole, although the observed differences were not statistically significant. Interestingly, the rate of chronic persistence of was higher in the metronidazole group (16%) than in the fusidic acid group (9%). This difference was not statistically significant, but it may prove to be of interest in the future when considering the risk of recurrences of CDAD. Response to therapy was rapid and within 5 days of the onset of therapy for both drugs (data not shown), which agrees with findings reported by other investigators. 13,14 No investigations that have focused on treatment of uncomplicated CDAD have provided statistically significant evidence that one antibiotic treatment is superior to another. 15 Moreover, most studies, especially those with alternative agents like bacitracin and teicoplanin have not been large enough to yield statistically significant data and have not dealt exclusively with patients experiencing a first episode of CDAD. 12,13,16 18 The impact of fusidic acid on CDAD has been examined in two earlier trials. 11,12 Cronberg et al. carried out an unblinded study to compare fusidic acid (various regimens of 0.5 1.5 g daily) with metronidazole and vancomycin for treatment of CDAD. 11 These investigators noted a cure rate of 75% for fusidic acid and similar rates for metronidazole and vancomycin, a level that is markedly lower than observed in our study. However, they also included patients with recurrent episodes of CDAD. Wenisch et al. compared the efficacy of fusidic acid, metronidazole, vancomycin and teicoplanin in treatment of CDAD in 126 patients. 12 The investigated regimens represented higher doses and a longer duration of treatment than applied in our study and led to clinical cure rates of 93 96%, which were higher than those we achieved with fusidic acid and comparable to our results with metronidazole. Wenisch et al. also observed a percentage rate of clinical recurrence of 28% in the fusidic acid group, which agrees with our findings but is higher than the rates these investigators reported for the other treatment agents (16% for metronidazole and vancomycin and 7% for teicoplanin). However, differences between the recurrence rates were not statistically significant, except for teicoplanin versus fusidic acid (P = 0.042) and they noted that the percentage of positive cytotoxin assays 30 days after treatment was not significantly higher for the patients who had taken fusidic acid than for those who had been given metronidazole. They also found that patients with endoscopically confirmed pseudomembranous colitis showed similar results regardless of whether they had received fusidic acid or metronidazole. It was not stated in the paper whether recurrent cases of CDAD had been enrolled. 214
Treatment of CDAD Other studies that have investigated metronidazole in comparison with other agents such as vancomycin, have found similar clinical cure rates for metronidazole as in our study. 13,19 Approximately 20% of all cases of CDAD usually resolve spontaneously when the precipitating antibiotic is withdrawn. 8 However, only 17% of the patients included in our study (considering both treatment groups) were undergoing antibiotic treatment at the time of enrolment and most of the subjects had underlying conditions that required specific treatment for CDAD. Also, since symptomatic patients with CDAD are considered as a source for nosocomial spread of C. difficile, it is reasonable to treat these patients at the time of laboratory diagnosis. 8 Both metronidazole and vancomycin have drawbacks. Nausea frequently occurs during treatment with metronidazole, and pregnancy has been considered to be a contraindication for the use of this drug. 19,20 Decreased susceptibility of C. difficile to metronidazole has also been observed. 21 24 Considering vancomycin, a significant association has been found between vancomycin treatment of CDAD and VRE infection or colonization. 24 This may result in vancomycin-resistant strains of C. difficile, which poses new challenges in the management of CDAD. 25 As a result of the growing concern that widespread use of vancomycin will lead to increased resistance in enterococci, this drug should be saved as an alternative for treatment of recurrent disease. 8,26 Fusidic acid has a limited spectrum of activity and is effective mainly against Gram-positive bacteria; most Bacteroides spp. are resistant or only modestly sensitive and Gram-negative bacteria are completely resistant to this drug. 9,27 In contrast, metronidazole has an antimicrobial spectrum focused primarily on anaerobic bacteria such as Bacteroides spp. 28 It has been suggested that the absence of Bacteroides spp. in the colonic microflora of patients with CDAD can give rise to chronic relapsing CDAD. 28,29 In other words, the presence of Bacteroides spp. in the gut may prevent colonization of C. difficile, which might be an advantage for fusidic acid. Although selection of resistant bacteria during treatment with fusidic acid has been documented previously both in vitro and in vivo, it does not seem to occur at high frequency in clinical practice. 30,31 The resistance rates of Gram-positive bacteria against fusidic acid have remained low in all parts of the world, even in countries with a high consumption of fusidic acid (Denmark) 32 with the exception of a clonal spread of an impetigo strain resistant to fusidic acid in Scandinavia, which, however, is not considered to be due to an overuse of fusidic acid. 33 A major impetus for our study was the knowledge that earlier clinical experience of fusidic acid for treatment of CDAD had been satisfactory. To minimize the ecological impact of the drugs, we used lower doses and a shorter duration of treatment than had been employed in previous investigations. Concerning costs for a 7 day course of treatment, metronidazole is the cheapest ( 8.2), followed by fusidic acid ( 13) and vancomycin (125 mg 4) ( 66). Few options are currently available to treat patients suffering from CDAD. Even though we did not reach the planned sample size mainly due to spontaneous cure of diarrhoea, already initiated treatment for CDAD and risk for non-compliance, our investigation does include a large number of patients. There was a tendency towards a higher frequency of clinical failure in the fusidic acid group on days 8 13, but no difference in clearance of toxin was seen at that time. Also, there were no differences in clinical and bacteriological outcome on days 35 40. Therefore, fusidic acid can be considered as an adequate alternative to metronidazole for treatment of initial episodes of CDAD. Conflict of interest statement None. Acknowledgements We are grateful to Stig Cronberg for help in initiating the study and Jan-Åke Nilsson for help with the statistics. We also thank all participating investigators at the Departments of Infectious Diseases at hospitals in southern Sweden: in Helsingborg, Katrin Arnvik, Anna-Karin Larsson, and Bushra Ibrahim-Baker; in Jönköping, Peter Iveroth and Jesper Svefors; in Karlskrona, Ingrid Lindblad and Carl-Johan Fraenkel; in Kristianstad, Karin Sjöbeck and Mattias Waldeck; in Lund, Karin Norlin and Gisela Otto; in Skövde, Lars Ljungström and Erik Backhaus; in Växjö, Per Dahl; in Örebro, Torbjörn Norén. This study was supported by grants from Region Skåne and the Scandinavian Society of Antimicrobial Chemotherapy, and by Leo Pharma AB, Malmö, Sweden, who also delivered the study medication. References 1. Wilcox, M. H. (1996). Cleaning up Clostridium difficile infection. Lancet 348, 767 8. 2. Jones, E. M. & MacGowan, A. P. (1998). Back to basics in management of Clostridium difficile infections. Lancet 352, 505 6. 3. Wilcox, M. H. (1998). Clostridium difficile setting the scene. Journal of Antimicrobial Chemotherapy 41, Suppl. C, 1 3. 4. Kyne, L., Sougioultzis, S., McFarland, L. V. et al. (2002). Underlying disease severity as a major risk factor for nosocomial Clostridium difficile diarrhoea. Infection Control and Hospital Epidemiology 23, 653 9. 5. Morris, A. M., Jobe, B. A., Stoney, M. et al. (2002). Clostridium difficile colitis: an increasingly aggressive iatrogenic disease? Archives of Surgery 137, 1096 100. 6. Simor, A. E., Bradley, S. F., Strausbaugh, L. J. et al. (2002). Clostridium difficile in long-term-care facilities for the elderly. Infection Control and Hospital Epidemiology 23, 696 703. 7. Borriello, S. P. (1998). Pathogenesis of Clostridium difficile infection. Journal of Antimicrobial Chemotherapy 41, Suppl. C, 13 9. 8. Wilcox, M. H. (1998). Treatment of Clostridium difficile infection. Journal of Antimicrobial Chemotherapy 41, Suppl. C, 41 6. 9. Collignon, P. & Turnidge, J. (1999). Fusidic acid in vitro activity. International Journal of Antimicrobial Agents 12, Suppl. 2, S45 58. 10. Oppenheimer, M., Kronvall, G., Karlsson, I. et al. (2000). Fusidic acid disk diffusion testing of Clostridium difficile can be calibrated using single-strain regression analysis. Scandinavian Journal of Infectious Diseases 32, 633 6. 11. Cronberg, S., Castor, B. & Thoren, A. (1984). Fusidic acid for the treatment of antibiotic-associated colitis induced by Clostridium difficile. Infection 12, 276 9. 12. Wenisch, C., Parschalk, B., Hasenhundl, M. et al. (1996). Comparison of vancomycin, teicoplanin, metronidazole, and fusidic acid for the treatment of Clostridium difficile-associated diarrhoea. Clinical Infectious Diseases 22, 813 8. 13. Teasley, D. G., Gerding, D. N., Olson, M. M. et al. (1983). Prospective randomised trial of metronidazole versus vancomycin for Clostridium-difficile-associated diarrhoea and colitis. Lancet ii, 1043 6. 14. Wilcox, M. H. & Howe, R. (1995). Diarrhoea caused by Clostridium difficile: response time for treatment with metronidazole and vancomycin. Journal of Antimicrobial Chemotherapy 36, 673 9. 215
M. Wullt and I. Odenholt 15. Gerding, D. N. (2000). Treatment of Clostridium difficileassociated diarrhoea and colitis. Current Topics in Microbiology and Immunology 250, 127 39. 16. Young, G. P., Ward, P. B., Bayley, N. et al. (1985). Antibioticassociated colitis due to Clostridium difficile: double-blind comparison of vancomycin with bacitracin. Gastroenterology 89, 1038 45. 17. Dudley, M. N., McLaughlin, J. C., Carrington, G. et al. (1986). Oral bacitracin vs vancomycin therapy for Clostridium difficile-induced diarrhoea. A randomised double-blind trial. Archives of Internal Medicine 146, 1101 4. 18. de Lalla, F., Privitera, G., Rinaldi, E. et al. (1989). Treatment of Clostridium difficile-associated disease with teicoplanin. Antimicrobial Agents and Chemotherapy 33, 1125 7. 19. Olson, M. M., Shanholtzer, C. J., Lee, J. T. et al. (1994). Ten years of prospective Clostridium difficile-associated disease surveillance and treatment at the Minneapolis VA Medical Centre, 1982 1991. Infection Control and Hospital Epidemiology 15, 371 81. 20. Barbut, F. & Meynard, J. L. (2002). Managing antibiotic associated diarrhoea. British Medical Journal 324, 1345 6. 21. Barbut, F., Decre, D., Burghoffer, B. et al. (1999). Antimicrobial susceptibilities and serogroups of clinical strains of Clostridium difficile isolated in France in 1991 and 1997. Antimicrobial Agents and Chemotherapy 43, 2607 11. 22. Wong, S. S., Woo, P. C., Luk, W. K. et al. (1999). Susceptibility testing of Clostridium difficile against metronidazole and vancomycin by disk diffusion and Etest. Diagnostic Microbiology and Infectious Disease 34, 1 6. 23. Brazier, J. S., Fawley, W., Freeman, J. et al. (2001). Reduced susceptibility of Clostridium difficile to metronidazole. Journal of Antimicrobial Chemotherapy 48, 741 2. 24. Pelaez, T., Alcala, L., Alonso, R. et al. (2002). Reassessment of Clostridium difficile susceptibility to metronidazole and vancomycin. Antimicrobial Agents and Chemotherapy 46, 1647 50. 25. Gerding, D. N. (1997). Is there a relationship between vancomycin-resistant enterococcal infection and Clostridium difficile infection? Clinical Infectious Diseases 25, Suppl. 2, S206 10. 26. Fekety, R. (1997). Guidelines for the diagnosis and management of Clostridium difficile-associated diarrhoea and colitis. American College of Gastroenterology, Practice Parameters Committee. American Journal of Gastroenterology 92, 739 50. 27. Verbist, L. (1990). The antimicrobial activity of fusidic acid. Journal of Antimicrobial Chemotherapy 25, Suppl. B, 1 5. 28. Freeman, C. D., Klutman, N. E. & Lamp, K. C. (1997). Metronidazole. A therapeutic review and update. Drugs 54, 679 708. 29. Tvede, M. & Rask-Madsen, J. (1989). Bacteriotherapy for chronic relapsing Clostridium difficile diarrhoea in six patients. Lancet i, 1156 60. 30. O Neill, A. J., Cove, J. H. & Chopra, I. (2001). Mutation frequencies for resistance to fusidic acid and rifampicin in Staphylococcus aureus. Journal of Antimicrobial Chemotherapy 47, 647 50. 31. Ravenscroft, J. C., Layton, A. M., Eady, E. A. et al. (2003). Short-term effects of topical fusidic acid or mupirocin on the prevalence of fusidic acid resistant (Fus R ) Staphylococcus aureus in atopic eczema. British Journal of Dermatology 148, 1010 7. 32. Turnidge, J. & Collignon, P. (1999). Resistance to fusidic acid. International Journal of Antimicrobial Agents 12, Suppl.2, S35 44. 33. Österlund, A., Edén, T., Olsson-Liljequist, B. et al. (2002). Clonal spread among Swedish children of a Staphylococcus aureus strain resistant to fusidic acid. Scandinavian Journal of Infectious Diseases 34, 729 34. 216