Clinical efficacy of oral vancomycin compared to oral and to intravenous metronidazole in Clostridium difficile infection, 2009

Abstract number: P1584

Leitner J., Schmid D., Kuo H.W., Allerberger F., Michl V., Tesik P., Wenisch C.

Objectives: Recent reports showed frequent treatment failure of metronidazole for the treatment of Clostridium difficile infection (CDI). Therefore we wanted to compare the efficacy of oral vancomycin with oral and intravenous metronidazole in inpatients with CDI.

Methods: The source population comprised CDI patients of a Viennese hospital fulfilling the case definition given by ESCD with the need of antimicrobial therapy. Study subjects were included consecutively according to their hospital admission between November 2008 and August 2009. Exclusion criterion was the lack of 30-day follow-up. The treatment regimes oral metronidazole (group I), intravenous metronidazole (group II) and oral vancomycin (group III) were compared with regards to duration of diarrhoea following the start of antibiotic therapy, treatment failure of the first chosen antibiotic therapy regime (defined by change to an alternative antibiotic regime), frequency of recurrent CDI (i.e. defined by recurrent episode following 2 weeks previous episode) and 30-day mortality. For univariable analyses c² test, c² test for contingency tables or Fisher's exact test and the t-test were used. The Charlson Index was used for co-morbidity scoring.

Results: After the exclusion of 25 patients due to missing follow up data, a total of 63 study subjects remained for analyses. The patient characteristics of the three therapy groups did not reveal relevant imbalances in age, sex or comorbidities. There were no significant differences regarding therapy response, recurrences, and duration of diarrhoea between the treatment regimes (Table 1). In contrast, 30-day mortality was significantly higher in patients on intravenous metronidazole compared to the two oral treatment regimens (p = 0.01 using c² test for contingency tables).

Conclusion: Intravenous metronidazol was associated with a poor outcome in C. difficile infection compared to oral vancomycin and oral metronidazole.

Univariable analyses

Outcome	Therapy regime
	Metronidazole p.o. N = 47	c2-test p	Metronidazole i.v. N = 11	c2-test p	Vancomycin N = 5
	n (%)	RR (95%CI)		n (%)	RR (95%CI)		n (%)	RR (95%CI)
No response to first choice antibiotic(s)	8 (17%)	0.43 (0.12–1.48)	0.24	1 (9.1% )	0.23 (0.03–1.96)	0.21	2 (40%)	1
Recurrent episode of diarrhoea (2 weeks since the onset of the previous episode)	2 (4.3%)	n.c.	1	1 (9.1%)	n.c.	1	0	1
30-day mortality	3 (6.4%)	n.c.	1	4 (36.4%)	n.c.	0.24	0	1
	Metronidazole p.o. N = 44	t-test p	Metronidazole i.v. N = 11	t-test p	Vancomycin N = 5
	Mean	Range		Mean	Range		Mean	Range
Duration of diarrhoea (days)	14	1–82	0.86	8.4	3–24	0.12	13.4	7–21