G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth Aikaterini Pistiki a , Irene Galani a , Emmanouel Pyleris b , Charalambos Barbatzas b , Mark Pimentel c , Evangelos J. Giamarellos-Bourboulis a,∗ a
4th Department of Internal Medicine, University of Athens, Medical School, Athens, Greece Department of Gastroenterology, Sismanogleion General Hospital, Athens, Greece c GI Motility Program, Cedars-Sinai Medical Center, Los Angeles, CA, USA b
a r t i c l e
i n f o
Article history: Received 31 July 2013 Accepted 17 December 2013 Keywords: Rifaximin Time–kill effect Enterobacteria Intestinal overgrowth
a b s t r a c t Rifaximin, a non-absorbable rifamycin derivative, has published clinical efficacy in the alleviation of symptoms in patients with irritable bowel syndrome (IBS). Small intestinal bacterial overgrowth (SIBO) is associated with the pathogenesis of IBS. This study describes for the first time the antimicrobial effect of rifaximin against SIBO micro-organisms from humans. Fluid was aspirated from the third part of the duodenum from 567 consecutive patients; quantitative cultures diagnosed SIBO in 117 patients (20.6%). A total of 170 aerobic micro-organisms were isolated and the in vitro efficacy of rifaximin was studied by (i) minimum inhibitory concentration (MIC) testing by a microdilution technique and (ii) time–kill assays using bile to simulate the small intestinal environment. At a breakpoint of 32 g/mL, rifaximin inhibited in vitro 85.4% of Escherichia coli, 43.6% of Klebsiella spp., 34.8% of Enterobacter spp., 54.5% of other Enterobacteriaceae spp., 82.6% of non-Enterobacteriaceae Gram-negative spp., 100% of Enterococcus faecalis, 100% of Enterococcus faecium and 100% of Staphylococcus aureus. For the time–kill assays, 11 E. coli, 15 non-E. coli Gram-negative enterobacteria and three E. faecalis isolates were studied. Rifaximin produced a >3 log10 decrease in the starting inoculum against most of the tested isolates at 500 g/mL after 24 h of growth. The results indicate that rifaximin has a potent effect on specific small bowel flora associated with SIBO. This conclusion should be regarded in light of the considerable time–kill effect at concentrations lower than those achieved in the bowel lumen after administration of conventional doses in humans. © 2014 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Small intestinal bacterial overgrowth (SIBO) is now widely recognised as an important cause of symptoms in a variety of gut disorders such as inflammatory bowel disease, hepatic encephalopathy and irritable bowel syndrome (IBS). In SIBO, the normal intestinal flora of the proximal intestine is altered both in quantity and quality; the flora adopts the characteristics of the large intestinal flora, and aerobic and anaerobic coliforms predominate at densities >103 CFU/mL and often exceeding 105 CFU/mL [1,2]. In a recent survey by our group, 320 consecutive patients admitted to an outpatient department underwent upper gastrointestinal tract endoscopy and fluid aspirated from the third part of the duodenum was quantitatively cultured for aerobes. SIBO by aerobic coliforms
∗ Corresponding author. Present address: 4th Department of Internal Medicine, Attikon University Hospital, 1 Rimini Street, 12462 Athens, Greece. Tel.: +30 210 58 31 994; fax: +30 210 58 32 446. E-mail address: [email protected] (E.J. Giamarellos-Bourboulis).
was diagnosed in 19.4% of patients. A positive link was also found between the presence of SIBO and a diagnosis of IBS [3]. The hypothesis that SIBO is important in IBS led to the concept that oral intake of non-absorbable antibiotics may inhibit SIBO and offer alleviation of symptoms [4,5]. Rifaximin is a non-absorbable, well-tolerated antibiotic derivative of rifamycin. It possesses a broad antimicrobial spectrum both against Gram-positive and Gram-negative bacteria colonisers of the human gut as well as against enteropathogens. Owing to its non-absorbable nature, it is widely used for the management of traveller’s diarrhoea and hepatic encephalopathy [6]. Promising results have also been shown when rifaximin is administered to patients with IBS related to SIBO; in these cases, a 14-day course of rifaximin alleviated considerably the symptoms of IBS, predominantly bloating, flatulence and abdominal pain [7]. However, in this latter prospective clinical study, diagnosis of SIBO was done by the lactulose breath test and not by direct aspiration and quantitative culture of the contents of the proximal small intestine where SIBO develops. Until now, no study has demonstrated the in vitro activity of rifaximin against isolates from subjects with SIBO. To this end, we
0924-8579/$ – see front matter © 2014 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
2
extended our previously conducted study in a larger number of patients [3] and investigated the in vitro effect of rifaximin against aerobe isolates from subjects with SIBO.
2. Patients and methods 2.1. Collection and sampling of biological material Duodenal aspirates were collected between September 2009 and December 2011 from consecutive patients undergoing upper gastrointestinal tract endoscopy in the Department of Gastroenterology of ‘Sismanogleion’ General Hospital (Athens, Greece). The study protocol was approved by the Ethics Committee of the hospital. Each patient was enrolled once in the study after providing written informed consent. Inclusion and exclusion criteria for the study and the sampling procedure from the third part of the duodenum are reported elsewhere [3]. Duodenal aspirates were transported to the laboratory within 1 h after collection and were quantitatively cultured under aerobic conditions. A 0.1 mL aliquot of the fluid was directly plated onto MacConkey agar (Becton Dickinson, Cockeysville, MD) and another 0.1 mL aliquot was consecutively diluted 1:10 four times into cation-adjusted Mueller–Hinton II broth (CA-MHB) (Becton Dickinson). A 0.1 mL aliquot of each dilution was plated onto MacConkey agar and the plates were incubated under aerobic conditions at 35 ◦ C. The number of colonies was counted in each dilution and was multiplied with the appropriate dilution factor. The number of bacteria was expressed as CFU/mL. Identification of Gram-negative bacteria was performed by API20E and API20NE systems (bioMérieux, Paris, France). Identification of Staphylococcus aureus was done by positive catalase and coagulase tests. Identification of Enterococcus spp. was performed by a negative catalase test and by bile and salt solubility. In this study, SIBO was diagnosed as the presence of large bowel-type flora in the small intestine at a density >103 CFU/mL [3]. Isolated species were stored in skim milk at −70 ◦ C until susceptibility testing. Prior to testing, isolates were subcultured twice on MacConkey agar.
2.2. Susceptibility testing Minimum inhibitory concentrations (MICs) of ampicillin, cefuroxime, gentamicin, levofloxacin and rifaximin were determined by the broth microdilution assay in a final volume of 0.1 mL of CA-MHB using one log-phase inoculum of 5 × 105 CFU/mL. Single colonies were suspended in CA-MHB and were incubated for 2 h in a shaking water-bath; this was adjusted to the test inoculum using 0.5 of the McFarland climax. Escherichia coli reference strain ATCC 25922 was run in parallel in all experiments. Water-soluble amorphous powders of ampicillin and gentamicin were purchased from AppliChem GmbH (Darmstadt, Germany) and of cefuroxime from Sigma Co. (St. Louis, MO). Levofloxacin powder was provided by Sanofi GmbH (Berlin, Germany) and rifaximin powder was provided by Alfa Wassermann SpA (Bologna, Italy). Owing to the poor solubility of rifaximin in water, the agent was first diluted in ethanol and was then added to the growth medium so that the final concentration of ethanol in the growth medium was 0.01%. Preliminary experiments showed that this concentration of ethanol did not affect visible bacterial growth. The tested concentrations of all antimicrobials ranged between 0.25 g/mL and 256 g/mL. Interpretation of results was done using the European Committee on Antimicrobial Susceptibility Testing (EUCAST) susceptibility breakpoints [8].
2.3. Time–kill assay The killing effect of rifaximin over time was studied against 11 E. coli, 15 non-E.-coli Gram-negative enterobacteria and three Enterococcus faecalis isolates. This was done in tubes with CA-MHB of a 10 mL final volume using an 8 × 105 CFU/mL log-phase starting inoculum of the studied strain prepared as mentioned above and concentrations of rifaximin >100 g/mL. Owing to the poor water solubility of rifaximin, the powder was first diluted in ethanol and was then added to CA-MHB containing 2% bile (Difco Oxgall, dehydrated fresh bile; Becton Dickinson, Le Pont-de-Claix, France). Bile was used to increase rifaximin solubility as reported elsewhere [9] but also to simulate the conditions of the small intestine. The presence of bile had no inhibitory effect on the growth of the isolates as proved by preliminary experiments (data not shown). A tube without antibiotic was also included in each experiment as a growth control. Tubes were left to incubate in a shaking water-bath at 37 ◦ C. At 0, 2, 4, 6 and 24 h of incubation, absolute colony counts were quantitatively measured as described above for the duodenal samples. Results were expressed as log10 CFU/mL. The lower limit of detection was 1 log10 CFU/mL. Any decrease in bacterial growth ≥3 log10 compared with the starting inoculum was considered a killing effect.
3. Results During the study period, 567 patients were subject to upper gastrointestinal tract endoscopy; SIBO was diagnosed by quantitative culture of the duodenal aspirate in 117 patients (20.6%). Using the Rome II criteria [3], of these 567 patients, 169 were classified as having IBS and 398 as non-IBS. IBS was found in 68 (58.1%) of 117 patients with SIBO and in 101 (22.4%) of 450 patients without SIBO (P < 0.0001 by 2 test). In 64 of the 117 patients with SIBO, SIBO was due to overgrowth of one isolate and in the other 53 patients it was due to overgrowth of two isolates. As a consequence, 170 isolates were studied in total, comprising E. coli (n = 48), Klebsiella pneumoniae (n = 32), Klebsiella oxytoca (n = 7), Enterobacter cloacae (n = 14), Enterobacter aerogenes (n = 9), Citrobacter freundii (n = 5), Proteus mirabilis (n = 2), Proteus vulgaris (n = 1), Serratia marcescens (n = 3), Pseudomonas aeruginosa (n = 10), Acinetobacter baumannii (n = 6), Stenotrophomonas maltophilia (n = 6), Moraxella spp. (n = 1), S. aureus (n = 12), E. faecalis (n = 11) and Enterococcus faecium (n = 3). No other bacterial species were isolated from any enrolled patient at densities ≥103 CFU/mL. Susceptibility testing results for all tested antimicrobials are shown in Table 1. No breakpoint is proposed by EUCAST for rifaximin. At the breakpoint of 32 g/mL proposed by others [10], rifaximin inhibited in vitro 85.4% of E. coli, 43.6% of Klebsiella spp., 34.8% of Enterobacter spp., 54.5% of other Enterobacteriaceae spp., 82.6% of non-Enterobacteriaceae Gramnegative spp., 100% of E. faecalis, 100% of E. faecium and 100% of S. aureus. However, reported levels of rifaximin in stools are very high, even reaching 8000 g/mL [11], i.e. a concentration where all of the isolates implicated in SIBO in the studied patients are inhibited. This concentration is far greater than the 32 g/mL breakpoint suggested by others [10]. This discrepancy between the used breakpoint and the stool concentrations prompted us to conduct time–kill assays of rifaximin against several SIBO isolates. These time–kill assays were run in the presence of bile to simulate the bowel environment. MICs of rifaximin for the 11 E. coli isolates used for the time–kill assays ranged between 4 g/mL and 64 g/mL. As shown in Fig. 1a, the time–kill effect of rifaximin was concentration-dependent. A bactericidal effect was observed in four isolates after 6 h of exposure to 250 g/mL and in nine
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
3
Table 1 In vitro activity of rifaximin and comparator agents against 170 isolates implicated in the pathogenesis of small intestinal bacterial overgrowth (SIBO) from 117 patients. Micro-organism (no. of isolates)
Antimicrobial agenta
MIC (g/mL) Range
% inhibited MIC50
MIC90
Escherichia coli (n = 48)
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 ≤0.25 to 4 ≤0.25 to 16 4–64
16 8 1 ≤0.25 16
>256 32 2 1 64
41.7 70.8 93.8 91.7 85.4
Klebsiella spp. (n = 39)c
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
2 to >256 2 to >256 0.5–128 ≤0.25 to 64 8–128
128 4 1 ≤0.25 64
>256 >256 4 16 128
7.7 66.7 79.5 84.6 43.6
Enterobacter spp. (n = 23)d
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
8–256 4–64 ≤0.25 to 4 ≤0.25 to 0.50 32–128
64 8 1 ≤0.25 64
256 16 2 ≤0.25 128
4.3 60.9 95.7 100 34.8
Other Enterobacteriaceae (n = 11)e
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 0.50–64 ≤0.25 to 0.50 8–128
128 16 2 ≤0.25 32
>256 >256 2 ≤0.25 64
9.1 45.5 90.9 100 54.5
Non-Enterobacteriaceae Gram-negative spp. (n = 23)f
Ampicillin Cefuroxime Gentamicin (4) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 0.50 to >256 ≤0.25 to 64 ≤0.25 to >256
>256 >256 8 1 16
>256 >256 >256 16 >256
Ampicillin (4) Cefuroxime Gentamicin (128) Levofloxacin Rifaximin (32)b
≤0.25 to 4 1 to >256 2 to >256 ≤0.25 to >256 ≤0.25 to 16
2 32 128 2 4
2 >256 256 64 16
Ampicillin (4) Cefuroxime Gentamicin (128) Levofloxacin Rifaximin (32)b
64 to >256 8 to >256 >256 4–256 8–32
256 8 >256 128 16
>256 >256 >256 256 32
100
Ampicillin Cefuroxime Gentamicin (1) Levofloxacin (1) Rifaximin (32)b
≤0.25 to 128 2 to >256 0.5–2 ≤0.25 to –64 ≤0.25
64 >256 2 32 ≤0.25
75.0 75.0 100
Enterococcus faecalis (n = 11)
Enterococcus faecium (n = 3)
Staphylococcus aureus (n = 12)
4 2 1 ≤0.25 ≤0.25
50.0g 58.8h 82.6 100 72.7 100 0.0 0.0
EUCAST, European Committee on Antimicrobial Susceptibility Testing; MIC, minimum inhibitory concentration; MIC50/90 , MIC required to inhibit 50% and 90% of the isolates, respectively. a The EUCAST breakpoint is provided in parentheses, where available. b The breakpoint for rifaximin is according to Ouyang-Latimer et al. [10]. c Species of Klebsiella pneumoniae and Klebsiella oxytoca are counted together. d Species of Enterobacter cloacae and Enterobacter aerogenes are counted together. e Species of Citrobacter freundii, Proteus mirabilis, Proteus vulgaris and Serratia marcescens are counted together. f Species of Acinetobacter baumannii, Pseudomonas aeruginosa, Stenotrophomonas maltophilia and Moraxella spp. are counted together. g No breakpoints are available from EUCAST for S. maltophilia and Moraxella spp. Only species of A. baumannii and P. aeruginosa have been counted. h No breakpoints are available from EUCAST for S. maltophilia. Only species of A. baumannii, P. aeruginosa and Moraxella spp. have been counted.
isolates after 24 h of exposure to 500 g/mL. Fig. 1b and c shows the characteristic time–effect of rifaximin against two E. coli isolates. The time–kill effect of rifaximin was also tested against three K. pneumoniae, three K. oxytoca, two E. aerogenes, three E. cloacae, two C. freundii and two S. marcescens at concentrations of 150 g/mL and 500 g/mL. MICs of rifaximin for these isolates ranged between 8 g/mL and 128 g/mL. No killing effect of rifaximin was found on C. freundii or E. aerogenes. In the other species, a killing effect of 500 g/mL rifaximin was found after 24 h of exposure. Characteristic time–kill curves are shown in Fig. 2. A similar killing effect of rifaximin was shown against all three isolates of E. faecalis studied (Fig. 3).
4. Discussion The results of the present study indicate that rifaximin has considerable antimicrobial efficacy against isolates derived from patients with SIBO. The antimicrobial effect was tested both by MIC testing and time–kill assays. Using MIC testing, rifaximin was compared with ampicillin, cefuroxime, gentamicin and levofloxacin. Rifaximin is almost totally excreted in the stool, reaching concentrations as high as 8000 g/g [11]. Based on the measured MICs, the ratio of tissue concentration/MIC for the SIBO isolates is very high. This ratio for the studied comparators is lower than that of rifaximin. This observation can easily explain the findings of one retrospective analysis on patients with SIBO and IBS treated with
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6 4
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
Fig. 1. Time–kill effect of rifaximin against 11 Escherichia coli isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) cumulative change in the growth of isolates after exposure to different concentrations of rifaximin; (b) effect of rifaximin on E. coli isolate 696 with a rifaximin MIC of 4 g/mL; and (c) effect of rifaximin on E. coli isolate 518 with rifaximin MIC of 64 g/mL. MIC, minimum inhibitory concentration.
Fig. 2. Time–kill effect of rifaximin against four Enterobacteriaceae spp. isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) Klebsiella pneumoniae isolate 760 with a rifaximin MIC of 8 g/mL; (b) Klebsiella oxytoca isolate 748 with a rifaximin MIC of 8 g/mL; (c) Enterobacter cloacae isolate 222 with a rifaximin MIC of 64 g/mL; and (d) Serratia marcescens isolate 147 with a rifaximin MIC of 64 g/mL. MIC, minimum inhibitory concentration.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
5
Fig. 3. Time–kill effect of rifaximin on three Enterococcus faecalis isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) E. faecalis isolate 361 with a rifaximin MIC of 0.25 g/mL; (b) E. faecalis isolate 380 with a rifaximin MIC of 4 g/mL; and (c) E. faecalis isolate 220 with a rifaximin MIC of 16 g/mL.
either rifaximin or other antimicrobials. This retrospective analysis showed that the clinical efficacy of rifaximin was greater than that of neomycin and -lactams in the relief of symptoms of SIBO [12]. It should, however, be underscored that the diagnosis of SIBO in this retrospective study was done by the lactulose breath test and not by direct quantitative culture of small intestinal fluid. The most striking finding of the present study was the killing effect of rifaximin over time against SIBO isolates. Testing was done in the presence of bile to mimic the conditions of the small intestine where SIBO takes place. In a recent study using one E. coli isolate from one patient with traveller’s diarrhoea, the solubility of rifaximin was increased 70–120-fold in solutions containing bile acids, which increased the antimicrobial effect of rifaximin [9,10]. One suggested mechanism was that solubilisation of rifaximin by bile acids reduces its size and facilitates its entrance into the bacterial cell. The studied inoculum for these time–kill assays was >105 CFU/mL, which is equal to the most strict cut-off point used by some authors for the definition of SIBO [1,3]. Under these conditions, the observed time–kill effect was mainly observed at concentrations of 500 g/mL, which is far lower than the reported stool concentrations [10,13]. This involved isolates of E. coli, Klebsiella spp., Enterobacter spp. and E. faecalis, which are the main SIBO pathogens. The reported time–kill effect of rifaximin on SIBO isolates explains the clinical efficacy of rifaximin on the symptoms of patients with IBS described in two phase 2 and two phase 3 double-blind, placebo-controlled clinical trials [14–16]. These trials were designed on the grounds that clinical benefit from rifaximin treatment would derive from the eradication of SIBO, which is implicated in the pathogenesis of IBS. Phase 2 trials were conducted on small numbers of patients. Patients were treated with oral rifaximin 400 mg three times daily. An improvement of global IBS symptoms was found accompanied by a decrease in the bloating score [14,15]. Benefit remained after 10 weeks of follow-up
[15]. The results of two identically designed, phase 3, double-blind, randomised trials (TARGET 1 and TARGET 2) using 550 mg of oral rifaximin three times daily for 2 weeks as therapy for patients with IBS without constipation were published recently [16]. The results showed a significant relief of global IBS symptoms during the first 4 weeks after treatment with rifaximin compared with placebo [40.8% vs. 31.2% (P = 0.01) in the TARGET 1 trial and 40.6% vs. 32.2% (P = 0.03) in the TARGET 2 trial]. Significant differences were also found in relief from bloating and abdominal pain [16]. To the best of our knowledge, this is the first study designed to validate the antimicrobial activity of rifaximin against isolates of SIBO. Previous reports on the antimicrobial activity of rifaximin refer to enteropathogens causing diarrhoea [17–19]. Even in these cases, pathogens were derived from stool samples. No study has been conducted against isolates from patients with SIBO coming not from stool but from the proximal part of the intestine where SIBO develops. The results of the present study indicate that rifaximin is potent in eliminating SIBO. This conclusion should be regarded in light of the considerable time–kill effect at the 500 g/mL concentration, which is lower than the concentration of rifaximin in the bowel lumen after administration of conventional doses in humans. Funding: This study was funded in part by a kind donation from Vianex S.A. (Athens, Greece) and in part by an unrestricted educational grant from Alfa Wassermann SpA (Bologna, Italy). Competing interests: None declared. Ethical approval: The study protocol received approval from the Ethics Committee of Sismanogleion General Hospital (Athens, Greece). References [1] Posserud I, Stotzer PO, Björnsson ES, Abrahamsson H, Simrén M. Small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Gut 2007;56:802–8.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
[2] Yamini D, Pimentel M. Irritable bowel syndrome and small intestinal bacterial overgrowth. J Clin Gastroenterol 2010;44:672–5. [3] Pyleris E, Giamarellos-Bourboulis EJ, Tzivras D, Koussoulas V, Barbatzas C, Pimentel M. The prevalence of overgrowth by aerobic bacteria in the small intestine by small bowel culture: relationship with irritable bowel syndrome. Dig Dis Sci 2012;57:1321–9. [4] Koo HL, DuPont HL. Rifaximin: a unique gastrointestinal-selective antibiotic for enteric diseases. Curr Opin Gastroenterol 2010;26:17–25. [5] Menees SB, Maneerattannaporn M, Kim HM, Chey WD. The efficacy and safety of rifaximin for the irritable bowel syndrome: a systematic review and metaanalysis. Am J Gastroenterol 2012;107:28–35. [6] Scarpinato C. Rifaximin, a poorly absorbed antibiotic: pharmacology and clinical potential. Chemotherapy 2005;51:36–66. [7] Meyrat P, Safroneeva E, Schoepfer AM. Rifaximin treatment for the irritable bowel syndrome with a positive lactulose hydrogen breath test improves symptoms for at least 3 months. Aliment Pharmacol Ther 2012;36:1084–93. [8] European Committee on Antimicrobial Susceptibility Testing. In: Clinical breakpoints v. 3.1. EUCAST; 2013 [accessed 16.12.13]. [9] Darkoh C, Lichtenberger LM, Ajami N, Dial EJ, Jiang ZD, DuPont HL. Bile acids improve the antimicrobial effect of rifaximin. Antimicrob Agents Chemother 2010;54:3618–24. [10] Ouyang-Latimer J, Jafri S, VanTassel A, Jiang ZD, Gurleen K, Rodriguez S, et al. In vitro antimicrobial susceptibility of bacterial enteropathogens isolated from international travelers to Mexico, Guatemala, and India from 2006 to 2008. Antimicrob Agents Chemother 2011;55:874–8. [11] Jiang ZD, Ke S, Palazzini E, Riopel L, DuPont H. In vitro activity and fecal concentration of rifaximin after oral administration. Antimicrob Agents Chemother 2000;44:2205–6.
[12] Yang J, Lee HR, Low K, Chatterjee S, Pimentel M. Rifaximin versus other antibiotics in the primary treatment and retreatment of bacterial overgrowth in IBS. Dig Dis Sci 2008;53:169–74. [13] Taylor DN, McKenzie R, Durbin A, Carpenter C, Haake R, Bourgeois AL. Systemic pharmacokinetics of rifaximin in volunteers with shigellosis. Antimicrob Agents Chemother 2008;52:1179–81. [14] Sharara AI, Aoun E, Abdul-Baki H, Mounzer R, Sidani S, Elhaj I. A randomized double-blind placebo-controlled trial of rifaximin in patients with abdominal bloating and flatulence. Am J Gastroenterol 2006;101: 326–33. [15] Pimentel M, Park S, Mirocha J, Kane SV, Kong Y. The effect of a nonabsorbed oral antibiotic (rifaximin) on the symptoms of the irritable bowel syndrome: a randomized trial. Ann Intern Med 2006;145:557–63. [16] Pimentel M, Lembo A, Chey WD, Zakko S, Ringel Y, Mareya SM, et al. Rifaximin therapy for patients with irritable bowel syndrome without constipation. N Engl J Med 2011;364:22–3. [17] Sierra JM, Ruiz J, Navia MM, Vargas M, Vila J. In vitro activity of rifaximin against enteropathogens producing traveler’s diarrhea. Antimicrob Agents Chemother 2001;45:643–4. ´ J, et al. In vitro [18] Sierra JM, Navia MM, Vargas M, Urassa H, Schellemberg D, Gascon activity of rifaximin against bacterial enteropathogens causing diarrhoea in children under 5 years of age in Ifakara, Tanzania. J Antimicrob Chemother 2001;47:904–5. [19] Gomi H, Jiang ZD, Adachi JA, Ashley D, Lowe B, Verenkar MO, et al. In vitro antimicrobial susceptibility testing of bacterial enteropathogens causing traveler’s diarrhea in four geographic regions. Antimicrob Agents Chemother 2001;45:212–6.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
ARTICLE IN PRESS International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents journal homepage: http://www.elsevier.com/locate/ijantimicag
In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth Aikaterini Pistiki a , Irene Galani a , Emmanouel Pyleris b , Charalambos Barbatzas b , Mark Pimentel c , Evangelos J. Giamarellos-Bourboulis a,∗ a
4th Department of Internal Medicine, University of Athens, Medical School, Athens, Greece Department of Gastroenterology, Sismanogleion General Hospital, Athens, Greece c GI Motility Program, Cedars-Sinai Medical Center, Los Angeles, CA, USA b
a r t i c l e
i n f o
Article history: Received 31 July 2013 Accepted 17 December 2013 Keywords: Rifaximin Time–kill effect Enterobacteria Intestinal overgrowth
a b s t r a c t Rifaximin, a non-absorbable rifamycin derivative, has published clinical efficacy in the alleviation of symptoms in patients with irritable bowel syndrome (IBS). Small intestinal bacterial overgrowth (SIBO) is associated with the pathogenesis of IBS. This study describes for the first time the antimicrobial effect of rifaximin against SIBO micro-organisms from humans. Fluid was aspirated from the third part of the duodenum from 567 consecutive patients; quantitative cultures diagnosed SIBO in 117 patients (20.6%). A total of 170 aerobic micro-organisms were isolated and the in vitro efficacy of rifaximin was studied by (i) minimum inhibitory concentration (MIC) testing by a microdilution technique and (ii) time–kill assays using bile to simulate the small intestinal environment. At a breakpoint of 32 g/mL, rifaximin inhibited in vitro 85.4% of Escherichia coli, 43.6% of Klebsiella spp., 34.8% of Enterobacter spp., 54.5% of other Enterobacteriaceae spp., 82.6% of non-Enterobacteriaceae Gram-negative spp., 100% of Enterococcus faecalis, 100% of Enterococcus faecium and 100% of Staphylococcus aureus. For the time–kill assays, 11 E. coli, 15 non-E. coli Gram-negative enterobacteria and three E. faecalis isolates were studied. Rifaximin produced a >3 log10 decrease in the starting inoculum against most of the tested isolates at 500 g/mL after 24 h of growth. The results indicate that rifaximin has a potent effect on specific small bowel flora associated with SIBO. This conclusion should be regarded in light of the considerable time–kill effect at concentrations lower than those achieved in the bowel lumen after administration of conventional doses in humans. © 2014 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction Small intestinal bacterial overgrowth (SIBO) is now widely recognised as an important cause of symptoms in a variety of gut disorders such as inflammatory bowel disease, hepatic encephalopathy and irritable bowel syndrome (IBS). In SIBO, the normal intestinal flora of the proximal intestine is altered both in quantity and quality; the flora adopts the characteristics of the large intestinal flora, and aerobic and anaerobic coliforms predominate at densities >103 CFU/mL and often exceeding 105 CFU/mL [1,2]. In a recent survey by our group, 320 consecutive patients admitted to an outpatient department underwent upper gastrointestinal tract endoscopy and fluid aspirated from the third part of the duodenum was quantitatively cultured for aerobes. SIBO by aerobic coliforms
∗ Corresponding author. Present address: 4th Department of Internal Medicine, Attikon University Hospital, 1 Rimini Street, 12462 Athens, Greece. Tel.: +30 210 58 31 994; fax: +30 210 58 32 446. E-mail address: [email protected] (E.J. Giamarellos-Bourboulis).
was diagnosed in 19.4% of patients. A positive link was also found between the presence of SIBO and a diagnosis of IBS [3]. The hypothesis that SIBO is important in IBS led to the concept that oral intake of non-absorbable antibiotics may inhibit SIBO and offer alleviation of symptoms [4,5]. Rifaximin is a non-absorbable, well-tolerated antibiotic derivative of rifamycin. It possesses a broad antimicrobial spectrum both against Gram-positive and Gram-negative bacteria colonisers of the human gut as well as against enteropathogens. Owing to its non-absorbable nature, it is widely used for the management of traveller’s diarrhoea and hepatic encephalopathy [6]. Promising results have also been shown when rifaximin is administered to patients with IBS related to SIBO; in these cases, a 14-day course of rifaximin alleviated considerably the symptoms of IBS, predominantly bloating, flatulence and abdominal pain [7]. However, in this latter prospective clinical study, diagnosis of SIBO was done by the lactulose breath test and not by direct aspiration and quantitative culture of the contents of the proximal small intestine where SIBO develops. Until now, no study has demonstrated the in vitro activity of rifaximin against isolates from subjects with SIBO. To this end, we
0924-8579/$ – see front matter © 2014 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
2
extended our previously conducted study in a larger number of patients [3] and investigated the in vitro effect of rifaximin against aerobe isolates from subjects with SIBO.
2. Patients and methods 2.1. Collection and sampling of biological material Duodenal aspirates were collected between September 2009 and December 2011 from consecutive patients undergoing upper gastrointestinal tract endoscopy in the Department of Gastroenterology of ‘Sismanogleion’ General Hospital (Athens, Greece). The study protocol was approved by the Ethics Committee of the hospital. Each patient was enrolled once in the study after providing written informed consent. Inclusion and exclusion criteria for the study and the sampling procedure from the third part of the duodenum are reported elsewhere [3]. Duodenal aspirates were transported to the laboratory within 1 h after collection and were quantitatively cultured under aerobic conditions. A 0.1 mL aliquot of the fluid was directly plated onto MacConkey agar (Becton Dickinson, Cockeysville, MD) and another 0.1 mL aliquot was consecutively diluted 1:10 four times into cation-adjusted Mueller–Hinton II broth (CA-MHB) (Becton Dickinson). A 0.1 mL aliquot of each dilution was plated onto MacConkey agar and the plates were incubated under aerobic conditions at 35 ◦ C. The number of colonies was counted in each dilution and was multiplied with the appropriate dilution factor. The number of bacteria was expressed as CFU/mL. Identification of Gram-negative bacteria was performed by API20E and API20NE systems (bioMérieux, Paris, France). Identification of Staphylococcus aureus was done by positive catalase and coagulase tests. Identification of Enterococcus spp. was performed by a negative catalase test and by bile and salt solubility. In this study, SIBO was diagnosed as the presence of large bowel-type flora in the small intestine at a density >103 CFU/mL [3]. Isolated species were stored in skim milk at −70 ◦ C until susceptibility testing. Prior to testing, isolates were subcultured twice on MacConkey agar.
2.2. Susceptibility testing Minimum inhibitory concentrations (MICs) of ampicillin, cefuroxime, gentamicin, levofloxacin and rifaximin were determined by the broth microdilution assay in a final volume of 0.1 mL of CA-MHB using one log-phase inoculum of 5 × 105 CFU/mL. Single colonies were suspended in CA-MHB and were incubated for 2 h in a shaking water-bath; this was adjusted to the test inoculum using 0.5 of the McFarland climax. Escherichia coli reference strain ATCC 25922 was run in parallel in all experiments. Water-soluble amorphous powders of ampicillin and gentamicin were purchased from AppliChem GmbH (Darmstadt, Germany) and of cefuroxime from Sigma Co. (St. Louis, MO). Levofloxacin powder was provided by Sanofi GmbH (Berlin, Germany) and rifaximin powder was provided by Alfa Wassermann SpA (Bologna, Italy). Owing to the poor solubility of rifaximin in water, the agent was first diluted in ethanol and was then added to the growth medium so that the final concentration of ethanol in the growth medium was 0.01%. Preliminary experiments showed that this concentration of ethanol did not affect visible bacterial growth. The tested concentrations of all antimicrobials ranged between 0.25 g/mL and 256 g/mL. Interpretation of results was done using the European Committee on Antimicrobial Susceptibility Testing (EUCAST) susceptibility breakpoints [8].
2.3. Time–kill assay The killing effect of rifaximin over time was studied against 11 E. coli, 15 non-E.-coli Gram-negative enterobacteria and three Enterococcus faecalis isolates. This was done in tubes with CA-MHB of a 10 mL final volume using an 8 × 105 CFU/mL log-phase starting inoculum of the studied strain prepared as mentioned above and concentrations of rifaximin >100 g/mL. Owing to the poor water solubility of rifaximin, the powder was first diluted in ethanol and was then added to CA-MHB containing 2% bile (Difco Oxgall, dehydrated fresh bile; Becton Dickinson, Le Pont-de-Claix, France). Bile was used to increase rifaximin solubility as reported elsewhere [9] but also to simulate the conditions of the small intestine. The presence of bile had no inhibitory effect on the growth of the isolates as proved by preliminary experiments (data not shown). A tube without antibiotic was also included in each experiment as a growth control. Tubes were left to incubate in a shaking water-bath at 37 ◦ C. At 0, 2, 4, 6 and 24 h of incubation, absolute colony counts were quantitatively measured as described above for the duodenal samples. Results were expressed as log10 CFU/mL. The lower limit of detection was 1 log10 CFU/mL. Any decrease in bacterial growth ≥3 log10 compared with the starting inoculum was considered a killing effect.
3. Results During the study period, 567 patients were subject to upper gastrointestinal tract endoscopy; SIBO was diagnosed by quantitative culture of the duodenal aspirate in 117 patients (20.6%). Using the Rome II criteria [3], of these 567 patients, 169 were classified as having IBS and 398 as non-IBS. IBS was found in 68 (58.1%) of 117 patients with SIBO and in 101 (22.4%) of 450 patients without SIBO (P < 0.0001 by 2 test). In 64 of the 117 patients with SIBO, SIBO was due to overgrowth of one isolate and in the other 53 patients it was due to overgrowth of two isolates. As a consequence, 170 isolates were studied in total, comprising E. coli (n = 48), Klebsiella pneumoniae (n = 32), Klebsiella oxytoca (n = 7), Enterobacter cloacae (n = 14), Enterobacter aerogenes (n = 9), Citrobacter freundii (n = 5), Proteus mirabilis (n = 2), Proteus vulgaris (n = 1), Serratia marcescens (n = 3), Pseudomonas aeruginosa (n = 10), Acinetobacter baumannii (n = 6), Stenotrophomonas maltophilia (n = 6), Moraxella spp. (n = 1), S. aureus (n = 12), E. faecalis (n = 11) and Enterococcus faecium (n = 3). No other bacterial species were isolated from any enrolled patient at densities ≥103 CFU/mL. Susceptibility testing results for all tested antimicrobials are shown in Table 1. No breakpoint is proposed by EUCAST for rifaximin. At the breakpoint of 32 g/mL proposed by others [10], rifaximin inhibited in vitro 85.4% of E. coli, 43.6% of Klebsiella spp., 34.8% of Enterobacter spp., 54.5% of other Enterobacteriaceae spp., 82.6% of non-Enterobacteriaceae Gramnegative spp., 100% of E. faecalis, 100% of E. faecium and 100% of S. aureus. However, reported levels of rifaximin in stools are very high, even reaching 8000 g/mL [11], i.e. a concentration where all of the isolates implicated in SIBO in the studied patients are inhibited. This concentration is far greater than the 32 g/mL breakpoint suggested by others [10]. This discrepancy between the used breakpoint and the stool concentrations prompted us to conduct time–kill assays of rifaximin against several SIBO isolates. These time–kill assays were run in the presence of bile to simulate the bowel environment. MICs of rifaximin for the 11 E. coli isolates used for the time–kill assays ranged between 4 g/mL and 64 g/mL. As shown in Fig. 1a, the time–kill effect of rifaximin was concentration-dependent. A bactericidal effect was observed in four isolates after 6 h of exposure to 250 g/mL and in nine
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
3
Table 1 In vitro activity of rifaximin and comparator agents against 170 isolates implicated in the pathogenesis of small intestinal bacterial overgrowth (SIBO) from 117 patients. Micro-organism (no. of isolates)
Antimicrobial agenta
MIC (g/mL) Range
% inhibited MIC50
MIC90
Escherichia coli (n = 48)
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 ≤0.25 to 4 ≤0.25 to 16 4–64
16 8 1 ≤0.25 16
>256 32 2 1 64
41.7 70.8 93.8 91.7 85.4
Klebsiella spp. (n = 39)c
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
2 to >256 2 to >256 0.5–128 ≤0.25 to 64 8–128
128 4 1 ≤0.25 64
>256 >256 4 16 128
7.7 66.7 79.5 84.6 43.6
Enterobacter spp. (n = 23)d
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
8–256 4–64 ≤0.25 to 4 ≤0.25 to 0.50 32–128
64 8 1 ≤0.25 64
256 16 2 ≤0.25 128
4.3 60.9 95.7 100 34.8
Other Enterobacteriaceae (n = 11)e
Ampicillin (8) Cefuroxime (8) Gentamicin (2) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 0.50–64 ≤0.25 to 0.50 8–128
128 16 2 ≤0.25 32
>256 >256 2 ≤0.25 64
9.1 45.5 90.9 100 54.5
Non-Enterobacteriaceae Gram-negative spp. (n = 23)f
Ampicillin Cefuroxime Gentamicin (4) Levofloxacin (1) Rifaximin (32)b
4 to >256 2 to >256 0.50 to >256 ≤0.25 to 64 ≤0.25 to >256
>256 >256 8 1 16
>256 >256 >256 16 >256
Ampicillin (4) Cefuroxime Gentamicin (128) Levofloxacin Rifaximin (32)b
≤0.25 to 4 1 to >256 2 to >256 ≤0.25 to >256 ≤0.25 to 16
2 32 128 2 4
2 >256 256 64 16
Ampicillin (4) Cefuroxime Gentamicin (128) Levofloxacin Rifaximin (32)b
64 to >256 8 to >256 >256 4–256 8–32
256 8 >256 128 16
>256 >256 >256 256 32
100
Ampicillin Cefuroxime Gentamicin (1) Levofloxacin (1) Rifaximin (32)b
≤0.25 to 128 2 to >256 0.5–2 ≤0.25 to –64 ≤0.25
64 >256 2 32 ≤0.25
75.0 75.0 100
Enterococcus faecalis (n = 11)
Enterococcus faecium (n = 3)
Staphylococcus aureus (n = 12)
4 2 1 ≤0.25 ≤0.25
50.0g 58.8h 82.6 100 72.7 100 0.0 0.0
EUCAST, European Committee on Antimicrobial Susceptibility Testing; MIC, minimum inhibitory concentration; MIC50/90 , MIC required to inhibit 50% and 90% of the isolates, respectively. a The EUCAST breakpoint is provided in parentheses, where available. b The breakpoint for rifaximin is according to Ouyang-Latimer et al. [10]. c Species of Klebsiella pneumoniae and Klebsiella oxytoca are counted together. d Species of Enterobacter cloacae and Enterobacter aerogenes are counted together. e Species of Citrobacter freundii, Proteus mirabilis, Proteus vulgaris and Serratia marcescens are counted together. f Species of Acinetobacter baumannii, Pseudomonas aeruginosa, Stenotrophomonas maltophilia and Moraxella spp. are counted together. g No breakpoints are available from EUCAST for S. maltophilia and Moraxella spp. Only species of A. baumannii and P. aeruginosa have been counted. h No breakpoints are available from EUCAST for S. maltophilia. Only species of A. baumannii, P. aeruginosa and Moraxella spp. have been counted.
isolates after 24 h of exposure to 500 g/mL. Fig. 1b and c shows the characteristic time–effect of rifaximin against two E. coli isolates. The time–kill effect of rifaximin was also tested against three K. pneumoniae, three K. oxytoca, two E. aerogenes, three E. cloacae, two C. freundii and two S. marcescens at concentrations of 150 g/mL and 500 g/mL. MICs of rifaximin for these isolates ranged between 8 g/mL and 128 g/mL. No killing effect of rifaximin was found on C. freundii or E. aerogenes. In the other species, a killing effect of 500 g/mL rifaximin was found after 24 h of exposure. Characteristic time–kill curves are shown in Fig. 2. A similar killing effect of rifaximin was shown against all three isolates of E. faecalis studied (Fig. 3).
4. Discussion The results of the present study indicate that rifaximin has considerable antimicrobial efficacy against isolates derived from patients with SIBO. The antimicrobial effect was tested both by MIC testing and time–kill assays. Using MIC testing, rifaximin was compared with ampicillin, cefuroxime, gentamicin and levofloxacin. Rifaximin is almost totally excreted in the stool, reaching concentrations as high as 8000 g/g [11]. Based on the measured MICs, the ratio of tissue concentration/MIC for the SIBO isolates is very high. This ratio for the studied comparators is lower than that of rifaximin. This observation can easily explain the findings of one retrospective analysis on patients with SIBO and IBS treated with
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6 4
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
Fig. 1. Time–kill effect of rifaximin against 11 Escherichia coli isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) cumulative change in the growth of isolates after exposure to different concentrations of rifaximin; (b) effect of rifaximin on E. coli isolate 696 with a rifaximin MIC of 4 g/mL; and (c) effect of rifaximin on E. coli isolate 518 with rifaximin MIC of 64 g/mL. MIC, minimum inhibitory concentration.
Fig. 2. Time–kill effect of rifaximin against four Enterobacteriaceae spp. isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) Klebsiella pneumoniae isolate 760 with a rifaximin MIC of 8 g/mL; (b) Klebsiella oxytoca isolate 748 with a rifaximin MIC of 8 g/mL; (c) Enterobacter cloacae isolate 222 with a rifaximin MIC of 64 g/mL; and (d) Serratia marcescens isolate 147 with a rifaximin MIC of 64 g/mL. MIC, minimum inhibitory concentration.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
5
Fig. 3. Time–kill effect of rifaximin on three Enterococcus faecalis isolates from patients with small intestinal bacterial overgrowth (SIBO): (a) E. faecalis isolate 361 with a rifaximin MIC of 0.25 g/mL; (b) E. faecalis isolate 380 with a rifaximin MIC of 4 g/mL; and (c) E. faecalis isolate 220 with a rifaximin MIC of 16 g/mL.
either rifaximin or other antimicrobials. This retrospective analysis showed that the clinical efficacy of rifaximin was greater than that of neomycin and -lactams in the relief of symptoms of SIBO [12]. It should, however, be underscored that the diagnosis of SIBO in this retrospective study was done by the lactulose breath test and not by direct quantitative culture of small intestinal fluid. The most striking finding of the present study was the killing effect of rifaximin over time against SIBO isolates. Testing was done in the presence of bile to mimic the conditions of the small intestine where SIBO takes place. In a recent study using one E. coli isolate from one patient with traveller’s diarrhoea, the solubility of rifaximin was increased 70–120-fold in solutions containing bile acids, which increased the antimicrobial effect of rifaximin [9,10]. One suggested mechanism was that solubilisation of rifaximin by bile acids reduces its size and facilitates its entrance into the bacterial cell. The studied inoculum for these time–kill assays was >105 CFU/mL, which is equal to the most strict cut-off point used by some authors for the definition of SIBO [1,3]. Under these conditions, the observed time–kill effect was mainly observed at concentrations of 500 g/mL, which is far lower than the reported stool concentrations [10,13]. This involved isolates of E. coli, Klebsiella spp., Enterobacter spp. and E. faecalis, which are the main SIBO pathogens. The reported time–kill effect of rifaximin on SIBO isolates explains the clinical efficacy of rifaximin on the symptoms of patients with IBS described in two phase 2 and two phase 3 double-blind, placebo-controlled clinical trials [14–16]. These trials were designed on the grounds that clinical benefit from rifaximin treatment would derive from the eradication of SIBO, which is implicated in the pathogenesis of IBS. Phase 2 trials were conducted on small numbers of patients. Patients were treated with oral rifaximin 400 mg three times daily. An improvement of global IBS symptoms was found accompanied by a decrease in the bloating score [14,15]. Benefit remained after 10 weeks of follow-up
[15]. The results of two identically designed, phase 3, double-blind, randomised trials (TARGET 1 and TARGET 2) using 550 mg of oral rifaximin three times daily for 2 weeks as therapy for patients with IBS without constipation were published recently [16]. The results showed a significant relief of global IBS symptoms during the first 4 weeks after treatment with rifaximin compared with placebo [40.8% vs. 31.2% (P = 0.01) in the TARGET 1 trial and 40.6% vs. 32.2% (P = 0.03) in the TARGET 2 trial]. Significant differences were also found in relief from bloating and abdominal pain [16]. To the best of our knowledge, this is the first study designed to validate the antimicrobial activity of rifaximin against isolates of SIBO. Previous reports on the antimicrobial activity of rifaximin refer to enteropathogens causing diarrhoea [17–19]. Even in these cases, pathogens were derived from stool samples. No study has been conducted against isolates from patients with SIBO coming not from stool but from the proximal part of the intestine where SIBO develops. The results of the present study indicate that rifaximin is potent in eliminating SIBO. This conclusion should be regarded in light of the considerable time–kill effect at the 500 g/mL concentration, which is lower than the concentration of rifaximin in the bowel lumen after administration of conventional doses in humans. Funding: This study was funded in part by a kind donation from Vianex S.A. (Athens, Greece) and in part by an unrestricted educational grant from Alfa Wassermann SpA (Bologna, Italy). Competing interests: None declared. Ethical approval: The study protocol received approval from the Ethics Committee of Sismanogleion General Hospital (Athens, Greece). References [1] Posserud I, Stotzer PO, Björnsson ES, Abrahamsson H, Simrén M. Small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Gut 2007;56:802–8.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
G Model ANTAGE-4243; No. of Pages 6 6
ARTICLE IN PRESS A. Pistiki et al. / International Journal of Antimicrobial Agents xxx (2014) xxx–xxx
[2] Yamini D, Pimentel M. Irritable bowel syndrome and small intestinal bacterial overgrowth. J Clin Gastroenterol 2010;44:672–5. [3] Pyleris E, Giamarellos-Bourboulis EJ, Tzivras D, Koussoulas V, Barbatzas C, Pimentel M. The prevalence of overgrowth by aerobic bacteria in the small intestine by small bowel culture: relationship with irritable bowel syndrome. Dig Dis Sci 2012;57:1321–9. [4] Koo HL, DuPont HL. Rifaximin: a unique gastrointestinal-selective antibiotic for enteric diseases. Curr Opin Gastroenterol 2010;26:17–25. [5] Menees SB, Maneerattannaporn M, Kim HM, Chey WD. The efficacy and safety of rifaximin for the irritable bowel syndrome: a systematic review and metaanalysis. Am J Gastroenterol 2012;107:28–35. [6] Scarpinato C. Rifaximin, a poorly absorbed antibiotic: pharmacology and clinical potential. Chemotherapy 2005;51:36–66. [7] Meyrat P, Safroneeva E, Schoepfer AM. Rifaximin treatment for the irritable bowel syndrome with a positive lactulose hydrogen breath test improves symptoms for at least 3 months. Aliment Pharmacol Ther 2012;36:1084–93. [8] European Committee on Antimicrobial Susceptibility Testing. In: Clinical breakpoints v. 3.1. EUCAST; 2013 [accessed 16.12.13]. [9] Darkoh C, Lichtenberger LM, Ajami N, Dial EJ, Jiang ZD, DuPont HL. Bile acids improve the antimicrobial effect of rifaximin. Antimicrob Agents Chemother 2010;54:3618–24. [10] Ouyang-Latimer J, Jafri S, VanTassel A, Jiang ZD, Gurleen K, Rodriguez S, et al. In vitro antimicrobial susceptibility of bacterial enteropathogens isolated from international travelers to Mexico, Guatemala, and India from 2006 to 2008. Antimicrob Agents Chemother 2011;55:874–8. [11] Jiang ZD, Ke S, Palazzini E, Riopel L, DuPont H. In vitro activity and fecal concentration of rifaximin after oral administration. Antimicrob Agents Chemother 2000;44:2205–6.
[12] Yang J, Lee HR, Low K, Chatterjee S, Pimentel M. Rifaximin versus other antibiotics in the primary treatment and retreatment of bacterial overgrowth in IBS. Dig Dis Sci 2008;53:169–74. [13] Taylor DN, McKenzie R, Durbin A, Carpenter C, Haake R, Bourgeois AL. Systemic pharmacokinetics of rifaximin in volunteers with shigellosis. Antimicrob Agents Chemother 2008;52:1179–81. [14] Sharara AI, Aoun E, Abdul-Baki H, Mounzer R, Sidani S, Elhaj I. A randomized double-blind placebo-controlled trial of rifaximin in patients with abdominal bloating and flatulence. Am J Gastroenterol 2006;101: 326–33. [15] Pimentel M, Park S, Mirocha J, Kane SV, Kong Y. The effect of a nonabsorbed oral antibiotic (rifaximin) on the symptoms of the irritable bowel syndrome: a randomized trial. Ann Intern Med 2006;145:557–63. [16] Pimentel M, Lembo A, Chey WD, Zakko S, Ringel Y, Mareya SM, et al. Rifaximin therapy for patients with irritable bowel syndrome without constipation. N Engl J Med 2011;364:22–3. [17] Sierra JM, Ruiz J, Navia MM, Vargas M, Vila J. In vitro activity of rifaximin against enteropathogens producing traveler’s diarrhea. Antimicrob Agents Chemother 2001;45:643–4. ´ J, et al. In vitro [18] Sierra JM, Navia MM, Vargas M, Urassa H, Schellemberg D, Gascon activity of rifaximin against bacterial enteropathogens causing diarrhoea in children under 5 years of age in Ifakara, Tanzania. J Antimicrob Chemother 2001;47:904–5. [19] Gomi H, Jiang ZD, Adachi JA, Ashley D, Lowe B, Verenkar MO, et al. In vitro antimicrobial susceptibility testing of bacterial enteropathogens causing traveler’s diarrhea in four geographic regions. Antimicrob Agents Chemother 2001;45:212–6.
Please cite this article in press as: Pistiki A, et al. In vitro activity of rifaximin against isolates from patients with small intestinal bacterial overgrowth. Int J Antimicrob Agents (2014), http://dx.doi.org/10.1016/j.ijantimicag.2013.12.008
