Combined azithromycin and metronidazole therapy is effective in inducing remission in pediatric Crohn's disease

Transcription

1 Journal of Crohn's and Colitis (2011) 5, available at Combined azithromycin and metronidazole therapy is effective in inducing remission in pediatric Crohn's disease Arie Levine a,, Dan Turner b a Pediatric Gastroenterology and Nutrition Unit, Wolfson Medical Center, Tel Aviv University, Israel b Pediatric Gastroenterology and Nutrition, Shaare Zedek Medical Center, The Hebrew University of Jerusalem, Israel Received 25 November 2010; received in revised form 27 December 2010; accepted 18 January 2011 KEYWORDS Crohn's disease; Inflammatory bowel disease; Child; Therapy; Antibiotics; Azithromycin; Apoptosis Abstract Background: Crohn's disease (CD) is characterized by an aberrant response to the gut microbiota. We aimed to assess whether azithromycin based therapy is effective in inducing remission in CD, due to its effect in inducing apoptosis and efficacy against biofilms and intracellular bacteria. Methods: Retrospective analysis of patients treated with an 8 week course of combined azithromycin and metronidazole. Patients were included if they had active CD defined as pediatric CD activity index (PCDAI) 10, and were not receiving any other medication for inducing remission in active disease. PCDAI score and CRP were recorded at baseline and 8 weeks thereafter. Results: Thirty two patients (mean age 13.1 ± 3.9, mean duration of disease 0.65 years) were included, of whom 21 (66%) entered clinical remission (PCDAI b 10) after 8 weeks of treatment. The mean age at treatment and duration of disease did not differ between patients entering remission and those unresponsive to therapy. CRP, normalized in 54% of patients with elevated CRP at baseline. Factors associated with lack of response were a more severe disease (reflected by higher PCDAI and CRP values at baseline), presence of arthritis and extensive disease (ileocolonic, or prominent upper intestinal disease). Conclusions: An 8 week course of azithromycin and metronidazole therapy may be effective in inducing clinical remission in mild-moderate luminal CD in children and young adults European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. 1. Introduction Corresponding author at: Pediatric Gastroenterology and Nutrition Unit, Wolfson Medical Center, Sackler School of Medicine, Tel Aviv University, Israel. Tel.: , ; fax: address: alevine@wolfson.health.gov.il (A. Levine). Growing evidence suggests that Crohn's disease (CD) results from a defective or inappropriate response to the gut microbiota. Bacteria have been shown to reside on and invade epithelial cells of inflamed mucosa and granulomas and to /$ - see front matter 2011 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. doi: /j.crohns

2 Azithromycin and metronidazole theraphy is effective in inducing remission in pediatric CD 223 replicate inside macrophage phagolysosomes. 1 8 Patients with CD often have circulating antibodies against bacterial antigens. 4,8 Several disease susceptibility genes in CD are associated with innate immunity, recognition of bacterial pathogens, and handling of intracellular bacteria Asecond derangement that may be seen in association with CD is defective apoptosis of activated T cells, possibly via defective NOD2/CARD15 that may prevent apoptosis of dendritic cells. 12,16 These data might suggest that constant bacterial triggering of the immune system combined with the inability to eradicate bacteria and down regulate T cell activation may be the underlying reason for disease activity. The major flaw in this proposed pathogenesis to date has been the relative lack of efficacy of antibiotics. Multiple previous studies failed to show a benefit of antibiotic therapy for inducing remission in CD, 18,19 with the exception of one small study. 17 Subsequently, recent reviews and guidelines did not recommend antibiotic therapy for induction of remission. 19,20 These studies focused primarily on the azole family of antibiotics (mainly metronidazole) and quinolones. Azithromycin has never been studied to date despite its several intriguing properties that make it an attractive candidate for inducing remission. It has excellent intracellular penetration and with its high luminal concentrations, it may be ideal for acting against the persistent intracellular bacteria in CD that, in turn, may trigger the on-going inflammation. It also acts on biofilms which may be resistant to many types of antibiotics, 7 and is a potent activator of apoptosis of T cells Combined with metronidazole, it covers a wide variety of bacteria that colonize the small and large intestine. Several years ago one of the authors (AL) encountered several patients with Crohn's colitis refractory to medical therapy who responded to azithromycin based therapy. The index case was a 7 year old girl with CD colitis from age 1 year, who had required a hemicolectomy at age 3. She had not responded previously to antibiotics. A colonoscopy at the time of treatment displayed active colitis. She received azithromycin based therapy for 8 weeks, with complete remission and mucosal healing, the remission lasted 1 year, and two subsequent relapses over the second year responded to a repeat course of azithromycin and metronidazole. This led us to use of azithromycin based therapy, using the same protocol, as the preferred antibiotic therapy in our practice. The aim of this study was to review our experience with a combination treatment of azithromycin and metronidazole for induction of remission in active CD, following a standardized protocol. 2. Methods This is a retrospective report of our experience with azithromycin based antibiotic therapy for active CD in children and young adults, over a five year period. This study was approved by the hospital's ethical committee. All underwent an ileocolonoscopy and small bowel imaging at diagnosis, and all except three underwent gastroscopy. CD was confirmed by established criteria based on clinical, radiological, endoscopic and histopathological findings. 24 Location of disease was defined by macroscopic involvement using the Montreal classification. 25 Children and young adults 21 years with active CD, defined as a PCDAI 10, 26 were included. Patients were excluded if they received any other medication for inducing remission (i.e. corticosteroids, biologics and nutritional therapy above 50% of their daily requirements) or had perianal or fistulizing disease. Concurrent maintenance therapy with immunomodulators was allowed, as they were not expected to induce remission in active disease before the time of our endpoint at 8 weeks. All patients had negative results on stool culture and parasites prior to therapy. All patients received azithromycin mg/kg day up to a maximal dose of 500 mg, once daily, for five consecutive days per week for 4 weeks, and three times a week for the following 4 weeks. This was used in conjunction with metronidazole mg/kg/day in two divided doses, given daily for 8 weeks. Patients intolerant to either medication (e.g. nausea or abdominal pain) had the dose reduced by 25%. Our primary endpoint was remission rate after 8 weeks of therapy (defined as PCDAIb10 or b7.5 without the height item) 25 and secondary endpoints were normalization of CRP (defined as 0.5 mg/dl), improvement rate (defined as a drop in the PCDAI of at least 12.5 points 25 ), change in standard blood tests from baseline and side effects Statistical analysis Data are presented as means (±standard deviation), or medians (interquartile range (IQR)), as appropriate for the distribution normality. Unpaired categorical data were compared using X 2 or Fisher's exact test as appropriate. Unpaired Student's t-test, or Wilcoxon rank sum test was used to compare unpaired continuous data, and paired Student's t-test for paired data, such as PCDAI score before and after treatment. Following a modified intention to treat (ITT) principle, all patients receiving at least one week of antibiotic therapy were included in the analysis. All comparisons were made using two sided significance levels of p b Statistical analyses were performed using SPSS V Results A total of 32 eligible children were treated with the antibiotic protocol, of whom 15 (47%) had mild disease (PCDAI points), 12 (37%) had moderate disease ( points) and 5 (16%) had severe disease (N37.5 points) (Table 1).These children and young adults comprise all patients treated that met inclusion and exclusion criteria. The mean PCDAI score dropped from 28±10 before starting therapy to 8.6 ±8.3, 8 weeks later (Pb0.001). Twenty one patients (66%) entered remission and 11 (34%) did not. Of the 11 children who did not enter remission, one discontinued treatment due to severe C. difficile infection (regarded as failure by the ITT principal), and all others were with mild disease (mean PCDAI score 20 ± 5); 7/11 (64%) improved. Thus, only 4 (13%) of the total 32 children, including the one with the C. difficile, did not improve at all. All blood tests improved from baseline to week 8 (CRP declined from 3.2 ±2.7 to 1.2 ±2.6 mg/dl (P=0.04), ESR from 44±19 to 26±12 (Pb0.001), albumin increased from 3.8 ±0.6

3 224 A. Levine, D. Turner Table 1 Characteristics of the 32 included children counts (%), medians (interquartile range) or means ± SD, are presented as appropriate for the data distribution. Total cohort (n =32) Remission (n =21) Non remission (n =11) P value Males 22 (69%) 15 (71%) 7 (64%) NS Age (years) 13.1 ± ± ± 3.7 NS Range Disease duration (years) 0.65 (0 7.3) 0.7 (0 1.7) 0.5 (0 7.3) New onset disease 11 (34%) 6 (29%) 5 (46%) NS Location (Montreal) L1 8 (25%) 6 (29%) 2 (18%) NS L2 7 (22%) 7 (33%) 0 (0%) NS L3 14 (44%) 7 (33%) 7 (64%) P=0.14 b L4 a 13 (41%) 6 (29%) 7 (64%) P=0.07 Extensive (L3+L4) 4 (13%) 1 (5%) 3 (27%) P=0.11 Known fistulizing or stricturing disease 6 (19%) 4 (19%) 2 (18%) NS Baseline PCDAI 28± 10 25± 10 34± 6 P = Mild 15 (47%) 13 (62%) 2 (18%) Moderate 12 (37%) 6 (29%) 6 (55%) P=0.025 Severe 5 (16%) 2 (9%) 3 (27%) Baseline weight (kg) 38± ± ± 11.9 NS Baseline CRP (mg/dl) 3.2 ± ± ±2.1 P = 0.16 Baseline ESR (mm/h) 44± 19 44± 20 46± 17 NS Baseline albumin (g/dl) 3.8 ± ± ±0.4 NS Baseline hemoglobin (g/dl) 11.3 ± ± ± 1 NS NS=not significant (i.e. PN0.2); CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; and PCDAI, pediatric Crohn's disease activity index; and Bold data denotes statistically significant variable. a Not mutually exclusive. b L3 versus L1 or L2. to 4 ±0.4 g/dl (P=0.013), and hemoglobin increased from 11.3±1.2 to 11.8 ±0.8 g/dl (P=0.001)). CRP was evaluated at baseline in 24/32 patients, and in 31/32 patients at 8 weeks. Forty seven percent of patients had normal values of CRP ( 0.5 mg/dl) after treatment compared with only 14% at baseline (Pb0.001). Among patients with a baseline CRP N0.5 mg/dl (24 of the total 32), 13 (54%) had a normal CRP at 8 weeks. Only one patient with CD colitis had a repeat colonoscopy after treatment which showed near complete mucosal healing. The main factor that predicted response was a more extensive or severe disease at baseline. Those who did not enter remission tended to have more extensive disease (small bowel and colonic and upper gastrointestinal involvement) than those who entered remission, but this did not reach significance, likely due to the small sample size (Table 1). The remission rates for L1 (i.e. ileal disease only), L2 (i.e. colonic disease) and L3 (i.e. ileocolonic) was 100% (7/ 7), 75% (6/8) and 50% (7/14) respectively. Patients who failed to enter remission had more severe disease at baseline as convincingly reflected by the PCDAI score and, numerically, by the CRP values (Table 1). Nonetheless, two of the five patients with severe disease at baseline, did manage to enter remission. Seven patients developed antibiotic associated adverse events. Only one was judged to be more than mild event in a patient who developed severe pseudomembranous colitis two weeks after commencing therapy. She required admission and eventually recovered. Two patients developed abdominal pain, and continued with reduced dose of azithromycin and metronidazole; both responded with complete remission. One patient discontinued metronidazole secondary to nausea. Two females developed fungal and perianal dermatitis each, responding to topical therapy. One patient developed antibiotic associated diarrhea, which ceased once the antibiotic was discontinued. 4. Discussion The results of this uncontrolled cohort suggest that combined azithromycin metronidazole based protocol may be effective in inducing remission in children and young adults with active CD. The remission rate was 66% and a significant proportion of the patients achieved normalization of CRP. This was especially true in patients with mild to moderate inflammatory disease involving primarily the terminal ileum and colon. Patients with extensive disease involving all segments of the GI tract, defined as L3+L4, had the lowest remission rate, 1/4 (25%). None of the patients with active joint disease entered a complete remission. Based on what we know today about the genetic basis of the disease and the presence of bacteria in macrophages and granulomas, it may be possible to achieve remission by addressing the initial trigger of bacterial adherence or invasion, instead of targeting the end response of the inflammatory process. If this is indeed the sequence of events, our present default medication, corticosteroids, may have a detrimental effect on disease course by further inhibiting the immune system, preventing mucosal healing, and allowing more bacteria to penetrate while triggering new clones of activated T cells. Indeed, in a propensity

4 Azithromycin and metronidazole theraphy is effective in inducing remission in pediatric CD 225 score-controlled analysis, previous use of corticosteroids has been shown to be a significant risk factor for complicated CD, independent of disease activity. 27 Antibiotics have been previously investigated in CD, with a relatively minor success. Metronidazole may cause a response in colonic disease, or if fistulae are present. 19,20 Quinolones have also been used, but with mixed success The nonabsorbable antibiotic rifaximin was recently shown to be effective in inducing remission in 52% of adult patients with mild to moderately active disease, using twice daily dosing. 28 The results of the rifaximin study and ours might indicate that antibiotics could play a more substantial role in the treatment of active CD. Factors such as the choice of antibiotic, dose used and site of disease may affect outcomes. Azithromycin has several unique properties that appear useful in the treatment of CD by providing qualities that could decrease bacterial triggering, while exerting a prolonged effect on activated T cells and neutrophils. It causes apoptosis of activated lymphocytes and neutrophils, while other antibiotics such as, beta-lactams, and quinolones do not. 21 Other medications that induce apoptosis such as corticosteroids and anti TNF alpha antibodies are also very successful in inducing a clinical remission and decreasing inflammatory markers. It has been postulated that the effect of azithromycin involves down regulation of Bcl xl 22 which can last up to a month after the last dose. 23 Another potential advantage of azithromycin, is its high luminal concentrations, similar to rifaximin; approximately half of an oral dose is not absorbed. 29 However, as opposed to rifaximin, azithromycin, has excellent penetration into cells and is effective in reducing the intracellular viabilities of Salmonella, Shigella, and different pathogenic E. coli strains. Approximately 77% of adherent invasive E. coli strains isolated from CD patients were sensitive to azithromycin, while 100% were sensitive to quinolones and none to clarithromycin. 30 Finally, azithromycin has a long half life and good safety profile; no adverse events were reported when administered to children with cystic fibrosis for 6 12 months in high doses These properties form the clinical and biologic basis for the apparent success of our treatment protocol. We recognize the multiple shortcomings of this report, including a retrospective design, lack of control group, and small number of patients. However, given the encouraging preliminary results, we believe that a protocol combining azithromycin and metronidazole is worth further evaluation for mild to moderately active pediatric CD involving the terminal ileum or colon. This study adds to the growing body of evidence that interventions directed at the microbiome may play a role in the emerging treatments for active CD. References 1. Sartor JB. Pathogenesis and immune mechanisms of chronic inflammatory bowel diseases. Am J Gastroenterol 1997;92: S5 S Martin HM, Campbell BJ, Hart CA, Mpofu C, Nayar M, Singh R, et al. Enhanced Escherichia coli adherence and invasion in Crohn's disease and colon cancer. Gastroenterology 2004 Jul;127(1): Ryan P, Kelly RG, Lee G, et al. Bacterial DNA within granulomas of patients with Crohn's disease detection by laser capture microdissection and PCR. Am J Gastroenterol 2004 Aug;99(8): Mow WS, Landers CJ, Steinhart AH, et al. High-level serum antibodies to bacterial antigens are associated with antibioticinduced clinical remission in Crohn's disease: a pilot study. Dig Dis Sci 2004;49: Glasser AL, Boudeau J, Branch N, Perruchot MH, Colombel JF, Darfeuille-Michaud A. Adherent invasive Escherichia coli strains from patients with Crohn's disease survive and replicate within macrophages without inducing host cell death. Infect Immun 2001 Sep;69(9): Bringer MA, Glasser AL, Tung CH, Meresse S, Darfeuille-Michaud A. The Crohn's disease-associated adherent-invasive Escherichia coli strain LF82 replicates in mature phagolysosomes within J774 macrophages. Cell Microbiol 2006;8: Martinez-Medina M, Naves P, Blanco J, et al. Biofilm formation as a novel phenotypic feature of adherent-invasive Escherichia coli (AIEC). BMC Microbiol 2009;21(9): Lodes MJ, Cong Y, Elson CO, Mohamath R, Landers CJ, Targan SR, et al. Bacterial flagellin is a dominant antigen in Crohn disease. J Clin Investig 2004;113: Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 2001 May 31;411(6837): Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 2001 May 31;411(6837): Barrett JC, Hansoul S, Nicolae DL, et al. Genome-wide association defines more than 30 distinct susceptibility loci for Crohn's disease. Nat Genet 2008;40: Zelinkova Z, Van Beelen AJ, Dekort S, et al. Muramyl dipeptideinduced gene expression in NOD2 mutant and wild type Crohn's disease patient-derived dendritic cells. Inflamm Bowel Dis 2008;14: Rioux JD, Xavier RJ, Taylor KD, et al. Genome-wide association study identifies new susceptibility loci for Crohn disease and implicates autophagy in disease pathogenesis. Nat Genet 2007;39: Kuballa P, Huett A, Rioux JD, Daly MJ, Xavier RJ. Impaired autophagy of an intracellular pathogen induced by a Crohn's disease associated ATG16L1 variant. PLoS ONE 2008;3(10): e Wehkamp J, Harder J, Weichenthal M, et al. NOD2 (CARD15) mutations in Crohn's disease are associated with diminished mucosal {alpha}-defensin expression. Gut 2004 Nov;53(11): Ina K, Itoh J, Fukushima K, et al. Resistance of Crohn's disease T cells to multiple apoptotic signals is associated with a Bcl-2/Bax mucosal imbalance. J Immunol 1999;163: Colombel JF, Lemann M, Cassagnou M, et al. A controlled trial comparing ciprofloxacin with mesalazine for the treatment of active Crohn's disease. Groupe d'etudes Thérapeutiques des Affections Inflammatoires Digestives (GETAID). Am J Gastroenterol 1999 Mar;94(3): Steinhart AH, Feagan BG, Wong CJ, Vandervoort M, Mikolainis S, Croitoru K, et al. Combined budesonide and antibiotic therapy for active Crohn's disease: a randomized controlled trial. Gastroenterology 2002;123: Sandborn WJ, Feagan BG, Lichtenstein GR. Medical management of mild to moderate Crohn's disease: evidence-based treatment algorithms for induction and maintenance of remission. Aliment Pharmacol Ther 2007;26: Travis SPL, Stange EF, Lemann M, et al. European evidence based consensus on the diagnosis and management of Crohn's disease:current management. Gut 2006;55: Kadota J, Mizunoe S, Kishi K, Tokimatsu I, Nagai H, Nasu M. Antibiotic-induced apoptosis in human activated peripheral lymphocytes. Int J Antimicrob Agents 2005;25 Mar(3):

5 226 A. Levine, D. Turner 22. Mizunoe S, Kadota J, Tokimatsu I, Kishi K, Nagai H, Nasu M. Clarithromycin and azithromycin induce apoptosis of activated lymphocytes via down-regulation of Bcl-xL. Int Immunopharmacol 2004 Sep;4(9): Culic O, Erakovic V, Cepelak I, Barisic K, Brajsa K, Ferencic Z, et al. Azithromycin modulates neutrophil function and circulating inflammatory mediators in healthy human subjects. Eur J Pharmacol 2002 Aug 30;450(3): Bousvaros A, Antonioli DA, Colletti RB, et al. Differentiating ulcerative colitis from Crohn disease in children and young adults: report of a working group of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition and the Crohn's and Colitis Foundation of America. J Pediatr Gastroenterol Nutr 2007 May;44(5): Satsangi J, Silverberg M, Vermiere S, Colombel JF. The Montreal classification of inflammatory bowel diseases; controversies, consensus and implications. Gut, 55; p Turner D, Griffiths AM, Walters TD, et al. Appraisal of the Pediatric Crohn's Disease Activity Index on four prospectively collected datasets: recommended cutoff values and clinimetric properties. Am J Gastroenterol 2010;105: Vernier-Massouille G, Balde M, Salleron J, et al. Natural history of pediatric Crohn's disease: a population based study. Gastroenterology 2008;135: Prantera C, Lochs H, Campieri M, et al. Antibiotic treatment of Crohn's disease: results of a multicentre, double blind, randomized, placebo-controlled trial with rifaximin. Aliment Pharmacol Ther 2006 Apr 15;23(8): Luke DR, Foulds G. Disposition of oral azithromycin in humans. Clin Pharmacol Ther 1997;61(6): Subramanian S, Roberts C, Hart CA, et al. Replication of colonic Crohn's disease mucosal Escherichia coli isolates within macrophages and their susceptibility to antibiotics. Antimicrob Agents Chemother 2008 February;52(2): Saiman L, Anstead M, Mayer-Hamblett N, et al. Effect of azithromycin on pulmonary function in patients with cystic fibrosis uninfected with Pseudomonas aeruginosa: a randomized controlled trial. JAMA 2010 May 5;303(17): Southern KW, Barker PM, Solis A. Macrolide antibiotics for cystic fibrosis. Cochrane Database Syst Rev 2004;2:CD Kabra SK, Pawaiya R, Lodha R, et al. Long-term daily high and low doses of azithromycin in children with cystic fibrosis: a randomized controlled trial. J Cyst Fibros 2010 Jan;9(1):17 23.