Are inflammatory bowel disease (IBD) and pouchitis a reactive enteropathy to group D streptococci (Enterococci)?

To the Editor:

Previously in inflammatory bowel disease (IBD) we raised the possibility that Enterococci (EC) are the missing link in the pathogenesis of IBD. We argued that IBD, ankylosing spondylitis, and psoriasis are a spectrum of the same disease.1 The evidence included a similar attack rate in identical twins, the 80% prevalence of EC in the stool, and a similar age of onset.1 Additionally, we noted the higher than expected prevalence of inflammatory arthritis and psoriasis in IBD patients (20% versus 1%) and the increasing association of the development of psoriasiform skin lesions in Crohn's disease patients treated with biologics.2 Specific genetic defects shared between these diseases have also raised the possibility that an overlap of their pathogenesis exists.3, 4 We know that pouchitis occurs at a much higher incidence in patients with IBD, immunologic skin disease, severe inflammatory joint disease when compared to patients who undergo J-pouch surgery for noninflammatory reasons like familial adenomatous polyposis (FAP).

This concept has profound implications in that if it is possible to avoid the causative agent the disease would be cured despite having the genetic predisposition. Out of desperation and frustration 1 our patients agreed to be treated using a new concept.

CASE REPORT

Our patient is a 52-year-old female with a past medical history of ulcerative colitis (UC) diagnosed at age 19. Her disease was severe and refractory to therapy. After her exacerbations were no longer controlled with medication including prednisone, she underwent a total colectomy with ileoanal anastomosis and J-pouch creation 11 years following initial diagnosis. Since surgery, her symptoms included recurrent episodes of pouchitis with ≈15 watery stools daily. She also developed spondyloarthritis following her UC diagnosis, which was being controlled with methotrexate, plaquenil, and adalimumab. Her pouchitis flares were marginally controlled with cyclical courses of ciprofloxacin and metronidazole. Each cycle consisted of a 4-week course of both antibiotics. Improvement was not reached until the end of the second week, and upon discontinuation of the regimen after the fourth week, her symptoms of pouchitis quickly returned. As her symptoms became refractory to this regimen, she was started on a 1-month trial of doxycycline 100 mg twice daily alone. In 5 days her clinical symptoms rapidly improved. Her stools became more formed and decreased in number. She began gaining weight, up to 106 pounds above her baseline of 96 pounds. Simultaneously she began a “sterile diet” in which she avoided raw fruits and vegetables to prevent recolonization with enterococcus. She was able to stop methotrexate and adalimumab for her arthritic complaints. Three months following her trial of doxycycline and a sterile diet, her symptoms continue to be well controlled.

DISCUSSION

We considered several antibiotics that are more effective than ciprofloxacin against Gram-positive bacteria. Sulfa has been used with modest success; however, along with penicillin they run a high risk of allergic reactions. Erythromycin has hepatic enzyme effects making drug interactions problematic. We chose doxycycline because it has been safely used for up to a year in the Gulf War study. We also used it in a 6-month study with rifampin to treat chlamydia in patients with post-chlamydial reactive arthritis.5 Patients need to be cautioned about upper gastrointestinal irritation and photosensitivity. Tetracyclines have other potential effects, including inhibition of metalloproteinases, but we believe it is acting as an antibiotic. To date we have no stool culture data.

The most novel aspect to this new approach tries to reduce or even eliminate alimentary exposure to EC, stressing cooked food, which presumably kills the bacteria. The problems with this approach include EC becoming resistant or ingesting resistant organisms. The possibility of multiple organisms causing IBD could lessen the efficacy of this approach. The risks of prolonged use of a cooked diet absent in fresh fruits and vegetables are unknown.

We believe a prospective controlled study initially of pouchitis patients taking doxycycline and a cooked diet versus the gold standard of ciprofloxacin and metronidazole with a standard diet should answer the question of the usefulness of this new approach.

These results need to be interpreted with caution, as pouchitis can undergo spontaneous remissions. No definite conclusions can be drawn from this preliminary experience. The major innovation in this work is not in replacing traditional antibiotics which do work, but in explaining why patients relapse after stopping them.

Stephanie Ingram MD*, Joseph M. McKinley MD*, Frank Vasey MD*, John D. Carter MD*, Patrick G. Brady MD*, * Internal Medicine, University of South Florida, Tampa, Florida.