HELICOBACTER PYLORI INFECTION

See also "Peptic Ulcer Disease," "Nonulcer Dyspepsia," and NSAIDs

I. Pathophysiology--Ass'd with chronic gasritis, PUD, and gastric cancer

II. Diagnosis

  1. Examination of biopsy specimens
    1. Pathologic exam (Gold standard but false-negatives can occur in pts on Proton-pump Inhibitors)
    2. Urease ("CLOTest") testing on tissue specimen
    3. Culture
    4. PCR
  2. Serology
    1. Identifies infection; can't definitively identify eradication
    2. Use of serology for test-of-cure
      1. Conversion from positive to negative H. pylori IgG at 18mos after tx was 60% sensitive and 100% specific for cure as defined by no H. pylori on gastric bx at 18mos in a series of 23 pts (JAMA 280:363, 1998--AFP)
      2. Can follow IgG and IgA levels of anti-H. pylori Ab's as an alternative to repeating biopsy or breath testing (Clin. Inf. Dis. 25:1038, 1997--JW)
  3. Urea breath test
    1. Highly sensitive & specific (> 90% for both, specificity is higher--can get false-negatives from Proton-pump Inhibitors, antibiotics, & bismuth)
    2. Can be used to identify eradication
    3. Pt ingests C14 or C13-labeled urea; if H. pylori is present in the stomach, it splits the urea into ammonia + CO2, which is detected in the expired breath
    4. Ongoing proton-pump inhibitor tx reduces sensitivity of breath testing for H. pylori, possibly for 2 weeks after cessation (Ann. Int. Med. 129:547, 1998--JW)
  4. Urea blood testing
    1. Pt ingests 13-C-labeled urea and 30min later blood is drawn for 13-C-bicarbonate
    2. 89% sensitive, 96% specific c/w histology in one study (Am. J. Gastroent. 94:1522, 1999--JW)
  5. Stool antigen test (EIA)
    1. Similar sensitivity/specificity to exams of biopsy specimens in kids (J. Peds. 136:823, 2000--JW)
    2. Sensitivity of 94% and specificity of 98% for persistence of H. pylori 35d after completion of tx (Ann. Int. Med. 136:280, 2002--JW)
  6. Urine antibody testing
  7. Culture--Not commonly used; H. pylori is difficult to culture so the test is insensitive
  8. Comparisons of diagnostic modalities for H. pylori infection
    1. In a study of 316 children 2-17yo undergoing EGD with biopsy, all of whom underwent urea breath testing, urine Ab, stool Ag, and serum Ab testing, the sensitivity of urea breath testing (c/w biopsy) was highest (96%, compared with 63% for urine Ab, 73% for stool Ag, and 89% for serum Ab); specificities for all were in the 93-97% range (J. Peds. 146:164, 2005--JW)
    2. Stool Ag testing in several stufies showed similar sensitivity and specificity to urea breath testing in adults (both for pre- and post-treatment diagnosis; Lancet 354:30, 1999; Am. J. Gastroent. 95:925, 2000; BMJ 320:148, 2000--JW)

III. Treatment

  1. Benefits of treatment
    1. See sections on specific clinical conditions, e.g. Dyspepsia (Nonulcer), Peptic Ulcer Disease
    2. For prevention of gastric cancer--Questionable benefit
      1. 1630 pts with H. pylori infection, 642 w/precancerous lesions at baseline (gastric atrophy, intestinal metaplasia, or gastric dysplasia), randomized to 2wks of (omeprazole), (amoxicillin/clavulanate + metronidazole), (omeprazole, amoxicillin/clavulanate, and metronidazole), or placebo, all x 2wks.  Over mean 7.5y f/u, no diff. b etween groups in incidence of gastric cancer in overall cohort BUT sig. reduction in incidence of gastric cancer in the subgroup of those w/no precancerous lesions at study entry (JAMA 291:187, 2004--abst)
      2. In a study in 3,365 pts 35-64yo with gastric histologic abnormalities (from gastritis to dysplasia) and positive Helicobater pylori serology randomized to H. pylori eradication tx (amoxicillin + omeprazole x 2wks) vs. placebo. At 7y, active-tx group had sig. lower incidence of progression of histologic abnormalities than placebo group; no sig. diff. incidence of gastric Ca. (J. Nat. Ca. INst. 98:974, 2006-JW).
      3. In a meta-analysis of 6 randomized trials of H. pylori looking at subsequent incidence of gastric Ca, over 4-10y f/u, H. pylori tx was associated with a sig. reduction in gastric Ca risk (RR 0.65), though many studies were not of high methodologic quality (Ann. Int. Med. 151:121, 2009-JW)
  2. Single-antibiotic regimens are less effective than two- or three-antibiotic regimens (Arch. Int. Med. 158:1651, 1998--AFP, and others)
  3. 7d regimens often less effective than 14d regimens.
  4. Antibiotic-resistant strains starting to be identified as of 2000
  5. Many regimens have been studied; these may be skewed by local patterns of antimicrobial resistance
  6. Factors to consider in selecting a regimen:
    1. Local resistance patterns
    2. Prior exposure to abx
    3. Patient compliance factors
    4. Cost
  7. Recommended by UW GI as of 1/2000: 10-14d of the following (Ass'd with > 90% eradication rates; eradication rates may be slightly higher with 14d tx per Alimentary and Pharmacology and Therapeutics. Issue 10 pp 1029-1033. 1996, cited in UW GI Guidelines):
    1. A Proton-pump Inhibitor (see link for doses)
    2. Clarithromycin 500mg BID
    3. Either Metronidazole 500mg BID or Amoxicillin 1g BID
      1. Though not mentioned in UW guidelines, using Bismuth subsalicylate 524mg BID instead of either of these was ass'd wtih similar eradcation rates in one randomized trial of 112 pts with documented H. pylori infection, with PUD, h/o PUD, or nonulcer dyspepsia who hadn't had previous tx for H. pylori (Am. J. Gastroent. 92:1483, 1997--AFP)
  8. Other specific regimens
  1. "Triple therapy"--Mean eradication rates 76-88%; 10-15d of: