Peptic ulcer: H. pylori: 14C urea breath tests, Giemsa or HPS stains helped diagnosis.

Clinical bottom line (level 1b)

  1. About half of patients suspected of having H. pylori infection had it.
  2. Patients who had a positive 14 C-urea breath test had H. pylori infection (LR 8 ), and those with a negative result were less likely to have it (LR 0.04).
  3. Patients with a positive Giemsa stain had an infection (LR 8 ), and those with a negative result were less likely to have it (LR 0.04).
  4. Patients with a positive HPS stain had an infection (LR 8 ), and those with a negative result were slightly less likely to have it (LR 0.25).
Fallone et al: Clinical Investigational Medicine 1995; 18: 177-185
Expires November 2002

The study

Setting: hospital endoscopy unit, Canada

50 patients (aged mean 49 years, 54% female) presentation at endoscopy unit needing gastroscopy

Excluded if
  • <16 years old
  • serious medical condition which would make gastric biopsy dangerous (eg. bleeding diathesis, massive gastrointestinal bleed)
  • lactating
  • pregnant
  • women of childbearing age who were not using adequate birth control
  • appropriate biopsies omitted at the time of endoscopy


    • Patients were not allowed to start any new acid-suppressing medications or antibiotics between the performance of endoscopy and the breath test.
    Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
    Reference standard:
    • Histological staining with Steiner silver. The presence of H. pylori was graded: 0- no organisms; 1- <10 organisms; 2- >10 but <50 organisms in several clumps per slide; 3- >50 organisms in several clumps per slide.
    Diagnostic test: Giemsa and hematoxylin-phloxin-saffron (HPS) stains; 14-C urea breath test- patients were fasted for 12 hours prior to the test, then they produced a fasting breath sample and were given the 14-C urea. The patients then cleaned their teeth and produced breath samples at 1, 2, 5, 10, 15, 20, 25 and 30 minutes after drinking the isotope. Infected patients would exhale 14-C labelled carbon dioxide and a positive test was defined as exhalation of at least 1.5% of the administered radioactive dose at 15 minutes, less than 1.5% was considered negative.

The evidence


diagnostic test H. pylori infection no infection LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
breath test 23 1 -
(43.3 to infinity) 		100% 0.04
(0.01 to 0.28) 		4.00%
Giemsa stain 23 1 -
(43.3 to infinity) 		100% 0.04
(0.01 to 0.28) 		4.00%
HPS stain 18 6 -
(43.3 to infinity) 		100% 0.25
(0.13 to 0.50) 		19.0%
total 24 26

Comments

  1. The paper realtes to older diagnostic techniques which are rarely needed today.
  2. This a relatively small study and I would interpret the 100% true positive test result with caution.

    Citation

    1. Fallone CA, Mitchell A, Paterson WG: Determination of the test performance of less costly methods of Helicobacter pylori detection. Clinical Investigational Medicine 1995; 18: 177-185
    Contributor: Clare Wotton and Musab Hayatli, November 1999
    Reviewer: Goutham Rao

    Clinical Question.
    Patient who need gastroscopy
    Intervention or Exposure urea breath test, Giemsa stain and Hematoxylin-phloxin-saffron (HPS)
    Comparison silver stain
    Outcome detecting H. pylori