Detection of H. Pylori by PCR in Gastric Biopsies
S Kiss, V Zsikla, M Baumann, C Triller, G Cathomas. Kantonales Institute for Pathology, Liestal, Switzerland
Background: The detection of bacteria by conventional histology is considered the gold standard for the diagnosis of H. pylori infection in gastric biopsies. We recently showed, however, that in about 20% of biopsies with characteristic inflammatory pattern but no bacteria detected by conventional histology, PCR can reveal an infection by H. pylori (Zsikla et al.; Am J Surg Pathol 2006;30:242). To validate this data in the daily pathology practice, we prospectively analyzed all biopsies for the presence of H. pylori by PCR, which were considered by the signing out pathologist having an inflammatory pattern compatible with Helicobacter infection but lacking bacteria by histology.
Design: The histology of all biopsies which have been analyzed by nested PCR for Helicobacter bacteria at the Kantonales Institute of Pathology, Liestal, Switzerland between the 1. 1. 2003 to the 31.12.2007 were re-evaluated and the inflammation graded according to the Sidney system. A nested PCR for H. pylori was performed as previously described, including a control PCR to verify an adequate DNA extraction.
Results: Of the total of 360 biopsies tested, H. pylori was detected by PCR in 147 (40.8%) biopsies. The total inflammatory score was significantly higher in the PCR positive compared to the PCR negative biopsies (2.3 versus 2.8; p<0.05). In biopsies with an inflammatory score of 4, PCR for Helicobacter was positive in 48.1% compared to 38.8% of biopsies with a score of 3; however, this difference is not statistical significant. Only in biopsies with an inflammatory score of 1, a significantly lower detection rate was observed compared to the other inflammatory scores (18.6% versus 46.2%; p<0.0001).
Conclusions: Our data show that in gastric biopsies with an appropriate inflammatory pattern but lack of Helicobacter by histology, PCR detects H. pylori in about 40% of cases, confirming our previous data and validating this procedure in the setting of clinical practice. However, even in higher inflammatory scores, in more than 50% of biopsies the aetiology of the inflammation remains unclear.
Wednesday, March 11, 2009 1:00 PM
Poster Session VI # 87, Wednesday Afternoon