Journal of Research in Medical Sciences (Jun 2007)
Are endoscopic findings predictive for the presence of H. pylori infection? What about indirect histologic findings?
Abstract
<font face="TimesNewRoman,Bold" size="1"><p align="left"><strong>BACKGROUND: </strong><font face="TimesNewRoman" size="2">It is still controversial whether certain endoscopic features can be used to diagnose Helicobacter pylori related gastritis. Our aim was to determine how macroscopic findings were related to histomorphological changes and the presence of H. pylori in patients undergoing endoscopy.</font></p></font><font face="TimesNewRoman,Bold" size="1"><p align="left"><strong>METHODS: </strong><font face="TimesNewRoman" size="2">The study population involved 501 consecutive gastrointestinal (GI) clinic admissions who underwent esphagogastroduodenoscopy for upper GI symptoms between October 2002 and March 2004. At least 2 antral and 2 body biopsies were obtained from each patient and were examined histologically for the presence of gastritis and were stained for H. pylori using modified Giemsa staining method. Endoscopic findings were reviewed retrospectively by two experts blinded to the H. pylori status and patients history. The endoscopic findings of gastritis, classified by a modification of the Sydney system and histological findings were determined by updated Sydney system. Statistical analysis was done using SPSS 11.</font></p></font><font face="TimesNewRoman,Bold" size="1"><p align="left"><strong>RESULTS: </strong><font face="TimesNewRoman" size="2">A total of 501 consecutive patients (256 females, 245males) ranging from 8 to 91 years (mean, 49.5 years) were studied. H. pylori was found in 326 patients (65.1%). Relative frequency of H pylori in females was 53% and in males was 47%. Rugal hypertrophy, raised erosion and bleeding were observed only in patients with H. pylori infection (specificity = 100%). Neutrophil activity also was observed only in patients with H. pylori infection. Among endoscopic findings, erythema showed a high sensitivity (81.3%) and positive predictive value (87.1%) for the diagnosis of H. pylori infection. Gastritis was present in 84.3 % of all patients and 97% (316/326) of those with H. pylori and 56.6% (99/175) of those without H. pylori. There was significant statistical correlation between H. pylori infection and gastritis (P<0.001). H. pylori was present in 76% (316/415) of gastritis patients and 5.1% (4/77) of patients without gastritis.</font></p></font><font face="TimesNewRoman,Bold" size="1"><p align="left"><strong>CONCLUSIONS: </strong><font face="TimesNewRoman" size="2">An accurate endoscopic assessment of gastritis according to the Sydney system along with the histological findings is valuable indicator of H. pylori infection.</font></p></font><font face="TimesNewRoman,Bold" size="1"><p align="left"><strong>KEY WORDS: </strong><font face="TimesNewRoman" size="2">Helicobacter pylori, gastritis, Sydney system, peptic ulcer.</font></p></font>