Prevalence of Helicobacter pylori in peptic ulcer patients in greater Rochester, NY: is empirical triple therapy justified?

OBJECTIVES: Among patients with peptic ulcer disease, the prevalence of Helicobacter pylori has been reported to range from 80% to 90%. Thus empirical cost-effective therapy has been suggested. We surveyed patients with peptic ulcer disease in Rochester, NY.
METHODS: From two teaching hospitals all patients who had duodenal ulcers (DU) and/or gastric ulcers (GU) on esophagogastroduodenoscopy (EGD) with antral biopsy for histology for H. pylori and for rapid urease (CLO) test were included in the study. We examined a total of 160 patients with DU and 145 patients with GU, age range 18-92 yr, obtaining clinical data, race, medication profile, and history of use of nonsteroidal antiinflammatory drugs (NSAIDs). An ulcer was defined if the lesion with loss of mucosal integrity was > or = 0.5 cm, with apparent depth. H. pylori was considered present if CLO test and/or histology were positive for H. pylori. To confirm the reliability of nonuse of NSAIDs, we randomly checked blood samples of 90 such patients from the ambulatory clinic for the presence of salicylates. To identify the sensitivity of the CLO test, we performed a serology test for H. pylori antibody in 100 subjects to compare the CLO test results. Also, 500 CLO test results were compared to the histology results for H. pylori.
RESULTS: Among 160 DU patients, 16 were NSAID users with negative H. pylori and excluded from the prevalence study. Of the remaining 144 patients with DU, H. pylori was present in 88 patients (61%). When these data were analyzed according to race, H. pylori was present in 54 (52%) of 104 whites compared to 34 of 40 (85%) nonwhites (blacks, Hispanics, Asians) (p < 0.01). Among 145 GU patients 18 were NSAID users with negative H. pylori and excluded from the prevalence analysis. Of the remaining 127 patients with GU, H. pylori was present in 87 patients (61%). Among them, H. pylori was present in 46 of 87 (53%) whites, whereas 31 of 40 nonwhites (78%) were H. pylori-positive (p < 0.01). Antral histology and CLO test for H. pylori were in agreement in 92% of cases. Serology and CLO test for H. pylori were in agreement in 87% of cases. None of the randomly screened patients, including 16 ulcer patients with negative H. pylori, showed presence of salicylate in blood.
CONCLUSION: In greater Rochester, NY, where the majority of our patients with EGD were whites, the prevalence of H. pylori among ulcer patients was lower compared to other regions, particularly among whites. This suggests that an additional causative factor or factors for peptic ulcers may be present. Hence, empirical antibiotic therapy of ulcer patients without confirming the presence of H. pylori may not be justified.