Background: To assess the value of serological biomarker testing as a substitute for esophagogastroduodenoscopy (EGDS) in pre-operative assessment of patients referred for bariatric surgery.
Methods: Sixty-five obese patients with a mean age of 43 years (range: 21-65) and a mean body mass index (BMI) of 44 (range: 36-59) were studied. The patients were tested with a four-biomarker panel: pepsinogen I and II, gastrin-17 (basal and stimulated), and Helicobacter pylori (HP) antibodies (GastroPanel®, Biohit Oyj, Finland). On the basis of the biomarker test, the patients were classified into the HS (healthy stomach) group (n = 22) with the normal biomarker profile and the NHS (non-healthy stomach) group (n = 43). The classification of patients into HS and NHS was evaluated against the gold standard, i.e. EGDS with biopsies.
Results: The concordance (Cohen's kappa) between the biomarker test and gastric histology was 0.68; 95% CI 0.504-0.854, with an overall agreement of 84.6% (95% CI 73.9-91.4%). In the NHS group, all 43 patients had biopsy-confirmed chronic gastritis: 39 non-atrophic HP-gastritis, 4 atrophic antrum gastritis (AGA) of moderate severity.In the HS group only 6 patients had mild superficial H.pylori negative gastritis. Of the 22 HS subjects with the normal biomarker profile, 20 (31% of all 65) had no complaints either, while the remaining two had reflux symptoms with esophagitis. In the NHS group 10 patients had esophagitis and 8 had also reflux symptoms.
Conclusions: The normal biomarker profile is an excellent surrogate for healthy stomach, implicating that pre-operative EGDS could have been avoided in 31% of our asymptomatic bariatric surgery patients who had the normal biomarker profile.
Keywords: Bariatric surgery; GastroPanel; Gastroscopy; Histological evaluation; Morbid obesity; Serological biomarkers; Sydney system (USS); Test accuracy.