Helicobacter pylori diagnostic tests in Colombian children

Abstract

Helicobacter pylori (H. pylori) infection is dynamic in the pediatric population, increasing with age. There are several diagnostic tests available and effective in identifying infected and uninfected people. However, there is still a need for a non-invasive, reliable, and tolerable test for children. 

Objective: To estimate the validity of H. pylori testing in children in Cali, Colombia. 

Patients and Method: A total of 236 symptomatic children under 10 years of age referred for clinical evaluation with upper endoscopy due to abdominal pain, gastroesophageal reflux, vomiting, dyspepsia, and diarrhea were included. Three diagnostic tests were used to determine H. pylori infection: [13C]-Urea breath test (UBT); H. pylori stool antigen (HpSA) test, and serum H. pylori antigen by ELISA. The validity was evaluated by sensitivity and specificity analysis, establishing the diagnosis of H. pylori infection based on the histological study of gastric mucosal biopsies as the gold standard or reference method. 

Results: The estimated prevalence of H. pylori in the three tests ranged from 1.12% to 27.3%, similar to that reported by histological evaluation (21.6%). The sensitivity found through conventional analysis (2x2) for the HpSA test was 97.9%, and for the UBT test and the serum H. pylori antigen was 87.5% and 88.2%, respectively. Higher sensitivity was found in all three tests when latent class analysis was used, especially in the HpSA test which was equal to the reference histopathological test (100%). 

Conclusion: The HpSA test showed the best discrimination in both infected and healthy children under 10 years of age in Cali, Colombia.