Helicobacter pylori infection in asymptomatic children: impact of epidemiologic factors on accuracy of diagnostic tests.

BACKGROUND

Childhood is a time of high risk for acquiring H. pylori infection. Accurate information on the epidemiology of the infection requires accurate diagnostic tests.

OBJECTIVE

To study the epidemiologic factors that influence the accuracy of diagnostic tests among asymptomatic children.

METHODS

Two rapid, noninvasive assays (FlexSure whole-blood test and OraSure saliva test) were evaluated using the 13C-urea breath test ((13)C-UBT) as the gold standard. The three diagnostic tests were conducted in children from 10 licensed day care centers from various locations in Houston, Texas. Demographic information and socioeconomic factors evaluated included age, gender, ethnicity, and crowded living conditions in each center. The three tests were conducted on each child on the same day. We examined the effect of epidemiologic factors on the sensitivity, the specificity, and the positive and negative predictive values of both tests.

RESULTS

Thirty-two percent of the 243 asymptomatic children who participated in the study tested positive for H. pylori using 13C-UBT. Among all the children, the sensitivities of FlexSure and OraSure were 47% (95% confidence interval, 35-57%) and 65% (95% confidence interval, 54-73%), respectively. Both tests had greater than 95% specificity. Positive and negative predictive values for FlexSure were 86% and 80%, respectively, and 98% and 86% for OraSure. Among children attending the more crowded day care centers, the sensitivities of FlexSure and OraSure were 40% and 63%, respectively, compared with 100% sensitivity for FlexSure and 80% sensitivity for OraSure among children attending the less-crowded centers. FlexSure and OraSure tests had the highest sensitivities among the 6-to 10-year-olds. Ethnicity had no effect on the sensitivity or the specificity of either test.

CONCLUSION

Epidemiologic factors influenced the accuracy of serologic tests for diagnosing H. pylori infection in children. The lowest sensitivities of both tests were among children who attended the more crowded centers and among the youngest and oldest age groups, reflecting an increased proportion of recent infections. The strategy for choosing a diagnostic test for H. pylori infection in children varies based on the epidemiologic characteristics of the studied population.