|
|
Back to 2016 Annual Meeting
Concordance between serology and urea breath test in detecting helicobacter pylori infection in an ethnically diverse morbidly obese bariatric patient population
hui zheng3, jennifer marie nichols3, christian adkisson3, tatyan Clarke2,1, Rohit Soans2,1, Michael A. Edwards*2,1
1Surgery, Temple University Hospital, Philadelphia, PA; 2Surgery, Division of General and Minimally Invasive Surgery, Lewis Katz School of Medicine, Temple University Health System, Philadelphia, PA; 3Surgery, Lewis Katz School of Medicine, Temple University, Philadelphia, PA
Background: Helicobacter pylori (HP) is a carcinogenic and prevalent infection in the US. The role of HP eradication prior to bariatric surgery is a topic of controversy. Equally controversial are the efficacy of the available non-invasive studies for the diagnosis of active HP infection, including antibody titer and urea breath test (UBT). As such, there are no standards for evaluation nor recommendations for treatment of HP in this population .The aim of this study was to determine the concordance between serology and UBT in accurately diagnosing active HP infection in an ethnically diverse obese patient population.
Methods: We performed a retrospective analysis of patients undergoing evaluation for bariatric surgery at our MBS-AQIP Comprehensive Accredited program between May 2011 and June 2015. All patients who were evaluated for HP infection by both HP antibody titer and urea breath test were included in the study.
Results: 76 patients were evaluated by both serology and UBT. HP prevalence by a positive serology was 37%. 82% were female and the mean age was 43.6 years. The ethnic distribution was 52.6% African American (AA), 30.3% Hispanic and 15.8% White. The concordance between serology and UBT were 79%, 53% and 68% for IgG, IgA and IgM, respectively. True negative rates were 96%, 82% and 78% for IgG, IgA and IgM, respectively, and true positive rates for all serology isotypes were below 50% (range 11-50%). The concordance (true positive and true negative) between IgG and UBT was significant (Figure 1). Among AA and White patients, concordance was highest between IgG and UB T (78%, 86%). For Hispanic patients, concordance was highest between IgM and UBT (82%). The positive predictive value (PPV) of the serology isotypes was poor (0-50%) among all ethnic groups. The negative predictive value (NPV) was 67-100%, with IgG having the highest NPV (92-100%) among ethnic groups (Figure 2). The overall concordance of IgG and UBT was significantly different across ethnic groups (p <.001).
Conclusion: Helicobacter pylori antibody titer has a poor PPV in our ethnically diverse, morbid obese patient population with a relatively high prevalence of HP infection and is not sufficient to make the diagnosis of active HP infection. Serology has a very high negative predictive value and is sufficient to rule-out active HP infection in this patient population.
 
Back to 2016 Annual Meeting
|
