|
Back to 2017 Posters
CURRENT SCREENING RECOMMENDATIONS FOR BARRETT'S ESOPHAGUS DO NOT DIFFERENTIATE ESOPHAGEAL CANCER PATIENTS FROM POPULATION CONTROLS
Katie S. Nason*1, Ryan M. Levy1, Inderpal S. Sarkaria1, Thomas Vaughan2, Galen E. Switzer3, James D. Luketich1
1Cardiothoracic Surgery, University of Pittsburgh, Pittsburgh, PA; 2Fred Hutchinson Cancer Center, Seattle, WA; 3Medicine, University of Pittsburgh, Pittsburgh, PA
Objective: Screening guidelines for Barrett’s esophagus to facilitate early diagnosis of esophageal adenocarcinoma continue to evolve. Based on moderate quality evidence, the AGA (2011) suggested screening patients with multiple risk factors. Our study aimed to determine whether clinically available risk factors could discriminate esophageal adenocarcinoma patients from population controls.
Methods: Patients with newly diagnosed esophageal adenocarcinoma (cases) were matched (age [+/- 5yrs], race, & sex) to population controls and completed a structured telephone interview (n=156 pairs). Presence of risk factors for Barrett’s esophagus that are clinically available to the primary physician included age 50 years or older, male sex, white race, chronic gastroesophageal reflux (GERD), and elevated body mass index (BMI). We stratified chronic GERD at two cut-points (5 and 10 or more years) and defined elevated BMI as 30 or greater. Presence of hiatal hernia and intra-abdominal body fat distribution are also in guideline, but were not included as they are not clinically available without additional testing. The number of risk factors were tallied and associations with case status assessed. Analysis was performed using statistical tests for paired data.
Results: We interviewed 156 cases and matched controls. The majority were 50 years or older (n=148 pairs; 95%), male (n=133 pairs; 85%). All were Caucasian. Both heartburn and volume regurgitation were reported by 62% of cases (n=96/153) and 48% of controls (n=73/153) while 24% (n=36) and 27% (n=41) reported one or the other symptom, respectively (p=0.02). Odds of symptoms were 2 times higher in cases than in controls (OR 2.1; 95% CI 1.1, 4.0). Symptom presence (present/absent) and duration were reported by 149 cases and 155 controls. Among cases, 62% (n=92) and 51% (n=76) reported GERD for 5 and 10 or more years, respectively vs 57% (n=88; p=0.483)
and 52% (n=80; p=1.00) of controls. Obesity was identified in 41% of cases (n=64) and 34% of controls (n=53; p=0.254). Median number of risk factors when both 5 & 10 or more years of GERD were considered was 4 for both groups (IQR 3-4; p=0.25 for 5 years and p=0.68 for 10 years of GERD). Odds of being a case did not significantly increase as number of risk factors increased (5+ years of GERD: OR 1.19 per additional risk factor; 95% CI 0.87, 1.65 & 10+ years GERD: OR 1.07; 95% CI 0.79, 1.45).
Conclusions: Esophageal cancer cases were twice as likely to report reflux symptoms in their lifetime compared to population controls. When considering other risk factors, however, increasing number of risk factors did not differentiate cancer patients from controls. This inability to discriminate suggests that clinically available predictors provide insufficient guidance for primary physician and gastroenterologists regarding which patients to screen for Barrett’s esophagus.
Back to 2017 Posters
|
