Jeannette Y. Wick, RPh, MBA, FASCP
Acid reflux is not just an uncomfortable condition. Chronic acid reflux can lead to Barrett’s esophagus (BE) which is considered a premalignant condition. Barrett's esophagus is strongly associated (about 0.5% per patient-year) with esophageal adenocarcinoma, a cancer that carries a poor prognosis and is increasing in incidence in industrialized nations.
Today, most cases of BE are diagnosed using costly endoscopy. Because of its cost, use of endoscopy as a population screening tool is unrealistic. A team of researchers from the United Kingdom aimed to develop a prescreening tool to aid decision making for diagnostic referrals. PLOS One
has published their results
in its April 2014 issue.
The researchers examined a prospective cohort of 1603 patients who underwent endoscopy, and identified risk factors; they used an external sample of patients (n = 477) who did not have BE (but did have symptoms of reflux or dyspepsia) to validate their findings. They evaluated associations between demographic factors, symptoms, and medication use with the presence of BE.
They developed 2 predictive models. The first, predictive for columnar lined epithelium (CLE) of any length, included age, sex, chest pain, abdominal pain, medication for ‘‘stomach’’ symptoms, and acid reflux. The second, using a stricter definition of intestinal metaplasia with segments >2 cm, included most of the same factors. They replaced acid reflux with heartburn and weighted factors somewhat differently.
The authors note that these 2 prediction models may identify around 20% of individuals unlikely to benefit from more invasive investigation for Barrett’s esophagus. They indicate that both prescreening tools are low cost and easy to use. Although 80% of patients in whom BE is a possibility would still need to be screened, these models would spare a significant number of patients the inconvenience and cost of endoscopy. The health care system could realize considerable savings.
Experts continue to debate the need for BE screening in the general population. Reliable predictive models like these 2 may make screening more cost effective.
Ms. Wick is a visiting professor at the University of Connecticut School of Pharmacy and a freelance writer from Virginia.