Treat or investigate?

Abstract

Although upper gastrointestinal endoscopy provides very useful information about organic diseases, such as ulcers or cancer, there are still some questions about the most cost-effective strategy to apply in gastroesophageal reflux disease (GERD). In other words, is it mandatory to perform endoscopy in all patients before starting anti-reflux therapy, or is it better to treat empirically with an antisecretory drug? There are several factors in support of an empirical approach for the management of GERD in clinical practice. Indeed, typical symptoms like heartburn or regurgitation are very specific for the diagnosis of GERD when they are dominant. However, heartburn and regurgitation are so frequent in the primary care setting that it is not practical or even feasible to investigate all patients. In contrast, the prevalence of esophagitis in the community is very low. At least half of the patients referred to an open-access endoscopy unit are 'endoscopy-negative' or show irrelevant findings. In fact, most patients with esophagitis have mild lesions at endoscopy, and severe complications (such as stenosis or ulcer) are very rare. Moreover, there is no evidence that mild-to-moderate esophagitis worsens with time. Last but not least, symptoms are what matter to the patient. They can adversely impact the quality of life, whether or not esophagitis is present at endoscopy. On the other hand, empirical therapy should not result in the masking of organic diseases or delaying the investigation of conditions in which late diagnosis and treatment could have an impact. The risk of masking malignancy is age, context and geographically dependent. However, there is a quite good consensus that an empirical approach for the management of GERD is acceptable if symptoms are typical, the patient is young (< 45 years old) and there are no alarm symptoms. Other concerns about empirical therapy in GERD include: (a) inappropriate prolonged use of drug therapy, (b) weakening of subsequent investigation, (c) side effects, and (d) a great likelihood of symptom recurrence, possibly resulting in postponement of endoscopy and increased direct and indirect costs (sick-leave days). Indeed, the long-term expense for different strategies depends on the cost of endoscopy (with or without sedation), the cost of physician visits and the symptom recurrence rate. If GERD is associated with a high rate of recurrences, savings may be relatively modest with an empirical strategy. On the contrary, reducing the cost of endoscopy and developing open-access endoscopy units without long waiting lists may be more cost-effective than promoting empirical treatment. In conclusion, a symptom-based strategy is certainly practical and feasible for the majority of GERD patients seen by general practitioners in the primary care setting. In these patients, a proton pump inhibitor (PPI) can be safely prescribed in empirical conditions, provided simple guidelines are followed. Symptom relief can also be used as an indication of healing in patients with mild or moderate esophagitis at initial endoscopy. In contrast, endoscopy is still indicated in patients with persistent or recurrent symptoms of GERD. The role of the PPI test for diagnosis of GERD should be further validated before being widely recommended.