Are we missing Gastro-Oesophageal Cancer at Endoscopy?

INTRODUCTION: Upper GI endoscopy is the principle investigation for the diagnosis of gastro-oesophageal cancer. However, cancers can be missed at endoscopy and we conducted this study to ascertain the miss-rate for initial upper GI endoscopy and whether errors were clinical or clerical. We also studied the impact of a missed cancer in terms of the chance of curative treatment.METHODS: A retrospective study was performed looking at all patients diagnosed with gastro-oesophageal cancer between January 2005 and June 2009 at Nottingham City Hospital. Each patient's electronic hospital record was searched for evidence of an OGD within a 2-year time frame prior to the histological diagnosis. In those with a history of negative endoscopy, initial clinical presentation and endoscopic findings were documented, as well as any follow-up planned. Findings and treatment outcome of the subsequent 'diagnostic' endoscopy were also recorded.RESULTS: The total number of gastro-oesophageal cancer diagnoses was 1103 patients over four and a half years. Exclusions left 1075 patients. Ninety four percent (n=1010) had their cancer diagnosis made on their first OGD. However, 65 patients had had a negative endoscopy performed within 2 years. Of those, 42 (65?) were appropriately followed up with a repeat OGD within 3 months, while the remaining 23 (2? of all cancers diagnosed) did not receive appropriate follow up. Clerical errors in follow-up accounted for 5/23 (23?), whilst the remainder were discharged and re-presented with suspicious symptoms ultimately leading to the cancer diagnosis. These 18/23 cases were therefore clinical errors. Median delay in diagnosis was 7.9 months. Of those delayed >6 months, their chance of being offered curative surgery reduced from 24? to 6?. There was a strong correlation betweenthe location of abnormal findings on the initial and diagnostic endoscopy.DISCUSSION: Curative treatment for gastro-oesophageal cancer depends on stage of disease with early tumours being more likely to be resectable and curable. Upper GI endoscopy is the gold standard investigation but we have demonstrated that cancers can be missed. Clerical errors occasionally occur and are unacceptable. Clinical errors are more common and endoscopists should have a high index of suspicion and should be prepared to repeat an endoscopy early where the findings are equivocal or where poor views are obtained.