Introduction 2-week WAIT (2WW) referrals for suspected upper gastrointestinal (UGI) cancer guidelines was introduced by the department of health in 2000. It was meant to allow patients presenting to their general practitioners with specific symptoms suggestive of UGI cancer rapid access to endoscopy. It was anticipated that this could lead to a 20% reduction in cancer related deaths for patients under the age of 75 years by the year 2010. However, only about 4% of patients have endoscopic evidence of UGI cancer. Anecdotal evidence suggests significant pick up rate of non-UGI cancers in this patient cohort.

Aims To audit the outcomes of 2WW referrals for UGI cancer to secondary care to establish the prevalence and predictors of both upper GI and other cancers in this cohort.

Methods All patients referred under the 2 week UGI cancer referral system to Nottingham University Hospital between April 2008 and 2009 were extracted using the hospital IT search engine. Data on patient demographics, indications of referral, smoking and alcohol use, endoscopic, radiological, laboratory results as well as outcomes were collected using the hospital computerised hospital records. Data on other cancers detected within a year of referral for endoscopy were recorded. Data were analysed using PASW17 (SPSS Inc, Chicago, Illinois, USA) and multivariate logistic regression analysis was used to determine predictors of cancer.

Results There were 56 (10.5%) UGI cancers in the 530 patients referred in this period. There were 31 (6.5%) non-UGI cancers in the 474 patients who did not have a cancer noted on endoscopy. Of these the majority were colorectal cancers (14), lung cancer (4) and lymphoma (4). On multivariate analysis only iron deficiency anaemia (IDA) was associated with an increased risk of non-UGI cancer (p=0.002, OR 2.9 with 95% CI 1.2 to 6.7). The predictors of an UGI cancer on multivariate analysis were dysphagia (p=0.01, OR 2.6, 95% CI 1.3 to 5.2), weight loss (p=0.01, OR 2.6, 95% CI 1.3 to 5.2), age over 60 (p=0.01, OR 5.2, 95% CI 1.9 to 13.9) and smoking (p=0.002, OR 2.7, 95% CI 1.4 to 5.2).

Conclusion 6.5% of patients referred under the 2WW referral for UGI cancer with no endoscopic evidence of cancer harbour cancers elsewhere. The presence of IDA was significantly associated with non-UGI cancers. As previously shown, dysphagia, weight loss and advanced age were significantly associated with UGI cancer.1 Referrals under 2WW need further refinement to identify those at risk for non-UGI cancer.