S114 What proportion of patients undergoing chest X-ray for NICE lung cancer symptoms are diagnosed with lung cancer over a two year follow-up period?. (December 2018)

• Record Type: Journal Article
• Authors: Aslam, R; Kennedy, MPT; Bhartia, B; Shinkins, B; Neal, RD; Callister, MEJ
• Is Part Of: Thorax. Volume 73(2018)Supplement 4
• Page Start: A71
• Page End: A72
• Publication Date: 2018-12
• Subjects: Chest -- Diseases -- Periodicals; Thorax; Chest -- Diseases; Periodicals; Periodicals; 617.54
• DOI: 10.1136/thorax-2018-212555.120
• Languages: English
• ISSNs: 0040-6376

Abstract

Abstract : Introduction/objectives: The main route to diagnosis for lung cancer patients remains chest X-ray (CXR) performed in response to respiratory symptoms. The positive predictive value (PPV) of symptoms for lung cancer has not been studied in large prospective datasets. The Leeds self-request CXR (SR-CXR) service offers CXR for patients 50 years+who fulfil criteria from the NICE Lung Cancer Guideline (CG24). Symptoms and smoking status are collected from all patients in a standardised manner. Methods: All SR-CXR requests between 2011 and 2015 were linked to a comprehensive database of lung cancer diagnoses (populated from NLCA and NCRAS). All cases had at least 2 years follow-up. PPVs were calculated by symptom, smoking behaviour, and time since CXR for cancer diagnosis (0–3, 4–12, 13–24 months). Results: Of 8, 986 SR-CXRs carried out, 143 patients (1.6%) were diagnosed with lung cancer over the following 2 years; PPVs are shown in table 1. The highest rates of cancer diagnosis occur within 3 months of the CXR (0.7%) equating to one case per 143 CXRs. As expected, rates of lung cancer diagnosis within 3 months were much higher in ever smokers (1.0%) than never smokers (0.1%) equating to one cancer diagnosis per 101 and 728 CXRs respectively. Haemoptysis and weight loss had the highest PPVs for months 0–3 and 4–12, although the actual numbers of cancers diagnosed were small. Overall, rates of lung cancer diagnosis between 4 and 12 months were low even in ever-smokers (0.6%). Rates of lung cancer diagnosis between 13 and 24 months were largely similar between symptoms (and remained so with longer follow-up to 48 months – data not shown) and probably reflect background risk in this population. Conclusions: Although PPVs for respiratory symptoms are relatively low, the low cost and harm of CXRs would suggest that a low threshold for CXR especially in ever-smokers remains justified. Investigation with CT instead of CXR for symptoms is likely to expedite diagnosis for some patients currently diagnosed after 3 months, but these data suggest the yield of such a strategy may be relatively low for symptoms other than haemoptysis and possibly weight loss.