FA05.01: NON-ONCOLOGICAL PULMONARY MORTALITY IS ASSOCIATED WITH OESOPHAGECTOMY FOR CANCER
Abstract Background Improvements in oesophageal cancer care have meant ever more patients are being declared cured. Whether oesophagectomy itself causes long-term non-cancer mortality is not known. This study was conducted to assess the timing and frequency of non-cancer causes of death after oesophagectomy in the UK with a population-based survey. Methods Hospital Episode Statistics provides contemporaneous admission data for all inpatient National Health Service encounters since 2000. A linked database was constructed of all HES encounters with oesophagectomy treatment codes, with Office for National Statistics mortality data, which included cause and date of death. Minimum follow-up was to 5 years. Independent variables potentially predictive of cause of death were entered into logistic regression analyses. Results There were 7204 oesophagectomy patients for which linked mortality data was available. A total of 302 died within 90 days, and a further 5874 died of primary cancer recurrence. Of the remaining 908 non-index-cancer deaths, 238 (26.2%) died of respiratory causes, 210 (23.1%) died of other cancers, 158 (17.4%) died of cardiac diagnoses and 64 (7%) died of cerebrovascular diagnoses. Survival patterns for those dying of cardiac and primary cancer recurrence were similar, with 80% occurring within 2 years of surgery. Non-cancer respiratory mortality was a later occurrence, with 80% occurring by 6 years. A pre-operative pulmonary diagnosis was associated with pulmonary mortality (OR 2.66 95% C.I. 1.49–4.77, P < 0.001), and a pre-operative ischaemic heart disease diagnosis were associated with post-operative cardiac death (OR 2.28 95% C.I. 1.13–4.59, P = 0.021). Long-term respiratory mortality was associated with inpatient respiratory complications in the index encounter (OR 2.60 95% C.I. 1.36 to 4.98, P = 0.004). Comparison to mortality rates after colectomy for cancer revealed 2-fold increased risk of non-cancer pulmonary death after oesophagectomy. Conclusion Oesophagectomy may increase the risk of non-cancer respiratory death in oesophageal cancer survivorship, with implications for peri-operative pathways and follow-up programs. Further work is needed to test whether this is a consequence of reconstruction, peri-operative complications, or progression of pre-morbid diagnoses. Disclosure All authors have declared no conflicts of interest.