Early Complications Following Oesophagectomy for Cancer in Relation to Long-Term Healthcare Utilisation: A Prospective Population-Based Cohort Study

BACKGROUND Little is known about how early postoperative complications after oesophagectomy for cancer influence healthcare utilisation in the long-term. We hypothesised that these complications also increase healthcare utilisation long after the recovery period. METHODS This was a prospective, nationwide Swedish population-based cohort study of patients who underwent curatively intended oesophagectomy for cancer in 2001-2005 and survived at least 1 year postoperatively (n = 390). Total days of in-hospitalisation, number of hospitalisations and number of visits to the outpatient clinic within 5 years of surgery were analysed using quasi-Poisson models with adjustment for patient, tumour and treatment characteristics and are expressed as incidence rate ratios (IRR) and 95% confidence intervals (CI). RESULTS There was an increased in-hospitalisation period 1-5 years after surgery in patients with more than 1 complication (IRR 1.5, 95% CI 1.0-2.4). The IRR for the number of hospitalisations by number of complications was 1.1 (95% CI 0.7-1.6), and 1.2 (95% CI 0.9-1.6) for number of outpatient visits in patients with more than 1 complication. The IRR for in-hospitalisation period 1-5 years following oesophagectomy was 1.8 (95% CI 1.0-3.0) for patients with anastomotic insufficiency and 1.5 (95% CI 0.9-2.5) for patients with cardiovascular or cerebrovascular complications. We found no association with number of hospitalisations (IRR 1.2, 95% CI 0.7-2.0) or number of outpatient visits (IRR 1.3, 95% CI 0.9-1.7) after anastomotic insufficiency, or after cardiovascular or cerebrovascular complications (IRR 1.2, 95% CI 0.7-1.9) and (IRR 1.1, 95% CI 0.8-1.5) respectively. CONCLUSION This study showed an increased total in-hospitalisation period 1-5 years after oesophagectomy for cancer in patients with postoperative complications, particularly following anastomotic insufficiency.