Associations between blood cultures after surgery for colorectal cancer and long-term oncological outcomes
Menée au Danemark à partir de données portant sur 21 349 patients atteints d'un cancer colorectal traité entre 2003 et 2013, cette étude évalue l'association entre la suspicion d'une infection ayant nécessité des cultures d'échantillons sanguins dans les 30 jours qui ont suivi l'intervention chirurgicale et la survenue de complications ou la survie à long terme
Background : Systemic inflammation for colorectal cancer may be associated with an increased risk of recurrence. This study investigated whether a clinically suspected infection, for which blood cultures were sent within 30 days after surgery for colorectal cancer, was associated with long-term oncological outcomes.
Methods : This register-based national cohort study included all Danish residents undergoing surgery with curative intent for colorectal cancer between January 2003 and December 2013. Patients who developed recurrence or died within 180 days after surgery were not included. Associations between blood cultures taken within 30 days after primary surgery and overall survival, disease-free survival and recurrence-free survival were analysed using Cox regression models adjusted for relevant clinical confounders, including demographic data, cancer stage, co-morbidity, blood transfusion, postoperative complications and adjuvant chemotherapy.
Results : The study included 21 349 patients, of whom 3390 (15·9 per cent) had blood cultures taken within 30 days after surgery. Median follow‐up was 5·6 years. Patients who had blood cultures taken had an increased risk of all‐cause mortality (hazard ratio (HR) 1·27, 95 per cent c.i. 1·20 to 1·35; P < 0·001), poorer disease-free survival (HR 1·22, 1·16 to 1·29; P < 0·001) and higher risk of recurrence (HR 1·15, 1·07 to 1·23; P < 0·001) than patients who did not have blood cultures taken.
Conclusion : A clinically suspected infection requiring blood cultures within 30 days of surgery for colorectal cancer was associated with poorer oncological outcomes.
British Journal of Surgery , résumé, 2018