Aim The aim of this study was to investigate the use of resection in a cohort of palliatively treated patients with stage IV rectal cancer. To avoid selection bias, particular attention was paid to correction for comorbidity and extent of disease. Method Patients with stage IV rectal cancer in two hospitals in Groningen were consecutively included over a 5-year period. Comorbidity was defined as major (dementia, cardiac failure or left ventricle ejection fraction < 30%, or severe chronic obstructive pulmonary disease), minor (diabetes, hypertension, mild renal disease or mild pulmonary disease) or none. The effect of patient and disease characteristics on survival was assessed using Kaplan–Meier and Cox regression analyses. Results Of 88 patients, 11 (13%) underwent elective surgical resection without chemotherapy, 15 (17%) received both elective resection and chemotherapy, 21 (24%) underwent palliative chemotherapy only and 41 (47%) had supportive care only. The extent of disease (P < 0.01), hospital (P = 0.02) and comorbidity (P = 0.04) were correlated with worse survival. Patients treated surgically survived for longer than patients treated nonsurgically, when the data were corrected for age, comorbidity, extent of disease and hospital [hazard ratio (HR) = 0.4 (95% CI = 0.2–0.7)]. Perioperative morbidity was seen in 38% of the patients, and 30-day mortality was 0%. Conclusion In this retrospective cohort, resection was associated with longer survival independently of the extent of distant metastases, age and comorbidity.