The aim of this study was to assess the prognostic role of co-morbidity in severe chronic obstructive pulmonary disease (COPD). A cohort of 270 COPD patients, mean (+/-SD) age 67+/-9 yrs, consecutively discharged from a University Hospital after an acute exacerbation was studied. Mean (+/-SD) forced expiratory volume in one second (FEV1) was 34+/-16% of predicted and FEV1/forced vital capacity (FVC) was 40.5+/-13.8%. The most common co-morbid diseases were: hypertension (28%), diabetes mellitus (14%), and ischaemic heart disease (10%). Clinical, electrocardiogram (ECG), and respiratory function data taken at the time of discharge were collected from the clinical records. The Charlson's index was used to quantify co-morbidity. Follow-up was conducted by means of telephone calls. Multivariate survival analysis was used to identify the independent predictors of death. The median survival of the cohort was 3.1 yrs. Death was predicted by the following variables: age (hazard rate (HR) 1.04; 95% confidence intervals (95% CI) 1.02-1.05), ECG signs of right ventricular hypertrophy (HR 1.76; 95% CI 1.30-2.38), chronic renal failure (HR 1.79; 95% CI 1.05-3.02), ECG signs of myocardial infarction or ischaemia (HR 1.42; 95% CI 1.02-1.96), FEV1 < 590 mL (HR 1.49; 95% CI 0.97-2.27). A score based upon these variables predicted mortality at 5 yrs with a sensitivity of 63% and a specificity of 77%. Selected co-morbid diseases and electrocardiogram signs of right ventricular hypertrophy play a major prognostic role in advanced chronic obstructive pulmonary disease. The clinical assessment of patients with chronic obstructive pulmonary disease should include these important and easily measurable variables.