The study compares responders and non-responders to post-operative patient-controlled analgesia (PCA) and evaluates factors that might differ between these two groups in order to identify non-responders during the early post-operative period. A prospective, randomized, double-blinded study design was used. Patients recovering from abdominal surgery were assigned to one of three treatment groups for a study period of 48 h. After titration of an individual loading dose, patients could self-administer 1 mL bolus doses(2 mg of morphine, 20 mg of tramadol or placebo) using a PCA device. Patients responding or not responding to the treatment were identified. In non-responders the escape medication was morphine.There were 96 responders and 65 non-responders. All responders showed similar pain scores, irrespective of the drug they received. Drug consumption of placebo responders was twice as high as that of opioid responders. Pain scores and analgesic consumption of non-responders were significantly higher compared with responders, although those patients received morphine. The loading dose correlated with subsequent analgesic consumption. Altogether, 89.2% of the non-responders were identified after the loading dose. Size of loading dose and pain scores during the first 30 min are useful for assessing the overall response to post-operative pain management. These factors may be valuable for predicting individual pain management.