Eleven controlled trials were included, of which eight examined pain outcomes (1,004 patients), and three investigated side-effects only. Jadad quality scores were between 3 and 5 out of 5 for all trials except one which scored 0.
Patients who received hydromorphone had significantly better analgesia than those who received morphine (d=-0.266, p=0.006, eight trials). Heterogeneity was high (Ι²=73.6%).
Subgroup analyses showed that there was a significant difference favouring hydromorphone for acute pain treatment (d=-0.228, p=0.012, four trials), but no difference for chronic cancer pain (four trials).
Side effects outcomes were reported, and suggested no significant difference between hydromorphone and morphine.
Sensitivity analyses showed that the overall treatment effect did not substantially change after removal of individual studies.