Am Fam Physician. 2007;76(11):1645
Clinical Question
Is patient-controlled opioid analgesia more effective than conventional analgesia for postoperative pain?
Evidence-Based Answer
Patient-controlled opioid analgesia is safe and provides a statistically significant improvement in analgesia in postoperative patients, but the clinical significance of the improvement is marginal.
Practice Pointers
Patient-controlled analgesia (PCA) using opioids (e.g., morphine) has become the standard of care at many institutions for the management of postoperative pain. However, previous systematic reviews in 1992 and 2001 found conflicting results regarding the effectiveness of this approach compared with traditional analgesia.1,2 The authors of this Cochrane review identified 55 studies with a total of 3,861 patients, and they compared these studies with the 16 studies1 and 33 studies2 found in the previous systematic reviews. None of the studies were double-blinded and, in general, the studies were of poor quality.
Most studies measured pain on a visual analog scale (e.g., a scale with zero meaning the patient had no pain, and 100 meaning that the patient had severe pain) at various points in time. A change of 10 to 15 points on a 100-point visual analog scale is generally considered to be clinically significant and detectable by patients as an improvement.3 In these studies, the average pain intensity during a time period or the final pain intensity at the end of a time period was measured. The study durations generally ranged from 24 to 72 hours.
Pain intensity during the first 24 hours was eight points lower in the PCA group than in a group receiving conventional analgesia (95% CI, −12 to −4 points), nine points lower between 25 and 48 hours (95% CI, −14 to −5 points), and 13 points lower between 49 and 72 hours (95% CI, −20 to −6 points).3 Patients in the PCA group used more opioids than those in the control group over 72 hours; however, with the exception of a small increase in pruritus, there was no difference in other adverse effects such as nausea, vomiting, sedation, or urinary retention. Overall, more patients were more satisfied with the mode of analgesia in the PCA group (84 versus 65 percent; number needed to treat = 5.3).