Skip navigation
Acute Pain | Chronic Pain | General

Patient-controlled analgesia in postoperative pain

Clinical bottom line:

Patient-controlled analgesia produces a modest improvement in pain relief over a 24 hour period compared with conventional analgesia. It is preferred by patients, and is not associated with additional side effects.

Patient preference for PCA had an NNT of 1.9 (1.6 to 2.5) for at least satisfactory analgesia when compared with conventional analgesia at 24 hours in patients with postoperative pain.

Check out an updated review on this topic

Patient-controlled analgesia (PCA) via electronically controlled pumps has advantages over other routes. It provides fast, easy drug administration, and pumps are easy to use and are time effective. It is thought that pain relief with PCA will be more effective and consistent, with less of the discomfort caused by conventional scheduled administration of drugs.

Systematic review

Ballantyne JC, Carr DB, Chalmers TC, Dear KB, Angelillo IF, Mosteller F. Postoperative patient-controlled analgesia: meta-analyses of initial randomized control trials. J Clin Anesth 1993; 5:182-93.

Included trials were randomised controlled trials of PCA (patient-controlled, intermittent, intravenous boluses without additional continuous infusion); postoperative pain; adult population; non-intensive care patients with no pre-existing medical condition.

Clinically relevant outcomes were examined in the analysis. Analgesic efficacy was assessed by visual analogue scale, and the 24 hour reading or mean reading over 24 hours was included in the analysis. Patient satisfaction was measured using the Eisenbach and Ellis scales or as a patient preference for PCA versus conventional therapy. The analysis examined the differences between experimental and control means and a standard error of the difference for all main outcomes except analgesic use. Where these were not available in the original trials they were calculated from an appropriate outcome.


Analgesic efficacy.

Eleven trials were included in the analysis. PCA was associated with significantly better pain relief than conventional analgesia. Although this difference was statistically robust, the clinical relevance of the improvement was modest.

Analgesic use.

Thirteen trials were included in the analysis. There was no significant difference between PCA and conventional analgesia.

Patient satisfaction.

Three trials were included in the analysis. These data had an NNT of 1.9 (1.6 to 2.5 with 95% confidence) for at least satisfactory analgesia with PCA compared with conventional analgesia at 24 hours.

Length of hospital stay.

Six trials were included in the analysis. There was no significant difference between PCA and conventional analgesia on length of hospital stay.

Adverse effects

Incidence data or mean scores were included in the analysis, depending on data presented in original trials. This included data on sedation, nausea and vomiting, respiratory depression and pruritus. No significant differences were reported in original trials or meta-analysis.

Postoperative morbidity.

Only one trial was designed to measure postoperative morbidity. This trial showed significant improvements with PCA in minute ventilation recovery, ambulatory ability and recovery of oral temperature.

Of the remaining trials, there may be a lack of sensitivity in trial design. However, three of three trials showed no significant difference in pulmonary function. One of two trials showed improvements in time to ambulation with PCA.

Duration of ileus and time to first enteral feeding were measured. One of four trials showed improvements with PCA in time to first feed and fluid intake. Three of three trials showed no significant difference between first passage of flatus or stool, comparing PCA and conventional administration.

Related topics