To the Editor:
—We refer to the recent study by Choinière et al., 1which contrasted the efficacy and costs of patient-controlled analgesia (PCA) with regularly administered intramuscular (IM) opioid therapy. The conclusion that PCA is more costly and does not have clinical advantages for pain management after hysterectomy deserves comment. The limitations of on-demand nurse-administered IM opioid therapy as a method of controlling postoperative pain are well recognized. 2PCA was introduced into clinical practice in the early 1980s as a means of overcoming these limitations. Personal control, rapid onset of pain relief, and timely effective analgesic therapy at the bedside are important aspects of PCA use. 3The technique adjusts for interpatient and intrapatient variations in opioid requirements. Today, many consider PCA therapy the “gold standard” of parenteral opioid administration for the control of postoperative pain. Consequently, alternative techniques of opioid administration must at least demonstrate comparative efficacy to PCA use. In our experience, a large majority of patients who have had a chance to compare IM injections and PCA prefer the latter. Unfortunately, patients in this study received only one form of treatment. Allowing patients to compare both techniques by a crossover design might have shown higher satisfaction with PCA. Furthermore, many patients report significant discomfort from repeated IM injections in the buttock area. 4Patients administered IM opioids in this study received at least 12 injections. Of interest, patients who received scheduled IM morphine required significantly more morphine in a 48-h period (132 ± 37 vs. 93 ± 50 mg;P < 0.0001) to obtain equivalent pain relief, significantly more rescue doses (30%vs. 0%), and significantly more morphine dose adjustments (63%vs. 15%;P < 0.0001). Given that both patient groups had similar outcomes in terms of analgesic efficacy, it should be noted that the IM patients received more morphine, required more rescue doses and dose adjustments, and consumed more nursing time than the PCA patients. This is perhaps a testament to the individualization, ease of administration, and overall success of PCA therapy. Although pain scores were equal in both groups, pain on movement was not measured during the first 24 h of this study. If it had been, a difference between the two techniques might have been observed. The assessment of efficacy of an analgesic technique should also include a measure of convenience of that technique to both patient and staff. The authors failed to comment on the ease of applicability of the IM regimen and on how successful the nursing staff was with regard to administering drug on time and dealing with apparently frequent problems of inadequate pain control in the IM group, particularly during the first 24 h postoperatively. In the somewhat artificial environment of a study, it may be possible to administer timely and appropriate IM injections. However, previous clinical experience indicates that this is a major issue in providing adequate postoperative pain relief and results in significant patient dissatisfaction. 5Nurse-administered IM opioid injections require adequate staffing levels to minimize delay between request and injection. In this climate of ever-shrinking health care dollars, we question whether there are sufficient numbers of nursing staff to follow this proposed IM protocol. Indeed, in many hospitals in the United States, the move seems to be toward providing higher numbers of less-skilled workers (medical assistants and licensed practical nurses rather than registered nurses) to care for patients in the postoperative setting. The burden imposed on the nursing staff in determining the success of such an IM regimen is not defined in this study. The saying “if it isn't broken don't fix it” may well apply to this study. IM injection of opioids has had its day and failed. Let us not return to the dark days of postoperative pain management without just cause or reason.