Current issues of ACP Journal Club are published in Annals of Internal Medicine


Subcutaneous and intravenous hydromorphone infusions were equally effective for chronic cancer pain

ACP J Club. 1991 July-Aug;115:16. doi:10.7326/ACPJC-1991-115-1-016

Source Citation

Moulin DE, Kreeft JH, Murray-Parsons N, Bouquillon AI. Comparison of continuous subcutaneous and intravenous hydromorphone infusions for management of cancer pain. Lancet. 1991 Feb 23;337:465-8.



To compare the effectiveness of continuous subcutaneous with intravenous hydromorphone infusion for chronic cancer pain.


Randomized, double-blind, double-dummy, crossover trial with 48-hour periods of infusion.


Hospital in London, Ontario, Canada.


53 patients with pain from advanced cancer had been started on continuous subcutaneous opioid infusions because of side effects from oral or rectal analgesics. 20 patients (10 men and 10 women; mean age 62 y; mean weight 61 kg) participated in the study; 23 patients refused and 10 were excluded because of metabolic encephalopathy or brain metastases. Principal causes of pain were bone metastases, soft-tissue infiltration of chest wall or retroperitoneum, bowel obstruction, and brachial plexopathy.


Using subcutaneous infusion, the maximum analgesic dosage of hydromorphone hydrochloride with minimal side effects was established before the trial. For each 48-hour trial period, new infusion sites were started on the anterior chest wall or abdomen; 2 programmable infusion pumps simultaneously delivered hydromorphone and saline placebo, intravenously and subcutaneously in random order. After a 24-hour washout period in which morphine was administered, the hydromorphone was started by the alternate route. Morphine injections were offered every 3 hours for breakthrough pain throughout the study; all other analgesics were continued unchanged.

Main outcome measures

Patients' visual analog estimates of pain intensity, pain relief, mood, and sedation level; mean plasma hydromorphone concentration.

Main results

2 patients died (from myocardial infarction and pulmonary embolism); 1 became encephalopathic, 1 required emergency radiotherapy, and 1 lost intravenous access. 15 patients completed the study. The mean hydromorphone infusion rate was 6.1 mg/h (range, 1 to 35 mg/h). Improvement in mean scores for pain, mood, and sedation occurred within 2 hours of beginning the infusions and was maintained (P < 0.02), but patients perceived no difference between intravenous- and subcutaneous-infusion periods (P > 0.1). The mean number of morphine injections for breakthrough pain over 48 hours was 4.8 for intravenous infusion and 5.3 for subcutaneous infusion. 4 patients had mild nausea. Steady-state hydromorphone concentration was reached by 24 hours by both methods. The mean adjusted plasma concentration of hydromorphone between 24 and 48 hours was higher for the intravenous route (27.2 vs 21.3 ng·mLmg, P = 0.02).


Patients with advanced cancer received equivalent relief from chronic pain when hydromorphone was infused subcutaneously or intravenously.

Source of funding: Health and Welfare Canada.

Address for article reprint: Dr. D.E. Moulin, Victoria Hospital, 375 South Street, London, Ontario N6A 4G5, Canada.


New routes of drug administration are needed to improve the management of patients with cancer pain. Drug therapy with continuous subcutaneous infusions is useful, particularly for patients who cannot tolerate oral opioids. Pain-management surveys show that most patients will require at least 2 routes of administration (oral and parenteral) during the course of their painful illness and one-third will require at least 3 alternative routes.

The well-designed study by Moulin and colleagues provides evidence for the safety and efficacy of subcutaneous hydromorphone compared with intravenous hydromorphone, although the small sample size does not exclude differences smaller than those detected, which could be clinically important. Also, the relative efficacy and safety of these 2 routes of administration still need to be compared for periods of infusion longer than 48 hours.

The simplicity of the subcutaneous route is that a 27-gauge butterfly needle can be changed readily at home. This is a practical advantage over the intravenous route for which trained professionals are needed. However, the broader use of implantable infusion devices to provide intravenous access makes the intravenous route an important option. Thus, I disagree with the authors' suggestion that continuous intravenous infusions "should be abandoned." Drug, dose, and route of administration tailored to the needs of the patient should be the rule.

This study provides reliable evidence to support the use of subcutaneous infusions for pain management in patients who require parenteral administration of opioids (1).

Kathleen M. Foley, MD
Memorial Sloan-Kettering Cancer CenterNew York, New York, USA


1. Coyle N, Adelhardt J, Foley KM, Portenoy RK. Character of terminal illness in the advanced cancer patient: pain and other symptoms during the last four weeks of life. Journal of Pain and Symptom Management. 1990;5:83-93.