Intraspinal morphine for chronic pain: a retrospective, multicenter study.

Intraspinal opioids are frequently used in the treatment of cancer and noncancer pain, but few studies have evaluated the efficacy of this technique. This multicenter, retrospective study surveyed physicians in the United States regarding their standard practices when using intraspinal opioids delivered via an implanted drug administration device. Thirty-five physicians (50.0%) responded, providing 429 usable patient forms (52.4%), which sought information about screening, outcomes, dosing, and adverse effects. Patients with malignant (32.7%) and noncancer (67.3%) pain had been treated for an average of 14.6 +/- 0.57 months (range, 8-94 months) at the time of form completion. For all patients, the mean percent relief was 61.0% +/- 1.35%. Patients with somatic pain tended to have greater relief, as measured by a global rating of pain relief, than did patients with other types of pain (Mann-Whitney test, P = 0.0003). After titration during the first 3 months, intrathecal morphine doses increased only twofold from 6.84 +/- 0.65 mg/day at 3 months to 13.19 +/- 1.76 mg/day at 24 months. Compared to those with noncancer pain, malignant pain patients had a higher average initial dose. The average dose used by cancer patients escalated quickly and then stabilized, whereas the average doses used by noncancer pain patients exhibited a more gradual, linear increase in dose. Long-term adverse drug effects were uncommon, but system malfunction, usually catheter related, occurred in 21.6% of patients. Prospective, randomized, controlled clinical studies of long-term efficacy and adverse effects are warranted.