Background: Intrathecal morphine forms granulomas that arise from the adjacent arachnoid membrane. The authors propose that these inflammatory cells exit the meningeal vasculature secondary to meningeal mast cell degranulation.
Methods: Three sets of experiments were accomplished in dogs: (1) ex vivo meningeal mast cell degranulation (histamine release was measured ex vivo from canine dura incubated with opiates); (2) in vivo cutaneous mast cell degranulation (flare areas on the dog abdomen were measured after subcutaneous opiates); and (3) in vivo granuloma pharmacology. Dogs with lumbar intrathecal catheters received infusion of intrathecal saline or intrathecal morphine. Intrathecal morphine dogs received (1) no other treatment (control); (2) twice-daily subcutaneous naltrexone; (3) intrathecal co-infusion of cromolyn; or (4) twice-daily subcutaneous cromolyn for the 24- to 28-day study course.
Results: Morphine but not fentanyl evoked dural histamine release, which was blocked by cromolyn but not naloxone. Wheal/flare was produced by subcutaneous morphine, methadone, hydromorphone, but not fentanyl, and was unaffected by naltrexone but prevented by cromolyn. Granulomas occurred in all dogs receiving intrathecal morphine (15 of 15); subcutaneous naltrexone had no effect on granulomas (six of six) but was reduced by concurrent intrathecal cromolyn (zero of five) or twice-daily subcutaneous cromolyn (one of five).
Conclusions: The pharmacology of cutaneous/dural mast cell degranulation and intrathecal granulomas are comparable, not mediated by opioid receptors, and reduced by agents preventing mast cell degranulation. If an agent produces cutaneous mast cell degranulation at concentrations produced by intrathecal delivery, the agent may initiate granulomas.