Long-term Respiratory Depression Induced by Intrathecal Morphine Treatment for Chronic Neuropathic Pain

Intrathecal opioid treatment has become a widely used approach in cancer and chronic pain, particularly for the treatment of patients with neuropathic pain, failed back syndrome, and mixed-type pain.1,2In contrast to the frequent reports of respiratory depression after postoperative intrathecal or epidural opioid administration,3–5there are only a few reports of severe drug-related complications under chronic intrathecal treatment using an intrathecal drug delivery system (IDDS) with a regular dosage. Particularly, to our knowledge, no case of a slowly increasing chronic respiratory depression after IDDS implantation has been reported.2,6–10We report the case of a 41-yr-old man referred to our pain clinic 6 yr after a motorcycle accident leading to a C4–C7 root transection with attributed medullar and cervical plexus lesion. Despite implantation of an IDDS (Isomed-60-ml; Medtronic, Inc., Minneapolis, MN) 8 months previously, he experienced intractable neuropathic pain, including deafferentation pain at the left upper limb, and tactile allodynia at the left chest after a coagulation of the dorsal root entry zone of the substantia gelatinosa (DREZ lesion) 1 yr ago. At time of presentation, the medication included 75 mg amitriptyline, 1,800 mg gabapentin, and an intrathecal infusion of 4 mg morphine per day since 1 yr. The pain relief was insufficient, with a mean pain level of 7 on a numeric rating scale (0–10). In addition, the patient reported increasing dyspnea, severe fatigue, sleep disorder, and depressed mood during the past 8 months. He was unable to walk more than 10 m, he needed permanent administration of oxygen, and partly assisted ventilation (continuous positive airway pressure) became necessary during the past 6 months. The medical and neurologic examination revealed a tired patient with complete paralysis of the left arm, accompanied by atrophy, anesthesia (C4–C7), Horner syndrome, and tactile allodynia of the left chest. Chest x-ray and computed tomography of the chest demonstrated a left elevated diaphragm as a consequence of phrenic nerve paralysis (figs. 1 and 2). Arterial blood gas analysis revealed respiratory acidosis (in arterial blood: partial pressure of oxygen [Pao²], 47.0 mmHg; partial pressure of carbon dioxide [Paco²], 65.1 mmHg; pH, 7.33; base excess, 5.3 mm; saturation, 80%). Pain started immediately after the accident and was treated by several combinations of opioids and other analgesics, which the patient did not remember in detail. An IDDS was implanted in January 2004, with an initial daily dose of 14 mg morphine and 0.15 mg clonidine. Nearly 2 weeks later, catheter leakage and dislocation provoked a withdrawal syndrome, and after replacement of the catheter with the previous dose of morphine, cardiopulmonary resuscitation became necessary. The patient recovered completely from this intervention. Subsequently, the morphine dosage was reduced to 2 mg/day. The exact time course of dose changes within the following months is unknown, but the dose finally increased to 4 mg morphine per day. During these last months, the patient's psychiatric state and general condition worsened significantly.

Because of the psychological symptoms, particularly the severe tiredness and depressed mood, and the reduced pulmonary function, we suspected chronic opioid intoxication, and consequently the daily intrathecal morphine dose was reduced from 4 to 1 mg within 3 weeks and subsequently was switched to a concomitant oral medication (12 mg/day hydromorphone). In addition, 0.6 mg/day clonidine was substituted for 5 weeks, and pregabalin (300 mg/day) was substituted for gabapentin. Under this medication, the patient reported a considerable improvement in pain level, tiredness, and psychological state, and the dyspnea and respiratory function recovered to normal (in arterial blood: Pao², 126.7 mmHg; Paco², 41.8 mmHg; pH, 7.424; base excess, 2.8 mm; saturation, 98%). Obviously, the morphine effects on respiration were facilitated by (1) consequences of the accident, including phrenic nerve paralysis, elevation of the diaphragm, and atelectasis, and (2) the reduced vigilance after dose escalation. However, the key role of intrathecal morphine for the chronic deterioration of the patient's condition was proven by complete recovery not only of the tiredness and other psychiatric symptoms but also by return to normal in all respiratory parameters and the physical capacity after morphine reduction and change to oral opioid treatment.

One reason for this case presentation was the remarkable fact that all involved physicians (neurosurgeons, neurologists, rehabilitation and pain specialists) did not recognize the correlation of increasing morphine dose without any analgesic improvements, the increasing fatigue, exercise dyspnea, and the deterioration of pulmonary function step-by-step for several months although respiratory depression with intrathecal opioids is well known. The missing anticipation of respiratory risk under long-term intrathecal morphine medication is matched by missing precautions in the cited European and German guidelines.11,12In consequence, physician awareness is apparently limited only to acute signs of intoxication (such as bradypnea, respiratory arrest). There is an increasing number of reports such as this one revealing potentially life-threatening side effects or persistent neurologic sequelae of IDDS, and there are no controlled trials evaluating the frequency of more moderate respiratory depression or increased sleep apnea syndrome in chronic pain patients. The current patient is one of several referred to our pain clinic and treated intrathecally because of a supposed resistance to therapy. These patients were mainly diagnosed in neurosurgical or orthopedic departments with a monodisciplinary approach to pain treatment.13Most of them—like the current patient—could be treated sufficiently without IDDS using multimodal nonmedical protocols and medical treatment, mainly including oral opioids. Therefore, treatment resistance should be diagnosed very cautiously. We recommend a reevaluation of intrathecal opioid treatment in chronic pain states considering that, in contrast to intrathecal spasmolytic treatment and oral opioid pain treatment,14–16no randomized controlled trials are available.

*University Hospital, BG-Kliniken Bergmannsheil, Bochum, Germany. andrea.scherens@rub.de