Intrathecal Morphine Pump for Neuropathic Cancer Pain

Introduction

Cancer pain has traditionally been treated following the World Health Organization (WHO) analgesic ladder. In all, 80% to 90% of cancer pain patients are said to be benefitted with this simplified approach. ¹ It is the remaining 10% to 20% of cases which tests the skill and knowledge of a pain physician. ² These include abdominal visceral pain, cancer-induced bony pain, and central and peripheral neuropathic pain. ³ Of all the difficult pain syndromes, neuropathic pain has attracted attention of most pain physicians. From neuromodulator drugs, peripheral and central neural blockade to surgical neuroablative procedures, all have been tried. ⁴ Opioids though generally considered being ineffective for neuropathic pain have now found to be efficacious in neuropathic pain; albeit in a higher dosage. This high dose may cause intolerable side effects. ⁴ Intrathecal drug delivery reduces the overall dose and the systemic side effects of morphine and thus may have a better tolerability profile and efficacy than upward titration of oral morphine to intolerable high doses. We report a similar case of severe neuropathic type of cancer pain which responded to epidural morphine, including the neuropathic component. Finally, a programmable intrathecal morphine infusing pump was implanted for the patient.

Case Report

A 61-year-female patient who was a known case of endometrial carcinoma diagnosed upon histopathological report following a total abdominal hysterectomy and bilateral salpingoophorectomy in March 2009. The patient then received 28 fractions of external beam radiotherapy 50 Gy over a period 40 days from August to September 2009, followed by 2 cycles of intrauterine brachytherapy. A positron emission tomography–computed tomography (PET-CT) scan done thereafter showed metastases involving left paraaortic lymph node involving left psoas muscle, left common iliac artery, left supraclavicular lymph node, right subpectoral lymph node, and left ureter with hydroureteronephrosis. Then she received 6 cycles of paclitaxel and carboplatin chemotherapy. After that, a PET scan was done which revealed increased Flu Deoxy glucose (FDG) uptake in left psoas, left ureter, and L2, 3, and 4 vertebrae with resolution of left supraclavicular lymph node. She then received palliative radiation therapy (RT) to vertebrae L1 to L5 from November 2009 to January 2010.

She was referred to us by our hospital’s radiotherapists for pain management in November 2010 as her pain was not being relieved with weak opioids. Patient had pain in her lower abdomen and right lower limb up to the knee joint so much so that any position change was not possible because of severe pain. Her lower abdominal pain was purely visceral in character, whereas her thigh pain was typically burning, movement related, and almost purely of neuropathic variety. Her baseline Visual Analog Scale (VAS) at rest was 7/10, on slight movement it increased to 10/10, and she was crying in pain most of the time. At presentation, she was taking tramadol 100 mg 4 times a day with a combination of ibuprofen 400 mg and paracetamol 500 mg 3 times a day. She was also taking gabapentin 400 mg twice daily.

We admitted the patient to our pain and palliative care ward and started her on oral morphine. We titrated doses of oral morphine upward from 5 mg 4th hourly to 60 mg 4th hourly in a matter of 72 hours. We prescribed nonsteroidal anti-inflammatory drugs ([NSAIDs]; aceclofenac 100 mg and paracetamol 500 mg twice daily) and gabapentin (600 mg thrice daily) in usual maximum tolerable doses as well. Over a week’s time, patient was quite comfortable lying in one position with VAS at rest being 1/10. As soon as she tried to move, her pain would return with a VAS of 10/10. She could not even imagine standing or walking with this pain. At rest, she felt dizzy and was oversedated. We also tried pregabalin (Lyrica, Pfizer Inc, US) 150 mg twice daily but that only added to her sedation and dizziness. We tried reducing morphine dose and increasing pregabalin dose to 300 mg twice daily but that did not help.

We got a magnetic resonance imaging (MRI) lumbosacral spine done in January 2011 to know the present status of her spinal column. The MRI revealed near-total destruction of vertebral bodies L2 to L4 and infiltration into the left psoas compartment. Further radiotherapy was ruled out by the radiotherapists.

Over the rounds in our pain ward, numerous possibilities were discussed for the patient. There was no role of epidural steroids because of involvement of lumbar plexus encased in the psoas sheath. Lumbar plexus block was also ruled out as it would give only short-term relief; moreover, direct infiltration of the tumor into the psoas compartment would make the procedure difficult and effect unreliable. Finally, we zeroed onto 2 modalities, either a lumbar plexus block or an epidural morphine with or without local anesthetic. Hence, an epidural catheter was placed to evaluate response to epidural morphine at L1 to L2 level. With 3 mg boluses of epidural morphine in a volume of 10 mL given every 6 hours, patient’s VAS score had come to 0 on lying down, 2 on changing position from lying down to sitting, and 3 on walking with support. Moreover, patient was very alert all the time and the systemic side effects of morphine had just disappeared. The patient was definitely more than happy. Local anesthetic bupivacaine was also tried in a concentration of 0.125% but that made the patient uncomfortable because of mild sensory loss and there was no added analgesia. As further curative treatment had been ruled out by primary physicians and patient’s general condition was good with life expectancy definitely above 6 months, we went ahead with planning for an externally programmable intrathecal morphine infusion pump (SynchroMed II; Medtronics Minnesota, US ).

After discussing the pros and cons of such an intervention with the patient and her relatives, an informed consent was taken. Taking all aseptic precautions, a programmable intrathecal pump with a reservoir volume of 40 mL was placed in the anterior abdominal wall of the patient with the catheter placed in the subarachnoid space at L1 to L2 level, all done under fluoroscopic guidance. The catheter tip was identified at T11 level. Initial dose was kept at 1 mg/d as the patient had been receiving 12 mg of morphine epidurally per day. Her pain relief in the initial 36 hours was only 60%. Gradually her dose was increased to 3.5 mg/d over a period of 2 weeks. With that dose, she had no pain on lying down and minimal pain on walking using a walker. The smile on her face was priceless. Along with the morphine infusion, she was discharged on advice to take paracetamol 500 mg thrice daily and gabapentin 400 mg twice daily. Refill date of the pump was 3 months from the discharge day and that saved the patient from unnecessary visits to the pain clinic which is a good 300 km from her home. On follow-up for pump refill in April 2011, patient’s VAS was stable with 0 at rest, 2 on changing position, and 3 on walking with support without any further increase in intrathecal morphine dose. Needless to say, the patient and doctors were more than satisfied.

Discussion

Cancer pain is mostly of mixed type due to multiple etiological factors responsible for it. It is traditionally managed with opioid medications with adjuvants to take care of inflammatory and neuropathic components. Adjuvants typically include NSAIDs and anticonvulsant medications like gabapentin, pregabalin, and amitryptiline. ⁵

Although WHO in its analgesic ladder recommends the use of morphine for all kinds of severe malignant pain, various authors are of the view that neuropathic pain is the least likely to respond to morphine in tolerable doses. Arner and Meyerson ⁶ in 1988 concluded that morphine was not effective for treating neuropathic and idiopathic forms of pain. Eide et al ⁷ in 1994 found that pain of postherpetic neuralgia that did not respond to oral morphine was relieved with ketamine. There are others who have found morphine to be effective for neuropathic pain as well albeit in higher doses than for nociceptive pain. Kupers et al ⁸ and Rowbotham et al ⁹ in 1991 found morphine to be efficacious for neuropathic pain of different origins. Indelicato and Portenoy ¹⁰ in 2002 tried rotation of opioids in chronic central and peripheral neuropathic pain and found them to be effective.

Morphine and gabapentin in combination have been found better than morphine alone for treating neuropathic pain. Gilron et al ¹¹ in a crossover trial conducted in 2005 concluded that gabapentin and morphine combined achieved better analgesia at lower doses of each drug than either as a single agent, with constipation, sedation, and dry mouth as the most frequent adverse effects. In the present case, we had tried high doses of gabapentin along with oral morphine but of no use. Moreover, side effects began to add up and resulted in excessive sedation and the pain scores were unchanged, especially upon movement.

Pregabalin is a novel congener of gabapentin which was found to be particularly effective in neuropathic pain. Siddall et al ¹² found that pregabalin 150 to 600 mg/d to be effective in relieving central neuropathic pain, improving sleep, anxiety, and overall patient status in patients with spinal cord injury. In our clinical experience also, we have found pregabalin to be efficacious in patients where gabapentin failed to provide symptomatic relief. A select few patients like the one in this report do not report any relief with either drug.

Interventional pain management in patients with cancer is indicated when either oral pharmacotherapy has failed or side effects of oral medications are intolerable. Nerve blocks, neuraxial drug administration, and neuromodulation therapies all have been in use for treating refractory cancer pain. ¹³ Intrathecal morphine pumps have been in use for quite some time, but its role in treatment of neuropathic pain is not established. Drugs like clonidine, local anesthetics, and ketamine have been added to morphine in these pumps when neuropathic pain was not resolved with morphine alone. ¹⁴ These medications add to side effects, complicate drug titration and filling, and may not be available as preservative-free preparations in all the countries, thus increasing risk of neurotoxicity due to preservative infusion into the intrathecal space.

Morphine has been the drug of first choice for use in intrathecal drug delivery systems. In a review article by Stearns et al ¹⁵ in the Journal of Supportive Oncology, the authors have described morphine as a first-line agent for trial in intrathecal infusions, and then adjuvants are tried depending upon the component of pain not amenable to morphine alone.

Overall, the use of implantable drug delivery systems (IDDSs) improves pain scores and quality of life in patients with cancer. A randomized clinical trial of IDDS plus comprehensive medical management (CMM) versus CMM alone showed better clinical success at 4 weeks for IDDS patients. ¹⁶ Direct drug delivery into the neuraxis is most likely to decrease systemic toxicity and increase maximum tolerable doses of morphine.

Conclusion

Refractory cancer pain is a common finding in pain practice. Use of intrathecal morphine pumps is limited, especially for neuropathic pain. This case report highlights the effectiveness of intrathecal morphine, even in neuropathic pain while reducing systemic toxicity with higher doses of oral morphine.