Department of Anesthesia, Pain, and Perioperative Medicine, Harvard Medical School, Brigham & Women’s Hospital, Boston, Massachusetts
May 17, 2002.
Address correspondence and reprint requests to John Loughrey, MB, MRCPI, FCARCSI, Department of Anesthesia, Pain, and Perioperative Medicine, Harvard Medical School, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Address e-mail to firstname.lastname@example.org.
Advances in technology for long-term intrathecal drug administration, along with evolving knowledge about the neurophysiology and pharmacology of spinal nociception, have created a new option for enhanced analgesia for patients suffering from chronic pain. As noted in a review involving 13,342 mainly US-based chronic pain patients, the use of long-term infusions of intrathecal medications has significantly increased (1). Intrathecal drug administration is cost-effective when compared with other analgesic modalities for patients with failed back surgery syndrome (2). The practice of using drugs other than morphine is also increasing despite little data on long-term safety of such adjuvant therapy.
Neuraxial administration of opioids can cause central nervous system (CNS) excitation, which can manifest as delirium or agitation (3). Significant neural toxicity can occur if inappropriate medications are infused intrathecally via an implanted reservoir. In evaluating patients who present with neuropsychiatric disorders and who have a spinal catheter in situ, the supraspinal effects of intrathecal infusions should be considered. We describe a patient who developed an acute dissociative mental state, which resolved on removal of intrathecal medication from his pump reservoir. Toxicology analysis revealed that the injectate was falsely positive for phencyclidine and methadone.
A 55-yr-old man suffering from chronic back and left leg pain was referred to the pain management center for consideration of a trial of intrathecal opiates. He had undergone numerous interventional pain management techniques at other centers over the previous 10 yr, including a trial of dorsal column stimulation, and was knowledgeable about sophisticated modalities of pain control. His intake of medications included 75–85 mg/d of oxycodone. His medical history was significant for previous lumbar discectomy, depression, and gastric bypass surgery.
After psychological evaluation, the patient underwent a trial of intrathecal opioid administration that resulted in significant pain reduction and was subsequently implanted with a Medtronic Synchromed® (Medtronic, Minneapolis, MN) intrathecal pump. Tolerance to intrathecal morphine developed rapidly, requiring progressive increases in the daily dose from 2 mg/d immediately after implantation to 13.8 mg/d 6 mo later. The infusion comprised of a complex-continuous pattern involving several dosage adjustments daily. During this time, the patient experienced no apparent mental status changes despite escalating intrathecal opioid doses and continuation of oral opioids.
Approximately 7 mo after the initiation of intrathecal therapy, the patient exhibited cognitive deterioration, presenting to the emergency room in a dissociative mental state. This had developed acutely over the previous 3 days and featured stammering speech, mild lethargy, blunting of affect, and both long- and short-term memory loss. There was no history of head trauma, and no definitive injuries were identified. The cognitive deficits were quite remarkable because the patient was not oriented to place or time and was unable to recognize previously familiar people. Visual field disturbance was also reported. His medications included oxycodone 70 mg, gabapentin 900 mg, clonazepam 3 mg, tizanidine 8 mg, trazodone 100 mg, and venlafaxine 150 mg per day, all in divided doses as recommended. Intrathecal morphine was infusing at 13.8 mg/d. Neurologic and psychiatric evaluations together with full hematologic, metabolic, and endocrine workup did not reveal a cause for his altered mental status. A brain magnetic resonance image including angiography failed to reveal an etiology for his cognitive decline and an electroencephalogram study showed rare bilateral theta slowing but no global changes suggestive of an encephalopathy. Toxicology analysis of blood and urine did not reveal the use of any other drugs other than those prescribed. Immediately on hospitalization, the dose of oral oxycodone was reduced to 40 mg, and the intrathecal morphine infusion was reduced to 10 mg/24 h. Gabapentin was tapered to 300 mg/d, clonazepam was substituted with lorazepam 0.5 mg at night, and the psychoactive medications venlafaxine and trazodone were discontinued.
By Day 5 after hospital admission, and despite these medication reductions, the patient showed no improvement in mental status, even on far smaller doses of systemic medications than he was accustomed to. A lumbar puncture and concurrent aspiration of the intrathecal medication reservoir was performed. There was no evidence of meningitis, and the cerebrospinal fluid (CSF) had a normal appearance and electrolyte count. Toxicology analysis of CSF was not performed. Surprisingly, a cloudy yellow discolored opaque fluid was aspirated from the intrathecal medication reservoir, although in the expected quantity based on anticipated pump volume contents. All previous reservoir aspirations for medication refill, including the most recent one 36 days previously, had revealed unremarkable aspirates. The specimen was found to be sterile, but toxicology analysis revealed its composition to be positive for opiates, methadone, and phencyclidine. No reasonable explanation was found to explain the presumed contaminants, and the thought of illicit intrathecal pump access was entertained.
After discarding the contaminated discolored aspirate, the pump was refilled with morphine 40 mg/mL, which was set to infuse at 10 mg/d. Improvement in the patient’s mental status was first noted 2–3 days later. Within approximately 1 wk, significant restoration of memory and cognitive function occurred, but baseline levels of higher cognitive function, personality, and memory were not regained. He had amnesia for the period 1 wk before hospitalization to 2–3 days after removal of the cloudy yellow fluid from his pump. The patient denied any history of interference with the intrathecal medication reservoir and was told that initial test results were positive for phencyclidine. The morphine infusion was further reduced to 8 mg/d, and venlafaxine was reintroduced without any change in reported pain or mental status. Five months later, the patient requested an explantation of the device, citing feelings of paranoia and vulnerability. The patient continued to complain of memory loss and inability to concentrate 7 mo later.
We report the case of a patient who, while receiving intrathecal morphine therapy via a Synchromed® pump, developed a dissociative mental state with significant cognitive decline. The administration of intrathecal morphine has previously been associated with paranoia (4), nystagmus (5), and also myoclonus (6). Neuraxially administered opioids may also cause CNS excitation, delirium, and agitation (3). Although there is extensive clinical experience with opioids used as intrathecal analgesics, with apparent small potential for neurotoxicity at standard doses, little systematic data are available on large concentration infusions over longer time periods such as in this case (7).
Whereas systemic polypharmacy was a potential cause of our patients’ acute mental status changes, we observed no improvement in the patients’ mental status after five days, despite rationalizing his stable regimen of systemic medications. Moreover, the temporal relationship between replacement of the contents of his intrathecal pump and our patients’ cognitive improvement makes it likely that the intrathecal infusion was the cause of his altered mental status. We then considered illicit substance contamination of the medication reservoir, which has previously been reported in the literature (8). Clinically, the patient’s presenting symptoms were consistent with those produced by a psychoactive medication such as phencyclidine (9), which was detected by our immunoassay. Also, the abnormal discoloration of the medication aspirated from the reservoir led to the suspicion of contamination. However, the concordance between the expected and the actual volume found in the pump reservoir make illicit interference with the device highly unlikely.
The issue of likely false positive toxicology screening results for phencyclidine and methadone with our standard immunoassay was then examined. The residual morphine from the medication reservoirs of two other patients with intrathecal pumps, but without any suspicion for pump manipulation, also tested positive for phencyclidine using the same immunoassay (DuPont ACA®, DuPont Company, Wilmington, DE). In vitro testing has confirmed that sigma opioids display similar binding properties to phencyclidine (10), raising the possibility that opioid assays may test falsely positive for phencyclidine-type compounds at large concentrations. Because the immunoassay technique used is primarily intended for use on human biological samples only, we also therefore disregarded the detection of methadone in our sample. Of note, phencyclidine was reported as being detected in a case of illicit drug use via an implanted intrathecal pump by Burton et al. (8), but the assay technique was not described.
The exact cause of the abnormal discoloration of the residual aspirate from the pump in our patient and resolution of symptoms after its removal remains unclear. The largest concentration of morphine commercially available for intrathecal use is 25 mg/mL, with larger concentrations, as in our case (40 mg/mL), requiring compounding from a powdered form by a pharmacist. We have been unable to compound stable solutions of morphine for intrathecal infusion at concentrations larger than 50 mg/mL. Morphine sulfate solutions may darken with age (11) and develop a yellowish discoloration (product information, Abbott Labs, Chicago, IL). It is possible that a larger concentration of morphine was actually present, which, when unstable in a solution, led to a cloudy yellow appearance. Morphine-induced CSF pH value changes have previously been noted (12), with implications for stability of analgesic compounding mixtures. We believe that our patients’ symptoms may have been caused by an abnormally large concentration of morphine in the pump.
We therefore submit that a heightened degree of awareness surrounding the potential for CNS side effects, both behavioral and pathophysiological, is required among physicians who care for patients with intrathecal analgesic regimes. In a patient who carries an implanted intrathecal medication reservoir, unexplained mental status changes should raise the possibility that an unexpected substance or a side effect from a known drug may be causing CNS changes. Toxicology analysis of pump reservoir medications should not be performed with immunoassay techniques because currently available commercial kits for routine toxicology analysis of biological samples may be associated with falsely positive results. To provide definitive analysis, liquid or gas chromatography, coupled with mass spectroscopy, are more appropriate techniques (13,14).
1. Hassenbusch SJ, Portenoy RK. Current practices in intraspinal therapy: a survey of clinical trends and decision making. J Pain Symptom Manage 2000; 20: S4–11.
2. de Lissovoy G, Brown RE, Halpern M, et al. Cost effectiveness of long-term intrathecal morphine therapy for pain associated with failed back surgery syndrome. Clin Ther 1997; 19: 96–112.
3. Rozan JP, Kahn CH, Warfield CA. Epidural and intravenous opioid induced neuroexcitation. Anesthesiology 1995; 83: 860–3.
4. Christie JM, Meade WR, Markowsky S. Paranoid psychosis after intrathecal morphine. Anesth Analg 1993; 77: 1298–9.
5. Ueyama H, Nishimura M, Tashiro C. Naloxone reversal of nystagmus associated with intrathecal morphine administration. Anesthesiology 1992; 76: 153.
6. DeConno F, Caraceni A, Martini C, et al. Hyperalgesia and myoclonus with intrathecal infusion of high-dose morphine. Pain 1991; 47: 337–9.
7. Hodgson PS, Neal JM, Pollock JE, Liu SS. The neurotoxicity of drugs given intrathecally (spinal). Anesth Analg 1999; 88: 797–809.
8. Burton AW, Conroy B, Garcia E, et al. Illicit substance abuse via an implanted intrathecal pump. Anesthesiology 1998; 89: 1264–7.
9. Showalter CV, Thornton WE. Clinical pharmacology of phencyclidine toxicity. Am J Psychiatry 1977; 134: 1234–7.
10. Quirion R, Hammer RP Jr, Herkenham M, Pert CB. Phencyclidine (angel dust)/sigma “opiate” receptor: visualization by tritium-sensitive film. Proc Natl Acad Sci USA 1981; 78: 5881–5.
11. Altman L, Hopkins RJ, Ahmed S, Bolton S. Stability of morphine sulphate in Cormed III (Kalex) intravenous bags. Am J Hosp Pharm 1990; 47: 2040–2.
12. Wagemans MF, Bakker EN, Zuurmond WW, et al. Intrathecal administration of high-dose morphine solutions decreases the pH of cerebrospinal fluid. Pain 1995; 61: 55–9.
13. Hirtz J. Importance of analytical methods in pharmacokinetic and drug metabolism studies. Biopharm Drug Dispos 1986; 7: 315–26.
14. Cone EJ, Dickerson S, Paul BD, Mitchell JM. Forensic drug testing for opiates. IV. Analytical sensitivity, specificity, and accuracy of commercial urine opiate immunoassays. J Anal Toxicol 1992; 16: 72–8.