Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine for pain treatment in cancer patients

Transcription

1 2013; 8(2): Original Research Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine for pain treatment in cancer patients Koji Kawamura 1) and Tetsushi Fukushige 2) 1) Tagawa Municipal Hospital, Palliative Care Unit, 2) Kurume University Hospital, Palliative Care Center Recived 7/30/2012, revised 10/15/2012, accepted 1/10/2013 Introduction: Few studies have examined neuropathological changes such as the degenerative necrosis and demyelination of spinal nerve cells accompanying intrathecal administration of opioids (ITO) to relieve refractory cancer pain. Previous studies have produced conflicting results as to whether or not ITO causes nerve tissue damage. The current study neuropathologically investigated autopsy specimens from patients who received ITO. Methods: Subjects were 7 patients who received continuous intrathecal analgesia and who were later autopsied (4 males, 3 females). Six patients were administered morphine and bupivacaine while 1 patient was administered fentanyl and bupivacaine. The duration of administration ranged from days. Results: Two patients who received long-term administration of morphine were found to have severe necrotic degeneration and gliosis of spinal neurons and demyelination in the dorsal horn and dorsal roots. However, neuropathological changes were not noted in Patient 4, who was briefly administered morphine, or in the patient who was administered fentanyl. Conclusion: The total dose of morphine used for ITO and the duration of its administration were suggested to be related to the extent of nerve tissue damage. Thus, nerve tissue damage due to ITO might be primarily associated with morphine. Palliat Care Res 2013; 8(2): Key words: cancer pain, intrathecal administration of opioids, neuropathological findings Introduction Since opioids were first administered to humans in the subarachnoid space in ), many patients with refractory cancer pain have received intrathecal administration of opioids. Various neuropathological studies on this method of pain relief have been performed, but most have focused on inflammatory mass formation at the catheter tip, and reports on spinal disorders and nerve damage such as degenerative necrosis and demyelination of spinal nerve cells using human autopsy material are extremely rare 2 4). In 1992, Sjöberg et al. 2) neuropathologically studied 15 patients who received long term administration of morphine and bupivacaine, a local anesthetic; they reported finding damage to the spinal cord and nerve roots. In 1994, Karlsson et al. 3) similarly reported damage to the spinal nerves. However, in 1997 Wagemans et al. 4) autopsied 10 patients with cancer who had received long term administration of morphine and bupivacaine, but they reported that there was no nerve tissue damage. Thus, there are contradictory reports as to whether or not this analgesia causes nerve tissue damage. The current study neuropathologically examined autopsy specimens from patients who received continuous intrathecal administration of opioids. Methods Subjects were 7 patients who received continuous intrathecal administration of opioids at Kurume University Hospital from July 2001 to March 31, 2010; these patients were later autopsied (4 males, 3 females; age range: years; mean age: 52 years). One patient had difficulty maintaining a standing position during puncture of the subarachnoid space, so preparations for continuous intrathecal administration of opioids were made under spinal anesthesia; local anesthesia was used for all of the other patients. Implanting of access ports was determined based on the patient s prognosis. In 4 patients with a prognosis of 1 month or longer, a kit for an epidural port from Deltec was used along with a polyurethane epidural catheter that was placed in the subarachnoid space; the port was placed on the right midclavicular line above the 8th to 9th rib. A grip- Corresponding author : Koji Kawamura Tagawa Municipal Hospital, Palliative Care Unit , Oaza hosii, Tagawa shi, Fukuoka, Japan TEL FAX E mail: k_kawamura0726@cg8.so net.ne.jp 192 Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine

2 per needle was used for port puncture and the catheter was replaced weekly. In 3 patients with a prognosis of less than 1 month, an epidural catheter from B. Braun made of polyamide resin was placed in the subarachnoid space. A subcutaneous tunnel of about 30 cm was created and the catheter was fixed in place. The site for catheter insertion was disinfected daily with povidone iodine. The insertion site was L3/4 in 6 patients and L2/3 in 1. The infusion was a mixture of an opioid and bupivacaine and physiological saline at a total volume of 12 ml/day. The infusion was continuously administered with a syringe pump at a rate of ml/hr. Rescue dose for breakthrough pain was 1/24 of the daily dosage amount of opioid. The drugs infused were morphine and bupivacaine (6 patients) or fentanyl and bupivacaine (1 patient). Autopsy was performed two to 10 hrs after death and specimens were collected. After confirming the site for catheter insertion and the location of the catheter tip, the spinal cord from the thorax (T8) to the lumbar region (L5) was removed along with the dura mater and fixed with 10% paraformaldehyde. After fixation, the spinal cord was sectioned every 1.5 cm. Specimens 5 mm thick were prepared and embedded in paraffin. Paraffin blocks were sectioned and subjected to hematoxylin eosin staining. Myelin sheath staining was performed to better visualize demyelinated lesions and glial fibrillary acidic protein (GFAP) immunostaining was performed to confirm gliosis. Results Characteristics of the 7 patients, the types of opioids administered, the site of catheter insertion, and the location of the catheter tip are shown in Table 1A. Autopsies were performed on all 7 patients; patients consisted of 4 males and 3 females ranging in age from years (mean age: 52 years). The primary cancer was rectal cancer in 2 patients, cervical cancer in 2, ureteral cancer in 1, prostate cancer in 1, and undifferentiated pleomorphic sarcoma in 1. The site of catheter insertion was L 3/4 in 6 patients and L 2/3 in 1. The catheter tip was located at L1 in 3 patients, L2 in 2, and unspecified in 2. Macroscopic irregularities such as cancer metastasis were not noted in any of the patients spinal cord specimens. Histologically, necrotic degeneration of neurons in the dorsal horn of the spinal cord was noted in Patient 2, who received long term administration of morphine (Fig. 1). Similarly, demyelination in the dorsal horn and dorsal roots was noted in Patient 3, who also received long term administration of morphine (Fig. 2). GFAP immunostaining of spinal cord specimens from Patient 2 is shown in Fig. 3. Gliosis was noted in the dorsal horn. The total dose of opioid and bupivacaine and duration of their administration are summarized along with neurohistopathological findings in Table 1B. Additionally, the range of dosage amounts of opioid and bupivacaine (mg/day or μg/day) (lowest dose peak dose) and infusion pump flow rate (infusion rate) (ml/h) are indicated in Table 1C. Severe necrotic degeneration and gliosis of spinal neurons and demyelination in the dorsal horn and dorsal roots were noted in Patients 2 and 3, who received long term administration of morphine. These neuropathological changes were most distinct near the tip of the catheter, but weak damage to the spinal cord was recognized over a wide range, including areas distant from the tip of the catheter. However, these neurohistopathological findings were not noted in patients who only received morphine for a brief period. Although Patient 7 received long term administration of fentanyl, necrotic degeneration and gliosis of spinal neurons and demyelination in the dorsal horn and dorsal roots were not noted. Discussion Several factors could possibly lead to nerve damage during intrathecal administration of opioids. These include the catheter itself, the type of drugs (morphine, bupivacaine, etc.) administered, direct effects of cancer invasion, therapeutic anticancer drugs, and radiation therapy. With regard to the catheter, those used in the current study were made of polyurethane or a polyamide resin. In experiments with dogs, polyurethane catheters were reported to cause inflammatory masses in the subdural space 5). On the other hand, in research in which Yaksh et al. 6) used a dog and administered mg/day of morphine to the subarachnoid space for 28 days, it was reported that even with a polyethylene catheter, an inflammatory mass formed at the tip. It is possible, therefore, that the formation of inflammatory mass is not related to the catheter material. A polyurethane catheter was used in Patient 1, who was found to have localized fibrosis and inflammatory cell infiltration in the subdural space. Experiments with dogs and human studies both suggest that a polyurethane catheter may be associated with the development of inflammatory mass. Inflammatory mass was not noted in Patients 2 or 3, who both received long term administration of morphine. That is to say, in this study the involvement of morphine was negatively correlated with inflammatory mass, while physical involvement of the catheter was strongly indicated to be pertinent to the formation of inflammatory mass. As for administration of drugs, morphine, fentanyl, and bupivacaine were administered to the subarachnoid space of the current patients. Severe necrotic degeneration and gliosis of spinal neurons and demyelination in the dorsal horn and dorsal roots were noted in patients who received long term (Patients 2 and 3), but not short term (Patients 4 6). Thus, animal experiments and human studies have both indicated that nerve damage due to intrathecal administration of opioids primarily involves morphine. Although Patient 7 received long term administration of fentanyl, no necrotic degeneration and gliosis of spinal neurons or demyelination in the dorsal horn and dorsal roots were noted. Since bupivacaine was administered at the same time as fentanyl we consider that neither fentanyl nor bupivacaine is neurotoxic. Only 1 of the current patients was administered fentanyl; future study with additional cases is needed. Mercadante et al. 7) reported that there were no neurological sequelae and that neurotoxicity did not readily develop if the dose of bupivacaine did not exceed mg/day. Additionally, Perren et al. 8) have reported that inflammatory mass forming in the spinal cord has been recognized in cases in which intrathecal administration of a mixture of bupivacaine and clonidine has been continued for three years or more. In these cases, the dosage of bupivacaine was mg/day. The dose of bupiva- Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine 193

3 Table 1 Characteristics of patients A. Autopsied patients Patient Age & sex Underlying condition 1 65 y.o. male Urothelial cancer, lt. ureter Type of opioid administered Site of catheter insertion morphine L3/4 L2 Location of catheter tip 2 53 y.o. female Cervical cancer morphine L3/4 unspecified 3 53 y.o. male Rectal cancer morphine L3/4 L y.o. male Prostate cancer morphine L2/3 unspecified 5 50 y.o. male Rectal cancer morphine L3/4 L y.o. female Cervical cancer morphine L3/4 L y.o. female Undifferentiated pleomorphic sarcoma fentanyl L3/4 L2 B. Neurohistopathological changes following intrathecal administration of an opioid and bupivacaine Ventral Total dose of Total dose of Days of Dorsal horn of nerve bupivacaine (mg) opioid (mg) administration the spinal cord roots Spinal cord 1 2, ( ) Demyelination, mild ( ) , ( ) Necrotic degeneration Demyelination, and gliosis of spinal severe neurons 3 13,420 5, ( ) Necrotic degeneration Demyelination, and gliosis of spinal severe neurons ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 7 1, ( ) ( ) ( ) C. Range of dosage amounts of opioid and bupivacaine and infusion pump flow rate Morphine hydrochloride (mg/day) Fentanyl (μg/day) Bupivicaine (mg/day) Infusion rate Lowest dose Peak dose Lowest dose Peak dose Lowest dose Peak dose (ml/h) , caine given to the current patients was about 5 60 mg/day, which was almost the same as daily dosage that reported by Mercadante et al. 7) Compared to the daily dosage administered by Perren et al. 8) ( mg/day), while our daily dosage (5 60 mg/day) was higher, the period of administration was shorter, at less than one year for all our cases. Long term administration of bupivacaine might cause neurotoxicity even in the dosage commonly used. Table 2 compares our autopsied patients with those in previous studies. There were no major differences in the doses of morphine and bupivacaine and duration of their administration. As in the current study, Sjöberg et al. 2) found nerve 194 Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine

4 Table 2 Comparison of neurohistopathological changes following intrathecal administration of morphine and bupivacaine in the current patients and previous reports Source. patients Dose of morphine (mg) Days of Pathological (mg) administration changes 10 41, Demyelination 22 3, , ne 18 5, , Demyelination Sjöberg (1992) ,900 Wagemans (1997) 10 7 Current study Dose of bupivacaine Fig. 1 HE staining (Micrograph from patient 2) te necrotic degeneration of nerve cells in the dorsal horn of the spinal cord (white arrows). Fig. 3 GFAP immunostaining (Micrograph from patient 2) te gliosis of the dorsal horn of the spinal cord. Many fibrillary astrocytes are found (white arrows). clinically reported to cause Lhermitte s sign9). Patients in the current study who were found to have severe nerve damage (Patients 2 and 3) Dorsal horn did not undergo chemotherapy. Patients 1 and 4 underwent radiation Dorsal root therapy while Patients 2 and 3 who did not undergo radiation therapy had severe damage. With regard to nerve damage due to radiation therapy, a report indicated that 0.18 of patients receiving irradiation at a dose of Gy develop myelitis10). clinical manifestations suggestive of myelitis were noted in our patients. Patients in the current study received a substantial radiation dose: Patient 1 received a radiation dose of 50 Gy while Patient 4 received a dose of 30 Gy. Localized fibrosis and inflammation of the subdural space were not Fig. 2 uxol fast blue staining (myelin sheath L staining) (Micrograph from patient 3) Demyelination of the dorsal horn and dorsal roots is apparent (black arrows). noted in Patients 2 and 3, who had severe nerve damage, so the inflammatory masses observed in our study were unlikely to have resulted from morphine and bupivacaine. Thus, we consider that the nerve damage in our patients was caused by long term administration of morphine to the subarachnoid space. In humans, continuous infusion of morphine intrathecally is clinically reported to cause dysfunc- damage, however a study by Wagemans et al. found no nerve dam- tion and paralysis of sensory nerves and motor nerves11, 12). A conflict- age. Further study is needed to clarify this discrepancy. ing report, however, indicated that high doses of morphine 4) Another factor to examiner is cancer metastasis. Cancer was found administered intrathecally over a long period of time (i.e. 90 days) did outside the dura mater in Patient 2. Metastases were large enough to not produce clinical neurologic deficits13). A study of autopsy speci- be identified microscopically but nerve damage was not noted in their mens from patients who were administered morphine intrathecally vicinity, so the cancer was not associated with nerve damage. Yet an- for a long period of time also noted no nerve damage14). Thus there other element to consider is chemotherapy. Patients 1, 5, and 7 under- are contradicting reports, and no consensus has been reached on this went chemotherapy. The anticancer drugs used were cisplatin, bleo- issue. In our experiment, spinal cord nerve damage resulting from the mycin, and mitomycin. Unlike the latter 2 drugs, cisplatin causes long term administration of morphine to the subarachnoid space was peripheral neuropathy and degeneration of the dorsal column and is histopathologically observed, but neurological deficit such as sensory Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine 195

5 impairment resulting from this damage were not recognized. Conclusion The total dose of morphine used for intrathecal analgesia and the duration of its administration were related to the extent of nerve damage. Animal experiments and human studies have both indicated that nerve damage due to intrathecal analgesia is primarily associated with morphine. In line with other reports, bupivacaine did not lead to nerve damage at the doses administered to patients in the current study. In addition, the current results suggested that long term administration of fentanyl may not cause nerve damage. References 1) Wang JK, Nauss LA, Thomas JE. Pain relief by intrathecally applied morphine in man. Anesthesiology 1979; 50: ) Sjöberg M, Karlsson PA, rdborg C, et al. Neuropathologic findings after long term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients. Anesthesiology 1992; 76: ) Karlsson PA, rdberg C, Sjöberg M, et al. Neuropathological findings after long term intrathecal infusion of buprenorphine and bupivacaine of pain treatment in cancer patients. Eur J Pain 1994; 15: ) Wagemans MFM, van der Valk P, Spoelder EM, et al. Neurohistopathological findings after continuous intrathecal administration of morphine or a morphine/bupivacaine mixture in cancer pain patients. Acta Anaesthesiol Scand 1997; 41: ) Coombs DW, Colburn RW, Allen CD, et al. Toxicity of chronic spinal analgesia in a canine model: neuropathologic observations with dezocine lactate. Reg Anesth 1990; 15: ) Yaksh TL, Horais KA, Tozier NA, et al. Chronically infused intrathecal morphine in dogs. Anesthesiology 2003; 99: ) Mercadante S. Problems of long term spinal opioid treatment in advanced cancer patients. Pain 1999; 79: ) Perren F, Buchser E, Chédel D, et al. Spinal cord lesion after long term intrathecal clonidine and bupivacaine treatment for the management of intractable pain. Pain 2004; 109: ) Walther PJ, Rossitch E Jr, Bullard DE. The development of Lhermitte s sign during cisplatin chemotherapy. Possible drug induced toxicity causing spinal cord demyelination. Cancer 1987; 60: ) Marcus RB Jr, Million RR. The incidence of myelitis after irradiation of the cervical spinal cord. Int J Radiat Oncol Biol Phys 1990; 19: ) Rawal N, Arnér S, Gustafsson LL, et al. Present state of extradural and intrathecal opioid analgesia in Sweden. A nationwide follow up survey. Br J Anaesth 1987; 59: ) Greenberg HS, Taren J, Ensminger WD, et al. Benefit from and tolerance to continuous intrathecal infusion of morphine for intractable cancer pain. J Neurosurg 1982; 57: ) Arnér S, Rawal N, Gustafsson LL. Clinical experience of long term treatment with epidural and intrathecal opioids a nationwide survey. Acta Anaesthesiol Scand 1988; 32: ) Coombs DW, Fratkin JD, Meier FA, et al. Neuropathologic lesions and CSF morphine concentrations during chronic continuous intraspinal morphine infusion. A clinical and post mortem study. Pain 1985; 22: Potential conflicts of interest; The author(s) indicated no potential conflicts interest. 196 Neurohistopathological findings after continuous intrathecal administration of opioids and bupivacaine

6 原著 がん性疼痛に対する持続くも膜下オピオイド 鎮痛法に伴う神経病理学的所見 河村 康司1), 福重 哲志2) 1) 田川市立病院 緩和ケア内科, 2) 久留米大学 緩和ケアセンター くも膜下オピオイド鎮痛法 (ITO) に伴う脊髄神経細胞の変性壊死や脱髄などの脊髄, 神経障害に関して病理組織学的に研 究された報告はきわめて少ない. ITO により神経組織障害が引き起こされたとする報告とないとする報告があり, 意見の一 致をみていない. 今回 ITO を施行された 7 症例 (男性 4 名, 女性 3 名) の剖検材料を用いて神経病理学的に検討した. モルヒ ネとブピバカインの投与が 6 例, フェンタニルとブピバカインの投与が 1 例であり投与期間は 日である. モルヒネ を長期間投与した 2 症例は高度の脊髄神経細胞の変性壊死, グリオーシスおよび後角, 後根の脱髄が認められたが, 短期間 しか投与しなかった 4 症例およびフェンタニル投与例ではこれらの神経病理学的変化は認められなかった. ITO で使用し たモルヒネの総投与量および投与期間と脊髄神経障害の程度には関連性があり, ITO による神経障害には主としてモルヒネ が関与していると考えられた. Palliat Care Res 2013; 8(2): Key words: がん性疼痛, くも膜下オピオイド鎮痛法, 神経病理学的所見 図 1 HE 染色 (症例 2) 脊髄後角の神経細胞の変性, 壊死がみられる (矢印). 後角 図 3 GFAP 染色 (症例 2) 脊髄後角のグリオーシス. 矢印は反応性に肥大し 突起の発達した星状膠細胞を示す. 後根 図 2 Luxol fast blue 染色 (髄鞘染色) (症例 3) 脊髄後角および後根の脱髄 (矢印) が認められる. 持続くも膜下鎮痛法に伴う神経病理学的所見 197

7 表 1 患者背景 A. 剖検症例 症例 年齢, 性別 原疾患 投与オピオカテーテルカテーテルイドの種類の穿刺部位の先端部位 1 65 歳, 男性 左尿管がん モルヒネ L3/4 L 歳, 女性 子宮頸がん モルヒネ L3/4 詳細不明 3 53 歳, 男性 直腸がん モルヒネ L3/4 L 歳, 男性 前立腺がん モルヒネ L2/3 詳細不明 5 50 歳, 男性 直腸がん モルヒネ L3/4 L 歳, 女性 子宮頸がん モルヒネ L3/4 L 歳, 女性 未分化多型肉腫フェンタニル L3/4 L2 B. くも膜下腔オピオイド ブピバカイン投与後の神経病理学的変化 ブピバカイン総量 (mg) 1 2, , ,771 オピオイド総量 (mg) 投与日数 脊髄前根 脊髄後根 脊髄 ( ) 脱髄, 軽度 ( ) 2,090 神経細胞の変性壊死 113 ( ) 脱髄, 高度 とグリオーシス 5,200 神経細胞の変性壊死 345 ( ) 脱髄, 高度 とグリオーシス 18 6 ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 42 ( フェンタニル ) 112 ( ) ( ) ( ) C. くも膜下腔オピオイド ブピバカイン投与量の幅および注入速度 塩酸モルヒネ (mg/ 日 ) フェンタニル (μg/ 日 ) ブピバカイン (mg/ 日 ) 注入速度最低投与量最高投与量最低投与量最高投与量最低投与量最高投与量 (ml/ 時 ) , 表 2 過去のくも膜下モルヒネ ブピバカイン投与後の神経病理学的研究報告と自験例との比較 参考文献 症例数 モルヒネのブピバカインの投与量 (mg) 投与量 (mg) 投与日数 病理学的変化 Sjöberg (1992) , , 脱髄 Wagemans (1997) , , なし 自験例 , , 脱髄 198 持続くも膜下鎮痛法に伴う神経病理学的所見