Lumbar Subarachnoid Ethylenediaminetetraacetate Induces Hindlimb Tetanic Contractions in Rats: Prevention by CaC1, Pretreatment; Observation of Spinal Nerve Root Degeneration B. C. Wang, MD, D. Li, MA, and D. E. Hillman, PhD
MB,
J. M. Hiller, PhD, E. J. Simon,
PhD,
G. Budzilovich,
MD,
Departments of Anesthesiology, Psychiatry, Pharmacology, Neuropathology, Physiology, and Biophysics, New York University School of Medicine, New York, New York
Disodium ethylenediaminetetraacetate (Na,EDTA) has replaced sodium bisulfite as the antioxidant in 2-chloroprocaine, Nesacaine CE. This study was undertaken to determine whether this new formulation has neurotoxic effects when administered in the subarachnoid space. Sprague-Dawley rats receiving subarachnoid injections of 1.5 mM or higher concentrations of Na,EDTA immediately initiated a circling behavior that was followed by the development of tetanic contractions of the hindlimbs lasting for 15-20 min. The tetanic contractions were followed by a brief period of hindlimb paralysis.
S
odium bisulfite, an antioxidant contained in 2-chloroprocaine, Nesacaine CE, has recently been replaced by a chelator, disodium ethylenediaminetetraacetate (EDTA) dihydrate. A number of adverse effects have been reported after enteral and parenteral administration of EDTA. These include teratogenicity (1);collagen fibril depletion (2); intestinal villi shortening (3); neural crest tumor formation and cell death in rats and mice (4); hepatic necrosis in calves (5); and contact allergy (6), vasculitis, cardiac tamponade, hemolytic anemia, and renal damage in humans (7). Moreover, systemic hypocalcemia and increased parathyroid hormone have been reported in humans after radiography with contrast media containing EDTA (8). Use of this type of contrast for coronary angiography has also caused ventricular fibrillation and repolarization changes (9). Ethylenediaminetetraacetate containing 2-chloroprocaineMPF has been used widely as an epidural anesthetic Accepted for pubiication A p d 23, 1992. Address correspondence to Dr. Wang, Department of Anesthesiology, New York University School of Medicine, 550 First Avenue, New York, NY 10016. 01992 by the International Anesthesia Research Society 0003-2999/92/$5.00
Pretreatment of rats by subarachnoid injections of 1 mM CaC1, prevented the development of tetanic contractionsand paralysis of the hindlimb. Histologic examination of animals receiving Na,EDTA revealed moderate to severe focal degenerative changes in spinal nerve roots. Control rats receiving subarachnoid injections of normal saline solution did not develop tetanic contraction nor pathological changes on light microscopy. These results suggest that the preservative used in Nesacaine-MPF may be neurotoxic. (Anesth Analg 1992;75:895-9)
and may enter the subarachnoid space through accidental dural puncture. The effects of EDTA in the subarachnoid space are unknown. A rat model was used to test the effects of EDTA in the subarachnoid space.
Methods With approval of our Animal Care and Research Committee, Sprague-Dawley rats weighing 500-600 g were anesthetized with 0.5 mL/kg intramuscular injection of an anesthetic mixture that contained ketamine (42.85 mg/mL), xylazine (8.57 mg/mL), and acepromazine (1.42 mg/mL). After lumbar laminectomy, with aseptic technique, PElO subarachnoid space (SAS) catheters (Becton Dickinson, Parsippany, N.J-) were implanted under direct vision with a technique recently reported (10). One week after surgery, correct position of the SAS catheter was verified by observing temporary hindlimb paralysis after induction of spinal block with 0.05 mL of 1.5% lidocaine. One or more days after this test, the animals were divided into five groups receiving SAS injections at 1-h intervals, as follows: group 1 (control Anesth Analg 1992;75:895-9
895
896
ANESTH ANALG 1!392;75:895-9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL SUBARACHNOID EDTA: TETANY, DEGENERATION
Table 1. Occurrence of Tetanic Contractions After Subarachnoid Injections of Different Concentrations of Na,EDTA in Rats Na,EDTA Rat No.
0.01%
0.025%
1 2 3
NE NE NE NE NE NE
NE NE NE NE NE NE
4
5 6
0.05%
0.1%
NE 3rd 2nd NE NE 1st
NE 2nd 1st 1st 1st 1st
EDTA, ethylenediaminetetraacetate; NE, no effect (no tetanic contraction); lst, 2nd, 3rd, injection number when tetanic contractions were observed.
animals, n = 6) received three SAS injections (at 1-h intervals) of normal saline solution (NS) (50 p L [0.05 mL], pH 5). Group 2 (EDTA only, n = 6) received (at 1-h intervals) three to four SAS injections of 0.05 mL of Na,EDTA (molecularweight 336.21, pH 4 . 5 4 . 8 ) of varied concentrations on different days, including 0.01% (0.3 mM), 0.025% (0.75 mM), 0.05% (1.5 mM), and 0.1% (3 mM). Group 3 (CaC1, + Na,EDTA, n = 7), at 1-h intervals, received three SAS injections of 148 pL of 1mM (0.01%)CaC1, (molecular weight 110.99, pH 5-5.7), which was immediately followed by 148 p L of 3 mM (0.1%) Na,EDTA. Additionally, at 1-h intervals, two other rats received the CaC1, + EDTA combination six times. Group 4 (histologic study, n = 3 [not counting the rat from group 21) received eight injections of 0.05 mL of 0.1% Na,EDTA at 1-h intervals and were killed 3 days later for pathological examination of the spinal cords. Group 5 (additional histologic study of saline solution control animals, n = 6) received 0.05 mL of NS at 1-h intervals for 8 h and were killed 5 days later for pathological examination.
Results There was no demonstrable motor abnormality after SAS injection of NS and 0.01% and 0.025% Na,EDTA. One minute after 0.1% Na,EDTA injection, five of six rats in group 2 circled around continuously and then developed tetanic contractions of the hindlimbs at a frequency of 100+/min. Four animals manifested this behavior after the first injection and one after the second injection. Three rats developed tetanic contractions of hindlimbs after 0.05% Na,EDTA injection: one after the first, one after the second, and one after the third injection. Hindlimb paralysis lasting 10-15 min usually followed the tetanic contractions. Table 1 summarizes the results in group 2. Tetanic contractions and other motor abnor-
Figure 1. Low-power magnification ( x 25) of hematoxylin-eosinstained section of rat spinal nerve root 6 days after receiving subarachnoid injection of Na,EDTA. Approximately one-third of the fibers show acute vacuolar and granular degeneration. A relatively intact portion of the root is shown in the right upper corner.
malities were not seen in group 3 rats when SAS Na,EDTA injections were preceded by CaCl,. Tetanic contractions developed in all of the group 4 animals between the fourth and eighth injections. Pathological examinations were performed in 10 animals: one from group 2, three from group 4, and six from group 5. The animal from group 2 was killed for histologic examination 6 days after receiving SAS Na,EDTA injection every hour for 3 h. This rat had tetanic contractions after receiving the second and third SAS injections containing 0.1% and 0.05% Na,EDTA, respectively. Before autopsy, this rat had already recovered from hindlimb paralysis and showed no sign of motor abnormality. Histologic examination revealed focal lesions characterized by destruction of axons and their myelin sheaths, resulting in cystic degeneration of nerve roots. Prominent macrophage activity was evident (Figures 1 and 2). The spinal cords of two of the three animals in group 4 revealed Wallerian degeneration of spinal nerve roots (Figures 3 and 4). In the histologic sections examined, the remaining rat in group 4 did not show degenerative changes. The spinal cord and spinal roots of group 5 rats did not show pathological changes after SAS injection of NS. There were no tetanic contractions in group 5.
Discussion Before the formulation change, Chloroprocaine CE, Nesacaine CE contained 0.2% sodium bisulfite, which was shown to be neurotoxic in rabbits (11-13) and rats (14). Subsequently, the manufacturer reduced the antioxidant content to 0.07% and eventu-
ANESTH ANALG 1992;75:89.5-9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL. SUBARACHNOID EDTA: TETANY, DEGENERATION
Figure 2. High-power magnification ( X 9 0 ) of Figure 1 showing focal lesions characterized by destruction of axons and their myelin sheaths, resulting in vacuolar and granular degeneration. Prominent macrophage activity was evident. Hematoxylin-eosin stain.
897
Figure 4. Spinal nerve root 3 days after intrathecal administration of Na,EDTA. There is diffuse Wallerian degeneration of individual nerve fibers characterized by formation of vacuoles containing fragments of axons and myelin sheaths. Hematoxylin-eosin stain, X50.
Tetunic Contractions
Figure 3. Spinal cord (top) and spinal nerve root (bottom) of rat 3 days after intrathecal administration of Na,EDTA. There is multifocal degeneration of nerve root fibers characterized by formation of vacuoles containing fragments of axons and myelin sheaths. Hematoxylin-eosin stain, x 25.
ally eliminated it altogether, replacing it with EDTA. Recent reports of severe back pain after epidural chloroprocaine (Nesacaine-MPF) prompted the authors to examine the effects of EDTA in the SAS to determine if the new formulation of chloroprocaineis appropriate. We did not pinpoint the etiology of the back pain but were led to believe that EDTA is the possible cause by the following findings: (a) SAS administration of EDTA induced tetanic contractions, probably as a result of hypocalcemia; (b) CaCl, pretreatment prevented the development of tetanic contractions; (c) microscopic evidence of spinal nerve root degeneration was observed, a condition that may be associated with nerve root irritation causing pain. These three findings are discussed next.
The fact that low concentrations (0.01% and 0.025%) of Na,EDTA did not cause tetanic contractions indicates a concentration-dependent effect. As shown in Table 1, four rats developed tetanic contractions after the first injection of 0.1% EDTA, whereas with 0.05% EDTA, only one developed tetanic contractions after the first injection, one after the second, and one after the third. This would suggest that cumulative dosage and duration of contact may influence the result. Because the SAS of the rat is very small, the volume of the injectate has to be limited. For this reason, we had to use concentrations of EDTA higher than the 0.011% contained in the commercial solution of chloroprocaine-MPF; however, in terms of total milligrams, the dose that we used was still only approximately 1/40 of the amount of EDTA contained in a single epidural dose of chloroprocaine-MPF in humans. In general, based on body weight, drug dosage for rats is usually much larger than that for humans. Ethylenediaminetetraacetate chelates calcium, which plays an important role in neuromuscular function, including repolarization of the myocardium (9). Hypocalcemia causes tetany; however, the exact mechanism for the production of tetanic contraction after SAS EDTA administration remains unresolved. In a previous study (15), tetanic contractions were also produced by EDTA administered in the epidural space. As expected, the latency period before the development of tetanic contractions after epidural EDTA was longer than that after SAS injection, because it takes time to pass through the dura.
898
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL SUBARACHNOID EDTA: TETANY, DEGENERATION
Calcium Chloride Pretreatment
Degeneration of the Spinal Nerve Roots
The fact that pretreatment with CaCl, prevented the development of tetanic contractions in group 3 suggests that hypocalcemia secondary to chelation could be the responsible factor. The preventive role of CaC1, was further supported by the fact that the two additional rats in group 3 did not develop tetanic contractions even after receiving, at 1-h intervals, eight injections of the combination of CaC1, + EDTA. The report of successful treatment of severe back pain after epidural Nesacaine-MPF with CaC1, (16) reinforced this support. Patients with parathyroid hormone deficiency, impaired calcitriol synthesis, hypomagnesemia and hypermagnesemia, and diseases of the kidneys and liver are especially at risk for hypocalcemia (17). Vitamin D is synthesized in the skin or absorbed by the gut, 25-hydroxylated in the liver, and then 1hydroxylated in the kidneys to its most active form, calcitriol. Therefore, diseases of the kidneys or liver could impair calcium homeostasis, and in patients with such ailments even a low concentration of EDTA may be unsafe. Myoclonic seizures, probably secondary to increased serotonergic activity in the spinal cord, have been reported in rats after a tenfold analgesic dose of intrathecal morphine. This effect was reduced by pretreatment with parachlorophenylalanine in doses that reduced spinal 5-hydroxytryptamine (serotonin) by approximately 60% (18). Speculatively, if EDTA produces tetanic contractions by acting on the serotonergic pathway in the spinal cord, it would be inadvisable to add EDTA to intrathecal morphine preparations. Opioid receptors may be affected by EDTA through calcium modulation. The binding of opioid receptors is affected by the valence of the cation (i.e., monovalent or divalent) and the site of binding (i.e., p, 6, K ) . In membrane suspensions from guinea pig brain or cerebellum, Paterson et al. (19) found that CaCl, inhibited equilibrium binding at p and K sites but potentiated binding at S sites. Kappa receptors have been linked to direct blockade of the calcium channel (19). The K-ligand dynorphin A decreases basal free intracellular calcium in rat cortical synaptosomes. Through calcium modulation, EDTA may affect the calcium channel (ZO), the secondary messenger, the muscarinic receptor (21), and the acetylcholine and norepinephrine neurotransmitter release (22,23). There is evidence suggesting that leucine enkephalin and acetylcholine are cotransmitters stored in and released from the same nerve terminals in the parasympathetic ganglia of the bladder (24).
Our preliminary observation that spinal nerve roots had degenerated lends support to similar results from previous investigations of epidural and SAS administration of EDTA (15,25).This would suggest that the reported severe back pain after epidural chloroprocaine-MPF (16,2630) could be due to spinal nerve root irritation caused by EDTA contained in the anesthetic solution. Unfortunately, we did not perform pathological studies on all of our animals. Further research, including studies of NesacaineMPF, are warranted. Recently, on the basis of autopsy findings in 15 patients, without substantiating evidence, Sjoberg et al. (31) have tried to exonerate low-concentration sodium metabisulfite and EDTA from causing neurotoxicity. We commend their efforts in the study but disagree with their interpretation of results. Their table of ”Pathoanatomic findings” listed several patients with neurilemoma, loss of axon, demyelination of spinal nerve roots, 10 patients with “puzzling” mononuclear infiltration (”five had radiotherapy of the spinal column and three had potentially neurotoxic cytostatic treatment”), one patient with degeneration of the dorsal column and dorsal horns with focal subpial vacuolization and demyelination at the level of a neurilemoma, and one patient with arterial thrombosis and medullary infarction of the spinal cord. One patient developed paraplegia and urinary retention 158 days after the start of the intrathecal treatment, and the authors called this ”an expected consequence of large daily dose (7-10 mg/h) of intrathecal bupivacaine given to cope with severe neurogenic pains.” They claim that ”the incidence and extent of the microscopic nerve root lesion were not related to the concentrations and cumulative doses of morphine (and its preservatives) or bupivacaine, the osmolarity and pH of the analgesic mixture, or the duration of treatment.” It appears that they attribute the pathological lesions to tumor invasion, radiation therapy, and treatment with potential neurotoxic agents. Without a control study for substantiation, it would be inappropriate to make such claims that might lead medical professionals to disregard the potential toxicity of bisulfite and EDTA. We have observed nearly all these focal pathological findings in our animal models with use of EDTA (15,25) in the absence of tumor invasion, irradiation, or treatment with neurotoxic agents (Figures 1 4 ) . Irreversible hindlimb paralysis has also been reported after intrathecal bisulfite administration in rabbits (11). We do not assume that preservatives are responsible for all of the lesions described by Sjoberg et al. (31), but we cannot rule out EDTA and bisulfite as the cause of pathology in some of the patients. Acciden-
ANESTH ANALG 1992;7589%9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL. SUBARACHNOID EDTA: TETANY, DEGENERATION
tal dural puncture, although rare, can be a nuisance in epidural anesthesia. Drugs intended for epidural administration could inadvertently enter the SAS. Therefore, incorporating EDTA in epidural anesthetics may be unwise before its safety is proved. Moreover, the fact that the cases of severe back pain were reported after epidural anesthesia with chloroprocaine containing EDTA suggests that complications can occur even without dural puncture. We thank Dr. Herman Turndorf, Professor and Chairman of the Department of Anesthesiology of New York University, for his continued support and encouragement. We also thank Dr. Sanford Miller for reading the manuscript.
References 1. Brownie CF, Noden D, Krook L, Haluska M, Aronson AL. Teratogenic effect of calcium edetate (CaEDTA)in rats and the protective effect of zinc. Toxicol Appl Pharmacol 1986;82:42643. 2. Braide VB, Grill V, DeIbello G. Calcium ethylenediaminetetraacetate toxicity in the rat: ultrastructural effects on skin collagen. Pharmacol Res Commun 1988;2013?-46. 3. Rosenblatt DE, Doyle DG, Aronson AL. Calcium ethylenediaminetetraacetate (CaEDTA) toxicity: time and dose dependent studies on intestinal morphology in the rat. Exp Mol Pathol 1978;28:215-26. 4. Nozue AT. Effects of EDTA in newborn mice with special reference to neural crest cells. Anat Anz 1988;166:209--17. 5. Bulgins MS, Maas J, Anderson BC, Hugh D, Lincoln SD, Lane VM. Death associated with parenteral administration of copper disodium edetate in calves. J Am Vet Med Assoc 1986;188: 406-9. 6. De Groot AC. Contact allergy to EDTA in a topical corticosteroid preparation. Contact Dermatitis 1986;15:250-2. 7. Peterson GR. Adverse effects of chelation therapy (letter). JAMA 1983;2502926. 8. Mellette LE, Gomez LS. Systemic hypocalcemia after clinical injections of radiographic contrast media: amelioration by omission of calcium chelating agents. Radiology 1983;147: 677-9. 9. Zukerman LS, Friehling TD, Wolf NM, Meister SG, Nahass G, Kowey PR. Effect of calcium-binding additives on ventricular fibrillation and repolarization changes during coronary angiography. J Am Coll Cardiol 1987;10:1249-53. 10. Wang BC, Li D, Hillman DE, Turndorf H. Lumbar subarachnoid catheterization in rats. Pharmacol Biochem Behav 1991; 38:685-8. 11. Wang BC, Hillman DE, Spielholz NI, Turndorf H. Chronic neurological deficits and Nesacaine-CE-an effect of the anesthetic, 2-chloroprocaine, or the antioxidant, sodium bisulfite. Anesth Analg 1984;63:445-7. 12. Ready LB, Plumer MH, Haschke RH, Austin E, Sumi M. Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology 1985;63:364-70.
899
13. Gissen AJ, Datta S, Lambert D. The chloroprocaine controversy: 11. Is chloroprocaine neurotoxic? Reg Anaesth 1984;9: 13-5. 14. Hersh EV, Havelin D, Condouris GA. Evaluation of subarachnoid 2-chloroprocaine and bisulfite in the rat. Reg Anaesth 1987;12:5-9. 15. Wang BC, Li D, Hiller JM, Simon EJ, Budzilovich G, Turndorf H. Epidural EDTA induces tetanic contractions in rats (abstract). Anesth Analg 1991;72(Suppl):S312. 16. Dirkes J. Treatment of Nesacaine-MPF induced back pain with calcium chloride (letter). Anesth Analg 1990;70:461-2. 17. Prielipp RC, Zaloga GP. Life-threatening hypocalcemia after abdominal aortic aneurysm repair in patients with renal insufficiency. Anesth Analg 1991;73:638-41. 18. Shohami E, Evron S. Intrathecal morphine induces myoclonic seizures in the rat. Acta Pharmacol Toxicol 1985;56:50-4. 19. Paterson SJ, Robson LE, Kosterlitz HW. Control of cations of opioid binding in guinea pig membranes (monovalent cations/ divalent cations/k, S and K-binding sitedguinea pig cerebellum). Proc Natl Acad Sci USA 1986;83:6216-20. 20. Welch SP, Singha AK, Dewey WL. The antinociception produced by intrathecal morphine calcium, A23187, U50488H, [D-Ala', N-Me-Phe4, Gly-Ollenkephalin and [D-Pen', D-Pen5]enkephalin after intrathecal administration of calcitonin gene-related peptide in mice. J Pharmacol Exp Ther 1989; 251:1-8. 21. Schwartz L, Atlas D. Synergy between membrane depolarization and muscarinic receptor activation lead to potentiation of neurotransmitter release. Brain Res 1989;503:62-7. 22. Clapp LH, Vivaudou MB, Walsh JV, Singer JJ. Acetylcholine increases voltage-activated CaZ+current in freshly dissociated smooth muscle cells. Proc Natl Acad Sci USA 1987;84:2092-6. 23. Cherubini E, North RA. Mu and kappa opioid inhibit transmitter release by different mechanisms. Proc Natl Acad Sci USA 1985;82:1860-3. 24. Kawatani M, Shioda S, Nakai Y , Takeshige C, De Groat WC. Ultrastructural analysis of enkephalinergic terminals in parasympathetic ganglia innervating the urinary bladder of the cat. J Comp Neurol 1989;288:81-91. 25. Wang BC, Li D, Hiller JM, Simon EJ, Budzilovich G . Subarachnoid disodium EDTA induces concentration dependent tetanic contraction in rats: prevention by CaC1, pretreatment (abstract). Anesthesiology 1991;75A651. 26. Hynson JM, Sessler DI, Glosten B. Back pain in volunteers after epidural anesthesia with chloroprocaine. Anesth Analg 1991;72:253-6. 27. Orkin FK, Bogetz MS. Back pain following uncomplicated epidural anesthesia with chloroprocaine (abstract). Anesthesiology 1989;71:A716. 28. Fibuch EE, Opper SE. Back pain following epidurally administered Nesacaine-MPF. Anesth Analg 1989;69:113-5. 29. Stevens RA, Chester WL, Artuso JD, Bray JG, Nellestein JA. Back pain after epidural anesthesia with chloroprocaine in volunteers: preliminary report. Reg Anaesth 1991;16:199-203. 30. Stevens RA, Barbara L, Urquhart BSN, Urmey WF, Kao TC. Back pain after chloroprocaine epidural anesthesia (abstract). Reg Anaesth 1992;17(Suppl 1):7. 31. Sjoberg M, Karlsson PA, Nordberg C, et al. Neuropathologc findings after long-term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients. Anesthesiology 1992;76:173-86.
MB,
J. M. Hiller, PhD, E. J. Simon,
PhD,
G. Budzilovich,
MD,
Departments of Anesthesiology, Psychiatry, Pharmacology, Neuropathology, Physiology, and Biophysics, New York University School of Medicine, New York, New York
Disodium ethylenediaminetetraacetate (Na,EDTA) has replaced sodium bisulfite as the antioxidant in 2-chloroprocaine, Nesacaine CE. This study was undertaken to determine whether this new formulation has neurotoxic effects when administered in the subarachnoid space. Sprague-Dawley rats receiving subarachnoid injections of 1.5 mM or higher concentrations of Na,EDTA immediately initiated a circling behavior that was followed by the development of tetanic contractions of the hindlimbs lasting for 15-20 min. The tetanic contractions were followed by a brief period of hindlimb paralysis.
S
odium bisulfite, an antioxidant contained in 2-chloroprocaine, Nesacaine CE, has recently been replaced by a chelator, disodium ethylenediaminetetraacetate (EDTA) dihydrate. A number of adverse effects have been reported after enteral and parenteral administration of EDTA. These include teratogenicity (1);collagen fibril depletion (2); intestinal villi shortening (3); neural crest tumor formation and cell death in rats and mice (4); hepatic necrosis in calves (5); and contact allergy (6), vasculitis, cardiac tamponade, hemolytic anemia, and renal damage in humans (7). Moreover, systemic hypocalcemia and increased parathyroid hormone have been reported in humans after radiography with contrast media containing EDTA (8). Use of this type of contrast for coronary angiography has also caused ventricular fibrillation and repolarization changes (9). Ethylenediaminetetraacetate containing 2-chloroprocaineMPF has been used widely as an epidural anesthetic Accepted for pubiication A p d 23, 1992. Address correspondence to Dr. Wang, Department of Anesthesiology, New York University School of Medicine, 550 First Avenue, New York, NY 10016. 01992 by the International Anesthesia Research Society 0003-2999/92/$5.00
Pretreatment of rats by subarachnoid injections of 1 mM CaC1, prevented the development of tetanic contractionsand paralysis of the hindlimb. Histologic examination of animals receiving Na,EDTA revealed moderate to severe focal degenerative changes in spinal nerve roots. Control rats receiving subarachnoid injections of normal saline solution did not develop tetanic contraction nor pathological changes on light microscopy. These results suggest that the preservative used in Nesacaine-MPF may be neurotoxic. (Anesth Analg 1992;75:895-9)
and may enter the subarachnoid space through accidental dural puncture. The effects of EDTA in the subarachnoid space are unknown. A rat model was used to test the effects of EDTA in the subarachnoid space.
Methods With approval of our Animal Care and Research Committee, Sprague-Dawley rats weighing 500-600 g were anesthetized with 0.5 mL/kg intramuscular injection of an anesthetic mixture that contained ketamine (42.85 mg/mL), xylazine (8.57 mg/mL), and acepromazine (1.42 mg/mL). After lumbar laminectomy, with aseptic technique, PElO subarachnoid space (SAS) catheters (Becton Dickinson, Parsippany, N.J-) were implanted under direct vision with a technique recently reported (10). One week after surgery, correct position of the SAS catheter was verified by observing temporary hindlimb paralysis after induction of spinal block with 0.05 mL of 1.5% lidocaine. One or more days after this test, the animals were divided into five groups receiving SAS injections at 1-h intervals, as follows: group 1 (control Anesth Analg 1992;75:895-9
895
896
ANESTH ANALG 1!392;75:895-9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL SUBARACHNOID EDTA: TETANY, DEGENERATION
Table 1. Occurrence of Tetanic Contractions After Subarachnoid Injections of Different Concentrations of Na,EDTA in Rats Na,EDTA Rat No.
0.01%
0.025%
1 2 3
NE NE NE NE NE NE
NE NE NE NE NE NE
4
5 6
0.05%
0.1%
NE 3rd 2nd NE NE 1st
NE 2nd 1st 1st 1st 1st
EDTA, ethylenediaminetetraacetate; NE, no effect (no tetanic contraction); lst, 2nd, 3rd, injection number when tetanic contractions were observed.
animals, n = 6) received three SAS injections (at 1-h intervals) of normal saline solution (NS) (50 p L [0.05 mL], pH 5). Group 2 (EDTA only, n = 6) received (at 1-h intervals) three to four SAS injections of 0.05 mL of Na,EDTA (molecularweight 336.21, pH 4 . 5 4 . 8 ) of varied concentrations on different days, including 0.01% (0.3 mM), 0.025% (0.75 mM), 0.05% (1.5 mM), and 0.1% (3 mM). Group 3 (CaC1, + Na,EDTA, n = 7), at 1-h intervals, received three SAS injections of 148 pL of 1mM (0.01%)CaC1, (molecular weight 110.99, pH 5-5.7), which was immediately followed by 148 p L of 3 mM (0.1%) Na,EDTA. Additionally, at 1-h intervals, two other rats received the CaC1, + EDTA combination six times. Group 4 (histologic study, n = 3 [not counting the rat from group 21) received eight injections of 0.05 mL of 0.1% Na,EDTA at 1-h intervals and were killed 3 days later for pathological examination of the spinal cords. Group 5 (additional histologic study of saline solution control animals, n = 6) received 0.05 mL of NS at 1-h intervals for 8 h and were killed 5 days later for pathological examination.
Results There was no demonstrable motor abnormality after SAS injection of NS and 0.01% and 0.025% Na,EDTA. One minute after 0.1% Na,EDTA injection, five of six rats in group 2 circled around continuously and then developed tetanic contractions of the hindlimbs at a frequency of 100+/min. Four animals manifested this behavior after the first injection and one after the second injection. Three rats developed tetanic contractions of hindlimbs after 0.05% Na,EDTA injection: one after the first, one after the second, and one after the third injection. Hindlimb paralysis lasting 10-15 min usually followed the tetanic contractions. Table 1 summarizes the results in group 2. Tetanic contractions and other motor abnor-
Figure 1. Low-power magnification ( x 25) of hematoxylin-eosinstained section of rat spinal nerve root 6 days after receiving subarachnoid injection of Na,EDTA. Approximately one-third of the fibers show acute vacuolar and granular degeneration. A relatively intact portion of the root is shown in the right upper corner.
malities were not seen in group 3 rats when SAS Na,EDTA injections were preceded by CaCl,. Tetanic contractions developed in all of the group 4 animals between the fourth and eighth injections. Pathological examinations were performed in 10 animals: one from group 2, three from group 4, and six from group 5. The animal from group 2 was killed for histologic examination 6 days after receiving SAS Na,EDTA injection every hour for 3 h. This rat had tetanic contractions after receiving the second and third SAS injections containing 0.1% and 0.05% Na,EDTA, respectively. Before autopsy, this rat had already recovered from hindlimb paralysis and showed no sign of motor abnormality. Histologic examination revealed focal lesions characterized by destruction of axons and their myelin sheaths, resulting in cystic degeneration of nerve roots. Prominent macrophage activity was evident (Figures 1 and 2). The spinal cords of two of the three animals in group 4 revealed Wallerian degeneration of spinal nerve roots (Figures 3 and 4). In the histologic sections examined, the remaining rat in group 4 did not show degenerative changes. The spinal cord and spinal roots of group 5 rats did not show pathological changes after SAS injection of NS. There were no tetanic contractions in group 5.
Discussion Before the formulation change, Chloroprocaine CE, Nesacaine CE contained 0.2% sodium bisulfite, which was shown to be neurotoxic in rabbits (11-13) and rats (14). Subsequently, the manufacturer reduced the antioxidant content to 0.07% and eventu-
ANESTH ANALG 1992;75:89.5-9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL. SUBARACHNOID EDTA: TETANY, DEGENERATION
Figure 2. High-power magnification ( X 9 0 ) of Figure 1 showing focal lesions characterized by destruction of axons and their myelin sheaths, resulting in vacuolar and granular degeneration. Prominent macrophage activity was evident. Hematoxylin-eosin stain.
897
Figure 4. Spinal nerve root 3 days after intrathecal administration of Na,EDTA. There is diffuse Wallerian degeneration of individual nerve fibers characterized by formation of vacuoles containing fragments of axons and myelin sheaths. Hematoxylin-eosin stain, X50.
Tetunic Contractions
Figure 3. Spinal cord (top) and spinal nerve root (bottom) of rat 3 days after intrathecal administration of Na,EDTA. There is multifocal degeneration of nerve root fibers characterized by formation of vacuoles containing fragments of axons and myelin sheaths. Hematoxylin-eosin stain, x 25.
ally eliminated it altogether, replacing it with EDTA. Recent reports of severe back pain after epidural chloroprocaine (Nesacaine-MPF) prompted the authors to examine the effects of EDTA in the SAS to determine if the new formulation of chloroprocaineis appropriate. We did not pinpoint the etiology of the back pain but were led to believe that EDTA is the possible cause by the following findings: (a) SAS administration of EDTA induced tetanic contractions, probably as a result of hypocalcemia; (b) CaCl, pretreatment prevented the development of tetanic contractions; (c) microscopic evidence of spinal nerve root degeneration was observed, a condition that may be associated with nerve root irritation causing pain. These three findings are discussed next.
The fact that low concentrations (0.01% and 0.025%) of Na,EDTA did not cause tetanic contractions indicates a concentration-dependent effect. As shown in Table 1, four rats developed tetanic contractions after the first injection of 0.1% EDTA, whereas with 0.05% EDTA, only one developed tetanic contractions after the first injection, one after the second, and one after the third. This would suggest that cumulative dosage and duration of contact may influence the result. Because the SAS of the rat is very small, the volume of the injectate has to be limited. For this reason, we had to use concentrations of EDTA higher than the 0.011% contained in the commercial solution of chloroprocaine-MPF; however, in terms of total milligrams, the dose that we used was still only approximately 1/40 of the amount of EDTA contained in a single epidural dose of chloroprocaine-MPF in humans. In general, based on body weight, drug dosage for rats is usually much larger than that for humans. Ethylenediaminetetraacetate chelates calcium, which plays an important role in neuromuscular function, including repolarization of the myocardium (9). Hypocalcemia causes tetany; however, the exact mechanism for the production of tetanic contraction after SAS EDTA administration remains unresolved. In a previous study (15), tetanic contractions were also produced by EDTA administered in the epidural space. As expected, the latency period before the development of tetanic contractions after epidural EDTA was longer than that after SAS injection, because it takes time to pass through the dura.
898
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL SUBARACHNOID EDTA: TETANY, DEGENERATION
Calcium Chloride Pretreatment
Degeneration of the Spinal Nerve Roots
The fact that pretreatment with CaCl, prevented the development of tetanic contractions in group 3 suggests that hypocalcemia secondary to chelation could be the responsible factor. The preventive role of CaC1, was further supported by the fact that the two additional rats in group 3 did not develop tetanic contractions even after receiving, at 1-h intervals, eight injections of the combination of CaC1, + EDTA. The report of successful treatment of severe back pain after epidural Nesacaine-MPF with CaC1, (16) reinforced this support. Patients with parathyroid hormone deficiency, impaired calcitriol synthesis, hypomagnesemia and hypermagnesemia, and diseases of the kidneys and liver are especially at risk for hypocalcemia (17). Vitamin D is synthesized in the skin or absorbed by the gut, 25-hydroxylated in the liver, and then 1hydroxylated in the kidneys to its most active form, calcitriol. Therefore, diseases of the kidneys or liver could impair calcium homeostasis, and in patients with such ailments even a low concentration of EDTA may be unsafe. Myoclonic seizures, probably secondary to increased serotonergic activity in the spinal cord, have been reported in rats after a tenfold analgesic dose of intrathecal morphine. This effect was reduced by pretreatment with parachlorophenylalanine in doses that reduced spinal 5-hydroxytryptamine (serotonin) by approximately 60% (18). Speculatively, if EDTA produces tetanic contractions by acting on the serotonergic pathway in the spinal cord, it would be inadvisable to add EDTA to intrathecal morphine preparations. Opioid receptors may be affected by EDTA through calcium modulation. The binding of opioid receptors is affected by the valence of the cation (i.e., monovalent or divalent) and the site of binding (i.e., p, 6, K ) . In membrane suspensions from guinea pig brain or cerebellum, Paterson et al. (19) found that CaCl, inhibited equilibrium binding at p and K sites but potentiated binding at S sites. Kappa receptors have been linked to direct blockade of the calcium channel (19). The K-ligand dynorphin A decreases basal free intracellular calcium in rat cortical synaptosomes. Through calcium modulation, EDTA may affect the calcium channel (ZO), the secondary messenger, the muscarinic receptor (21), and the acetylcholine and norepinephrine neurotransmitter release (22,23). There is evidence suggesting that leucine enkephalin and acetylcholine are cotransmitters stored in and released from the same nerve terminals in the parasympathetic ganglia of the bladder (24).
Our preliminary observation that spinal nerve roots had degenerated lends support to similar results from previous investigations of epidural and SAS administration of EDTA (15,25).This would suggest that the reported severe back pain after epidural chloroprocaine-MPF (16,2630) could be due to spinal nerve root irritation caused by EDTA contained in the anesthetic solution. Unfortunately, we did not perform pathological studies on all of our animals. Further research, including studies of NesacaineMPF, are warranted. Recently, on the basis of autopsy findings in 15 patients, without substantiating evidence, Sjoberg et al. (31) have tried to exonerate low-concentration sodium metabisulfite and EDTA from causing neurotoxicity. We commend their efforts in the study but disagree with their interpretation of results. Their table of ”Pathoanatomic findings” listed several patients with neurilemoma, loss of axon, demyelination of spinal nerve roots, 10 patients with “puzzling” mononuclear infiltration (”five had radiotherapy of the spinal column and three had potentially neurotoxic cytostatic treatment”), one patient with degeneration of the dorsal column and dorsal horns with focal subpial vacuolization and demyelination at the level of a neurilemoma, and one patient with arterial thrombosis and medullary infarction of the spinal cord. One patient developed paraplegia and urinary retention 158 days after the start of the intrathecal treatment, and the authors called this ”an expected consequence of large daily dose (7-10 mg/h) of intrathecal bupivacaine given to cope with severe neurogenic pains.” They claim that ”the incidence and extent of the microscopic nerve root lesion were not related to the concentrations and cumulative doses of morphine (and its preservatives) or bupivacaine, the osmolarity and pH of the analgesic mixture, or the duration of treatment.” It appears that they attribute the pathological lesions to tumor invasion, radiation therapy, and treatment with potential neurotoxic agents. Without a control study for substantiation, it would be inappropriate to make such claims that might lead medical professionals to disregard the potential toxicity of bisulfite and EDTA. We have observed nearly all these focal pathological findings in our animal models with use of EDTA (15,25) in the absence of tumor invasion, irradiation, or treatment with neurotoxic agents (Figures 1 4 ) . Irreversible hindlimb paralysis has also been reported after intrathecal bisulfite administration in rabbits (11). We do not assume that preservatives are responsible for all of the lesions described by Sjoberg et al. (31), but we cannot rule out EDTA and bisulfite as the cause of pathology in some of the patients. Acciden-
ANESTH ANALG 1992;7589%9
REGIONAL ANESTHESIA AND PAIN MANAGEMENT WANG ET AL. SUBARACHNOID EDTA: TETANY, DEGENERATION
tal dural puncture, although rare, can be a nuisance in epidural anesthesia. Drugs intended for epidural administration could inadvertently enter the SAS. Therefore, incorporating EDTA in epidural anesthetics may be unwise before its safety is proved. Moreover, the fact that the cases of severe back pain were reported after epidural anesthesia with chloroprocaine containing EDTA suggests that complications can occur even without dural puncture. We thank Dr. Herman Turndorf, Professor and Chairman of the Department of Anesthesiology of New York University, for his continued support and encouragement. We also thank Dr. Sanford Miller for reading the manuscript.
References 1. Brownie CF, Noden D, Krook L, Haluska M, Aronson AL. Teratogenic effect of calcium edetate (CaEDTA)in rats and the protective effect of zinc. Toxicol Appl Pharmacol 1986;82:42643. 2. Braide VB, Grill V, DeIbello G. Calcium ethylenediaminetetraacetate toxicity in the rat: ultrastructural effects on skin collagen. Pharmacol Res Commun 1988;2013?-46. 3. Rosenblatt DE, Doyle DG, Aronson AL. Calcium ethylenediaminetetraacetate (CaEDTA) toxicity: time and dose dependent studies on intestinal morphology in the rat. Exp Mol Pathol 1978;28:215-26. 4. Nozue AT. Effects of EDTA in newborn mice with special reference to neural crest cells. Anat Anz 1988;166:209--17. 5. Bulgins MS, Maas J, Anderson BC, Hugh D, Lincoln SD, Lane VM. Death associated with parenteral administration of copper disodium edetate in calves. J Am Vet Med Assoc 1986;188: 406-9. 6. De Groot AC. Contact allergy to EDTA in a topical corticosteroid preparation. Contact Dermatitis 1986;15:250-2. 7. Peterson GR. Adverse effects of chelation therapy (letter). JAMA 1983;2502926. 8. Mellette LE, Gomez LS. Systemic hypocalcemia after clinical injections of radiographic contrast media: amelioration by omission of calcium chelating agents. Radiology 1983;147: 677-9. 9. Zukerman LS, Friehling TD, Wolf NM, Meister SG, Nahass G, Kowey PR. Effect of calcium-binding additives on ventricular fibrillation and repolarization changes during coronary angiography. J Am Coll Cardiol 1987;10:1249-53. 10. Wang BC, Li D, Hillman DE, Turndorf H. Lumbar subarachnoid catheterization in rats. Pharmacol Biochem Behav 1991; 38:685-8. 11. Wang BC, Hillman DE, Spielholz NI, Turndorf H. Chronic neurological deficits and Nesacaine-CE-an effect of the anesthetic, 2-chloroprocaine, or the antioxidant, sodium bisulfite. Anesth Analg 1984;63:445-7. 12. Ready LB, Plumer MH, Haschke RH, Austin E, Sumi M. Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology 1985;63:364-70.
899
13. Gissen AJ, Datta S, Lambert D. The chloroprocaine controversy: 11. Is chloroprocaine neurotoxic? Reg Anaesth 1984;9: 13-5. 14. Hersh EV, Havelin D, Condouris GA. Evaluation of subarachnoid 2-chloroprocaine and bisulfite in the rat. Reg Anaesth 1987;12:5-9. 15. Wang BC, Li D, Hiller JM, Simon EJ, Budzilovich G, Turndorf H. Epidural EDTA induces tetanic contractions in rats (abstract). Anesth Analg 1991;72(Suppl):S312. 16. Dirkes J. Treatment of Nesacaine-MPF induced back pain with calcium chloride (letter). Anesth Analg 1990;70:461-2. 17. Prielipp RC, Zaloga GP. Life-threatening hypocalcemia after abdominal aortic aneurysm repair in patients with renal insufficiency. Anesth Analg 1991;73:638-41. 18. Shohami E, Evron S. Intrathecal morphine induces myoclonic seizures in the rat. Acta Pharmacol Toxicol 1985;56:50-4. 19. Paterson SJ, Robson LE, Kosterlitz HW. Control of cations of opioid binding in guinea pig membranes (monovalent cations/ divalent cations/k, S and K-binding sitedguinea pig cerebellum). Proc Natl Acad Sci USA 1986;83:6216-20. 20. Welch SP, Singha AK, Dewey WL. The antinociception produced by intrathecal morphine calcium, A23187, U50488H, [D-Ala', N-Me-Phe4, Gly-Ollenkephalin and [D-Pen', D-Pen5]enkephalin after intrathecal administration of calcitonin gene-related peptide in mice. J Pharmacol Exp Ther 1989; 251:1-8. 21. Schwartz L, Atlas D. Synergy between membrane depolarization and muscarinic receptor activation lead to potentiation of neurotransmitter release. Brain Res 1989;503:62-7. 22. Clapp LH, Vivaudou MB, Walsh JV, Singer JJ. Acetylcholine increases voltage-activated CaZ+current in freshly dissociated smooth muscle cells. Proc Natl Acad Sci USA 1987;84:2092-6. 23. Cherubini E, North RA. Mu and kappa opioid inhibit transmitter release by different mechanisms. Proc Natl Acad Sci USA 1985;82:1860-3. 24. Kawatani M, Shioda S, Nakai Y , Takeshige C, De Groat WC. Ultrastructural analysis of enkephalinergic terminals in parasympathetic ganglia innervating the urinary bladder of the cat. J Comp Neurol 1989;288:81-91. 25. Wang BC, Li D, Hiller JM, Simon EJ, Budzilovich G . Subarachnoid disodium EDTA induces concentration dependent tetanic contraction in rats: prevention by CaC1, pretreatment (abstract). Anesthesiology 1991;75A651. 26. Hynson JM, Sessler DI, Glosten B. Back pain in volunteers after epidural anesthesia with chloroprocaine. Anesth Analg 1991;72:253-6. 27. Orkin FK, Bogetz MS. Back pain following uncomplicated epidural anesthesia with chloroprocaine (abstract). Anesthesiology 1989;71:A716. 28. Fibuch EE, Opper SE. Back pain following epidurally administered Nesacaine-MPF. Anesth Analg 1989;69:113-5. 29. Stevens RA, Chester WL, Artuso JD, Bray JG, Nellestein JA. Back pain after epidural anesthesia with chloroprocaine in volunteers: preliminary report. Reg Anaesth 1991;16:199-203. 30. Stevens RA, Barbara L, Urquhart BSN, Urmey WF, Kao TC. Back pain after chloroprocaine epidural anesthesia (abstract). Reg Anaesth 1992;17(Suppl 1):7. 31. Sjoberg M, Karlsson PA, Nordberg C, et al. Neuropathologc findings after long-term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients. Anesthesiology 1992;76:173-86.
