Nerve injury and regional anaesthesia Alain Borgeat and Stephan Blumenthal

Transcription

1 Nerve injury and regional anaesthesia Alain Borgeat and Stephan Blumenthal Purpose of review In recent years there has been a renewed interest in regional anaesthesia, particularly peripheral nerve blockade, not only to improve the patient s well being, but also to meet the requirements of modern orthopaedic surgery. Nerve injury in this context is the complication most feared by the patient, the anaesthesiologist and the surgeon. Recent findings To date, data dealing with the incidence of nerve injury in regional anaesthesia have almost exclusively been retrieved from close claims analysis. Recently, prospective, well controlled studies have shown that severe neurologic complications rarely occur: for the upper extremity, an incidence of 0.2 1% has been reported. New insights into the mechanisms of local anaesthetic neurotoxicity have demonstrated that ropivacaine has the least potential for neurotoxicity. Administration of the lowest possible concentrated solution of local anaesthetic is likely to be even less neurotoxic. The role of local anaesthetics in the development of apoptosis is nowadays well recognized. The consequences of other factors, such as nerve stretching and compression, in the pathology of nerve damage are emphasized. Summary Significant advances have been made in regional anaesthesia in the past 10 years. The introduction of catheter techniques has cleared the way for better regional anaesthetic and analgesic blocks. Studies dealing with placement of perineural catheters show that the catheter does not increase neurological complications. Properly performed, regional anaesthesia is a safe form of anaesthesia and the benefits far outweigh the risks. Keywords regional anaesthesia, local anaesthetics, nerve stretching and compression, neurologic complications Curr Opin Anaesthesiol 17: # 2004 Lippincott Williams & Wilkins. Department of Anaesthesia, Orthopaedic University Hospital Balgrist, Zurich, Switzerland Correspondence to Prof. Alain Borgeat, Chief of Staff Anaesthesia, Orthopaedic University Hospital Balgrist, Forchstrasse 340, CH-8008 Zurich, Switzerland Tel: ; fax: ; aborgeat@balgrist.unizh.ch Current Opinion in Anaesthesiology 2004, 17: # 2004 Lippincott Williams & Wilkins Introduction There has been renewed interest in regional anaesthesia in the past 10 years due to the availability of new drugs, new materials and results demonstrating some major advantages linked to the administration of central or peripheral blocks. The use of continuous perineural catheters, in particular, has been associated with better pain control [1], higher patient satisfaction [2], dramatic reduction in postoperative nausea and vomiting [3], shorter hospital stay [4], and positive effects on surgical outcome [4]. Each year the number of blocks performed increases, automatically leading to an increased rate of complications. Among these, nerve injury is the adverse event most feared by the patient, the anaesthesiologist, and the surgeon. Incidence of nerve injury after regional anaesthesia There are relatively few published reports on complications associated with the use of peripheral nerve blocks. Until recently, data dealing with the incidence of nerve injury in context with regional anaesthesia were almost exclusively retrieved from the close claims analysis, a structured evaluation of adverse anaesthetic outcomes collected from the closed malpractice insurance claim files. In 1999, Cheney et al. [5] reported that 16% of claims resulted from nerve injury. The most frequent sites of injury were the ulnar nerve (28%), the brachial plexus (20%), lumbosacral nerve roots (16%), and the spinal cord (13%). Ulnar nerve (85%) injuries most likely occurred in association with arm mal-positioning during general anaesthesia, whereas spinal cord (58%) and lumbosacral nerve root (92%) injuries were most likely linked to regional anaesthetic techniques. The same research group performed a follow-up investigation updating their previous results [6]. The authors disclosed that permanent nerve injuries accounted for 19% (n = 193) of regional anaesthesia claims compared with 8% (n = 273) of other surgical anaesthesia claims. Lumbosacral nerve root injuries and paraplegia were the most common injuries observed in the regional anaesthesia group, whereas peripheral nerve and brachial plexus injuries as well as quadriplegia were more often associated with the performance of general anesthesia. The number of claims involving the median, femoral, and sciatic nerves was similar among regional and general anaesthesia techniques. Information resulting from these investigations, however, is blunted by the lack of denominator data. Recent data from prospective studies have shown that the incidence of neurological complications after regional anaesthesia is low. In a 417

2 418 Regional anaesthesia multi-centre trial, Capdevila et al. [7] reported three neuropathies (0.21%) after 1416 continuous peripheral nerve blocks of the upper and lower extremities. Borgeat et al. [8] observed an incidence of severe long-term complications of 0.4% after 510 interscalene blocks (286 single shots and 234 with a perineural catheter). The placement of an interscalene catheter was not associated with an increased incidence of minor or severe neurological complications. These results were confirmed by the same research group in a prospective study which included 700 interscalene catheters [9 ]. Severe neurological complications occurred in 0.2% of cases. Not a single severe neurological complication was observed in a prospective study involving 500 popliteal catheters (263 single shots and 237 with a perineural catheter) [10]. Although nerve injuries are a common concern during the performance of peripheral nerve blocks, postoperative neurological complications may actually be more frequent after general anaesthesia than after peripheral nerve blocks. Indeed, Lynch et al. [11] observed an incidence of severe neurological complications of 4.3% after 417 total shoulder arthroplasties. Of those with a neurological complication, only 17% had an interscalene block. The surgical population in this study was quite similar to the one investigated in the two abovementioned shoulder investigations [8,9 ], in which the incidence was 0.4 and 0.2%, respectively. These first studies show that when peripheral nerve blocks are well performed, the incidence of neurological complications does not increase when compared with a general anaesthetic, emphasizing the importance of proper patient positioning and stretching of the plexus as a causal factor during surgery. Moreover, to date, there is no evidence that the perineural catheter represents an increased risk for neurological complications. Mechanisms of neural damage There are different mechanisms associated with neural damage and their particular role in cases of nerve injury might be difficult to assess. Pathology of peripheral nerve lesions Three different types of peripheral nerve injuries are recognized. (1) Axonal sideration (neurapraxy): this condition is a mild degree of neural insult that results in a decrease in neural conduction velocity. It is reversible. (2) Axonal interruption with sheath conservation (axonotmesis): this condition occurs when only the axon is physically disrupted with preservation of the endoneurial and other supporting connective tissue structures. After a sideration of approximately 6 weeks, the regeneration phenomenon starts at a rate of 1 2 mm/24 h. The recovery shows a more favourable course in young and healthy patients as well as in cases of distal plexus nerve damage. This type of lesion is most frequently associated with compression related to positioning, stretching of the plexus and following administration of toxic drugs or highly concentrated local anaesthetics. (3) Fascicular interruption (neurotmesis): this condition represents the greatest degree of nerve disruption and is a complete disruption of all supporting connective tissue structures. The nerve is completely severed and there is no continuity. The injury carries a poor prognosis of complete recovery. Neurotoxicity of local anaesthetics All local anaesthetics are potentially neurotoxic and neural blockade can be seen as a reversible expression of this. It is known that the administration of local anaesthetics of higher concentration as well as intraneural injections, especially when epinephrine has been added, can result in severe nerve damage [12]. Local anaesthetics are unquestionably neurotoxic, since, when placed directly on nerve fibres, myelinated or nonmyelinated, high concentrations of lidocaine have caused irreversible conduction loss [13,14]. Lambert et al. [13] exposed desheathed bullfrog sciatic nerves for 15 min to 5% or 1.5% lidocaine, with or without 7.5% dextrose, 0.5% tetracaine, and 0.75% bupivacaine (the latter two without dextrose), assessing recovery from conduction blockade. The authors found that exposure to 5% lidocaine, with or without dextrose, or to 0.5% tetracaine resulted in irreversible total conduction blockade, whereas 1.5% lidocaine or 0.75% bupivacaine caused 25 50% residual blockade. Nerves exposed to Ringer s solution, 7.5% dextrose or 0.06% tetracaine had an insignificant residual block. The relationship between lidocaine concentration and nonreversible block after drug washout in frog sciatic nerve was investigated by Bainton and Strichartz [14]. The author was able to demonstrate that a nonreversible block starts at a concentration as low as 40 mm (equivalent to a 1% lidocaine concentration) and increases gradually with increasing concentrations, finally resulting in complete ablation of the compound action potential at 80 mm (*2%). Extended storage of frogs at 48C and the presence of nifedipine make the nerve more resistant to lidocaine. The crucial role of local anaesthetic concentration in peripheral nerve block has been confirmed by Nakamura et al. [15 ]. This research group examined the lidocaine dosage requirements necessary for sciatic nerve block in rats, using a fourfold range of volume. It was found that blocks of greater depth and longer duration result from injection of smaller volumes and, correspondingly, higher lidocaine concentrations containing the same dose.

3 Nerve injury and regional anaesthesia Borgeat and Blumenthal 419 The hypothesis that exposure to local anaesthetics produces morphological changes in growing neurons has been investigated by different groups. Nassogne et al. [16,17] exposed fetal mouse brain cocultures to cocaine. The authors found that exposure to cocaine results selectively in loss of neurites followed by neuronal death. From this study, it became evident that disappearance of neurons was caused by apoptosis (a regulated, programmed cellular self-destruction). Kim et al. [18] investigated the interaction of local anaesthetics with membrane phospholipids which could affect various neuronal as well as nonneuronal cellular activities by exposing a neuroblastoma cell line to dibucaine. Apoptosis was demonstrated by direct visualization of morphological nuclear changes. It was also shown that at dibucaine concentrations which induced apoptosis, the membrane fluidity was significantly altered, indicating that fluidity may be a major target for dibucaine s cytotoxicity. Radwan et al. [19] compared four different local anaesthetics to test the hypothesis that exposure to these drugs produces morphological changes in growing neurons. The authors used isolated dorsal root ganglion neurons from chick embryos to assess quantitatively the growth cone collapse assay. At 20 h after washout, bupivacaine and ropivacaine showed insignificant percentage growth cone collapse, whereas that for lidocaine and mepivacaine was significantly higher than the control values. The possibility that increased cytoplasmic calcium may play a role in the neurotoxicity of local anaesthetics was investigated by Johnson et al. [20]. Neurons derived from dorsal root ganglion were exposed to different concentrations of lidocaine and bupivacaine. In this model it was shown that lidocaine higher than 2.5% increased cytoplasmic calcium to toxic levels within 60 min, whereas bupivacaine and lower concentrations of lidocaine transiently altered calcium homeostasis, but without neurotoxic effect. Increased glutamate concentrations of the cerebrospinal fluid have been reported after intrathecal exposure to tetracaine [21]. Yamashita et al. [22 ] compared the effects of 2% tetracaine, 10% lidocaine, 2% bupivacaine, and 2% ropivacaine on the release of glutamate in the cerebrospinal fluid. Characteristic histopathological changes (vacuolization) were most prominent with lidocaine and tetracaine, followed by bupivacaine, and least with ropivacaine. As a possible alternative short-acting intrathecal agent, prilocaine was tested in a rat model [23]. Intrathecal infusion of 2.5% prilocaine in saline was compared with 2.5% lidocaine in saline. Spinal cord specimens showed the same degree of damage between the two groups. Therefore, it does not seem that substitution of prilocaine for lidocaine is likely to reduce the risk. The comparative neurotoxicity of intrathecal and epidural lidocaine in rats was investigated by Kirihara et al. [24. ]. Different concentrations of lidocaine were given through either the epidural or the intrathecal catheter. The authors found that persistent functional impairment occurred only after intrathecal lidocaine. Histologic damage in the nerve roots and the spinal cord was less severe after epidural than after intrathecal lidocaine injection, confirming the clinical impression that neurological complications are less frequent after epidural anaesthesia than after spinal anaesthesia [25]. Aldrete [26], in a well conducted review, summarized the most recent facts and precautions recommended to reduce the occurrence of neurologic deficits and arachnoiditis following neuraxial anaesthesia. Lastly, Poyraz et al. [27] showed that lidocaine and octanol have different modes of action at tetrodotoxin-resistant Na + channels of peripheral nerves. This observation was explained by different binding sites at the Na + channel. Two clinical consequences emerge from these investigations. First, among the available local anaesthetics, ropivacaine seems to have the least potential for neurotoxicity. Second, the concept of drug volume should be favoured over the concept of drug concentration. This implies that the least concentrated solution of local anaesthetic to achieve an efficient block should be administered. Physical stress induced neurotoxicity Neurological complications are more and more recognized to be potentially secondary to patient positioning or manipulation during surgery [28]. Ikeda et al. [29] demonstrated the vulnerability of the gradually elongated nerve to compression. Rabbit sciatic nerves were gradually elongated to 30 mm at a rate of either 2.0 or 4.0 mm/day. Immediately after elongation was completed the nerve was compressed (comparable in daily practice to surgical manipulation and compression exercised by the retractor) for 30 min at forces from 15 to 60 g/0.1 cm 2. In the group with elongation to 4.0 mm/ day and compressed with the lower force, 15 g/0.1 cm 2, electrophysiologic and histologic examinations were compatible with axonotmesis. The authors concluded that whereas mild compression does not cause nerve injury to the intact nerve, it may cause severe damage to the gradually elongated nerve. Chowet et al. [30 ] investigated whether a commonly performed positioning like wrist extension for intra-arterial catheter placement could lead to nerve injury. Surprisingly, in all awake healthy volunteers the authors found a conduction block involving the median nerve after between 30 and 60 min of positioning. Although the effects were transient, it is recommended that wrists be returned promptly to the neutral position following arterial line placement. Indirect nerve compression may easily lead to neurodegeneration. This has been shown by Damaser et al. [31. ]. The vagina of rats was distended by slow inflation of a

4 420 Regional anaesthesia balloon with water up to 3 ml. Histological examination of the pudendal nerve demonstrated a high number of nerve fascicles with degeneration. These investigations demonstrate that nerves are fragile structures which need to be handled with care. These studies also emphasize the importance of mechanisms such as stretching, traction, compression or a combination of these in producing neurological damage. Role of epinephrine Some anaesthesiologists add epinephrine to local anaesthetics during peripheral block procedures to reduce the drug concentration and to prolong the duration of the block. This practice may generate reduced nerve blood flow. It was demonstrated that epinephrine 1: reduced nerve blood flow by 35% [32]. In a human study it was shown that epinephrine 1: prolonged the action of 1% lidocaine during superficial peroneal nerve block, most likely explained by a decrease in local blood flow [33]. Thus, both drugs are susceptible to decreased neural blood flow. Sinnott et al. [34 ] tried to explain the mechanism by which epinephrine potentiates the action of lidocaine during peripheral nerve block. Sciatic nerve blocks were performed in rats with 0.1 ml of either 0.5% lidocaine or 0.5% lidocaine with epinephrine. Intraneural lidocaine content was regularly assessed. The authors found that adding epinephrine to lidocaine solutions did not increase the amount of intraneural lidocaine at early time points, but later (10 60 min after injection) a fourfold higher content of lidocaine was measured. It was hypothesized that the initial vasoconstriction of epineural compartment vessels delays the rapid removal of lidocaine and allows more drug to enter the deeper perineural compartment early in the block. Epinephrine decreases neural blood flow and at the same time facilitates the penetration of local anaesthetics into the nerve. This association is ambivalent; it may be beneficial in certain conditions, but may also potentiate the neurotoxicity of each local anaesthetic. Conclusion Significant advances have been made in regional anaesthesia in the past 10 years. The introduction of catheter techniques has opened the way for better regional anaesthesia conditions in the operating room and allowed anaesthesiologists to exercise their expertise during the first 3 5 postoperative days on the ward. Properly performed, regional anaesthesia is a safe form of anaesthesia and the benefits far outweigh the risks. The occurrence of nerve injury is still possible and should constantly be kept in mind. However, we now have better needles, catheters, and local anaesthetics, as well as recommendations to perform regional blocks according to the state of the art. The incidence of neurologic damage after regional anaesthesia is low and should become even less frequent in the future. References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as:. of special interest of outstanding interest 1 Borgeat A, Schappi B, Biasca N, Gerber C. Patient-controlled analgesia after major shoulder surgery. Anesthesiology 1997; 87: Borgeat A, Tewes E, Biasca N, Gerber C. Patient-controlled interscalene analgesia with ropivacaine after major shoulder surgery PCIA vs PCA. Br J Anaesth 1998; 81: Borgeat A, Ekatodramis G, Schenker CA. Postoperative nausea and vomiting in regional anesthesia. Anesthesiology 2003; 98: Capdevila X, Barthelet Y, Biboulet P, et al. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999; 91: Cheney FW, Domino KB, Caplan RA, Posner KL. Nerve injury associated with anesthesia. Anesthesiology 1999; 90: Lee LA, Posner KL, Domino KB, et al. Injury associated with regional anesthesia in the 1980s and 1990s: a closed claims analysis. Anesthesiology (in press). 7 Capdevila X, Pirat P, Branchereau S, et al. Continuous peripheral nerve blocks in 1416 patients: a prospective multicenter descriptive study measuring incidences and characteristics of non infectious adverse events [abstract]. Anesthesiology 2002; 96:A Borgeat A, Ekatodramis G, Kalberer F, Benz C. Acute and nonacute complications associated with interscalene block and shoulder surgery. Anesthesiology 2001; 95: Borgeat A, Dullenkopf A, Ekatodramis G, Nagy L. Evaluation of the lateral modified approach for continuous interscalene block after shoulder surgery. Anesthesiology 2003; 99: This prospective study looks at the incidence of neurologic complications after 700 interscalene catheters. This investigation showed that the incidence of severe long-term complications was low (0.2%). 10 Borgeat A, Blumenthal S, Karovic D, et al. Clinical evaluation of a modified posterior anatomical approach to performing the popliteal block. Reg Anesth Pain Med 2004; 29: Lynch NM, Cofield RH, Silbert PL, Hermann RC. Neurologic complications after total shoulder arthoplasty. J Shoulder Elbow Surg 1996; 5: Selander D. Neurotoxicity of local anesthetics: animal data. Reg Anesth 1993; 18: Lambert LA, Lambert DH, Strichartz GR. Irreversible conduction block in isolated nerve by high concentrations of local anesthetics. Anesthesiology 1994; 80: Bainton CR, Strichartz GR. Concentration dependence of lidocaine-induced irreversible conduction loss in frog nerve. Anesthesiology 1994; 81: Nakamura T, Popitz-Bergez F, Birknes J, Strichartz GR. The critical role of concentration for lidocaine block of peripheral nerve in vivo. Anesthesiology 2003; 99: In this study the influence of volume and concentration of lidocaine on functional blockade was investigated. Blocks of greater depth and longer duration result from injection of smaller volumes and, correspondingly, higher lidocaine concentrations containing the same dose. 16 Nassogne MC, Louahed J, Evrard P, Courtoy PJ. Cocaine induces apoptosis in cortical neurons of fetal mice. J Neurochem 1997; 68: Nassogne MC, Evrard P, Courtoy PJ. Selective direct toxicity of cocaine on fetal mouse neurons: teratogenic implications of neurite and apoptotic neuronal loss. Ann N Y Acad Sci 1998; 846: Kim M, Lee YS, Mathews HL, Wurster RD. Induction of apoptotic cell death in a neuroblastoma cell line by dibucaine. Exp Cell Res 1997; 231: Radwan IAM, Saito S, Goto F. The neurotoxicity of local anesthetics on growing neurons: a comparative study of lidocaine, bupivacaine, mepivacaine and ropivacaine. Anesth Analg 2002; 94: Johnson ME, Saenz JA, Da Silva AD, et al. Effect of local anesthetic on neuronal cytoplasmic calcium and plasma membrane lysis (necrosis) in a cell culture model. Anesthesiology 2002; 97: Ohtake K, Matsumoto M, Wakamatsu H, et al. Glutamate release and neuronal injury after intrathecal injection of local anesthetics. Neuroreport 2000; 11:

5 Nerve injury and regional anaesthesia Borgeat and Blumenthal Yamashita A, Matsumoto M, Matsumoto S, et al. A comparison of the neurotoxic effects on the spinal cord of tetracaine, lidocaine, bupivacaine, and ropivacaine administered intrathecally in rabbits. Anesth Analg 2003; 97: The effects of different local anaesthetics on glutamate concentrations in cerebrospinal fluid and histopathological outcome were investigated. Lidocaine as compared with bupivacaine and ropivacaine showed a nonsignificant increase in glutamate concentrations and the worst histopathologic score. 23 Kishimoto T, Bollen AW, Drasner K. Comparative spinal neurotoxicity of prilocaine and lidocaine. Anesthesiology 2002; 97: Kirihara Y, Saito Y, Sakura S, et al. Comparative neurotoxicity of intrathecal and epidural lidocaine in rats. Anesthesiology 2003; 99: This study substantiates the clinical impression that neurologic complications are less frequent after epidural anaesthesia than after spinal anaesthesia. 25 Horlocker TT, Abel MD, Messick J, Schroeder DR. Small risk of serious neurologic complications related to lumbar epidural catheter placement in anesthetized patients. Anesth Analg 2003; 96: Aldrete JA. Neurologic deficits and arachnoiditis following neuroaxial anesthesia. Acta Anesthesiol Scand 2003; 47: Poyraz D, Bräu ME, Wotka F, et al. Lidocaine and octanol have different modes of action at terodotoxin-resistant Na + channels of peripheral nerves. Anesth Analg 2003; 97: Boardman ND, Cofield RH. Neurologic complications of shoulder surgery. Clin Orthop 1999; 368: Ikeda K, Yokoyama M, Tomita K, Tanaka S. Vulnerability of the gradually elongated nerve to compression injury. Hand Surg 2001; 6: Chowet AL, Lopez JR, Brock-Utne JG, Jaffe RA. Wrist hyperextension leads to median nerve conduction block. Anesthesiology 2004; 100: This study shows that wrist hyperextension for arterial line placement and stabilization results in profound impairment of median nerve function. Although the effects are transient, it is recommended that wrists be returned promptly to the neutral position following arterial line placement. 31. Damaser MS, Broxton-Kim C, Ferguson C, et al. Functional and neuroanatomical effects of vaginal distension and pudendal nerve crush in the female rat. J Urology 2003; 170: This study shows that the pudendal nerve is vulnerable to injury during vaginal distension. 32 Partridge BL. The effects of local anesthetics and epinephrine on rat sciatic nerve blood flow. Anesthesiology 1991; 75: Bernards CM, Kopacz DJ. Effect of epinephrine of lidocaine clearance in vivo: a microdialysis study in humans. Anesthesiology 1999; 91: Sinnott CJ, Cogswell LP, Johnson A, Strichartz GR. On the mechanism by which epinephrine potentiates lidocaine s peripheral nerve block. Anesthesiology 2003; 98: In this study it was demonstrated that adding epinephrine to lidocaine does not increase the amount of intraneural lidocaine at early time points, but later (10 60 min after injection) a fourfold higher content of lidocaine was measured.