Spinal anesthesia with lidocaine is ideal for ambulatory surgery because of its short duration of action. However, transient neurologic symptoms (TNS) occur in 0-40% of patients. The incidence of TNS with mepivacaine, which has a similar duration of action, is unknown.


Sixty ambulatory patients undergoing knee arthroscopy received spinal anesthesia in a randomized, double-blinded manner, with either 45 mg 1.5% mepivacaine or 60 mg 2% lidocaine. An L3-L4 midline approach was used with a 27-gauge Whitacre needle and a 20-gauge introducer. The local anesthetic was injected over approximately 30 s with the aperture of the Whitacre needle in a cephalad direction. Two to 4 days after operation, each patient was questioned about the development of TNS. In addition, the two groups were compared for time to regression of sensory and motor blockade and time to discharge milestones.


Three patients receiving lidocaine were lost to follow-up. None of the 30 patients in the mepivacaine group developed TNS, whereas 6 of 27 (22%) in the lidocaine group did (P = 0.008). Time to regression to the L5 sensory level and to complete resolution of motor block were similar in both groups. The times to discharge milestones were also comparable.


The incidence of TNS is greater with 2% lidocaine than with 1.5% mepivacaine for patients having unilateral knee arthroscopy under spinal anesthesia. Mepivacaine seems to be a promising alternative to lidocaine for outpatient surgical procedures because of its similar duration of action. Further studies are warranted to determine the optimal dose of intrathecal mepivacaine for ambulatory surgery and the incidence of TNS with other doses and concentrations of intrathecal mepivacaine.

SEVERE or permanent neurologic injury stemming from local anesthetic toxicity during spinal anesthesia is a rare occurrence. [1,2]However, transient neurologic symptoms (TNS) may be a clinical sign of mild, temporary neurologic dysfunction associated with the local anesthetic, lidocaine. Several studies have investigated the cause of TNS, including its association with lidocaine concentration, [3]osmolarity, [4]and additives such as epinephrine [5]and dextrose concentration. [4]Hampl et al. [6]compared the incidence of TNS after either lidocaine or bupivacaine spinal anesthesia and found a statistically significant higher incidence with 5% lidocaine; however, because of bupivacaine's long duration of action, it is not an ideal alternative to lidocaine, especially for short procedures and ambulatory patients.

Mepivacaine, an amide local anesthetic with physicochemical properties similar to those of lidocaine, has been used for spinal anesthesia since 1956. [7]El-Shirbiny et al. [8]reported experiences with 20,000 mepivacaine spinal anesthetics and concluded that this local anesthetic “could be considered one of the best for spinal anesthetic blocks.” Despite this and other studies, the popularity of lidocaine forced mepivacaine into obscurity.

Given the recent reports of TNS associated with varying concentrations of lidocaine in spinal anesthesia, we decided to re-evaluate mepivacaine as a spinal anesthetic, specifically with regard to the incidence of TNS, in a randomized, double-blind comparison with isobaric 2% lidocaine in ambulatory patients undergoing unilateral knee arthroscopy. We tested the hypothesis that spinal anesthesia with 1.5% mepivacaine would result in a lower incidence of TNS in the ambulatory surgical population than spinal anesthesia with 2% lidocaine.

After obtaining institutional review board approval and informed, written consent, 60 patients classified as American Society of Anesthesiologists physical status I-III who were undergoing unilateral knee arthroscopy were studied. Power analysis using a reported incidence of TNS at 16%[5]for lidocaine spinal blocks revealed that 75 patients in each group were required to detect a 50% reduction in TNS with 95% confidence limits. However, an interim analysis was performed after 60 patients because the institutional review board was concerned that the incidence of TNS at our institution was unknown. The study was terminated after the interim analysis because of the observed difference.

Patients in whom spinal anesthesia was contraindicated or who had a known hypersensitivity to amide local anesthetics were excluded. The patients were assigned using a random-number table to receive either 2% 3 ml (60 mg) lidocaine (group L; lidocaine HCl; Abbott Laboratories, N. Chicago, IL), or 1.5% 3 ml (45 mg) mepivacaine (group M; polocaine, Astra Pharmaceuticals, Westborough, MA). These doses were chosen based on clinical experience at our institution suggesting comparable duration and extent of sensory blockade.

After a peripheral intravenous infusion was begun, the patients each received 1 to 5 mg midazolam (Versed; Roche, Manati, Puerto Rico) intravenously for anxiolysis. Patients received oxygen via nasal cannulae at a rate of 2–4 l/min and were monitored with continuous pulse oximetry, a noninvasive blood pressure device, and electrocardiography. They were placed in the right or left lateral decubitus position at the discretion of the anesthesiologist. The skin and subcutaneous tissue of the L3-L4 interspace were anesthetized with 1–3 ml of the same local anesthetic used for the spinal. A 20-gauge, 1.25-inch introducer (Becton-Dickinson, Franklin Lakes, NJ) was inserted, followed by a 27-gauge, 3.5-inch Whitacre needle (Becton-Dickinson) until free-flowing cerebrospinal fluid could be visualized. Group L then received 3 ml 2% lidocaine and group M received 3 ml of 1.5% mepivacaine, injected over approximately 30 s with the aperture of the Whitacre needle facing cephalad. The number of attempts and degree of difficulty of the block (1 = easy, 2 = moderate, 3 = difficult) were recorded. After injection, the patients were immediately turned to lay in the supine position. An investigator blinded to the local anesthetic assessed sensory level to pinprick and motor response using a modified Bromage scale (0 = full movement, 1 = movement of knees only, 2 = movement of ankles only, and 3 = no movement) at regular intervals until resolution of the block.

The incidence of hypotension (systolic blood pressure < 90 mmHg) and bradycardia (heart rate < 50 beats/min), as well as the use of ephedrine and atropine, both during and after operation, were recorded. Intraoperative sedation, when administered, consisted of further doses of midazolam or propofol (Zeneca, Wilmington, DE). No narcotics were used during operation.

On arrival in the postanesthesia care unit, data were collected by a nurse who was blinded to the patient anesthetic group, on the times to ingest food, void, sit in a chair, and be discharged from the hospital.

Patients were contacted 2 days after operation and asked a standard series of questions (see appendix 1) by one of the investigators who was blinded to the group assignment. Transient neurologic symptoms were defined as back pain or dysesthesia that radiated to the buttocks, thighs, hips, or calves and began within the first 24 h after surgery. Localized pain or tenderness at the injection site or lower back without radiation was not considered TNS. This definition was essentially the same used by Pollock et al. [5] 

Statistical Analysis

Groups were compared using the Fisher exact test to compare the incidence of TNS, and analysis of variance to compare time for the block to regress to the L5 level, time to achieve a Bromage score of 0, and times to achieve discharge milestones. Alpha was set at 0.025 for the interim analysis.

Of the 60 patients enrolled in the study, three in the lidocaine group were lost to follow-up. Because of clinical responsibilities, block regression data were collected in 22 lidocaine and 19 mepivacaine patients. There were no differences between groups with regard to patient age, sex, height, weight, or American Society of Anesthesiologists physical status (see Table 1). In addition, there were no differences with regard to the number of attempts, degree of difficulty of the spinal anesthetic, duration of the operative procedure, or amount of sedation given. No paresthesias were elicited in any patient. All of the blocks in both groups provided a sufficient sensory level within 10 min and adequate surgical anesthesia. There was no difference in the incidence of perioperative bradycardia or hypotension between groups (three in group M, one in group L). None of the patients experienced a postdural puncture headache.

None of the 30 patients in the mepivacaine group developed TNS, whereas 6 of 27 (22.2%) patients in the lidocaine group did (P = 0.008;Table 2). Data were also analyzed by assuming that TNS did not develop in the three lidocaine patients who were lost to follow-up. This more conservative statistical analysis still resulted in a significant difference between groups (P < 0.025).

All patients with TNS were followed by telephone interview until symptoms resolved. Symptoms were treated with nonsteroidal anti-inflammatory agents and resolved within 1–5 days. Within the lidocaine group, patients in whom TNS developed were older than those without TNS (49 +/- 6 yr [range, 38–55 yr] vs. 35 +/- 12 yr [range, 18–59 yr]; P < 0.01).

The time for the sensory block to regress to the L5 level and the time for complete resolution of motor block (Bromage score = 0) were similar in both groups (Table 3). The times required to achieve discharge milestones (sit in a chair, tolerate food, and void) and the time to discharge and the total postanesthesia care unit time did not differ between groups (Table 4).

This study shows that TNSs are significantly more common after spinal anesthesia with 60 mg 2% lidocaine than with 45 mg 1.5% mepivacaine in ambulatory surgery patients undergoing unilateral knee arthroscopy (22.2% vs. 0%; P = 0.008). Further, the comparable time of onset and duration of action make mepivacaine a suitable alternative to lidocaine for ambulatory surgical procedures. At our institution, mepivacaine has become a commonly used alternative to lidocaine for spinal anesthesia in the ambulatory setting.

Transient neurologic symptoms after spinal anesthesia with hyperbaric 5% lidocaine were first described in 1993 by Schneider et al., [9]and was thought to be a mild manifestation of previously reported serious neurotoxic complications after administration of large doses of intrathecal lidocaine. [10]Subsequent studies have followed, eliminating osmolarity and dextrose concentration of the local anesthetic solution [4]and lidocaine concentration [3,5]as contributing factors.

Other studies have suggested that flexion of the hips and knees during lithotomy positioning [9]and knee arthroscopy [5]may contribute to the development of TNS. Schneider et al. [9]proposed that patient positioning may be a contributory factor in the development of TNS because all four patients in their initial report were in the lithotomy position. The lithotomy position may cause stretching of the sacral nerves, predisposing them to the described effects of local anesthetics. Pollack et al. [5]found a significantly higher incidence of TNS after lidocaine spinal anesthesia in patients undergoing knee arthroscopy compared with those undergoing hernia repair. During knee arthroscopy, the operative hip and knee are flexed during portions of the procedure, perhaps mimicking the effects of the lithotomy position on sacral nerves. Hogan [11]postulated that the small size of the S3 through S5 nerve roots may predispose these roots to the neurotoxic effects of local anesthetics; stretching may affect smaller nerve roots to a greater extent. The present study was not designed to evaluate patient positioning in the development of TNS; however, TNS did not develop in the patients who received mepivacaine despite the same positioning.

Within the lidocaine group, the mean age of the patients in whom TNS developed was significantly greater than that of those who received lidocaine but in whom TNS did not develop (49 yr compared with 35 yr). Three original case reports [9,12,13]that described TNS included eight patients whose mean age was 48 yr, with a range of 36–63 yr. Our data suggest that younger patients may not be as prone to the development of TNS. One potential mechanism may be that the nerve fibers of younger patients are better able to tolerate the effects of stretching compared with those of older patients, and this protects them from developing TNS.

Current data indicate that the incidence of TNS after spinal anesthesia with 2% or 5% lidocaine is greater compared with bupivacaine, [4,5]as it is now with 1.5% mepivacaine. Mepivacaine, an amide local anesthetic used widely for brachial plexus and peripheral nerve anesthesia, has not been popular for spinal anesthesia. The reasons for this are unclear. Despite many reports of successful application of mepivacaine for spinal anesthesia in the 1960s, [8,14–17]its recent use has been overshadowed by lidocaine and bupivacaine. Although the incidence of TNS after spinal anesthesia with bupivacaine is much lower than that after spinal anesthesia with lidocaine, the clinical duration of action of bupivacaine precludes its use as a routine spinal anesthetic in the ambulatory setting. Ben-David et al. [18]explored the possibility of using lower doses of bupivacaine to provide surgical anesthesia in ambulatory patients. The results appeared to be unsatisfactory because more than one third of the patients receiving low doses of bupivacaine felt mild discomfort or pain during surgery.

The financial implications of a timely discharge from the ambulatory postanesthesia care unit can be significant. If spinal anesthesia is to remain a viable alternative for outpatient surgery, drugs or techniques that allow a rapid discharge, with few side effects, must be used. Further studies on the dose-response relations for mepivacaine, specifically with regard to the attainment of specific discharge criteria, are warranted.

The purpose of this study was to investigate the incidence of TNS after mepivacaine spinal anesthesia with the hope of finding an acceptable alternative to lidocaine for short ambulatory surgery procedures. Although the results are striking, there are several limitations to the study. First, the sample size (n = 57) is relatively small for a clinical study. It should not be concluded that TNS occurs only with lidocaine. We are aware of at least three cases of TNS after spinal anesthesia with mepivacaine at our institution. Two occurred after the completion of this study, and the other occurred in a nonambulatory patient with four previous episodes of TNS after spinal anesthesia with lidocaine. Power analysis required 75 patients in each group to detect a 50% reduction in the incidence of TNS. However, because of concerns regarding the incidence of TNS with mepivacaine, we decided to do an interim analysis after 60 patients. Given the significant nature of the results, the study was terminated.

A second limitation is that two different doses and concentrations of local anesthetics were used. This was done based on clinical experience with each drug. Sixty milligrams of 2% lidocaine was our standard for spinal anesthesia for unilateral knee arthroscopy. Forty-five milligrams of 1.5% mepivacaine seemed to provide anesthesia of similar duration. The possibility remains that TNS may be equally frequent with higher concentrations or larger doses of mepivacaine.

In conclusion, the incidence of TNS is greater with 2% lidocaine than with 1.5% mepivacaine for patients having unilateral knee arthroscopy under spinal anesthesia. Further, mepivacaine is a promising alternative to lidocaine for outpatient surgical procedures. Further studies are warranted to determine the dose-response characteristics of intrathecal mepivacaine and the effect of dose and concentration on the incidence of TNS.

Questionnaire Used to Evaluate Transient Neurologic Symptoms after Operation

1. If 0 is no pain and 10 the worst pain imaginable, how would you rate your pain after surgery? Knee Pain? Did patient volunteer back pain? YES (if yes, go to 4) or NO?

2. If patient doesn't volunteer that he/she has back pain: Did you have discomfort anywhere other than the surgical site? YES or NO? If Yes, where?

3. Did you have back pain after surgery? YES or NO (If no, go to 9) If Yes,

4. If 0 is no pain and 10 the worst pain imaginable, how would you rate your back pain?

5. Where was the pain in your back?

6. Did the pain radiate anywhere? YES or NO. Hips R L Both; Buttocks R L Both; Thighs R L Both; Calves R L Both; Sacrum R L Both

7. How long did the pain last?

8. Did the pain prevent you from sleeping, sitting, lying down, walking, going to work? Circle.

9. Had you ever had back problems or similar sensations before this surgery? YES or NO?

Phillips O, Ebner H, Nelson A, Black M: Neurologic complications following spinal anesthesia with lidocaine: A prospective review of 10,440 cases. Anesthesiology 1969; 30:284-9.
Matsushige D, Horlocker T, McGregor D, Schroeder D: Neurologic complications of spinal anesthesia. Reg Anesth 1996; 3A:A730.
Hampl K, Schneider M, Pargger H, Gut J, Drewe J, Drasner K: A similar incidence of transient neurologic symptoms after spinal anesthesia with 2% and 5% lidocaine. Anesth Analg 1996; 83:1051-4.
Hampl K, Schneider M, Thorin D, Ummenhofer W, Drewe J: Hyperosmolarity does not contribute to transient neurologic symptoms after spinal anesthesia with hyperbaric 5% lidocaine. Reg Anesth 1995; 20:363-8.
Pollack J, Neal J, Stephenson C, Wiley C: Prospective study of the incidence of transient neurologic symptoms in patients undergoing spinal anesthesia. Anesthesiology 1996; 84:1361-7.
Hampl K, Schneider M, Ummenhofer W, Drewe J: Transient neurologic symptoms after spinal anesthesia. Anesth Analg 1995; 81:1148-53.
Ekenstam B, Egner B, Ulfendahl H, Dhuner K, Oljelund O: Trials with Carbocaine. Br J Anesth 1956; 28:503.
El-Shirbiny A, Rasheed M, Elmaghraby A, Motaweh M: Experiences with Carbocaine in spinal anesthesia. Acta Anesth Scand 1966; 23:442-8.
Schneider M, Ettlin T, Kaufmann M, Schumacher P, Urwyler A, Hampl K, Hochstetter A: Transient neurologic toxicity after hyperbaric subarachnoid anesthesia with 5% lidocaine. Anesth Analg 1993; 76:1154-7.
Rigler M, Drasner K, Krejcie T, Yelich S, Scholnick F, DeFontes J, Bohner D: Cauda equina syndrome after continuous spinal anesthesia. Anesth Analg 1991; 72:275-81.
Hogan Q: Size of human lower thoracic and lumbosacral nerve roots. Anesthesiology 1996; 85:37-42.
Sjostrom S, Blass J: Severe pain in both legs after spinal anesthesia with hyperbaric 5% lignocaine solution. Anaesthesia 1994; 49:700-2.
Pinczower G, Chadwick H, Woodland R, Lowmiller M: Bilateral leg pain following lidocaine spinal anesthesia. Can J Anesth 1995; 42:217-20.
Dunn R, Gee H, Carnes M, Fabian L: Spinal anesthesia with mepivacaine hydrochloride. Anesth Analg 1963; 42:49-54.
Lipton E, Sennott F, Batt B: Mepivacaine for spinal anesthesia in vaginal delivery. Am J Obstet Gynecol 1966; 96:333-6.
Siker E, Wolfson B, Stewart W, Pavilack P, Pappas A: Mepivacaine for spinal anesthesia: Effect of changes in concentration and baricity. Anesth Analg 1966; 45:191-6.
Henschel E, Remus C, Mustafa K, Jacoby J: Isobaric mepivacaine in spinal anesthesia. Anesth Analg 1967; 46:475-9.
Ben-David B, Levin H, Solomon E, Admoni H, Vaida S: Spinal anesthesia in ambulatory surgery: The effect of saline dilution. Anesth Analg 1996; 83:716-20.