Spinal anesthesia can cause transient but distressing postoperative back pain that involves the buttocks or legs, termed transient neurologic symptoms (TNS) (1). Lidocaine is more likely to cause TNS than bupivacaine or tetracaine (2), but the TNS rate after spinal anesthesia with mepivacaine is less well established (3). A prospective, randomized, blinded comparison of lidocaine versus mepivacaine spinal anesthesia for knee arthroscopy in ambulatory patients demonstrated a 22% rate of TNS after lidocaine, but no TNS was observed among 30 patients given isobaric 1.5% mepivacaine (P = 0.008) (4). Subsequent trials of isobaric mepivacaine confirmed small rates of TNS, ranging from 0% to 7.5% (5–7). However, 30%–37% of patients given spinal anesthesia with hyperbaric 4% mepivacaine developed TNS (8,9). The largest study of TNS after mepivacaine spinal anesthesia included 100 patients (9). There are no large series studies of mepivacaine for spinal anesthesia published within the last 30 yr. Studies of mepivacaine spinal anesthetics published in the 1960s, including one with 20,000 patients (10), did not address TNS. Review articles judged that mepivacaine had the same risk of inducing TNS as lidocaine (3) and called for additional evaluation of mepivacaine for spinal anesthesia (11).
At the Hospital for Special Surgery, 2% standard dose spinal lidocaine has been virtually abandoned in favor of mepivacaine. In this prospective cohort study of 1210 patients, we tested the hypothesis that spinal (or combined spinal epidural [CSE]) anesthesia with isobaric 1.5% mepivacaine has a small rate of TNS (defined as similar to the 0%–7.5% rate reported for 1.5% mepivacaine, and dissimilar to the 30%–37% rate reported for 4% hyperbaric mepivacaine). The primary outcome variable was the rate of TNS.
All patients who received a plain 1.5% mepivacaine spinal (or CSE) anesthetic for ambulatory surgery at the Hospital for Special Surgery Ambulatory Surgery Center were eligible for entry in this prospective cohort study. Glucose was not added to the mepivacaine. Data were collected prospectively by a trained research assistant. The study was approved by the hospital IRB before patient enrollment. Consent was not obtained, as per our IRB standard, because patient care was not changed by this cohort study. It is institutional practice for patients to be contacted by telephone to evaluate their recovery from anesthesia and surgery, including questioning about headache and backache. Patients were enrolled from 11/05/01 to 8/23/02.
Anesthetic and demographic data were recorded by the anesthesiologist. Anesthetic management (choice of spinal versus CSE, type and extent of sedation, fluid and hemodynamic management) was determined by the anesthesiologist. An effective anesthetic was defined as one sufficient for surgery without general anesthesia. A safe anesthetic was defined as one that did not cause a long-term complication such as permanent neurologic deficit. Patients undergoing reconstruction of the anterior cruciate ligament (ACL) were typically (316 of 329) placed in Trendelenburg position with the knees maximally flexed, at the surgeons' request. Patients undergoing knee arthroscopy were often (613 of 781) placed with the knee flexed and the hip variably flexed, neutral or extended.
Postoperative analgesia was determined by the surgical team, and may or may not have included nonsteroidal antiinflammatory drugs (NSAIDs).
Attempts were made to contact patients twice, ideally at days 1–4 and again at days 6–9. If the first successful contact was made on or after day 7, no further contacts were made. A single coinvestigator used a standardized questionnaire to determine presence of symptoms of backache and headache (see Appendix). The investigator who called the patients was not blinded to the anesthetic management. Patients who reported symptoms were questioned about possible radiation of backache and possible positional nature of headache. TNS was defined as new onset of back pain that radiated bilaterally to buttocks or distally (12). Back pain that did not meet these criteria was not viewed as TNS. Specifically, exacerbation of preexisting back pain, point tenderness, back pain that did not radiate, unilateral radiating pain, or new pain in extremities not linked to back pain, were not termed TNS. Postdural puncture headache (PDPH) was defined as new onset of headache with postural features (13). Patients were asked about their recovery from the anesthetic, including any complications. As is customary at this hospital, patients with any complaints or apparent complications related to their anesthetic (including but not limited to TNS and PDPH) were referred to the attending anesthesiologist who performed or supervised their anesthetic. Treatment, if any, of the perianesthetic complications was determined by the attending anesthesiologist. Typical treatment of TNS consisted of analgesics including NSAIDs unless contraindicated. Patients were followed until resolution of their symptoms.
Initial statistical analysis was performed with Statview for Windows, version 5.0 (Cary, NC). Statistical consultation was obtained and additional data analyses were performed using SPSS version 10 from SPSS Inc. (Chicago, IL). The associations of the main variables of interest (age, mepivacaine dose, procedure, position, diagnosis) with the outcome variable (TNS) were analyzed using t-tests, nonparametric tests, and contingency table analysis as appropriate. Fisher's exact test was always used when numbers analyzed were <5 (StatXact version 6; Cytel Software Corp., Cambridge, MA). The α level was set at 0.05 for these tests. Variables for which the association with the outcome was statistically significant were then entered into the backwards stepwise logistic regression, with a cutoff of 0.06. The dependent variable was TNS. Independent variables were age, mepivacaine dose, procedure (ACL or not ACL), position, and diagnosis.
Enrollment consisted of 1273 patients; data are presented for the 1210 patients successfully contacted postoperatively (95% follow-up rate). All patients enrolled were tracked. Of patients contacted at least once, 832 of 1210 contacts were made on days 1–4. Of patients contacted twice, 873 of 1027 calls were made on days 6–9.
Spinal anesthesia was administered to 838 patients (69%). CSE anesthesia was administered to 372 patients (31%). Table 1 describes characteristics of patients successfully contacted. The most frequently administered dose of mepivacaine was 45 mg (3 mL), administered to 582 patients. The mean dose of mepivacaine was 50 ± 8 (range, 30–70) mg. For patients given a CSE, the median mepivacaine dose was 52.5 mg (25%ile 45 mg, 75%ile 60 mg). For patients given a spinal alone (not a CSE), the median mepivacaine dose was 45 mg (25%ile 45 mg, 75%ile 52.5 mg). Subarachnoid fentanyl was given to 46 of 1210 patients. Epinephrine was not added to mepivacaine for intrathecal use. Of 372 epidural catheters placed, 162 were used during the operation. Epidural fentanyl alone was given to 13 patients. Epidural fentanyl + lidocaine was given to 33 patients. Epidural lidocaine was given to 123 patients (including those given fentanyl + lidocaine), mepivacaine to 12 patients, bupivacaine to 10 patients, and mixtures of the preceding local anesthetics were given to 4 patients.
After isobaric mepivacaine spinal anesthesia, 78 of 1210 (6.4%) patients developed TNS (95% confidence intervals 5.1%–8%). An additional 22 patients (1.8%) had new onset of unilateral pain (back pain radiating to only one side). Inadequacy for surgery of the spinal anesthetic was noted in 1.7% of the patients. None of the CSE anesthetics were inadequate for surgery. None of the patients had permanent neurologic sequelae.
Characteristics of patients with TNS are given in Table 2, with subgroup analysis in Table 3. Prone and supine positions were combined for Table 3. TNS rates separately were 0 of 24 (prone) and 19 of 223 (supine) (not significant). The rate of TNS among patients undergoing ACL reconstruction (6 of 329) was less than the rate observed among patients not undergoing ACL reconstruction (72 of 881) (P < 0.0001). The rate of TNS was not altered by perioperative use of NSAIDs (5.8% TNS if NSAIDs were used, 7.4% TNS if NSAIDs were not used, not significant by contingency table, Fisher's exact test). The total rate of any postoperative back pain was 33%.
Median visual analog scale (VAS) score for patients with TNS, when at its worst, was 7 (25%ile, 5; 75%ile, 8; range, 2–10). Median duration was 3 days (25%ile, 1 day; 75%ile, 5 days; range, 1–15). The most distal extension of TNS-related pain was the buttocks (24%), hips (24%), thighs (28%), or calves (23%).
Univariate analysis indicated that TNS was significantly associated with age (P < 0.001), mepivacaine dose (P = 0.002), and procedure (patients undergoing ACL reconstruction were less likely to develop TNS) (P < 0.001). Multivariate analysis, logistic regression indicated that significant predictors were age (P = 0.003) and procedure (ACL or not ACL) (P < 0.001).
PDPH developed in 1.2% of the patients. The rate of PDPH was not affected by age, body mass index, or anesthetic technique (spinal versus CSE) (evaluated by unpaired t-test). Gender, approach (midline versus paramedian), or number of passes also did not influence the rate of PDPH (evaluated by contingency table, Fisher's exact test). Median VAS score for patients with PDPH was 8.5, when at its worst (25%ile, 7; 75%ile, 10). Median duration was 6.5 days (25%ile, 3 days; 75%ile, 8 days).
This cohort study of 1210 ambulatory patients was performed because mepivacaine spinal anesthesia remains controversial, as evidenced by statements that there are no reliable alternatives to lidocaine for spinal anesthesia (14). Intrathecal mepivacaine was effective as a spinal anesthetic for the surgical procedures studied. None of the CSEs, and only 1.7% of the spinal anesthetics, were inadequate for surgery. No permanent neurologic sequelae occurred after mepivacaine spinal anesthesia. The upper limit of the 95% confidence interval for long-term adverse outcomes after spinal anesthesia with 1.5% mepivacaine is 25/10,000 (15), similar to the 17/10,000 rate after spinal anesthesia with other local anesthetics (16).
The incidence of TNS (6.4% overall) was not influenced by intraoperative position, weight, gender, approach (midline versus paramedian), or perioperative use of NSAIDs. All of our patients were outpatients, and none was placed in lithotomy position. Multivariant analysis indicated that age, but not mepivacaine dose predicted TNS. It is not clear why patients undergoing ACL reconstruction were less likely to develop TNS.
Reported risk factors for development of TNS include lithotomy position, outpatient status, and obesity (2). Lithotomy and knee arthroscopy positions may have a more frequent incidence of TNS compared with supine position (1). Rates of TNS after lidocaine spinal anesthesia for operations in the supine position include no TNS in obstetrical patients (17) and 5% TNS for hernia repair (18). However, a 26% rate of drug-specific TNS can occur in the supine position (19). It is not clear why this investigation did not demonstrate an effect of intraoperative position on TNS rate. Other investigators did not find weight or outpatient status to be predictive of TNS (12). TNS was more common in older patients after mepivacaine spinal anesthesia in a prospective randomized trial (4), but an epidemiologic investigation, which did not include subarachnoid mepivacaine, did not find age to be linked to TNS (2).
It has been reported that TNS developed in 0%–7.4% of ambulatory patients undergoing knee arthroscopy after spinal anesthesia with 1.5% mepivacaine (7). Two trials report frequent rates of TNS (30%–37%) after spinal anesthesia with hyperbaric 4% mepivacaine (8,9). One can speculate that these disparate rates of TNS are the result of different effects of 1.5% isobaric versus 4% hyperbaric mepivacaine.
TNS rates depend on the local anesthetic used for spinal anesthesia (1–3). Chloroprocaine has recently been advocated as a TNS-free spinal anesthetic (20). Bupivacaine has a 0%–1% rate of TNS (3). Procaine was associated with a less frequent rate of TNS than lidocaine (6% versus 31%) (21). TNS developed in 33% of patients given 50 mg of lidocaine for knee arthroscopy, but TNS developed in only 3.6% of patients given 20 mg of lidocaine + 25 μg of fentanyl (22).
TNS was defined as new onset of back pain that radiated bilaterally to buttocks or distally (4,7,12). Because TNS can develop on the second postoperative day (9), the study definition did not require symptoms to begin on the first postoperative day. Other definitions are more inclusive [pain or dysesthesia in legs or buttocks with or without back pain (2)], less inclusive [radiation down the lower leg (23)], or mixed [e.g., broader definition but shorter time frame (18)]. TNS occurs after general anesthesia (24), raising concerns that use of a broad definition may lessen the specificity of the diagnosis. A definition of TNS that requires bilateral symptoms may not exclude many patients. In a trial that did not require bilateral symptoms, all patients with TNS had bilateral symptoms (19).
A major limitation of this work is that this was a cohort study, i.e., a prospective observational investigation without a comparison group. Cohort studies are useful to determine complication rates after a procedure (25). Our conclusions would have been stronger with a prospective randomized trial. A prospective randomized trial used a similar patient population and TNS definition to demonstrate a 22% rate of TNS after intrathecal lidocaine and 0% TNS after intrathecal mepivacaine (4).
Subarachnoid administration of 1.5% mepivacaine (as well as 2% lidocaine, 0.5% bupivacaine, chloroprocaine, or fentanyl) is “off-label” use in the United States. Some preparations of 1.5% mepivacaine are labeled “not for spinal anesthesia.” Food and Drug Administration (FDA) approval of mepivacaine as a spinal anesthetic would require studies that have never been done for this practice. The absence of FDA approval does not imply lack of safety or deviation from medicolegal standard of care.
Future research should include prospective comparisons of mepivacaine to chloroprocaine, small-dose lidocaine + fentanyl, or small-dose bupivacaine + fentanyl. Direct comparison of 1.5% isobaric to 4% hyperbaric mepivacaine is also needed, but 4% hyperbaric mepivacaine is not available in the United States. A larger series of mepivacaine spinal anesthetics would determine the rate of rare complications with narrower confidence intervals.
This prospective cohort study of mepivacaine spinal anesthesia in 1210 patients documented frequent success and a small rate of TNS. Intrathecal mepivacaine is likely to be a safe and effective alternative for spinal anesthesia in the ambulatory setting.
The authors thank Jennifer Gordon and Jane Lipnitsky for their help as research assistants, and also Margaret G. E. Peterson, PhD, for statistical advice.
Transient Neurologic Symptoms Questioins
1. Did you have back pain after surgery? Yes No
2. When did the pain start?
3. If 0 is no pain and 10 is the worst imaginable pain, how would you rate your pain at its worst?
4. Did the pain radiate anywhere? Hips R L BButtocks R L BThighs R L B Calves R L B
5. How long did the pain last? ______ days after onset.
6. Did the pain prevent you from carrying on normal activities? Yes No
7. Did anything make the pain better? Medications? Other?
1. Pollock JE. Transient neurologic symptoms: etiology, risk factors, and management. Reg Anesth Pain Med 2002;27:581–6.
2. Freedman JM, Li DK, Drasner K, et al. Transient neurologic symptoms after spinal anesthesia: an epidemiologic study of 1,863 patients. Anesthesiology 1998;89:633–41.
3. Liu SS, McDonald SB. Current issues in spinal anesthesia. Anesthesiology 2001;94:888–906.
4. Liguori GA, Zayas VM, Chisholm MF. Transient neurologic symptoms after spinal anesthesia with mepivacaine and lidocaine. Anesthesiology 1998;88:619–23.
5. Pawlowski J, Sukhani R, Pappas AL, et al. The anesthetic and recovery profile of two doses (60 and 80 mg) of plain mepivacaine for ambulatory spinal anesthesia. Anesth Analg 2000;91:580–4.
6. Salazar F, Bogdanovich A, Adalia R, et al. Transient neurologic symptoms after spinal anaesthesia using isobaric 2% mepivacaine and isobaric 2% lidocaine. Acta Anaesthesiol Scand 2001;45:240–5.
7. Zayas VM, Liguori GA, Chisholm MF, et al. Dose response relationships for isobaric spinal mepivacaine using the combined spinal epidural technique. Anesth Analg 1999;89:1167–71.
8. Salmela L, Aromaa U. Transient radicular irritation after spinal anesthesia induced with hyperbaric solutions of cerebrospinal fluid-diluted lidocaine 50 mg/ml or mepivacaine 40 mg/ml or bupivacaine 5 mg/ml. Acta Anaesthesiol Scand 1998;42:765–9.
9. Hiller A, Rosenberg PH. Transient neurological symptoms after spinal anaesthesia with 4% mepivacaine and 0.5% bupivacaine. Br J Anaesth 1997;79:301–5.
10. el-Shirbiny AM, Rasheed MH, Elmaghraby A, Motaweh M. Experiences with Carbocaine in spinal anaesthesia. Report of 20,000 cases. Acta Anaesthesiol Scand Suppl 1966;23:442–8.
11. Gaiser RR. Should intrathecal lidocaine be used in the 21st century? J Clin Anesth 2000;12:476–81.
12. Silvanto M, Tarkkila P, Makela ML, Rosenberg PH. The influence of ambulation time on the incidence of transient neurologic symptoms after lidocaine spinal anesthesia. Anesth Analg 2004;98:642–6.
13. Turnbull DK, Shepherd DB. Post-dural puncture headache: pathogenesis, prevention and treatment. Br J Anaesth 2003;91:718–29.
14. Rowlingson JC. To avoid “transient neurologic symptoms”: the search continues. Reg Anesth Pain Med 2000;25:215–7.
15. Ho AM, Dion PW, Karmakar MK, Lee A. Estimating with confidence the risk of rare adverse events, including those with observed rates of zero. Reg Anesth Pain Med 2002;27:207–10.
16. Horlocker TT, McGregor DG, Matsushige DK, et al. A retrospective review of 4767 consecutive spinal anesthetics: central nervous system complications. Perioperative Outcomes Group. Anesth Analg 1997;84:578–84.
17. Aouad MT, Siddik SS, Jalbout MI, Baraka AS. Does pregnancy protect against intrathecal lidocaine-induced transient neurologic symptoms? Anesth Analg 2001;92:401–4.
18. Pollock JE, Neal JM, Stephenson CA, Wiley CE. Prospective study of the incidence of transient radicular irritation in patients undergoing spinal anesthesia. Anesthesiology 1996;84:1361–7.
19. Keld DB, Hein L, Dalgaard M, et al. The incidence of transient neurologic symptoms (TNS) after spinal anaesthesia in patients undergoing surgery in the supine position: hyperbaric lidocaine 5% versus hyperbaric bupivacaine 0.5%. Acta Anaesthesiol Scand 2000;44:285–90.
20. Kouri ME, Kopacz DJ. Spinal 2-chloroprocaine: a comparison with lidocaine in volunteers. Anesth Analg 2004;98:75–80.
21. Hodgson PS, Liu SS, Batra MS, et al. Procaine compared with lidocaine for incidence of transient neurologic symptoms. Reg Anesth Pain Med 2000;25:218–22.
22. Ben-David B, Maryanovsky M, Gurevitch A, et al. A comparison of minidose lidocaine-fentanyl and conventional-dose lidocaine spinal anesthesia. Anesth Analg 2000;91:865–70.
23. Tong D, Wong J, Chung F, et al. Prospective study on incidence and functional impact of transient neurologic symptoms associated with 1% versus 5% hyperbaric lidocaine in short urologic procedures. Anesthesiology 2003;98:485–94.
24. Hiller A, Karjalainen K, Balk M, Rosenberg PH. Transient neurological symptoms after spinal anaesthesia with hyperbaric 5% lidocaine or general anaesthesia. Br J Anaesth 1999;82:575–9.
© 2005 International Anesthesia Research Society
25. Carey TS, Boden SD. A critical guide to case series reports. Spine 2003;28:1631–4.