Epinephrine Increases the Neurotoxic Potential of Intrathecally Administered Lidocaine in the Rat

The study was approved by the Committee on Animal Research of the University of California, San Francisco. All experiments were conducted on male Sprague-Dawley rats (weight, 200–250 g).

Animals were anesthetized by intraperitoneal injection of methohexital (40–60 mg/kg), and catheters were introduced into the subarachnoid space using previously described modifications 14,15of the method of Yaksh and Rudy 16: 32-gauge polyurethane catheters (Micor, Allison Park, PA) were passed through a slit in the atlanto-occipital membrane and advanced 11 cm to lie with their tip caudal to the conus medullaris. Rats were allowed to recover for 24 h before the study began.

To assess sensory function, the tail-flick test was performed at the proximal, middle, and distal portions of the tail, as previously described. 14,15To prevent tissue damage, the heat stimulus was terminated if no response occurred by 8 s (cutoff). Anesthesia in the perineum, hind limbs, or trunk was defined as the absence of a vocal response and withdrawal to skin clamp.

Rats were divided into four groups to receive a 1-h intrathecal infusion (1 μl/min) of one of four test solutions. Group L (n = 20) received 5% lidocaine in saline; group LE (n = 20) received 5% lidocaine with epinephrine (0.2 mg/ml) in saline; group E (n = 20) received epinephrine (0.2 mg/ml) in saline; and group S (n = 20) received saline.

All solutions were prepared immediately before injection. The lidocaine solutions were prepared by dissolving crystalline lidocaine hydrochloride (Sigma Chemical, St. Louis, MO) in preservative-free normal saline (Abbott Laboratories, North Chicago, IL). The epinephrine solutions were prepared by adding the appropriate volume of a 1:1000 epinephrine solution (Abbott Laboratories) to the lidocaine solution or preservative-free normal saline.

Rats were placed in a horizontal acrylic restraint, and baseline tail-flick latency was assessed immediately before infusion. Infusions were administered at a rate of 1 μ/min for 1 h using a mechanical infusion pump. A segment of calibrated polyethylene tubing was inserted between the syringe and the intrathecal catheter, and the infusion was monitored by observing the movement of a small air bubble within the tubing. During infusion of the test solution, tail-flick latency was assessed every 10 min until the animal failed to respond to the heat stimulus. The extent of anesthesia level was determined by a skin clamp applied progressively cephalad until a response was elicited. Animals were evaluated for persistent elevation in tail-flick latency 4 and 7 days after infusion.

Animals were killed by injection of an overdose of pentobarbital 7 days after infusion. They were perfused intracardially with a phosphate-buffered glutaraldehyde-paraformaldehyde fixative. The spinal cord and nerve roots were dissected, immersed in the same glutaraldehyde solution used for perfusion fixation, and embedded in glycol methacrylate. The embedded tissue was sectioned at 6 mm rostral and 12 mm caudal to the conus medullaris, using a JB-4 microtome (1-μm transverse sections; Energy Beam Sciences, Agawam, MA). The tissue was treated with 4% osmium tetroxide and stained with toluidine blue. Histopathologic evaluation was performed using light microscopy by a neuropathologist blinded to the intrathecal solution received and to the results of sensory testing.

Tail-flick latencies at the proximal, middle, and distal portions of the tail were averaged to give a mean tail-flick latency. The extent of anesthesia during infusions was scored on a scale of 1–5, where 1 = tail, 2 = perineum, 3 = hind limb, 4 = lower trunk, and 5 = upper trunk. To assess the effect of drugs infused on sensory function, average tail-flick latencies were converted to percent maximal possible effect, calculated as [(tail-flick latency − baseline)/(cutoff − baseline)]× 100.

Specimens obtained 6 mm rostral to the conus were used for qualitative examination of the spinal cord. Specimens obtained 12 mm caudal to the conus were used for quantitative analysis of nerve injury. There were approximately 25 fascicles per cross-section, and each was assigned an injury score of 0–3, where 0 = normal, 1 = mild, 2 = moderate, and 3 = severe (table 1). The injury score for each animal was then calculated as the average score of all fascicles present in the cross-section.

To ensure that the four groups were equivalent before administration of the test solutions, baseline tail-flick latencies were compared using one-way analysis of variance. To assess the effect of infusion, baseline latencies for the control (saline) group were compared with their respective values obtained 4 and 7 days after infusion using repeated-measures analysis of variance. To determine the effect of the test solutions, the percent maximal effect data for tail flick for all groups were compared using a two-way analysis of variance with repeated measures over time (day 4 and day 7), with comparisons of all pairs performed with the Tukey Kramer test. The effect of epinephrine on block height was tested by ordinal logistic regression. Nerve injury severity for lidocaine versus  lidocaine with epinephrine, lidocaine versus  saline, and epinephrine versus  saline was evaluated using the Mann–Whitney U test with Dunn correction for multiple comparisons. Analyses were performed with GB-Stat (Dynamic Microsystems, Silver Spring, MD). For all comparisons, P < 0.05 was considered significant.

One animal given 5% lidocaine (group L) failed to develop perineal anesthesia during infusion and was excluded from data analysis. At necropsy, this catheter was noted to be extradural. During infusion, the level of sensory block ranged from the perineum to the upper trunk and was significantly higher in animals given lidocaine with epinephrine than in those given lidocaine alone.

There was no significant difference in baseline tail-flick latencies for the four groups. Assessment of sensory function on days 4 and 7 in saline-treated animals were similar to baseline. Tail-flick latencies (percent maximal effect) in animals treated with lidocaine and lidocaine plus epinephrine differed significantly from latencies in epinephrine- and saline-treated animals (fig. 1). The tail-flick latencies for animals given lidocaine with epinephrine were significantly greater than the latencies of animals receiving lidocaine alone. Sensory function in epinephrine- and saline-treated animals did not differ.

Sections obtained from animals treated with lidocaine, lidocaine with epinephrine, and epinephrine demonstrated moderate to severe injury in 32, 46, and 3% of fascicles, respectively. Those obtained from saline-treated animals revealed moderate injury in less than 0.5% of fascicles and no severe injury. The nerve injury scores for lidocaine with epinephrine were significantly greater than for lidocaine alone, and both differed significantly from saline or epinephrine. However, there was no significant difference between epinephrine and saline (fig. 2).

Qualitative light microscopic examination revealed focal damage of spinal posterior column in four animals in group L and one animal in group LE; however, there was no apparent spinal cord injury in the remaining animals, nor in any of the those in the epinephrine- or saline-treated groups.

The current experiments demonstrated that adding epinephrine significantly increases sensory impairment and morphologic damage induced by intrathecal lidocaine. However, although epinephrine affected lidocaine-induced injury, we found no significant functional or morphologic effect of epinephrine administered in the absence of anesthetic. The latter finding is consistent with previous results in the literature 17but must be interpreted cautiously. First, animals might have had minor functional impairment that went undetected by observation and limited sensory testing. Second, the incidence of moderate to severe fascicle injury with epinephrine was slightly higher than with saline. It is therefore possible that failure to detect a significant difference reflects lack of statistical power. Finally, morphologic changes were assessed only by light microscopic examination, and epinephrine might have an effect detectable by more sensitive methods. However, the clinical significance of such findings, if present, would be insignificant compared with the major sensory deficits and histologic damage produced by lidocaine and the profound potentiation of injury by the addition of epinephrine.

Four of the animals in the lidocaine group had focal damage in the posterior column compared with one animal in the group that received lidocaine with epinephrine, an observation inconsistent with the functional and quantitative histologic data. However, this difference was not statistically significant and may reflect the limited examination of the spinal cord, which was confined to a single evaluation at one level.

That epinephrine alone had no demonstrable effect may provide insight into the mechanism of its potentiation. Although such findings do not exclude epinephrine impacting directly on neural tissue, they are more consistent with an indirect effect perhaps mediated by epinephrine-induced vasoconstriction. Specifically, whereas vasoconstriction per se  might be well tolerated, a reduction in blood flow could delay absorption of lidocaine, increasing anesthetic exposure and hence toxicity. Although controversy exists, the available laboratory and clinical data provide some support for this concept. Experimentally, subarachnoid epinephrine has been shown to produce regional dural vasoconstriction 18while not affecting spinal cord blood flow. 8,18However, when administered with lidocaine, epinephrine attenuates both spinal cord and dural hyperemia induced by subarachnoid anesthetic. 19Epinephrine has also been shown to prolong spinal block in dogs 20and primates, 21although this may result, in part, from an antinociceptive effect derived from the α-adrenergic agonism of epinephrine. 22The impact of epinephrine on lidocaine uptake has been inconsistent, as some laboratory studies report significant reductions by epinephrine 9while others do not. 20,21Perhaps most pertinent, data from clinical investigations demonstrate the expected reduction in peak plasma lidocaine concentrations 7as well as a delay in recovery from spinal anesthesia. 7,23–26 

Surprisingly, in contrast to these data and despite its clinical use to prolong sensory block, some studies have not found epinephrine to extend the duration of a lidocaine spinal anesthetic. 27,28Two factors appear to account for these conflicting results. First, there is wide interpatient variability, which would promote a type II error, a concept supported by data from a crossover study in volunteers. 25Second, prolongation by epinephrine tends to preferentially occur at the lower segments of the neuraxis, and some studies have focused attention on thoracic dermatomes. For example, Spivey 28reported negative results but only examined the effect of epinephrine on two-segment regression. Although a study by Chambers et al.  27is often quoted for its negative findings, this interpretation is based on their data for two- and four-segment regression and duration of T12 sensory block. However, in the same study, Chambers et al.  reported statistically significant differences with respect to complete resolution of sensory impairment and recovery of motor function. In fact, the investigators never concluded that epinephrine lacked effect, but rather that there was no “clinically useful prolongation of spinal anesthesia.” This circumstance contrasts sharply with enhancement of lidocaine neurotoxicity, where the effect of epinephrine would be greatest for those neural elements most susceptible to injury.

Recommendations regarding the amount of epinephrine to be used for spinal anesthesia generally vary between 0.2 and 0.5 mg (0.2–0.5 ml of 1:1000 epinephrine). 29However, although there are no data regarding practice patterns, it is likely that the majority of clinicians use a dose between 0.2 and 0.3 mg with lidocaine administered as a 2.5% solution. Thus, when expressed as a concentration, the epinephrine administered in the current studies (0.2 mg/ml) likely exceeds that used by most clinicians. This difference might potentially impact the ability to extrapolate from the present data. However, it is the subarachnoid concentration that determines effect, and the subarachnoid concentration of epinephrine relative to that of lidocaine after injection is determined exclusively by the ratio of their respective doses in the administered solution. When considered on this basis, the amount of epinephrine used in the present studies (0.2 mg/50 mg lidocaine) is at the low to mid region of the clinical dose range. Moreover, the available data suggest that higher doses of epinephrine are needed with lidocaine to provide a clinically useful effect for most surgical procedures. For example, the most recent study examining the effect of epinephrine on duration of lidocaine spinal anesthesia compared 0.2, 0.4, and 0.6 mg, and only the 0.6-mg dose prolonged the time to two-segment regression. 24As previously noted, this finding is consistent with those in the literature demonstrating that lower doses of epinephrine are only effective at prolonging lidocaine spinal anesthesia at more caudal segments. 25,27,28 

Although adequate data concerning nerve blood flow of the cauda equina are lacking, the results of the current study suggest that ischemia is unlikely to play an important role with respect to epinephrine’s enhancement of lidocaine neurotoxicity. However, these results may not be applicable to peripherally injected anesthetic. 10,11In contrast to spinal administration, epinephrine and lidocaine both reduce peripheral-nerve blood flow, and the combination appears to be synergistic. 30It therefore remains to be determined whether such reduction is sufficient to induce ischemic injury or to contribute to anesthetic neurotoxicity.

The current studies used a commercially available solution of epinephrine containing the preservative sodium bisulfite. This was chosen to model anesthetic practice and thereby assess the clinical impact of using epinephrine with lidocaine for spinal anesthesia. Furthermore, it would have been difficult to conduct these experiments without an antioxidant because epinephrine is unstable in solution, undergoing rapid decomposition to adrenochrome. Nonetheless, we cannot rule out the possibility that potentiation of injury was caused by bisulfite, a preservative previously implicated in neurologic injury 12,13and with documented toxicity in some experimental models. 31–33However, despite these concerns, we believe it is unlikely that bisulfite contributed to recent deficits with spinal anesthesia or to the enhanced toxicity we observed. First, previous clinical deficits were associated with possible intrathecal administration of intended epidural doses of chloroprocaine. 12,13In addition to the much larger volume injected, the anesthetic solution contained 0.2% sodium bisulfite, which was not diluted before administration. In contrast, when epinephrine is used with lidocaine for spinal anesthesia, the final concentration of bisulfite is substantially lower because the epinephrine is added to a larger volume of (preservative-free) anesthetic. Second, the concentration of bisulfite used in earlier toxicity studies were generally an order of magnitude greater than we administered and, despite these high concentrations, evidence for bisulfite toxicity was inconclusive. 34,35Third, autopsy examinations after long-term intrathecal infusion of bisulfite-containing solutions for cancer pain fail to reveal neuropathologic findings suggestive of toxicity. 36,37Finally, we found no significant functional or morphologic effect of the epinephrine solution alone. Again, this does not eliminate the possibility that bisulfite potentiates lidocaine-induced damage, but it does make it far less likely.

The present findings perhaps shed some light on the etiology and significance of transient neurologic symptoms after spinal anesthesia. After a 1993 report of four patients with pain, dysesthesia, or both after spinal anesthesia with lidocaine, 38numerous case reports 39–42and clinical studies 43–49documented that these symptoms commonly follow lidocaine spinal anesthesia but are relatively rare with other anesthetics. There has been speculation and considerable concern that transient neurologic symptoms and neurologic deficits after spinal anesthesia represent opposite points on a single spectrum of toxicity. However, in contrast to the enhancement of injury we observed, data from a clinical trial 47and a large-scale epidemiologic study 49indicate that epinephrine has no effect on the incidence of transient neurologic symptoms after lidocaine spinal anesthesia. This discrepancy between our results and the clinical data argues against a common mechanism. At a minimum, the inconsistency indicates that the presence of transient neurologic symptoms is not an appropriate surrogate marker for major toxicity and, conversely, raises doubt that transient neurologic symptoms can be effectively investigated using existing animal models of neurotoxicity.

In conclusion, our results suggest that the neurotoxicity of intrathecal lidocaine is enhanced by the coadministration of epinephrine. The mechanism of this enhanced toxicity remains to be established, but it is perhaps related to an effect of vasoconstriction on anesthetic exposure. Although extrapolation to clinical practice must be made with caution, recommendations for maximum safe intrathecal dose of lidocaine may need to consider whether the solution contains a vasoconstrictor. However, considering the already narrow therapeutic index of spinal lidocaine and the existence of viable alternatives to lidocaine with epinephrine (e.g. , bupivacaine), the present findings call further into question the practice of using a vasoconstrictor to prolong lidocaine spinal anesthesia.

The authors thank Ricardo Ciriales, Department of Anesthesia and Perioperative Care, University of California, San Francisco, for excellent technical assistance.