Continuous epidural anesthesia is often used for patients undergoing infrainguinal peripheral vascular surgery (1). The advantages of epidural anesthesia over general anesthesia include attenuation of the stress response to surgery, reduced incidence of deep venous thrombosis, and increases in graft blood flow of as much as 50% (2). Additionally, the epidural catheter can be used for postoperative analgesia.
During prolonged vascular procedures under epidural anesthesia, large amounts of local anesthetic, often exceeding the maximal recommended dose, may be administered either by repeated doses or continuous infusion. Although hepatic clearance of a single dose of lidocaine is rapid (3), prolonged, continuous administration of lidocaine may lead to toxic plasma levels, particularly in elderly patients. Hepatic blood flow in elderly patients is reduced in direct proportion to age-related decreases in cardiac output. Additionally, even in the absence of hypotension, epidural anesthesia causes a further, independent decrease in hepatic blood flow (4).
We designed this study to test two hypotheses. First, during prolonged epidural lidocaine anesthesia, elderly patients would develop high, potentially toxic, plasma lidocaine concentrations. Second, because dopamine increases hepatic blood flow during epidural anesthesia (4), concurrent IV, small-dose dopamine would reduce plasma lidocaine concentrations during prolonged epidural anesthesia in elderly patients.
With institutional review board approval and informed consent, we studied 20 patients over the age of 65 who were scheduled for peripheral vascular surgery under epidural anesthesia. Patients were excluded if they had been in an intensive care unit preoperatively, had a contraindication to epidural anesthesia, had congestive heart failure or a left ventricular ejection fraction <35%, had hepatic disease with transaminases greater than 20% above normal, had renal disease with creatinine greater than 1.6 mg/dL, or were taking β-blockers.
After beginning standard physiologic monitoring, including an arterial catheter, and recording of baseline values, epidural catheters were inserted 3 cm into the epidural space at the L2-3 interspace. A maximum of 2 mg of midazolam and 100 μg fentanyl were used to facilitate these procedures. All patients received supplemental oxygen by nasal cannula or face mask to maintain SpO2 > 96%. After successful insertion of the epidural catheter, patients were randomly assigned (double-blinded) to receive an IV infusion of either dopamine at 2 μg · kg−1 · min−1 (dopamine group) or placebo (normal saline, control group). Both infusions were run continuously until 5 h after surgery. Five minutes after the beginning of the unknown IV infusion, 20 mL of 2% plain lidocaine was injected through the epidural catheter. Thirty minutes later, an epidural infusion of 2% plain lidocaine at 10 mL/h was begun and continued until the end of surgery. The epidural infusion was supplemented with 5-mL boluses for inadequate analgesia. If the anesthetic level rose above the T8 dermatome, the epidural infusion was decreased to 5 mL/h until the level returned to T8 or below. Phenylephrine, in 100-μg doses, was used when the systolic blood pressure was <90 mm Hg or more than 40% below baseline. Throughout the procedure, patients were monitored closely for signs and symptoms of local anesthetic toxicity. Arterial blood gases, hemoglobin and hematocrit, serum glucose, and other appropriate laboratory tests were performed according to our usual routine or as indicated.
Arterial blood samples were taken for total plasma lidocaine concentration (gas-liquid chromatographic assay) just before injecting the first epidural dose (baseline), at 5, 15, 30, 60, 90, and 120 min, and hourly thereafter. A final sample was taken when the lidocaine infusion was stopped at the end of surgery. After the termination of the lidocaine infusion, samples for total plasma lidocaine concentration were taken at 30, 60, and 90 min and at 2, 3, 4, and 5 h.
Comparisons between the two groups were made for mean rate of lidocaine administration (mg/h), maximal lidocaine plasma concentration, lidocaine plasma concentration at each hour, and the cumulative dose at each hour by using analysis of covariance, adjusting for patient age and weight where significant. A random effects regression model was used to compare groups for the increase in plasma lidocaine concentration during infusion and to estimate the average slope by group and overall. Kaplan-Meier survival analysis and Cox proportional hazards regression were used to compare plasma lidocaine half-time for the groups. Correlation between plasma lidocaine concentration and age, weight, and lidocaine dose was assessed with Spearman correlation coefficients. The significance level for each hypothesis was 0.05, and a Bonferroni correction for multiple comparisons was used when appropriate. We determined that a sample size of 10 per group would ensure 80% power to detect 2.0 μg/mL plasma lidocaine differences, assuming a standard deviation of 1.2 μg/mL (5). Results were reported as mean ± SD or with confidence intervals unless otherwise indicated.
Table 1 gives patient demographics, the duration of surgery, and intraoperative fluid balance for the two groups. Only weight was significantly different between the groups, therefore plasma lidocaine concentrations were adjusted for weight. Within 30 min of the initial 20-mL bolus of 2% lidocaine, all the patients developed adequate anesthetic levels (median dermatome = T6, range T8 to T2). After the initial lidocaine bolus, mean arterial blood pressure decreased similarly in both groups, 21% ± 12% in the control group and 30% ± 13% in the dopamine group. Phenylephrine, in 100-μg doses, was required for two patients (9 doses, total) in the control group and 5 patients (14 doses, total) in the dopamine group during the first 30 min. Thereafter, no further phenylephrine was required, and blood pressures and heart rates were similar for both groups.
Baseline plasma lidocaine levels showed trace amounts, 0.37 ± 0.32 and 0.32 ± 0.25 μg/mL for the control and dopamine groups, respectively. We assume these values resulted from the lidocaine used to facilitate epidural, arterial line, and venous access placement. Plasma lidocaine concentrations rose quickly after the initial epidural bolus injection of 400 mg. Within 5 min, plasma lidocaine levels had reached 89% ± 12% (control) and 80% ± 15% (dopamine) of the 30-min levels. At 30 min, plasma lidocaine levels were 3.2 ± 1.1 μg/mL for the control group and 3.2 ± 1.0 μg/mL for the dopamine group.
Throughout the study, there was no difference in plasma lidocaine concentration between the two groups. Plasma lidocaine concentrations increased continuously during the epidural lidocaine infusion (Fig. 1). The slope of the increase was not different between the groups, averaging 0.58 μg · mL−1 · h−1 (95% confidence interval: 0.47–0.68 μg · mL−1 · h−1). During the epidural infusion, the mean maximal plasma lidocaine concentration was 5.8 ± 2.3 μg/mL in the control group and 5.7 ± 1.2 μg/mL in the dopamine group. However, there was considerable individual variability; overall maximal plasma lidocaine concentrations ranged from 2.9 μg/mL to 9.5 μg/mL. There was no correlation between age and maximal plasma lidocaine concentration.
Although the plasma lidocaine concentrations were never different between the two groups, the amount of lidocaine required to maintain adequate analgesia differed significantly. The average lidocaine infusion rate was the total dose required for surgery divided by the length of surgery; control patients required 242 ± 72 mg/h and dopamine patients 312 ± 60 mg/h (P < 0.03) of the 2% lidocaine solution. Considering the infusion dose alone, control patients required 174 ± 64 mg/h and dopamine patients required 252 ± 61 mg/h (P < 0.02) We found a strong correlation (R = 0.7) between hourly lidocaine dose and maximal plasma concentration (P < 0.01). However, as noted earlier, plasma lidocaine concentration was not different between the two groups.
Although the duration of surgery was similar for both groups, there was considerable individual variability (Table 1); however, at the end of the fourth hour of surgery, there were still 10 patients in both groups in surgery. Thereafter, there was always a difference of at least two patients between the groups. The total dose of lidocaine, initial dose, infusion dose, and boluses, at the end of Hour 4 for the control group was 1088 ± 191 mg and for the dopamine group, 1228 ± 168 mg (P < 0.05, Fig. 2).
The total dose of lidocaine administered for surgery was 1650 ± 740 mg for the control group and 1940 ± 400 for the dopamine group. Despite these large doses and the fact that several patients developed plasma lidocaine concentrations in excess of 6.0 μg/mL, no patient exhibited any signs or symptoms of local anesthetic toxicity.
Although the lidocaine infusion was terminated at the end of surgery, on admission to the postanesthesia care unit, all patients had residual anesthetic levels ranging from L1 to T8. Plasma lidocaine samples taken during 5 h in the postanesthesia care unit were used to estimate the effect of dopamine on the decay of plasma lidocaine concentration. For each patient, the end of surgery constitutes the beginning of this analysis. Plasma lidocaine concentrations at the end of surgery were similar for the two groups 5.5 ± 1.2 μg/mL and 5.6 ± 1.1 μg/mL for the control and dopamine groups, respectively. Estimated context-sensitive plasma elimination half-times by using both the Cox model and Kaplan-Meier analysis were not different for the two groups, averaging 4 ± 1 h for both groups. At no time during the elimination phase were the plasma concentrations different between the groups. The mean slope of the terminal decay for both groups was 0.59 μg · mL−1 · h−1 (95% CI: 0.69–0.48 μg · mL−1 · h−1).
In this study, we had anticipated finding high concentrations of plasma lidocaine during prolonged epidural anesthesia in the control patients and demonstrating that a concurrent small-dose infusion of dopamine would result in lower plasma lidocaine levels in the dopamine patients. Instead, we found similarly high plasma lidocaine levels for both the control and dopamine patients, but a higher lidocaine dose requirement for the dopamine patients. We hypothesize that enhanced elimination caused the similarity between the plasma lidocaine levels, despite the significantly higher doses administered to the dopamine group.
In elderly patients, lidocaine clearance is reduced because of age-related decreases in cardiac output and concomitantly reduced hepatic blood flow, as well as lowered liver metabolism (6). Additionally, cardiac output may be further impaired in the elderly because of other disease or medications (7). Abernethy and Greenblatt (8) showed that male volunteers over the age of 65 had a 35% reduction in lidocaine clearance. Bowdle et al. (9) measured plasma lidocaine levels after a single 400 mg epidural bolus in elderly patients and did not find toxic plasma levels. However, in a simulation of repeated dosing, 200 mg every 45 min, they projected the plasma lidocaine concentration would approach 10 μg/mL after the third injection. Our elderly patients received, on average, over 260 mg/h of lidocaine for more than six hours, but the maximal plasma concentration was 6.2 ± 3.0 μg/mL, and only one patient had a plasma lidocaine level above 8.0 μg/mL.
Epidural anesthesia reduces hepatic blood flow even when normotension is maintained (4,10). However, a continuous,small-dose dopamine infusion, acting on dopamine receptors in the mesenteric vessels, restores hepatic perfusion and increases hepatic metabolism (4,11–13). Additionally, small-dose dopamine increases both lidocaine rate of transfer to and elimination from isolated perfused rat liver (14). Based on the above reasoning, we expected to see lower plasma lidocaine levels in the patients treated with dopamine, but these patients had the same plasma lidocaine levels as the control patients while requiring more epidural lidocaine to maintain the same level of anesthesia. Through the first four hours of anesthesia, the dopamine group required 13% more lidocaine. Over the entire duration of anesthesia administration for all patients, on an hourly basis, the dopamine group required 29% more lidocaine to maintain the same anesthetic level.
The only explanation we have for our findings is that dopamine did, in fact, increase hepatic elimination, thereby increasing lidocaine egress from the epidural space, increasing lidocaine dose requirements, and resulting in no measurable differences in plasma lidocaine levels between the two groups. The strong correlation between lidocaine dose and plasma level, despite no significant difference in mean plasma lidocaine concentration between the groups, tends to support this hypothesis. As to other possible mechanisms, we have not been able to find any data in the literature to support dopamine-mediated alterations in epidural blood flow that would explain the higher dose requirements. This would be particularly appealing in concert with increased hepatic elimination. Finally, dopamine itself is a mediator of nociception (15). Axons containing dopamine in the spinal cord have a role in pain modulation (16), and dopamine antagonists have antinociceptive effects (17–19). However, because the epidural block is peripheral to the site of dopamine’s effect on nociception, it is difficult to implicate this mechanism as the cause of the increased lidocaine requirements in the dopamine group.
It is unfortunate that the wide range of operative times, total lidocaine doses, and plasma lidocaine levels prevented the demonstration of a significant difference in lidocaine elimination times after surgery. The hypothesis that dopamine-enhanced hepatic elimination drives an increase in lidocaine flux causing the increased dose requirement in the dopamine group is appealing. A clearly more rapid terminal elimination phase in the dopamine group would have further strengthened that hypothesis.
In this study, elderly patients received a continuous epidural infusion of lidocaine averaging more than 200 mg/h for as long as nine hours without developing symptoms or signs of local anesthetic toxicity, despite one-third of the patients’ having plasma lidocaine concentrations greater than 6 μg/mL. The concurrent administration of small-dose dopamine (2 μg · kg−1 · min−1) appears to have increased hepatic elimination, causing as much as 45% increase in the epidural lidocaine infusion requirement after the induction bolus, without a significant difference in plasma lidocaine levels. Although the number of subjects in our study was relatively small, we believe the results indicate that the use of a dopamine infusion during prolonged lidocaine epidural anesthesia does not reduce plasma lidocaine concentration.
We thank Fanny Shutway, RN, for her valuable assistance with this study.
1. Yeager M, Glass D, Neff R, Brinck-Johnses T. Epidural anesthesia and analgesia in high risk surgical patients. Anesthesiology 1987; 66:729–36.
2. Cousins M, Wright C. Graft, muscle, skin blood flow after epidural block in vascular surgical procedures. Surg Gynecol Obstet 1971; 133:59–64.
3. Bowdle T, Freund P, Slattery J. Propranolol reduces bupivacaine clearance. Anesthesiology 1987; 66:36–3.
4. Tanaka N, Nagata N, Hamakawa T, Takasaki M. The effect of dopamine on hepatic blood flow in patients undergoing epidural anesthesia. Anesth Analg 1997; 85:286–90.
5. Takasaki M, Kajitani H. Plasma lidocaine concentrations during continuous epidural infusion of lidocaine with and without epinephrine. Can J Anaesth 1990; 37:166–9.
6. Brandfonbrener M, Landowne M, Shock N. Changes in cardiac output with age. Circulation 1955;557–66.
7. Conrad K, Beyers J, Finley P, Burnham L. Lidocaine elimination: effects of metoprolol and of propranolol. Clin Pharmacol Therap 1983; 33:133–8.
8. Abernethy D, Greenblatt D. Impairment of lidocaine clearance in elderly male subjects. J Cardiovasc Pharmacol 1983; 5:1093–6.
9. Bowdle T, Freund P, Slattery J. Age-dependent lidocaine pharmacokinetics during lumbar peridural anesthesia with lidocaine hydrocarbonate or lidocaine hydrochloride. Reg Anesth 1986; 11:123–7.
10. Kennedy WF, Everett GB, Cobb LA, Allen GD. Simultaneous systemic and hepatic hemodynamic measurements during high peridural anesthesia in normal man. Anesth Analg 1971; 50:1069–77.
11. Angehrn W, Schmid E, Althuaaus F, et al. Effects of dopamine on hepatosplanchnic blood flow. J Cardiovasc Pharmacol 1980; 2:257–65.
12. Kurokawa T, Nonami T, Ujita T, et al. Dopamine compensates for deterioration of hepatic hemodynamics and metabolism during occlusion and reperfusion of the hepatic artery. Arch Surg 1995; 130:1085–9.
13. Winso O. Effects of dopamine on the portal circulation after therapeutic hepatic artery ligation. Acta Anaesthesiol Scand 1988; 32:458–63.
14. Shimadzu K, Nishi S, Kariya N, et al. The pharmacokinetic change of lidocaine by catecholamine using isolated perfused rat liver. Life Sci 1998; 62:2399–405.
15. Peterfreund R, Kosofsky B, Fink J. Cellular localization of dopamine D2 receptor messenger RNA in the rat trigeminal ganglion. Anesth Analg 1995; 81:1181–5.
16. Ridet J, Geffard M, Privat A. Light and electron microscopic studies of the effects of p-chloroamphetamine on the monoaminoergic innervation of the rat spinal cord. J Comparat Neurol 1994; 343:281–96.
17. Bittencourt A, Takahashi R. Mazindol and lidocaine are antinociceptives in the mouse formalin model: involvement of dopamine receptors. Eur J Pharmacol 1997; 330:109–13.
18. Schreiber S, Backer M, Weizman R, Pick C. Augmentation of opioid antinociception by the atypical antipsychotic drug risperidone in mice. Neuroscience 1997; 228:25–8.
19. Wang H, Xu SF. Effect of D1 and D2 dopamine receptor antagonists on acupuncture analgesia. Sheng Li Hsueh Pao 1993; 45:61–8.