Volume 104

Number 2, Part 1 Prolonged verapamil elimination with chronic therapy

Goodman AH: Cardiovascular action of verapamil in the dog with particular reference to myocardial contractility and atrioventricular conduction. Cardiovasc Res l&623, 1976.

21. Melville KI, Benfey BG: Coronary vasodilatory and cardiac adrenergic blocking effects of iproveratril. Can J Physiol Pharmacol 43:339, 1965.

22. Nayler WG, McInnes I, Swann JB, Price JM, Carson V, Race D, Lowe TE: Some effects of iproveratril (Isoptin) on the cardiovascular system. J Pharmacol Exp Ther 161:247, 1968.

23. Nayler WG, Szeto J: Effect of verapamil on contractility, oxygen utilization, and calcium exchangeability in mammali- an heart muscle. Cardiovasc Res 6:120, 1972.

24. Chew YC, Hecht HS, Collett JT, McAllister RG, Singh BN:

Influence of severity of ventricular dysfunction on hemody- namic responses to intravenously administered verapamil in ischemic heart disease. Am J Cardiol 47:917, 1981.

25. Eichelbaum M, Birkel P, Grube U, Gutgemann, Somogyi A: Effects of verapamil on P-R intervals in relation to verapamil plasma levels following single I.V. and oral administration during chronic treatment. Klin Wochenschr 58:919, 1980.

26. Keefe DL, Harapat SR, Kates RE: Relationship between pharmacodynamics and myocardial concentration of verap- amil. Am J Cardiol 47:406, 1981.

27. Schwartz JB, Keefe D, Peters F, Kates R, Harrison DC: Concentration dependent heart rate suppression with verap- amil. Am J Cardiol 47:406, 1981.

Excessive serum lidocaine levels during maintenance infusions: Mechanisms and prevention

Clinical and pharmacokinetic data were reviewed in 72 patients who developed excessive iidocaine serum levels during maintenance infusions. Fifty-one of the 72 (70%) were cardiac patients who had mean iidocaine excretory clearances less than one half of normal. Forty percent of these became toxic in spite of a reduced infusion rate (30 wg/kg/min). Seven patients with normal excretory mechanisms became toxic when they received large doses of iidocaine. The remaining 14 cases lacked an identifiable cause to explain the development of higher than therapeutic serum levels. inordfnateiy high serum levels of monoethyigiycinexylidide (MEGX), an active iidocaine metaboitte, were found in seven patients, but in only one was MEGX greater than iidocaine. Prolonged infusions (24 hours or greater) were not clearly associated with the worst iidocaine elimination clearances. Lidocaine toxicity was life-threatening or significantly complicated the management of 15 patfents. Based on the data presented, guidelines are offered as an approach to the prevention of toxicity from maintenance iidocaine infusions. (AM HEART J 104:203, 1982:)

Richard Davison, M.D., Michele Parker, R.N., and Arthur J. Atkinson, Jr., M.D. Chicago, Ill.

Intravenous lidocaine is a safe and effective agent for the acute control of ventricular ectopic activity. Fatal toxicity is very rare and has usually been associated with bolus administration. In contrast,

From the Department of Medicine and Pharmacology, Northwestern University, and the Department of Nursing, Northwestern Memorial Hospital.

Received for publication Oct. 2, 1980; revision received March 12, 1981; accepted June 2, 1981.

Reprint requests: Richard Davison, M.D., Northwestern Memorial Hospi- tal, Wesley-&B, 250 E. Superior St., Chicago, IL 60611.

adverse effects during the continuous infusion of lidocaine are not uncommon and may be life-threat- ening. The available literature that is pertinent to this topic is quite limited and consists almost exclu- sively of isolated case reports and incidence studies.1-4 The following discussion is based on clini- cal and pharmacokinetic data derived from 72 patients who developed excessive serum levels dur- ing maintenance therapy with lidocaine. It is an attempt to gain insight into the mechanisms respon- sible for lidocaine toxicity and to generate recom- mendations that will help lessen its incidence.

0002-8703/82/080203 + 06$00.60/O 13 1982 The C. V. Mosby Co. 203

204 Davison, Parker, and Atkinson August, 1982

American Heart Journal

Table I. Clinical characteristics of patients

CHF-S OD tiNX Total

Number of patients Mean age, years In-hospital mortality Primary diagnosis:

Acute myocardial infarction Arrhythmia CHF Pulmonary embolism Renal failure Chronic lung disease Shock Postoperative

51 7 14 72 68.5 (+10.3)* 74.1 (k10.3) 64.1 (kl1.3) 68.9 (k10.6)

51$ 43% 14"; 43°C.

20 5

17 2 3 1 1 2

1 7 28 1 5 11

- - 17 - - 2 1 - 4 2 2 5

- - 1 2 - 4

*Standard deviation in parentheses. Abbreviations: CHF-S = congestive heart failure-shock; OD = overdose; UNX = unexpected.

METHODS

During the period between April, 1973, and March, 1976, 2287 patients were admitted to the Coronary and Medical Intensive Care Units at Northwestern Memorial Hospital. Nine-hundred and two received lidocaine for the treatment of ventricular ectopic activity. One or more intravenous boluses of lidocaine were followed by a con- tinuous infusion administered by an electronic pump (IMED or IVAC). The house staff was encouraged to order infusion rates according to the patient’s weight (15 to 55 wg/kg/min), and to secure blood samples for lidocaine levels whenever toxicity was suspected. All blood samples were obtained by individual venipuncture or were drawn from an indwelling arterial catheter. A total of 362 blood lidocaine levels were obtained and 88 were found to be above the accepted therapeutic level (2 to 5 &ml). A review of these patients’ hospital records indicated that in 16 instances the high lidocaine blood level may not have accurately reflected a true steady-state with the biophase, either because the time of sampling was poorly docu- mented or because the level was obtained less than 1 hour after a bolus injection. The remaining 72 cases form the basis of this report. All were personally followed during their acute illness by one of the investigators.

Patients. For the purpose of analysis, four categories were established based on factors that are known to predispose to lidocaine toxicity: (1) Congestive heart failure-shock (CHF-S). Patients were considered to have clinically diagnosable CHF if they presented at least three of the four following findings: (a) pulmonary congestion by physical examination or chest x-ray examination; (b) cardiomegaly by chest x-ray examination; (c) dependent pitting edema; and (d) an audible S, gallop. Shock was felt to be present when any two of the following three items were noted: (a) arterial systolic blood pressure (cuff method) of 80 mm Hg or less; (b) urinary flow of less than 30 c&r; and (c) pale, clammy skin. (2) Liuer disease. Abnormal liver function tests and a history of preexisting liver disease identified the patients in this group. (3) Overdose (00). Lidocaine dosage was felt to have been

excessive whenever the rate of its administration exceeded 55 pg/kg/min. (4) Unexpected (UNX). This category was composed of patients that did not fulfill any of the above criteria yet were found to have lidocaine blood levels in excess of therapeutic levels. The lidocaine elimination clearance was calculated for each patient as follows: lidocaine elimination clearance (ml/min/kg) = lidocaine dose (pg/kg/min)/lidocaine serum levels &g/ml). Data were analyzed using the one-way analysis of variance test.

Assay procedures. Serum levels of lidocaine and mono- ethylglycinexylidide (MEGX), a biologically active metabolite of lidocaine,6 were determined by gas chroma- tography, as were measurements of plasma concentrations of lidocaine and MEGX. Trimecaine (mg, 45 to 47’ C) was used as an internal standard for the assay and was made up in 1:4 chloroform-benzene to a concentration of 1.0 pg/ml. To a 15 ml glass-stoppered conical centrifuge tube were added 1 ml plasma, 5 ml internal standard solution, and 0.2 ml of 5N NaOH. After mixing, the tube was centrifuged at 800 x g for 3 minutes. The organic phase was transferred to a 5 ml pear-shaped flask and evapo- rated to dryness at 25 mm Hg in a 25’ C water bath. After dissolving the residue in 50 ~1 of benzene, a 5 ~1 aliquot was injected into a Varian Series 1440 gas chromatograph. The chromatograph was equipped with a flame ionization detector and fitted with a 2.0 m x 2 mm (i.d.) stainless steel coiled tube packed with 6:l SE-30:OV-17 coated on SO/l00 mesh Chromosorb solid support (Chemical Re- search Services, Addison, Ill.). The temperature of the injection port was 270’ C, of the column 210” C, and of the detector 250” C. The flow rate of nitrogen carrier gas was 40 mllmin. The hydrogen and air flows for the detector were 30 ml/min and 200 ml/min, respectively. A Hewlett- Packard Model 3352 B Laboratory Data System was used to determine peak areas and to calculate plasma concen- trations of lidocaine and MEGX. Chromatography time was 10 minutes, with an overall analysis time of approxi- mately 1 hour since samples were analyzed in duplicate. Over a 6-month period, repeated assay of 5.0 pg/ml

Volume 104

Number 2, Part 1 Lidocaine toxicity during maintenance infusions 205

Table II. Maintenance lidocaine infusions: Dose, duration, serum levels, and metabolism*

CHF-S

Body weight (kg) 66.9 ( c 13.3) Lidocaine dose

gglkglmin 32.8 ( I!X 12.3) mglmin 2.2 ( + 0.84)

Lidocaine serum level &g/ml) 7.3 (k2.0) Duration of infusion (hr)t 34 Lidocaine elimination clearances (ml/min/kg)$

All infusions 4.6 Infusions < 24 hours 4.65 Infusions > 24 hours 4.80

MEGX serum level (kg/ml) 1.5 (k1.5) MEGX/lidocaine ratio 0.17 (kO.26)

*All values are mean ? one standard deviation. iFrom start of infusion to time of sample with high blood level. fLidocaine elimination clearance = lidocaine dose (ag/kg/min)/lidocaine serum level (@g/ml).

UNX OD

69.7 (2 11.4) 53 (t0.5)

37.6 (k 11.1) 61.1 ( +- 4.6) 2.6 (f 1.0) 3.2 (~2.0) 7.2 (+ 2.2) 6.5 ( k2.0)

29 16

9.9 5.7 - 5.24 - 5.93

1.26 (of: 1.0) 1.26 (k 1.06) 0.21 (20.18) 0.19 (kO.17)

Abbreviations same as in Table 1.

lidocaine and MEGX plasma standards under routine not attain statistical significance 07 > 0.05). Only 2 operating conditions gave a mean value of 5.3 a/ml for of 7 OD patients received lidocaine for over 24 hours, both compounds with a precision of 11% (S.D.) for lidocaine and 22% for MEGX.

while prolonged infusions were common in the CHF-S (29 of 54,57 % ) and UNX (7 of 14,50 % ) groups.

RESULTS

Patient population. The clinical characteristics of the patients discussed are summarized in Table I. Of the 51 individuals in the CHF-S group, 45 had CHF, one presented in shock without evidence of CHF, and another five fulfilled criteria for both. Only two instances were found in which liver disease was prominent, therefore no further analysis of these cases was attempted. Seven patients (10% of the total population) received excessive amounts of lido- Caine for their body weights. The 14 cases labeled “unexpected” (19% of total population) were those in whom no obvious explanation could be found for the high serum lidocaine levels.

Lidocaine infusion-dose and duration. This infor- mation is summarized in Table II. Although a statistically significant difference existed between the OD category and CHF-S group in absolute amount of lidocaine administered (pg/min), no such difference was noted in serum levels of the drug between all groups. The OD category had a signifi- cantly smaller mean body weight (p < 0.05). This explains the greater dose of lidocaine (hg/kg/min) received by the patients in this group (p < 0.05). The duration of the lidocaine infusions was mea- sured from their initiation to the time when an elevated serum level was documented. Although the mean infusion time was shorter in the OD group (16 hours f 10) than in CHF-S (35 hours f 30) and the UNX (29 hours + 19) categories, this difference did

Lidocaine excretion and metabolltes (Table II). The lidocaine elimination clearance was calculated for each patient and the results are displayed in Fig. 1. Predictably, elimination clearances of patients with CHF clustered in the lower range, and significantly differed (p < 0.05) from those of patients in the OD group. Cases in which lidocaine toxicity occurred unexpectedly were found throughout the spectrum of lidocaine clearances. Since the rate of lidocaine elimination has been shown to decrease after 24 hours of continuous administration: the patients in each group were divided into those that had received up to 24 hours of intravenous lidocaine mainte- nance, and those with infusions exceeding 24 hours. When the mean lidocaine elimination clearance for those patients were compared within the CHF-S and UNX categories, no statistically significant differ- ence was demonstrated (p > 0.05).

The mean levels of MEGX were quite similar for the three groups and the same was true for the ratio between the mean MEGX and the lidocaine serum levels. Five of the 72 patients demonstrated a MEGX serum level that was equal to or greater than 50% of the lidocaine level. The only instance where the MEGX level was greater than the lidocaine level (9.8 and 5.7 gm/cc, respectively), was found in a CHF patient.

Clinical manifestations of lidocaine toxicity. Lido- Caine toxicity was suspected in 83% of our patients (60 of 72) on the basis of clinical manifestations that subsided when the infusion was reduced or discon-

206 Davison, Parker, and Atkinson August, 1982

Amsrtcan Heart Journal

13 12

11

10

NURSER g 8

PATIENTS 7 6

5 4

3 2

1

LIDOCAINE CLEARANCE kc/min/Kgl

Fig. 1. Frequency and distribution of lidocaine elimination clearances in 72 patients with excessive lidocaine serum levels.

Table III. Morbidity due to lidocaine toxicity

Hypotension Cardiac arrest Apnea Coma Agitation* Syncope

*Requiring tranquilizers and/or restraints.

tinued. In three other cases the patients were either comatose or paralyzed and therefore could not be clinically evaluated for toxicity. Levels were obtained in nine asymptomatic patients to establish whether an inadequate antiarrhythmic response was due to subtherapeutic levels of lidocaine. In the majority of cases, toxicity was manifested by rela- tively mild central nervous system symptomatology such as dizziness, dysarthria, mild confusion, tinni- tus, paresthesias, etc. In 15 instances (25 % ) morbid- ity of considerable import was observed and is detailed in Table III. In all of these instances, no other cause was found for the clinical manifestations which did not recur after the lidocaine infusion was reduced or discontinued. Lidocaine could not be implicated in any of the deaths that occurred within the group.

DISCUSSION

Importance of conslderlng lldocaine dosage per body wetght. Following initial loading, either by repeated bolus injections or by a brief high dose infusion,7’a a constant infusion of lidocaine is recommended in order to maintain blood and tissue levels. Excessive blood levels will develop whenever the rate of

administration of lidocaine exceeds the rate of elim- ination. This phenomenon is well exemplified by the seven patients (OD) in our series who received lidocaine infusions at rates greater than the recom- mended maximum (55 clg/kg/min). An impairment in lidocaine removal was excluded by the demon- stration of a lidocaine elimination clearance that was well within the range reported in normal individuals (Thompson et als: 11.1 ml/min/kg; Zito and ReidlO: 10.9 ml/min/kg). Although the elimination clear- ances are conventionally expressed as a function of body weight, it is a widespread practice to calculate lidocaine maintenance doses in milligrams per minute, without adjustment for weight.“-I3 We believe that this unwise conduct fosters a form of lidocaine toxicity that is entirely preventable. The risks associated with “fixed” infusion rates are illustrated by the study of Lie et alI* Fifteen percent of 107 patients receiving lidocaine infusions-3 mg/ min, regardless of body weight-experienced side effects. This is an immoderately high incidence of toxicity considering that the elderly and the serious- ly ill had been excluded.

Impaired cardiocirculatory function predisposes to

lidocaine toxicity. In 70% of our cases (51 of 72) the excessive accumulation of lidocaine was associated with significant impairment of cardiovascular func- tion. Passive venous congestion of the liver or a reduction in its blood supply such as seen in shock states, are factors known to interfere with the disposition of lidocaine. The mean lidocaine clear- ance of our cardiac patients was calculated to be less than one half of normal. Yet 75% of these individu- als received more than the 20 wg/kg/min recom- mended maximum maintenance dose of lidocaine in

Volume 154

Number 2, Part 1 Lidocaine toxicity during maintenance infusions 207

this situaLion.15 Assuming that the present series is representative of the population at risk, it is con- cluded that excessive lidocaine serum levels can most often be attributed to circulatory impairment that is overlooked or, at best, underestimated. Because the extent of the impairment in lidocaine elimination cannot be assessed clinically with any degree of accuracy, this type of toxicity can best be avoided if lidocaine maintenance is routinely started at a dose of 10 pg/kg/min in patients presenting any evidence of cardiovascular dysfunction.

Consideration of instances of unexpected lidocaine

toxicity. Plasma levels of MEGX were assayed in all but one of the cardiac patients. We were unable to confirm previous observations suggesting that CHF promotes the accumulation of MEGX, a metabolite that could contribute to the induction of toxic phenomena.16 Approximately one out of every five patients that developed above-therapeutic lidocaine serum levels did so for reasons that were not clini- cally apparent. The lidocaine clearances of these “unexpected” cases of toxicity offered a wide range of values and suggest that this is a heterogeneous group. It can be speculated that they represent cases of unsuspected cardiac or hepatic disease, or less likely, examples of aberrant lidocaine metabolism.

Consideration of prolonged infusions and of CNS compromise. Recent evidence indicates that pro- longed infusion of lidocaine-24 hours or greater- are associated with a decline in the capacity of the liver to excrete this drug. 6* l7 If this phenomenon had played an important part in our series, we would have expected that those individuals receiving lido- Caine for the longest period would also have the poorest lidocaine elimination clearances. Since such association was not demonstrated, the contribution of extended infusions to the genesis of lidocaine toxicity remains uncertain and requires further investigation. Conforming past observations, we found that the clinical manifestations of lidocaine toxicity are usually mild and most commonly referred to the central nervous system (CNS). None of our patients experienced seizures, a side effect almost universally listed as common with lidocaine toxicity.ls l8 On the other hand, as many as 25% of the patients that were clinically toxic suffered man- ifestations that were either life-threatening or that significantly complicated their management. Espe- cially at risk were those cases in whom lidocaine was administered in the face of a compromised mental status. In this setting two unfortunate patterns were seen to develop. If the early signs of toxicity were clouded by the underlying disease state, lidocaine accumulation was allowed to continue to the point of

severe cardiorespiratory depression. Alternatively, the toxic manifestations were recognized but misin- terpreted as evidence of progression of the neurolog- ical pathology, and misguided therapeutic ventures were sometimes undertaken.

Because routine blood samples were not obtained, it is almost certain that not all instances of excessive serum levels of lidocaine occurring within our study population were identified. Furthermore, the serum level at which clinical manifestations of lidocaine toxicity become evident varies considerably between individuals. Therefore we suspect that the relative contribution of each of the mechanisms discussed depends upon the characteristics of the populations studied and the method of sampling.

Conclusions. The following guidelines summarize our conclusions and are offered as an approach to the prevention of toxicity from maintenance lido- Caine infusions: (1) The dose of lidocaine should be routinely calculated on the basis of body weight. There is no justification for protocols that call for a fixed number of milligrams to be infused per minute. (2) Since the clinical quantification of circu- latory insufficiency is quite inaccurate, patients exhibiting this condition should initially receive the lowest recommended infusion rate (10 pg/kg/min). (3) The administration of lidocaine to patients with preexisting neurological dysfunction is especially fraught with risk. If serum levels are not easily obtainable, alternate antiarrhythmic medication should be considered. (4) When lidocaine is given “prophylactically” to arrhythmia-free patients, the possibility of adjusting its infusion rate to the lowest effective dose is lost. In this setting, toxic or subther- apeutic levels can best be avoided by routine serum level determinations. (5) Until the clinical signifi- cance of the increase in lidocaine levels occurring with prolonged infusions is ascertained, it seems wise to attempt a reduction in the administration rate after 24 hours. (6) Any change in the hemody- namic or mental status of a patient receiving lido- Caine is due to early toxicity until proven otherwise. The onset of lidocaine toxicity is always insidious and often unexpected.

The authors gratefully acknowledge the Medical Records Staff, Northwestern Memorial Hospital, for their help and support; Maria Rivera for expert secretarial assistance; and Melvin B. Meyer (Astra, Worcester, Mass.), for providing the supplies of Trimecaine used in this study.

REFERENCES

1. The Boston Collaborative Drug Surveillance Program: Drug- induced convulsions. Lancet 1~671, 1972.

2. Pfeifer HJ, Greenblatt DJ, Koch-Weser J: Clinical use and toxicity of intravenous lidocaine. AM HEART J 92:168, 1976.

3. Jewitt DE, Kishon Y, Thomas M: Lignocaine in the manage-

208 Davison, Parker, and Atkinson Augusl. 1982

American Heart Journal

4.

5.

6.

7.

8.

9.

10.

ment of arrhythmias after acute myocardial infarction. Lan- cet 1:266, 1968. Cheng TO, Wadhmer K: Sinus standstill following intrave- nous lidocaine administration. JAMA 223:790, 1973. Strong JM, Atkinson AJ Jr: Simultaneous measurement of plasma concentration of lidocaine and its desethylated metabolite by mass fragmentography. Anal Chem 44:2287, 1977. LeLorier J, Moisan R, Gagne J, Caille G: Effect of the duration of infusion on the disposition of lidocaine in dogs. J Pharmacol Exp Ther 203:507, 1977. Wyman MG, Lalka D, Hammersmith L, Cannon DS, Gol- dreyer BN: Multiple bolus technique for lidocaine adminis- tration during the first hours of an acute myocardial infarc- tion. Am J Cardiol 41:313, 1978. Levy RA, Charuzi Y, Mandel WJ: Lignocaine: A new tech- nique for intravenous administration. Br Heart J 39:1026, 1977. Thompson PD, Melmon KL, Richardson JA, Cohn K, Stein- brunn W, Cudihee R, Rowland M: Lidocaine pharmaeokinet- its in advanced heart failure, liver disease, and renal failure in humans. Ann Intern Med 78:499, 1973. Zito RA, Reid PR: Lidocaine kinetics predicted by indocya- nine green clearance. N Engl J Med 298:1160. 1978.

11.

12.

13. 14.

15.

16.

17.

18.

Harrison DC: Should lidocaine be administered routinely to all patients after acute myocardial infarction? Circulation 58:581, 1978. Rothfeld EL, Parsonnet J, McGorman W, Linden S: Harbin- gers of paroxysmal ventricular tachycardia in acute myocar- dial infarction. Chest 71:142, 1977. Gazes PC: The first few days. ER Medicine 91:128, 1977. Lie KI, Wellens HS, van Capelle FJ, Durrer D: Lidocaine in the prevention of primary ventricular fibrillation. N Engl J Med 291:1324, 1974. Collinsworth KA, Kalman SM, Harrison DC: The clinical pharmacology of lidocaine as an antiarrhythmic drug. Circu- lation 50:1217, 1974. Halkin H, Meffin P, Melmon KL, Rowland M: Influence of congestive heart failure on blood levels of lidocaine and its active monodeethylated metabolite. Clin Pharmacol Ther 17:669, 1975. LeLorier J, Grenon D, Latour Y, Caille G, Dumont G, Brosseau A, Solignac A: Pharmacokinetics of lidocaine after prolonged intravenous infusions in uncomplicated myocardi- al infarction. Ann Intern Med 87:700, 1977. Crampton RS, Oriscello RG: Petit and grand ma1 convulsions during lidocaine hydrochloride treatment of ventricular tachycardia. JAMA 204:201, 1968.