The pharmacokinetic and pharmacodynamic interaction between propafenone and lidocaine

Although propafenone is a known substrate and inhibitor of the cytochrome P450 4-hydroxylation pathway of debrisoquin (CYP2D6 isozyme), its effects on other hepatic mixed- function oxidative isozymes have not been extensively evaluated. We studied the influence of propafenone on the disposi- tion of continuously infused lidocaine in 12 healthy male volunteers. Placebo or propafenone (225 mg every 8 hours) was orally administered for 4 days before and during lidocaine administration (2 mg/kg/hr for 22 hours). In the 11 (92%) subjects phenotyped as extensive metabolizers, propafenone significantly increased the lidocaine area under the plasma concentration time curve (81.7 ± 16.2 versus 76.3 ± 15.6 p.g hr/m1; p s 0.05) and reduced systemic fidocaine clearance (9.53 ± 1.77 versus 10.27 ± 2.24 ml/min/kg; p 5_ 0.05), but did not significantly affect volume of distribution at steady state (2.48 ± 0.33 versus 2.64 ± 0.45 L/kg; p = 0.10) or mean residence time (4.37 ± 0.92 versus 4.47 ± 0.87 hours; difference not significant) compared with placebo, respectively. Adverse central nervous sys- tem effects were significantly worse in severity and duration during the propafenone phase (p 5. 0.05). Propafenone minimally inhibits the metabolism of lidocaine. This suggests that the ability of pro- pafenone to inhibit metabolic pathways exclusive of the CYP2D6 isozyme may be limited. In addition, potentiation of disturbing central nervous system adverse effects may occur during combination therapy of propafenone and fidocaine. (CLIN PHARMACOL TILER 1993;53:38-48.)

Michael R. Ujhelyi, PharmD, Eleanor A. O'Rangers, PharmD, Chengde Fan, BS, Jeffrey Kluger, MD, Chantal Pharand, PharmD, and Moses S. S. Chow, PharmD Stors and Hartford, Conn.

Propafenone is a unique antiarrhythmic agent that has a molecular structure and a chiral carbon similar to the p-adrenergic blocker propranolo1.1 It is exten- sively metabolized through several metabolic path- ways, but the cytochrome P450I1D6 (CYP2D6) isozyme predominates. The CYP2D6 isozyme hydrox- ylates propafenone to its major metabolite, 5-hydroxy- propafenone, whereas other mixed-function oxidative pathways are responsible for the formation of the other major metabolite of propafenone, N-depropyl- propafenone. Expression of the CYP2D6 isozyme is

genetically regulated, and approximately 7% of the

From the Schools of Pharmacy and Medicine, University of Con- necticut, Stors and Hartford, and the Departments of Pharmacy Service and Cardiology, Hartford Hospital, Hartford.

Supported by a grant-in-aid from Knoll Pharmaceuticals. Received for publication April 20, 1992; accepted Sept. 28, 1992. Reprint requests: Michael R. Ujhelyi, PharmD, University of Cin-

cinnati Medical Center, College of Pharmacy, 3223 Eden Ave., Cincinnati, OH 45267-0004.

13/1/43076

white population is enzyme deficient and therefore are phenotypically classified as poor metabolizers.2-5 The result is a fourfold higher oral clearance (CL.) of pro- pafenone in extensive metabolizers compared with phenotyped poor metabolizers (1115 -± 1238 versus 264 -1- 48 ml/min).2 Furthermore, patients phenotyped as extensive metabolizers can saturate the CYP2D6 isozyme at therapeutic doses, such that a dosage in- crease in propafenone results in a disproportionate in- crease in the plasma propafenone concentration."

The complex nature of the metabolism of pro- pafenone may potentially be responsible for interfer- ing with the metabolism of other substrates. It has been well documented that propafenone can inhibit the oxidative metabolism of other agents that also un- dergo polymorphic oxidative metabolism through the CYP2D6 isozyme, such as propranolol, metoprolol, and debrisoquin.29.1° The mechanism of this specific interaction is thought to be caused by the ability of propafenone to saturate or competitively inhibit this isozyme, thereby inhibiting the oxidative metabolism of substrates for this pathway.

38 0009-9236/93/$1.00 + 0.10

VOLUME 53

NUNMER1

Propafenone may also inhibit the metabolism of agents that undergo oxidative metabolism but do not cosegregate with debrisoquin. One investigation showed a 50% increase in warfarin concentrations during concomitant propafenone administration." Studies in rats have also shown that propafenone is an inhibitor of global metabolic oxidation, reducing the systemic clearance (CL) of antipyrine by 23% to 40%." These findings suggest that propafenone may inhibit the oxidative metabolism of substrates that do not go through the CYP2D6 pathway.

In addition, propafenone may further decrease glo- bal hepatic metabolic function by yet another mecha- nism. It has been documented that propafenone ex- hibits 13-adrenergic blocking properties at serum con- centrations above 0.500 Rg/ml. 2'3'8 B-Adrenergic blockers can significantly reduced the hepatic CL of drugs whose metabolic fate is highly dependent on he- patic blood flow."

Lidocaine is rapidly metabolized by way of oxida- tive N-deethylation, which has been related to the CYP3A4 isozyme.14 Given as a single dose, lidocaine has a high hepatic extraction ratio (62% to 81%), so that its metabolic CL is dependent on both oxidative metabolism and liver blood flow.15 Agents that inhibit oxidative metabolism of the hepatic cytochrome P450 system, decrease hepatic blood flow, or both are known to decrease the systemic CL of lidocaine.".16 Given the unique metabolic characteristics of pro- pafenone described above, we hypothesize that propafenone can significantly inhibit lidocaine metab- olism, thereby increasing lidocaine plasma concentra- tions. Because lidocaine has a narrow therapeutic to toxic plasma concentration ratio, we investigated the effects of propafenone on the disposition of lidocaine.

METHODS Subjects. Twelve male subjects between the ages of

22 and 40 years (mean weight, 82.7 8.1 kg) were recruited for this study. The study was approved by the Hartford Hospital Institutional Review Committee. After providing written informed consent, subjects had a complete physical examination and electrocar- diogram performed. A detailed medical history and laboratory tests (hematology and chemistry profiles) were also obtained. Subjects were included into the study if they were healthy, within 15% of ideal body weight, and nonsmokers. Subjects were excluded if they had any of the following: (1) known heart, lung, kidney, or liver diseases, (2) significant ECG abnor- malities, (3) abnormal chemistry or hematology serum tests, (4) abnormal findings on physical examination

Propafenone with lidocaine 39

(5) known allergies to lidocaine or related compounds (i.e., local anesthetics), (6) history of seizure disor- ders, (7) use of concurrent medications, including over the counter preparations, (8) history of alcohol or illicit drug abuse, or (9) participation in any drug study within the past 1 month.

Study design. This was a randomized, single pla- ceboblinded, two-way crossover study with a 3-week washout period. Approximately 1 week before the first study period, the subjects ingested 30 mg dex- tromethorphan (Benylin DM, 10 mg/5 ml, Parke Davis) to determine their CYP2D6 isozyme phenotype (poor metabolizer or extensive metabolizer). Urine samples were collected 4 hours later and frozen at 70° C for future analysis.

Subsequently, the 12 subjects were randomly as- signed to receive either placebo or oral 225 mg pro- pafenone every 8 hours for 6 days. On the evening of the fourth day of propafenone or placebo therapy, the subjects were admitted to an ECG-monitored hospital bed and had one intravenous catheter placed into each forearm. The subjects then fasted for 12 hours (over- night) and were forbidden to consume caffeine- or al- cohol-containing beverages. The diet, following fast- ing, was low in protein (. 45 gm/day) and fat ( .30 gm/day) to control for food-induced changes in liver blood flow." In addition, the subjects remained sed- entary for the study duration and were not allowed to ambulate (except for restroom use) to limit posture- or exercise-induced changes in liver blood flow. I 8 On the morning of the fifth day, the subjects received a lido- caine loading dose (1.5 mg/kg over 10 minutes) intra- venously through an indwelling venous catheter, which followed the scheduled morning dose of pro- pafenone or placebo. Immediately after the loading dose, a continuous infusion of lidocaine (2 gm in 500 ml of 5% dextrose in water) was started at 2 mg/kg/hr and continued for 22 hours. Plasma samples were ob- tained from the contralateral arm at time 0 (baseline) before propafenone and lidocaine administration and at the following times after administration: 1, 2, 4, 8, 10, 12, 16, 18, 22, 22.17, 22.33, 22.5, 22.75, 23, 24, 25, 26, 28, 30, 32, and 34 hours, timed from the beginning of the lidocaine loading dose. Serial ECG and blood pressure measurements were performed be- fore the propafenone or placebo morning dose then every 2 (approximate peak) and 8 (trough) hours after the dose until the end of the study at time 34 hours.' Propafenone or placebo therapy was continued until day 6, and the last dose was at time 24 hours.

For the duration of the study period, the subjects were carefully monitored and questioned in a non-

directed manner about adverse reactions (the subjects were blinded to treatment throughout the entire study duration). Adverse events were scored from 0 to 4: 0

for no side effects; 1 if side effects occurred only dur- ing the lidocaine loading dose and dissipated within 1

hour; and 2, 3, or 4 if the effects were mild, moder- ate, or severe, respectively, and persisted beyond the first hour of lidocaine.

The 2 weeks after subjects were discharged from the hospital were a washout period. On the third week the subjects were crossed over to the second part of the study to receive the alternate therapy and to have the same procedures repeated.

Drug analysis. All blood samples collected were centrifuged within 60 minutes of collection. The se- rum was harvested and stored at 70° C until analy- sis. Assays for determination of propafenone and pro- pafenone's major metabolites (5-hydroxypropafenone and N-depropylpropafenone) were performed by use of HPLC (model 6000A, Waters Chromatography Div., Millipore, Milford, Mass.) with a Colica V60 (250 mm x 4.6 mm) column and ultraviolet variable wavelength detector set at 254 nm (LDC-Milton Roy, Riviera Beach, Fla.). Plasma samples were extracted by adding 1 p.g internal standard (111115, Knoll Phar- maceuticals, Whippany, N.J.) and 0.1 ml of 0.2N so- dium hydroxide to 1 ml plasma. The samples were then shaken (for 10 minutes) with 5 ml ether. After centrifugation (at 2000 rpm for 5 minutes), the or- ganic layer was eluted off and dried under air flow at 40° C. The dried solute was reconstituted with 100 ill of mobile phase (53% methane, 26% dichlo- romethane, 18% hexane, 3% deionized water, and 0.1% ammonia) and 50 ill was injected into the HPLC system. Lidocaine serum concentrations used a HPLC method that has been described previously.19 Intraday and interday coefficients of variation were less than 8% for lidocaine, propafenone, 5-hydroxypropafe- none, and N-depropylpropafenone at 5.00 and 0.20, 1.00 and 0.200, 1.00 and 0.200, and 0.500 and 0.100 jig/ml, respectively. The urine concentration of dex- tromethorphan and its major metabolite dextrorphan were determined by use of an HPLC assay as de- scribed previously 20

Pharmacokinetic analysis. Pharmacokinetic param- eters for lidocaine and propafenone, including CL0

and CL, mean residence time (MRT), volume of dis- tribution at steady state (Vss), and area under the plasma concentrationtime curve (AUC), were deter- mined by use of model-independent pharmacokinetics. AUC for lidocaine was calculated by use of the trape-

zoidal method, with extrapolation from the last con- centration time point to infinity using the terminal elimination rate constant. AUC for propafenone was calculated by use of the trapezoidal method from t = 0 to t = 34. The elimination rate constant of lidocaine was calculated from a least-squares regression of the terminal portion of the lidocaine concentrationtime profile. MRT corrected for infusion duration and V55 of lidocaine were calculated from the following equa- tions: (AUMC/AUC) (T/2), and (Dose AUMC/ AUC2) (Dose T/2 AUC), respectively, where T is equal to the infusion duration of lidocaine (22 hours), Dose is equal to the total dose infused (45.2 mg/kg), and AUMC is the area under the moment time curve from zero extrapolated to infinity. Lido- caine CL was calculated as the quotient of lidocaine dose and AUC(0-00). Propafenone apparent CL0 was calculated as the quotient of propafenone dose (225 mg) and AUC of the 24- to 32-hour dosing interval. Lidocaine concentration at steady state was the mean concentration of the 18- and 22-hour time points, where the difference between these points was less than 10%. The ratio of propafenone to 5-hydroxypro- pafenone plasma AUC values were calculated and used to estimate the activity and affinity of CYP2D6 isozyme for propafenone .21

Blood pressures were obtained by use of an auto- mated inflation cuff (Critikon Inc., Tampa, Fla.). ECG intervals (PR, QRS, JT, and RR) were measured by a digitizing pad interfaced with a computer pro- gram (Sigma Scan, Jandel Scientific, Corte Madera, Calif.). The reported intervals were an average of three consecutive beats measured from either leads II, aVF or V5 at a recording paper speed of 50 mm/sec.

Extensive metabolizers were classified by a dextro- methorphan/dextrorphan urine molar ratio 0.30.20 Lidocaine pharmacokinetic parameters during pro- pafenone were compared with placebo by use of two- way ANOVA, incorporating subjects, treatment, and treatment order into the statistical model. Only sub- jects phenotyped as extensive metabolizers were in- cluded in the statistical analysis of the pharmacoki- netic parameters to maintain a homogenous population with respect to hepatic metabolism. The change in CL and Vss from placebo to propafenone were related to propafenone average steady-state serum concentra- tions by use of regression and correlation analysis. ECG and blood pressure values between treatment groups were compared by use of two-way ANOVA with repeated measures. The maximum change in ECG intervals (PR, QRS, JTc, and RR) from placebo

CLIN PHARMACOL THER

40 Ujhelyi et al. JANUARY 1993

VOLUME 53 NUMBER 1

10

0.1

PROPAFENONE

0 PLACEBO

Propafenone with lidocaine 41

to propafenone was related to the propafenone con- centration at the respective time point by use of re- gression and correlation analysis. The degree of ad- verse effects between the treatment groups were compared by use of the Wilcoxon signed-rank test. All parametric data are expressed as mean values ± SD. The level of statistical significance was set at a p value of 0.05 with use of a two-tailed test.

RESULTS A total of 14 subjects were enrolled into the study.

One subject was dropped from the study because of noncompliance with study protocol. The other subject requested to be withdrawn from the study because of intolerable central nervous system side effects (dizzi- ness and paresthesia) during the lidocaine propafenone combination. Therefore, the data from 12 subjects are presented.

Pharmacokinetics. One of the 12 white subjects (8%) was classified as a poor metabolizer (dextro- methorphan/dextrorphan molar ratio of 1.8). Fig. 1

depicts the lidocaine concentrationtime curve during placebo and propafenone. In the subjects who were extensive metabolizers, propafenone produced a small

but statistically significantly reduction in systemic CL of 7% (-0.74 ± 1.05 ml/min/kg; p s 0.05) with a concomitant small increase in AUC(0-00) of 7% (5.0 ± 6.6 lig hr/ml; p 0.05) compared with pla- cebo (Table I). Propafenone also decreased Vss by 6% (-0.16 ± 0.27 L/kg; p = 0.10), but statistical significance was not achieved. Propafenone had no ef- fect on MRT. The observed changes in CL and Vss did not correlate with propafenone average steady- state concentrations (r2 s 0.02; difference not signifi- cant). Systemic lidocaine CL between placebo and propafenone study periods decreased or did not change in 10 extensive metabolizers by 0% to 25%, whereas it increased in one extensive metabolizer by 8% (Fig. 2). In the subject who was a poor metabo- lizer, lidocaine CL and Vss increased by 21% and 59%, respectively (Fig. 2).

During lidocaine treatment, the AUC, CL., and steady-state trough concentration of propafenone ranged from 4.55 to 34.67 lig hr/ml, 381 to 2919 ml/min/70 kg, and 0.058 to 0.875 i.tg/ml, respectively (Table II). The subject who was a poor metabolizer had the lowest oral propafenone CL and highest AUC and steady-state trough concentration (Table II and

0 5 10 15 22 24 26 28 30 32 34

TIME (hr)

Fig. 1. Lidocaine concentrationtime profile during the placebo and propafenone study phases.

42 Ujhelyi et al.

15 r-

14

13

12

9

7

Fig. 3). In two extensive metabolizers, propafenone elimination was not first order or log linear. These two extensive metabolizers, who displayed nonlinear pro- pafenone pharmacokinetics (dextromethorphan/dex- trorphan molar ratios of 0.165 and 0.025), had similar propafenone and metabolic concentration-time pro- files as the subject who was a poor metabolizer (Fig. 3).

3.5 -

1.5 / PL PROP

CLIN PHARMACOL THER JANUARY 1993

AUC(0-.0), Area under the plasma concentration-time curve from time 0 to infinity; CL, total body clearance; MRT, mean residence time; Vss, steady-state volume of distribution; Css, steady-state plasma concentration.

*Mean ± SD. tp 0.05. tp = 0.10.

Pharmacodynamics. Average systolic and diastolic blood pressures during the placebo-lidocaine phase were 136 -± 10 and 85 ± 9 mm Hg, respectively, which was not significantly different from the values during the propafenone-lidocaine phase of 135 ± 10

and 83 -± 9 mm Hg, respectively (difference not significant). The propafenone-lidocaine combination significantly prolonged the PR and QRS intervals

Extensive metabolizers (n = 11) Poor metabolizer (n = I) Placebo* Propafenone* Placebo Propafenone

AUC(0-cc) 76.3 ± 15.6 81.7 ± 16.21: 78.5 65.1 CL (ml/min/kg) 10.27 ± 2.24 9.53 ± 1.771: 9.59 11.56 MRT (hr) 4.47 ± 0.87 4.37 ± 0.92 3.07 4.05 Vss (L/kg) 2.64 -± 0.45 2.48 ± 0.33* 1.77 2.81 Cs, (14/m1) 3.38 ± 0.57 3.41 ± 0.54 3.45 2.53

Treatment Group Treatment Group

Fig. 2. Change in lidocaine phannacokinetic parameters from placebo (PL) to propafenone

(PROP) in the extensive metabolizers (open circles) and in the poor metabolizer (solid squares).

Table I. Pharmacokinetic parameters of lidocaine in the subjects who were extensive metabolizers and in the subject who was a poor metabolizer

6 )' PL PROP

VOLUME 53

NUMBER 1

-J

cp

6/ \

*AIN /

0.1 aNN 2 ---

*Mean ± SD.

on average by 10% and 15%, respectively, compared with the placebo-lidocaine combination (Fig. 4; p < 0.0001). Propafenone had no significant effect on the JTe or RR interval (difference not significant). The maximum change in PR interval from the placebo-lidocaine to the propafenone-lidocaine study periods did not correlate with propafenone concen- trations (r2 -= 0.09; difference not significant). In

Time (hr)

Fig. 3. Concentration-time profile of propafenone (open symbols) and 5-hydroxypropafenone (solid symbols) in the extensive metabolizers with linear pharmacokinetics (circles), the extensive metabolizers with nonlinear pharmacokinetics (triangles), and the poor metabolizer (squares).

Table II. Pharmacokinetic parameters of propafenone, 5-hydroxypropafenone, and N-depropylpropafenone at steady state in extensive metabolizers with linear and nonlinear pharmacokinetics and in the poor metabolizer

Extensive metabolizers

Propafenone with lidocaine 43

II

contrast, the maximum change in QRS interval sig- nificantly correlated with propafenone concentrations Ir2 = 0.39 (y = 9.85x + 12.5); p = 0.031 (Fig. 5).

Two subjects who had the highest propafenone con- centrations (one poor metabolizer and one extensive metabolizer) experienced adverse effects to pro- pafenone (dysgeusia, paresthesia, and lightheaded-

Propafenone CL (ml/min/70 kg) 1668 ± 659 712 ± 160 381 Propafenone AUC (p.,g hr/ml) 9.58 ± 3.87 22.88 ± 3.9 34.67 5-Hydroxypropafenone AUC hr/ml) 4.66 ± 1.68 1.85 0.67 3.13 N-Depropylpropafenone AUC (p,g hr/ml) 0.83 ± 0.68 0.99 ± 0.10 1.44 Propafenone AUC/5-hydroxypropafenone AUC 2.15 ± 0.72 12.83 ± 2.57 11.07

Linear Nonlinear pharmacokinetics pharmacokinetics

(n = 9)* (n = 2)* Poor metabolizer (n = I)

0 5 10 15 20 25 30 35

ness) before the lidocaine phase, whereas one subject (an extensive metabolizer) experienced adverse effects to placebo (lethargy and decreased exercise tolerance). Most subjects experienced adverse effects to lidocaine during both placebo and propafenone phases. Lido- caine adverse effects were primarily related to the cen- tral nervous system (lightheadedness, dizziness, pares- thesia, lethargy, and somnolence). The central nervous system adverse effects observed during the placebo- lidocaine phase were almost solely related to the bolus infusion and had a median severity rank of 1

(range, 1 to 2), which was significantly less than the median rank of 2 (range, 0 to 4) occurring during the propafenone-lidocaine phase (p = 0.02). The most se- vere central nervous system effects during pro- pafenone-lidocaine phase were observed in the same subjects (ranks of 3 to 4) who had mild side effects to propafenone alone.

DISCUSSION Studies have shown higher efficacy rates with com-

bination antiarrhythmic therapy compared with single agents.22 Thus it is possible that lidocaine and pro-

CLIN PHARMACOL THER JANUARY 1993

pafenone will be used concomitantly. It is expected that the concomitant use of these agents would be used to treat refractory arrhythmias or during the pro- cess of converting intravenous lidocaine therapy to oral propafenone therapy. With the potential of pro- pafenone to inhibit the oxidative metabolism of sev- eral substrates and the narrow therapeutic to toxic ra- tio of lidocaine, we investigated the pharmacokinetic and pharmacodynamic interaction potential between these two agents.

Lidocaine pharmacokinetics. The data from this study suggest that propafenone minimally impairs the metabolism of lidocaine. Overall, less than a 10% de- crease in systemic CL of lidocaine was observed dur- ing concomitant propafenone administration. The re- duction in CL occurred with a concomitant decrease in Vss. Others have reported similar findings with cimet- idine and propranolol. Cimetidine, a known inhibitor of global oxidative metabolism, decreased the CL and Vss of single-dose and continuously infused lidocaine by 14% to 30% and 7% to 21%, respectively.16,23-25 Propranolol, known to reduce liver blood flow, de- creased lidocaine CL by a magnitude similar to that of

44 Ujhelyi et al.

210

195

180

165

\ 150

u) QRS Propafenone

0 135 QRS Placebo

a) PR Propafenone 120

0 PR Placebo

CD 105 1.1.1

90

75 -

0 5 10 15 20 25 30

Time (hr)

Fig. 4. The QRS and PR intervals versus time profile during propafenone and placebo study phases.

VOLUME 53 NUMBER 1

50

45

ao

35

30

25

20

15

100

35

30

25

20

15

10

5 0

cimetidine when lidocaine was given as a single dose or as a continuous infusion, but it did not influence li- docaine Vss.13

Long-term administration of lidocaine (>20 hours) yields a significant reduction in hepatic extraction of lidocaine by 30% to 50% and thus systemic clear- ance. 13,15,25,26 Because hepatic extraction is reduced with long-term administration, it is expected that the metabolism of lidocaine will be more sensitive to he- patic enzyme inhibition but may still be affected by changes in liver blood flow.13 Therefore two possible mechanisms exist in which propafenone could affect the metabolism of lidocaine: (1) inhibition of oxida- tive metabolism and (2) a reduction in global liver blood flow. Propafenone is known to decrease the CL of several hepatically metabolized substrates, includ- ing metoprolol, propranolol, debrisoquin, antipyrine, and warfarin.2.9- 12 The first three agents are substrates for the CYP2D6 isozyme, whereas the latter two agents are oxidatively metabolized through other met- abolic pathways, suggesting that propafenone may in- hibit several metabolic pathways.

Propafenone with lidocaine 45

Propafenone also possesses B-adrenergic blocking properties. B-blockers, such as propranolol, are known to reduce lidocaine CL. /3 This is most likely attributable to a propranolol-induced reduction in car- diac output and subsequent decline in liver blood flow.27 However, the potency of propafenone as a 13-blocker is approximately one-twentieth of that seen with propranolo1.3 To produce a 10% reduction in the maximal exercise heart rate, propafenone serum con- centrations above 1.0 itg/m1 are needed.3 In this study three subjects (two extensive metabolizers and one poor metabolizer) had propafenone concentrations near or above 1.0 1.i.g/ml, which were three to four times greater than the other subjects. However, pro- pafenone concentrations did not correlate with changes in lidocaine CL, where the three subjects with the highest propafenone concentrations had either a small reduction (4% to 10%) or an increase (21%) in lidocaine CL.

On the basis of this limited data, we postulated that propafenone would inhibit the metabolism of lido- caine. It was also expected that the reduction in lido-

0.2 0.4 0.6 0.8 1 1.2 1.4 16 Propafenone (ug/mL)

Fig. 5. Regression analysis of the maximum change in PR and QRS intervals in relationship to propafenone concentrations.

-

R2=0.09, p=N8

0.2 0.4 1 0.6 0.8 1.2 1.4 16

caine CL would be greater in subjects with higher propafenone concentrations because greater 13-block-

ing effects would be evident. However, propafenone produced a small (albeit statistically significant) 7% reduction in lidocaine CL in subjects who were exten- sive metabolizers, whereas lidocaine CL was in- creased by 21% in the subject who was a poor metab- olizer. Unfortunately, the present study cannot de- termine the mechanism of this disparity between the extensive metabolizers and the one poor metabolizer. The findings of this study suggest the following:

the inhibitory affinity of propafenone for the he- patic isozyme(s) responsible for the metabolism of li- docaine is limited and is likely much lower than its in- hibitory affinity for the CYP2D6 isozyme, and/or

the 13-blocking effects of propafenone at the con- centrations seen are probably insufficient to produce a reduction in liver blood that would lead to a 10% or greater reduction in lidocaine CL.

Propafenone did not significantly alter the volume of distribution of lidocaine. It is possible that pro- pafenone may increase lidocaine distribution by de- creasing protein binding through competitive displace- ment because both propafenone and lidocaine are 60% to 85% bound to al-acid glycoprotein.14.23.28 How- ever, we observed a trend toward a lower lidocaine Vss during propafenone-lidocaine therapy, suggesting that this type of protein binding interaction is unlikely. Other agents that have been reported to decrease lido- caine CL either decreased (cimetidine) or had no ef- fect on (propranolol) the Vss of lidocaine.13,23.24

Propafenone pharmacokinetics. It has been well documented that propafenone exhibits nonlinear phar- macokinetics. Saturation of the metabolism of pro- pafenone can occur at doses as low as 300 mg/day, al-

though it is not commonly seen until daily doses exceed 600 Mg.3'6'8 Our data showed that saturation of this enzyme occurred at a daily dose of 675 mg/day in two subjects who were phenotyped with dextro- methorphan as extensive metabolizers. With saturation of the CYP2D6 isozyme, the metabolic molar ratio of propafenone/5-hydroxypropafenone increases to val- ues observed in poor metabolizers.21 Therefore the two extensive metabolizers of dextromethorphan ap- peared as poor metabolizers of propafenone.

The disposition of propafenone in the remaining nine poor metabolizers during lidocaine was similar to previous reports of propafenone alone.2-4.24 It appears that lidocaine had little effect on the disposition of propafenone, although propafenone trough concentra- tions in the present study appeared to be lower during

the lidocaine infusion compared to before (time 0

hour) and after (time 34 hour) lidocaine (Fig. 3). However, it is impossible to determine with certainty the influence of lidocaine on the metabolism of pro- pafenone in this study because our study was not spe- cifically designed to elucidate such changes.

Pharmacodynamics. Pharmacodynamic changes from the placebo-lidocaine phase to the propafenone- lidocaine phase were limited to increases in PR and QRS intervals. On average, the PR and QRS intervals over the 34-hour study period were increased during the propafenone-lidocaine phase by a magnitude simi- lar (10% to 20%) to that reported by other investiga- tors with propafenone alOne.2'8'29'30 The PR and QRS intervals before lidocaine (time 0 hour) were similar to the intervals during steady-state lidocaine administra- tion (time 18 hour) and after lidocaine administration (time 34 hour). Therefore the changes in ECG values are likely attributable to propafenone, with minimal contributing effects from lidocaine.31 The maximum change in QRS interval was related to propafenone concentrations, although this relationship had a con- siderable amount of variability that may be attribut- able to the active metabolite. Others have reported similar findings.2'29

Another pharmacodynamic effect of propafenone observed in this study was adverse central nervous system reactions. Both subjects who complained of side effects on propafenone before the lidocaine phase of the study had the highest average steady-state pro- pafenone concentrations (1.02 and 0.762 ig/m1). Other reports have described similar findings.2 More important, however, was the potentiation of disturbing central nervous system adverse effects during concom- itant propafenone and lidocaine therapy. The intensity of this pharmacodynamic interaction resulted in the withdrawal of one subject (not included in data re- sults) from the study. Because propafenone and lido- caine can elicit adverse central nervous system effects by themselves, it is not surprising that the concomitant use of these agents magnifies these adverse reactions. Interestingly, this pharmacodynamic interaction was not related to changes in lidocaine concentrations be- tween the placebo and propafenone phases. To our knowledge this is the first report describing this inter- action, although the significance of this observation may be limited by the small sample size.

The findings of this study show that oral pro- pafenone has little effect on the disposition and metab- olism of continuously infused lidocaine. The slight decrease in systemic lidocaine CL produced by pro-

CLIN PHARMACOL THER

46 Ujhelyi et al. JANUARY 1993

VOLUME 53

NUMBER I

pafenone is clinically insignificant. This suggests a limited ability for propafenone to inhibit metabolic pathways that do not involve the CYP2D6 isozyme. However, the occurrence of disturbing central nervous system side effects are increased in frequency and se- verity when these antiarrhythmic agents are combined. Therefore this antiarrhythmic combination may be poorly tolerated by some persons.

We thank Carol Campbell, CCRN, Linda Freeman- Bosco, CCRN, and Ellen Gleason, CCRN, for expert nurs- ing and technical assistance, as well as Michael B. Bottorff, PharmD, for his critical review and suggestions.

References Chow MSS, Lebsack C, Hilleman D. Drug review. Propafenone: a new antiarrhythmic agent. Clin Pharm 1988;7:869-77. Siddoway LA, Thompson KA, McAllister B, et al. Polymorphism of propafenone metabolism and disposi- tion in man: clinical and pharmacokinetic conse- quences. Circulation 1987;75:785-91. Lee TJ, Kroemer HK, Silberstein DJ, et al. The role of genetically determined polymorphic drug metabolism in the beta- blockade produced by propafenone. N Engl J Med 1990;322:1764-8. Boriani G, Strocchi E, Capucci A, et al. Relationships between debrisoquine hydroxylation and propafenone pharmacokinetics. Drug Invest 1990;2:114-9. Funck-Brentano C, Kroemer HK, Pavlou H, Woosley RL, Roden DM. Genetically determined interaction be- tween propafenone and low dose quinidine: role of ac- tive metabolites in modulating net drug effect. Br J Clin Pharmacol 1989;27:435-44. Giani P, Landolina M, Giudici V, et al. Pharmacokinet- ics and pharmacodynamics of propafenone during acute and chronic administration. Eur J Clin Pharmacol 1988;34:187-94. Capucci A, Boriani G, Marchesini B, et al. Minimal ef-

fective concentration values of propafenone and 5-hy- droxypropafenone in acute and chronic therapy. Cardio- vase Drugs Ther 1990;4:281-7. Zoble RG, Kirsten EB, Brewington J, the Propafenone Research Group. Pharmacokinetic and pharmacody- namic evaluation of propafenone in patients with ven- tricular arrhythmia. CLIN PHARMACOL THER 1989;45:535- 41

Kowey PR, Kirsten EB, Fu CHJ, Mason WD. Interac- tion between propranolol and propafenone in healthy volunteers. J Clin Pharmacol 1989;29:512-7. Wagner F, Kalusche D, Trenk D, Jahnchen E, Roskamm H. Drug interaction between propafenone and metoprolol. Br J Clin Pharmacol 1987;24:213-20. Kates RE, Yee YG, Kirsten EB. Interaction between

Propafenone with lidocaine 47

warfarin and propafenone in healthy volunteer subjects. CLIN PHARMACOL THER 1987;42:305-11. Chong MT, Svensson CK. Dose-dependent effects of propafenone on in vivo drug metabolism in the rat. Res Commun Chem Pathol Pharmacol 1987;58:147-56. Ochs HR, Carstens G, Greenblatt DJ. Reduction in li- docaine clearance during continuous infusion and by coadministration of propranolol. N Engl J Med 1980;303:373-7. Bargetzi MJ, Aoyama T, Gonzalez FJ, Meyer UA. Li- docaine metabolism in human liver microsomes by cy- tochrome P450IIIA4. CLIN PHARMACOL THER 1989;46: 521-7. LeLorier J, Mosian R, Gagne J, Caille G. Effect of the duration of infusion on the disposition of lidocaine in dogs. J Pharmacol Exp Ther 1977;203:507-11. Bauer LA, Edwards WAD, Randolph RP, Blouin RA. Cimetidine-induced decrease in lidocaine metabolism. Am Heart J 1984;108:413-6. Svensson CK, Mauriello PM, Barde SH, Middleton E, Lalka D. Effect of carbohydrates on estimated hepatic blood flow. CLIN PHARMACOL THER 1984;35:660-5. Svensson CK, Mauriello PM, Barde SH, et al. Effect of carbohydrates on estimated liver blood flow. CLIN PHAR-

MACOL THER 1984;25:660-5. Chow MSS, Kluger J, DiPersio DM, Lawrence R, Fieldman A. Antifibrillatory effects of lidocaine and bretylium immediately postcardiopulmonary resuscita- tion. Am Heart J 1985;110: 938-43. Evans WE, Relling MV, Petros WP, Meyer WH, Mirro J, Crom WR. Dextromethorphan and caffeine as probes for simultaneous determination of debrisoquin oxidation and N-acetylation phenotypes in children. CLIN PHARMA-

COL THER 1989;45:568-73. Latini R, Belloni M, Bernasconi R, et al. Identification of propafenone metaboliser phenotype from plasma and urine excretion data. Eur J Clin Pharmacol 1992;42: 111-4. Luderitz B, Mletzko R, Jung W, Manz M. Combina- tion of antiarrhythmic drugs. J Cardiovasc Pharmacol 1991;17(suppl 6):S48-S52. Jackson JE, Bentley JB, Glass SJ, Fukui T, Gandolfi J, Plachetka JR. Effects of histamine-2 receptor blockade on lidocaine kinetics. CLIN PHARMACOL THER 1985;37: 544-8. Feely J, Wilkinson GR, McAllister CB, Wood AD. In- creased toxicity and reduced clearance of lidocaine by cimetidine. Ann Intern Med 1982;96:592-4. Powell JR, Foster J, Patterson JH, Cross R, Wargin W. Effect of duration on lidocaine infusion and route of ci- metidine administration on lidocaine pharmacokinetics. Clin Pharm 1986;5:993-8. Bauer LA, Brown T, Gibaldi M, et al. Influence of long-term infusions on lidocaine kinetics CLIN PHARMA-

COL THER 1982;31:433-7.

Branch RA, Shand DG, Wilkinson GR, Nies AS. The reduction of lidocaine clearance by d/-propranolol: an example of hemodynamic drug interaction. J Pharmacol Exp Ther 1973;184:515- 9. Harron DWG, Brogden RN. Propafenone. A review of its pharmacodynamic and pharmacokinetic properties and therapeutic use in the treatment of arrhythmias. Drugs 1987;34:617-47. Haefeli EW, Vozeh S, Ha HR, Follath F. Comparison

of the pharmacodynamic effects of intravenous and oral propafenone. CL1N PHARMACOL THER 1990;48:245-54. Pritchett ELC, Smith WM, Kirsten EB. Pharmacoki- netic and pharmacodynamic interactions of propafenone and cimetidine. J Clin Pharmacol 1988;28:619-24. Feld GK, Nademanee K, Singh BN, Kirsten E. Hemo- dynamic and electrophysiologic effects of combined in- fusion of lidocaine and propafenone in humans. J Clin Pharmacol 1987;27:52-9.

Bound volumes available to subscribers

Bound volumes of CLINICAL PHARMACOLOGY & THERAPEUTICS are available to subscribers (only) for the 1993 issues from the Publisher, at a cost of $52.50 for domestic, $69.18 for Canadian, and $65.50 for international subscribers for Vol. 53 (January-June) and Vol. 54 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue of the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact MosbyYear Book, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, USA; phone (800)325-4177, ext. 4351, or (314)453-4351.

Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.

CLIN PHARMACOL THER

48 Ujhelyi et al. JANUARY 1993