Abstract
Introduction
Tizanidine, approved by the Food and Drug Administration (FDA) in 1996, is often used as an antispastic agent when oral treatment is indicated. Tizanidine is an imidazoline derivative with activity at both spinal and supraspinal levels. The exact mechanism of action has not been fully elucidated, but its pharmacodynamic effects are primarily linked to its central α2-adrenoceptor agonist properties, although its imidazoline receptor binding may play a role.1,2 Tizanidine appears to act predominantly presynaptically in the spinal cord by reducing release of the excitatory amino acid glutamate and aspartate from the presynaptic terminal of spinal interneurons and it may facilitate the action of the inhibitory neurotransmitter glycine. 3
There are relatively few effective and approved alternatives in the United States that may be used as oral antispastic agents. It is therefore a promising agent with regard to both its efficacy and tolerability, when compared to other myorelaxant drugs.
It was reported that tizanidine decreases blood pressure (BP) and heart rate (HR) in animals1,4,5 and humans,2,6 an effect which has been possibly linked to its imidazoline receptor binding.4,5 Tizanidine has a narrow therapeutic index, often making optimal patient dosing difficult. 7
Interestingly, abnormalities in the electrocardiography (ECG) were reported in several cases. Luciani et al 8 described a case of tizanidine poisoning (120 mg: suicidal attempt), in which significant electrocardiographic changes were observed. The ECG showed sinus bradycardia 34 beats/min and PQ interval of 240 ms (atrioventricular [AV] block). 8 Bes et al 9 reported a case of ECG changes during treatment with tizanidine (bradycardia and ventricular extrasystoles), which were assumed to be due to severe hypokalemia (2 mmol/L) associated with diuretic treatment. Moreover, several cases of bradycardia were noted with tizanidine,10,11 including 1 accompanied with acute right heart failure 12 and a number of syncopes9,10 and sudden death. 13 Recently, a fatal torsades de pointes–related cardiac arrest in a 27-year-old woman was reported. Tizanidine was identified as a culprit drug in that case, along with azithromycin. 14
According to the classification of the University of Arizona Center for Education and Research on Therapeutics (CERT), tizanidine was classified as “A drug that may prolong the QT interval but at this time lacks substantial evidence for causing Torsades de Pointes” (www.torsades.org).
As excessive QT prolongation was clearly associated with an increased risk of triggering Torsades de pointes (TdP), the aim of our study was to evaluate the effects of tizanidine on cardiac ventricular repolarization. The present study was carried out at 3 levels; in vitro (cellular), ex vivo (isolated heart), and in vivo (whole animal).
Methods
Experiments were performed in accordance with our institutional guidelines on animal use in research. Animals were housed and maintained in compliance with the Guide to the Care and Use of Experimental Animals of the Canadian Council on Animal Care.
Patch-Clamp Experiments: Cell Culture, Transfection, and Whole Cell Voltage-Clamp Recordings
Experiments were performed on either human ether-a-go-go-related gene (HERG) stably transfected in human embryonic kidney cells HEK293 to recapitulate the rapid potassium (IKr) current or in Chinese hamster ovary (CHO) cells transiently transfected with 2 μg (each) of KCNQ1+KCNE1 complementary DNAs (cDNAs) to recapitulate the slow potassium (IKs) current. The CHO cells were transfected using the calcium–phosphate method. Green fluorescent protein (GFP) was coexpressed to assess transfection efficiency and to identify expressing CHO cells. HEK293 cells were maintained in minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS, Invitrogen, Burlington, ON, Canada), 1% penicillin–streptomycin (Invitrogen), 1%
Recordings were performed on either HEK293 or CHO cells in 35-mm Petri dishes mounted on stage of an inverted microscope (Olympus IX51). Currents were recorded in the whole cell configuration of the patch-clamp technique using an Axopatch 200A amplifier (Molecular Devices-Axon Instruments, Union City, California). Voltage-clamp was controlled by the pCLAMP software package (version 9.0, Molecular Devices-Axon Instruments). Cells were superfused with the bath solution containing (mmol/L): 145 NaCl, 4 KCl, 1.8 CaCl2, 1 MgCl2, 10 glucose, and 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid ([HEPES] pH 7.35, adjusted with NaOH). Micropipettes were pulled from borosilicate glass capillaries (Warner Instruments, Hamden, Connecticut) on a horizontal patch electrode puller and heat polished to obtain a tip resistance between 1 and 3 MΩ when filled with the following intracellular solution containing (mmol/L): 110 KCl, 1 MgCl2, 5 1,2-Bis(2-aminophenoxy) ethane-N’,N’,N’,N’-tetraacetic acid tetrapotassium salt (BAPTA-K4), 5 K2 adenosine triphosphate (K2ATP), and 10 HEPES (pH 7.2, adjusted with KOH). The liquid junction potential between the patch pipette and the bath solution was corrected by −5 mV. The recordings were made 10 minutes after obtaining the whole cell configuration to allow the currents to stabilize and the contents of the patch electrode to diffuse adequately. Currents were generated from a holding potential of −80 mV, using either 1-second (HERG) or 5-second (KCNQ1+KCNE1) depolarizing steps from −40 to +60 mV in 10 mV increments, and tail currents were measured at −40 mV. The series resistance was compensated >80% to improve whole cell voltage-clamp measurements. Currents were filtered at 1 kHz using a 4-pole Bessel filter (−3 dB/octave) and sampled at 2 kHz. All experiments were performed at room temperature (22°C-23°C).
Tizanidine hydrochloride solution (0.1-100 μmol/L) was prepared daily by dissolving required amount of the drug (Sigma-Aldrich, St Louis, Missouri) in a sample of the Tyrode solution perfusing the cells. Tizanidine IC50 on HERG current was estimated using tail current maximal amplitude measured at −40 mV after a voltage step to +20 mV (n = 6 cells, each concentration), normalized to baseline, plotted as a function of tizanidine concentration, and fitted to the Hill equation.
Student paired t test was performed on the magnitude of tizanidine effect on the maximal tail current amplitude of HERG or KCNQ1+KCNE1 currents using statistical tools in SigmaPlot (Jandel Scientific Software, San Rafael, California). Differences were considered significant at a P value <.05.
Langendorff Retroperfusion Experiments
Male Hartley guinea pigs (Charles River Laboratories, Montréal, QC, Canada) weighing 250-350 g were anticoagulated by intraperitoneal (ip) injection of heparin sodium (400 IU). Thirty minutes later, animals were killed by cervical dislocation, and the hearts were rapidly extirpated and immersed in cold (4°C) Krebs-Henseleit buffer containing (in mmol/L) glucose 5, KCl 4.7, CaCl2 1.2, NaHCO3 25, NaCl 118.5, MgSO4 2.5, and KH2PO4 1.2. This solution was continuously gassed with 95% oxygen plus 5% carbon dioxide (pH 7.4, 37°C).
Each heart was cannulated and retrogradely perfused via the aorta with the buffer at a constant pressure equivalent to 70 mm Hg, using a custom-made Isolated Heart IH-SR double warming coil heart perfusion system from Hugo Sachs Elektronik-Harvard Apparatus, March-Hugstetten, Germany. Hearts were electrically stimulated at a basic pacing cycle length (BCL) of 250 ms (4 Hz; approximately the natural sinus rate for guinea pigs and the slowest possible to work with when one wishes to avoid AV node ablation) with a small coaxial stimulation electrode connected to a programmable stimulator module. Monophasic action potential (MAP)-tip recording electrodes were securely positioned on the surface (epicardium) of each ventricle to obtain visually adequate signals (amplitude >25 mV, stable phase 4). Both MAP signals were continuously recorded (digital sampling rate, 1 kHz), along with perfusion pressure, and stored on hard disk for analysis. These values were averaged by the use of a routine designed specifically for this purpose and incorporated in the analysis algorithm of the ISOHEART software package from Hugo-Sachs Elektronik. At least 12 complexes were used for each measurement. Monophasic action potential signals were recorded at a BCL of 250 ms. Then, to assess rate dependency, BCL was changed to 200 ms, and the heart was paced for 1 minute before the MAP was recorded. Thereafter, perfusion was performed with buffer containing tizanidine 1µmol/L for a period of 30 minutes at a BCL of 250 ms. To assess rate-dependent effects of the drug, MAP signals were recorded again at BCL of 200 ms. Perfusion with buffer containing no drug was then restarted to assess reversibility of drug effects.
Student paired t test was performed on the magnitude of tizanidine effect on monophasic action potential duration measured at 90% repolarization (MAPD90) at BCL of 250 and 200 ms, using statistical tools in SigmaPlot (Jandel Scientific Software). Differences were considered significant at a P value < .05.
Wireless Cardiac Telemetry Experiments
Male Hartley guinea pigs (Charles River Laboratories, Montréal, QC, Canada) weighing 250-350 g (n = 6 animals) were surgically implanted with wireless cardiac telemetry radio transmitters (Model TAIICTA-F40, Data Sciences International [DSI], St. Paul, Minnesota). Aseptic surgery was performed to implant a telemeter in the peritoneal cavity of each animal, following the general procedures recommended by the manufacturer (DSI). Animals were anesthetized by isoflurane inhalation (4 L/min of isoflurane 3% to induce anesthesia and 1 L/min to maintain it). A midline abdominal incision was made in the skin and muscle layers.
The body of the telemeter was laid gently onto the intestines and attached to the overlying muscle wall with nonabsorbable sutures. The muscle wall was then closed with absorbable sutures. The telemeter is equipped with two leads for sensing the heart’s electrical activity; these leads ran through small punctures made in the muscle wall to exteriorize them from the peritoneal cavity. A small trochar was slid between the skin and muscle of the upper abdomen and chest to form 2 narrow subdermal tracks for the leads to lie in. The final position of the telemeter leads was with the negative lead tip near the right shoulder and the positive lead tip below the left axilla, on the fifth left rib; this simulates a conventional lead II ECG. The abdominal skin was then closed with surgical staples, and the animals were allowed to recover for 2 weeks. Immediately after the surgery, animals were administered ketoprofen 1 mg/kg once daily subcutaneously (SC) for 3 days to reduce postoperative pain and inflammation. They were also administered a single dose of enrofloxacin 2.5 mg/kg SC to prevent perioperative infections.
Electrocardiographic data were collected continuously using the Dataquest A.R.T. acquisition system (version 4.1) from DSI. Continuous recording was started 90 minutes prior to the ip injection of tizanidine 0.25 mg/kg (dissolved in saline solution), and continued up to 24 hours after the injection. Electrocardiographic signals were sampled at 1000 Hz. Both RR and QT intervals were analyzed using the Ponemah electrophysiology platform (DSI) using the ECG module. QT was corrected using the Van de Water formula (QTcvW = QT − 0.087 × [RR − 1000]), which has been shown to be more “conservative” for correcting the QT interval in conscious guinea pigs and dogs15,16 than the Bazett or the Fridericia formulas, which both tend to overcorrect the QT at rapid heart rates.
Student paired t test was performed on the magnitude of tizanidine effect on the prolongation of the QTcvW using statistical tools in SigmaPlot (Jandel Scientific Software). Differences were considered significant at a P value <.05.
Results
Left upper panels of Figure 1 show typical currents elicited in a HERG-transfected HEK293 cell under baseline conditions and after a 10-minute exposure to tizanidine 1 µmol/L. In this cell, the drug caused a ∽20% to 25% reduction of tail currents. Lower panel of Figure 1 shows concentration dependence of the effect of tizanidine on HERG tail current. The estimated IC50 for tizanidine on HERG tail current was >100 µmol/L. Right upper panels of Figure 1 show typical currents elicited in a KCNQ1+KCNE1-transfected CHO cell under baseline conditions and after a 10-minute exposure to tizanidine 1 µmol/L. In this cell, the drug caused a small ∽3% to 4% reduction of activating currents. The average reduction of KCNQ1+KCNE1 activating currents under tizanidine 1 µmol/L was 5.23% ± 4.54% (n = 5 cells, P > .05 vs baseline).
Left upper panels show typical currents elicited in a HERG-transfected HEK293 cell under baseline conditions and after a 10-minute exposure to tizanidine 1 µmol/L. Right upper panels show typical currents elicited in a KCNQ1+KCNE1-transfected CHO cell under baseline conditions and after a 10-minute exposure to tizanidine 1 µmol/L. Lower panel shows HERG tail current amplitude measured at
Figure 2 shows typical recordings of ventricular epicardial MAPs of Langendorff retroperfused isolated guinea pig hearts at baseline (solid tracing), and after a 30-minute exposure to tizanidine 1 µmol/L (dashed tracing). The effects of the drug on MAPD90 at BCL of 200 and 250 ms are shown in Table 1 . Tizanidine 1 µmol/L caused the typical reverse-rate–dependent prolongation of MAPD90 associated with IKr-blocking drugs. Indeed, the drug prolonged MAPD90 by 8.22 ± 2.03 ms (6.7%) and 11.70 ± 3.08 ms (8.5%) at BCL 200 and 250 ms, respectively (both P < .05 vs baseline).
Typical recordings of ventricular epicardial monophasic action potentials of Langendorff retroperfused isolated guinea pig hearts paced at a basic cycle length (BCL) of 200 ms, at baseline (solid tracing), and after a 30-minute exposure to tizanidine 1 µmol/L (dashed tracing).
Effects of Tizanidine 1 µmol/L on MAPD90 at Basic Pacing Cycle Lengths (BCL) of 200 and 250 ms
NOTE: MAPD90 = monophasic action potential duration measured at 90% repolarization; ms = millisecond.
a P < .05 vs baseline.
Upper panel of Figure 3 shows typical recordings of lead II ECG in conscious and unrestrained guinea pigs, before and 90 minutes after a single ip injection of tizanidine 0.25 mg/kg. In this animal, tizanidine caused a 11-ms prolongation of the QT interval. However, the drug also caused sinus bradycardia, as shown by a prolonged RR interval under tizanidine (from 292 ms at baseline to 324 ms under the drug). When the QT is corrected for the heart rate using the Van de Water formula (QTcvW = QT − 0.087 × [RR − 1000]), QTcvW increases from 212 ms at baseline to 223 ms, 90 minutes after a single ip injection of tizanidine 0.25 mg/kg. Lower panel of Figure 3 shows the individual effect of tizanidine 0.25 mg/kg ip on the QTcvW in conscious and unrestrained guinea pigs (n = 6) implanted with cardiac telemetry radio transmitters. On average (n = 6), the QTcvW was prolonged from 209.73 ± 1.67 ms at baseline to 221.65 ± 1.52 ms, 90 minutes after a single ip injection of tizanidine 0.25 mg/kg, giving an average maximal prolongation of the QTcvW of 11.93 ± 1.49 ms (P < .05 vs baseline). Not surprisingly tizanidine also showed a 38.00 ± 7.29 ms prolongation of the RR interval (heart rate slowing; P < .05 vs baseline). It also prolonged the QRS segment by a small 1.83 ± 0.48 ms (P < .05 vs baseline). In contrast, tizanidine had no significant effect on QRS amplitude, as summarized in Table 2 .
Upper panel shows typical lead II guinea pig ECG tracings recorded at baseline and 90 minutes after a single intraperitoneal injection of tizanidine 0.25 mg/kg. Vertical lines indicate where cursors were placed for QT interval measurements. Lower panel shows the individual effect of tizanidine 0.25 mg/kg ip on the QTcVW in conscious and unrestrained guinea pigs (n = 6) implanted with cardiac telemetry radio transmitters. ECG indicates electrocardiography; ip, intraperitoneal.
Maximal Effects of a Single 0.25 mg/kg ip Dose of Tizanidine on ECG Parameters Recorded in Guinea Pigs
NOTE: ECG = electrocardiograph; ip = intraperitoneal.
a P < .05 vs baseline.
Discussion
Our results indicate that tizanidine prolongs cardiac repolarization by blocking IKr. Our estimated IC50 for HERG is higher than 100 µmol/L. However, HERG current in heterologous expression systems does not equal IKr in native cardiac myocytes, with intact regulation and auxiliary subunits. Usually the IC50 in native cells tends to be significantly lower. Data on blood and tissue concentrations of toxic tizanidine doses are absent from the literature; however, peak plasma concentrations in humans following single and multiple doses do not exceed 0.025 mg/L, 17 which correspond to 86 nmol/L, and which seem to be far from the micromolar range. However, a recent postmortem analysis of tizanidine distribution in a 57-year-old woman who committed suicide revealed a tizanidine concentration of 2.34 mg/L in a heart blood sample. 18 This rather corresponds to 8.064 µmol/L. Our patch-clamp data show that tizanidine 1 µmol/L causes a 18.96% ± 2.13% reduction of HERG current amplitude. Moreover, our Langendorff retroperfusion data also show that tizanidine 1 µmol/L is sufficient to cause a 11.70 ± 3.08 ms increase in MAPD90 when pacing the hearts at a cycle length of 250 ms (4 Hz). More interestingly, our data also show that a single ip tizanidine dose of 0.25 mg/kg in guinea pigs causes a 11.93 ± 1.49 ms increase in the QTc.
In view of the present findings, it is always intriguing to realize that similar reductions of HERG current may prolong the QT interval to a different extent, as shown by Crumb et al. 19 For instance, 14% HERG blockade by olanzapine was reported to prolong the QTc by an average of only 1.7 ms. Comparable (15.1%) HERG blockade by haloperidol was rather shown to cause a 4 times higher (7.1 ms) average QTc prolongation. 19
Results obtained in this study clearly indicate that tizanidine possesses direct electrophysiological effects on a major ionic current involved in cardiac repolarization, namely IKr. Isolated heart experiments demonstrated correlated effects of the drug on ventricular muscle. Indeed, this is the first study demonstrating the basic mechanism by which tizanidine can prolong the QT interval.
The cardiac toxicity induced by tizanidine may be explained, on one hand, by increased plasma concentrations of the drug (due to either physiologically reduced clearance of tizanidine or inhibition of its metabolism by other drugs) and, on the other hand, by combined administration of action potential-lengthening agents. Another thing to consider with tizanidine is its own “heart-slowing effect.” Indeed, bradycardia is a well-known risk factor for excessive QT prolongation and triggering of TdP. Moreover, as we show in Table 1, tizanidine exhibits the typical reverse-rate–dependent MAPD-prolonging effect of IKr blockers, meaning increased MAPD prolongation as the heart rate decreases. Tizanidine may therefore promote its QT-prolonging effect via its own heart-slowing properties.
Although tizanidine’s effects are moderate and observable at high concentrations, paying attention to QTc may still be worthwhile. It is well known that tizanidine hydrochloride is a “small” molecule, with a molecular weight of 290.18 Da and a protein-binding ratio of approximately 30%. 11 The excretion route is 70% through the kidneys and the Cmax for a healthy adult taking 2 mg is reported to be 1.83 ng/mL. 11 However, in a patient with renal failure, the area under the curve is approximately 7 times that of a healthy adult, and a high Cmax (about twice) is found in pharmacokinetics studies. 11 According to FDA’s warnings, 20 in the case of renal impairment, tizanidine clearance is reduced by more than 50% in elderly patients with creatinine clearance <25 mL/min, compared to healthy elderly participants. This would be expected to lead to a longer duration of clinical effect. 20
As a matter of fact, many case reports suggest a propensity of tizanidine to cause arrhythmogenic effects after overdose.8,11,14 Moreover, drug metabolism studies have clearly demonstrated that CYP1A2 is the principal enzyme involved into the biotransformation of tizanidine in vitro and in vivo.21,22 Various drugs, such as ciprofloxacin, fluvoxamine, rofecoxib, as well as oral contraceptives containing ethinyl estradiol and gestodene, are well-known inhibitors of CYP1A2.22–25 Under conditions of decreased CYP1A2 activity, plasma concentrations of tizanidine are expected to rise significantly. Indeed, 3- to 12-fold increases in tizanidine plasma concentrations were noticed during the coadministration of drugs cited above.22–25 In fact, very recently, it was reported that coadministration of tizanidine and azithromycin, another potential QT-prolonging drug, was the likely cause of a fatal torsades de pointes–related syncope. 14
QT prolongation alone does not always explain TdP but with other risk factors, including hypokalemia, slow heart rates (which appear to be frequent with tizanidine), preexisting cardiac diseases (ventricular hypertrophy, heart failure, previous antiarrhythmic therapy), female gender, baseline QTc >0.46 seconds or coadministration of drugs that prolong the QT interval, may predispose to drug-induced TdP.26–28 The presence of risk factors, particularly hypokalemia, reduces the “repolarization reserve” of the human heart, and explains why some patients develop TdP after receiving IKr-blocking drugs for long periods of time, with no prior evidence of proarrhythmia.29,30 In that sense, tizanidine administration would probably be dangerous in patients with narrowed repolarization reserve, where it could represent the final “hit” on repolarization causing TdP.
Study Limitations
The guinea pig is an interesting animal model for studying drug-induced prolongation of cardiac repolarization and long QT syndrome, as it is the only rodent to express both IKr and IKs, as in the human heart. However, the relative contribution of IKs to cardiac repolarization in the guinea pig appears to be higher than in humans, especially at rest. Therefore, the use of the guinea pig for studying the MAPD- and QT-prolonging effect of drugs (which are almost always IKr blockers) may lead to underestimate the importance of the effects tizanidine may have on cardiac repolarization. Indeed, a strong guinea pig IKs may have compensated the loss of repolarization reserve incurred by tizanidine via IKr blockade.
Conclusions
Our study indicates that lengthening of cardiac repolarization is to be expected in patients during chronic treatment with tizanidine. Block of the rapid component of the cardiac delayed rectifier current (HERG), lengthening of MAPD90, and QTc prolongation were observed at supraclinical concentrations of the drug. Therefore, clinical attention to QT interval prolongation and triggered ventricular tachyarrhythmias (torsades de pointes) should be warranted when prescribing tizanidine in patients with impaired renal/liver function or those with congenital long QT syndrome or concomitant risk factors for QT prolongation such as electrolytes disturbances (hypokalemia, hypomagnesemia), bradycardia, and obviously, the combined use of other QT-prolonging drugs.
Footnotes
The author(s) declared no conflicts of interest with respect to the authorship and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This study was supported by a grant from the Corporation de l’Institut de cardiologie de Québec to BD. PV is the recipient of a Master studentship award from the Fonds d’Enseignement et de Recherche (FER) de la Faculté de pharmacie de l’Université Laval. BD and CS are recipients of Scholarship awards from the Fonds de la recherche en santé du Québec (FRSQ).