Antiarrhythmic Properties of Phenytoin

BACKGROUND: Phenytoin has long been used to treat epilepsy and for some time as an antiarrhythmic drug (AAD). It is known that the diastolic calcium leakage through dysfunctional cardiac ryanodine receptors (RyR2) is a mechanism for arrhythmias in heart failure. Recent evidence suggests that phenytoin inhibits dysfunctional RyR2, reduces the calcium leak during diastole in heart failure, and may improve cardiac systolic function. This indicates the potential for repurposing phenytoin as an AAD in patients with heart failure. METHODS: A systematic search of MEDLINE, Embase, and the Cochrane Library databases was performed in March 2019. The search was limited to the studies published in the English language from 1946 to 2019. Studies on the antiarrhythmic effects of phenytoin in adults compared to no treatment or other AADs were included. Studies were excluded if there was insucient clinical data regarding antiarrhythmic effects, dosing and administration of phenytoin and other ADDs. Conference abstracts, editorials, case studies and review articles were also excluded. RESULTS: A total of 157 non-duplicate titles were screened, and 25 articles underwent full-text review. 13 studies met the inclusion criteria, representing a total of 985 patients. Phenytoin was found to be effective in treating arrhythmias associated with digitalis toxicity, and in suppressing premature ventricular contractions (PVCs). In a recent animal study, phenytoin inhibited diastolic calcium leak through dysfunctional RyR2 in failing sheep hearts and improved cardiac systolic function without affecting normal functional RyR2. CONCLUSION: Phenytoin has an acceptable safety prole when used as an AAD. It has some utility in treating digitalis-induced arrhythmias and suppressing PVCs, however, further study is needed to determine its ecacy as an antiarrhythmic in heart failure patients given new evidence of its RyR2 stabilising properties. arrhythmia (VT, VF) while no AADs. Phenytoin was administered both orally (PO) and intravenously (IV) and phenytoin serum level in both responder and non-responder groups were within the therapeutic limits at the time of EPS (19.5 ± 4.7 µg/ml and 18.2 ± 4.8 µg/ml, respectively). Of the total 64 participants, only 11% of the patients had a negative EPS (responders: no inducible arrhythmias after phenytoin administration) and 89% had a positive EPS after phenytoin administration (non-responders). The responder group continued on maintenance phenytoin and non-responders were divided in 2 sub-groups to either receive AADs other than phenytoin or phenytoin either alone or in combination with other AADs. After a total of 3 years follow up, it was shown the long-term success was limited in the responder group. Only 3% of the study participants had responded to phenytoin therapy in both initial study and long-term follow up. In the non-responder group, long-term use of phenytoin either alone or combined with another AAD was shown to be ineffective. There was 71% (5 of the 7) mortality rate in the responders to the phenytoin therapy, but only 2 of 5 deaths were due to an arrhythmia, and the authors concluded that no death in this group could be attributed directly to phenytoin failure. This trial showed phenytoin is not an effective AAD for treating ventricular arrhythmias suppressing ventricular arrhythmias at EPS. this study had ventricular arrhythmias refractory 3 AADs in different on the atrial, AV nodal, and His-Purkinje system refractory periods. AH interval represented the AV nodal conduction time (the onset of the low atrial depolarization to the onset of the His deection), and HV interval represented the His-Purkinje conduction time (the onset of the His deection to the onset of ventricular depolarization). The patients received IV DPH (5–10 mg/kg, 100 mg/5 minutes). They concluded that DPH had no effect on His-Purkinje conduction time (HV interval) in both sinus rhythm (SR) and during a variety of paced atrial rates but the In another similar study by Damato et al. [10], 13 patients underwent EPS to identify the impacts of DPH on atrioventricular (PH interval) and intraventricular conduction (HQ interval) over different paced heart rates (HRs). They also assessed the ecacy of DPH in treating existing arrhythmias in the study participants (PVCs and atrial tachycardia). All patients in this study were on maintenance digitalis treatment. Phenytoin was administered IV in all patients (dose 250–750 mg) to either terminate the arrhythmia or reach the maximum dose of 1000 mg. There were no serious side effects requiring alteration of the treatment. DPH was found to be effective in reducing burden of PVCs. It also enhanced AV conduction (shorten the PH interval) over various paced HRs in the majority of patients but did not affect intraventricular conduction time. Tsuchioka et al. [11], investigated the electrophysiological effects of DPH in a randomized controlled trial. They compared the effects of intravenous DPH in 20 patients who had sinus node (SN) dysfunction with 20 patients without SN dysfunction. SN dysfunction was dened as persistent sinus bradycardia, documented episodes of sino-atrial block and/or sinus arrest, and bradycardia with ectopic supraventricular tachyarrhythmia (SVTs). All the participants underwent EPS both before and after phenytoin administration (DPH, 5 mg/kg, maximum 250 mg/10 min). The PA (interval from the pacing spike to the A wave), AH and HV intervals were measured both during baseline rhythm and atrial pacing. There were no statistically signicant changes in either group, in the PA, AH and HV intervals, nor in atrial, AV nodal and ventricular refractory periods, after DPH administration. DPH had also no effect on mean SCL (Sinus Cycle Length) and mean SACT (Sinoatrial conduction time) in the two groups. Rai et al. [18] in their study, investigated the ecacy of phenytoin in the rare Andersen Tawil Syndrome (ATS) which is a familial periodic paralysis affecting the heart and skeletal system. 7 siblings with the diagnosis of ATS based on cardiac arrhythmias and genetic studies were included in the study. Patients with symptomatic ventricular tachycardia or frequent PVC associated with left ventricle failure (ejection fraction < 60%) were initially treated with oral propranolol and in the event of treatment failure, underwent left sympathetic cardiac denervation (LSCD). Persistent ventricular arrhythmia (PVCs > 25%/24 hours) or ongoing symptoms despite LSCD was considered as treatment failure and managed in a stepwise approach by a trial of different medications including propranolol, ecainide, verapamil, spironolactone and nicorandil for a duration of 3–6 months per each therapy. 3 patients, who failed to respond to the previous steps, were considered for treatment with intravenous fosphenytoin followed by oral phenytoin (5mg/kg/day). After one month of phenytoin administration, the burden of PVCs reduced to < 1% /24 hours in 2 patients and < 8% /24 hours in the third patient. There was no report of signicant adverse events, however, considering the small number of participants and short term follow up in this study, the long-term safety of using phenytoin in this particular group of patients could not be assessed. The authors suggested that phenytoin should only be used in ATS patients with ventricular arrhythmias or ectopic-induced left ventricular failure who are resistant to other therapies. tricyclic antidepressant arrhythmias investigate the ecacy of intravenous phenytoin in reversal of conduction degree AV block, intraventricular conduction delay (IVCD), and both rst-degree AV block and intraventricular conduction delay in combination.

authors (EM and SB) to identify the studies, which are appropriate to the study question. Full text of these studies was assessed independently for eligibility against the inclusion and exclusion criteria and discrepancies were resolved by consensus. (Fig. 1) Data Collection Process Data were collected independently by the two review authors. Data elds included study type, study year, sample size, patient clinical characteristics, types of arrhythmia, symptoms, mode and dose of AAD administration, short and long-term outcomes including arrhythmia burden, response to treatment, adverse drug reactions, hospitalization, and mortality. All disagreements were resolved by consensus.

Risk of Bias
The risk of bias was assessed independently by the two reviewer authors. We used Cochrane tool for randomised trials and the ROBINS-I tool for non-randomised studies [7] (Table 1, 2). Disagreements were resolved by consensus.

Results
As the result of database search, a total of 157 non-duplicate studies were screened. After reviewing the studies against the prede ned inclusion and exclusion criteria, 25 studies underwent full-text review. Subsequently, 13 clinical trials representing a total number of 985 patients met inclusion and exclusion criteria and appropriately answer the study question. Of the nal 13 clinical trials, 9 studies were non-randomised clinical trials and 4 were randomised controlled clinical trials. In 4 studies phenytoin was administered intravenously, in 6 studies a combination of intravenous and oral, and in 3 studies it was given orally. The therapeutic dose in all the studies was 5-10 mg/kg/day with the targeted therapeutic level of 10-20 µg/ml. (Table 3)  These studies suggest that phenytoin is an effective antiarrhythmic medication in treating arrhythmias (especially in drug induced ventricular arrhythmias), with an acceptable side effect pro le.

Studies of Electrophysiological Effects of Phenytoin
Epstein et al. [8] studied the e cacy of phenytoin in treating inducible ventricular arrhythmia. 64 adults with refractory ventricular arrhythmias who had failed on average 3 AADs were included in this trial. All participants underwent a baseline electrophysiology study (EPS), followed by a repeat study after administration of an AAD. They all had inducible ventricular arrhythmia (VT, VF) while receiving no AADs. Phenytoin was administered both orally (PO) and intravenously (IV) and phenytoin serum level in both responder and non-responder groups were within the therapeutic limits at the time of EPS (19.5 ± 4.7 µg/ml and 18.2 ± 4.8 µg/ml, respectively). Of the total 64 participants, only 11% of the patients had a negative EPS (responders: no inducible arrhythmias after phenytoin administration) and 89% had a positive EPS after phenytoin administration (nonresponders). The responder group continued on maintenance phenytoin and non-responders were divided in 2 sub-groups to either receive AADs other than phenytoin or phenytoin either alone or in combination with other AADs. After a total of 3 years follow up, it was shown that the longterm success was limited in the responder group. Only 3% of the study participants had responded to phenytoin therapy in both initial study and long-term follow up. In the non-responder group, long-term use of phenytoin either alone or combined with another AAD was shown to be ineffective. There was 71% (5 of the 7) mortality rate in the responders to the phenytoin therapy, but only 2 of 5 deaths were due to an arrhythmia, and the authors concluded that no death in this group could be attributed directly to phenytoin failure. This trial showed that phenytoin is not an effective AAD for treating ventricular arrhythmias as assessed by suppressing inducible ventricular arrhythmias at EPS. However, most of the patients included in this study had ventricular arrhythmias that were refractory to 3 AADs and therefore the outcomes might not be representative of patients in different clinical settings.
Caracta et al. [9] investigated the electrophysiological properties of diphenylhydantoin (DPH) in 14 patients. All patients underwent a baseline EPS and a repeat study 10 minutes after phenytoin infusion. His bundle recording was used and correlated with the phenytoin blood levels to determine the impacts of DPH on the atrial, AV nodal, and His-Purkinje system refractory periods. AH interval represented the AV nodal conduction time (the onset of the low atrial depolarization to the onset of the His de ection), and HV interval represented the His-Purkinje conduction time (the onset of the His de ection to the onset of ventricular depolarization). The patients received IV DPH (5-10 mg/kg, 100 mg/5 minutes). They concluded that majority of patients (7 out of 14 in SR and 10 out of 14 during atrial pacing) demonstrated enhanced AV nodal conduction (AH interval) following the drug administration.
In another similar study by Damato et al. [10], 13 patients underwent EPS to identify the impacts of DPH on atrioventricular (PH interval) and intraventricular conduction (HQ interval) over different paced heart rates (HRs). They also assessed the e cacy of DPH in treating existing arrhythmias in the study participants (PVCs and atrial tachycardia). All patients in this study were on maintenance digitalis treatment. Phenytoin was administered IV in all patients (dose 250-750 mg) to either terminate the arrhythmia or reach the maximum dose of 1000 mg. There were no serious side effects requiring alteration of the treatment. DPH was found to be effective in reducing burden of PVCs. It also enhanced AV conduction (shorten the PH interval) over various paced HRs in the majority of patients but did not affect intraventricular conduction time.
Tsuchioka et al. [11], investigated the electrophysiological effects of DPH in a randomized controlled trial. They compared the effects of intravenous DPH in 20 patients who had sinus node (SN) dysfunction with 20 patients without SN dysfunction. SN dysfunction was de ned as persistent sinus bradycardia, documented episodes of sino-atrial block and/or sinus arrest, and bradycardia with ectopic supraventricular tachyarrhythmia (SVTs). All the participants underwent EPS both before and after phenytoin administration (DPH, 5 mg/kg, maximum 250 mg/10 min). The PA (interval from the pacing spike to the A wave), AH and HV intervals were measured both during baseline rhythm and atrial pacing. There were no statistically signi cant changes in either group, in the PA, AH and HV intervals, nor in atrial, AV nodal and ventricular refractory periods, after DPH administration. DPH had also no effect on mean SCL (Sinus Cycle Length) and mean SACT (Sinoatrial conduction time) in the two groups.

Studies of Phenytoin E cacy compared with other AADs/Placebo
Karlson et al. [12], compared the antiarrhythmic e cacy of phenytoin with procainamide. They enrolled 81 inpatients following acute myocardial infarction (AMI), who developed ventricular arrhythmia within the rst 8 hours of their hospital stay. 42 patients were randomized to receive procainamide and 39 patients to receive phenytoin. Phenytoin was administered IV followed by oral maintenance dose (5 mg/kg/day). The treatment was stopped in 4 patients in each group due to either adverse effects or therapeutic failure and they were excluded from the study. There were 9 hospital deaths (5 in procainamide and 4 in phenytoin group) but only 1 of them occurred during the study period (in the procainamide group). The treatment failure rate during the rst 2 hours of medication administration was higher in the phenytoin group (66%) compared to the procainamide group (34%) (P < 0.05). It was concluded that procainamide is a more effective AAD compared to phenytoin, in treating early-onset, ventricular arrhythmias complicating AMI.
The e cacy of phenytoin in treating cardiac arrhythmias was investigated in another study by Eddy et al. [13]. 37 patients (30 adults and 7 children) with acute onset arrhythmias of less than 2 days duration were enrolled in the study. 21 patients were being treated for AMI, 2 with rheumatic heart disease, 7 with congenital heart disease, 1 with ischemic heart disease, 1 with myocarditis, and 5 without any other known cardiac pathology. Other AADs such as lignocaine, procainamide, and propranolol had been tried unsuccessfully on some patients with ventricular arrhythmias. Phenytoin was given both IV and orally (5 mg/kg/day), and it was effective in treating 18 of the 21 cases with a variety of post AMI arrhythmias including SVTs (restoring sinus rhythm (SR) in 6 of 9 patients), ventricular arrhythmias (restoring SR in 10 of 12 patients) and reducing the burden of PVCs. In the other 16 patients with arrhythmias due to various causes, only 6 patients had a satisfactory response. Of the 5 patients who received oral phenytoin, 3 responded favourably. In the responders to IV phenytoin, the arrhythmias terminated in the rst 10 minutes of infusion. The authors concluded that phenytoin is effective in treating ischaemic-induced arrhythmias, including in some patients that were refractory to other AADs such as lignocaine.
Kemp et al. [14] investigated the e cacy of DPH in treating non-digitalis-induced ventricular ectopic rhythms. They enrolled 10 patients from a random group of patients with PVCs detected during 12-lead electrocardiographic tracing. None of the participants had been previously treated with digoxin. They randomized 5 patients to the treatment group (DPH initially 100 mg QID orally, followed by 100 mg TDS for maintenance) and 5 patients to the control group (received placebo). It was reported that the number of PVCs in the DPH group was signi cantly reduced compared to the placebo group, throughout the three months study period. There was also some inconsistent reduction in the frequency of PVCs in the control group. The small number of patients enrolled in the study, the method of detecting PVCs and unclear randomization process without blinding, raises concerns about the reliability of the ndings in this study.
In a study by Bernstein et al. [15] the e cacy of oral DPH was investigated in 60 patients with symptomatic recurrent cardiac arrhythmias refractory to other AADs including quinidine, procainamide and digitalis. Patients enrolled in this study had a variety of arrhythmias including paroxysmal atrial brillation, paroxysmal and chronic atrial utter, premature atrial systole and PVCs. They all were commenced on DPH (300 mg/day, oral) and reviewed on fortnightly basis for an average total follow up of 16 months. The majority of the patients (62%), restored and maintained SR throughout the follow up period. DPH was discontinued in 28% of the patients due to either treatment failure or side effects. It was concluded that DPH is an effective AAD in treating recurrent cardiac arrhythmias.

Studies of Phenytoin E cacy as a Prophylactic Antiarrhythmic Agent
Lovell et al. [16], investigated the long-term e cacy of phenytoin in treating arrhythmias after myocardial infarction and its impact on prognosis. They enrolled 568 patients admitted with AMI, 285 were randomised to receive phenytoin at non-therapeutic dose (3 or 4 mg/day), and 283 patients were randomized to receive therapeutic doses of phenytoin (300 or 400 mg/day). The patients were followed at six-weekly interval visits, for a total of 12 months. In the rst 6 months, there was a lower rate of palpitations in treatment group (24%) compared to the control group (32%) (P < 0.05). This, however, might re ect a detection bias, as the visiting practitioners were not blinded. There was no difference in one-year survival rate between the two groups.
In another clinical trial, by Seuffert et al. [17], they studied the e cacy of DPH in preventing and treating arrhythmias during general anaesthesia in patients admitted for an elective inguinal hernia repair, with no previous cardiovascular comorbidities. Patients were grouped according to the anaesthetic agent used during the operation. The patients who received cyclopropane were randomized into two groups. The rst group (11 patients) received IV DPH (5 mg/kg) and the second group (9 patients) received an equivalent amount of IV uid, prior to the procedure. DPH was also given to another 10 patients who developed arrhythmias during administration of other anaesthetic agents such as halothane and methoxy urane. There was no report of arrhythmias in 8 of the 11 patients in the treatment group throughout the entire general anaesthesia compared to only 1 of the 9 patients in the control group with no arrhythmia (P = 0.01). DPH also restored sinus rhythm in all the 10 patients who developed arrhythmia with other anaesthetic agents. This study demonstrated potential use of DPH in preventing and treating arrhythmias during general anaesthesia. However, the outcomes of this trial are based on a small number of participants in the study with no blinding in the assessment of outcomes.
Study of Phenytoin E cacy in a rare syndrome Rai et al. [18] in their study, investigated the e cacy of phenytoin in the rare Andersen Tawil Syndrome (ATS) which is a familial periodic paralysis affecting the heart and skeletal system. 7 siblings with the diagnosis of ATS based on cardiac arrhythmias and genetic studies were included in the study. Patients with symptomatic ventricular tachycardia or frequent PVC associated with left ventricle failure (ejection fraction < 60%) were initially treated with oral propranolol and in the event of treatment failure, underwent left sympathetic cardiac denervation (LSCD). Persistent ventricular arrhythmia (PVCs > 25%/24 hours) or ongoing symptoms despite LSCD was considered as treatment failure and managed in a stepwise approach by a trial of different medications including propranolol, ecainide, verapamil, spironolactone and nicorandil for a duration of 3-6 months per each therapy. 3 patients, who failed to respond to the previous steps, were considered for treatment with intravenous fosphenytoin followed by oral phenytoin (5mg/kg/day). After one month of phenytoin administration, the burden of PVCs reduced to < 1% /24 hours in 2 patients and < 8% /24 hours in the third patient. There was no report of signi cant adverse events, however, considering the small number of participants and short term follow up in this study, the long-term safety of using phenytoin in this particular group of patients could not be assessed. The authors suggested that phenytoin should only be used in ATS patients with ventricular arrhythmias or ectopic-induced left ventricular failure who are resistant to other therapies.

Studies on the E cacy of Phenytoin in treating Drug-induced Arrhythmias
Hagerman et al. [19] enrolled 10 patients, with tricyclic antidepressant (TCA) induced arrhythmias to investigate the e cacy of intravenous phenytoin in reversal of the conduction defects. All the patients presented to the emergency department had a variety of arrhythmias including rst degree AV block, intraventricular conduction delay (IVCD), and both rst-degree AV block and intraventricular conduction delay in combination. Phenytoin was administered under constant ECG monitoring (5-7 mg/kg, IV, 50 mg/min, maximum dose 500 mg). All the patients achieved normalization of their conduction defect within 14 hours of phenytoin administration. The authors concluded that phenytoin is a useful drug in treating cardiac conduction defects due to TCA toxicity, however, in current clinical practice it would generally not be deemed necessary to treat asymptomatic rst-degree AV block or IVCD in this setting.
Karliner et al. [20] conducted a study to identify the e cacy of phenytoin in 54 medical and surgical patients referred for management of abnormal cardiac rhythms to hospital either on the wards or in the emergency department. They were initially selected without considering prior use of digitalis or other AADs. Subsequently, 23 patients were found to have clinical suspicion for digitalis-induced arrhythmias including AF, PAT, PVCs and SVTs. DPH was administered intravenously (250 mg), with repeated dose if the arrhythmia failed to respond or recurred within two hours. This was followed by DPH maintenance therapy (100 mg TDS, PO or IMI). Success was de ned as persistent restoring of SR in SVTs and recurrent ventricular tachycardia, or signi cant reduction of the number of PVCs. Most of the arrhythmias, especially digitalis-induced arrhythmias, responded to treatment with DPH. It was concluded that DPH is effective in treating different cardiac arrhythmias, especially arrhythmias appeared to be digitalis-induced.

Discussion
The results of this current review demonstrate the usage of phenytoin in a variety of clinical settings with a wide range of outcomes. All the studies used a similar dosage of phenytoin for treating cardiac arrhythmias. The effective dose was 5-10 mg/kg/day with the targeted phenytoin serum level of 10-20 µg/ml, which is in keeping with other data suggesting the therapeutic blood level of DPH for the treatment of most ventricular arrhythmias is generally between 10-18 µg/ml. [21] Phenytoin is considered a safe medication with minimal hemodynamic effects in humans [22]. In clinical trials investigating the toxic effects of oral phenytoin, there was no report of cardiovascular adverse effects or serious electrocardiographic changes, and other phenytoin toxicities appeared at supra-therapeutic phenytoin serum levels (almost twofold higher than upper therapeutic level) [23,24]. Guldiken et al. [25] published a systematic review which concluded that phenytoin is safe in both oral, and IV administration when it is given with an infusion rate of 50 mg/min and less in young patients and a rate of less than 25 mg/min in elderly patients, and that phenytoin infusion rate was more important than total dose in the development of cardiovascular adverse effects. IV administration of phenytoin with an infusion rate slower than 50mg/min was not associated with any cardiovascular mortality in 1593 patients. There is other evidence recommending an infusion rate of less than 50 mg/min to prevent cardiovascular complications including hypotension and bradycardia. [26] In several human and animal studies, phenytoin appears to be an effective medication in treating digitalis-related arrhythmias and was proposed as the drug of choice. [27][28][29] Depolarization in cardiomyocytes normally starts with the opening of the rapid sodium channels, which is followed by an increase in the intracellular sodium level. Subsequently, voltage-gated calcium channels will cause calcium entry and release of calcium from the sarcoplasmic reticulum, resulting in muscle contraction [30]. Digoxin is a cardiac glycoside, which inhibits the sodium-potassium-ATPase pump, and blocks the sodium-potassium exchange. This will allow calcium accumulation inside the cardiomyocytes by inhibiting calcium export from the myocyte via the sodium-calcium antiporter [31,32]. The increased intracellular calcium increases contractility at therapeutic dosages and increases likelihood of after-depolarizations and arrhythmias at toxic levels. Digoxin also increases vagal activity, reduces sinus node activity, and prolongs AV nodal conduction [33]. Digitalis toxicity enhances ventricular automaticity, prolongs both atrioventricular and interventricular conduction time, and has suppressive effects on the sinus node [28]. Phenytoin is a class 1B AAD and has Na channel blocking effects [28]. It was demonstrated that DPH inhibits digitalis binding to the sodium-potassium-ATPase pump, antagonizes digitalis-induced delayed after depolarization (DAD), and also reverses the potassium e ux caused by cardiac glycosides [28,34]. These are the likely mechanisms by which phenytoin may have reduced digitalis induced arrhythmias in the study by Karliner et al. [20].
We previously mentioned that phenytoin does not prolong interventricular conduction (HQ interval) interval and enhances AV conduction in some studies, however, its exact mechanism in shortening the AH interval is unclear [9,10]. Bigger et al. [35] in an animal study, investigated the impacts of lidocaine on the electrophysiological properties of cardiomyocytes. It was shown that lidocaine reduced both action potential time and refractory period in Purkinje and ventricular muscle bres and suppressed the automaticity in Purkinje bres. These physiological properties of phenytoin and lidocaine contribute to their e cacy in treating digitalis toxicity with the electrophysiological manifestations of enhanced ventricular automaticity, prolongation of AV conduction and IV conduction [9,36]. In contrast to these ndings, studies on other AADs suggested that therapeutic doses of procainamide, quinidine and propranolol have a different effect on the IV conduction time, and almost invariably prolonged the His-Purkinje time interval both at sinus and paced atrial rates. [28,36,37] In some animal and human studies, phenytoin was reported to be effective in treating TCA induced arrhythmias [19,38]. TCAs have a "quinidinelike" effect and prolong phase 0 depolarization of the cardiac action potential in the myocardium. TCA induced cardiac conduction abnormality is due to the blockade of rapid sodium channels in the His-Purkinje system and myocardium which prolongs both repolarization and absolute refractory times. Management of TCA induced cardiotoxicity includes serum alkalinization and adjunctive treatments such as lidocaine, phenytoin, and magnesium [39]. Foianini et al. [40] in their review paper, favored using lidocaine over phenytoin in TCA cardiotoxicity. There were concerns with regards to using phenytoin due to the risk of exacerbating TCA-induced hypotension. It was also shown that lidocaine and TCAs competitively binds to sodium channels, but fast on/off kinetics of lignocaine, may potentially unbind more sodium channels. TCAs also have different kinetics in blocking sodium channels and lidocaine appeared to be more effective in treating cardiotoxicities related to TCAs with slower block to recovery time such as amitriptyline and nortriptyline [41]. In an animal study by Chopra, Laver et al. [42], it was shown that amitriptyline activates cardiac ryanodine receptors (RyR2) and causes an early release of calcium from the sarcoplasmic reticulum (SR), which may be responsible for ventricular arrhythmias. Considering phenytoin's effects in blocking dysfunctional RyR2 receptors [6], it might be effective in treating certain TCA-induced cardiotoxicities including amitriptyline. The evidence for using phenytoin in TCA-induced cardiotoxicity, however, is con icting and additional studies are required.
We reviewed several studies demonstrating the e cacy of phenytoin in treating frequent PVCs [10,13,17], and its usefulness in treating arrhythmias unrelated to drug toxicity [27]. Phenytoin was also successfully used in 19 consecutive patients to suppress PVCs recorded on ambulatory ECG monitoring after surgical correction of congenital heart diseases [43]. However, it is not clear how phenytoin decreases the ventricular ectopic activity. Gupta et al. [44], in an experimental animal study reported that DPH had a direct vasodilatory effect on the coronary vessels, decreased coronary vascular resistance, and improved coronary blood ow, which could be responsible for the e cacy of DPH in improving ventricular ectopic activity (along with its Na-and RyR2 channel blocking inhibiting actions).
There are relatively few head-to-head clinical trials comparing the antiarrhythmic e cacy of phenytoin with other AADs. We previously mentioned the overall superiority of procainamide in treating post myocardial infarction arrhythmias compared with phenytoin in the study by Karlson et al. [12]. An animal experiment showed similar ndings in favouring procainamide's e cacy in preventing ischaemic induced ventricular brillation compared with phenytoin [45]. In the study by Bernstein et al. [15], phenytoin was effective in maintaining sinus rhythm over a 16 months follow up period in the majority patients refractory to other AADs including quinidine, procainamide and digoxin.
Tsuchioka et al. in their study [11] showed that DPH has suppressive effects on sinus node function and should be used cautiously in the treating ventricular arrhythmias in patients with sinus node dysfunction. Phenytoin was also shown in another study to have a potential effect in decreasing ventricular automaticity and the authors recommended DPH should not be used in patients with complete heart block [46].
There are con icting data with regards to the e cacy of phenytoin as assessed by EPS. In the study by Epstein et al. [8] phenytoin had limited e cacy in treating ventricular tachycardia as assessed by EP study. Only 11% of patients responded with suppression of inducible ventricular tachycardia. Similar ndings were shown in the study by Fogoros et al. [47], where the success rate of using phenytoin in suppressing inducible ventricular tachyarrhythmia was 13%. However, for the patients who had success with phenytoin, after 12-month follow up, the actuarial recurrence was 0%. These ndings may suggest that the inducibility of VAs in EPS's is not a good predictor of subsequent clinical VAs in some populations or that failure of an AAD in suppressing inducible VAs during EPS's may not be indicative of their long-term e cacy.
There are little data available comparing modern AADs with phenytoin and in recent years, usage of phenytoin as an AAD has been limited to some case reports of refractory arrhythmias [48][49][50]. Based on new evidence from animal studies that phenytoin can stabilise Ca 2+ leak form abnormally phosphorylated RyR2s in heart failure [6], it is worth reconsidering the use of phenytoin as an AAD.
There are some limitations to this study, which affect the strength of the evidence included in the present systematic review. There were a small number of randomised clinical trials with small number of participants applicable to the research question, and most of the studies included were outdated. Most of the RCTs were also not double blinded, controlled RCTs. The risk of bias was mild to moderate in non-randomised clinical trials and mild or unclear in randomized controlled trials reviewed in the present study.

Conclusion
Phenytoin has been shown to have low rates of cardiac toxicity when given at the doses used to treat arrhythmias. There is some evidence it is useful at treating arrhythmias in the setting of digitalis toxicity and this may be due to it inhibiting digitalis binding to the Na-K ATPase pump. It may also have some utility in treating/suppressing PVCs in limited studies reviewed here. Phenytoin has not been well studied to reduce spontaneous arrhythmias in heart failure and given recent evidence of its ability to reduce Ca 2+ leak from phosphorylated RyR2s in this setting, further clinical studies are needed to answer this question.