In this article, the authors discuss the hotly debated topic of whether antiarrhythmic therapy should be initiated in-hospital or out-of-hospital. They recommend in-hospital initiation of antiarrhythmic medications for patients with significant structural heart disease, conduction disease, and/or QT prolongation. However, further studies are needed before a final decision on guidelines can be made for clinical practice. ABSTRACT: Background. Initiation of antiarrhythmic therapy for atrial fibrillation is a key step in the treatment of this disorder. Much controversy remains as to the risks and benefits of initiating therapy as an inpatient versus an outpatient. Objective. To explore the various issues of debate and to determine the importance and validity of these various issues when it comes to the evaluation of patients for in- versus out-of-hospital initiation of antiarrhythmic therapy for atrial fibrillation. Methods. A MEDLINE search of English language journal articles since 1966 and a hand search of bibliographies included in pertinent retrieved articles was undertaken. Articles used included review articles, retrospective studies, and meta-analyses. Results. The literature is full of articles for and against outpatient initiation of antiarrhythmic therapy. One side feels that the risks of antiarrhythmic therapy initiation are serious enough in all patients and easy enough to reverse or ameliorate if the patient is in the safety of the monitored hospital setting. The other side argues that these complications are infrequent enough except in certain commonly identifiable patients, that not all need hospitalization during antiarrhythmic initiation. The issues at the heart of the dispute include: the presence or absence of underlying heart disease; the period of monitoring after initiation of therapy; the choice of antiarrhythmic agent used; and even the seriousness and prevalence of the arrhythmia which can be induced. Conclusions. The issue of in- versus out-of-hospital initiation of antiarrhythmic therapy for atrial fibrillation remains a widely disputed topic. Many factors come under consideration when this topic is studied. At present, we recommend that patients with significant structural heart disease, conduction disease, and/or QT prolongation be strongly considered for in-hospital initiation of antiarrhythmic medications. Further prospective studies are necessary to assess the magnitude of the difference of initiating antiarrhythmic therapy as an inpatient versus as an outpatient. Atrial fibrillation is one of the most common arrhythmias encountered. Its prevalence increases with age and it has come to be recognized as a major cause of embolic events, especially involving the cerebrovasculature. Intense clinical research has been devoted over the past decade to the classification, mechanisms, treatment, outcomes, and other factors related to atrial fibrillation. With an increase in the understanding of the underlying mechanisms and causes of atrial fibrillation, stronger and more targeted treatment strategies have emerged and now serve as the mainstay of therapy. The general approach to the treatment of atrial fibrillation focuses on three main factors. The first of these factors is control of the ventricular rate. This is usually accomplished through the use of calcium channel or beta-receptor blockade. The second factor emphasizes the reduction in the risk of thromboembolism through anticoagulation. The third and sometimes most challenging factor pertains to establishing and maintaining sinus rhythm. This third factor begins with the establishment of sinus rhythm either by chemical or electrical cardioversion, or even through spontaneous conversion by the patient. Once sinus rhythm is obtained, the next step is to maintain it. This is usually accomplished by continuing the patient on antiarrhythmic therapy. Current practice has usually dictated that patients with atrial fibrillation in whom antiarrhythmic therapy is being initiated are hospitalized and monitored closely. This is because of the fear of the proarrhythmic effects caused by antiarrhythmic therapy. There is no question that continuous monitoring allows for the immediate detection and termination of life-threatening arrhythmias, but the strategy of hospitalization for this purpose requires a sufficient risk of lethal proarrhythmia to be warranted. The assessment of this risk requires the examination of numerous factors including patient risk factors, the particular anti-arrhythmic medication, the time period to the manifestation of proarrhythmic effects, and numerous other factors. The purpose of this study was to retrospectively assess whether the practice of hospitalizing patients for a period of usually 72 hours is warranted by examining the period after which a patient is started on antiarrhythmic therapy while still in the hospital. This 72-hour period was examined in order to determine whether any correlations with respect to patient risk factors and the particular medications used during the antiarrhythmic initiation phase for treatment of atrial fibrillation could be drawn.1-3 A MEDLINE search of the English language articles published from January 1, 1966 through December 1, 2000, was performed. Articles were found by intersecting the terms antiarrhythmia agents, atrial fibrillation, and initiation of therapy. All types of articles were included in the search. Titles obtained from the search were scanned on the computer monitor and articles which appeared to focus on the issue of in-hospital versus out-of-hospital initiation of therapy were selected. Copies of these articles were then obtained and reviewed. These articles obtained included review articles, retrospective studies, and meta analyses. Other articles also used included reports from scientific meetings and statements from medical/scientific societies. Additional articles were obtained through a hand search of the bibliographies of the retrieved articles. Inpatient versus outpatient initiation of antiarrhythmic therapy continues to be a widely debated issue in the medical community. Numerous studies make the case for inpatient initiation citing certain outcomes and cost analyses,1,4-7 while others use different outcomes and cost analyses to make the argument for outpatient initiation of antiarrhythmics.8-10 The issue of debate comes when the risks of proarrhythmia and other side effects, which are known to occur, are discussed. One side feels that these risks are serious enough in all patients and easy enough to reverse or ameliorate if the patient is in the safety of the monitored hospital setting. The other side argues that these complications are infrequent enough except in certain commonly identifiable patients that not all need hospitalization during antiarrhythmic initiation. Arguments for the two sides are backed up by varying medical, legal, risk-benefit, and cost-analysis points of view (Table 1). Proarrhythmia. Depending on the prevalence of underlying diseases in the study populations, the incidence of proarrhythmia has been cited in the range from 1-8%.5,6,11,12 Significant arrhythmic complications include new or increased ventricular arrhythmias, prolonged QT intervals (which can predispose to ventricular arrhythmias), and bradyarrhythmias.6 Maisel et al.4 identified 64/597 trials (with a total of 417 patients) in which patients developed one of the above three arrhythmic complications. Maisel s review further points out that bradyarrhythmias, which may or may not be symptomatic, are the most common form of electrophysiologic adverse response. These are the identified arrhythmias which have been known to develop in patients during initiation of antiarrhythmics. These arrhythmias, if detected, can prompt changes in management, whether it be electrolyte adjustments, antiarrhythmic dose adjustment, antiarrhythmic discontinuation, further electrophysiologic testing, or even more invasive steps like pacemaker implantation. Patients underlying risk factors. An issue at the center of the debate over inpatient vs. outpatient management focuses on the patients underlying heart disease and whether certain patients are more likely to be at greater risk for proarrhythmia than others; and whether we can use the degree or type of underlying heart disease as a marker to screen which patients will need inpatient monitoring at the initiation of antiarrhythmic therapy. There has been agreement among different authors2,4-6 about high risk patients in whom antiarrhythmic therapy should be initiated in-hospital. These patients include those with: 1) advanced left ventricular dysfunction; 2) prolonged QT interval at baseline; 3) a prior documented proarrhythmic response; and 4) a tendency to symptomatic bradycardia.6 Numerous studies document the increased incidence of proarrhythmia in patients with structural heart disease. Maisel et al.4 noted that the risk of adverse events was greatest among patients with a history of a prior myocardial infarction. In an analysis of 185 patients Prystowsky13 found that only 4% of adverse events occurred in patients without structural heart disease. Flaker et al.14 further argues that patients without a history of congestive heart failure had no increased risk of cardiac mortality during antiarrhythmic therapy. These reports support the notion that most adverse effects, especially proarrhythmia, tend to occur in the presence of underlying cardiac structural disease. Period of monitoring. Another debated issue is the period of monitoring after antiarrhythmic initiation. Most patients are monitored for a period of 72 hours. Chung et al.6 reported that all of their 120 patients studied, except for three, met criteria for the presence or absence of any significant arrhythmia complication within three days of initiation of hospital monitoring. Chung further reports that 70% of all events occur within the first 24 hours of therapy. Maisel et al.4 concur that the greatest risk for cardiac adverse events occurs during the first 24-72 hours. Thibault8 takes a different view in his meta-analysis, stating that only about half of the arrhythmic events occur within three days of initiation of therapy and that this makes the process of hospitalizing all patients at initiation of therapy a low-yield and expensive process. Another point to consider is the fact that proarrhythmic risks tend to be higher after patients have converted to sinus rhythm or are in the process of doing so.5,13 Although this is an important point, most studies do not focus on the patients rhythm at initiation of antiarrhythmic therapy. Specific antiarrhythmic drug used. The choice of antiarrhythmic medication used also plays a role in this debate. The two major groups of antiarrhythmics used for atrial fibrillation are the Class I and III agents. Class I agents include IA agents (Quinidine, Procainamide, and Disopyramide) and IC agents (Flecainide and Propafenone). Class III agents include Sotalol, amiodarone, and Dofetilide. Of the studies performed, varying numbers of instances of proarrhythmia were noted with the Class IA and IC agents. In a meta-analysis, Maisel et al.4 found 13/189 trials showing proarrhythmic effects for Procainamide, 16/179 trials for Quinidine, 1/20 trials for Disopyramide, 14/110 trials for Propafenone, and 0/2 trials for Flecainide. Thibault8 reported that out of 17 studies using Flecainide which he evaluated, 30/1,634 patients in these studies experienced proarrhythmic effects with Flecainide. One study which directly compared Flecainide versus Quinidine15 found that the incidence of proarrhythmia in both treatment groups was very low and that no mortality was noted from proarrhythmia during their 12-month trial period. These results were similar to those found for Flecainide in an earlier study by Pritchett et al.16 In terms of the class III agents, Maisel et al.4 found 8/72 trials of Sotalol and 1/25 trials of amiodarone showing proarrhythmic side effects. Thibault8 found proarrhythmic effects in 6 out of 260 patients in 5 trials of Sotalol, and 10 patients out of 816 with proarrhythmic effects in 13 studies of amiodarone. With respect to the newer class III agent Dofetilide, Singh et al.17 in the SAFIRE-D trial randomized 325 patients to different dosages of dofetilide versus placebo and found that 11 patients in the dofetilide arm had to be withdrawn due to serious adverse events. One patient suffered sudden death and the other 10 were withdrawn secondary to prolongation of QT or QTc interval. Seven of these patients had prolongation of the QT interval within 72 hours. Singh further notes that there were seven proarrhythmic events noted in the dofetilide-treated patients; three (1.2%) of the events were considered to be related to dofetilide. The DIAMOND-CHF18 study examined the use of dofetilide in patients with ventricular dysfunction and congestive heart failure. It showed that treatment with dofetilide does not increase mortality when it is initiated in the hospital with dosage determined on the basis of renal function and QTc response to therapy. Cost analysis. The cost effectiveness of initiating antiarrhythmics in the hospital is yet another important point of discussion. When assessing cost effectiveness, it is important to focus on the issues discussed above, which include the period of monitoring, the patients underlying conditions and the antiarrhythmic used. A cost analysis study performed by Simons et al.7 examined 57 drug trials involving 2,822 patients undergoing initiation of antiarrhythmic therapy. They found that based on a 72-hour average adverse event rate of 0.63%, inpatient therapy initiation cost $19,231 per year of life saved for a 60-year-old patient with an average life expectancy. Most experts in the United States consider interventions that cost 19 Even when adverse event rates were varied to account for model hypothetical scenarios, the cost per year of life saved was still less than $50,000. Based on these results, Simon et al. conclude that the cost effectiveness of a 72-hour hospitalization for initiation of antiarrhythmic therapy of supraventricular tachycardias compares favorably with other widely accepted medical and surgical therapies. Even though this is the conclusion reached, the authors caution that the risk of proarrhythmia is related to the presence or absence of underlying heart disease and that this should be accounted for in the final analysis when it becomes more clear what the magnitude of the increase in risk really is. In an article by Prystowsky,9 the low frequency of proarrhythmia is used as an argument against the cost effectiveness of in-hospital antiarrhythmic initiation. Prystowsky argues that patients without structural heart disease, sinus node dysfunction, or AV conduction abnormalities who have a normal baseline QT interval need not be hospitalized for initiation of antiarrhythmics. Present recommendations. Although there are no clear cut guidelines on in- versus out-of-hospital initiation of antiarrhythmic drugs for atrial fibrillation, we currently concur with Prystowsky s recommendations that in-hospital initiation of antiarrhythmics occur for the following high-risk groups: those with significant structural heart disease, conduction disease, and/or QT prolongation. Conclusion and future directions. The issue over in- versus out-of-hospital initiation of antiarrhythmic therapy continues to conjure up debate. Many facets of the argument need to be considered and discussed before a final decision on guidelines can be made for clinical practice. More research needs to be undertaken on the topics of discussion in this area to better delineate the more important factors and include them in the final decision-making process of whether to hospitalize patients at the initiation of antiarrhythmic therapy. These studies need to prospectively assess the magnitude of the difference in initiating antiarrhythmic therapy as an inpatient versus as an outpatient by observing the adverse events which occur and evaluating their impact and degree of harm to the patient. Improvements in the understanding of the basic mechanisms of action of the antiarrhythmics, the mechanisms of arrhythmias, as well as their interplay with patient factors, will allow for the development of safer and more effective antiarrhythmic regimens in the future.