Original Contribution

A new algorithm for the initial evaluation and management of supraventricular tachycardia

H.C. Tyler Richmond MD, Lee Taylor, III MD, Michael H. Monroe MD, Laszlo Littmann MD*

Department of Internal Medicine, Carolinas Medical Center, Charlotte, NC 28232, USA

Received 9 September 2005; revised 28 November 2005; accepted 1 December 2005

Abstract Interpretations by physicians and those generated by electrocardiograph computer softwares have poor ability to recognize different types of supraventricular tachycardia . Therefore, we developed and tested a new SVT algorithm based on easily identifiable morphological characteristics and a simple dichotomous yes/no format regarding initial electrocardiographic manifestation and response pattern. The algorithm was then tested by medical house staff during the initial evaluation of 50 adult ED and cardiac intensive care unit patients suspected of having SVT. For a wide representation of SVTs, the new algorithm gave an overall diagnostic accuracy rate of 90%. Adenosine use was limited to 54% of the cases. No patient developed hemodynamic instability after algorithm-dictated interventions were carried out. Electrocardiograph computer-generated diagnoses correctly identified the specific type of SVT in 38% of the cases. This study shows the effectiveness of the proposed new algorithm in the rapid bedside evaluation and management of SVTs and confirms that computer-generated diagnoses are unreliable. D 2006

Introduction

Recognizing, diagnosing, and properly treating supra- ventricular tachycardia (SVT) are common tasks that emergency medicine and other physicians face. Rapid diagnosis and appropriate management can help decrease the potential morbidity and mortality associated with this class of arrhythmias. Thus, recognition and management of SVT have remained an integral component of physician education in the form of Advanced cardiac life support [1].

* Corresponding author. Tel.: +1 704 355 3165; fax: +1 704 355 7626.

E-mail address: laszlo.littmann@carolinashealthcare.org (L. Littmann).

Despite this training, physicians still have a poor ability to properly diagnose SVT. Two studies suggested that medical house staff and attending physicians diagnose Narrow complex tachycardias incorrectly in approximately 40% of cases [2,3]. Overuse of adenosine is common [4] and with certain forms of SVT may actually decrease diagnostic success [2,4]. In addition, physicians are often misled by the frequently incorrect computer-generated interpretations of SVTs [5]. Correct SVT recognition may be problematic because the current ACLS algorithm for SVT requires a positive identification of the type of SVT at the initiation of the algorithm and omits frequently misdiagnosed SVTs such as sinus tachycardia [1].

In response to these areas of concern, we developed a new algorithm based on the following principles: (1) initial cate- gorization of the SVT according to its electrocardiographic

0735-6757/$ – see front matter D 2006 doi:10.1016/j.ajem.2005.12.004

(ECG) manifestation rather than a specific diagnosis; (2) in- clusion of all relatively frequent forms of SVT; and (3) clear distinction between SVT and Wide complex tachycardia of uncertain etiology. We used a dichotomous yes/no flow sheet format dependent on the initial ECG manifestation of the tachycardia and the subsequent response pattern to maneu- vers that block the AV node. The use of adenosine was only permitted under specific, well-defined circumstances. Fig. 1 shows our new SVT algorithm with detailed instructions on the appropriate use of various AV nodal blocking agents incorporated in footnotes [6,7].

This pilot study was designed as a prospective investiga- tion to determine whether our algorithm could aid medical house staff in diagnosing and treating SVTs at the bedside.

Methodology

During a 7-month period, a sample of 50 patients was collected from our university-associated community teaching

hospital with its 114000-visits-per-year ED and 20-bed cardiac intensive care unit. We educated emergency medi- cine, family practice, internal medicine, and physical medi- cine and rehabilitation residents on the patient entry criteria and use of the algorithm. Patients presenting with symptom- atic tachycardia were entered into the study by medical house staff at their discretion. If a patient was suspected of having an SVT, an unused printed copy of the algorithm was obtained by each resident and reviewed. The initial ECGs were recorded using GE-Marquette ECG carts (Milwaukee, WI) fitted with the 12-SL interpretative software.

The diagnosis of SVT was allowed only under the following specific circumstances: (1) rate N 100/min; (2) QRS duration V 0.11 second; or (3) wide complex tachycardia (QRS duration z 0.12 second) with a typical bundle-branch block morphology plus constant P-QRS relationship or a preexisting bundle-branch block or an in- traventricular conduction disturbance with a perfectly iden- tical QRS morphology plus a constant P-QRS relationship [8].

Fig. 1 New algorithm for the initial evaluation and management of SVT.

All other wide complex tachycardias were excluded to follow the appropriate ACLS guidelines for wide complex tachycardia of uncertain type or ventricular tachycardia [1]. Of note, the house staff were encouraged to exclude patients with obvious sinus tachycardia of an identifiable cause and those who had a known history of Atrial fibrillation and who presented with an increased ventricular rate. Once a resident determined eligibility for the study, he or she followed the algorithm with all physiological, pharmaco- logic, and electrical interventions performed accordingly. Rhythm strips and/or further ECGs were encouraged and obtained at the discretion of each resident.

Starting at the top, the algorithm (Fig. 1) first branches with the question of whether the rhythm is regular. For a rhythm determined to be regular, the next distinction is whether it is sinus tachycardia. Identifying characteristics of sinus tachycardia (slightly irregular rhythm; upright P waves in leads I, II, and aVF; short normal PR intervals; the presence of premature atrial complexes; and heart rate modification by autonomic maneuvers) are offered for review. If the tachycardia is determined to be sinus, the algorithm termi- nates with bidentify and treat the cause.Q A further attempt to immediately establish the mechanism of the remaining regular tachycardias is not required. Instead, instruction is given to apply physiological or pharmacologic maneuvers that are intended to result in AV nodal block. Choices to induce AV block are listed and repeated attempts to block down the AV node are encouraged. Only if AV block cannot be achieved by any means is D/C cardioversion offered. If AV block is achieved, then the next question is whether the tachycardia is terminated. If the tachycardia is terminated with AV block, then the general diagnosis of reentrant tachycardia (AV nodal reentry or AV macro-reentry) is made and the initial treatment has been accomplished. If, however, AV block does not result in termination of the tachycardia, then atrial flutter or ectopic atrial tachycardia is unmasked as the diagnosis and standard ACLS treatment options are presented for consideration.

Going back to the original branch point, if analysis of the ECG reveals an irregular rhythm, then the operator should determine if group beatings are present. If they are present, then the most likely diagnosis is a variable blocking of either atrial flutter or ectopic atrial tachycardia and standard treatment options are given for consideration. If there is no

group beating present (ie, the tachycardia is irregularly irregular), then one is asked to look for discrete P waves. The presence of distinct P waves or sinus beats suggests multifocal atrial tachycardia (MAT), whereas their absence confirms a diagnosis of AF. For both types of arrhythmias, generally accepted treatment options are provided.

While following the said algorithm in evaluating and treating the patients, the house staff marked a decision- intervention line tracing their path on the printed form. On the back of the page, each resident made note of the following: (1) patient identification data; (2) clinical diag- nosis; (3) tachycardia diagnosis after use of the algorithm;

(4) general description of clinical course and events, includ- ing untoward effects of therapy and hemodynamic problems;

(5) general comments, problems, or recommendations for the algorithm; and (6) patient disposition. We reviewed the data forms biweekly to ensure that the entries met the appropriate requirements. After 50 samples were collected, an electrophysiologist reviewed all ECGs and rhythm strips. His interpretations were used as the correct diagnoses.

Results

A total of 24 medical house staff participated in the study. There were 25 cases entered in the cardiac intensive care unit, 23 in the ED, and 2 in hospital wards. In Table 1, the tachycardia diagnoses as determined by the house staff’s use of the algorithm are compared with the electro- physiologist’s interpretation of the ECGs and rhythm strips. A wide variety of SVTs were encountered, and, appropri- ately, no case of ventricular tachycardia was entered. The medical house staff diagnoses were correct in 45 of the 50 cases (90%; see Table 1). Adenosine was used on 27 occasions (54%). There was no instance of hemody- namic instability after the algorithm-dictated interventions were carried out, and resident comments on the algorithm were uniformly positive.

The ECG computer-generated diagnoses were correct in only 24 of the 50 cases (48%); of these, a specific diagnosis (ie, other than SVT) was given in only 19 (38%). In 2 cases, an undetermined rhythm was noted; in 1 case, a wide QRS tachycardia was reported. The computer interpretation was incorrect on 23 occasions (46%). Through the implementa-

tion of the algorithm, the house staff rejected all 23 incorrect diagnoses generated by the ECG computer. Of these, another incorrect diagnosis was reached on 2 occasions, whereas the correct diagnosis was reached on 21 occasions.

Discussion

Our proposed and tested SVT algorithm provides an organized, stepwise process in the interpretation of SVTs and will likely improve medical decision making and patient management.

Existing SVT algorithms

To our knowledge, this is the first prospective evaluation of the clinical utility of an SVT algorithm in the literature. Despite the enormous scale on which the current ACLS guidelines are being taught, we are not aware of any attempt to verify its efficacy or usefulness. Other SVT algorithms have been published, but they appear too cumbersome for general use and subsequently have not been tested for their didactic or clinical value [6,9,10].

Construction of the new SVT algorithm

The new SVT algorithm was constructed to follow easily recognizable morphological characteristics and a physiolog- ically based logical path. It first separates regular and irregular SVTs–a generally easy task but one that is sometimes aided by the use of calipers. Of the regular SVTs, sinus tachycardia is the most common. Positive identification of this rhythm through reminders of its typical manifestations is a required step to avoid the use of adenosine and to emphasize the fact that sinus tachycardia is rarely a primary arrhythmia. Residents were instructed to always consider atrial flutter with 2:1 block, especially in patients who have a regular tachycardia with rates between 130 and 160 beats/ min. A second, hidden P wave can sometimes be found by halving of the apparent P-P interval.

Maneuvers are then offered to block down the AV node. Many strategies to block the node are listed with encour- agement to try other strategies if one proves unsuccessful. Instructions are given on the appropriate use of the various AV nodal blocking agents. If AV block results in tachycardia termination, then AV nodal conduction must be an integral part of the tachycardia circuit defining AV nodal or AV macro-reentrant tachycardia as the etiology. If AV nodal block did not result in tachycardia termination, then the site of the regular SVT must have been above the node; ie, atrial flutter or ectopic atrial tachycardia with constant blocking ratio was the diagnosis.

Of the irregular tachycardias, the first distinction is between regular rhythms that appear irregular because of variable AV block and truly irregular ones. Group beatings are defined as episodes of regular SVT separated by pauses that almost always represent atrial tachycardia or atrial flutter with variable block. The irregularly irregular SVTs are most

often AF. Multifocal atrial tachycardia, however, may also manifest as an irregularly irregular SVT and is frequently incorrectly diagnosed as AF both by computer interpretation software and by physicians. A search for distinct sinus beats is required to avoid this misdiagnosis and should prompt the physician to consider the diagnosis of MAT if found.

In our SVT algorithm, some therapy occurs simulta- neously with interpretation, but, overall, we did not focus on details of the acute treatment of various types of SVT. The reentrant paroxysmal SVTs were terminated as part of the evaluation process. A common treatment box was offered for atrial tachycardia/atrial flutter with constant blocking ratio (regular SVT), for atrial tachycardia/atrial flutter with variable blocking ratio (irregular SVT with group beating), and for AF (irregularly irregular SVT without distinct sinus beats). This treatment box included consideration of anti- coagulation, Rate control, and antiarrhythmic management, without specifying the various antiarrhythmic treatment choices. With both sinus tachycardia and MAT, physicians were encouraged to identify and treat the underlying cause of these typically secondary tachyarrhythmias.

Clinical utility of the new algorithm

This pilot study, although not statistically evaluated, suggests that our proposed algorithm is effective in the rapid bedside diagnosis and management of SVT. It disallows consideration of wide complex tachycardia of uncertain etiology from the outset through a more rigorous definition of SVT. It requires no a priori diagnosis, only the ability to answer yes/no questions based on Standard ECG observa- tions. Imperative for any guideline, it follows a logical pattern that is conducive for quick comprehension and recall. Furthermore, the use of adenosine is potentially decreased by separating out rhythms (namely, sinus tachycardia, MAT, and AF) whose diagnosis could be hampered by its use [2,4]. For patients presenting with new-onset symptomatic SVT, the bedside diagnostic accuracy rate was 90% when used by a representative group of house staff. A wide variety of supraVentricular tachyarrhythmias were entered and, of note, no primary ventricular rhythm was included. It is not known, however, how many attempts at inclusion were aborted secondary to a failure to meet the strict SVT definition. Sinus tachycardia was correctly diagnosed without the use of adenosine in all 6 cases, and 19 of

20 reentrant tachycardias were correctly identified and terminated with the appropriate use of AV nodal blocking maneuvers (mostly adenosine). Overall, the use of adeno- sine was not necessary to elucidate the correct diagnosis or to initiate treatment in almost half of the cases.

Of the 5 incorrect diagnoses made by the house staff,

2 occurred secondary to the occasional difficulty in distinguishing between the presence of P waves and that of AF waves in irregular rhythms without group beatings. Two other incorrect diagnoses were the result of misreading the presence or absence of group beatings in rapid irregular

rhythms. The last incorrect diagnosis occurred through inappropriate use of the algorithm where a clearly irregular rhythm with group beatings was considered to be a regular rhythm. None of these said patterns of mistakes warrants a change in the algorithm as one cannot account for all reader abilities in such a format.

Limitations

Weaknesses of the study include the lack of electrophys- iological study to verify final diagnoses, the small sample size, user discretion on entry, and the lack of a controlled design. It is also not known how the algorithm would have worked with attending physicians. Based on well-described ECG principles, we feel that electrophysiological study is not necessary to make the general diagnostic distinctions called for by the algorithm [6,9-12]. The user discretion was to avoid an abundance of obvious rhythms (trivial sinus tachycardia, AF) that would falsely inflate the overall accuracy rate. Unfortunately, an effective controlled trial is difficult to design and to implement as teaching of the algorithm may induce bias. A future controlled design could best be approached with a follow-up sequential study where house staff would collect electrocardiograms demonstrating SVT based on their use of the current ACLS SVT guidelines, followed by the teaching and use of our proposed new algorithm to collect cases. Relative usefulness could then be determined by comparing the 2 accuracy rates.

Summary

The proposed new algorithm appears to be effective in the rapid bedside evaluation and management of SVTs and may prove valuable in the setting of an ED.

References

1. The American Heart Association in collaboration with the Interna- tional Liaison Committee on Resuscitation (ILCOR). Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascu- lar care: an international consensus on science. Circulation 2000; 102(Suppl):I1- I384.
2. Conti JB, Belardinelli L, Curtis AB. Usefulness of adenosine in diagnosis of tachyarrhythmias. Am J Cardiol 1995;75:952 – 5.
3. O’Rourke DJ, Palac RT, Schindler JT, et al. Clinical utility of adenosine in difficult to diagnose tachyarrhythmias. Clin Cardiol 1999;22:633 – 9.
4. Knight BP, Zivin A, Souza J, et al. Use of adenosine in patients hospitalized in a university medical center. Am J Med 1998;105: 275 – 80.
5. Bogun F, Anh D, Kalahasty G, et al. Misdiagnosis of atrial fibrillation and its clinical consequences. Am J Med 2004;117:636 – 42.
6. Blomstrom-Lundquist C, Scheinman MM, Aliot EM, et al. ACC/ AHA/ESC guidelines for the management of patients with supraven- tricular arrhythmias–executive summary: a report of the American College of Cardiology, American Heart Association Task Force on Practice Guidelines, and the European Society of Cardiology Committee for Practice Guidelines. J Am Coll Cardiol 2003;42: 1493 – 531.
7. Littmann L, Anderson JD, Monroe MH. Adenosine and Aggrenox: a hazardous combination. Ann Intern Med 2002;137:e- 76.
8. Littmann L, McCall MM. Ventricular tachycardia may masquerade as supraventricular tachycardia in patients with pre-existing bundle branch block: a potential cause of misdiagnosis and inappropriate management. Ann Emerg Med 1995;26:98 – 101.
9. Bar FW, Brugada P, Dassen WRM, et al. Differential diagnosis of tachycardia with narrow QRS complex (shorter than 0.12 second). Am J Cardiol 1984;54:555 – 60.
10. Jones GK. A practical approach to narrow complex tachycardia. Int Med 1996;81 – 94.
11. Littmann L, Tenczer J, Fenyvesi T. Atrioventricular nodal reentrant Paroxysmal supraventricular tachycardia. Arch Intern Med 1984; 144:129 – 31.
12. Ganz LI, Friedman PL. Supraventricular tachycardia. N Engl J Med 1995;332:162 – 73.