Arrhythmia? Does This Patient with Palpitations Have a Cardiac

Home , Cannon A waves, Hypovolemia, Palpitations

Does This Patient With Palpitations Have a Cardiac Arrhythmia?

Paaladinesh Thavendiranathan; Akshay Bagai; Clarence Khoo; et al. Online article and related content current as of November 17, 2009. JAMA. 2009;302(19):2135-2143 (doi:10.1001/jama.2009.1673)

http://jama.ama-assn.org/cgi/content/full/302/19/2135

Supplementary material eFigures and eTables http://jama.ama-assn.org/cgi/content/full/302/19/2135/DC1 Correction Contact me if this article is corrected. Citations Contact me when this article is cited. Topic collections Physical Examination; Cardiovascular System; Radiologic Imaging; Diagnosis; Echocardiography; Arrhythmias; Cardiac Diagnostic Tests Contact me when new articles are published in these topic areas. CME course Online CME course available. CME course Online CME course available.

Subscribe Email Alerts http://jama.com/subscribe http://jamaarchives.com/alerts

Permissions Reprints/E-prints [email protected] [email protected] http://pubs.ama-assn.org/misc/permissions.dtl

Downloaded from www.jama.com by guest on November 17, 2009 THE RATIONAL CLINICIAN’S CORNER CLINICAL EXAMINATION

Does This Patient With Palpitations Have a Cardiac Arrhythmia?

Paaladinesh Thavendiranathan, MD Context Many patients have palpitations and seek advice from general practitioners. Akshay Bagai, MD Differentiating benign causes from those resulting from clinically significant cardiac ar- Clarence Khoo, MD rhythmia can be challenging and the clinical examination may aid in this process. Objective To systematically review the accuracy of historical features, physical ex- Paul Dorian, MD amination, and cardiac testing for the diagnosis of cardiac arrhythmia in patients with Niteesh K. Choudhry, MD, PhD palpitations. Data Source, Study Selection, and Data Extraction MEDLINE (1950 to Au- CLINICAL SCENARIO gust 25, 2009) and EMBASE (1947 to August 2009) searches of English-language ar- A 58-year-old woman presents to the ticles that compared clinical features and diagnostic tests in patients with palpitations emergency department with intermit- with a reference standard for cardiac arrhythmia. Of the 277 studies identified by the tent episodes of palpitations. She search strategy, 7 studies were used for accuracy analysis and 16 studies for diagnos- describes “heart fluttering” that usu- tic yield analysis. Two authors independently reviewed articles for study data and quality and a third author resolved disagreements. ally lasts less than 5 minutes, which is associated with a sense of “impending Data Synthesis Most data were obtained from single studies with small sample sizes. doom,” sweating, and paresthesia in A known history of cardiac disease (likelihood ratio [LR], 2.03; 95% confidence inter- both hands. She is unable to tell val [CI], 1.33-3.11), having palpitations affected by sleeping (LR, 2.29; 95% CI, 1.33- 3.94), or while the patient is at work (LR, 2.17; 95% CI, 1.19-3.96) slightly increase whether the rhythm is regular or the likelihood of a cardiac arrhythmia. A known history of panic disorder (LR, 0.26; irregular and denies a regular rapid- 95% CI, 0.07-1.01) or having palpitations lasting less than 5 minutes (LR, 0.38; 95% pounding sensation in the neck. CI, 0.22-0.63) makes the diagnosis of cardiac arrhythmia slightly less likely. The pres- There is no associated presyncope or ence of a regular rapid-pounding sensation in the neck (LR, 177; 95% CI, 25-1251) syncope. She has a history of panic or visible neck pulsations (LR, 2.68; 95% CI, 1.25-5.78) in association with palpita- disorder but is otherwise healthy and tions increases the likelihood of a specific type of arrhythmia (atrioventricular nodal takes no medications. Her pulse rate reentry tachycardia). The absence of a regular rapid-pounding sensation in the neck and rhythm are palpably normal and makes detecting the same arrhythmia less likely (LR, 0.07; 95% CI, 0.03-0.19). No other features significantly alter the probability of clinically significant arrhythmia. Di- the rest of her physical examination agnostic tests for prolonged periods of electrocardiographic monitoring vary in their along with a 12-lead electrocardio- yield depending on the modality used, duration of monitoring, and occurrence of typi- gram is normal. cal symptoms during monitoring. Loop monitors have the highest diagnostic yield (34%- 84%) for identifying an arrhythmia. WHY IS THE CLINICAL Conclusions While the presence of a regular rapid-pounding sensation in the neck EXAMINATION FOR or visible neck pulsations associated with palpitations makes the diagnosis of atrio- PALPITATIONS IMPORTANT? ventricular nodal reentry tachycardia likely, the reviewed studies suggest that the clini- Palpitations are a common, unpleas- cal examination is not sufficiently accurate to exclude clinically significant arrhythmias ant, and often alarming awareness of in most patients. Thus, prolonged electrocardiographic monitoring with demonstra- heartbeats,1 with a prevalence as high tion of symptom-rhythm correlation is required to make the diagnosis of a cardiac ar- as 16% in medical outpatients.2 They rhythmia for most patients with recurrent palpitations. often pose a clinical challenge3 JAMA. 2009;302(19):2135-2143 www.jama.com because of the wide differential diag- Author Affiliations: Division of Cardiology, Depart- Corresponding Author: Niteesh K. Choudhry, nosis (BOX). Palpitations may occur due ment of Medicine (Dr Thavendiranathan) and Divi- MD, PhD, Brigham and Women’s Hospital, Har- to a change or abnormality in heart sion of Cardiology, St Michael’s Hospital (Drs Bagai vard Medical School, 1620 Tremont St, Ste and Dorian), University of Toronto, Toronto, Ontario, 3030, Boston, MA 02120 (nchoudhry@partners Canada; Division of Cardiology, Department of Medi- .org). cine, University of British Columbia, Vancouver, Canada The Rational Clinical Examination Section Editors: CME available online at (Dr Khoo); and Division of Pharmacoepidemiology and David L. Simel, MD, MHS, Durham Veterans Affairs www.jamaarchivescme.com Pharmacoeconomics, Department of Medicine, Medical Center and Duke University Medical Center, and questions on p 2160. Brigham and Women’s Hospital and Harvard Medi- Durham, NC; Drummond Rennie, MD, Deputy cal School, Boston, Massachusetts (Dr Choudhry). Editor.

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA

Because a minority of patients have tant to consider clinically significant ar- Box. Differential Diagnosis palpitations while being examined by rhythmias associated with sudden car- of Palpitations their physician, the challenge is to cap- diac death. Atrial fibrillation, flutter, Arrhythmia ture a recording of the cardiac rhythm atrial tachycardia, and ventricular during symptoms. While event moni- tachycardia (eFigure 1A, B, I) tend to Defined as atrial fibrillation or flutter, atrioventricular node reentry tors have been designed to facilitate this occur later in life and are often associ- tachycardia or atrioventricular re- process, the diagnostic yield varies with ated with structural heart disease. Some entry tachycardia, atrial tachycar- the frequency of symptoms and dura- arrhythmias such as atrioventricular dia, ventricular tachycardia, pre- tion of the monitored period. Arrhyth- node reentry tachycardia (eFigure 1C) mature ventricular contractions or mias also may occur in individuals who may be more common in women than premature atrial contractions, or have no symptoms at all.7 Therefore, the men.9-11 multifocal atrial tachycardia presence of an arrhythmia on diagnos- A history of panic disorder should be The causes are primary electrical tic testing does not confirm that it is the explored.12 The details of a family his- abnormality or electrical abnor- cause of a patient’s symptoms.7 To be tory of palpitations should be re- mality secondary to structural car- certain, their symptoms must be cor- corded, especially if family members diac disease or comorbid medical related with an electrocardiographi- have established diagnoses such as ar- conditions. cally documented rhythm distur- rhythmogenic right ventricular cardi- Sinus tachycardia bance. Similarly, if the patient repeatedly omyopathy13 or atrial fibrillation.14 Any The causes include hyperthyroid- has a normal cardiac rate and rhythm history of previous cardiac disease may ism, anxiety or panic disorder, fe- during typical symptoms, one can reas- predispose patients to more clinically ver, hypovolemia, stimulants (caf- sure the patient that the cause is likely significant arrhythmias4,6 and suggest feine, alcohol), medications, blood nonarrhythmic. the need for a more aggressive search loss, pheochromocytoma, hypo- While palpitations are usually be- for a cardiac cause. glycemia, and idiopathic. nign, they may be a manifestation of Patients should be asked to tap out Normal sinus rhythm life-threatening conditions. More im- the rhythm of their palpitations, or to The cause is heightened cardiac portantly, recurrent palpitations can be choose from cadences tapped by the perception for an unclear reason. associated with significant disability, in- physician, to identify the regularity and cluding impaired work performance speed of the palpitations. Single skipped and the inability to perform house- beats or a sensation of the heart stop- hold duties.4,8 However, using diagnos- ping and then starting with a pound- rhythm, such as an arrhythmia (ie, an tic tests such as event monitors and ing, flipping, or jumping sensation, es- abnormal, disordered, or disorga- echocardiograms for every person with pecially while sitting quietly or lying in nized heartbeat), due to an appropri- palpitations can be costly and of low di- bed and lasting only for brief periods, ate increase in normal sinus rate, agnostic yield. Therefore, we re- have traditionally been attributed to or with a normal sinus rate and rhythm viewed the utility of clinical history, premature atrial or ventricular extra sys- due to heightened sensitivity and physical examination, and resting rou- toles.15,16 An irregular heartbeat, both perception of one’s heartbeats tine electrocardiography as screening in rhythm and strength, that begins and (eFigure 1 is available at http://www tests for identifying patients with pal- terminates abruptly suggests atrial .jama.com). pitations whose symptoms are likely or fibrillation. In one study, primary cardiac dis- unlikely to be due to a cardiac arrhyth- The association of polyuria and pal- ease (43%) and anxiety or panic dis- mia. pitations may indicate supraventricu- order (31%) were the most common lar tachycardia because increased atrial causes in patients presenting with HOW TO EVALUATE A PATIENT pressures stimulate production of na- palpitations to the emergency depart- WITH PALPITATIONS triuretic peptides.17 A regular rapid- ment, admitted to the hospital, or Patient History pounding sensation in the neck may sig- attending a medical clinic.4 Among Most demographic and historical fea- nify atrioventricular node reentry patients with cardiac disease, palpita- tures do not significantly influence the tachycardia when the contraction of the tions were attributable to arrhythmia likelihood of clinically significant ar- atria against closed atrioventricular in 91% of cases. Thus, the pretest rhythmias.5,6 Patient age may be im- valves produces increased right atrial probability of cardiac arrhythmia in a portant because supraventricular tachy- pressures and reflux of blood into the similar patient population would be cardias, particularly ones that use a superior vena cava.18 Associated shirt 39%. In 2 other studies, 19% of bypass tract (atrioventricular reentry flapping, defined as visible movement patients presenting with palpitations tachycardia) (eFigure 1D), may be first of patient’s clothes during the epi- were found to have a clinically sig- experienced earlier in life.9,10 In young sode, also has been described with both nificant arrhythmia.5,6 athletes with palpitations, it is impor- atrioventricular node reentry tachycar-

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA dia and atrioventricular reentry tachy- that may give rise to an arrhythmia. applied at the time of symptoms (event cardia.19 Presyncope or syncope may When a patient is examined while hav- recorders).32 Traditionally, intermit- represent more clinically significant ar- ing palpitations or the examiner de- tent event recorders store electrocar- rhythmias such as ventricular tachy- tects an asymptomatic arrhythmia, cer- diographic monitoring for several min- cardia (eFigure 1I).16 However, syn- tain physical examination features may utes once activated by the patient and cope can occasionally result from acute be useful. Atrial fibrillation is sug- hence cannot capture asymptomatic ar- vasodilatation and/or rapid heart rate gested by a pulse that is not regular and rhythmias or those associated with loss with low cardiac output that occurs at has no repeating pattern (ie, irregu- of consciousness.32 Newer loop record- the beginning of a supraventricular larly irregular), the presence of a pulse ers provide continuous, real-time out- tachycardia20,21 or due to conversion deficit (ie, obtaining a lower pulse rate patient electrocardiographic monitor- pauses occurring at the end, espe- at the wrist than at the apex), or the aus- ing and can automatically detect cially in patients with underlying si- cultation of variable first heart sound asymptomatic arrhythmias in addi- nus node disease. Conditions such as intensity. These findings are due to beat- tion to being activated by the pa- hyperthyroidism could be associated to-beat variation in stroke volume that tient.33,34 Intermittent event recorders with sinus tachycardia or atrial fibril- occurs during atrial fibrillation. The allow for prolonged monitoring (weeks lation (eFigure 1G, I). Similarly, pal- presence of cannon A waves on the to months) in patients who have infre- pitations associated with a psychiatric jugular venous pressure suggests an ar- quent symptoms. These devices may diagnosis such as panic disorder could rhythmia associated with atrioventric- have a higher specificity because the pa- suggest sinus tachycardia. However, it ular dissociation such as ventricular tient activates the recording during is essential to rule out clinically signifi- tachycardia.31 A cannon A wave is a symptoms. Specifically, loop moni- cant arrhythmias before attributing pal- prominent wave in the jugular venous tors save information for a predeter- pitations to the patient’s psychiatric pressure that occurs due to the con- mined period prior to the patient trig- condition.4,8,22 traction of the right atrium against a ger, and hence, can help identify the Onset during catecholamine ex- closed tricuspid valve. initiation sequence for arrhythmias. cess, such as during exercise, may sug- These stored events can be transmit- gest ventricular tachycardia or sinus Diagnostic Tests ted through a telephone for physician tachycardia (more commonly).23 Pal- Standard 12-lead electrocardiography review. pitations starting during sleep or states is the initial test in patients with pal- An electrophysiologic study is an in- of increased vagal tone (eg, at termi- pitations and may identify the arrhyth- vasive test of the electrical conduction nation of exercise) can be associated mia or provide insight into underly- system of the heart. Although often per- with vagal-mediated atrial fibrillation ing structural and electrical abnormality formed for diagnostic and therapeutic or, less likely, certain subtypes of that may be a precipitant for arrhyth- purposes in patients with a known ar- long-QT syndromes.24 Other triggers for mias. Patients with electrical or struc- rhythmia or who have presented tachycardias include alcohol or caf- tural abnormalities on 12-lead electro- with syncope or resuscitated sudden feine consumption.25,26 cardiography may warrant a more cardiac death, it is occasionally per- While patients with QT prolonga- aggressive search for a cardiac cause of formed as a diagnostic test in patients tion and associated arrhythmias usu- palpitations. with palpitations in whom there is high ally present with syncope, a medica- The prototypical clinical event moni- suspicion for cardiac origin.35 tion review is warranted. Drugs that tor is the Holter monitor that continu- Exercise treadmill testing with a stan- prolong QT and predispose patients to ously and simultaneously records 2 or dard Bruce protocol may be useful in torsades de pointes and other ventricu- 3 electrocardiographic leads. At the end patients whose palpitations typically oc- lar arrhythmias include antiarrhyth- of the monitoring period (typically 24 cur during exercise or are provoked by mics, antimicrobials, antihistamines, or 48 hours), the data are analyzed for cardiac ischemia.15,36 psychotropic drugs, and other miscel- arrhythmias (eFigure 1) and are cor- When palpitations occur infre- laneous drugs such as motility drugs, related with symptoms recorded by the quently or are associated with serious diuretics via electrolyte depletion, and patient. The Holter monitor detects events such as syncope that cannot be protease inhibitors for human immu- asymptomatic arrhythmias and may identified using intermittent event nodeficiency virus.27-30 capture arrhythmias in patients who are recorders, implantable loop recorders unable to trigger the device (eg, dur- (implanted under the skin in the left Physical Examination ing syncope). Frequently, patients may parasternal region) can record the Most patients with episodic palpita- not experience their usual symptoms patient’s electrocardiogram continu- tions are examined when asymptom- during monitoring and the test is ously for prolonged periods (several atic. Typically, the purpose of the physi- nondiagnostic. months to years).37,38 Patients keep a cal examination in this setting is to Intermittent event recorders can be diary of their symptoms for symptom- identify structural heart abnormalities worn continuously (loop recorders) or rhythm correlation. The device can

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA also be triggered with an external several presenting complaints but did RESULTS activator. not provide separate data for the sub- Only 7 studies met inclusion criteria Echocardiography may identify group with palpitations; (3) focused for the assessment of diagnostic structural heart diseases that may be a only on comparison between specific accuracy (TABLE 1; eFigure 2 is a precipitant for arrhythmias.15,39 While arrhythmias or used the presence of ar- flow diagram illustrating the identifi- the presence of structural heart dis- rhythmias as opposed to palpitations as cation of articles and is available at ease increases the likelihood of a clini- inclusion criteria; or (4) did not re- http://www.jama.com).4-6,11,12,18,42 Pal- cally significant arrhythmia and sug- quire symptom rhythm correlation for pitations were the predominant pre- gests the need for a more aggressive the diagnosis of arrhythmia. From the senting complaint in these studies search for an arrhythmic substrate, it results of the same literature search, (99.4% of the included patients). The does not prove that the patient’s palpi- studies were identified providing data majority of the data was extracted from tations are secondary to an arrhyth- on the diagnostic yield of the various the 2 level 1 studies.5,6 Only 1 study mia. tests (eg, electrocardiography and loop assessed a limited number of physical monitoring). examination signs in patients with pal- METHODS The data extracted were the num- pitations.6 No study evaluated a com- Search Strategy ber of patients enrolled, symptoms, bination of historical and physical ex- and Data Collection signs or tests assessed, the number of amination features or the precision of Structured MEDLINE (1950 to Au- patients with and without arrhythmia any historical or physical examination gust 25, 2009) and EMBASE (1947 to for each clinical parameter, and the fre- feature. The reference standards in the August 2009) literature searches were quency of typical symptoms and clini- included studies were electrophysi- performed to identify English- cally significant arrhythmias (when pre- ological study,18 24-hour Holter moni- language articles relevant to the preci- sent). From this, the likelihood ratios tor,12,42 intermittent event record- sion or accuracy of the clinical exami- (LRs) were calculated for the indi- ers,5,6 and in 2 studies a combination nation for patients with palpitations. vidual findings described, along with of methods.4,11 Among studies that used Search terms included palpitations, heart the 95% confidence intervals (CIs). loop recorders (both for diagnostic ac- racing, heart pounding, physical exami- Where possible, the LRs were sepa- curacy and yield data), only 1 study34 nation, medical history taking, profes- rately calculated for detecting any ar- had the automatic trigger feature to rec- sional competence,“sensitivity and speci- rhythmias and clinically significant ar- ord asymptomatic arrhythmias. ficity,” reproducibility of results, observer rhythmias. An arrhythmia was defined Only 2 of the 7 diagnostic accuracy variation, “diagnostic tests, routine,” de- as any rhythm with a heart rate of 60/ studies distinguished clinically insig- cision support techniques, Bayes theo- min or less, or 100/min or greater, nificant and significant arrhyth- rem, and mass screening. Two authors and/or that was not normal sinus mias,5,6 although only 1 allowed for the independently reviewed the abstracts of rhythm. Clinically significant arrhyth- calculation of LRs for both types of the search and retrieved potentially rel- mias were those that likely require spe- rhythm disturbances.5 Other studies evant articles and a third author re- cific management including ventricu- looked only at clinically significant ar- solved disagreements. Additional ar- lar tachycardia, atrioventricular node rhythmias11,18,42 or did not differenti- ticles were identified by reviewing the reentry tachycardia, atrioventricular re- ate between the 2.4,12 reference lists of retrieved articles and entry tachycardia, atrial fibrillation, Seven studies examined the utility of expert suggestions.15,16,40 atrial flutter, atrial tachycardia, junc- the features on history for diagnosing Articles reporting original empiri- tional tachycardia, or ventricular ecto- an arrhythmia as the cause of palpita- cal studies evaluating historical fea- pic beats occurring in salvos. tions (TABLE 2). Most of the data are tures, physical examination, or diag- The yield of the various diagnostic obtained from studies with small sample nostic tests against a reference standard tests was calculated and defined as the sizes. Although several features in- for the diagnosis of palpitations sec- number of patients who had any ar- crease the likelihood that a patient’s ondary to an arrhythmia were in- rhythmia or clinically significant ar- palpitations were secondary to an ar- cluded. Acceptable reference stan- rhythmia during monitoring. When- rhythmia, most have 95% CIs cross- dards included clinical event monitors, ever available, separate data were ing unity and thus may not be clini- intermittent event recorders, implant- provided for the subgroup of patients cally useful. able loop recorders, in-hospital telem- who had their typical symptoms dur- The only findings with an LR of 2.00 etry, 12-lead electrocardiographic ing the monitoring period. or greater for any arrhythmia were a his- monitoring during symptoms, or elec- Articles were graded for method- tory of cardiac disease (LR, 2.03; 95% trophysiological study. Excluded stud- ological quality using standard meth- CI, 1.33-3.11) and palpitations af- ies (1) focused primarily on nonar- ods with a threshold of more than 100 fected by sleeping (LR, 2.29; 95% CI, rhythmic diagnoses in patients with patients distinguishing level 1 from level 1.33-3.94; which are presumably pal- palpitations; (2) enrolled patients with 2 studies.41 pitations that are severe enough to wake

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA patients up from sleep) or while the pa- ever, a second study11 found that regular (LR, 1.52; 95% CI, 1.04-2.24) tient was at work (LR, 2.17; 95% CI, neck fullness is not useful for distin- or those that were affected by sleeping 1.19-3.96). Although description of pal- guishing atrioventricular node reen- (LR, 1.83; 95% CI, 1.03-3.27). pitations as either regular (LR, 1.66; try tachycardia from other arrhyth- Although no study specifically 95% CI, 1.20-2.29) or irregular (LR, mias (LR, 0.85; 95% CI, 0.44-1.64), assessed the accuracy of the associ- 1.65; 95% CI, 1.22-2.22) had little value but the presence of visible neck pul- ated symptom of shirt flapping in in the likelihood of cardiac arrhyth- sations may be useful (LR, 2.68; 95% patients with palpitations, 1 study19 mia, this information may be helpful in CI, 1.25-5.78). More recently, the found that in 326 patients with docu- the right context because certain ar- description of palpitations in the mented arrhythmias, the proportion rhythmias are typically regular while neck in patients with documented of patients with atrioventricular node others are irregular. narrow complex tachycardia was reentry tachycardia (58%) who The 2 factors with an LR of 0.50 or shown to distinguish atrioventricular reported shirt flapping was greater less for any arrhythmia were an under- node reentry tachycardia from atrio- than that reported by patients with lying history of panic disorder (LR, ventricular reentry tachycardia with other arrhythmias such as atrioven- 0.26; 95% CI, 0.07-1.01) and duration an LR of 2.41 (95% CI, 1.54-3.76).10 tricular reentry tachycardia (44%), of palpitation less than 5 minutes (LR, From the reviewed studies, no other ventricular tachycardia (32%), atrial 0.38; 95% CI, 0.22-0.63) (Table 2). features appear to be useful for rul- flutter (17%), and atrial fibrillation However, these observations are based ing in or ruling out a clinically sig- (13%). on single studies and the upper bound nificant arrhythmia. Based on a single study6 (TABLE 3), of the 95% CI for a history of panic dis- In these 7 studies,4-6,11,12,18,42 all pa- the presence of resting bradycardia order included 1.00. tients including those who had no (Ͻ60/min) during the examination in- The presence of an associated symptoms during the monitored pe- creases the likelihood of a clinically sig- regular rapid-pounding sensation in riod were included in the accuracy nificant arrhythmia (LR, 3.00; 95% CI, the neck (LR, 177; 95% CI, 25-1251) analysis. Only 1 study provided data on 1.27-7.08). No other physical exami- increased the likelihood that the a subgroup of 81 patients who had their nation findings, including the pres- patient’s symptoms of palpitations typical symptoms during the monitor- ence of murmurs, have been evalu- are due to atrioventricular node reen- ing period.5 For these patients, the most ated in patients presenting with try tachycardia.18 The absence of an useful feature for detecting an arrhyth- palpitations associated regular rapid-pounding mia was the occurrence of palpita- No studies reported on the sensitiv- sensation in the neck significantly tions at work (LR, 2.38; 95% CI, 1.03- ity and specificity of baseline 12-lead decreased the likelihood of atrioven- 5.50). Potentially useful features for electrocardiographic abnormalities in tricular node reentry tachycardia detecting clinically significant arrhyth- predicting a cardiac arrhythmia as a (LR, 0.07; 95% CI, 0.03-0.19). How- mias include palpitations described as cause of symptoms. Nevertheless, base-

Table 1. Accuracy Studies Assessing Clinical Features for the Diagnosis of a Cardiac Arrhythmia in Patients With Palpitations No. of Any Arrhythmia, Study Sourcea Patients Patient Population Reference Standard No. (%) Levelb Gürsoy et al,18 1992 244 Patients with palpitations referred for Electrophysiological study NA 3 electrophysiological study Barsky et al,42 1994 145 Patients referred for ambulatory monitoring 24-h Holter monitoring 28 (19) 3 due to palpitations Barsky et al,12 1994 131 Patients referred for ambulatory monitoring 24-h Holter monitoring 27 (21) 3 due to palpitations Weber and Kapoor,4 1996 190 Patients with palpitations at least once in Electrocardiography, 75 (39) 4 the past 3 mo presenting to Holter monitoring, emergency department, general telemetry, loop monitoring, practitioner, or admitted to hospital electrophysiological study Summerton et al,5 2001 139 Patients with palpitations in past 3 mo Event recorder 42 (30) 1 presenting to general practitioner Hoefman et al,6 2007 127 Patients with history of palpitations and/or Loop monitoring 83 (65) 1 light-headedness presenting to a general practitioner Sakhuja et al,11 2009 239 Patients with palpitations referred for Electrophysiological study, 224 (94) 4 electrophysiological study, telemetry, Holter cardioversion, or ablation monitoring, 12-lead electrocardiography Abbreviation: NA, data not available. a All studies were of prospective cohort design. b Methodologic quality of studies determined using standard methods with a threshold of more than 100 patients distinguishing level 1 from level 2 studies.41

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA line electrocardiography is typically per- A total of 16 studies4-6,12,34,37,42,44-52 pro- agnostic yield refers to the occurrence formed prior to other diagnostic vided diagnostic yield data (eTable 1 is of a cardiac arrhythmia detected dur- tests.39,43 available at http://www.jama.com). Di- ing the monitoring period in either all

Table 2. Accuracy of Clinical Features for the Diagnosis of Arrhythmia Likelihood Ratio (95% CI)

Any Arrhythmiaa Significant Arrhythmia No./Total Finding Source (%) Positive Negative Positive Negative Demographic and historical features Cardiac disease Hoefman et al6 NA NA NA 0.42 (0.06-3.06) 1.07 (0.96-1.20) Weber and Kapoor4 61/190 (32) 2.03 (1.33-3.11) 0.71 (0.57-0.88) NA NA Male sex Weber and Kapoor4 74/190 (39) 1.73 (1.21-2.48) 0.69 (0.54-0.89) NA NA Summerton et al5 46/139 (33) 1.63 (1.02-2.58) 0.76 (0.56-1.03) 1.20 (0.69-2.11) 0.90 (0.65-1.26) Hoefman et al6 33/127 (26) NA NA 1.37 (0.71-2.66) 0.88 (0.65-1.19) Age Ͼ60 y Summerton et al5 33/139 (24) 1.70 (0.95-3.06) 0.83 (0.66-1.05) 1.89 (1.03-3.47) 0.77 (0.56-1.06) Hoefman et al6 32/127 (25) NA NA 1.43 (0.74-2.78) 0.87 (0.64-1.18) Smoking (Ͼ11 cigarettes/d) Summerton et al5 16/133 (12) 0.78 (0.27-2.26) 1.03 (0.91-1.17) 0.77 (0.26-2.25) 1.03 (0.91-1.17) Anxiety disorder Summerton et al5 46/133 (35) 0.98 (0.59-1.63) 1.01 (0.77-1.32) 0.92 (0.49-1.72) 1.04 (0.78-1.39) Family history of palpitations Summerton et al5 29/133 (22) 0.86 (0.41-1.77) 1.04 (0.87-1.26) 1.07 (0.49-2.37) 0.98 (0.78-1.24) Alcohol use (Ͼ10 drinks/wk) Summerton et al5 21/130 (16) 0.76 (0.30-1.92) 1.05 (0.90-1.23) 1.02 (0.38-2.79) 1.00 (0.83-1.19) Panic disorder Barsky et al12 32/131 (24) 0.26 (0.07-1.01) 1.30 (1.10-1.53) NA NA Any psychiatric disorder Barsky et al42 36/145 (25) NA NA 0.67 (0.29-1.58) 1.12 (0.91-1.37) Description of palpitations Regular Summerton et al5 67/139 (48) 1.66 (1.20-2.29) NA 1.38 (1.02-1.86) 0.55 (0.27-1.14) Irregular Weber and Kapoor4 90/190 (47) 1.65 (1.22-2.22) 0.62 (0.46-0.84) NA NA Duration Ͼ5 min Hoefman et al6 50/101 (50) NA NA 0.79 (0.46-1.36) 1.23 (0.81-1.86) Weber and Kapoor4 127/190 (67) 1.52 (1.25-1.85) 0.38 (0.22-0.63) NA NA Duration Ͼ60 s Hoefman et al6 77/101 (76) NA NA 1.02 (0.79-1.32) 0.95 (0.40-2.24) Summerton et al5 95/135 (70) 1.15 (0.93-1.43) 0.69 (0.36-1.31) 1.17 (0.93-1.48) 0.63 (0.27-1.44) Continuous symptoms Summerton et al5 102/139 (73) 1.06 (0.86-1.30) NA 0.93 (0.71-1.23) 1.20 (0.62-2.31) Heart rate Ͼ100/min Summerton et al5 71/139 (51) 0.91 (0.63-1.31) NA 1.08 (0.78-1.50) 0.86 (0.44-1.68) Patient setting during palpitations or precipitating factors Affected by sleeping Summerton et al5 36/138 (26) 2.29 (1.33-3.94) 0.70 (0.53-0.93) 2.44 (1.43-4.14) 0.63 (0.42-0.93) Occurring at work Summerton et al5 31/136 (23) 2.17 (1.19-3.96) 0.76 (0.60-0.98) 1.54 (0.78-3.04) 0.86 (0.65-1.14) Affected by caffeine Summerton et al5 16/138 (12) 1.84 (0.74-4.61) 0.91 (0.79-1.07) 2.06 (0.78-5.39) 0.89 (0.72-1.09) Occurring during holiday Summerton et al5 20/137 (15) 1.56 (0.69-3.53) 0.92 (0.78-1.09) 0.79 (0.25-2.49) 1.04 (0.88-1.24) Occurring during weekend Summerton et al5 29/137 (21) 1.43 (0.74-2.76) 0.90 (0.73-1.11) 0.72 (0.27-1.88) 1.08 (0.89-1.32) Affected by alcohol Summerton et al5 16/137 (12) 1.36 (0.53-3.49) 0.96 (0.83-1.10) 1.94 (0.74-5.12) 0.90 (0.74-1.09) While lying in bed Summerton et al5 84/137 (61) 1.30 (1.01-1.68) 0.61 (0.35-1.07) 1.02 (0.73-1.44) 0.97 (0.57-1.65) Affected by exercise Summerton et al5 33/139 (24) 0.74 (0.36-1.50) 1.09 (0.90-1.32) 0.78 (0.33-1.82) 1.07 (0.87-1.33) Affected by breathing Summerton et al5 38/138 (28) 0.52 (0.25-1.08) 1.23 (1.02-1.49) 0.52 (0.20-1.33) 1.20 (0.98-1.48) While resting Hoefman et al6 71/122 (58) NA NA 1.02 (0.69-1.50) 0.97 (0.56-1.69) Associated symptoms Regular rapid-pounding Gürsoy et al18 50/190 (26) NA NA 177 (25-1251) 0.07 (0.03-0.19) sensation in neck Neck fullness Sakhuja et al11 48/239 (20) NA NA 0.85 (0.44-1.64) 1.04 (0.90-1.20) Visible neck pulsations Sakhuja et al11 23/239 (10) NA NA 2.68 (1.25-5.78) 0.87 (0.76-1.00) Dizzy spells Summerton et al5 72/137 (53) 0.93 (0.65-1.33) 1.08 (0.75-1.57) 1.34 (0.96-1.88) 0.67 (0.39-1.18) Chest pain Summerton et al5 34/137 (25) 0.81 (0.42-1.59) 1.07 (0.87-1.30) 0.92 (0.42-1.98) 1.02 (0.81-1.30) Dyspnea Summerton et al5 17/139 (12) 0.31 (0.07-1.29) NA 0.27 (0.04-1.96) 1.12 (1.01-1.25) Vasovagal symptoms Hoefman et al6 49/127 (39) NA NA 1.72 (1.11-2.65) 0.63 (0.39-1.03) (pale and/or sweaty) Presyncope Hoefman et al6 72/127 (57) NA NA 1.04 (0.71-1.51) 0.95 (0.57-1.61) Abbreviations: CI, confidence interval; NA, data not available or cannot be calculated. aIncludes clinically significant and nonsignificant arrhythmias (premature atrial or ventricular contractions and sinus tachycardia).

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA of the patients included in a study or Table 3. Accuracy of Physical Examination Features for the Diagnosis of Arrhythmias in the only in the subgroup of patients who Study by Hoefman et al6 had symptoms during monitoring. Significant Arrhythmia, The diagnostic yield of 12-lead LR (95% CI) No./Total electrocardiography performed dur- Physical Examination (%) Positive Negative 4,6 ing symptoms ranged from 3% to 26% Abnormal heart rate Ͻ60/min 17/127 (13) 3.00 (1.27-7.08) 0.78 (0.60-1.02) for any arrhythmias and 2%6 for clini- or Ͼ100/min cally significant arrhythmias. The yield Obesity 15/126 (12) 1.55 (0.54-4.44) 0.93 (0.77-1.13) of the 24-hour Holter monitoring was Hypertension on examination 42/127 (33) 1.01 (0.54-1.90) 1.00 (0.73-1.36) 34%44 for any arrhythmia and ranged Abbreviations: CI, confidence interval; LR, likelihood ratio. between 3% and 24%44,45 for clinically significant arrhythmias. One study46 used a 48-hour Holter monitor as a ref- Our review focused on studies of tient is 40%. While the nonspecific de- erence standard and had a diagnostic patients presenting with palpitations scription of symptoms as “heart rac- yield of 21% for any arrhythmia and 0% rather than patients with conditions, ing” and the inability to tell the cadence for clinically significant arrhythmias. such as hyperthyroidism, who had of the rhythm are not helpful, the pre- The diagnostic yield for loop moni- palpitations as part of their symptom vious diagnosis of panic disorder (LR, tors ranged from 34%6 to 84%47 for any complex. As such, our results are 0.26) and the duration of less than 5 arrhythmia and from 8%48 to 36%47 for most relevant for patients without an minutes (LR, 0.38) decreases the like- clinically significant arrhythmias. A obvious underlying medical problem lihood of any arrhythmia. The post- 2-week loop recorder had a greater yield or structural heart disease that might test probability of any arrhythmia based than a 1-week recorder, however, using cause their palpitations. Also, the only on the patient’s previous history it for 3 weeks had minimal49 or no ad- data does not provide information of panic disorder would be 15%. The ditional yield.47 The use of loop record- about the specific kind of arrhythmia absence of a regular rapid-pounding ers with an automatic trigger function experienced. sensation in the neck decreases the like- for asymptomatic arrhythmias had a A significant proportion of the pa- lihood of atrioventricular node reen- slightly higher yield for clinically sig- tients in the included studies did not try tachycardia (LR, 0.07). nificant arrhythmias.34 The yield for have symptoms during the monitored While many physicians would not event recorders ranged from 30%5 to period and only a few features have pursue further testing initially be- 60%50 for any arrhythmia and from been evaluated in the 81 patients with cause of the relatively low posttest prob- 17%50 to 19%5 for a clinically signifi- symptoms during monitoring. There- ability, it is essential to recognize that cant arrhythmia (eTable 1). Among pa- fore, the majority of the LRs were cal- patients with panic disorders may also tients with typical symptoms during culated using all included patients have clinically significant arrhyth- monitoring (eTable 2), the yield of in- rather than only those with symptoms mias22,42; and it may be prudent to per- termittent event recorders was higher during monitoring. Similarly, except in form long-term electrocardiographic for any arrhythmia and for clinically sig- 3 studies (eTable 2),5,6,51 the yield data monitoring to rule out clinically significant arrhythmias.5,6,51 Based on 1 was extracted from studies in which nificant arrhythmias before attribut- study,37 the yield of implantable loop some of the patients did not experi- ing the symptoms to panic disorder. recorders was 73% for clinically sig- ence their symptoms during monitor- Structural cardiac assessment could be nificant arrhythmias during a mean ing. Because some of these patients may deferred unless the patient has persis- (SD) monitoring period of 279 (228) actually have an arrhythmia, we could tent palpitations, develops more alarm- days. have underestimated the true yield of ing symptoms such as syncope, or a the diagnostic tests. clinically significant arrhythmia is iden- LIMITATIONS OF Finally, there are no published data tified on electrocardiographic moni- PUBLISHED STUDIES evaluating combinations of features or toring. Only 7 studies provided data on diag- the precision of the clinical examina- nostic accuracy and only 2 of these5,6 tion. As a result, multiplying together BOTTOM LINE were of high methodological quality individual LRs or applying them se- When evaluating patients with palpi- (level 1). Many of the studies had small quentially may substantially overesti- tations, the presence of underlying sample sizes and in some cases data mate posttest probability. medical conditions should be care- from subgroups in single studies were fully considered. In the emergency de- relied on to evaluate the accuracy of SCENARIO RESOLUTION partment, primary cardiac diagnoses are clinical parameters. Therefore, cau- The pretest probability of any cardiac the most common reason for palpitation should be taken in interpreting our arrhythmia in this patient based on a tions (43%), but anxiety or panic dis- results. study4 that would have enrolled this pa- orders are also frequent (31%).4

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA

A known history of cardiac disease Acquisition of data: Thavendiranathan, Bagai, Khoo. dias in patients without pre-excitation. J Am Coll Analysis and interpretation of data: Thavendiranathan, Cardiol. 2009;53(25):2353-2358. (LR, 2.03; 95% CI, 1.33-3.11) and Bagai, Dorian, Choudhry. 11. Sakhuja R, Smith LM, Tseng ZH, et al. Test char- palpitations affected by sleeping (LR, Drafting of the manuscript: Thavendiranathan, Bagai, acteristics of neck fullness and witnessed neck pulsa- Dorian, Choudhry. tions in the diagnosis of typical AV nodal reentrant 2.29; 95% CI, 1.33-3.94) or while at Critical revision of the manuscript for important in- tachycardia. Clin Cardiol. 2009;32(8):E13-E18. work (LR, 2.17; 95% CI, 1.19-3.96) tellectual content: Thavendiranathan, Bagai, Khoo, 12. Barsky AJ, Cleary PD, Sarnie MK, Ruskin JN. Panic slightly increase the likelihood of a Choudhry. disorder, palpitations, and the awareness of cardiac Statistical analysis: Thavendiranathan, Bagai, activity. J Nerv Ment Dis. 1994;182(2):63-71. cardiac arrhythmia while palpitations Choudhry. 13. Hermida JS, Minassian A, Jarry G, et al. Familial lasting less than 5 minutes (LR, 0.38; Administrative, technical, or material support: incidence of late ventricular potentials and electro- Thavendiranathan, Khoo, Dorian, Choudhry. cardiographic abnormalities in arrhythmogenic right 95% CI, 0.22-0.63) and a known his- Study supervision: Dorian, Choudhry. ventricular dysplasia. Am J Cardiol. 1997;79(10): tory of panic disorder make the diag- Financial Disclosures: None reported. 1375-1380. nosis less likely (LR, 0.26; 95% CI, Additional Information: eTable 1 and eTable 2 and 14. Marcus GM, Smith LM, Vittinghoff E, et al. A first- eFigure 1 and eFigure 2 are available at http://www degree family history in lone atrial fibrillation patients. 0.07-1.01). The presence of a regular .jama.com. Heart Rhythm. 2008;5(6):826-830. rapid-pounding sensation in the neck Additional Contributions: We thank Emmy Hoef- 15. Abbott AV. Diagnostic approach to palpitations. man, MD (Department of General Practice, Aca- Am Fam Physician. 2005;71(4):743-750. as opposed to neck fullness in associa- demic Medical Center, University of Amsterdam, Am- 16. Zimetbaum P, Josephson ME. Evaluation of pa- tion with palpitations increases the sterdam, the Netherlands) for providing necessary tients with palpitations. N Engl J Med. 1998;338 information regarding her study; Luigi Casella, MD, (19):1369-1373. likelihood that the patient has atrio- and Chi Ming Chow, MD (both with the Division of 17. Abe H, Nagatomo T, Kobayashi H, et al. Neuro- ventricular node reentry tachycardia Cardiology, St Michael’s Hospital, University of humoral and hemodynamic mechanisms of diuresis (LR, 177; 95% CI, 25-1251), whereas Toronto, Toronto, Ontario, Canada) for providing the during atrioventricular nodal reentrant tachycardia. Pac- electrocardiogram strips; David Simel, MD, MHS (Duke ing Clin Electrophysiol. 1997;20(11):2783-2788. its absence makes atrioventricular University Medical Center and Durham Veteran Af- 18. Gürsoy S, Steurer G, Brugada J, Andries E, Brugada node reentry tachycardia less likely fairs Medical Center, Durham, North Carolina) for help- P. Brief report: the hemodynamic mechanism of pounding with question synthesis, data extraction, manu- ing in the neck in atrioventricular nodal reentrant (LR, 0.07; 95% CI, 0.03-0.19). The script composition, and editing; and Camilla Wong, tachycardia. N Engl J Med. 1992;327(11):772- presence of visible neck pulsations MD (Department of Geriatrics, St Michael’s Hospital, 774. University of Toronto, Toronto, Ontario, Canada) and 19. Laurent G, Leong-Poi H, Mangat I, et al. Influ- also increases the likelihood of atrio- Albert Sun, MD (Department of Cardiology, Duke Uni- ence of ventriculoatrial timing on hemodynamics ventricular node reentry tachycardia versity Medical Center, Durham, North Carolina) for and symptoms during supraventricular tachycardia reviewing and commenting on the manuscript. No fi- (LR, 2.68; 95% CI, 1.25-5.78). [published online ahead of print September 3, nancial compensation was received by any of the in- 2009]. J Cardiovasc Electrophysiol. 2008;20(2): dividuals listed in this section. Because of the limitations of the lit- 176-181. erature and the consequences of miss- 20. Goldreyer BN, Kastor JA, Kershbaum KL. The hemodynamic effects of induced supraventricular ing an important rhythm disturbance, REFERENCES tachycardia in man. Circulation. 1976;54(5):783- no clinical examination features ap- 1. Mayou R, Sprigings D, Birkhead J, Price J. Char- 789. pear to be sufficiently accurate to ex- acteristics of patients presenting to a cardiac clinic with 21. Leitch JW, Klein GJ, Yee R, Leather RA, Kim palpitation. QJM. 2003;96(2):115-123. YH. Syncope associated with supraventricular clude other clinically significant ar- 2. Barsky AJ, Ahern DK, Bailey ED, Delamater BA. Pre- tachycardia: an expression of tachycardia rate or rhythmias, especially in high-risk dictors of persistent palpitations and continued vasomotor response? Circulation. 1992;85(3): patients. Therefore, when a clinically medical utilization. J Fam Pract. 1996;42(5):465- 1064-1071. 472. 22. Lessmeier TJ, Gamperling D, Johnson-Liddon significant arrhythmia is suspected, fur- 3. Hoefman E, van Weert HC, Reitsma JB, Koster RW, V, et al. Unrecognized paroxysmal supraventricular ther testing including evaluating car- Bindels PJ. Diagnostic yield of patient-activated loop tachycardia: potential for misdiagnosis as panic recorders for detecting heart rhythm abnormalities in disorder. Arch Intern Med. 1997;157(5):537- diac structure with transthoracic ech- general practice: a randomised clinical trial. Fam Pract. 543. ocardiography and attempting to 2005;22(5):478-484. 23. Varma N, Josephson ME. Therapy of “idio- 4. Weber BE, Kapoor WN. Evaluation and outcomes pathic” ventricular tachycardia. J Cardiovasc establish symptom-rhythm correla- of patients with palpitations. AmJMed. 1996; Electrophysiol. 1997;8(1):104-116. tion with prolonged electrocardio- 100(2):138-148. 24. Coumel P. Clinical approach to paroxysmal graphic monitoring should be under- 5. Summerton N, Mann S, Rigby A, Petkar S, Dhawan atrial fibrillation. Clin Cardiol. 1990;13(3):209- J. New-onset palpitations in general practice: assess- 212. taken. The selection of the monitoring ing the discriminant value of items within the clinical 25. Lampert R, Joska T, Burg MM, Batsford WP, type depends on the frequency of the history. Fam Pract. 2001;18(4):383-392. McPherson CA, Jain D. Emotional and physical pre- 6. Hoefman E, Boer KR, van Weert HC, Reitsma JB, cipitants of ventricular arrhythmia. Circulation. symptoms. If the symptoms occur daily, Koster RW, Bindels PJ. Predictive value of history tak- 2002;106(14):1800-1805. a Holter monitor may be of reasonable ing and physical examination in diagnosing arrhyth- 26. Hansson A, Madsen-Hardig B, Olsson SB. mias in general practice. Fam Pract. 2007;24(6): Arrhythmia-provoking factors and symptoms at the diagnostic yield, with emphasis on the 636-641. onset of paroxysmal atrial fibrillation: a study based importance of accurate diary record- 7. Brugada P, Gürsoy S, Brugada J, Andries E. Inves- on interviews with 100 patients seeking hospital tigation of palpitations. Lancet. 1993;341(8855): assistance. BMC Cardiovasc Disord. 2004;4:13. ing. If symptoms occur more infre- 1254-1258. 27. De Ponti F, Poluzzi E, Cavalli A, Recanatini M, quently, an intermittent event re- 8. Barsky AJ, Cleary PD, Coeytaux RR, Ruskin JN. The Montanaro N. Safety of non-antiarrhythmic drugs corder such as a loop monitor is a more clinical course of palpitations in medical outpatients. that prolong the QT interval or induce torsade de Arch Intern Med. 1995;155(16):1782-1788. pointes: an overview. Drug Saf. 2002;25(4):263- appropriate test. 9. Porter MJ, Morton JB, Denman R, et al. Influence 286. of age and gender on the mechanism of supraven- 28. Yap YG, Camm AJ. Drug induced QT prolon- Author Contributions: Dr Thavendiranathan had full tricular tachycardia. Heart Rhythm. 2004;1(4): gation and torsades de pointes. Heart. 2003;89 access to all of the data in the study and takes re- 393-396. (11):1363-1372. sponsibility for the integrity of the data and the ac- 10. Gonza´ lez-Torrecilla E, Almendral J, Arenal A, et al. 29. Roden DM. Drug-induced prolongation of the curacy of the data analysis. Combined evaluation of bedside clinical variables and QT interval. N Engl J Med. 2004;350(10):1013- Study concept and design: Thavendiranathan, Bagai, the electrocardiogram for the differential diagnosis of 1022. Khoo, Dorian, Choudhry. paroxysmal atrioventricular reciprocating tachycar- 30. Ray WA, Murray KT, Meredith S, Narasimhulu

Downloaded from www.jama.com by guest on November 17, 2009 PALPITATIONS AND CARDIAC ARRHYTHMIA

SS, Hall K, Stein CM. Oral erythromycin and the Exercise Testing Guidelines). Circulation. 2002; 44. Scalvini S, Zanelli E, Martinelli G, Baratti D, Giordano risk of sudden death from cardiac causes. N Engl J 106(14):1883-1892. A, Glisenti F. Cardiac event recording yields more di- Med. 2004;351(11):1089-1096. 37. Giada F, Gulizia M, Francese M, et al. Recurrent agnoses than 24-hour Holter monitoring in patients 31. Cook DJ, Simel DL. Does this patient have unexplained palpitations (RUP) study comparison of with palpitations. J Telemed Telecare. 2005;11 abnormal central venous pressure? In: The Rational implantable loop recorder versus conventional diag- (suppl 1):14-16. Clinical Examination. New York, NY: McGraw-Hill; nostic strategy. J Am Coll Cardiol. 2007;49(19): 45. Barsky AJ, Ahern DK, Delamater BA, Clancy SA, 2009:125-135. 1951-1956. Bailey ED. Differential diagnosis of palpitations: pre- 32. Crawford MH, Bernstein SJ, Deedwania PC, et al. 38. Seidl K, Rameken M, Breunung S, et al; liminary development of a screening instrument. Arch ACC/AHA guidelines for ambulatory electrocardiog- Reveal-Investigators. Diagnostic assessment of recur- Fam Med. 1997;6(3):241-245. raphy: executive summary and recommendations: a rent unexplained syncope with a new subcutane- 46. Kinlay S, Leitch JW, Neil A, Chapman BL, Hardy report of the American College of Cardiology ously implantable loop recorder. Europace. 2000; DB, Fletcher PJ. Cardiac event recorders yield more /American Heart Association task force on practice 2(3):256-262. diagnoses and are more cost-effective than 48-hour guidelines (Committee to Revise the Guidelines for Am- 39. Blomström-Lundqvist C, Scheinman MM, Aliot Holter monitoring in patients with palpitations: a con- bulatory Electrocardiography). Circulation. 1999; EM, et al; American College of Cardiology; American trolled clinical trial. Ann Intern Med. 1996;124 100(8):886-893. Heart Association Task Force on Practice Guidelines; (1 pt 1):16-20. 33. Joshi AK, Kowey PR, Prystowsky EN, et al. First European Society of Cardiology Committee for Prac- 47. Zimetbaum PJ, Kim KY, Josephson ME, Goldberger experience with a Mobile Cardiac Outpatient Telem- tice Guidelines; Writing Committee to Develop Guide- AL, Cohen DJ. Diagnostic yield and optimal etry (MCOT) system for the diagnosis and manage- lines for the Management of Patients With Supraven- duration of continuous-loop event monitoring ment of cardiac arrhythmia. Am J Cardiol. 2005; tricular Arrhythmias. ACC/AHA/ESC guidelines for the for the diagnosis of palpitations: a cost-effective- 95(7):878-881. management of patients with supraventricular ar- ness analysis. Ann Intern Med. 1998;128(11):890- 34. Olson JA, Fouts AM, Padanilam BJ, Prystowsky rhythmias–executive summary: a report of the Ameri- 895. EN. Utility of mobile cardiac outpatient telemetry for can College of Cardiology/American Heart Associa- 48. Wu CC, Hsieh MH, Tai CT, et al. Utility of patient- the diagnosis of palpitations, presyncope, syncope, and tion Task Force on Practice Guidelines and the European activated cardiac event recorders in the detection of the assessment of therapy efficacy. J Cardiovasc Society of Cardiology Committee for Practice Guide- cardiac arrhythmias. J Interv Card Electrophysiol. 2003; Electrophysiol. 2007;18(5):473-477. lines (Writing Committee to Develop Guidelines for 8(2):117-120. 35. Zipes DP, DiMarco JP, Gillette PC, et al. Guide- the Management of Patients With Supraventricular 49. Hoefman E, van Weert HC, Boer KR, Reitsma J, lines for clinical intracardiac electrophysiological and Arrhythmias). Circulation. 2003;108(15):1871- Koster RW, Bindels PJ. Optimal duration of event re- catheter ablation procedures: a report of the Ameri- 1909. cording for diagnosis of arrhythmias in patients with can College of Cardiology/American Heart Associa- 40. Zwietering PJ, Knottnerus JA, Rinkens PE, Kleijne palpitations and light-headedness in the general tion Task Force on Practice Guidelines (Committee on MA, Gorgels AP. Arrhythmias in general practice: di- practice. Fam Pract. 2007;24(1):11-13. Clinical Intracardiac Electrophysiologic and Catheter agnostic value of patient characteristics, medical his- 50. Arjona Barrionuevo JdeD, Baro´ n-Esquivias G, Ablation Procedures), developed in collaboration tory and symptoms. Fam Pract. 1998;15(4):343- Nu´n˜ ez Rodrı´guez A, et al. Utility of cardiac event with the North American Society of Pacing and 353. recorders in diagnosing arrhythmic etiology of Electrophysiology. J Am Coll Cardiol. 1995;26 41. Simel DL, Keitz S. Update: primer on precision and palpitations in patients without structural heart (2):555-573. accuracy. In: The Rational Clinical Examination. New disease. Rev Esp Cardiol. 2002;55(2):107-112. 36. Gibbons RJ, Balady GJ, Bricker JT, et al; Ameri- York, NY: McGraw-Hill; 2009:9-16. 51. Zimetbaum P, Kim KY, Ho KK, Zebede J, Josephson can College of Cardiology/American Heart Associa- 42. Barsky AJ, Cleary PD, Coeytaux RR, Ruskin JN. ME, Goldberger AL. Utility of patient-activated car- tion Task Force on Practice Guidelines (Committee to Psychiatric disorders in medical outpatients complain- diac event recorders in general clinical practice. Am J Update the 1997 Exercise Testing Guidelines). ACC ing of palpitations. J Gen Intern Med. 1994;9(6): Cardiol. 1997;79(3):371-372. /AHA 2002 guideline update for exercise testing: sum- 306-313. 52. Fogel RI, Evans JJ, Prystowsky EN. Utility and cost mary article: a report of the American College of Car- 43. Rutten FH, Kessels AG, Willems FF, Hoes AW. of event recorders in the diagnosis of palpitations, pre- diology/American Heart Association Task Force on Electrocardiography in primary care: is it useful? Int J syncope, and syncope. Am J Cardiol. 1997;79 Practice Guidelines (Committee to Update the 1997 Cardiol. 2000;74(2-3):199-205. (2):207-208.

Downloaded from www.jama.com by guest on November 17, 2009