The Prevalence and Prognostic Significance of Near Syncope and Syncope
A prospective study of 395 cases in an emergency department (the SPEED Study)
; ; ; ; ; ; ; ; ;
Background: The prognostic significance of near-syncope has not yet been adequately characterized.
Methods: We collected prospective data on a consecutive series of patients seen in an emergency department with syncope (brief loss of consciousness, usually with loss of muscle tone) or near-syncope (a feeling that syncope is about to occur, but without actual loss of consciousness or muscle tone). We report on the prevalence, etiology, and prognosis of such events (the SPEED study). Patients were followed up at 30 days and at 6 months after the event.
Results: From 17 July to 31 October 2011, 395 patients were seen in the emergency department for a chief complaint of syncope or near-syncope (3% of all emergency patients). Their median age was 70 years, and 55% were men. 62% had experienced syncope, and 38% near-syncope. The patients with near-syncope were younger than those with syncope ( 63 vs. 72 years, p < 0.014) and were also more commonly male (63% vs. 49%, p = 0.006). The two patient groups did not differ significantly with respect to their measured laboratory values and vital parameters or their accompanying medical conditions. Hospitalizations were more common for syncope than for near-syncope (86% vs. 70%, p < 0.001). Etiologies were similarly distributed in the two patient groups, with the main ones being reflex syncope, orthostatic syncope, cardiac syncope, and syncope of uncertain origin. In all, 123 of 379 patients (32%) had further undesired events within 30 days of the event. Multivariable logistic regression revealed that age, heart rate, and renal dysfunction were independent predictors of undesired events, while the type of syncope was not.
Conclusion: Patients with near-syncope do not differ to any large extent from patients with syncope with respect to the features studied. The diagnostic evaluation should be similar for patients in the two groups.
Syncope is defined as a transient loss of consciousness caused by short-term global cerebral underperfusion, which can be accompanied by a loss of muscle tone (1, 2). Although many causes of syncope are benign and self-limiting, some events are associated with significant morbidity and mortality (1, 3–5). Up to 3% of emergency department consultations are due to syncope (6–8), and in Germany more than 70% of the affected patients are admitted as inpatients (5). During a 30-day follow-up period, serious events such as a need for intervention (e.g. cardioversion, defibrillator or pacemaker implantation, endoscopic intervention) or death are common (occurring in between 5 and 23% of cases) (4, 9–13).
Although current guidelines provide detailed procedures for syncope (1, 2), this is not the case for near syncope. In near syncope, patients report “impending syncope” but do not lose consciousness or muscle tone. The fact that there is no clear definition of near syncope leads to a lack of clarity in diagnosis; data on the frequency and prognosis of near syncope is therefore lacking. In individual cases patients with near syncope have been enrolled in syncope studies (6, 14–18). Patients with near syncope are probably excluded from most studies as a result of the lack of a clear definition of it (4, 8, 9, 19).
In clinical practice patients are often asked whether they lost consciousness completely. The authors have questioned their colleagues on this subject, and the prevailing opinion among those asked was that not having lost consciousness was associated with a more favorable prognosis in this population of patients. This investigation aimed to analyze the frequency and causes of near syncope in a German emergency department and the significance of these for prognosis.
Study design, patients
Between 17 July and 31 October 2011 the authors gathered prospective data in order to evaluate consecutive patients with syncope or near syncope (the Syncope Patient Evaluation in the Emergency Department [SPEED] Study) in the emergency department of a hospital providing the maximum range of services (with approximately 90 000 emergency patients per year). No study-specific interventions were performed. Data collection was qualitative and took data protection into account, and patients gave their consent using the General Conditions of Contract (Clause 14, paragraph 4). In this form the project was rated ethically sound by the Ethics Committee and was registered with clinicaltrials.gov (ID: NCT01916070). For follow-up, patients’ consent to oral questioning was obtained. The Declaration of Helsinki was followed.
The variables recorded in the research were those stated in current recommendations (1, 2, 4, 5). Concomitant diseases were assessed using the Charlson Comorbidity Index (CCI) (20, 21), and incompletely documented cases and cases of loss of consciousness unrelated to syncope were excluded (eFigure 1). The following diagnostic steps were taken according to standard procedure in all patients (5, 22):
- Clinical history taken
- Physical examination
- Vital signs recorded (including 12-lead ECG)
- Routine laboratory tests:
– Blood count without differential
– Sodium level
– Potassium level
– Glucose level
– Urea level
– Creatinine level
– International normalized ratio (INR)
– Activated partial thromboplastin time (aPTT)
All other examinations were performed as indicated. Data was extracted prospectively by a study assistant and entered into an electronic database.
All patients aged 18 years or over who presented to the emergency department with syncope or near syncope were included. The definition of syncope is short-term loss of consciousness as part of transient global cerebral underperfusion, accompanied by loss of muscle tone (1). Patients with near syncope report impending syncope with loss of consciousness or muscle tone but do not meet the criteria for syncope (15, 23).
All patients who had a progressive disorder of consciousness, had taken mind-altering substances, or presented with loss of consciousness unrelated to syncope (e.g. hyperventilation, hypoglycemia, epileptic seizure) were excluded (1).
The final, gold-standard diagnosis was established independently by two experienced emergency physicians who were familiar with all patient records. A substantial lack of consensus is to be expected in the diagnosis of near syncope in particular, as there are no clear criteria defining it. The agreement between the two physicians was 90% (kappa 0.77). Agreement on the etiology of syncope was 69% (kappa 0.59). In cases of disagreement between the two emergency physicians, the final gold-standard diagnosis was established by a cardiologist.
In addition to describing the characteristics of the patient population, clinical endpoints occurring within 30 days following the index event were quantified and named (the combined endpoint was need for invasive interventions or serious adverse events) (5, 6, 24–26). The available treatment records were consulted for this purpose during follow-up. The prospectively defined endpoints were as follows:
- Sudden death
- Myocardial infarction
- Intervention on coronary vessels or heart valves
- Implantation of pacemaker or automatic implantable cardioverter-defibrillator (AICD)
- Electrical cardioversion
- High-grade valvular heart disease
- Dangerous arrhythmia
- Decompensated heart failure
Non-cardiac endpoints included the following parameters:
- Endoscopy with intervention
- Blood transfusion
- Aortic dissection
- Pulmonary embolism
- Severe infection
- Craniocerebral trauma
- Repeat presentation within 30 days
- Severe electrolyte disorders
- Acute kidney failure
- Other life-threatening events (5)
Follow-up questions were asked by telephone 30 days and 6 months after the index event by a study assistant, using a structured questionnaire. If individual patients could not be reached, questions were asked by mail.
The recorded data are presented as frequencies, mean and standard deviation, 95% confidence interval or median, and first and third quartile (Q1, Q3). The Shapiro–Wilk test was used to test for normal distribution; normally distributed data was compared using Student’s t-test for independent samples, and non-normally distributed data using nonparametric tests (the U-test).
Discrete data was compared using the chi-square test. The p-values shown here make clear the differences between the prospectively defined groups. They have not undergone Bonferroni correction and should not be interpreted as significance tests in the narrow sense. Survival analyses were performed using Kaplan–Meier analysis. Statistical tests, including univariate and multivariable logistic regression analysis, were performed using standard software (SPSS version 20).
Of a total of 12 872 patients who presented to the emergency department during the study period, 408 were screened. Of these, 13 were excluded (patients who did not meet the criteria for syncope or near syncope, and incompletely documented cases; eFigure 1). The symptoms of the remaining 395 patients met the criteria for syncope (n = 242) or near syncope (n = 153).
Demographic characteristics are summarized in Table 1 (median age: 70 years; 48% of patients were aged over 70 years [n = 191] and 24% over 80 years [n = 93]; 55% were male ([n = 215]). The proportion of men was higher among near syncope patients than among syncope patients (63% [n = 97] versus 49% [n = 119], p = 0.006). There were no differences in terms of comorbidities, vital signs, or laboratory parameters (Table 1). Patients with near syncope were more often admitted by physicians in private practice (n = 28; 18%) or paramedics (n = 45, 29%), while those with syncope were more often admitted by emergency physicians (n = 117; 48%) (Figure 1). The underlying causes were comparable in both patient groups (eFigure 2). 86% of patients with syncope (n = 209) and 71% of those with near syncope (n = 109) were admitted as inpatients after presenting to the emergency department (p <0.001) (Figure 2). During the 30-day follow-up period an adverse event occurred in 82 syncope patients (34%) and 41 near syncope patients (27%) (combined endpoint; p = 0.15). The endpoint death after 30 days occurred in 2% of cases in both patient groups (Figure 2).
At 30 days, the health of 379 patients (96%) was successfully inquired after. The two patient groups did not differ in terms of the type of endpoint in 30-day follow-up (Table 2). Pacemaker or AICD implantation was indicated due to arrhythmia in 4% (n = 16) of the total cohort. The most frequent non-cardiac endpoint was a diagnosis of severe infection or sepsis, which occurred in 5% of patients (n = 20). One patient (0.3%) died as a result of acute myocardial infarction during follow-up.
The probability of an adverse event after 30 days was 34% (82 of 242) in the syncope group and 27% (41 of 153; p = 0.135) in the near syncope group. 6% of patients with syncope (n = 14) and 5% of those with near syncope (n = 8; p = 0.44) died within a follow-up period of six months following the index event (95% response rate) (Figure 3a). The combined endpoint had occurred in 149 patients (38%) after six months. The probability of event-free survival after six months was 58% (n = 134) for syncope patients, and 67% (n = 95; p = 0.09) in patients with incomplete loss of consciousness (Figure 3b).
Predictive factors for onset of the combined endpoint within a 30-day follow-up period were analyzed using univariate and multivariable logistic regression analysis (Table 3). While the type of syncope is no predictor of the occurrence of an endpoint, age, heart rate, and kidney function are independent predictive factors for an adverse event during follow-up.
Duration, follow-up care
The follow-up care received by syncope and near syncope patients following their emergency department treatment can be seen in eFigure 3. 51 of the 395 patients (13%) were discharged to outpatient care following their emergency department treatment. Of the 315 (80%) patients admitted as inpatients, 121 (31%) were treated in one of the follow-up units attached to the emergency department, 197 (62%) patients received further inpatient treatment in other specialized units, and 25 (6%) refused further treatment. Syncope patients were hospitalized for longer (median: three days [Q1 to Q3; 1–7]) than near syncope patients (median: one day [Q1 to Q3; 0–7; p = 0.006]).
The most important results of this prospective study, which was conducted in the emergency department of a hospital providing the maximum range of services and examined the frequency and prognostic significance of near syncope, can be summarized as follows:
- The constellation of symptoms “syncope” or “near syncope” is common (accounting for 3% of all emergency department patients). Of these patients, 62% lose consciousness completely, while the other 38% experience near syncope.
- A serious adverse event occurs within 30 days following the index event in 34% and 27% of patients with syncope and near syncope respectively; 2% of syncope patients and near syncope patients die within 30 days.
- There is no difference in the frequency or type of endpoint within the 30-day follow-up period between syncope and near syncope patients. After six-month follow-up there was a trend towards less favorable prognosis for syncope patients.
The adverse event rate in syncope and near syncope patients in this analysis, 32%, is high. Italian working groups report an adverse event rate of 5 to 6% after 10 days (13, 27), while US emergency physicians state a rate of 11.5% after seven days (6). These differences may be due to the differing structures of the health-care systems concerned (5): Germany’s two-tiered specialized physician structure means that fewer patients with only mild medical problems present to the emergency departments of facilities that provide the maximum range of services (5), whereas in other countries emergency departments dispense primary care (28–30). Because of the differing prevalence of adverse events, diagnosis strategies developed in other systems cannot be transferred to Germany (1, 9, 31).
A cohort study conducted in Boston, USA shows comparable adverse event rates (4). It is possible that the adverse event rate found in this study, which is higher than in an earlier publication by the authors’ own working group (5), can be explained by syncope as a constellation of symptoms not having been identified retrospectively in cases in which other diagnoses were given (e.g. syncope in third-degree AV block). This cohort study was therefore able to examine the patient population of a German hospital providing the maximum range of services and to draw valuable conclusions on the challenges faced by physicians in clinical practice in an emergency department.
Grossman et al. were the first to investigate the prognostic significance of near syncope (15): although patients with near syncope were less likely to be hospitalized than those with syncope (49% versus 69%; p = 0.001), their frequency of serious adverse events was similar (20% versus 23%; p = 0.40). In one smaller study (n = 93) the adverse event rate in near syncope patients was twice as high as in syncope patients (30% versus 17%; ).
Our results confirm those of the analysis by Grossman et al. (14, 15). The significantly lower hospitalization rate of near syncope patients may cause the probability of adverse events in them being underestimated. However, an absence of loss of consciousness is certainly not associated with better prognosis and requires the same level of thoroughness in diagnosis as complete loss of consciousness. When the duration of cerebral underperfusion is shorter, e.g. thanks to appropriate counteracting measures being taken promptly, complete loss of consciousness does not occur. Although the comorbidities of younger patients are comparable to those of older patients, better compensation abilities may explain incomplete loss of consciousness (14, 15).
Our results show that emergency department patients present different medical problems than those of specialized clinics (31, 32). Standardized syncope diagnosis strategies that are suitable for specialized syncope units or cardiology clinics (31, 32) cannot necessarily be transferred to the challenges that arise with emergency department patients. A suspected benign etiology for syncope, such as reflex syncope (neurally mediated syncope caused by inadequate cardiovascular reflex responses [33, 34]), must be critically queried in the context of emergency care in an emergency department (5, 15). For older patients in particular, reflex syncope should be considered a symptom that may potentially indicate a serious disease and therefore requires careful evaluation (35–37). This is also true for near syncope patients: 8% (n = 12) of reflex syncope patients present a severe infection, and 15% (n = 12) suffer other serious adverse events during short-term (30-day) follow-up. In summary, it can be established that the benign prognosis postulated for reflex syncope does not hold true for the heterogeneous patient population of an emergency department (33, 34, 38, 39).
This study has the following limitations:
The study was single-center, so the extent to which its findings can be applied generally must be interpreted critically. The reproducibility of earlier findings, the prospective nature of the study, its large patient population, and the completeness of follow-up (95% at six months) indicate high data quality and highlight how pertinent the analysis is.
Case numbers were not calculated prospectively. The fact that there was no statistical difference in the primary endpoint between the two types of syncope examined cannot be taken as indicating clinical equivalence. Ideally, this might be investigated in the future, in valid, multicenter research.
Because the definition of near syncope is unclear, patients suffering from general reduced performance may have been included erroneously. However, the fact that two independent physicians with access to all examination findings during the follow-up period agreed on the gold-standard diagnosis leads us to conclude that data quality was high. In this respect this study contrasts favorably with many others.
The 30-day adverse event rate in patients presenting to an emergency department in a hospital providing the maximum range of services with syncope or near syncope—32%—is very high. The prognosis for near syncope was not substantially different from the prognosis for syncope. The types of serious adverse events are also comparable. In order to avoid potentially endangering patients with near syncope, its cause should therefore be just as comprehensively investigated as that of syncope.
The authors would like to thank all employees of the Emergency Department at Nuremberg Hospital for their assistance with patient recruitment, and Ms. Rafaela Kljaic for her help in collecting data.
Conflict of interest statement
Prof. Christ has received lecture fees from Roche, Novartis Pharma, Alere, and Radiometer. Roche and Novartis Pharma have funded research projects of Prof. Christ’s.
Prof. Müller has received trial funding (third-party funds) from Brahms, the Swiss National Science Foundation, and the Swiss Heart Foundation.
Ms. Greve, Ms. Geier, Dr. Popp, Prof. Bertsch, PD Dr. Singler, Dipl.-Sozialwirt Meier, Mr. Smolarsky and Prof. Mang declare that no conflict of interest exists.
Manuscript received on 9 August 2013, revised version accepted on 21 January 2014.
Prof. Dr. med. Michael Christ
Klinik für Notfall- und Internistische Intensivmedizin
Prof. Ernst Nathan Str. 1
90419 Nürnberg, Germany
Greve; Geier; Popp, M.D.; Smolarsky; Prof. Christ
Institute for Clinical Chemistry, Laboratory Medicine, and Transfusion Medicine, Nuremberg Hospital: Prof. Bertsch
Master Program M. Sc. Medical Process Management, Friedrich-Alexander-University Erlangen-Nürnberg:
Geier, Prof. Mang
Department of Geriatrics, Nuremberg Hospital: Singler, M.D.
Chairholder at the Department of Health Economics, Friedrich-Alexander-University Erlangen-Nürnberg: Meier
Departement of Internal Medicine, Universitätsspital Basel, Switzerland: Prof. Müller
|1.||Moya A, Sutton R, Ammirati F, et al.: Guidelines for the diagnosis and management of syncope (version 2009). Eur Heart J 2009; 30: 2631–71. CrossRef MEDLINE PubMed Central|
|2.||Smars PA, Decker WW, Shen WK: Syncope evaluation in the emergency department. Curr Opin Cardiol 2007; 22: 44–8. MEDLINE|
|3.||Huff JS, Decker WW, Quinn JV, et al.: Clinical policy: critical issues in the evaluation and management of adult patients presenting to the emergency department with syncope. Ann Emerg Med 2007; 49: 431–44. CrossRef MEDLINE|
|4.||Grossman SA, Fischer C, Lipsitz LA, et al.: Predicting adverse outcomes in syncope. J Emerg Med 2007; 33: 233–9. CrossRef MEDLINE PubMed Central|
|5.||Güldner S, Langada V, Popp S, Heppner HJ, Mang H, Christ M: Patients with syncope in a German emergency department: description of patients and processes. Dtsch Arztebl Int 2012; 109(4): 58–65. VOLLTEXT|
|6.||Quinn JV, Stiell IG, McDermott DA, Sellers KL, Kohn MA, Wells GA: Derivation of the San Francisco Syncope Rule to predict patients with short-term serious outcomes. Ann Emerg Med 2004; 43: 224–32. CrossRef MEDLINE|
|7.||Sun BC, Emond JA, Camargo CA: Direct medical costs of syncope-related hospitalizations in the United States. Am J Cardiol 2005; 95: 668–71. CrossRef MEDLINE|
|8.||Kapoor WN: Evaluation and outcome of patients with syncope. Medicine (Baltimore) 1990; 69: 160–75. CrossRef MEDLINE|
|9.||Thiruganasambandamoorthy V, Hess EP, Turko E, Tran ML, Wells GA, Stiell IG: Defining abnormal electrocardiography in adult emergency department syncope patients: the Ottawa Electrocardiographic Criteria. CJEM 2012; 14: 248–58. MEDLINE|
|10.||Sun BC, Derose SF, Liang LJ, et al.: Predictors of 30-day serious events in older patients with syncope. Ann Emerg Med 2009; 54: 769–78. CrossRef MEDLINE PubMed Central|
|11.||Quinn J, McDermott D, Stiell I, Kohn M, Wells G: Prospective validation of the San Francisco Syncope Rule to predict patients with serious outcomes. YMEM 2006; 47: 448–54. MEDLINE|
|12.||Reed MJ, Newby DE, Coull AJ, Jacques KG, Prescott RJ, Gray AJ: The Risk stratification Of Syncope in the Emergency department (ROSE) pilot study: a comparison of existing syncope guidelines. Emerg Med J 2007; 24: 270–5. CrossRef MEDLINE PubMed Central|
|13.||Dipaola F, Costantino G, Perego F, et al.: San Francisco Syncope Rule, Osservatorio Epidemiologico sulla Sincope nel Lazio risk score, and clinical judgment in the assessment of short-term outcome of syncope. Am J Emerg Med 2010; 28: 432–9. CrossRef MEDLINE|
|14.||Grossman SA, Babineau M, Burke L, et al.: Applying the Boston syncope criteria to near syncope. J Emerg Med 2012; 43: 958–63. CrossRef MEDLINE|
|15.||Grossman SA, Babineau M, Burke L, et al.: Do outcomes of near syncope parallel syncope? Am J Emerg Med 2012; 30: 203–6. CrossRef MEDLINE|
|16.||Militianu A, Salacata A, Seibert K, et al.: Implantable cardioverter defibrillator utilization among device recipients presenting exclusively with syncope or near-syncope. J Cardiovasc Electrophysiol 1997; 8: 1087–97. CrossRef MEDLINE|
|17.||Natarajan B, Nikore V: Syncope and near syncope in competitive athletes. Curr Sports Med Rep 2006; 5: 300–6. CrossRef MEDLINE|
|18.||zur Muhlen von F, Quan W, D’Agate DJ, Cohen TJ: A study of carotid sinus massage and head-up tilt table testing in patients with syncope and near-syncope. J Invasive Cardiol 2002; 14: 477–82. MEDLINE|
|19.||Martin TP, Hanusa BH, Kapoor WN: Risk stratification of patients with syncope. YMEM 1997; 29: 459–66. MEDLINE|
|20.||Getchell WS, Larsen GC, Morris CD, McAnulty JH: Epidemiology of syncope in hospitalized patients. J Gen Intern Med 1999; 14: 677–87. CrossRef PubMed Central|
|21.||Charlson ME, Pompei P, Ales KL, MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40: 373–83. CrossRef MEDLINE|
|22.||Christ M, Greve Y: Der Patient mit Synkope – Realität von Diagnostik und Management in Notaufnahmen. Ther Umsch 2012; 70: 49–55. CrossRef MEDLINE|
|23.||Scharenbrock CG, Buggs AL, Furgerson JL, Manthey DE: A prospective evaluation of near syncope and syncope in the elderly. Acad Emerg Med 1999; 6: 532.|
|24.||Quinn JV, McDermott D: Medical decisionmaking and the San Francisco Syncope Rule. Ann Emerg Med 2006; 48: 762. CrossRef MEDLINE|
|25.||Colivicchi F, Ammirati F, Melina D, et al.: Development and prospective validation of a risk stratification system for patients with syncope in the emergency department: the OESIL risk score. Eur Heart J 2003; 24: 811–9. CrossRef MEDLINE|
|26.||Reed MJ, Newby DE, Coull AJ, Prescott RJ, Gray AJ: Diagnostic and prognostic utility of troponin estimation in patients presenting with syncope: a prospective cohort study. Emerg Med J 2010; 27: 272–6. CrossRef MEDLINE|
|27.||Costantino G, Perego F, Dipaola F, et al.: Short- and long-term prognosis of syncope, risk factors, and role of hospital admission: results from the STePS (Short-Term Prognosis of Syncope) study. J Am Coll Cardiol 2008; 51: 276–83. CrossRef MEDLINE|
|28.||Richardson LD, Hwang U: Access to care: a review of the emergency medicine literature. Acad Emerg Med 2001; 8: 1030–6. CrossRef MEDLINE|
|29.||Rust G, Ye J, Baltrus P, Daniels E, Adesunloye B, Fryer GE: Practical barriers to timely primary care access: impact on adult use of emergency department services. Arch Intern Med 2008; 168: 1705–10. CrossRef MEDLINE|
|30.||Schumacher JR, Hall AG, Davis TC, et al.: Potentially preventable use of emergency services: the role of low health literacy. Med Care 2013; 51: 654–8. CrossRef MEDLINE|
|31.||Brignole M, Ungar A, Casagranda I, et al.: Prospective multicentre systematic guideline-based management of patients referred to the Syncope Units of general hospitals. Europace 2010; 12: 109–18. CrossRef MEDLINE|
|32.||Ungar A, Del Rosso A, Giada F, et al.: Early and late outcome of treated patients referred for syncope to emergency department: the EGSYS 2 follow-up study. Eur Heart J 2010; 31: 2021–6. CrossRef MEDLINE|
|33.||Colman N, Nahm K, Ganzeboom KS, Shen WK, Reitsma J, Linzer M, et al.: Epidemiology of reflex syncope. Clin Auton Res 2004; 14 Suppl 1: 9–17. CrossRef MEDLINE|
|34.||Soteriades ES, Evans JC, Larson MG, et al.: Incidence and prognosis of syncope. N Engl J Med 2002; 347: 878–85. CrossRef MEDLINE|
|35.||Grubb BP, Karabin B: Syncope: evaluation and management in the geriatric patient. Clin Geriatr Med 2012; 28: 717–28. CrossRef MEDLINE|
|36.||Grubb BP: Neurocardiogenic syncope. N Engl J Med 2005; 352: 1004–10. CrossRef MEDLINE|
|37.||Kapoor WN: Syncope in older persons. J Am Geriatr Soc 1994; 42: 426–36. MEDLINE|
|38.||Grossman SA, Fischer C, Kancharla A, et al.: Can benign etiologies predict benign outcomes in high-risk syncope patients? J Emerg Med 2011; 40: 592–7. CrossRef MEDLINE|
|39.||Brignole M, Disertori M, Menozzi C, et al.: Management of syncope referred urgently to general hospitals with and without syncope units. Europace 2003; 5: 293–8. CrossRef MEDLINE|