Enhanced Detection of Paroxysmal Atrial Fibrillation by Early and Prolonged Continuous Holter Monitoring in Patients With Cerebral Ischemia Presenting in Sinus Rhythm
Background and Purpose—Diagnosis of paroxysmal atrial fibrillation is difficult but highly relevant in patients presenting with cerebral ischemia yet free from atrial fibrillation on admission. Early initiation and prolongation of continuous Holter monitoring may improve diagnostic yield compared with the standard of care including a 24-hour Holter recording.
Methods—In the observational Find-AF trial (ISRCTN 46104198), consecutive patients presenting with symptoms of cerebral ischemia were included. Patients free from atrial fibrillation at presentation received 7-day Holter monitoring.
Results—Two hundred eighty-one patients were prospectively included. Forty-four (15.7%) had atrial fibrillation documented by routine electrocardiogram on admission. All remaining patients received Holter monitors at a median of 5.5 hours after presentation. In those 224 patients who received Holter monitors but had no previously known paroxysmal atrial fibrillation, the detection rate with early and prolonged (7 days) Holter monitoring (12.5%) was significantly higher than for any 24-hour (mean of 7 intervals: 4.8%, P=0.015) or any 48-hour monitoring interval (mean of 6 intervals: 6.4%, P=0.023). Of those 28 patients with new atrial fibrillation on Holter monitoring, 15 (6.7%) had been discharged without therapeutic anticoagulation after routine clinical care (ie, with data from 24-hour Holter monitoring only). Detection rates were 43.8% or 6.3% for short supraventricular runs of ≥10 beats or prolonged episodes (>5 hours) of atrial fibrillation, respectively. Diagnostic yield appeared to be only slightly and not significantly increased during the first 3 days after the index event.
Conclusions—Prolongation of Holter monitoring in patients with symptoms of cerebral ischemic events increases the rate of detection of paroxysmal atrial fibrillation up to Day 7, leading to a relevant change in therapy in a substantial number of patients. Early initiation of monitoring does not appear to be crucial. Hence, prolonged Holter monitoring (≥7 days) should be considered for all patients with unexplained cerebral ischemia.
Atrial fibrillation is a frequent cause of ischemic stroke and patients with atrial fibrillation are at increased risk for sustaining a second stroke.1,2 A diagnosis of atrial fibrillation in patients with ischemic stroke usually results in a change in therapy, because oral anticoagulation is the most effective strategy to prevent secondary ischemic events in these patients.3–5 Identification of atrial fibrillation after cerebral ischemia can therefore be expected to lower morbidity from recurrent stroke.
Diagnosis of atrial fibrillation on physical examination or routine electrocardiogram is straightforward when the arrhythmia is persistent or permanent. Paroxysmal atrial fibrillation, however, has been reported to be asymptomatic in up to 50% of patients6,7 and can therefore be difficult to detect. However, the risk of thromboembolism purported by atrial fibrillation is the same whether paroxysmal or permanent.8 Different approaches to detect atrial fibrillation in patients after an ischemic stroke have been investigated. In a recent review on noninvasive electrocardiographic monitoring, Liao et al9 reported that prolonged monitoring is likely to result in higher detection rates. However, published data are limited to a maximum of 72 hours of continuous monitoring, whereas longer timespans have only been monitored by automatically triggered event recorders. Also, the probability of detecting atrial fibrillation may be higher in the acute phase due to clustering of episodes of the arrhythmia,6,10 making the early initiation of monitoring advantageous.
We report feasibility and detection rates of atrial fibrillation with continuous 7-day Holter monitoring applied early in the emergency department in a prospectively collected cohort of patients presenting with symptoms of ischemic stroke.
Materials and Methods
In a single-center prospective observational trial termed Find-AF (ISRCTN 46104198), consecutive patients presenting to the emergency department of the University of Göttingen between March 2009 and February 2010 with symptoms of stroke or transient ischemic attack were asked on admission to give preliminary consent for participation to allow for early biomarker sampling. All patients suspected to have acute cerebral ischemia are admitted for at least 24 hours; therefore, only inpatients were included. The primary objective of Find-AF is the identification of factors associated with incident atrial fibrillation after an ischemic stroke, specifically novel biomarkers, to guide diagnosis and therapy. Results on biomarkers will be reported in a separate publication. Patients who were found to have other definitive diagnoses (eg, intracranial bleeding) causative for their symptoms were excluded; all others were asked to confirm preliminary consent by signature. Exclusion criteria were age <18 years or inability or unwillingness to consent. Because this trial was noninterventional and all patients underwent the same diagnostic workup for atrial fibrillation, we deemed consent by first-degree relatives an acceptable alternative to enroll patients when patients were unable to consent themselves.
The study complies with the Declaration of Helsinki, the protocol was approved by the responsible ethics committee, and all patients gave written informed consent.
Data Collection and Clinical Evaluation
Baseline characteristics were recorded by a standardized questionnaire, including a detailed medical history and baseline medication. All patients underwent serial biomarker sampling at 0 hour, 6 hours, and 24 hours after presentation and received a carotid ultrasound and cerebral scan (CT or MRI). After written informed consent was obtained, a Holter monitor (CardioMem CM 3000; getemed Medizin- und Informationstechnik, Teltow, Germany) was applied by specifically trained study personnel (R.C.L. and B.H.). These devices are capable of recording a period of up to 10 days with a single charge of batteries on a secure digital storage card. Patients and relatives were instructed in the correct handling of the monitors. The Holter monitors were collected after 7 days. Patients discharged earlier were instructed to send back the devices after the whole monitoring period had elapsed.
Electrocardiographic recordings were analyzed offline by 2 investigators (M.W.-K. and R.S.) blinded to clinical data of the patients using dedicated analysis software (CardioDay; getemed Medizin- und Informationstechnik). Day 4 of the recording was defined as a 24-hour period that would have been recorded as part of standard care. This period was analyzed first and a detailed report for this interval only was forwarded to the treating neurologist. Remaining days of the recording were analyzed at a later time point with a focus on the detection of atrial fibrillation. Briefly, heart rate and RR variability plots were checked for patterns suggestive of atrial fibrillation, >20 electrocardiographic strips representing fastest and slowest heart rates were inspected and arrhythmias, and supraventricular tachycardias and supraventricular premature complexes as detected by the automated software algorithm were scanned. In cases of low-quality recordings, we deviated from this algorithm to more intensive analyses up to manual review of the whole recording period in several cases.
When previously undiagnosed atrial fibrillation was detected, the respective primary care physician was informed. Because 7-day Holter electrocardiograms were, except for the standard 24-hour interval, analyzed offline with some delay, results of prolonged monitoring did not influence treatment recommendations at discharge from our hospital.
Definitions and Statistical Analyses
Continuous data are given as mean±SD unless otherwise stated. Categorical variables are given as absolute number (percent). Etiology of stroke was classified according to the widely used Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification scheme.11 Stroke severity was approximated by video-trained physicians applying the National Institutes of Health Stroke Scale12 and the modified Rankin Scale.13 Presence of atrial fibrillation was defined as at least 1 period of >30 seconds’ duration of an absolute arrhythmia without detectable P waves and without a pattern more consistent with an alternative diagnosis as recommended previously.6,14 To add detail, we also calculated overall detection rates for shorter (>10 supraventricular ectopic complexes in a row)15 as well as longer (> 5 continuous hours16) periods of atrial fibrillation. Detected episodes were verified by a specialist in electrophysiology (J.S.) blinded to clinical data. Detection rates were calculated as a fraction of all patients who had received 7-day Holter monitoring (including those with inadequate quality of recordings), except for those with previously known atrial fibrillation and those with a final diagnosis other than cerebral ischemia (Figure 1). Differences between detection rates for different monitoring intervals were tested by χ2 test or Fisher exact test if applicable. To avoid arbitrarily setting 1 specific day as the standard 24-hour monitoring interval for comparison, we calculated the mean of the individual detection rates of all 7 24-hour intervals. The same was done for all 6 48-hour intervals. Statistical tests were performed with SPSS Statistics 17.0.0 (Chicago, Ill).
Two hundred eighty-one consecutive patients were included of whom 44 (15.7%) had atrial fibrillation at presentation. All remaining patients (n=237) underwent prolonged Holter monitoring. One patient withdrew consent without specific reasons (Figure 1). Of all patients, 8 (2.9%) received a final diagnosis other than ischemic stroke or transient ischemic attack at discharge and were therefore eliminated from analyses (1 of which had presented with atrial fibrillation). Clinical characteristics for the 229 patients without and those 43 with atrial fibrillation at baseline and with a final diagnosis of cerebral ischemia are given in the Table.
Holter recordings were started after a median of 5.5 hours (interquartile range, 3.5 to 8.4 hours) after admission and 9.5 hours (interquartile range, 6.0 to 16.3 hours) after symptom onset. Prolonged monitoring was generally well tolerated, resulting in a median recording time of 6.7 days (interquartile range, 4.4 to 7 days). Due to artifacts, interruptions for clinical procedures, and early detachment of the monitoring leads by patients, only 69% of all patients had recordings of at least 5 full days of evaluable material. In 5 patients, the overall quality of recordings was so low that they were essentially not evaluable. Although the procedure was well tolerated by the majority of patients, Holter recorders were reported to be cumbersome by several patients and also in some instances by nursing staff. Although we did not implement a standardized adverse event reporting system, study personnel were in close contact with patients and their treating physicians and nursing staff over the monitoring period. The only adverse event that was occasionally reported was skin irritations at the site of electrode placement. Over the whole study period, no technical problems with the devices occurred except for 1 defect of 1 individual electrode cable that was promptly replaced by the manufacturer.
Detection of Atrial Fibrillation
Detection rates for every single day of and cumulative detection rates over the monitoring period are depicted in Figure 2. There was no recognizable pattern of detection rates favoring any given point (eg, the first 24 hours) of the monitoring period. In patients without previously known paroxysmal atrial fibrillation, an overall detection rate of 12.5% with 7-day monitoring (corresponding to 10.3% of all patients with cerebral ischemia) was significantly higher than 4.8% for any 24 hours (P=0.015) or 6.4% for any 48-hour monitoring interval (P=0.023). Importantly, 15 of those 28 patients with atrial fibrillation detected by Holter monitoring (6.7% of all patients in which Holter was performed or 5.5% of all patients with cerebral ischemia) had been discharged without a recommendation for oral anticoagulation after routine clinical workup, including results of the 24-hour Holter report defined as standard of care. None of these patients had hard contraindications for oral anticoagulation. Short periods of supraventricular arrhythmia of >10 beats in a row were detected frequently during 7-day Holter monitoring, totaling 43.8% of all patients. Rates of atrial fibrillation with a duration >5 hours were detected in 6.3% of patients undergoing Holter monitoring (or 5.2% of the whole study population).
Interestingly, there were only 5 patients with a reported history of paroxysmal atrial fibrillation who did not present with atrial fibrillation and therefore did receive an electrocardiographic monitor. Of these, 1 was not evaluable, whereas of the remaining 4 patients, only 1 had evidence of atrial fibrillation on 7-day Holter monitoring, although all others had supraventricular runs of >10 beats.
Among those 8 patients with a final diagnosis other than cerebral ischemia, 1 presented in permanent atrial fibrillation, whereas none of the other had evidence of atrial fibrillation in their 7-day Holter recording, and only 1 had supraventricular runs of ≥10 beats.
To our knowledge, this is the first report on technical details of prolonged and early continuous Holter monitoring in patients with cerebral ischemia. We are able to document a substantial increase in detection rates with prolongation of monitoring. A significant number of atrial fibrillation episodes obviously escape detection during routine clinical care and detection by Holter results in a change in therapy in many of these cases.
Our observation that the rate of detection for the complete 7-day period compares favorably with 24 hours or even 48 hours of monitoring is in line with a recent meta-analysis in which a trend toward higher rates of detection with prolongation of monitoring was proposed.9 The only available data on detection rates for 4-day17 and 7-day18 monitoring intervals were obtained with automatically triggered event recorders and after a median of 10 or 55 days after the index event, respectively. Reported detection rates of 7.7% and 5.7% are consistent with a low diagnostic yield when monitoring is not performed continuously and at an early time point, which is corroborated by the results of the current study. Higher detection rates (up to 23%) than in our observation have only been reported with event monitoring up to 21 days19 or repeatedly for 7 days over the course of 6 months.20 Such high rates, however, may to a large extent be the result of more sensitive definitions for atrial fibrillation (eg, in the study by Tayal et al, only 3 of 13 patients reported as having atrial fibrillation had episodes lasting >30 seconds). The rate of change in recommended therapy of 6.7% of all patients screened transfers into a number needed to screen of 15 patients to change 1 secondary prophylactic regimen to oral anticoagulation with a resultant relative risk reduction for secondary ischemic events of approximately two thirds.21
We would have expected to see a higher rate of atrial paroxysms early after the index event.10 However, there was a nonsignificant trend toward more episodes on Days 2 and 3 only and the absolute difference was moderate. Furthermore, there was no evidence for a higher detection rate on Day 1. Because interpretation is also hampered by the fact that actual detection rates toward the end of the monitoring period might have been actually higher, because an increasing proportion of patients discontinued monitoring over time, we are reluctant to interpret these findings as evidence for higher diagnostic yield in the first 3 days after an ischemic event. Starting monitoring very early may therefore not be crucial for a diagnostic algorithm to detect paroxysmal atrial fibrillation, a finding that should be confirmed in larger cohorts. Despite the fact that most recorders were applied in the emergency room (and the remaining in our stroke unit), we still had a median lag of 5:30 hours from admission and 9:30 hours from symptom onset until the start of the recording and can therefore not draw conclusions about the very first hours after presentation with symptoms of a cerebral ischemic event.
Our data strongly argue for an extension of the monitoring interval in patients with ischemic cerebral events, whereas starting the monitoring period as early as possible may not be essential. What remains speculative is the optimum duration of monitoring. Because the overall rate of detection continued to rise up to Day 7, even longer monitoring periods may still further increase the diagnostic yield. An analysis from the TRENDS trial22 found atrial fibrillation episodes in 28% of at-risk patients receiving a pacemaker or implantable defibrillator over a follow-up of 1.1 years and we speculate that rates may be even higher in patients after an ischemic stroke. Alternative methods for long-term monitoring will therefore have to be investigated in the future, for example, implantable loop recorders23 or telemetrically transmitted daily electrocardiograms.24,25
Another relevant and incompletely answered question is what duration of an atrial fibrillation episode actually indicates an increased risk of thromboembolisms. A minimum duration of 5.5 hours has been reported from the TRENDS trial.16 As reported previously, detection rates for such longer episodes were approximately half of those according to the standard definition in our study. On the other hand, a recent population-based analysis demonstrated that even short runs of supraventricular activity as well as a high load of supraventricular premature complexes were associated with an adverse outcome, namely hospitalizations for atrial fibrillation, but also stroke and overall mortality.15 It is striking in this regard that we found a very high proportion of patients with supraventricular runs of ≥10 beats. Such supraventricular runs were associated with a rate of atrial fibrillation hospitalizations of >10 events per 1000 patient-years in the study by Binici et al.15 Combining these observations with the high rate of supraventricular runs in our study population gives rise to the speculation that such supraventricular runs may be a surrogate for longer episodes of atrial fibrillation that may in fact be a more common cause of cerebral ischemia than has been acknowledged until now.
Although our data principally argue for the performance of prolonged Holter monitoring, the method was reported to be cumbersome both by nursing staff and patients and the analysis was time-consuming. We would therefore favor the development of a more targeted approach in the future by subjecting only patients at higher risk of having paroxysmal atrial fibrillation to prolonged Holter monitoring. Identification of these patients should be possible in the future by readily applicable clinical scores, some of which have already been proposed,26 although not without room for improvement.27
Prolongation of continuous Holter monitoring in patients with a cerebral ischemic event increases the rate of detection of paroxysmal atrial fibrillation and results in a change in therapy in a substantial number of patients. Early initiation of monitoring does not seem to be crucial. Prolonged Holter monitoring should be considered for all patients with otherwise unexplained cerebral ischemia.
Sources of Funding
This work was supported by grants from the German Federal Ministry of Education and Research (German Heart Failure Network, TP 7 [FKZ 01GI0205], including a scholarship to R.S.) and clinical trial program ALDO-DHF (FKZ 01KG0506).
↵*K.G. and R.W. contributed equally and share last authorship.
- Received May 28, 2010.
- Revision received August 4, 2010.
- Accepted August 19, 2010.
Grau AJ, Weimar C, Buggle F, Heinrich A, Goertler M, Neumaier S, Glahn J, Brandt T, Hacke W, Diener HC. Risk factors, outcome, and treatment in subtypes of ischemic stroke: the German stroke data bank. Stroke. 2001; 32: 2559–2566.
Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991; 22: 983–988.
Crystal E, Connolly SJ. Role of oral anticoagulation in management of atrial fibrillation. Heart. 2004; 90: 813–817.
Kirchhof P, Auricchio A, Bax J, Crijns H, Camm J, Diener HC, Goette A, Hindricks G, Hohnloser S, Kappenberger L, Kuck KH, Lip GY, Olsson B, Meinertz T, Priori S, Ravens U, Steinbeck G, Svernhage E, Tijssen J, Vincent A, Breithardt G. Outcome parameters for trials in atrial fibrillation: recommendations from a consensus conference organized by the German Atrial Fibrillation Competence Network and the European Heart Rhythm Association. Europace. 2007; 9: 1006–1023.
Page RL, Wilkinson WE, Clair WK, McCarthy EA, Pritchett EL. Asymptomatic arrhythmias in patients with symptomatic paroxysmal atrial fibrillation and paroxysmal supraventricular tachycardia. Circulation. 1994; 89: 224–227.
Hohnloser SH, Pajitnev D, Pogue J, Healey JS, Pfeffer MA, Yusuf S, Connolly SJ. Incidence of stroke in paroxysmal versus sustained atrial fibrillation in patients taking oral anticoagulation or combined antiplatelet therapy: an Active W substudy. J Am Coll Cardiol. 2007; 50: 2156–2161.
Liao J, Khalid Z, Scallan C, Morillo C, O'Donnell M. Noninvasive cardiac monitoring for detecting paroxysmal atrial fibrillation or flutter after acute ischemic stroke: a systematic review. Stroke. 2007; 38: 2935–2940.
Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993; 24: 35–41.
Lyden P, Brott T, Tilley B, Welch KM, Mascha EJ, Levine S, Haley EC, Grotta J, Marler J. Improved reliability of the NIH Stroke Scale using video training. NINDS tPA Stroke Study group. Stroke. 1994; 25: 2220–2226.
van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988; 19: 604–607.
Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, Halperin JL, Le Heuzey JY, Kay GN, Lowe JE, Olsson SB, Prystowsky EN, Tamargo JL, Wann S, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Hunt SA, Nishimura R, Ornato JP, Page RL, Riegel B, Priori SG, Blanc JJ, Budaj A, Camm AJ, Dean V, Deckers JW, Despres C, Dickstein K, Lekakis J, McGregor K, Metra M, Morais J, Osterspey A, Zamorano JL. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: 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 (writing committee to revise the 2001 guidelines for the management of patients with atrial fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006; 114: e257–354.
Binici Z, Intzilakis T, Nielsen OW, Kober L, Sajadieh A. Excessive supraventricular ectopic activity and increased risk of atrial fibrillation and stroke. Circulation. 2010; 121: 1904–1911.
Glotzer TV, Daoud EG, Wyse DG, Singer DE, Ezekowitz MD, Hilker C, Miller C, Qi D, Ziegler PD. The relationship between daily atrial tachyarrhythmia burden from implantable device diagnostics and stroke risk: the TRENDS study. Circ Arrhythm Electrophysiol. 2009; 2: 474–480.
Jabaudon D, Sztajzel J, Sievert K, Landis T, Sztajzel R. Usefulness of ambulatory 7-day ECG monitoring for the detection of atrial fibrillation and flutter after acute stroke and transient ischemic attack. Stroke. 2004; 35: 1647–1651.
Tayal AH, Tian M, Kelly KM, Jones SC, Wright DG, Singh D, Jarouse J, Brillman J, Murali S, Gupta R. Atrial fibrillation detected by mobile cardiac outpatient telemetry in cryptogenic TIA or stroke. Neurology. 2008; 71: 1696–1701.
Wallmann D, Tuller D, Wustmann K, Meier P, Isenegger J, Arnold M, Mattle HP, Delacretaz E. Frequent atrial premature beats predict paroxysmal atrial fibrillation in stroke patients: an opportunity for a new diagnostic strategy. Stroke. 2007; 38: 2292–2294.
Ziegler PD, Glotzer TV, Daoud EG, Wyse DG, Singer DE, Ezekowitz MD, Koehler JL, Hilker CE. Incidence of newly detected atrial arrhythmias via implantable devices in patients with a history of thromboembolic events. Stroke. 2010; 41: 256–260.
Brignole M, Vardas P, Hoffman E, Huikuri H, Moya A, Ricci R, Sulke N, Wieling W, Auricchio A, Lip GY, Almendral J, Kirchhof P, Aliot E, Gasparini M, Braunschweig F, Botto GL. Indications for the use of diagnostic implantable and external ECG loop recorders. Europace. 2009; 11: 671–687.
Gaillard N, Deltour S, Vilotijevic B, Hornych A, Crozier S, Leger A, Frank R, Samson Y. Detection of paroxysmal atrial fibrillation with transtelephonic EKG in TIA or stroke patients. Neurology. 2010; 74: 1666–1670.
Suissa L, Bertora D, Lachaud S, Mahagne MH. Score for the Targeting of Atrial Fibrillation (STAF): a new approach to the detection of atrial fibrillation in the secondary prevention of ischemic stroke. Stroke. 2009; 40: 2866–2868.
Stahrenberg R, Wachter R, Gröschel K. A risk score to predict future atrial fibrillation derived from patients with stroke initially presenting with atrial fibrillation? Stroke. 2010; 41: e169.