Both the American Clinical Neurophysiology Society Reference Sinha, Sullivan and Sabau1 and the Canadian Society of Clinical Neurophysiologists Reference Dash, Dash and Primrose2 recommend that a single channel electrocardiogram (ECG) should be included as an extra channel of the electroencephalogram (EEG) recording, principally to detect contamination of EEG by apiculate waveforms, pulse artifact, or movement artifact.
However, another value of recording and interpreting the ECG, as part of the EEG record, Reference Kendirli, Aparci and Kendirli3 is the detection of arrhythmias, notably atrial fibrillation (AF) Reference Bontempo and Goralnick4 and pauses, as might occur in sick sinus syndrome or conduction block, which may be relevant in the differentiation of syncope versus seizures. Reference Viljoen, Smith and Chin5
The detection of AF is of great clinical importance. Reference Kirchhof, Benussi and Kotecha6 AF is the commonest cardiac arrhythmia in the general population, and it frequently indicates underlying heart disease and poses an important risk for cardiogenic stroke, other systemic embolism, myocardial infarction, heart failure, and increased mortality. Reference Odutayo, Wong, Hsaio, Hopewell, Altman and Emdin7,Reference Boriani and Proietti8 Recognition of AF and consequent appropriate therapy can prevent or ameliorate much of its associated morbidity and mortality. Reference Yaghi and Kamel9,Reference Boriani, Diemberger, Martignani, Biffi and Branzi10
The incidence of AF is age-dependent, with an abrupt increase after age 65, rising to over 10% of persons over 80 years. Reference Andrade, Khairy, Dobrev and Nattel11 AF can also occur in young adults, especially in those with hypertension, hyperthyroidism, valvular heart disease, and with lifestyle factors such as smoking and abuse of alcohol and recreational pharmaceuticals. Reference Aggarwal, Selvendran, Raphael and Vassiliou12 Therefore, special attention is needed to record and accurately interpret the ECG channel in high-risk groups.
We undertook a review of the prevalence of AF as recorded and interpreted, as part of the EEG report, by GBY in over 4500 records from our EEG laboratories over 7 years. The purpose was to show the prevalence of AF in various age groups in patients referred to a community EEG laboratory in Canada.
Methods
All recordings were done with Natus EEG technology, with 18 channels of EEG and 1 ECG channel (long lead I) recording with a sampling rate of 240 Hz/channel and low and high frequency filter settings at 1 and 70 Hz, respectively. Recordings were 20 min in duration.
The study size of 4529 EEG reports included patients from 0 to 100 years of age, with the first EEG report dating back to early 2013. These included consecutive reports over 7 years reported by GBY. Only one recording per patient was considered. The EEG reports were reviewed, rather than the actual traces. Several recordings were reviewed to select suitable figures for the paper. Grouping of the reports was made based on the decade of age at the time of the EEG recording. For each report, age, gender, and occurrence of AF were noted. With the grouping of the reports based on the decade established, the number of patients noted to have AF was determined and calculated as an overall percentage of each age category.
AF was diagnosed when there were no “p waves” in the ECG trace and the R−R interval was irregularly irregular, compatible with ACC/AHA Guidelines. Reference Fuster, Ryden and Cannom13 Atrial flutter was included as AF, as their clinical significance is similar and the two rhythms often occur in the same individual. Reference Fuster, Ryden and Cannom13 Examples of recordings showing AF/atrial flutter are shown in Figure 1.
FIGURE 1: (A) is a recording from an 81-year-old man who developed encephalopathy while in hospital. The recording is of low voltage showing a predominance of beta activity and some electrode artifacts related to perspiration. The ECG channel shows atrial fibrillation with a minor interventricular conduction defect. (B) is from an 89-year-old woman with impaired consciousness. The EEG shows diffuse slowing and ECG channel shows atrial fibrillation with an ectopic beat. (C) is of an 89-year-old woman with “confusion.” There is excessive beta activity and slowing in the delta range in the left posterior head. The ECG channel shows rapid atrial fibrillation. (D) is of an 82-year-old man with abnormal mental status. There is intermittent focal slowing in the left posterior head. The ECG shows atrial fibrillation with occasional premature ventricular beats.
Results
The 7-year spread of EEG reports comprised 4529 patients, 43 of whom had AF. As shown in Table 1, there were a total of two cases of AF between the ages of 0 and 60 out of a total of 2948 EEG reports. The remaining 41 cases of AF were seen in patients above the age of 60, out of the remaining 1580 reports (Figure 2). Of the 43 patients with AF, 29 were male and 14 were female. Note that the female to male ratio increases in older age groups (Figure 3).
FIGURE 2: The percentage of atrial fibrillation per decade.
FIGURE 3: The number of cases AF per decade as related to gender.
Table 1: Prevalence of atrial fibrillation in each decade
Overall prevalence of AF in each grouping of decade is shown in Table 1 and Figure 2. Throughout the first 60 years of life, there is an insignificant prevalence of AF, with a couple of cases in a few of the decades. However, it is seen that past the age of 60, the prevalence of AF rises almost exponentially. The initial notable change in prevalence is seen in the rise from 0% to 0.9% between the (50s) and (60s) age grouping. This figure continues to increase, eventually reaching a value of 8% prevalence in the 90–100 age category.
The most significant increase of AF prevalence is noted in the rise from the 60–69 to the 70–79 age group, with a 200% increase in the 70s decade compared to the 60s decade. However, the decades onwards demonstrate a slower rate of growth with an 85% increase from the 70s to the 80s, and a 60% increase from the 80s to the 90s.
Discussion
Our study confirms the previously noted age dependency of AF in the population. Reference Frost, Vukelic Andersen, Godtfredsen and Mortensen14 In our study, 1.14% of patients over 20 years of age had AF, similar to 1.12% found in a large prevalence study of the US population, which also showed an exponential rise after age 60 years. Reference Naccarelli, Varker, Lin and Schulman15 Another survey of 8.3 million German patients with insurance claims found a higher prevalence of 2.132%, with a similar rise with age over 70 years and gender breakdown, but this population was more selective and included multiple sampling times. Reference Wilke, Groth and Mueller16
The finding of AF in routine EEG recordings is more than doubly important, as older patients often have other comorbidities, such as hypertension, congestive heart failure, diabetes mellitus, and prior stroke or transient ischemic attack, which further increase the risk of stroke. Reference Chao, Chia-Jen and Ta-Chuan17 AF carries a 4 to 5-fold increase in the risk of ischemic stroke, even in the absence of comorbidities. Reference Meschia, Bushnell and Boden-Abala18 In patients over 70 years of age, the average risk of stroke/year is about 3.5%, but the risk can vary by 20 times with comorbidities. Reference Meschia, Bushnell and Boden-Abala18 AF occasionally occurs in individuals under 35 years of age, with or without comorbidities or structural heart disease, with about a 4.5% overall risk, over time, of stroke, or systemic embolism. Reference Wultzler, von Ulmstein and Attenasio19 Once found the risk of stroke can be markedly lowered by the use of anticoagulants, especially the newer oral anticoagulants, and optimal control of comorbidities. Reference January, Wann and Calkins20
We found AF to be twice as common in men than women, a higher male:female ratio than in other publications. Reference Westerman and Wenger21–Reference Odening, DeiB and Dilling-Boer23 This may reflect the fact that we had relatively fewer patients in the >80-year age groups, as in advanced age the rate of AF increases more for women than it does for men and more women than men survive into advanced age. Reference Wultzler, von Ulmstein and Attenasio19
Limitations
Our study has obvious limitations: it is retrospective and involves a single EEGer supervising two laboratories. The number of patients with AF per decade is small, but this reflects the frequency in the population. It is possible that some cases of AF were missed, for example, if portions of the ECG trace could not be seen and if the AF were truly paroxysmal.
Conclusions and Recommendations
The ECG channel provides useful information in itself, especially for the detection of AF. Thus, the finding of AF is of great importance to the referring physician, and it is vitally important for the EEGer to draw attention to the finding of AF in the report. This should prompt the referring doctor to confirm the findings with 12-lead ECG or Holter monitoring and initiate further investigation and treatment or appropriate referrals, even though the detection of AF is peripheral/incidental to the reason for the EEG request.
While our project was focused on the detection of AF, the ECG channel (most commonly a long lead I) may also show problems in cardiac conduction, such as sinus pauses or heart block, as well as abnormal rates, which may be clinically relevant. It is extremely important that EEGers pay close attention to the ECG trace and to be familiar with the common rhythm disturbances and comment on them when they are found.
Acknowledgements
We thank Deana Kerr for retrieving the records and reports.
Conflict of Interest
The authors each declare no conflicts of interest or funding sources for this paper.
Open access
