Implications for Stroke Prevention - Circulation

DOI: 10.1161/CIRCULATIONAHA.112.126656

Stepwise Screening of Atrial Fibrillation in a 75-Year Old Population: Implications for Stroke Prevention

Running title: Engdahl et al.; Screening of Atrial Fibrillation

Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

Johan Engdahl, MD, PhD1; Lisbeth Andersson, RN1; Maria Mirskaya, RN1; Mårten Rosenqvist, MD, PhD2

1

Dept De p ooff Medicine, pt Meedi d ciine n , Hallands Hospital Halmstad Halmstad, d, Ha H Halmstad, lmstad, Sweden;; 2De Dept D pt of Clinical Science, Karolinska Karo Ka roli ro linnska li nska k Institute, Ins nsti titu ti tute tu tee, Da D Danderyds nder nd eryd er ydss Sjukhus, Sjukkhu hus, s, Stockholm, Sto ockho ckho holm lm m, Sweden S ed Sw den

Address for Add f Correspondence: C d Johan Engdahl, MD, PhD Department of Medicine Hallands Hospital Halmstad SE-301 85 Halmstad, Sweden Tel: +46-35-131000 Fax: +46-35-131559 E-mail: [email protected]

Journal Subject Codes: [5] Arrhythmias, clinical electrophysiology, drugs; [8] Epidemiology; [193] Clinical studies; [121] Primary prevention; [64] Primary and Secondary Stroke Prevention; [70] Anticoagulants

1


Abstract:

Background—Atrial fibrillation (AF) is a frequent source of cardiac emboli in patients with ischemic stroke. AF may be asymptomatic and therefore undiagnosed. Screening for silent AF seems suitable in risk populations, little is however known on the yield and cost-effectiveness of such screening. Methods and Results—All inhabitants in the municipality of Halmstad, Sweden age 75-76 were Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

invited to a stepwise screening program for AF. As a first step, participants recorded a 12-lead ECG and reported their relevant medical history. Those with sinus rhythm on 12-lead ECG, no history of AF and at least two risk factors according to CHADS2 were invited too a 2 week wee eekk recording ecording period using a hand-held ECG asked to record 20 or 30 seconds twice daily and if palpitations pa alp pit itat atiions at ions ooccurred. ccur urrred. ur re 1330 inhabitants were inv invited vit ited d of whom 8488 (64 (64%) 4%) participated. Previously uundiagnosed nddiag dia nosed silent sile si lentt AF le AF was was found fouund fo und inn 100 (1%) (1% %) among amo ong 848 84 48 individuals in ndiivi viddua duals als w who ho re recorded eco ord rded ed 112-lead 2-leead E 2ECG. CG. Among with AF, (43%) not OAC treatment. Among Am mon ng 81 ppatients atie at ienntss wi ith h kknown nown no wn A F, 335 5 (4 (43% 3%)) we 3% were re no ot oon ot n OA AC tr rea eatm tm ment. t A mong 4403 mong 03 ppersons erssonns er ns with at least tw two wo ri risk skk ffactors accto ors ffor orr sstroke, t ok tr oke, e,, w who ho ccompleted ompl om plet pl eted et ed tthe he hhand-held andan d heeld E dECG CG eevent vent ve nt rrecording, ecor ec o ding, 30 (7.4%) were diagnosed with paroxysmal AF. Thus 75/848 (9%) of the screened population were candidates for new OAC treatment, of those 57 actually started OAC treatment. Conclusions—Stepwise risk factor-stratified AF screening in a 75-year old population yields a large share of candidates for OAC treatment on AF indication.

Key words: atrial fibrillation, screening, anticoagulation, stroke prevention

2


Introduction Atrial fibrillation (AF) is the most common clinical arrhythmia with a prevalence steeply increasing with age. The prevalence of AF is often reported to be 6-8% in patients aged 75 years1, 2. AF is also a frequent source of cardiac emboli and a common etiology of ischemic stroke. The risk of ischemic stroke is increased in patients with AF3, 4. This risk can effectively be reduced by oral anticoagulation treatment (OAC)5. AF is sometimes symptomatic, but the correlation with symptoms is weak, thus AF can Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

be present with a lack of symptoms6-8. Often, an ischemic stroke is the first clinical sign of AF. Ischemic stroke associated with AF is known to be particularly severe and more frequently fatal than han other ischemic strokes9, 10. AF is present in 25-30% of patients sustaining an an acute accutte ischemic isch is chem ch e ic em stroke troke9, 11, 12. Thee aim Th a m of this ai this study was to explore, byy sstepwise teppwise ECG sc te cre r en nin ng, g, the prevalence of screening, prev previously vio i usly nott ddiagnosed i gnos ia gnosed ed d aasymptomatic symp sy mpto to omat matic AF,, ssuitable uitaable ffor or O OAC AC ttreatment reaatme atment n inn a popu nt ppopulation opu ulaati tioon aaged geed 7755766 years yea ears r and rs and to to study study stu udy to what wha hatt extent ex xte tent nt they they hey started staarte st tedd OAC OAC treatment. trrea eatm tmen tm entt. en t.

Methods Population Halmstad is a municipality in the south-west part of Sweden with 92 000 inhabitants. All individuals born in 1934 and 1935 were invited to participate by mail. If there was no response in 4-6 weeks, a reminder was sent. If there was no response or an active declination, no further contacts were made. Index visit At the index visit, all participants had to sign an informed consent and were asked to report their

3


medical history including presence of AF, antithrombotic treatment and thromboembolic risk factors according to the CHADS2 risk classification13. If a patient reported a diagnosis of AF, this had to be confirmed by ECG recordings in the medical records. The accuracy of the self-reported medical history was confirmed only in patients with AF. However, a random subset of 80 out of 727 patients with the questionnaire as the sole source of medical history was cross-checked against medical records in hospital, in primary care and against prescriptions. One of the 80 patients had erroneously omitted that he was treated for Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

hypertension, in the remaining 79 cases, medical history was reported correctly. The index visit also included recording of a 12-lead ECG. The first 100 12-lead ECGs eted ted by y a sstudy tuddy tu dy were interpreted by a study nurse and a cardiologist; the following were interpre interpreted nurse who consulted a cardiologist on demand. The ECG interpretations were also checked by and dom m ssamples ampl am p ess vviewed ie iewed by a cardiologist. ECGs ECG Gs w ere interpreted d onl ly re regarding rhythm and random were only atee. Patients w ho hhad ad aatrio-ventricular trio tr io-v io -v ventr entric icuula ic ular bblock locck ggrade rad de II oorr II IIII or or a hheart eart ea rt rrate atte be bel low 40 low 40/m /min /m in oorr rate. who below 40/min ab bove ove 14 140/ 0//mi minn we were re rreferred effer e redd fo or fu fur rtheer ev rthe eva alua uati ua t on ti on.. above 140/min for further evaluation. a tiici ar cipa pant pa nt had ad a ppacemaker acem ac em mak ker oorr IC ICD D im impl mp an ant, t, m ed dic ical al rrecords ecor ec ords or ds w eree st er stud udie ud i d with If thee pparticipant implant, medical were studied regard to the presence of atrial high rate episodes (mode switch) caused by AF. If present and lasting more than 30 seconds, EGM recordings were studied. If a 12-lead ECG revealed previously undiagnosed AF the patient was offered a work-up consisting of blood pressure measurement, blood samples of fasting plasma glucose and thyroid stimulating hormone at a study nurse visit and an echocardiogram at a cardiologist visit. Serum glucose was not analysed in previously known individuals with diabetes. Patients with a previously diagnosed AF without OAC treatment were offered this work-up if not previously performed. After this work-up, the patient was recommended anticoagulation treatment unless

4


there were contraindications. OAC treatment was managed within routine health care and initiated in our OAC clinic. Extended ECG recording Participants with at least one additional risk factor beside their age (i.e. CHADS2 –score >=2), no history of AF and sinus rhythm on the 12-lead ECG at the index visit, were asked to make additional ECG recordings. These were made by a handheld unit, recording ECG via lead I by application of the users’ thumbs (Zenicor Medical Systems AB, Sweden. www.zenicor.se). Via Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

an in-built mobile phone, the ECG is transmitted to a website. The participant was instructed to record 20 or 30 seconds of ECG twice daily during two weeks. The duration of the recording was eco ord der er.. Al Alll decided by a study nurse who judged the participants ability to handle the ECG re recorder. handheld ECGs were interpreted by a cardiac research nurse and a cardiologist. AF was defined 30 seconds seco se co ond ndss orr at at least l ast two separate recordingss wi le w ith at least 10 sseconds e onnds each of irregular ec ass 30 with hytthm h without ut visible vissib ble l p-waves. p-w -wav avees. av es. The The he Zenicor Zenicoor ECG G system syssteem has haas been beeen be en validated validdat ated ed in in previous preevio evio ious us rhythm 14 15 epo port rtss14, rt . Pa P Patients ati tien ents en ts w with itth AF Fw were eree offe er ooffered ffe fere reed a wo work work-up rk k-u up an and nd ooffered ffe fere fe r d tr re trea treatment e tm ea tmen entt as ddescribed esscr crib bed aabove. bo ove. reports

In cas ases e w es eree in er iinterpretation t rp te pre reta tati tion ti o ooff ha on hhandheld ndhe nd held he ld E CGs G w as hhampered am mpe p reed by ppoor o r si oo sign gnal gn al qquality, uality, the cases were ECGs was signal participants were offered an additional 48-hour Holter recording. In participants who displayed runs of suspected AF on event recording not qualifying according to the definition above, another two-week period of event recording were offered according to the judgement of the investigating cardiologist. A study flow chart is depicted in figure 1. Medical records from inhabitants who did not participate in the screening process were analysed with respect to AF diagnosis, presence of anticoagulation treatment and risk factors according to CHADS2. Both hospital and primary care records were studied. Ethics

5


The study was approved by the regional health research ethics board at Lund University and conducted according to the declaration of Helsinki. Inhabitants who did not participate in the screening procedure were informed via letter and newspaper advertising that we intended to study their medical records in order to characterise this subgroup. They were given the possibility to withdraw their participation also in this part of the study. Statistical methods Continuous variables are reported as mean and range. Selected proportions are reported with a Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

95% confidence interval. For continuous variables, student t-test was used. For proportions, Fishers exact test was used. Two-tailed tests were applied. A p-value of < 0.05 was regarded as significant. ignificant. In the tables, p-values of < 0.05 are listed.

Results Resu Re sult su ltss lt Off 11330 inhabitants invited the index The O 3 0 inhabi 33 bita taants in invi viite tedd to pparticipation, arrti t ci c pation,, 8848 48 ((64%) 644%)) aattended tteende tt d d th de he in inde deex scr sscreening creen reenin ingg vi in vvisit. siit. t. T hee cardiac ECG caard rdia iacc research ia rese re seear arch ch nurse nuurse ursee spent spe p nt 30 30 minutes minu mi nute tees at index ind ndex e vvisit ex issit pper er ppatient atie at ient ie ntt iincluding nclu nc luddinng ng 112-lead 2--le lead ad E CG CG registration ECG interpretation patient. egistration and and 40 40 minutes minu mi nutees at handheld nu hand n he h ld d ECG ECG rrecording e or ec ordi ding di ng iincluding nclu nc lu udi ding ng E CG in nte terp rpre reta re tati ta tion ti on per patient t. The cardiologist spent 5-10 minutes per patient for second opinion on handheld ECG recordings and 60 minutes per visit including echocardiography among patients with newly diagnosed AF. Characteristics including prevalence of AF among attending and not attending inhabitants are described in Table. A previous diagnosis of AF was confirmed in 81/848 (9.6%, 95% CI 7.811.7). In the group who did not attend the screening, the prevalence of AF was 39/352 (11.1%, 95% CI 8.2-14.8) (n.s.). Non-attendants had a higher prevalence of diabetes, heart failure and previous stroke (Table). Among the 81 patients who were previously diagnosed with AF in the screened group, 35

6


(43%) were not receiving anticoagulation treatment at study entry. The corresponding figure of the non-screened group was 56% (n.s.). Of these 35 patients with previously known AF, 17/35 (52%) started anticoagulation treatment. ECG recording – 12-lead ECG Previously unknown AF was diagnosed in 10 patients (1.2%, 95% CI 0.5-1.9) with a 12-lead ECG. The mean heart rate among these 10 patients was 83/min ranging from 64/min to 102/min. Their mean CHADS2 –score was 1.8. Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

One participant of 848 was diagnosed with newly detected AV block III on 12-lead ECG and received a pacemaker implant. Extended handheld ECG recording Among the 848 participants there were 419 (49%) with no previous AF, sinus rhythm on 12-lead EC CG at iindex ndex nd ex vis issit aand nd a CHADS2 –score of at least lea e st 2. Of these pa participants articcip pants an 16 declined further ECG visit participation part tic i ipation or deceased, decea eassed, d, leaving lea eavi ving ng 403 403 0 who who underwent underw wentt ECG ECG event even even nt recording reeco ord ding ngg w with ithh th it the he ha hhand-held nd-h nd -hel h ld ECG. EC CG. These The hese se 403 403 participants par arti tici cippant pantss in total tota to tall recorded ta reeco ord rded ed 12 12 380 38 80 ECG EC CG tracings trrac a in ngs lasting lasstiing n 20 20 or 30 30 seconds. seeco ond ndss. The The h mean numbe number er of rrecordings ecor ec ordi or ding di nggs pe pperr pa pati patient tiien e t wa wass 31 31. 1. 40 4 ppatients a ieent at n s re reco recorded cord co r ed rd d lless esss th es than a 228 an 8 ti time times mes but only me six patients recorded less than 20 times. All patients with ambulatory ECG recordings were included in the final analysis. Ten of the 403 recordings had to be completed with a 48 hourHolter recording due to difficulties in interpreting the hand-held ECG recording and most often with a suspicion of AF. Six of these ten recordings revealed paroxysmal AF. Due to short episodes of irregular heart rhythm on hand-held ECG raising suspicion of AF but not fulfilling our criteria, 4 participants undertook another period of two weeks ECG event recording. One of these four recordings revealed paroxysmal AF. Thus, 30/403 (7.4%, 95% CI 5.2-10.4) were diagnosed with AF previously unknown. The

7


mean CHADS2–score of these 30 patients was 2.5 including 6 patients with previous stroke. A description of patient flow and ECG diagnostics is shown in figure 2. Most patients with newly detected silent paroxysmal AF were diagnosed during the first days of their two-week ECG registration period and 22 of the 24 patient diagnosed with AF on handheld ECG had multiple recordings with AF runs. The duration of ECG recording necessary for detection of AF is shown in figure 3. The yield of different methods to identify patients with an indication for OAC treatment Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

and the proportion actually starting OAC treatment is shown in figure 4. Prevalence of AF on nfi firm r ed rm At baseline, 81/848 (9.6%, 95% CI 7.6-11.6) of participants had a previously con confirmed diagnosis of AF. Another 10 patients with AF diagnosed with 12-lead ECG and 30 were diiag agno nossed no sed on o handheld han an ndheld dh or Holter ECG, thus the total totaal prevalence in to in thee screened scre sc r ened population was diagnosed 121/84 8 8 (14.3%, (14.3% 3% %, 95 95 CI CI 12.1-16.8). 12.1 .11-16 -16.8) 6.8)). Am Amo ong par pparticipants articiipa ipants tss without wittho hout ut a previously pre revi viouusl vi slyy know kknown now wn AF 121/848 Among diag di agno ag nose no se,, 40/767 40/7 40 /767 67 7 (5.2%, (5. 5.22%, 95% 95% CI 3.8-7.7) 3.88-7 3. 8-7.7) 7.7) were wer eree diagnosed diag di ag gnooseed with with th new new w AF. AF However, Howe Ho weve we v r, ve r, only onlly 403 403 off diagnose, hese 767 pa arttic icip ip pan ants ts w erre ex exam am min i ed e w itth ex exte tend te nd ded hhandheld an ndh dhel eldd EC el ECG G re reco cord co rdin rd i g. in these participants were examined with extended recording. Work-up in patients with newly diagnosed and previously diagnosed AF Among the 40 patients with newly diagnosed AF, 38 underwent echocardiography. Left ventricular Ejection Fraction (LVEF) was slightly reduced (48%) in one patient and normal (> 50%) in the remaining patients. Mean LVEF was 60%. A majority (26/38) of these patients had enlarged left atria, defined as an area in apical four-chamber view of 24 cm2 or larger. Mean left atrium area was 29 cm2. None of these patients revealed significant valvular disease. Among patients leaving blood samples for glucose, 7/41 (17%) displayed elevated fasting glucose levels, ranging from 6.4 to 7.4 mmol/l. No patients were diagnosed with previously

8


unknown abnormal level of thyroid stimulating hormone.

Discussion In this study, stepwise risk factor-stratified AF screening in a 75-76 year old population identified a total prevalence of 14%, of which 62% had no OAC treatment. Among participants who were examined with extended handheld ECG recording, 30/403 were diagnosed with previously unknown paroxysmal AF. The amount of OAC treatment on AF indication more than Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

doubled among the screened participants. Screening for AF might become an effective method to prevent stroke by initiation of OAC treatment. Patient demographics More than 60% of our community’s inhabitants aged 75 and 76 participated in the study. Since our invitation process invitation combination ou ur in invi vita vi tati ta tion ti on pro ro ocess ce merely included an invita ati tion onn by letter in co omb mbin in nat atio i n with the fact that the study media wee ar the tuddy was nott accompanied acco ac omp pan nie iedd by a m edia ccampaign, edi am mpaaignn, w aree ppleased leease sedd wi with th th he pparticipation. he arti ar tici ciipa pati tion ti on. In aan n among 75-year old persons population were AF F pprevalence reva re vale leenc ncee study sttud dy am amon ong 75 75-y -yea -y earr ol ea ld pe per rson onss by Tveit on Tve veit it eett al al.,., 882% 2% ooff the the po popu pula pu lattion la ti n w eree er examined16. Ho Howe However, weve we ver, ve r, the the h Norwegian Nor o we wegi gian gi a sstudy an tudy tu dy uused seed te telephone tele leph le phhon onee reminders remi re mind mi ndder erss an andd ev eeven en hhome omee visits for om ECG recording. Interestingly, inhabitants not attending the AF screening programme had a higher burden of cardiovascular risk factors than those attending since they had higher mean CHADS2-score, affected by higher prevalence of diabetes, heart failure and stroke. There was no significant difference in baseline AF prevalence among participants and non-participants. ECG recording A single 12-lead ECG-recording in a 75-year old population revealed only 1% of newly diagnosed persistent or permanent AF, a figure also reported from Tveit et al16. Fitzmaurice et

9


al.17 found 2% of new AF using this method. Intermittent ECG recording yielded 7% new AF diagnoses in our study, comparable to the yield seen in extended ECG recordings in patients with ischemic stroke18, 19. This finding not only underlines the importance of age in AF prevalence, but also that most patients with AF have paroxysmal arrhythmia implicating that a single ECG recording with sinus rhythm has a low negative predictive value in excluding a diagnosis of AF. Hence, among the total of 121 patients with AF in this study, only 35 (29%) had persistent or permanent arrhythmia. Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

There are plenty of data on different methods of intermittent ECG-recording to detect paroxysmal AF, most of it derives from studies on patients with cryptogenic ischemic stroke, on patients who underwent AF ablation or from studies on antiarrhythmic drugs i.e.. ppatient atie at ient nt populations with previously diagnosed AF or patients with generally high cardiovascular risk. St tud udie iess on ie n aambulant m ul mb ulaant ant intermittent ECG recordings gs inn the general po popu ulaati tioon are scarce. Studies population C onntinuous nt ECG CG Gm onnitoorin orin ingg, g, w hiich wou w ould be reg gardeed as ““Gold Gold Gold d stan sstandard” tanddard” rd” fo forr EC CG sc creeen enin i g, in Continuous monitoring, which would regarded ECG screening, reveal eve veal al ppreviously revi re viou vi ousl slyy und uundiagnosed ndiag gno noseed pa paro paroxysmal roxy ro xysm smal sm al A AF F in aass much much aass 20 220-30% - 0% nnew -3 ew wA AF F di diag diagnoses agnnosses ag ses in 20, 21 populations wi with th h hhigh ighh ca ig ard dio iova v sc scul u ar rrisk ul iskk20 is . Th Thee ev evid evidence den ence ce ffor or tthe hee eelevated leva le vate va tedd ri te risk sk ooff is iischemic chemic cardiovascular

stroke in connection to brief AF episodes is mainly derived from device studies21, 22. The AF episodes detected in this study are of larger recording proportion than the episodes detected in device studies. Since the stroke risk is similar in paroxysmal and in permanent or persistent AF23, 24

, we hypothesize that patients diagnosed with silent paroxysmal AF in this study has a stroke

risk similar to patients with clinical evident AF. Further long-term evaluation of our patients will reveal the clinical course of their AF disease. Technical development has provided several ways of ambulatory ECG recording. Shortterm Holter recordings of 24-48 h was previously the standard method but is hampered by low

10


diagnostic yield, particularly when looking for paroxysmal AF. In addition, Holter monitoring most often generate a large share of ECG information without diagnostic interest. The efficacy of detecting silent paroxysmal AF by different ambulant ECG monitoring strategies has been outlined by Kirchhof and collegues25. Event recorders and loop recorders on the other hand are activated by the patient when symptoms occur. They can also detect and store asymptomatic arrhythmias, particularly when the recorder is continuously monitoring which in turn requires continuously attachment of the recorder to the patient which might affect patient compliance Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

during longer recordings26. Event recorders not continuously attached to the patient, like the ones used in this study, must be activated and attached by the patient. Event and loop recorders with intermittent ntermittent and continuously ambulatory ECG recording have demonstrated a bbetter ette teer di diag diagnostic agno ag nost no stiic yield in comparison to Holter recordings when it comes to detecting paroxysmal AF in stroke patients paatiien ents ts15. Ri R Rizos izzoss et al. al. reported that automated analyze ana n lyze of continuous continuo uo ous E ECG CG recorded in a stroke CG uunit nitt among among patients pat atie ientts with ie with h ischemic ische sche hemi micc stroke stro stro oke orr TIA TIA almost almost almo s tripled st tri ripl p ed the pl thee diagnostic dia i gnnos osti ticc yield ti yiel yi eldd with el with regard reg garrd to silent Holter o ddetection ettec e ti tion onn ooff si ileent pparoxysmal arrox oxyssma mall AF in n ccomparison ompaari om r soon to o 224-hour 4-ho 4hour ur H ollte terr rrecording eccord cord rdin in ng27. H High iggh gh diagnostic yield yieeld is is demonstrated d mo de monsstr trat ated at d by by Mobile Mobi Mo bile bi le Cardiac Card Ca rdia rd iacc Outpatient ia O tp Ou tpat atie at ieent T ele leeme metr tryy (M tr (MCO COT) CO T) aand n nd Telemetry (MCOT) implantable loop recorders in preliminary reports. These two modalities are however expensive and implantable loop recorder requires minor surgery. The optimal ambulatory ECG method is yet to be defined; the choice of this study is directed by patient compliance and cost effectiveness. Further screening studies will reveal if there are more suitable ambulatory ECG modalities. Work-up Work-up in patients with newly detected AF yielded a low prevalence of pathological findings with the exception of 12% elevated fasting glucose levels. No patient had newly detected thyroid

11


disease or structural heart disease besides the more or less expected finding of enlarged left atria. Since we only measured blood pressure at one visit, no patient was diagnosed with hypertension in the work up. Initiation of OAC Patients with a newly diagnosed AF were more inclined to initiate OAC treatment than patients with a known diagnosis of AF. Some of the patients with known AF without OAC treatment had previously been treated with OAC in connection to a cardioversion, after which the OAC Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

treatment was withheld if sinus rhythm seemingly persisted. Patients with known AF without symptoms seemed less declined to restart OAC treatment after its termination. The change in 2010 AF guidelines28 to recommend long-term OAC after f cardioversion if theree aare ree thromboembolic hromboembolic risk factors present was not always applied in patients treated according to previous pr rev vio iouus us recommendations. rec ecoomme menndations. Patients with newlyy diagnosed me diaagnosed AF were weere r on on the th other hand easily motivated m ottiva tiv ted to ccommence om mmen mence OA OAC C tr trea treatment, eaatm ment, des despite spite tthat haat mo most st ooff th them em w were erre wi with without thoout th out symp ssymptoms. ymp m to tom ms ms. Undertreatment with with AF Unde Un dert de rtre reat atm ment w men ithh OA it OAC C in ppatients attie ient nts wi nt w th hA F aand nd tthromboembolic hrrom ombo bo oem embo boli bo licc ri rrisk s factors sk fac acttorrs rs is is very veery y common. Am Among mon ongg pa pati patients tien ents w en with ith kn it know known ow wn AF iin n ou ourr st study, tud udy,, 443% 3% w were ere no er nott re rece receiving ceiv ce i in iv ng OA OAC C at study entry. According to nationwide Swedish inpatient-statistics, half of patients with AF are never treated with OAC29. Similar figures are reported from Go et al.30 and Waldo et al31. A markedly better guideline adherence with 85% of patients with AF and risk factors treated with OAC was reported from Tveit et al16. Thus, the widespread OAC undertreatment in patients with AF contributes to an unnecessary high stroke incidence. AF prevalence The baseline prevalence of AF the 75-year old population in this study (9.6%) is higher than reported from most other studies. A prevalence of 6-8% is often reported in this age group 1, 2, 32,

12


but higher prevalence figures are reported from Nordic countries 16, 33 and from the UK 17. After including the share of patients who underwent extended ECG recording, the prevalence of AF rose to 14% in our study. Since only half of the screened population was examined with handheld ECG recorder, it is not controversial to speculate that such ECG recording in the entire screened population would have further increased the prevalence. Screening programmes for AF, mainly in the primary care setting, have been reported from the UK 17, 34. In a large randomised UK study17 in patients aged above 65, primary care Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

centres were randomized to systematic or opportunistic screening which was compared to routine care. In patients invited to systematic screening, 53% registered ECG and a new AF diagnosis was noted in 52/2357 (2%). The UK study from Fitzmaurice et al. lack data on OAC OAC trea ttreatment, reatm atm tmen e t, en both in patients with known AF and in patients newly diagnosed with AF. Data on OAC is of mpo port rtaance rt ance ffor or ccalculations alcculations al cu of cost effectivenesss ssince innce the majori ity off co cost s s for AF stem from importance majority costs tro oke care. 35 F Furthermore, urtther ermo mo oree, single sinngle si nglee rrecordings e or ec orddinggs of 112-lead 2-lead ad E ECG, CG, aass uused CG sedd inn tthe se he UK UK st sstudy, udyy, hhave ud avee av stroke severe eve vere re limitations lim imit itat it atio at ionns in in detecting dettect ctin ng paroxysmal paro pa roxxysm ro xysm smal al AF. AF. F Whet Whether the herr sc screening cre reen en nin i g fo forr AF in in patients paati t en ents ts with wit ithh ri rrisk s ffactors sk a to ac tors r aand rs n iinitiation nd niti ni tiaati ti tion on ooff OA OAC C tr treatment will significantly reduce the incidence of stroke and be cost effective remains to be shown in further studies. However, based on the study from the UK17, both American Heart Association and American Stroke Association Primary Prevention of Stroke Guidelines from 201136 and the 2012 Focused Update of AF Guidelines from the European Society of Cardiology37 recommend opportunistic screening of AF in individuals at least 65 years of age in the primary care setting by pulse palpation followed by ECG recording in case of irregular pulse. As proposed by our study, systematic screening with extended ECG recording in a 75year-old population detect a considerable share of high-risk patients with untreated silent AF,

13


partly due to the higher prevalence of AF at age 75 years rather than 65 years and partly due to the extended ECG recording. The most favourable and cost-effective method for screening of AF is subject to further studies. Limitations This study has several limitations. Since our study was carried out in a single community, the results are probably not reproducible in all populations. The generalizability to individuals of other ages, races/ethnicity is uncertain. Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

The benefit from OAC treatment in patients with AF is so far studied among patients diagnosed on standard (i.e. 12-lead) ECG recordings and the benefit in patients diagnosed with shorter However, horter episodes of AF in single-lead ECG recordings remains less studied. Howe weeveer, tthe he following data suggest that these short AF episodes carry a risk similar to permanent and persistent pe erssis iste tentt AF: AF: F An increased -A n in ncrrease easeed ri risk sk ffor orr sstroke trooke tr oke inn device deevicce patients pattieent ntss with with h short sho hort rt episodes eppiso soode dess of AF AF is i reported repo re porrted d21, 22. - Short Shor Sh o t ep or epis issod o ess ooff AF iiss a co ccommon mmon mm on ffinding indi in ding di ng in ppatients a ie at ient n s su nt suff ffer ff erin er ingg fr in rom “cr cryp cr y togenic” episodes suffering from “cryptogenic” stroke38-41. Since handheld ECG recording was intermittent, episodes of AF may have remained undiagnosed. Data on risk factors according to CHADS2 was self-reported in participants without a diagnosis of AF and collected from medical records in non-participants. Both methods of data collection have limitations. A more comprehensive invitation procedure might have increased participation further. Persons might have been more willing to take part in an established and routinely performed screening programme rather than taking part in a clinical study.

14


For instance, 83% of invited 65-year old men accepted to participate in aortic abdominal aneurysm screening in the Uppland region in Sweden 42. Implications Undiagnosed AF is often the aetiology behind “cryptogenic” stroke. It is a challenge of considerable proportions to diagnose patients with silent AF and offer them OAC treatment. Unfortunately, OAC is withheld among half of patients with already known AF and risk factors. This study implies than patients with previously diagnosed and not yet diagnosed AF can get Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

better stroke prevention within a screening programme.

Conclusions Stepwise risk factor-stratified AF screening in a 75-year old population yields a large share of ca and ndid idat id ates at ess for for OAC OAC treatment on AF indication. Persons Perrsons not participating Pe parttic i ipat attin ingg had more candidates ca ard dio i vascular ar rrisk i k factors is f ctors fa ors than than th hos osee par pparticipating. articiipating. g. Pat attieent ntss with with h pparoxysmal arox ar oxxysmaal AF cconstitute onst on sttittute ute th thee cardiovascular those Patients ma ajo jori rity ri ty ooff th thee AF ppopulation. opuulat op ulatio o n. R epeeate ep eateed ha hand ndhe nd h ld he dE CG rrecording ecorrdi ecor d ng ng ddetected eteectted et ted ne new w AF inn 7% % ooff majority Repeated handheld ECG participants,, and and the the total to otaal prevalence prev pr eval ev allen ence ce of of AF was was 14% 14% inn the the population pop opul u attio ul ionn who who pa art rtic icip ic ipat ip a ed in the at participated screening programme. Most patients with newly diagnosed AF were willing to commence OAC treatment.

Acknowledgments: We thank Eva Mellberg for her work with administration of the study, including patient invitation procedures. Contributors: JE and MR conceived the project and designed it. JE, LA and MM made additional upgrades on the design and were responsible for data collection. JE analysed the data. JE wrote the first draft of the paper. All authors assisted in revising the paper and approved the final draft. JE is the guarantor.

Funding Sources: Grants were received from the Scientific Council of the Halland Region,

15


Southern Regional Health Care Committee and from the Swedish Heart and Lung Foundation. The researchers were independent from the funders. The funders had no role in conducting the study, writing the paper, or the decision to submit the paper for publication.

Conflict of Interest Disclosures: All authors have completed the Unified Competing Interest form at www. Icmje.org/coi_disclosure.pdf (available on request from the corresponding author). Dr Engdahl has received lecture fees from AstraZeneca and Boehringer Ingelheim and consultant fees from Sanofi Aventis. Dr Rosenqvist has received lecture fees from Sanofi Aventis, Merck Sharpe & Dome, Bayer, Boehringer Ingelheim, Pfizer and Medtronic, consultant fees from Sanofi Aventis, Merck Sharpe & Dome, Nycomed, Bristol Meyers Squibb, Bayer, Medtronic Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

and research grants from Sanofi Aventis, Merck Sharpe & Dome Boehringer Ingelheim.

References: 1. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, Singer DE. Prevalence of diag agno nose no sedd at se atri r all ffibrillation ri ibri ib r llation in adults: national im mpl plic i ations for rhythm m ma m nagement and stroke diagnosed atrial implications management pr rev eveenti ent onn: th the An AnTi Tico oag agul ulation on and n R issk Fact c or orss iin n Atr triaal Fibr rillllaation on ((ATRIA) A RI AT R A) Study dy. JA JJAMA. MA. prevention: AnTicoagulation Risk Factors Atrial Fibrillation Study. 20 001;285:237 01 37 70--23 2375 5. 2001;285:2370-2375. 2. N acca carellli GV GV, Va Var rker er H Linn J,, S ch hulma mann KL KL.. IIncreasing ncrreasi sinng ng pprevalence reevalenc ncce off atr tria tr iall ffibrillation ibbrillla lati tion on aand nd Naccarelli Varker H,, Li Schulman atrial flut fl utte terr in the te the United Uni nite tedd States. te Stat St ates at e . Am es mJC ardi ar diiol ol.. 22009;104:1534-1539. 009; 00 9 10 9; 104: 4:15 4: 15534 34-1 -153 153 539. 9. flutter Cardiol. 3. Olesen JB, JB B, Lip Lip GY, GY Hansen H ns Ha nsen e ML, en ML, L, Hansen Han anse senn PR, PR Tolstrup Tols To lsstrup p JS, JS, S, Lindhardsen Lin i dh har ards dsen ds en J, J, Selmer Selm Se lmer lm err C, C, Ahlehofff O, Olsen AM, Gislason GH, Torp-Pedersen C. Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study. BMJ.. 2011;342:d124. 4. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22:983-988. 5. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146:857-867. 6. Patten M, Maas R, Karim A, Muller HW, Simonovsky R, Meinertz T. Event-recorder monitoring in the diagnosis of atrial fibrillation in symptomatic patients: subanalysis of the SOPAT trial. J Cardiovasc Electrophysiol. 2006;17:1216-1220. 7. Roche F, Gaspoz JM, Da Costa A, Isaaz K, Duverney D, Pichot V, Costes F, Lacour JR, Barthelemy JC. Frequent and prolonged asymptomatic episodes of paroxysmal atrial fibrillation revealed by automatic long-term event recorders in patients with a negative 24-hour Holter.

16


Pacing Clin Electrophysiol. 2002;25:1587-1593. 8. Nergardh A, Frick M. Perceived heart rhythm in relation to ECG findings after direct current cardioversion of atrial fibrillation. Heart. 2006;92:1244-1247. 9. Indredavik B, Rohweder G, Lydersen S. Frequency and effect of optimal anticoagulation before onset of ischaemic stroke in patients with known atrial fibrillation. J Intern Med. 2005;258:133-144. 10. Saxena R, Lewis S, Berge E, Sandercock PA, Koudstaal PJ. Risk of early death and recurrent stroke and effect of heparin in 3169 patients with acute ischemic stroke and atrial fibrillation in the International Stroke Trial. Stroke. 2001;32:2333-2337.


11. Marini C, De Santis F, Sacco S, Russo T, Olivieri L, Totaro R, Carolei A. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a population-based study. Stroke. 2005;36:1115-1119. Paroxysmal 12. Rizos T, Wagner A, Jenetzky E, Ringleb PA, Becker R, Hacke W, Veltkamp R. P arox ar oxys y ma ys m l stroke transient atrial fibrillation is more prevalent than persistentt atrial fibrillation in acute acute strok ke an aand d tr tran ansi an sien si ent ischemic schemic attack patients. Cerebrovasc Dis. 2011;32:276-282. 13. Gage G ge BF,, Waterman Ga Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical stroke: National cl lin nic ical al cclassification lass la s ificcat ss atiion io schemes for predicting str rok okee: results from th tthee Na Nat ti tional Registry of Atrial Fibrillation. 2001;285:2864-2870. Fi ibr briilla ill tion. JJAMA. AMA AM A. 20 2001 0 ;2 01 ;285 85:2 :2 286 8644 28 42870 70. 14. Doliwa V,, Rose Rosenqvist detection 14 4. D oliwa PS, PS, S Frykman Fry ykm man nV R o enq qvi vist st M. M. Short-term Shhort--teerm m ECG ECG for for out ouut off hospital hoosppittal det tecctio ct on of ssilent illen atrial episodes. Scand Cardiovasc at tri rial al ffibrillation i ri ib rill lllat atio ionn ep pis isoode odes. es. Sc cannd Ca Card rddio ova vassc JJ.. 22009;43:163-168. 0009; 9;43 43:1 43 :1663-1 :1 -1668. -1 Sobocinski Anggardh Rooth Frykman V,, vo Arbin Wallen 15. Doliwa S oboc ob ocin oc insk in skii P, sk P A ngga ng gard ga r h Ro rd oot othh E, Fr F ykma yk m n Ku ma Kull lll V vvon n Ar Arbi binn M, bi M W alle al lenn H, le Rosenqvist Europace. Rose Ro senq nqvi vist st M. M Improved Impr Im prov oved ed screening scr cree eeni ning ng for for silent sil ilen entt atrial atri at rial al fibrillation fib ibri rill llaati tion on aafter fter ft er iischaemic scha sc haem emic ic sstroke. trok tr okee Eu Euro ropa pace ce 2012;14:1112-1116. 16. Tveit A, Abdelnoor M, Enger S, Smith P. Atrial fibrillation and antithrombotic therapy in a 75-year-old population. Cardiology. 2008;109:258-262. 17. Fitzmaurice DA, Hobbs FD, Jowett S, Mant J, Murray ET, Holder R, Raftery JP, Bryan S, Davies M, Lip GY, Allan TF. Screening versus routine practice in detection of atrial fibrillation in patients aged 65 or over: cluster randomised controlled trial. BMJ. 2007;335:383. 18. Alhadramy O, Jeerakathil TJ, Majumdar SR, Najjar E, Choy J, Saqqur M. Prevalence and predictors of paroxysmal atrial fibrillation on Holter monitor in patients with stroke or transient ischemic attack. Stroke. 2010;41:2596-2600. 19. Tagawa M, Takeuchi S, Chinushi M, Saeki M, Taniguchi Y, Nakamura Y, Ohno H, Kitazawa K, Aizawa Y. Evaluating patients with acute ischemic stroke with special reference to newly developed atrial fibrillation in cerebral embolism. Pacing Clin Electrophysiol.

17


2007;30:1121-1128. 20. Bloch Thomsen PE, Jons C, Raatikainen MJ, Moerch Joergensen R, Hartikainen J, Virtanen V, Boland J, Anttonen O, Gang UJ, Hoest N, Boersma LV, Platou ES, Becker D, Messier MD, Huikuri HV. Long-term recording of cardiac arrhythmias with an implantable cardiac monitor in patients with reduced ejection fraction after acute myocardial infarction: the Cardiac Arrhythmias and Risk Stratification After Acute Myocardial Infarction (CARISMA) study. Circulation. 2010;122:1258-1264. 21. Healey JS, Connolly SJ, Gold MR, Israel CW, Van Gelder IC, Capucci A, Lau CP, Fain E, Yang S, Bailleul C, Morillo CA, Carlson M, Themeles E, Kaufman ES, Hohnloser SH. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 2012;366:120-129.


22. Glotzer TV, Hellkamp AS, Zimmerman J, Sweeney MO, Yee R, Marinchak R, Cook J, Paraschos A, Love J, Radoslovich G, Lee KL, Lamas GA. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke: report of the Atrial Diagnostics Ancillary Study of the MOde Selection Trial (MOST). Circulation. 2003;107:1614-1619. 23. Friberg L, Hammar N, Rosenqvist M. Stroke in paroxysmal atrial fibrillation: report fibrillation n: re epo ort ffrom rom ro m tthe he Stockholm Cohort of Atrial Fibrillation. Eur Heart J. 2010;31:967-975. 24. Hohnloser SH,, Pajitnev D, Pogue J, Healey JS JJS,, Pfeffer MA, Yusuf S, Connolly SJ. Incidence off stroke versus sustained atrial fibrillation patients str trok okee in paroxysmal ok parox oxxysmal ys fib ibriillation in patien e ts taking en tak akin ing oral anticoagulation orr combined Coll Cardiol. combinedd antiplatelet a ti an tipl p ate pl tele leet therapy: ther th e ap er py: an an ACTIVE ACTI AC TIVE VE W Substudy. Suubsttud udy. y. J Am mC olll Ca ol C rddio iol. l. 20 22007;50:215607;5 07 ;50: 0 21156 0: 5 2161. 21661. 61 25. Cosio F,, 255. Kirchhof Kirc Ki r hh rc hhof of P, P, Bax Bax J, J, Blomstrom-Lundquist Blo loms msstr trom om-L om -Lun -L unndq dqui uisst C, ui C, Calkins Calk Ca l ins ins H, Camm Cam amm m AJ, A , Cappato AJ Capppat Capp atoo R, C osio oF Crijns H,, Di Diener HC, Goette Lip GY, Nattel Page Ravens U,, Crij ijns ns H D en ner H C, Goe ette A, IIsrael s ael CW sr CW, Ku Kuck ck KH, H L ip G Y N Y, atte at tell S, S P agee RL ag RL, Ra Rave vens ns U Schotten U,, St Steinbeck Vardas Waldo A,, We Wegscheider Willems S,, Br Breithardt Stei einb ei nbbec eckk G, G V ardaas P, W ar a do A al W gsch gs c ei ch eide derr K, W de illle l ms S B eith ei thar th ardt ar dt G. Early and and comprehensive comp co mpre rehe hens nsiv ivee management mana ma nage geme ment nt of of atrial atri at rial al fibrillation: fib ibrril illa lati tion on:: executive exec ex ecut utiv ivee summary summ su mmar aryy of tthe he pproceedings roce ro ceed edin ings gs from the 2nd AFNET-EHRA consensus conference 'research perspectives in AF'. Eur Heart J. 2009;30:2969-2977c. 26. Seet RC, Friedman PA, Rabinstein AA. Prolonged rhythm monitoring for the detection of occult paroxysmal atrial fibrillation in ischemic stroke of unknown cause. Circulation.124:477486. 27. Rizos T, Guntner J, Jenetzky E, Marquardt L, Reichardt C, Becker R, Reinhardt R, Hepp T, Kirchhof P, Aleynichenko E, Ringleb P, Hacke W, Veltkamp R. Continuous stroke unit electrocardiographic monitoring versus 24-hour holter electrocardiography for detection of paroxysmal atrial fibrillation after stroke. Stroke. 2012;43:2689-2694. 28. Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, Van Gelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O, Angelini A, Atar D, Colonna P, De Caterina R, De Sutter J, Goette A, Gorenek B, Heldal M, Hohloser SH, Kolh P, Le Heuzey JY, Ponikowski P, Rutten FH. Guidelines for the management of atrial fibrillation: the Task Force

18


for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J. 2010;31:2369-2429. 29. Friberg L, Rosenqvist M, Lip GY. Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study. Eur Heart J. 2012;33:1500-1510. 30. Go AS, Hylek EM, Borowsky LH, Phillips KA, Selby JV, Singer DE. Warfarin use among ambulatory patients with nonvalvular atrial fibrillation: the anticoagulation and risk factors in atrial fibrillation (ATRIA) study. Ann Intern Med. 1999;131:927-934. 31. Waldo AL, Becker RC, Tapson VF, Colgan KJ. Hospitalized patients with atrial fibrillation and a high risk of stroke are not being provided with adequate anticoagulation. J Am Coll Cardiol. 2005;46:1729-1736. Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

32. Rietbrock S, Heeley E, Plumb J, van Staa T. Chronic atrial fibrillation: Incidence, prevalence, and prediction of stroke using the Congestive heart failure, Hypertension, Age >75, Diabetes mellitus, and prior Stroke or transient ischemic attack (CHADS2) risk stratification str trat atif at i ic if icat atio ionn scheme. cheme. Am Heart J. 2008;156:57-64. 33. Stefansdottir H, Aspelund T, Gudnason V, Arnar DO. Trends in the incidence and prevalence 2011;13:1110-1117. of atrial fibrillation in Iceland and future projections. Europace. 2011; ;13:1110-1117. 34. Wheeldon NM, Tayler Anagnostou E,, Co Wales Oakley GD. Screening for 34 4. Wheeld W h don N M T M, ayle ay leer DI DI,, An Anag a no ag nost stou ou E Cook okk D,, Wa Wale l s C, O le akle ak leyy GD D. Sc Scre r en re enin ingg fo or at aatrial ra ri fibrillation primary care. fi ibrril illation in pr prim imarry ca im aree. Heart. Hear He art. t 11998;79:50-55. t. 99 98;79::50-555. 35. Ericson Bergfeldt L,, Bj Bjorholt Atrial cost illness Sweden. 35 5. Er Eric i so ic sonn L, L B ergf er g elldtt L gf Bjor orrho holt lt II.. At Atri riaal ri al ffibrillation: ibri ib riillaatio on:: tthe he co cos st ooff il ill lness in S weeden. de . Eur E ur J Health Econ. 2011;12:479-487. Heal alth th E conn. 20 co 2011 1;1 ;122:47 799 48 87. 36. Goldstein S, Cr Creager MA, 36 Gold Go ldst stei einn LB, LB Bushnell Bush Bu shne nell ll CD, CD Adams Ada dams ms RJ, RJ Appel App ppel el LJ, LJ Braun Brau Br aunn LT, LT Chaturvedi Chat Ch atur urve vedi di S Crea eage gerr MA Culebras A, Eckel RH, Hart RG, Hinchey JA, Howard VJ, Jauch EC, Levine SR, Meschia JF, Moore WS, Nixon JV, Pearson TA. Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2011;42:517-584. 37. Camm AJ, Lip GY, De Caterina R, Savelieva I, Atar D, Hohnloser SH, Hindricks G, Kirchhof P, Bax JJ, Baumgartner H, Ceconi C, Dean V, Deaton C, Fagard R, Funck-Brentano C, Hasdai D, Hoes A, Knuuti J, Kolh P, McDonagh T, Moulin C, Popescu BA, Reiner Z, Sechtem U, Sirnes PA, Tendera M, Torbicki A, Vahanian A, Windecker S, Vardas P, Al-Attar N, Alfieri O, Angelini A, Blomstrom-Lundqvist C, Colonna P, De Sutter J, Ernst S, Goette A, Gorenek B, Hatala R, Heidbuchel H, Heldal M, Kristensen SD, Le Heuzey JY, Mavrakis H, Mont L, Filardi PP, Ponikowski P, Prendergast B, Rutten FH, Schotten U, Van Gelder IC, Verheugt FW. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: An update of the 2010 ESC Guidelines for the management of atrial fibrillation * Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 2012.

19


38. 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. 39. Rizos T, Rasch C, Jenetzky E, Hametner C, Kathoefer S, Reinhardt R, Hepp T, Hacke W, Veltkamp R. Detection of paroxysmal atrial fibrillation in acute stroke patients. Cerebrovasc Dis. 2010;30:410-417. 40. Stahrenberg R, Weber-Kruger M, Seegers J, Edelmann F, Lahno R, Haase B, Mende M, Wohlfahrt J, Kermer P, Vollmann D, Hasenfuss G, Groschel K, Wachter R. Enhanced detection of paroxysmal atrial fibrillation by early and prolonged continuous holter monitoring in patients with cerebral ischemia presenting in sinus rhythm. Stroke. 2010;41:2884-2888.


41. 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. 42. Wanhainen A, Svensjo S, Tillberg M, Mani K, Bjorck M. [Abdominal aortic aaneurysm neur ne u ys ur ym screening way]. creening in Uppsala. Good experiences from the first four years--the rest of Sweden Swe weedeen on o its its w ay]] ay Lakartidningen. 2010;107:2232-2236.

Table Clinical Characteristics Ta abl ble 1. Cl C in niccal C hara ha raactter eris isti tics cs nn,, (% (%))

Male Maale gender gende der Previously d diagnosed iaagn gnos osed os ed A AF F Hear He Heart artt Fa Fail Failure ilur uree Hypertension Diabetes Mellitus Previous Stroke/TIA CHADS2 –score (mean)*

Participating P arrti t cippatinng nn=848 =848 3364 36 4 (4 (43% (43%) 3%)) 8 ((9%) 81 9%)) 9% 30 (4%) (4% 4%)) 446 (53%) 91 (11%) 80 (9%) 1.85

Non-participating N on-p on -paart articippat pating nn=352 n=35 =35 3522 1499 (4 14 (42% (42%) %) 39 ((11%) 11%) 11 % 34 (10%) (10 10%) %) 185 (53%) 60 (17%) 49 (14%) 2.08

* CHADS2 –score was calculated regardless of diagnosis of AF

Figure Legends:

Figure 1. Study design and flow of participants.

Figure 2. Study flow with regard to ECG diagnostics. 20

p

1 n=419 (55%)

Individuals In ndivvidualls w with ith ssinus in nus s rhythm rh hythm m and an nd CHAD DS2-sco CHADS -score ore = 1 8 (45%) (45% %) n=338

No AF on Handheld ECG n=373 New AF on Event recorder ECG n=30 (7.4%)


Figure 3


Figure 4

Stepwise Screening of Atrial Fibrillation in a 75-Year Old Population: Implications for Stroke Prevention Johan Engdahl, Lisbeth Andersson, Maria Mirskaya and Mårten Rosenqvist Circulation. published online January 23, 2013; Downloaded from http://circ.ahajournals.org/ by guest on February 16, 2018

Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2013 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2013/01/23/CIRCULATIONAHA.112.126656

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/