Heart 1997;78:250-254

Left atrial appendage function in patients withatrial flutter

Heyder Omran, Werner Jung, Rami Rabahieh, Dean MacCarter, Stefan Illien,Barbara Rang, Andreas Hagendorff, Rainer Schimpf, Berndt Luderitz

AbstractObjective-To determine whether echo-cardiographic markers of thromboem-bolic risk differ between patients withpure atrial flutter and patients with atrialflutter and intermittent atrial fibrillation.Design- Patients with atrial flutter werefollowed up prospectively for 12 months toidentify intermittent atrial fibrillation.After the follow up period, transthoracicand multiplane transoesophageal echo-cardiography were performed to assessleft atrial chamber and appendage size,peak emptying velocities, and emptyingfraction of the left atrial appendage. Thepresence of spontaneous echo contrastwas also determined.Setting-Tertiary cardiac care centre.Patients-20 consecutive patients withatrial flutter; 11 healthy subjects in sinusrhythm served as controls.Results-Intermittent atrial fibrillationwas documented in 11 patients by Holtermonitoring or surface ECG; atrial fibril-lation was not found in the other ninepatients. Compared with the patients withpure atrial flutter, patients with atrialflutter and intermittent atrial fibrillationhad larger left atrial chamber (mean (SD)4*5 (0.6) v 3-8 (0.5) cm; 95% confidenceinterval 0*2 to 1*2; P = 0.01) andappendage areas (6.7 (2.2) v 4-8 (4-9) cm;95% CI 0-4 to 3-5; P = 0.02), lower leftatrial appendage emptying fractions (33(11)% v 52 (ll)%; 95% CI 8 to 29; P =0.008), and also lower left atrialappendage emptying velocities (0.44(0.21) v 0 79 (0.27) mIs; 95% CI 0-13 to0-56; P = 0.005). In addition, a higherincidence of spontaneous echo contrast(11% v 36%) was observed in patients withatrial flutter and intermittent atrial fibril-lation.Conclusions-Left atrial appendage func-tion is depressed and spontaneous echocontrast more frequent in patients withatrial flutter and intermittent atrial fibril-lation, as opposed to patients with pureatrial flutter. These data support the con-cept that patients with atrial flutter andintermittent atrial fibrillation have anincreased risk of thromboembolic eventsand should therefore receive adequateanticoagulant treatment.

(Heart 1997;78:250-254)

Keywords: atrial flutter; left atrial appendage; throm-boembolism

The risk of thromboembolic events is assumedto be lower in patients with atrial flutter thanin patients with atrial fibrillation.' The reasonfor a lower risk in these patients has beenattributed to the presence of synchronousatrial mechanical contraction and organisedleft atrial appendage function. This suppos-edly prevents stasis of the blood in the atrialchamber and appendage, thereby reducing thelikelihood of the development of thrombi.Accordingly, patients with both atrial flutterand additional intermittent atrial fibrillationare considered to have a higher risk of embolicevents than patients with pure atrial flutter.

Recently it has been shown that trans-oesophageal echocardiography may be used toassess left atrial appendage function and bloodflow in the left atrium.3 In patients with atrialfibrillation, reduced left atrial appendagevelocities, and the presence of spontaneousecho contrast, an echogenic swirling pattern ofblood flow-indicating a low flow state in theleft atrium-have been shown to be markers ofthromboembolic risk.45 However, in contrastto patients with atrial fibrillation, there have asyet been no systematic studies assessing leftatrial appendage function and spontaneousecho contrast in patients with atrial flutter.The aim of this study was to describe left

atrial appendage function in patients withatrial flutter and to determine whether leftatrial appendage function and the incidence ofspontaneous echo contrast differ betweenpatients with pure atrial flutter and patientswith atrial flutter and intermittent atrial fibril-lation.

MethodsSTUDY PATIENTSThe study group consisted of 20 consecutivenon-anticoagulated adult patients with atrialflutter. All patients were followed up for 12months in our outpatient clinic to determinethe presence of intermittent atrial fibrillation.Holter monitoring was performed on a threemonths schedule and upon the occurrence ofsymptoms. Eleven patients in sinus rhythmwithout cardiac disease served as a controlgroup (eight female, three male, age 63 (SD 3)years). After the follow up period both trans-thoracic and transoesophageal echocardiogra-phy were performed after written informed

Department ofMedicine, Division ofCardiology, UniversityofBonn, Bonn,GermanyH OmranW JungR RabahiehS IllienB RangA HagendorffR SchimpfB LuderitzTeletronics, 7400South Tucson Way,Englewood, Colorado90112, USAD MacCarterCorrespondence to:Dr H Omran, Department ofMedicine Cardiology,University of Bonn,Sigmund-Freud-Str 25,53105 Bonn, Germany.Accepted for publication10 April 1997

250

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from

Left atrial appendage function in patients with atrialflutter

Patient characteristics and clinical data

Atnialflutter withinterminttent

Atial flutter atnialfibrillationNumber 9 11Men/women 7/2 8/3Age (years) 66 (5) 67 (9)Duration of last episode (days) 6 (6) 4 (4)Heart rate (beats/min) 89 (40) 98 (36)Aetiology

Valvar disease 2Coronary heart disease 2Cardiomyopathy 3Idiopathic 8 2*Other 1 1

*P < 0 05 patients with pure atrial flutter v patients with atrial flutter and intermittent atrial fib-rillation.

consent had been obtained from all patientsduring an episode of atrial flutter.

ECHOCARDIOGRAPHIC STUDIESAll studies were conducted with commerciallyavailable equipment (Vingmed 800 C,Vingmed Sound, Horton, Norway). To allowoff-line quantitative assessment of theechocardiographic data, studies were recordedon videotape with selected cineloops andvelocity spectra digitally transferred to aMacintosh Power PC computer for sub-sequent analysis, as described previously.6

For transthoracic echocardiography, a 3-25MHz transducer was used and all patientswere examined in the left lateral decubitusposition. A one-lead electrocardiogram wasrecorded continuously. The M mode left atrialdimension was measured at end systole in theparasternal long axis view, and left ventricularejection fraction was determined according tothe recommendations of the North AmericanSociety of Echocardiography.7

Topical lignocaine spray and viscous ligno-caine solution were used to anaesthetise theoropharynx before transoesophageal echocar-diography, which was performed with a 5MHZ multiplane transducer. The imagingplane and gain settings were adjusted toachieve optimal visualisation of the appendageand of spontaneous echo contrast. The samplevolume of the pulsed Doppler was placed 1 cminto the orifice of the left atrial appendage andthe profile of the velocities recorded. Care wastaken to minimise the angle of the incidence of

Figure 1 Left atrialappendage flow velocityprofile in sinus rhythm (A)and in a patient withatrialflutter (B).

the Doppler beam for flow assessment.Cineloops of the left atrium and the left atrialappendage were stored.

ECHOCARDIOGRAPHIC DATA ANALYSISEchocardiographic evaluations were per-formed in a single blinded manner, with theresults confirmed by two independentobservers following the original examination.The data were analysed by means of the evalu-ation software provided by the manufacturer(Echodisp, Vingmed Sound, Horton,Norway). The cineloops of the left atrium andleft atrial appendage were examined forthrombi and the presence of spontaneous echocontrast. Left atrial appendage area was mea-sured before cardioversion by tracing a linestarting from the top of the limbus of the leftupper pulmonary vein along the appendage'sendocardial border. Maximum appendagearea was determined during five heart cyclesand averaged. The pattern of the left atrialappendage velocity profile was described andpeak emptying and filling wavelets were mea-sured during seven consecutive cycles each,and maximum velocities then averaged. Insinus rhythm, five peak velocities were aver-aged. Since peak emptying and filling veloci-ties did not differ significantly, only emptyingvelocities are presented. Left atrial appendageejection fraction was calculated using the fol-lowing formula: (LAAmax - LAAmin)/LAAmax,where LAA.max and LAA,,,in are the maximumand minimum left atrial appendage area.

STATISTICAL ANALYSISDescriptive data are presented as the mean(SD). Continuous variables between groupswere compared by Mann-Whitney U test forunpaired observations. Nominal data werecompared by the Fisher's exact test. A P value< 0 05 was considered statistically significant(StatView 4 0, Abacus Inc, Berkeley,California, USA).

ResultsPATIENTSIn 11 patients, intermittent atrial fibrillationwas documented by Holter monitoring (mean

251

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from

Oinran, _7ung, Rabahieh, MacCarter, Illien, Raug, et al

Figure 2 Left atrial appendageflow velocity profile in a patient with pure atrialflutter and an atrioventricular conduction rate of 2:1 (A), in a patientwith varying atriovetntricular conduction rate (B), and in a patient with atrialflutter and intermittent aerialfibrillation (C).

(SD) 1 5 (0 7) episodes). The other ninepatients did not show atrial fibrillation andHolter monitoring was negative during the fol-low up period. Five patients of the latter groupunderwent successful catheter ablation ofatrial flutter. None of these patients had palpi-tations during a follow up period of sixmonths. Data of both groups are provided inthe table.

LEFT ATRIAL CHAMBER AND APPENDAGE SIZE,IEFT VENTRICULAR EJECTION FRACTION, ANDINCIDENCE OF SPONTANEOUS ECHO CONTRASTPatients with pure atrial flutter and controls didnot differ for left ventricular ejection fraction(64 (5)% v 62 (9)%, NS), left atrial diameter(3 8 (0 5) v 3 6 (0 4) cm, NS), and left atrialappendage area (4 8 (4 9) v 5 0 (5 0) cm, NS).

In contrast, compared with patients withpure atrial flutter, patients with atrial flutter andintermittent atrial fibrillation had significantlylarger left atria (4 5 (0 6) v 3 8 (0 5) cm; 95%confidence interval (CI) 0-2 to 1 2; P = 0 01)and left atrial appendages (6 7 (2 2) v 4 8 (4 9)cm; 95%/, CI 0 4 to 3 5; P = 0 02), and lowerleft ventricular ejection (45 (5)% v 64 (5)%,95% CI 13 to 23; P = 0 003).

Spontaneous echo contrast was not observedin the left atrial chamber and appendage in thecontrol group. One patient with pure atrial flut-ter had spontaneous echo contrast in the leftatrium. On the other hand, four of 11 patients

with atrial flutter and intermittent atrial fibrilla-tion had spontaneous echo contrast in the leftatrium and appendage. Thrombi were notfound in patients with atrial flutter or in thecontrol group.

QUALITATIVE DESCRIPTION OF THE LEFT ATRIALAPPENDAGE VELOCITY PROFILEIn control subjects the profile of left atrialappendage velocities showed a quadriphasicpattern with two emptying and two filling waves(fig 1A). The larger of emptying wavesoccurred after the P wave on the electrocardio-gram.

Patients with atrial flutter showed a differentprofile of left atrial appendage velocities. Thepattern consisted of alternating emptying andfilling wavelets (fig 1B). The number of empty-ing and filling wavelets during one heart cyclecoincided with the number of flutter waves inthe same heart cycle and was determined by theatrioventricular conduction rate. In fourpatients with an atrioventricular conductionrate of 4:1 there were four emptying and fillingwaves during one heart cycle, whereas in 13patients with a 3:1 or 2:1 atrioventricular con-duction rate there were three or two filling andemptying waves, respectively. In three patientswith irregular atrioventricular conduction rate,the number of left atrial appendage emptyingand filling waves varied in congruence to theatrioventricular conduction rate (fig 2).

Figure 3 Left aerialappendage emptvingfraction (A) and left atrialappendage velocities (B)in patients zvith sinuiisrhythmn (SR), those withpure atrialflutter (AFlu,),and those with atrialflutter and intermittentatrialfibrillationl(AFlu + AF). LAA EF,left atrial appendageemptying fraction; LAAv,left atrial appendageemptying velocities.

NS P = 0.003~~~~~TI k~~~~~~70 -

60-

50

40 -U-LU

< 30--J

20

10

0

1.2 r

P = 0.01 P = 0.005

1.0 W-

0.8 X

E

-j

0.6 -

0.4 K

0.2

oSR AFIu AFIu + AF

252

SR AFlu AFlu + AF

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from

Left atrial appendage function in patients with atrial flutter

QUANTITATIVE DESCRIPTION OF LEFT ATRIALAPPENDAGE CONTRACTION AND THE LEFFTATRIAL APPENDAGE VELOCITY PROFILEPeak emptying velocities of the left atrialappendage were significantly higher in patientswith pure atrial flutter than in control subjects(P = 0-01) or in patients with atrial flutterand intermittent atrial fibrillation (P = 0005)(fig 3). Left atrial appendage emptying frac-tion did not differ between patients with pureatrial flutter and control subjects. However,left atrial appendage emptying fraction was

significantly lower in patients with atrial flutterand intermittent atrial fibrillation than inpatients with pure atrial flutter (P = 0008)(fig 3).

DiscussionThe risk of thromboembolic events is assumedto be lower in patients with atrial flutter thanin patients with atrial fibrillation.' It is specu-lated that the lower risk of embolic events inpatients with atrial flutter is due to organisedmechanical atrial chamber and appendagefunction which prevents the formation ofthrombi. However, to the best of our knowl-edge there have not yet been systematic stud-ies analysing left atrial appendage function inpatients with atrial flutter and comparingpatients with pure atrial flutter to patients withintermittent atrial fibrillation.

LEFT ATRIAL APPENDAGE FUNCTION IN ATRIALFLUTTERRecently it has been shown that left atrialappendage function may be assessed by trans-oesophageal echocardiography. Atrial fibrilla-tion results in a fibrillatory pattern of the leftatrial appendage flow velocity profile and mayresult in depressed left atrial appendage con-

traction.3 In contrast, atrial flutter is associatedwith a regular pattern of atrial contractionresulting in an organised saw-type pattern ofthe left atrial appendage flow with alternatingfilling and emptying wavelets of the left atrialappendage.8 The results of our study endorsethese previous observations. In both the studyof Santiago et a18 and our study the flowwavelets coincide with the flutter waves on thesurface electrocardiogram and vary with theatrioventricular conduction rate, indicatingthat emptying and filling flows of the left atrialappendage are caused by active atrial chamberand appendage contraction. This assumptionis supported by the finding that Fourier trans-formation of the pulsed Doppler tracings ofthe left atrial appendage in atrial flutter shows aclose correlation between the dominant peakfrequency and the flutter emptying wave cyclelength.9 However, it is likely that passive fillingand emptying of the left atrial appendage mayalso occur, since peak flow velocities variedand decreased slightly at the end of diastole.Thus the results of this study confirm theassumption that atrial flutter causes an organ-ised contraction of the left atrial chamber andappendage. Furthermore, the results of thisstudy provide evidence that left atrialappendage function is similar between patients

with pure atrial flutter and healthy subjects insinus rhythm, since left atrial appendage emp-tying fractions did not differ between bothgroups. In addition, we were able to show thatpeak emptying velocities of the left atrialappendage are even higher in patients withpure atrial flutter than in patients with sinusrhythm.

COMPARISON OF PATIENTS WITH PURE ATRIALFLUTTER AND PATIENTS WITH ATRIAL FLUTTERAND INTERMITTENT ATRIAL FIBRILLATIONThe results of this study show that patientswith pure atrial flutter and patients with atrialflutter and intermittent atrial fibrillation bothhad an organised pattern of left atrialappendage flow. However, in contrast topatients with pure atrial flutter and healthysubjects in sinus rhythm, peak emptying veloc-ities and emptying fractions of the left atrialappendage were lower in patients with atrialflutter and intermittent atrial fibrillation.These findings suggest that left atrialappendage function is depressed in the latterpatient group. Our results also show thatpatients with atrial flutter and intermittentatrial fibrillation have larger left atria and leftatrial appendages. It may be assumed that theincrease in size of the left atrium and left atrialappendage is caused by depressed left atrialchamber and appendage function. However, itis also possible that enlargement itself maycause depressed left atrial appendage function.The assumption is supported by the fact thatmost patients with atrial flutter and intermit-tent atrial fibrillation suffer from underlyingheart disease and show decreased left ventricu-lar ejection fractions. Furthermore, we wereable to demonstrate a higher incidence ofspontaneous echo contrast in patients withatrial flutter and intermittent atrial fibrillation.Recently it has been shown that spontaneousecho contrast is associated with blood stasisand thrombus formation in the left atrialchamber and appendage. 1'l l Therefore, spon-taneous echo contrast has been suggested as amarker of increased thromboembolic risk.No thrombus was found in our study group.

However, 36% of patients with atrial flutterand intermittent atrial had spontaneous echocontrast in the left atrium. It may be assumedthat the pre-existing thromboembolic milieu inthe left atrial chamber and appendage may beaggravated during episodes of atrial fibrilla-tion. Alternatively, it may also be possible thatthrombi form during atrial flutter as a conse-quence of the thromboembolic milieu. Arecent report of the finding of a thrombus inone of seven patients with atrial flutter doesnot mention whether any of the patients hadintermittent atrial fibrillation.'2

CLINICAL IMPLICATIONSIn contrast to patients with pure atrial flutter,patients with atrial flutter and intermittentatrial fibrillation are prone to depressed leftatrial appendage function and often showspontaneous echo contrast in the left atrium.Thus the findings of this study support theconcept that patients with intermittent atrial

253

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from

Omran, _Jung, Rabahieh, MacCarter, Illien, Rang, et al

fibrillation have a higher risk of thromboem-bolic events and should receive adequate anti-coagulant treatment.

Because of the risk of thromboembolic com-plications in patients with atrial flutter andintermittent atrial fibrillation, the currentAmerican College of Chest Physicians guide-lines state that anticoagulation should be givento this particular patient group before car-dioversion."3 However, occasionally it may bedifficult to decide if intermittent atrial fibrilla-tion is present. Our study suggests that trans-oesophageal echocardiography may beperformed in this setting to determine left atrialappendage function and to search for sponta-neous echo contrast, in order to assess the riskof thromboembolism before cardioversion.

STUDY LIMITATIONSThe number of patients investigated was lim-ited; however, measurements were assessed ona prospective basis. Although patients wereprospectively followed up by Holter monitor-ing for determining intermittent atrial fibrilla-tion, short or asymptomatic episodes of atrialfibrillation may have been missed. Assessingthe presence of spontaneous echo contrast issubjective; however, it has been shownrecently that interobserver variability in thediagnosis of spontaneous echo contrast islow.'4

1 Arnold AZ, Mick MJ, Mazurek RP, Loop FD, TrohmanRG. Role for prophylactic anticoagulation for direct cur-rent cardioversion in patients with atrial fibrillation oratrial flutter. J7Am Coll Cardiol 1992;19:851-5.

2 Kinch JW, Davidoff R. Prevention of embolic events aftercardioversion of atrial fibrillation. Arch Intern Med1995;155:1353-60.

3 Pollick C, Taylor D. Assessment of left atrial appendagefunction by transesophageal echocardiography. Implica-tions for the development of thrombus. Circulation 1991;84:223-31.

4 Leung DYC, Black IW, Cranney GB, Hopkins AP, WalshWT. Prognostic implications of left atrial spontaneousecho contrast in nonvalvular atrial fibrillation. _J Am CollCardiol 1994;24:755-62.

5 Jones EF, Calafiore P, McNeil JJ, Tonkin AM, DonnanGA. Atrial fibrillation with left atrial spontaneous con-trast detected by transesophageal echocardiography is apotent risk factor for stroke. Am J Cardiol 1996;78:425-9.

6 Omran H, Jung W, Rabahieh R, Schimpf R, Wolpert C,Hagendorff A, et al. Left atrial chamber and appendagefunction after internal atrial defibrillation: a prospectiveand serial transesophageal echocardiographic study. JAmColl Cardiol 1997;29:131-8.

7 Henry WL. Report of the American Society of Echo-cardiography Committee on nomenclature and standardsin two-dimensional echocardiography. Circulation 1980;62:212-17.

8 Santiago D, Warshofsky M, Mandri GL, Tullio MD,Coromilas J, Reiffel J, et al. J Am Coll Cardiol 1994;24:159-64.

9 Grimm RA, Chandra S, Klein AL, Stewart WJ, Black IW,Kidwell GA, et al. Characterization of left atrialappendage Doppler flow in atrial fibrillation and flutterby Fourier analysis. Am HeartrJ 1996;132:286-96.

10 Black IW, Hopkins AP, Lee LCL, Walsh WF, JacobsonBM. Left atrial spontaneous echo contrast: a clinical andechocardiographic analysis. J Am Coll Cardiol 1991;18:398-404.

11 Fatkin D, Kelly RP, Feneley MP. Relations between leftatrial appendage blood flow velocity, spontaneousechocardiographic contrast and thromboembolic risk invivo. JAm Coll Cardiol 1994;23:961-9.

12 Black IW, Hopkins AP, Lee LCL, Walsh WF. Evaluationof transesophageal echocardiography before cardiover-sion of atrial fibrillation and flutter in nonanticoagulatedpatients. Am HeartJ_ 1993;126:375-81.

13 Laupacis A, Albers G, Dunn M, Feinberg W. Anti-thrombotic therapy in atrial fibrillation. Chest 1992;102(suppl) :426-33S.

14 Kronik G, Stollberger C, Schuh M, Abzieher F, Slany J,Schneider B. Interobserver variability in the detection ofspontaneous echo contrast, left atrial thrombi, and leftatrial appendage thrombi by transesophageal echocardio-graphy. Br HeartJ 1995;74:80-3.

254

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from

atrial flutter.Left atrial appendage function in patients with

Hagendorff, R. Schimpf and B. LüderitzH. Omran, W. Jung, R. Rabahieh, D. MacCarter, S. Illien, B. Rang, A.

doi: 10.1136/hrt.78.3.2501997 78: 250-254 Heart 

http://heart.bmj.com/content/78/3/250Updated information and services can be found at:

These include:

serviceEmail alerting

box at the top right corner of the online article. Receive free email alerts when new articles cite this article. Sign up in the

CollectionsTopic Articles on similar topics can be found in the following collections

(2127)Echocardiography (4779)Clinical diagnostic tests

(8842)Drugs: cardiovascular system

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on April 29, 2017 - Published by http://heart.bmj.com/Downloaded from