3The solid curve indicates the projected increase in AF prevalence if no further increase in AF incidence occurs. The dotted curve indicates the increase in AF prevalence if the increase in incidence rate continues to rise to a similar extent as it rose between 1980 and 2000.
6Nathan H, Eliakim M. The junction between the left atrium and the pulmonary veins. An anatomic study of human hearts. Circulation 1966;34:412– 422.MechanizmusJalife J, Berenfeld O, Mansour M. Mother rotors and fibrillatory conduction: amechanism of atrial fibrillation. Cardiovasc Res 2002;54:204 –216.Ectopic foci vs. circus movement activity. Moreover, these investigators estimated thatmaintenance of fibrillation in the canine atrium required aThe exact mechanisms underlying AF are still poorly critical number of four to six wavelets. Some support ofunderstood despite many years of research and speculation. this idea came from the pharmacological experiments ofAs early as 1907, Winterberg  surmised that AF was Wang and co-workers [17,18] in which termination ofdue to multiple rapidly firing foci distributed throughout atrial fibrillation by class IC antiarrhythmic drugs wasthe atria. In 1914, Mines  advanced the circus move- preceded by a decrease in the mean number of wavelets.ment theory of reentry. Thereafter, and until the late Subsequent experiments in dogs , as well as more19509s, all of the hypotheses proposed to explain AF were recent intra-operative mapping studies in humans ,variations of the circus movement and ectopic focus have provided important insight into the characteristics oftheories [13,14]. However, following publication of the wave front propagation during fibrillation, and have givenmultiple wavelet hypothesis by Moe and Abildskov , it support to Moe’s idea that multiple wavelets distributedbecame generally accepted that AF was the result of the randomly throughout the atria gave rise to the seeminglyrandom propagation of multiple wavelets across the atria, a chaotic activation patterns observed in the electrocarprocessthat was thought to be independent of the initiating diograms of patients with AF. Finally, the hypothesis wasevent. emboldened by the clinical observation that chronic AFcould be cured in some patients by the placement of2.2. Multiple wavelets vs. circus movement multiple surgical lesions (MAZE) to compartmentalize theatria into regions presumably unable to sustain the multipleMoe and Abildskov  strongly argued that both the wavelets . Indeed, this theory is virtually universallyectopic focus and circus movement reentry notions were accepted by most clinical electrophysiologists today.inadequate descriptions of AF. For them it was hard to However, none of the above-mentioned studies gavebelieve that either mechanism could be endowed with the answers to many critical questions about the origin of thenecessary stability to persist for many years as AF turbulent activity that gave rise to the multiple wavelets incommonly does. Thus, they proposed that the mechanism the experiments. For example, while the computer simulaofAF, whether occurring spontaneously in man or pro- tions of Moe et al.  suggested that 15–30 waveletsduced experimentally, was fundamentally different from were needed at a given time to keep the fibrillatory processflutter and from automatic ectopic discharge. In 1964, Moe going, the experiments of Allessie et al.  could showet al. published their classical computer model study on only 4–6 wavelets propagating on the surface of the dogAF . Based on that numerical study and on their heart. In the presence of such a small number of wavelets,previous experimental results , they postulated that AF one would expect that, at any given time, a large amountcould exist as a stable state, self-sustained and independent of tissue in both atria would be recovered from previousof its initiating agency, and also that an independent excitation, which would lead to coalescence of waveletssurvival of the arrhythmia was possible only in the and eventual termination of AF. In other words, it ispresence of inhomogeneous atrial repolarization enhanced reasonable to speculate that, in the experiments of Allessieby adequate vital activity. The idea of Moe et al. was et al. , the arrhythmia could have been in fact the resultsimple. They considered fibrillation to be a fundamentally of a single (or a small number of) high frequency sourceturbulent and self-sustaining process, which takes place in that was (were) hidden from view. Additional questionsa non-homogeneous excitable medium. Such a process include the following: is spontaneously occurring fibrillacouldbe initiated by an impulse propagating through the tion the result of a ‘‘mother rotor’’ that breaks andmedium at a time when some of its components have fractionates into multiple independent offspring? What arerecovered while others remain partially or fully refractory the fundamental structural and electrophysiological characasa result of a preceding activation. Accordingly, some teristics that sustain the wavelets and which enable theirelements in the medium mayMoe GK, Rheinboldt WD, Abildskov JA. A computer model of atrial fibrillation.Am Heart J 1964:200 –220.6
10Circulation 2004 Covariates Significantly Associated With Survival p HRAge at enrollment*<0.00011.06Coronary artery disease<0.00011.65Congestive heart failure<0.00011.83Diabetes<0.00011.56Stroke or transient ischemic attack<0.00011.54Smoking<0.00011.75First episode of atrial fibrillation0.00671.27Sinus rhythm<0.00010.54Warfarin use<0.00010.47Digoxin use<0.00011.50Rhythm-control drug use0.00051.41Circulation 2004
11Patients Without Recurrence (%) Ritmus-kontroll?CTAF Study100Amiodarone (n = 201)8060Propafenone (n = 101)Patients Without Recurrence (%)40Sotalol (n = 101)200.0100200300400500600Days of Follow-upRoy D, et al. New Engl J Med. 2000;342:913–920.
40CHADS2 score Condition Points C Congestive heart failure 1 H Hypertension (or treated hypertension) A Age >75 years D Diabetes S2 Prior Stroke or TIA2
41CHADS scoreTreatment with Coumadin reduces the risk of stroke in atrial fibrillation by about 2/3 (66%,)and treatment with aspirin reduces the risk by 1/4 (25%). The risk that anticoagulation willproduce a major bleeding episode varies depending on the patient's medical problems and thecare with which the protime or INR (i.e., the "thinness" of the blood) is monitored, butgenerally is less than 1 - 2% per year.Based on all this information, the developers of the CHADS model recommend stronglyconsidering therapy with Coumadin for anybody whose CHADS score is 1 or higher. Certainly,most patients with CHADS scores of 2 or higher stand a strong chance of benefitting fromanticoagulation.Precautions in using the CHADS modelThe CHADS model was developed by studying the records of 1733 Medicare beneficiaries aged65 to 95, and in this age group the model appears very accurate at predicting stroke risk.How well the CHADS model works for younger patients is not entirely clear, however, since thedata was gathered for elderly patients. The model is not valid at all for individuals with thevalve disease called mitral stenosis - these patients have a very high risk of stroke, andvirtually all of them should take anticoagulation therapy.41