Clin. Cardiol. 14,652-656 (1991)
Changes in Left Atrial Size in Patients with Lone Atrial Fibrillation G U I L L E R M O SOSA S U A R E Z , M.D., S I E V E N
LAMPERT, M.D.,
F.A.C.C., S H M U E L
RAVID,M.D., BERNARD LOWN,M.D.,
F.A.C.C.
Lown Cardiovascular Center, Department of Nutrition, Harvard School of Public Health, and the Cardiovascular Division, Department of Medicine, Brigham & Women's Hospital, Boston, Massachusetts, USA
Summary: A retrospective analysis was performed on 23 subjects with lone atrial fibrillation who were followed for an average of 6.2 years (1.1 - 12.8 years). In all patients, underlying organic heart disease was excluded based on history. physical exam, electrocardiogram, echocardiogram, and Doppler ultrasound interrogation. All patients had at least two echocardiographic studies during the period of observation. Atrial fibrillation was chronic in 1 I subjects and paroxysmal in 12. All echocardiographic measurements were obtained by averaging the measurements of two blinded investigators. Left atrial size increased an average of 5.6 mm which translates into a 14.7% increase over the baseline measurement. This increase in size was not associated with a change in left ventricular mass or fractional shortening as determined by echocardiography. Subjects with chronic atrial fibrillation had a larger percent increase lhan subjects with paroxysmal atrial fibrillation (18.9 vs. 10.8%)).although this relative change in size failed to reach statistical significance. The only variable which significantly contributed to the change in left atrial size was the duration of follow-up. We conclude that atrial fibrillation occurring in patients with lone atrial fibrillation may cause a slow and progressive increase in left atrial size independen@of changes in left ventricular size or function. Key words: atrial fibrillation, left atrium, echocardiography, ~OIIOW-UP study
Introduction Atrial fibrillation is frequently associated with left atrial enlargement. It remains unresolved whether the ar-
Address for reprints: Steven Larnpert, M.D. Lecturer in Cardiology Harvard School of Public Health 2 I Longwood Avenue Brookline, MA 02146, USA Receivcd: February 12. 1991 Accepted with revision: April 23, 1991
rhythmia by itself causes left atrial dilatation or whether chamber enlargement is a consequence of the very factors which frequently result in the fibrillation, namely increased left atrial pressure due either to structural heart disease or to left ventricular dysfunction. When mitral stenosis and atrial fibrillation coexist, the relative contribution of each to the left atrial enlargement is difficult to assess. Some investigators have postulated that the degree of obstruction was the critical determinant of left atrial enlargement.'-3 Other researchers disagreed, suggesting that the enlarged left atrium is a direct consequence of the atrial A recent prospective study utilizing regression analysis suggested that the severity of mitral stenosis accounted for most of the change in left atrial size, although atrial fibrillation made a small but independent contribution.' In the absence of valvular heart disease, atrial fibrillation is often associated with left atrial enlargement. In a heterogenous cohort free of valvular heart disease,x left atrial size was related to the duration of the arrhythmia and increased during a 6-month follow-up period. An ideal cohort for determining whether atrial fibrillation causes an increase in left atrial size exists among patients with lone atrial fibrillation. Evans and Swann' did not find left atrial enlargement in 20 patients with this disorder nor i n four patients followed for 10 years. A recent prospective study on 15 patients followed for 20 months suggested that the increase in left atrial size was a consequence of atrial fibrillation.'" The present study was designed to determine the influence of atrial fibrillation on changes in echocardiographically determined left atrial size in a population free of structural heart disease and followed for one year or longer. To define this effect better, a comparison was made between the paroxysmal and sustained type of atrial fibrillation.
Methods Patient Selection
We reviewed the records in 436 patients with atrial fibrillation examined by our group between 1982 and 1986. We selected 23 patients who were free of structural heart disease based on a normal history, physical examination, echocardiogram and Doppler ultrasound interrogation. At
G. S. Suarez et al.:Left atrial size in lone atrial fibrillation
least two echocardiographic examinations were required for inclusion in this study. Exclusion Criteria Patients were excluded for the following reasons: a history of rheumatic, ischemic, or hypertensive heart disease, valvular heart disease diagnosed by either physical exam or Doppler echocardiography, congestive heart failure, pre-excitation syndromes, hyperthyroidism, excessive alcohol intake, or cardiomyopathy. Patient Categories
In all patients the presence of atrial fibrillation was documented electrocardiographically. Atrial fibrillation was chronic in 1 1 patients and paroxysmal in 12 patients. Paroxysmal atrial fibrillation was defined as at least two documented episodes of atrial fibrillation with an intervening period of normal sinus rhythm. In this group the echocardiogram was obtained in normal sinus rhythm. The average age was 63.0k9.1 years and was not different in the two groups. There were 18 men and 5 women equally distributed in the two groups. Follow-Up Patients were followed for an average of 6.2 years with chronic atrial fibrillation were followed for 7.0 years compared with 5.4 years in the paroxysmal group (p =NS). ( I . I - 12.8 years), those
Echocardiographic Studies Equipment Echocardiographic studies performed prior to 1982 utilized an IREX M-mode ultrasound system. Only M-mode studies were performed prior to 1982; thereafter all studies includd two-dimensional guided M-mode echocardiogram. After 1985, all studies were performed on a Hewlett Packard 77020 ultrasonoscope equipped with a 2.5 MHz transducer and M-mode, two-dimensional echocardiography, and both pulse wave and continuous wave Doppler echocardiographic capabilities. Number of Echocardiographic Studies All patients underwent an average of 3.0 (2.0-5.0) echocardiographic studies during the follow-up period. All patients underwent at least one recent Doppler echocardiographic study to further exclude the presence of valvular heam disease. Echocardiographic Measurements M-mode tracings were of good quality and allowed accurate measurement of cardiac chambers. The following
653
measurements, adhering to the criteria of the American Society of Echocardiography,!’ were made in all subjects: left atrial anterior-posterior dimension, including the posterior aortic wall at end ventricular systole; left ventricular end-diastolic dimension; left ventricular end-systolic dimension; septa1 and posterior wall thickness. Left atrial size was adjusted for body surface area, calculated at the time of each echocardiographic study. The fractional shortening (FS) was calculated for each echocardiographic study according to the following formula: FS (%) = [(LVD-LVS)/LVD] x 100 Left ventricular mass was calculated for each echocardiographic study according to the anatomically corrected regression equation of Devereaux:12 Left ventricular mass (g)= 1.04 [(LVD+IVS+LVPW) 3-(LVD) 31- 13.6 where LVD= left ventricular diastolic diameter, LVS =left ventricular systolic diameter, IVS =interventricular septum thickness, and LVPW =posterior wall thickness. Left ventricular mass was then adjusted for height.I3 All echocardiographic studies were interpreted by two readers (SL; SR), who were blinded with regard to the diagnosis, temporal sequence, and subject. Pearson correlation coefficients were calculated for each pair of measurements. The correlation coefficient for left atrial size was .96. The correlation coefficients for left ventricular diastolic diameter, systolic diameter, interventricular septum, and posterior wall were 0.83,0.75,0.50, and 0.45, respectively. The two observers’ measurements were then averaged. After confirming the reliability of the two readers, the initial and final echocardiograms were selected for all subsequent analysis. Statistical Analysis Data were expressed as the mean and standard deviation. The unpaired Student’s t-test was used to compare between group measurements. Within subjects changes in echocardiographic measurements were analyzed with a paired Student’s t-test. Multiple regression analysis was performed to evaluate the relative contribution of age, duration of follow-up, baseline echocardiographic measurements, and type of atrial fibrillation on the change in left atrial size. Statistical significance was defined as p < 0.05.
In the entire cohort, the left atrial size increased an average of 5.6f6 mm during the follow-up period (p
Changes in Left Atrial Size in Patients with Lone Atrial Fibrillation G U I L L E R M O SOSA S U A R E Z , M.D., S I E V E N
LAMPERT, M.D.,
F.A.C.C., S H M U E L
RAVID,M.D., BERNARD LOWN,M.D.,
F.A.C.C.
Lown Cardiovascular Center, Department of Nutrition, Harvard School of Public Health, and the Cardiovascular Division, Department of Medicine, Brigham & Women's Hospital, Boston, Massachusetts, USA
Summary: A retrospective analysis was performed on 23 subjects with lone atrial fibrillation who were followed for an average of 6.2 years (1.1 - 12.8 years). In all patients, underlying organic heart disease was excluded based on history. physical exam, electrocardiogram, echocardiogram, and Doppler ultrasound interrogation. All patients had at least two echocardiographic studies during the period of observation. Atrial fibrillation was chronic in 1 I subjects and paroxysmal in 12. All echocardiographic measurements were obtained by averaging the measurements of two blinded investigators. Left atrial size increased an average of 5.6 mm which translates into a 14.7% increase over the baseline measurement. This increase in size was not associated with a change in left ventricular mass or fractional shortening as determined by echocardiography. Subjects with chronic atrial fibrillation had a larger percent increase lhan subjects with paroxysmal atrial fibrillation (18.9 vs. 10.8%)).although this relative change in size failed to reach statistical significance. The only variable which significantly contributed to the change in left atrial size was the duration of follow-up. We conclude that atrial fibrillation occurring in patients with lone atrial fibrillation may cause a slow and progressive increase in left atrial size independen@of changes in left ventricular size or function. Key words: atrial fibrillation, left atrium, echocardiography, ~OIIOW-UP study
Introduction Atrial fibrillation is frequently associated with left atrial enlargement. It remains unresolved whether the ar-
Address for reprints: Steven Larnpert, M.D. Lecturer in Cardiology Harvard School of Public Health 2 I Longwood Avenue Brookline, MA 02146, USA Receivcd: February 12. 1991 Accepted with revision: April 23, 1991
rhythmia by itself causes left atrial dilatation or whether chamber enlargement is a consequence of the very factors which frequently result in the fibrillation, namely increased left atrial pressure due either to structural heart disease or to left ventricular dysfunction. When mitral stenosis and atrial fibrillation coexist, the relative contribution of each to the left atrial enlargement is difficult to assess. Some investigators have postulated that the degree of obstruction was the critical determinant of left atrial enlargement.'-3 Other researchers disagreed, suggesting that the enlarged left atrium is a direct consequence of the atrial A recent prospective study utilizing regression analysis suggested that the severity of mitral stenosis accounted for most of the change in left atrial size, although atrial fibrillation made a small but independent contribution.' In the absence of valvular heart disease, atrial fibrillation is often associated with left atrial enlargement. In a heterogenous cohort free of valvular heart disease,x left atrial size was related to the duration of the arrhythmia and increased during a 6-month follow-up period. An ideal cohort for determining whether atrial fibrillation causes an increase in left atrial size exists among patients with lone atrial fibrillation. Evans and Swann' did not find left atrial enlargement in 20 patients with this disorder nor i n four patients followed for 10 years. A recent prospective study on 15 patients followed for 20 months suggested that the increase in left atrial size was a consequence of atrial fibrillation.'" The present study was designed to determine the influence of atrial fibrillation on changes in echocardiographically determined left atrial size in a population free of structural heart disease and followed for one year or longer. To define this effect better, a comparison was made between the paroxysmal and sustained type of atrial fibrillation.
Methods Patient Selection
We reviewed the records in 436 patients with atrial fibrillation examined by our group between 1982 and 1986. We selected 23 patients who were free of structural heart disease based on a normal history, physical examination, echocardiogram and Doppler ultrasound interrogation. At
G. S. Suarez et al.:Left atrial size in lone atrial fibrillation
least two echocardiographic examinations were required for inclusion in this study. Exclusion Criteria Patients were excluded for the following reasons: a history of rheumatic, ischemic, or hypertensive heart disease, valvular heart disease diagnosed by either physical exam or Doppler echocardiography, congestive heart failure, pre-excitation syndromes, hyperthyroidism, excessive alcohol intake, or cardiomyopathy. Patient Categories
In all patients the presence of atrial fibrillation was documented electrocardiographically. Atrial fibrillation was chronic in 1 1 patients and paroxysmal in 12 patients. Paroxysmal atrial fibrillation was defined as at least two documented episodes of atrial fibrillation with an intervening period of normal sinus rhythm. In this group the echocardiogram was obtained in normal sinus rhythm. The average age was 63.0k9.1 years and was not different in the two groups. There were 18 men and 5 women equally distributed in the two groups. Follow-Up Patients were followed for an average of 6.2 years with chronic atrial fibrillation were followed for 7.0 years compared with 5.4 years in the paroxysmal group (p =NS). ( I . I - 12.8 years), those
Echocardiographic Studies Equipment Echocardiographic studies performed prior to 1982 utilized an IREX M-mode ultrasound system. Only M-mode studies were performed prior to 1982; thereafter all studies includd two-dimensional guided M-mode echocardiogram. After 1985, all studies were performed on a Hewlett Packard 77020 ultrasonoscope equipped with a 2.5 MHz transducer and M-mode, two-dimensional echocardiography, and both pulse wave and continuous wave Doppler echocardiographic capabilities. Number of Echocardiographic Studies All patients underwent an average of 3.0 (2.0-5.0) echocardiographic studies during the follow-up period. All patients underwent at least one recent Doppler echocardiographic study to further exclude the presence of valvular heam disease. Echocardiographic Measurements M-mode tracings were of good quality and allowed accurate measurement of cardiac chambers. The following
653
measurements, adhering to the criteria of the American Society of Echocardiography,!’ were made in all subjects: left atrial anterior-posterior dimension, including the posterior aortic wall at end ventricular systole; left ventricular end-diastolic dimension; left ventricular end-systolic dimension; septa1 and posterior wall thickness. Left atrial size was adjusted for body surface area, calculated at the time of each echocardiographic study. The fractional shortening (FS) was calculated for each echocardiographic study according to the following formula: FS (%) = [(LVD-LVS)/LVD] x 100 Left ventricular mass was calculated for each echocardiographic study according to the anatomically corrected regression equation of Devereaux:12 Left ventricular mass (g)= 1.04 [(LVD+IVS+LVPW) 3-(LVD) 31- 13.6 where LVD= left ventricular diastolic diameter, LVS =left ventricular systolic diameter, IVS =interventricular septum thickness, and LVPW =posterior wall thickness. Left ventricular mass was then adjusted for height.I3 All echocardiographic studies were interpreted by two readers (SL; SR), who were blinded with regard to the diagnosis, temporal sequence, and subject. Pearson correlation coefficients were calculated for each pair of measurements. The correlation coefficient for left atrial size was .96. The correlation coefficients for left ventricular diastolic diameter, systolic diameter, interventricular septum, and posterior wall were 0.83,0.75,0.50, and 0.45, respectively. The two observers’ measurements were then averaged. After confirming the reliability of the two readers, the initial and final echocardiograms were selected for all subsequent analysis. Statistical Analysis Data were expressed as the mean and standard deviation. The unpaired Student’s t-test was used to compare between group measurements. Within subjects changes in echocardiographic measurements were analyzed with a paired Student’s t-test. Multiple regression analysis was performed to evaluate the relative contribution of age, duration of follow-up, baseline echocardiographic measurements, and type of atrial fibrillation on the change in left atrial size. Statistical significance was defined as p < 0.05.
In the entire cohort, the left atrial size increased an average of 5.6f6 mm during the follow-up period (p
