- Email: [email protected]
Atrial fibrillation in mitral stenosis: histologic, hemodynamic and metabolic factors
International Elsevier
143
Journal of Cardiologv, 5 (1984) 143-152
IJC 00105
Atria1 fibrillation in mitral stenosis: histologic, hemodynamic and metabolic factors Donald
V. Unverferth
‘, Richard H. Fertel ‘, Barbara Carl V. Leier 4
J. Unverferth
3 and
’ Division of Cardiology; ’ Department of Pharmacology; ’ Department of Pathology; 4 Division of Cardiology; Ohio State University College of Medicine, Columbus, OH, U.S.A. (Received
Unverferth
12 April 1983; revision received 11 August 1983; accepted 12 September 1983)
DV, Fertel RH, Unverferth
stenosis: histologic, 5:143-152.
hemodynamic
and
BJ, Leier CV. Atria1 fibrillation in mitral metabolic factors. Int J Cardiol 1984;
We examined the histologic, hemodynamic and metabolic factors associated with rheumatic mitral stenosis. Eighteen patients comprised three groups: Group I - 7 patients in sinus rhythm; Group II - 5 patients in intermittent atrial fibrillation; Group III - 6 patients in chronic atrial fibrillation. The left atrial dimension was determined by echocardiography. Left atrial pressure, mitral valve gradient, mitral valve area and the presence or absence of calcium in the mitral valve were determined at catheterization. The left atrial appendage was removed during open heart surgery and the tissue was analyzed for cell size, percent fibrosis and content of cyclic AMP and GMP. There was no difference between the groups in pulmonary capillary wedge pressure, mitral valve gradient, mitral valve area or the presence of calcium. The Group I left atrial dimension (51 f 2 mm, xk SE) was significantly smaller than that of Group III (56 k 2 mm, P < 0.05). Group II was not different from Groups I or III. Although the concentration of cyclic AMP did not differ among the groups, the cyclic GMP was significantly depressed in Group III (0.15 * 0.02 fmol/pg protein) when compared to Group I (0.24 k 0.05 fmol/pg protein, P -c0.01). Group II had intermediate values which did not differ from Groups I or III. The percent fibrosis was greatest in Group III (34.8 f 1.8%) and least in Group I (27.2 f 2.8%, P < 0.05). There was no difference in cell size among the groups. Although atrial fibrillation may lead to some of
This study was supported in part by the Central Ohio Heart Chapter of the American Heart Association, the James D. Casto Research Fund and the Eagles Cardiovascular Research Fund. Reprint requests to: Donald V. Unverferth, M.D., 657 Means Hall, 1655 Upham Drive, Columbus, OH 43210. U.S.A.
0167-5273/84/$03.00
0 1984 Elsevier Science Publishers B.V.
144
these irregularities, a depressed cyclic GMP, increased fibrosis and increased left atrial dimension may play a role in the pathogenesis of irreversible atrial fibrillation.
Introduction Rheumatic mitral valve disease is characterized by an irregular clinical deterioration from the progression of valve disease [l] and from the onset of atria1 dysrhythmias [1,2]. The loss of an appropriately timed atria1 contraction often triggers a decrease of cardiac output resulting in previously asymptomatic patients becoming disabled [1,3]. The pathogenesis of atria1 fibrillation in rheumatic mitral valve disease has been the subject of a review by Noble and Fisch [4] who hypothesized that the combination of mitral valve deformity and atrial inflammation secondary to rheumatic carditis results in left atria1 dilatation, fibrosis of the atria1 wall and degeneration of the atria1 muscle. This results in a marked disparity of the rates of recovery and conduction throughout the atria. The resulting premature atria1 activation may precipitate atria1 dysrhythmias, including atria1 fibrillation. Finally, they suggest that prolonged atria1 fibrillation results in disuse atrophy, further inhomogeneity in refractoriness and conduction and irreversible atria1 fibrillation [4]. Metabolic factors and especially cyclic nucleotides are known to play a role in ischemia-induced ventricular fibrillation [5--91. However, the role of cyclic nucleotides in atria1 arrhythmias is not known. The purpose of this study was to examine further the histologic and hemodynamic factors and the cyclic nucleotide levels associated with atrial fibrillation in rheumatic heart disease. Such associations may help elucidate the genesis of this dysrhythmia.
Materials and Methods Eighteen patients with rheumatic heart disease and predominant mitral stenosis were the subjects for the study. The average age was 48.8 k 8.6 (x* SD) years and there were 12 women and 6 men. Written informed consent was obtained from all subjects. Two of the patients in this series required open commissurotomy; 16 had mitral valve replacement. The patients were divided into 3 groups based on the analysis of their heart rhythm during the 3 months prior to open heart surgery. Group I consisted of 7 patients who were in sinus rhythm; Group II, 5 patients who had intermittent atria1 fibrillation; and Group III, 6 patients with chronic atria1 fibrillation persisting for at least the 3 months prior to surgery. The left atrial dimension was determined by M-mode echocardiography. Echocardiograms were performed prior to open heart surgery with an SKF Echoline 20A echocardiograph and an Irex recorder. A single echocardiographic technician performed the echocardiograms using identical techniques. Blinded echocardiographic interpretation was performed in accordance with the recommendations of the American Society of Echocardiography [lo]. Cardiac catheterization was performed on all patients between 2 days and 6
145
weeks prior to mitral valve replacement. None of the patients had a coronary artery obstruction measuring > 50% of the diameter of the vessel. Left and right, heart pressures were measured using a Hewlett Packard Model 4568C pressure unit and recorder at a cut-off frequency of 0 to 250 Hz. Cardiac outputs were determined in duplicate by indicator dilution using indocyanine green. The pulmonary capillary wedge pressure and the left ventricular pressure were recorded simultaneously. The planimetered mitral valve gradient and the mitral valve area were calculated by the method of Gorlin and Gorlin [ll]. In addition, the presence or absence of calcium in the mitral valve was determined by fluoroscopy prior to the injection of angiographic dye. Open heart surgery for mitral valve replacement was performed using a standard cardioplegic solution of 25 mEq of sodium bicarbonate, 15 g dextrose, 20 mEq of potassium chloride, 20 U of regular insulin and 1 g of procainamide in 1 liter of Plasmalyte 148@. While the patient was on cardiopulmonary bypass the left atria1 appendage was excised and divided into two parts. One half of the appendage was submerged in liquid nitrogen within 45 set of excision. The other half of the appendage was cut into l-mm cubes, placed in 2% glutaraldehyde and reserved for histologic analysis. The frozen biopsies from each procedure were stored in liquid nitrogen ( - 200°C) until analysis. The biopsies were then placed in 1 ml of 5% trichloroacetic acid and homogenized. The tissue was sonicated for 30 set and then centrifuged at 300 g for 15 min. Cyclic nucleotides (cyclic AMP and cyclic GMP) were assayed by a radioimmunoassay procedure [12] modified to produce increased sensitivity [13]. The reproducibility of this assay in our laboratory has been previously reported [14]. The l-mm cubes of glutaraldehyde fixed tissue were embedded in paraffin and light microscopic slides of 3-5 pm thickness were prepared. The slides were stained with hematoxylin and eosin and Masson’s trichrome stains using the methods of Hrapchik and Sheehan [15]. A stereoscope was used to assess the amount of fibrosis. A grid of 243 cross points was laid on each area at a magnification of 400 x . The number of cross points lying on fibrosed areas was divided by the total number of cross points evaluated (X 100) to obtain the percent fibrosis. Three areas were evaluated and the percent fibrosis for the three was averaged. There was a < 3% variability on repeated measures of the fibrosis. The microscopist was blinded to the patient group. The sections were also examined using a 43 X objective lens and an ocular micrometer disc. The diameter of 50 cells from each atrial appendage was determined by measuring the shortest distance across the cell at the nucleus. The cell diameter of each patient’s atrial appendage represents the mean of 50 measurements. Variability on repeated measurements was < 1.2 pm.
statistics Hemodynamic factors, atrial diameter, cyclic nucleotides, percent fibrosis and cell diameter were related between groups by analysis of variance with appropriate after tests.
146
Results The Group I left atria1 dimension (51 + 2 mm, ,%?k SE) was significantly smaller than Group III (56 f 2 mm, P < 0.05). Group II was not different from Groups I or III (Fig. 1). There were no significant differences among the groups in pulmonary capillary wedge pressure, mitral valve gradient or mitral valve area (see Fig. 2). There was no difference among the groups in fluoroscopically evident calcium in the mitral valve. The number of patients with calcium in each group was: Group I 3/7; Group II - 3/5; and, Group III - 3/6. MVA
MVG
PCW
LA we
. 62 -
.
!36-
mm
24.
2L ..
50.
:
22 . :
-. 20. i’:
:
.
.
-
mmHg I6
-
.
l
.
t-c . . . . . .. . .
14
IO
I
I
It
In
1 urn LNShSJ -NS-
lnm I- N!+NS-
LNS--ILNS--I
-N6-
-pco.o5-
1.5
:
.
mmHg
7
.
l
i 28 .
.
54.
26
32
: .
.
.
.
:
.+.
1.3
.
cm I. I
+
.9
:
.7
l
.. : .
t_ I
urn
~NSNS-J
-NS-
Fig. 2
Fig. 1
Fig. 1. Relationship of atria1 size to rhythm. The graph presents the individual data points and the mean. Group I (sinus rhythm) had significantly smaller left atria1 size by echocardiography than Group III (atrial fibrillation). Group II (intermittent atria1 fibrillation) did not have a significantly different left atrial size from Groups I or III. Fig. 2. Relationship of hemodynamics to atrial rhythm. There were no significant differences between the groups in pulmonary capillary wedge pressure (PCW), mitral valve gradient (MVG) or mitral valve area (MVA).
1.:.
-.- 1: L
CAMP &GM=
cGMP
CAMP
5.4
26
. .
.
4.6
3.6
:
-
l
fm,
3.0
:
l
l
.
.
I
JIIJJ
LI’6*NSJ -NS-
:
.-
+
giLn
:
2.2
22
24
I6
.
14
. .
.
l i
&-
l-NS *NSJ Lpco.01~
Fig. 3. Relationship of cychc nucleotides to atrial rhythm. groups in the levels of cyclic AMP (CAMP), the cGMP when compared to the levels of cGMP in Group I (sinus of cGMP. The ratio of cAMP/cGMP was significantly
I
am
l-NSLNSJ l- pco.01 d
Although there were no differences between the was depressed in Group III (atria1 fibrillation) rhythm). Group II was intermediate in its level greater in Group III than this ratio in Group I.
147 7.
FIBROSIS
CELL
DIAMETER
.
1
II
Ill
I
LNSJLNSJ
II
LNS-l
~p-ao.05~
m LNSJ
-NS-
Fig. 4. Relationship of histologic factors to atrial rhythm. The percent fibrosis in Group III was significantly greater than that in Group I. The fibrosis in Group II was intermediate and not different from Groups I or III. The right side of the illustration demonstrates that there is no difference in left atria1 appendage cell diameter between the groups. There was a significant difference between the cyclic nucleotide concentrations of the groups (Fig. 3). Group III had a significantly depressed level of cyclic GMP (0.15 + 0.02 fmol/pg protein, x+ SE) when compared to Group I (0.24 f 0.05 fmol/pg protein, P -c0.01). Also, the ratio of cyclic AMP to cyclic GMP was significantly greater in Group III (22.8 & 2.8) than in Group I (11.2 f 2.0, P < 0.01). Group
II had
intermediate
values
which
were
not significantly
different
from
those
I or III. There was no significant difference in the concentration of cyclic AMP among the groups. There was an increase in the percent fibrosis from 27.2 + 2.8% in Group I to 34.8 * 1.8% (P < 0.05) in Group III (Fig. 4). Group II was not significantly different from Group I or Group III. There was no difference in cell size between the groups (Fig. 4). of Groups
Discussion This study utilized the surgically removed left atria1 appendage for histologic and metabolic analyses. The ability to rapidly preserve tissue from surgery for histologic and metabolic studies has obvious advantages over the autolyzed tissue obtained at autopsy. The left atria1 appendage is a remnant of the primitive left atrium and is true left atria1 tissue [16]. Removal of the left atria1 appendage is not deleterious [17] and has been used previously for research studies [18-201. The atria1 appendages of 18 patients were studied. The number of observations in each group is small, however, and caution is advised in interpreting the results. Our study was designed to examine factors which may be associated with atria1 fibrillation in rheumatic mitral stenosis. Despite the small sample size, several relationships were noteworthy including an increased left atria1 size, depressed cyclic GMP with an elevated cyclic AMP/cyclic GMP ratio and the increased percent fibrosis in the patients with atria1 fibrillation. The factors which did not bear an
148
apparent relationship to these atria1 dysrhythmias included the pulmonary capillary wedge pressure, mitral valve pressure gradient, mitral valve area, calcium in the mitral valve and the cell diameter of the muscle cells of the left atria1 appendage. Factors not Associated with Rhythm The highest left atria1 pressures and greatest pressure gradients across the mitral valve were not seen in the patients with atria1 fibrillation. These pressures may have changed after the onset of atria1 fibrillation. Selzer has recorded a fall of atria1 pressure with the onset of fibrillation and ascribed this fall to a decrease of cardiac output [2]. Because of this effect of atria1 fibrillation on pressures, no statement can be made concerning this factor from our study. Calcification [21] and a decreased orifice size of the mitral valve [22] are associated with more symptoms and greater hemodynamic compromise. However, calcification was not more common in those with atria1 fibrillation and the mitral valve area was not significantly decreased compared to the other two groups. Thus, valvular characteristics did not appear to be major factors in the genesis of atria1 fibrillation. Another important negative finding of this study was that the left atrial appendage cell size did not relate to the presence of atrial fibrillation. Normal left atria1 cell size is 6 to 8 pm in diameter and 20 to 30 pm in length [23,24]. Although the cell diameter of these patients was increased (12 to 17 pm), there were no significant differences among the groups. Chronic and excessive hypertrophy of ventricular heart muscle cells causes abnormalities of cell metabolism (inefficient energy utilization [25] and depressed enzyme activity [26-281) and morphology (decreased ratio of mitochondria to myofibrils [29]). There is no information, however, that hypertrophy alone predisposes to dysrhythmias. Our study suggests that atria1 cell hypertrophy is not an important factor in the genesis of atria1 fibrillation. The finding that there was no cellular atrophy with chronic atria1 fibrillation differs from the conclusions of Bailey et al. [30]. Although he did not use quantitative techniques for the measurement of cell size, his study did employ a semiquantitative estimate of muscle mass. His study suggested that atria1 fibrillation leads to disuse atrophy and a self-perpetuation of chronic atria1 fibrillation. Atria1 cell size was not decreased in our patients and thus there was no atrophy. Overall atria1 muscle mass may be decreased, however, due to cell death. Factors Associated with Rhythm Despite the lack of a relationship of left atria1 cell size to rhythm, the left atria1 dimension was related. The patients in atria1 fibrillation had the largest left atria. These results support previous conclusions that the presence of atrial fibrillation is closely related to left atria1 dilatation in patients with mitral valve disease [31,32]. The study of Henry et al. [31] showed that atria1 fibrillation was rare below an atria1 dimension of 40 mm and common above 45 mm. Our study would suggest a higher
149
critical atria1 dimension of 53 mm. Both studies agree, however, on the importance of left atria1 dimension in the development of atria1 fibrillation. The relationship of atria1 enlargement with atria1 fibrillation does not prove causation. However, electrophysiologic studies by Hordof et al. [33] and Ten Eick and Singer [34] have convincingly demonstrated that cells from dilated or diseased human atria have depressed maximum diastolic potential and slow response action potentials known to be associated with slow conduction, unidirectionai block and re-entry [35]. The presence of slow response action potentials in the atria could permit the initiation and propagation of re-entrant arrhythmias. Similarly, the spontaneous rhythms that can be generated by slow response action potentials could, under appropriate conditions, give rise to ectopic atria1 rhythms [33]. Cyclic nucleotides play an important role in myocardial contractility. A rise of cyclic AMP improves cardiac function [36,37] while cyclic GMP is a negative inotrope [38,39]. In addition, cyclic AMP is known to play an important role in the production of dysrhythmias. Podzuweit et al. [5] have demonstrated that with profound ischemia there is a rise of cyclic AMP followed by ventricular fibrillation. If cyclic AMP does not rise with ischemia then there is no fibrillation [38]. Other investigators have also noted that the rise of cyclic AMP is related to dysrhythmias [6-B], that fibrillation alone does not raise cyclic AMP [9] and that the prevention of the rise of cyclic AT is protective against fibrillation [9]. Despite this well-documented importance of cyclic AMP in the genesis of ventricular dysrhythmias, there was no change of cyclic AMP in the atria1 appendages of patients in atria1 fibrillation. No previous studies have investigated the role of cyclic GMP in dysrhythmia production. However, in our study the fibrillating left atria1 appendage cells (which would be hypopolarized according to previous studies [33,34]) had decreased levels of cyclic GMP. It has been demonstrated that acetylcholine, which stimulates guanylate cyclase and increases cyclic GMP, increases the resting potential of the cell [40]. Thus, there is indirect evidence that a depressed level of cyclic GMP plays a role in the hypopolarization of the cell which in turn leads to dysrhythmias. In addition, the atria1 appendage cells of patients in atria1 fibrillation demonstrated an elevated cyclic AMP/cyclic GMP ratio. This ratio is elevated in ischemia-related dysrhythmias as well as in mitral stenosis-related atria1 fibrillation and may be a more important factor than the level of either cyclic nucleotide alone. The atria of patients with rheumatic heart disease have deposits of intercellular collagen which disrupt tissue geometry [30]. Davies and Pomerance noted this association but also observed that those patients who had a greater duration of atria1 fibrillation had a greater amount of atria1 fibrosis [41]. They hypothesized that fibrillation induced fibrosis [41]. These observations are consistent with our study which has shown fibrosis in the left atria1 appendage of patients with rheumatic heart disease. Fibrosis was found in greatest concentration in the group with atria1 fibrillation and least in those patients in sinus rhythm. Although fibrillation may play some role in the genesis of fibrosis, the presence of fibrosis causes an inhomogeneity of electrical conduction which may lead to irreversible atria1 fibrillation [30].
150
Con&ding Observations This study is in basic agreement with the Noble and Fisch hypothesis of the pathogenesis of atria1 fibrillation. One modification, however, is that the term “cellular atrophy” be replaced. Our study suggests that there may be cellular drop-out with replacement fibrosis but that the remaining cells become hypertrophic. Ten Pick et al. [32] and Hordof et al. [33] have added the concept of hypopolarization of diseased atrial cells which may be related to the depressed levels of cyclic GMP found in our study. The least understood concept is the role of metabolic factors in the pathogenesis of atria1 fibrillation. Further studies of the cyclic nucleotides, adenine nucleotides, calcium metabolism and others are needed to complete these concepts.
Acknowledgements The authors would like to thank Marion Lewis for her assistance in the handling of the tissue specimens, Pam Newton for technical assistance, Dr. James Kilman and Dr. Jack Vasko for supplying atrial appendage tissue and Tami Smith for the preparation of the manuscript.
References 1 SeLzer A, Cohn KE. Natural history of mitral stenosis: a review. Circulation 1972;45:878-890. 2 Seizer A. Effects of atrial fibrillation upon the circulation in patients with mitral stenosis. Am Heart J 1960;59:518-526. 3 Mitchell JH, Shapiro W. Atrial function and the hemodynamic consequences of atria1 fibrillation in man. Am J Cardiol 1969;23:556-566. 4 Noble RJ, Fisch C. Factors in the genesis of atria1 fibrillation in rheumatic valvular disease. Cardiovasc Chn 1972;5:97-114. 5 Podzuweit T, Dalby AJ, Cherry GW, Opie LH. Cyclic AMP levels in ischaemic and non-ischaemic myocardium following coronary artery ligation: relation to ventricular fibrillation. J Mol Cell Cardiol 1978;10:81-94. 6 Sugiura M, Ogawa K, Yamazaki N. Concentration of myocardial cyclic AMP and ventricular fibrillation induced by aminophylline. Jap Heart J 1979;20:177-182. 7 Podzuweit T. Catecholamine-cyclic-AMP-Ca*+-induced ventricular tachycardia in the intact pig heart. Basic Res Cardiol 1980;75:772-779. 8 Cobb PB, Witkowski FX, Sobel BE. Mechanisms contributing to malignant dysrhythmias induced by ischemia in the rat. J Clin Invest 1979;61:109-119. 9 Lubbe WF, McFadyen ML, Muller CA, Worthington M, Opie LH. Protective action of amiodarone against ventricular fibrillation in the isolated perfused rat heart. Am J Cardiol 1979;43:533-540. 10 Sahn DJ, DeMaria A, Kiss10 J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072-1083. 11 Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac valves, and central circulatory shunts. Am Heart J 1951;41:42-45. 12 Steiner AL, Parker CW, Kipnis DM. A radioimmunoassay for measurement of cyclic nucleotides. J Biol Chem 1972;247:1106-1113. 13 Miyadhi M, Mizudhi N, Sebo R. Increased sensitivity of the cyclic nucleotide radioimmunoassay. Ann Biochem 1977:77:429-435.
151
14 Unverferth DV, Fertel RH, Altschuld R, Magorien RD, Lewis RP, Leier CV. Biochemical measurements of endomyocardial biopsies. Cath Cardiovasc Diag 1981;7:55-64. 15 Sheehan DC, Hrapchak BB. Connective tissue and muscle fiber stains. In: Sneehan OC, Hropchak BB, eds. Theory and practice of histotechnology. St. Louis: CV Mosby Company, 1973. 16 Langman J. Cardiovascular system. In: Langman J, ed. Medical embryology. Baltimore: Williams & Wilkins, 1975. 17 Thomas TV. Left atria1 appendage and valve replacement. Am Heart J 1972;84:838-839. 18 Quintiliani R, Klimek J, Nightingale CH. Penetration of cephapirin and caphalothin into the right atria1 appendage and pericardial fluid of patients undergoing open-heart surgery. J Infect Dis 1979;139:348-352. 19 Martin P. Atria1 inotropic responses to brief vagal stimuli: frequency-force interactions. Am J Physiol 1980;239:H333-H341. 20 Green ER, Subramanian S, Faden H, Quintiliani R, Nightingale CH. A comparison of the penetration characteristics of cephapirin and caphalothin into right atria1 appendage, muscle, fat, and pericardial fluid of pediatric patients undergoing open-heart operation. Ann Thorac Surg 1981;31:155-160. 21 Wooley CF, Baba N, Kilman JW, Ryan JM. Thrombotic calcific mitral stenosis. Morphology of the calcific mitral valve. Circulation 1974;49:1167-1174. 22 Harvey RM, Ferrer MI, Samet P, et al. Mechanical and myocardial factors in rheumatic heart disease with mitral stenosis. Circulation 1955;11:531-551. 23 Legato MJ, Bull MB, Ferrer MI. Atria1 ultrastructure in patients with fixed intra-atria1 block. Chest 1974;65:252-261. 24 Legato MJ. Ultrastructure of the atria], ventricular and purkinje cell with special reference to the genesis of arrhythmias. Circulation 1973;47:178-189. 25 Strauer BE, Tauchert M. Evidence for an inefficient energy utilization in cardiac hypertrophy. Studies on the isolated human ventricular myocardium. Klin Wochenschr 1973;51:322-326. 26 Shiverick KT, Hamrell BB, Alpert NR. Structural and functional properties of myosin associated with the compensatory cardiac hypertrophy in the rabbit. J Mol Cell Cardiol 1976;8:837-851. 27 Wikman-Coffelt J, Walsh R, Fenner C, Kamiyama T, Sale1 A, Mason DT. Effects of severe hemodynamic pressure overload on the properties of canine left ventricular myosin: mechanism by which myosin ATPase activity is lowered during chronic increased hemodynamic stress. J Mol Cell Cardiol 1976;8:263-270. 28 Wikman-Coffelt J, Fenner C, Coffelt RJ, Sale1 A, Kamiyama T, Mason DT. Chronological effects of mild pressure overload on myosin ATPase activity in the canine right ventricle. J Mol Cell Cardiol 1975;7:219-224. 29 Goldstein MA, Sordahl LA, Schwartz A. Ultrastructural analysis of left ventricular hypertrophy in rabbits. J Mol Cell Cardiol 1974;6:265-273. 30 Bailey GWH, Braniff BA, Hancock W, Cohn KE. Relation of left atria1 pathology to atria1 fibrillation in mitral valvular disease. Ann Intern Med 1968;69:13-20. 31 Henry WL, Morganroth J, Pearlman AS, et al. Relation between echocardiographically determined left atria1 size and atria1 fibrillation. Circulation 1975;53:273-279. 32 Probst P, Goldschlager N, Selzer A. Left atrial size and atria1 fibrillation in mitral stenosis. Factors influencing their relationship. Circulation 1973;48:1282-1287. 33 Hordof AJ, Edie R, Malm JR, Hoffman BF, Rosen MR. Electrophysiologic properties and response to pharmacologic agents of fibers from diseased human atria. Circulation 1976;54:774-779. 34 Ten Eick RE, Singer DH. Electrophysiological properties of diseased human atrium. I. Low diastolic potential and altered cellular response to potassium. Circ Res 1979;44:545-557. 35 Wit AL, Rosen MR, Hoffman BF. Relationship of normal and abnormal electrical activity of cardiac fibers to the genesis of arrhythmias. Re-entry. Am Heart J 1974;88:664-670. 36 Hicks MJ, Shigekawa M, Katz AM. Mechanism by which CAMP dependent protein kinase stimulates calcium transport in cardiac sarcoplasmic reticulum. Circ Res 1979;44:384-391. 37 Brooker G. Oscillation of the CAMP concentration during the myocardial contraction cycle. Science 1973;182:933-934. 38 Katsuki S, Arnold WP, Murad F. Effects of sodium nitroprusside, nitroglycerin, and sodium azide on levels of cyclic nucleotides and mechanical activity of various tissues. J Cycl Nucl Res 1977;3:239-247.
152
39 Kohlhardt M, Haap K. 8-Guanosine 3’5’ monophosphate mimics the effect of acetylcholine on slow response action potential and contractile force in mammalian atria1 myocardium. J Mel Cell Cardiol 1978;10:573-586. 40 Ten Eick R, Nawrath H, McDonald TF, Trautwein W. On the mechanism of the negative inotropic effect of acetylcholine. Eur J Physiol 1976;361:207-213. 41 Davies MJ, Pomerance A. Pathology of atria1 fibrillation in man. Br Heart J 1972;34:520-525.
143
Journal of Cardiologv, 5 (1984) 143-152
IJC 00105
Atria1 fibrillation in mitral stenosis: histologic, hemodynamic and metabolic factors Donald
V. Unverferth
‘, Richard H. Fertel ‘, Barbara Carl V. Leier 4
J. Unverferth
3 and
’ Division of Cardiology; ’ Department of Pharmacology; ’ Department of Pathology; 4 Division of Cardiology; Ohio State University College of Medicine, Columbus, OH, U.S.A. (Received
Unverferth
12 April 1983; revision received 11 August 1983; accepted 12 September 1983)
DV, Fertel RH, Unverferth
stenosis: histologic, 5:143-152.
hemodynamic
and
BJ, Leier CV. Atria1 fibrillation in mitral metabolic factors. Int J Cardiol 1984;
We examined the histologic, hemodynamic and metabolic factors associated with rheumatic mitral stenosis. Eighteen patients comprised three groups: Group I - 7 patients in sinus rhythm; Group II - 5 patients in intermittent atrial fibrillation; Group III - 6 patients in chronic atrial fibrillation. The left atrial dimension was determined by echocardiography. Left atrial pressure, mitral valve gradient, mitral valve area and the presence or absence of calcium in the mitral valve were determined at catheterization. The left atrial appendage was removed during open heart surgery and the tissue was analyzed for cell size, percent fibrosis and content of cyclic AMP and GMP. There was no difference between the groups in pulmonary capillary wedge pressure, mitral valve gradient, mitral valve area or the presence of calcium. The Group I left atrial dimension (51 f 2 mm, xk SE) was significantly smaller than that of Group III (56 k 2 mm, P < 0.05). Group II was not different from Groups I or III. Although the concentration of cyclic AMP did not differ among the groups, the cyclic GMP was significantly depressed in Group III (0.15 * 0.02 fmol/pg protein) when compared to Group I (0.24 k 0.05 fmol/pg protein, P -c0.01). Group II had intermediate values which did not differ from Groups I or III. The percent fibrosis was greatest in Group III (34.8 f 1.8%) and least in Group I (27.2 f 2.8%, P < 0.05). There was no difference in cell size among the groups. Although atrial fibrillation may lead to some of
This study was supported in part by the Central Ohio Heart Chapter of the American Heart Association, the James D. Casto Research Fund and the Eagles Cardiovascular Research Fund. Reprint requests to: Donald V. Unverferth, M.D., 657 Means Hall, 1655 Upham Drive, Columbus, OH 43210. U.S.A.
0167-5273/84/$03.00
0 1984 Elsevier Science Publishers B.V.
144
these irregularities, a depressed cyclic GMP, increased fibrosis and increased left atrial dimension may play a role in the pathogenesis of irreversible atrial fibrillation.
Introduction Rheumatic mitral valve disease is characterized by an irregular clinical deterioration from the progression of valve disease [l] and from the onset of atria1 dysrhythmias [1,2]. The loss of an appropriately timed atria1 contraction often triggers a decrease of cardiac output resulting in previously asymptomatic patients becoming disabled [1,3]. The pathogenesis of atria1 fibrillation in rheumatic mitral valve disease has been the subject of a review by Noble and Fisch [4] who hypothesized that the combination of mitral valve deformity and atrial inflammation secondary to rheumatic carditis results in left atria1 dilatation, fibrosis of the atria1 wall and degeneration of the atria1 muscle. This results in a marked disparity of the rates of recovery and conduction throughout the atria. The resulting premature atria1 activation may precipitate atria1 dysrhythmias, including atria1 fibrillation. Finally, they suggest that prolonged atria1 fibrillation results in disuse atrophy, further inhomogeneity in refractoriness and conduction and irreversible atria1 fibrillation [4]. Metabolic factors and especially cyclic nucleotides are known to play a role in ischemia-induced ventricular fibrillation [5--91. However, the role of cyclic nucleotides in atria1 arrhythmias is not known. The purpose of this study was to examine further the histologic and hemodynamic factors and the cyclic nucleotide levels associated with atrial fibrillation in rheumatic heart disease. Such associations may help elucidate the genesis of this dysrhythmia.
Materials and Methods Eighteen patients with rheumatic heart disease and predominant mitral stenosis were the subjects for the study. The average age was 48.8 k 8.6 (x* SD) years and there were 12 women and 6 men. Written informed consent was obtained from all subjects. Two of the patients in this series required open commissurotomy; 16 had mitral valve replacement. The patients were divided into 3 groups based on the analysis of their heart rhythm during the 3 months prior to open heart surgery. Group I consisted of 7 patients who were in sinus rhythm; Group II, 5 patients who had intermittent atria1 fibrillation; and Group III, 6 patients with chronic atria1 fibrillation persisting for at least the 3 months prior to surgery. The left atrial dimension was determined by M-mode echocardiography. Echocardiograms were performed prior to open heart surgery with an SKF Echoline 20A echocardiograph and an Irex recorder. A single echocardiographic technician performed the echocardiograms using identical techniques. Blinded echocardiographic interpretation was performed in accordance with the recommendations of the American Society of Echocardiography [lo]. Cardiac catheterization was performed on all patients between 2 days and 6
145
weeks prior to mitral valve replacement. None of the patients had a coronary artery obstruction measuring > 50% of the diameter of the vessel. Left and right, heart pressures were measured using a Hewlett Packard Model 4568C pressure unit and recorder at a cut-off frequency of 0 to 250 Hz. Cardiac outputs were determined in duplicate by indicator dilution using indocyanine green. The pulmonary capillary wedge pressure and the left ventricular pressure were recorded simultaneously. The planimetered mitral valve gradient and the mitral valve area were calculated by the method of Gorlin and Gorlin [ll]. In addition, the presence or absence of calcium in the mitral valve was determined by fluoroscopy prior to the injection of angiographic dye. Open heart surgery for mitral valve replacement was performed using a standard cardioplegic solution of 25 mEq of sodium bicarbonate, 15 g dextrose, 20 mEq of potassium chloride, 20 U of regular insulin and 1 g of procainamide in 1 liter of Plasmalyte 148@. While the patient was on cardiopulmonary bypass the left atria1 appendage was excised and divided into two parts. One half of the appendage was submerged in liquid nitrogen within 45 set of excision. The other half of the appendage was cut into l-mm cubes, placed in 2% glutaraldehyde and reserved for histologic analysis. The frozen biopsies from each procedure were stored in liquid nitrogen ( - 200°C) until analysis. The biopsies were then placed in 1 ml of 5% trichloroacetic acid and homogenized. The tissue was sonicated for 30 set and then centrifuged at 300 g for 15 min. Cyclic nucleotides (cyclic AMP and cyclic GMP) were assayed by a radioimmunoassay procedure [12] modified to produce increased sensitivity [13]. The reproducibility of this assay in our laboratory has been previously reported [14]. The l-mm cubes of glutaraldehyde fixed tissue were embedded in paraffin and light microscopic slides of 3-5 pm thickness were prepared. The slides were stained with hematoxylin and eosin and Masson’s trichrome stains using the methods of Hrapchik and Sheehan [15]. A stereoscope was used to assess the amount of fibrosis. A grid of 243 cross points was laid on each area at a magnification of 400 x . The number of cross points lying on fibrosed areas was divided by the total number of cross points evaluated (X 100) to obtain the percent fibrosis. Three areas were evaluated and the percent fibrosis for the three was averaged. There was a < 3% variability on repeated measures of the fibrosis. The microscopist was blinded to the patient group. The sections were also examined using a 43 X objective lens and an ocular micrometer disc. The diameter of 50 cells from each atrial appendage was determined by measuring the shortest distance across the cell at the nucleus. The cell diameter of each patient’s atrial appendage represents the mean of 50 measurements. Variability on repeated measurements was < 1.2 pm.
statistics Hemodynamic factors, atrial diameter, cyclic nucleotides, percent fibrosis and cell diameter were related between groups by analysis of variance with appropriate after tests.
146
Results The Group I left atria1 dimension (51 + 2 mm, ,%?k SE) was significantly smaller than Group III (56 f 2 mm, P < 0.05). Group II was not different from Groups I or III (Fig. 1). There were no significant differences among the groups in pulmonary capillary wedge pressure, mitral valve gradient or mitral valve area (see Fig. 2). There was no difference among the groups in fluoroscopically evident calcium in the mitral valve. The number of patients with calcium in each group was: Group I 3/7; Group II - 3/5; and, Group III - 3/6. MVA
MVG
PCW
LA we
. 62 -
.
!36-
mm
24.
2L ..
50.
:
22 . :
-. 20. i’:
:
.
.
-
mmHg I6
-
.
l
.
t-c . . . . . .. . .
14
IO
I
I
It
In
1 urn LNShSJ -NS-
lnm I- N!+NS-
LNS--ILNS--I
-N6-
-pco.o5-
1.5
:
.
mmHg
7
.
l
i 28 .
.
54.
26
32
: .
.
.
.
:
.+.
1.3
.
cm I. I
+
.9
:
.7
l
.. : .
t_ I
urn
~NSNS-J
-NS-
Fig. 2
Fig. 1
Fig. 1. Relationship of atria1 size to rhythm. The graph presents the individual data points and the mean. Group I (sinus rhythm) had significantly smaller left atria1 size by echocardiography than Group III (atrial fibrillation). Group II (intermittent atria1 fibrillation) did not have a significantly different left atrial size from Groups I or III. Fig. 2. Relationship of hemodynamics to atrial rhythm. There were no significant differences between the groups in pulmonary capillary wedge pressure (PCW), mitral valve gradient (MVG) or mitral valve area (MVA).
1.:.
-.- 1: L
CAMP &GM=
cGMP
CAMP
5.4
26
. .
.
4.6
3.6
:
-
l
fm,
3.0
:
l
l
.
.
I
JIIJJ
LI’6*NSJ -NS-
:
.-
+
giLn
:
2.2
22
24
I6
.
14
. .
.
l i
&-
l-NS *NSJ Lpco.01~
Fig. 3. Relationship of cychc nucleotides to atrial rhythm. groups in the levels of cyclic AMP (CAMP), the cGMP when compared to the levels of cGMP in Group I (sinus of cGMP. The ratio of cAMP/cGMP was significantly
I
am
l-NSLNSJ l- pco.01 d
Although there were no differences between the was depressed in Group III (atria1 fibrillation) rhythm). Group II was intermediate in its level greater in Group III than this ratio in Group I.
147 7.
FIBROSIS
CELL
DIAMETER
.
1
II
Ill
I
LNSJLNSJ
II
LNS-l
~p-ao.05~
m LNSJ
-NS-
Fig. 4. Relationship of histologic factors to atrial rhythm. The percent fibrosis in Group III was significantly greater than that in Group I. The fibrosis in Group II was intermediate and not different from Groups I or III. The right side of the illustration demonstrates that there is no difference in left atria1 appendage cell diameter between the groups. There was a significant difference between the cyclic nucleotide concentrations of the groups (Fig. 3). Group III had a significantly depressed level of cyclic GMP (0.15 + 0.02 fmol/pg protein, x+ SE) when compared to Group I (0.24 f 0.05 fmol/pg protein, P -c0.01). Also, the ratio of cyclic AMP to cyclic GMP was significantly greater in Group III (22.8 & 2.8) than in Group I (11.2 f 2.0, P < 0.01). Group
II had
intermediate
values
which
were
not significantly
different
from
those
I or III. There was no significant difference in the concentration of cyclic AMP among the groups. There was an increase in the percent fibrosis from 27.2 + 2.8% in Group I to 34.8 * 1.8% (P < 0.05) in Group III (Fig. 4). Group II was not significantly different from Group I or Group III. There was no difference in cell size between the groups (Fig. 4). of Groups
Discussion This study utilized the surgically removed left atria1 appendage for histologic and metabolic analyses. The ability to rapidly preserve tissue from surgery for histologic and metabolic studies has obvious advantages over the autolyzed tissue obtained at autopsy. The left atria1 appendage is a remnant of the primitive left atrium and is true left atria1 tissue [16]. Removal of the left atria1 appendage is not deleterious [17] and has been used previously for research studies [18-201. The atria1 appendages of 18 patients were studied. The number of observations in each group is small, however, and caution is advised in interpreting the results. Our study was designed to examine factors which may be associated with atria1 fibrillation in rheumatic mitral stenosis. Despite the small sample size, several relationships were noteworthy including an increased left atria1 size, depressed cyclic GMP with an elevated cyclic AMP/cyclic GMP ratio and the increased percent fibrosis in the patients with atria1 fibrillation. The factors which did not bear an
148
apparent relationship to these atria1 dysrhythmias included the pulmonary capillary wedge pressure, mitral valve pressure gradient, mitral valve area, calcium in the mitral valve and the cell diameter of the muscle cells of the left atria1 appendage. Factors not Associated with Rhythm The highest left atria1 pressures and greatest pressure gradients across the mitral valve were not seen in the patients with atria1 fibrillation. These pressures may have changed after the onset of atria1 fibrillation. Selzer has recorded a fall of atria1 pressure with the onset of fibrillation and ascribed this fall to a decrease of cardiac output [2]. Because of this effect of atria1 fibrillation on pressures, no statement can be made concerning this factor from our study. Calcification [21] and a decreased orifice size of the mitral valve [22] are associated with more symptoms and greater hemodynamic compromise. However, calcification was not more common in those with atria1 fibrillation and the mitral valve area was not significantly decreased compared to the other two groups. Thus, valvular characteristics did not appear to be major factors in the genesis of atria1 fibrillation. Another important negative finding of this study was that the left atrial appendage cell size did not relate to the presence of atrial fibrillation. Normal left atria1 cell size is 6 to 8 pm in diameter and 20 to 30 pm in length [23,24]. Although the cell diameter of these patients was increased (12 to 17 pm), there were no significant differences among the groups. Chronic and excessive hypertrophy of ventricular heart muscle cells causes abnormalities of cell metabolism (inefficient energy utilization [25] and depressed enzyme activity [26-281) and morphology (decreased ratio of mitochondria to myofibrils [29]). There is no information, however, that hypertrophy alone predisposes to dysrhythmias. Our study suggests that atria1 cell hypertrophy is not an important factor in the genesis of atria1 fibrillation. The finding that there was no cellular atrophy with chronic atria1 fibrillation differs from the conclusions of Bailey et al. [30]. Although he did not use quantitative techniques for the measurement of cell size, his study did employ a semiquantitative estimate of muscle mass. His study suggested that atria1 fibrillation leads to disuse atrophy and a self-perpetuation of chronic atria1 fibrillation. Atria1 cell size was not decreased in our patients and thus there was no atrophy. Overall atria1 muscle mass may be decreased, however, due to cell death. Factors Associated with Rhythm Despite the lack of a relationship of left atria1 cell size to rhythm, the left atria1 dimension was related. The patients in atria1 fibrillation had the largest left atria. These results support previous conclusions that the presence of atrial fibrillation is closely related to left atria1 dilatation in patients with mitral valve disease [31,32]. The study of Henry et al. [31] showed that atria1 fibrillation was rare below an atria1 dimension of 40 mm and common above 45 mm. Our study would suggest a higher
149
critical atria1 dimension of 53 mm. Both studies agree, however, on the importance of left atria1 dimension in the development of atria1 fibrillation. The relationship of atria1 enlargement with atria1 fibrillation does not prove causation. However, electrophysiologic studies by Hordof et al. [33] and Ten Eick and Singer [34] have convincingly demonstrated that cells from dilated or diseased human atria have depressed maximum diastolic potential and slow response action potentials known to be associated with slow conduction, unidirectionai block and re-entry [35]. The presence of slow response action potentials in the atria could permit the initiation and propagation of re-entrant arrhythmias. Similarly, the spontaneous rhythms that can be generated by slow response action potentials could, under appropriate conditions, give rise to ectopic atria1 rhythms [33]. Cyclic nucleotides play an important role in myocardial contractility. A rise of cyclic AMP improves cardiac function [36,37] while cyclic GMP is a negative inotrope [38,39]. In addition, cyclic AMP is known to play an important role in the production of dysrhythmias. Podzuweit et al. [5] have demonstrated that with profound ischemia there is a rise of cyclic AMP followed by ventricular fibrillation. If cyclic AMP does not rise with ischemia then there is no fibrillation [38]. Other investigators have also noted that the rise of cyclic AMP is related to dysrhythmias [6-B], that fibrillation alone does not raise cyclic AMP [9] and that the prevention of the rise of cyclic AT is protective against fibrillation [9]. Despite this well-documented importance of cyclic AMP in the genesis of ventricular dysrhythmias, there was no change of cyclic AMP in the atria1 appendages of patients in atria1 fibrillation. No previous studies have investigated the role of cyclic GMP in dysrhythmia production. However, in our study the fibrillating left atria1 appendage cells (which would be hypopolarized according to previous studies [33,34]) had decreased levels of cyclic GMP. It has been demonstrated that acetylcholine, which stimulates guanylate cyclase and increases cyclic GMP, increases the resting potential of the cell [40]. Thus, there is indirect evidence that a depressed level of cyclic GMP plays a role in the hypopolarization of the cell which in turn leads to dysrhythmias. In addition, the atria1 appendage cells of patients in atria1 fibrillation demonstrated an elevated cyclic AMP/cyclic GMP ratio. This ratio is elevated in ischemia-related dysrhythmias as well as in mitral stenosis-related atria1 fibrillation and may be a more important factor than the level of either cyclic nucleotide alone. The atria of patients with rheumatic heart disease have deposits of intercellular collagen which disrupt tissue geometry [30]. Davies and Pomerance noted this association but also observed that those patients who had a greater duration of atria1 fibrillation had a greater amount of atria1 fibrosis [41]. They hypothesized that fibrillation induced fibrosis [41]. These observations are consistent with our study which has shown fibrosis in the left atria1 appendage of patients with rheumatic heart disease. Fibrosis was found in greatest concentration in the group with atria1 fibrillation and least in those patients in sinus rhythm. Although fibrillation may play some role in the genesis of fibrosis, the presence of fibrosis causes an inhomogeneity of electrical conduction which may lead to irreversible atria1 fibrillation [30].
150
Con&ding Observations This study is in basic agreement with the Noble and Fisch hypothesis of the pathogenesis of atria1 fibrillation. One modification, however, is that the term “cellular atrophy” be replaced. Our study suggests that there may be cellular drop-out with replacement fibrosis but that the remaining cells become hypertrophic. Ten Pick et al. [32] and Hordof et al. [33] have added the concept of hypopolarization of diseased atrial cells which may be related to the depressed levels of cyclic GMP found in our study. The least understood concept is the role of metabolic factors in the pathogenesis of atria1 fibrillation. Further studies of the cyclic nucleotides, adenine nucleotides, calcium metabolism and others are needed to complete these concepts.
Acknowledgements The authors would like to thank Marion Lewis for her assistance in the handling of the tissue specimens, Pam Newton for technical assistance, Dr. James Kilman and Dr. Jack Vasko for supplying atrial appendage tissue and Tami Smith for the preparation of the manuscript.
References 1 SeLzer A, Cohn KE. Natural history of mitral stenosis: a review. Circulation 1972;45:878-890. 2 Seizer A. Effects of atrial fibrillation upon the circulation in patients with mitral stenosis. Am Heart J 1960;59:518-526. 3 Mitchell JH, Shapiro W. Atrial function and the hemodynamic consequences of atria1 fibrillation in man. Am J Cardiol 1969;23:556-566. 4 Noble RJ, Fisch C. Factors in the genesis of atria1 fibrillation in rheumatic valvular disease. Cardiovasc Chn 1972;5:97-114. 5 Podzuweit T, Dalby AJ, Cherry GW, Opie LH. Cyclic AMP levels in ischaemic and non-ischaemic myocardium following coronary artery ligation: relation to ventricular fibrillation. J Mol Cell Cardiol 1978;10:81-94. 6 Sugiura M, Ogawa K, Yamazaki N. Concentration of myocardial cyclic AMP and ventricular fibrillation induced by aminophylline. Jap Heart J 1979;20:177-182. 7 Podzuweit T. Catecholamine-cyclic-AMP-Ca*+-induced ventricular tachycardia in the intact pig heart. Basic Res Cardiol 1980;75:772-779. 8 Cobb PB, Witkowski FX, Sobel BE. Mechanisms contributing to malignant dysrhythmias induced by ischemia in the rat. J Clin Invest 1979;61:109-119. 9 Lubbe WF, McFadyen ML, Muller CA, Worthington M, Opie LH. Protective action of amiodarone against ventricular fibrillation in the isolated perfused rat heart. Am J Cardiol 1979;43:533-540. 10 Sahn DJ, DeMaria A, Kiss10 J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072-1083. 11 Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac valves, and central circulatory shunts. Am Heart J 1951;41:42-45. 12 Steiner AL, Parker CW, Kipnis DM. A radioimmunoassay for measurement of cyclic nucleotides. J Biol Chem 1972;247:1106-1113. 13 Miyadhi M, Mizudhi N, Sebo R. Increased sensitivity of the cyclic nucleotide radioimmunoassay. Ann Biochem 1977:77:429-435.
151
14 Unverferth DV, Fertel RH, Altschuld R, Magorien RD, Lewis RP, Leier CV. Biochemical measurements of endomyocardial biopsies. Cath Cardiovasc Diag 1981;7:55-64. 15 Sheehan DC, Hrapchak BB. Connective tissue and muscle fiber stains. In: Sneehan OC, Hropchak BB, eds. Theory and practice of histotechnology. St. Louis: CV Mosby Company, 1973. 16 Langman J. Cardiovascular system. In: Langman J, ed. Medical embryology. Baltimore: Williams & Wilkins, 1975. 17 Thomas TV. Left atria1 appendage and valve replacement. Am Heart J 1972;84:838-839. 18 Quintiliani R, Klimek J, Nightingale CH. Penetration of cephapirin and caphalothin into the right atria1 appendage and pericardial fluid of patients undergoing open-heart surgery. J Infect Dis 1979;139:348-352. 19 Martin P. Atria1 inotropic responses to brief vagal stimuli: frequency-force interactions. Am J Physiol 1980;239:H333-H341. 20 Green ER, Subramanian S, Faden H, Quintiliani R, Nightingale CH. A comparison of the penetration characteristics of cephapirin and caphalothin into right atria1 appendage, muscle, fat, and pericardial fluid of pediatric patients undergoing open-heart operation. Ann Thorac Surg 1981;31:155-160. 21 Wooley CF, Baba N, Kilman JW, Ryan JM. Thrombotic calcific mitral stenosis. Morphology of the calcific mitral valve. Circulation 1974;49:1167-1174. 22 Harvey RM, Ferrer MI, Samet P, et al. Mechanical and myocardial factors in rheumatic heart disease with mitral stenosis. Circulation 1955;11:531-551. 23 Legato MJ, Bull MB, Ferrer MI. Atria1 ultrastructure in patients with fixed intra-atria1 block. Chest 1974;65:252-261. 24 Legato MJ. Ultrastructure of the atria], ventricular and purkinje cell with special reference to the genesis of arrhythmias. Circulation 1973;47:178-189. 25 Strauer BE, Tauchert M. Evidence for an inefficient energy utilization in cardiac hypertrophy. Studies on the isolated human ventricular myocardium. Klin Wochenschr 1973;51:322-326. 26 Shiverick KT, Hamrell BB, Alpert NR. Structural and functional properties of myosin associated with the compensatory cardiac hypertrophy in the rabbit. J Mol Cell Cardiol 1976;8:837-851. 27 Wikman-Coffelt J, Walsh R, Fenner C, Kamiyama T, Sale1 A, Mason DT. Effects of severe hemodynamic pressure overload on the properties of canine left ventricular myosin: mechanism by which myosin ATPase activity is lowered during chronic increased hemodynamic stress. J Mol Cell Cardiol 1976;8:263-270. 28 Wikman-Coffelt J, Fenner C, Coffelt RJ, Sale1 A, Kamiyama T, Mason DT. Chronological effects of mild pressure overload on myosin ATPase activity in the canine right ventricle. J Mol Cell Cardiol 1975;7:219-224. 29 Goldstein MA, Sordahl LA, Schwartz A. Ultrastructural analysis of left ventricular hypertrophy in rabbits. J Mol Cell Cardiol 1974;6:265-273. 30 Bailey GWH, Braniff BA, Hancock W, Cohn KE. Relation of left atria1 pathology to atria1 fibrillation in mitral valvular disease. Ann Intern Med 1968;69:13-20. 31 Henry WL, Morganroth J, Pearlman AS, et al. Relation between echocardiographically determined left atria1 size and atria1 fibrillation. Circulation 1975;53:273-279. 32 Probst P, Goldschlager N, Selzer A. Left atrial size and atria1 fibrillation in mitral stenosis. Factors influencing their relationship. Circulation 1973;48:1282-1287. 33 Hordof AJ, Edie R, Malm JR, Hoffman BF, Rosen MR. Electrophysiologic properties and response to pharmacologic agents of fibers from diseased human atria. Circulation 1976;54:774-779. 34 Ten Eick RE, Singer DH. Electrophysiological properties of diseased human atrium. I. Low diastolic potential and altered cellular response to potassium. Circ Res 1979;44:545-557. 35 Wit AL, Rosen MR, Hoffman BF. Relationship of normal and abnormal electrical activity of cardiac fibers to the genesis of arrhythmias. Re-entry. Am Heart J 1974;88:664-670. 36 Hicks MJ, Shigekawa M, Katz AM. Mechanism by which CAMP dependent protein kinase stimulates calcium transport in cardiac sarcoplasmic reticulum. Circ Res 1979;44:384-391. 37 Brooker G. Oscillation of the CAMP concentration during the myocardial contraction cycle. Science 1973;182:933-934. 38 Katsuki S, Arnold WP, Murad F. Effects of sodium nitroprusside, nitroglycerin, and sodium azide on levels of cyclic nucleotides and mechanical activity of various tissues. J Cycl Nucl Res 1977;3:239-247.
152
39 Kohlhardt M, Haap K. 8-Guanosine 3’5’ monophosphate mimics the effect of acetylcholine on slow response action potential and contractile force in mammalian atria1 myocardium. J Mel Cell Cardiol 1978;10:573-586. 40 Ten Eick R, Nawrath H, McDonald TF, Trautwein W. On the mechanism of the negative inotropic effect of acetylcholine. Eur J Physiol 1976;361:207-213. 41 Davies MJ, Pomerance A. Pathology of atria1 fibrillation in man. Br Heart J 1972;34:520-525.