CLINICAL RESEARCH
Europace (2013) 15, 1557–1561
doi:10.1093/europace/eut099
Atrial fibrillation
Morphological and mechanical examination
of the atrial ‘intima’
David Schwartzman 1*, Karen Schoedel 2, Donna Beer Stolz 3, and Elena Di Martino 4
1
Heart and Vascular Institute, University of Pittsburgh, UPMC Presbyterian, B535, Pittsburgh, PA 15213 2582, USA; 2Department of Pathology, University of Pittsburgh, Pittsburgh, PA,
USA; 3Department of Cell Biology, University of Pittsburgh, Pittsburgh, PA, USA; and 4Department of Civil Engineering and Centre for Bioengineering Research and Education, University
of Calgary, Calgary, AB, Canada
Received 2 March 2013; accepted after revision 23 March 2013; online publish-ahead-of-print 21 April 2013
Aims
To examine morphology and mechanical properties of the atrial ‘intima’, which we defined as the tissue interposed
between atrial endocardium and myocardium, in patients without known cardiovascular disease.
.....................................................................................................................................................................................
Methods
Post-mortem right and left atrial tissue was obtained from male infants (,1 year, n ¼ 4), children (10 –19 years, n ¼ 4),
and results
and adults (58–69 years, n ¼ 7). Using light microscopy and an ocular micrometer, atrial intimal (AIT) thickness was measured. Intimal collagen bundle thickness was measured using electron microscopy. Passive atrial wall stiffness was measured using a planar biaxial testing device. Among infants, left AIT (0.2 + 0.2 mm) and right (0.2 + 0.1 mm) AIT were not
significantly different (P ¼ 0.84). Among children, left AIT (0.6 + 0.2 mm) was significantly greater than right
(0.2 + 0.1 mm) AIT (P ¼ 0.03), and left AIT was marginally greater than in infants (P ¼ 0.07). Among adults, with the exception of the appendage region, left AIT (1.0 + 0.2 mm) was markedly greater than right AIT (0.3 + 0.1 mm; P , 0.05),
and left AIT was significantly greater than that in other age groups (P , 0.05). There were no differences in right AIT
among age groups. Left intimal collagen bundle thickness was greater in adults (0.0512 + 0.0056 mm) than infants
(0.0432 + 0.0071 mm) or children (0.0435 + 0.0013 mm), and bundles were less organized. Wall stiffness was attributable primarily to the intima (1245 + 132, vs. 260 + 45 N/m2 for the remaining atrial wall).
.....................................................................................................................................................................................
Conclusion
The left atrial intima, but not the right, thickens with age, becomes more disorganized ultrastructurally, and is responsible
for the majority of atrial wall stiffness.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atrium † Atrial fibrillation † Ablation † Ageing † Intima † Artery † Vascular
Introduction
The intima of a blood vessel is the tissue interposed between its endothelium and the muscular layer.1 Significant differences in intimal
morphology are apparent between arteries and veins, and in association with ageing. Prominent associations with ageing intimal thickening, ultrastructural disorganization, and stiffening.2 The atria too have
an ‘intima’ of sorts, and data addressing morphology and mechanical
properties of this tissue, including differences between right (RA) and
left atria (LA) and the impact of ageing, are scarce.
Methods
This analysis was sanctioned by the Institutional Review Board as well as
the Committee for Research Involving the Dead at the University of
Pittsburgh.
Subjects
Cardiac tissue was obtained at autopsy from male patients who had
died of non-cardiovascular causes (trauma or malignancy) within
24 h of the accession. No patient had known antemortem cardiac
or blood vessel disease, including hypertension, coronary or peripheral atherosclerosis, dyslipidaemia, arrhythmia, diabetes mellitus, or
abnormal cardiac structure/function. No patient was receiving medicines with prominent cardiac/vascular actions. Tissues were not
placed in formalin prior to examination. Three patient groups were
prospectively sought:
† Group 1: n ¼ 4, ages at death 1 day, 2 days, 22 days, and 11 months. No
cardiac or vascular pathology apparent during autopsy. This group is
hereinafter termed ‘Infants’.
† Group 2: n ¼ 4, ages at death 10, 15, 16, and 19 years. No cardiac or
vascular pathology apparent during autopsy. This group is hereinafter
termed ‘Children’.
* Corresponding author. Tel: +1 412 647 2762; fax: +1 412 647 7979, E-mail address: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2013. For permissions please email: [email protected].
1558
D. Schwartzman et al.
What’s new?
† The morphological examination of atrial intima was detailed
first.
† In adults, left atrial intimal thickness was markedly greater than
right.
† Left, but not right, intimal thickness increases with age.
† The intima is responsible for the majority of atrial wall mechanical stiffness.
† Group 3: n ¼ 7, ages at death 58– 69 years. Autopsy findings were remarkable for mild concentric left ventricular hypertrophy in one
subject and moderate coronary atherosclerosis in two subjects. This
group is hereinafter termed ‘Adults’.
Histological analysis
Specimen acquisition
Gross inspection of the cardiac chambers was performed to exclude
obvious pathology. Transmural sections of fresh cardiac tissue, designated as ‘specimens’, were taken. Specimens from Infants and Children
were limited to the atrial septum encompassing both RA and LA endocardiums (below); this limitation was by the request of the Director of Paediatric Pathology, so as to preserve the hearts for other teaching purposes.
Specimens from Adults were from a variety of cardiac sites, summarized
in Figure 1.
Light microscopy
In acquiring specimens, care was taken to preserve a perpendicular orientation to the endocardium. Specimens were formalin-fixed and embedded in paraffin blocks. Five micrometre thick sections of tissue were
mounted onto slides. Separate slides from each block were stained
with Haematoxylin and Eosin, Trichrome, and Verhoeff– Van Gieson.
As has been described previously, the atrial wall interposed between cavitary blood and the myocardial compartment is comprised of several
layers, in order of proximity to blood: (i) endothelial, comprised of a
single row of cells adjacent to blood, (ii) subendothelial, comprised of a
thin layer of delicate, irregular collagen fibrils, (iii) fibrillar, comprised of
regular elastic and collagen fibres, (iv) smooth muscle, which is sparse
and discontinuous, and(v) subendocardial, consisting of coarse elastic
and collagen fibrils, as well as capillaries (Figure 2). In actuality, the term
‘layers’ is a misnomer, because these are generally not sharply demarcated.3 Given this and the purposes of this study, we combine these
layers and hereinafter term the structure the ‘atrial intima’(AI).
An ocular micrometre (Olympus) was used to measure atrial intimal
thickness (AIT; Figure 2). Measurements were performed at 10 separate
sites on each slide, sampling the entire specimen, and the results averaged
across all stains. At each site of measurement of AIT, myocardial compartment thickness (MCT) was also measured by the same means (Figure 2).
Similar measurements were made on left ventricular (LV) specimens.
The degree of interstitial fibrosis within the myocardial compartment
was assayed semiquantitatively using the trichrome slides, according to
the following grading scheme: grade 1 ¼ none-mild; grade 2 ¼ moderate; grade 3 ¼ severe.
Electron microscopy
Atrial specimens from three patients were studied: one infant (aged 2
days), one child (aged 16 years), and one adult (aged 64 years). Specimens
which were previously paraffin-embedded were rehydrated using
Figure 1 Schematic of regions from which specimens were
obtained for analysis of adult group subjects. In the RA, Region 1
was free wall (trabeculated topography), and 2/3 posterior/medial
walls (smooth topographies). In the LA, Regions 4 – 7 were atrial
body and 9/10 superior pulmonary venoatrial junctions (smooth
topographies), and Region 8 proximal appendage (trabeculated
topography). Specimens were also obtained from the left ventricular free wall contiguous to the LA. Specimens from infants and children were from the interatrial septum only (Regions 3 and 4). As
shown in the inset, this region was located anterior to the fossa
ovalis, which is a space filled with fat (F), permitting separate assessment of RA, and LA walls from this single section.
Figure 2 Photomicrograph of a typical LA specimen (trichrome
stain, ×40 magnification) illustrating individual ‘layers’ and their
summation as the AI. AIT, atrial intimal thickness; MCT, atrial myocardial compartment thickness (incomplete in this image).
decreasing concentrations of ethanol at 15 min intervals, washed in
phosphate-buffered saline (PBS) three times, post-fixed in 1% osmium
tetroxide, 1% Fe6CN3 for 1 h, washed three times in PBS, then dehydrated through a 30 – 100% ethanol series using several changes of
1559
Morphological and mechanical examination of the atrial ‘intima’
Polybed 812 embedding resin (Polysciences). Tissue was embedded in by
inverting Polybed 812-filled BEEM capsules on top of the cells. Blocks
were cured overnight at 378C, then for two additional days at 658C.
Tissue was pulled off the slides and ultrathin sections (60 nm) were
obtained on a microtome and post-stained in 4% uranyl acetate for
10 min and 1% lead citrate for 7 min. Sections were viewed on a transmission electron microscope (JEOL, JEM 1210) at 80 kV equipped with an
charge-coupled device camera (Advanced Microscopy Techniques).
Collagen fibril diameters in the fibrillar layer of the intima (Figure 2)
were measured at .20 sites across the specimen, and the results averaged.
Analytical methods
Data are presented as mean + standard deviation, unless otherwise
stated. Comparison of continuous variables was performed utilizing a
Mann– Whitney rank-sum test, as appropriate. Comparisons of categorical variables were performed utilizing a x2 test. Correlation analysis was
performed utilizing the Pearson product-moment method. For each statistical test, a P , 0.05 was considered to be significant.
Results
Histological analysis
Atrial wall passive stiffness analysis
In each Adult patient, specimens from LA regions 5, 6, and 7 (Figure 1)
were tested. Using blunt technique, the LA intima was readily peeled
from the remaining atrial wall (Figure 3). In all specimens undergoing
mechanical analysis, this dissection plane was confirmed microscopically.
For each specimen, mechanical analysis of the intima and the remaining
atrial wall (comprised mainly of the myocardial compartment plus adventitium) samples were performed separately. The technique used for
mechanical analysis has been detailed previously.4 In brief, each tissue
sample was cut into 1 cm2, placed in Tyrode’s solution (378C), and
mounted in the device using suture line pairs attached to the tissue
edge on all four sides. Four graphite markers were glued to the surface
of the sample, forming a 5 × 5 mm2 matrix at its centre. Once constrained in the test fixture, planar biaxial tension was applied to the
sample via the suture lines. An initial small tension was applied to
ensure that there was no tissue buckling. In-plane deformation of the
sample, as assayed by marker displacement, was tracked using a digital
camera (resolution 0.02 mm/pixel). The testing apparatus applied a
tension of 30 N/m (corresponding to that realized under physiological
conditions in the LA), with an interval of 20 s allowed to reach peak
tension. The edges of the sample were permitted to expand freely, resulting in a uniform strain state in the central (intermarker) region. Each test
consisted of 10 contiguous cycles to allow for mechanical preconditioning, with data from the tenth cycle used for analysis. Tissue stiffness was
measured using the tangent modulus of elasticity (TME, Pascal [Pa]).4,5
Light microscopy
Among infants, right (0.2 + 0.1 mm) and left (0.2 + 0.2 mm) AIT
were similar (P ¼ 0.84). Among children, right AIT (0.2 + 0.1 mm)
was significantly less than left AIT (0.6 + 0.2 mm; P ¼ 0.03).
Differences between infants and children in right (P ¼ 0.2) and left
(P ¼ 0.07) AIT were of borderline significance. Adult data are summarized in Table 1: left AIT was significantly greater than right AIT
for all regions except the appendage. In the atrial regions common
to both groups (Regions 3 and 4 in Figure 1), differences in right
AIT between children and adults were not significant (P ¼ 0.3),
whereas left AIT was significantly greater (Figure 4).
In adults, the within-specimen correlation between AIT and MCT
was not significant (r ¼ 0.27; P ¼ 0.43). Myocardial compartment fibrosis grade in LA (2.5 + 0.3) and RA (2.3 + 0.5) specimens were
similar (P ¼ 0.68), and each significantly greater than in LV specimens
(1.1 + 0.1; all P , 0.05). Adult atrial myocardial compartment fibrosis grades were significantly greater than the corresponding atrial
regions in infants (LA 1.1 + 0.3; RA 1.2 + 0.2, both P , 0.01) and
children (LA 1.5 + 0.4; RA 1.4 + 0.4; both P , 0.05).
Electron microscopy
Data from these subjects are summarized in Table 2. Right atrium
bundle diameters were similar among subjects. In the infant and
child, RA and LA diameters were similar, whereas in the adult LA
Table 1 Adult intimal thickness and MCT
Region (Figure 1)
IT
MCT
................................................................................
Figure 3 Gross specimens from LA, showing rightward (A) and
leftward (B)-facing vantages: (A) the AI is partially peeled from the
remaining atrial wall (W). rs, right superior venoatrial junction
region; ri, right inferior venoatrial junction region. (B) The AI has
been peeled and completely removed from the remaining atrial
wall. Ls, left superior venoatrial junction; li, left inferior venoatrial
junction.
1 (RA)
0.3 + 0.1
0.9 + 0.8
2 (RA)
3 (RA)
0.3 + 0.2
0.3 + 0.1
2.9 + 0.8
2.7 + 1.0
4 (LA)
1.0 + 0.2a,b,c
3.1 + 1.3
5 (LA)
6 (LA)
0.8 + 0.2a
0.9 + 0.3a
3.2 + 1.4
3.1 + 1.5
7 (LA)
1.2 + 0.9a
4.6 + 1.2
8 (LAA)
9 (PV)
0.3 + 0.1
0.7 + 0.1a
1.1 + 1.4
2.0 + 0.8
10 (PV)
0.8 + 0.2a
2.4 + 0.5
0.04 + .05a
18 + 3
LV
Measurements are in millimetres.
a
P , 0.05 vs. Region 3 AIT.
b
P , 0.05 vs. Region 4 AIT in infants.
c
P , 0.05 vs. Region 4 AIT in children.
1560
D. Schwartzman et al.
Figure 4 Representative adult atrial photomicrographs (Haematoxylin and Eosin stain) demonstrating difference between left (LA,
left) and right (RA, right) intimal (AI) thicknesses. MC, myocardial
compartment.
Table 2 Intimal collagen fibril diameters
RA diameters
Figure 5 Representative LA electron microscopic images of the
intima (fibrillar layer) from members of adult (left) and children
(right) groups. Diameter measurements are shown for selected
bundles. In addition to the larger diameter, the bundles in the
adult are individually and collectively less well organized.
LA diameters
................................................................................
Infant
0.0374 + 0.0058a
Child
Adult
a
0.0369 + 0.0052
0.0358 + 0.0079a
0.0432 + 0.0071a
0.0435 + 0.0013a,b
0.0512 + 0.0056
Measurements are in micrometres.
a
P , 0.05 vs. adult LA.
b
P , 0.05 vs. child RA.
bundle diameter was significantly greater. In association with the
increased diameter was the subjective appearance of disorganization,
both within and between bundles (Figure 5).
Atrial wall passive stiffness analysis
From each member of the adult group, LA specimens from Regions
5, 6, and 7 were assayed. We did not observe significant inter-regional
or inter-patient differences, so the data for all specimens were combined. The TME of the intima alone (1245 + 132 kPa) was markedly
greater than that of the remaining atrial wall (260 + 45 kPa;
P , 0.001).
Discussion
Among tissues obtained from patients without known cardiovascular
disease, AIT increased with age in the left but not the RA. In adults, the
thickening was diffuse (excepting the appendage), and the magnitude
was not correlated with contiguous MCT. There was associated enlargement and disorganization of compositional collagen bundles.
The intima was the major contributor to LA wall stiffness.
Intimal ageing of arteries, a process which is distinct from atherosclerosis, is conserved in evolution.6 With advancing age, arteries
dilate and stiffen in association with progressive intimal thickening
and ultrastructural disorganization. To our knowledge, the present
report offers the first detailed characterization of atrial intimal
morphology. Age-related changes in the left, but not right, atrium
were similar to those observed in arteries. This is consistent with vascular ageing, wherein the intima of veins is less affected than that of
arteries. In addition, it parallels the gradual increase in atrial volume
and overall wall thickness with age, phenomena which are also
more prominent in LA than RA.7 Although considered a phenomenon of ‘healthy’ ageing, it is clear that processes occurring within
the intima predispose to disease throughout the arterial wall, possibly
mediated by molecular signalling.6 Although we have no data to substantiate it, we wonder whether a similar phenomenon might be operative in the atrium of certain patients, with intimal processes
facilitating disease in the contiguous myocardial compartment, yielding disease phenotypes such as atrial fibrillation.
The underlying cause of the observed differences in AIT between
RA and LA cannot be discerned from our data, but there are some
possibilities. First, these tissues have different embryologic origins,
which may, based on properties which are inherent or produce disparate responses to environmental stimuli, resulted in differences in
intimal ageing.8 Second, it is likely that RA and LA tissues age in different mechanical environments. We are not aware of prior reports
which have quantified the relative forces acting on RA and LA walls.
Accuracy would be difficult to achieve, given that forces attributable
to both pressure and flow act heterogeneously (regionally and transmurally) within geometrically complex chambers, the walls of which
are comprised of materials with variant properties, which are tethered to veins and ventricles, and experience complex morphological
variation (much of which is passive and due to mechanical actions of
the ventricles) during the cardiac mechanical cycle.9,10 As a highly
simplistic illustration, we calculated LA and RA wall stress utilizing
the average total wall thicknesses (AIT + MCT) and material properties measured in our adult specimens, assuming normal mean atrial
pressures (RA ¼ 4 mmHg, LA ¼ 12 mmHg) and identical chamber
shapes (sphere) and dimensions (25 mm radius). Wall stress in the
LA (358 kPa) was markedly greater than that in the RA (65 kPa).
1561
Morphological and mechanical examination of the atrial ‘intima’
This finding is consistent with that of Hunter et al.,11 who demonstrated a marked increase in calculated LA wall stress in association
with a pressure change of 10–20 mmHg. If mechanical environment
is an important determinant of intimal ageing, it might be reasonable
to view the intima as an ‘endoskeleton’ which accumulates thickness
over time in an effort to ameliorate stress. Mechanical explanations
may also underlie the relative lack of intimal thickening in the trabeculated left appendage region, given its role as an LA reservoir, as
well as the trivial intimal thickness observed in the LV, given the thickness of its muscular wall. Finally, it is possible that differences in the
chemical contents of pre- vs. post-pulmonary blood somehow influence intimal ageing.
Left atriaum intimal thickening may have important implications.
First, given its impact on wall stiffness, the intima likely plays an important role in the atrial contribution to age-associated diminishment
in left heart mechanical efficiency.12 Second, thickness of electrically
‘inert’ intimal tissue interposed between an endocardial mapping
electrode and the myocardial compartment might impact the information content of electrographic information which is recorded.
This may be of particular importance for signals generated by relatively small myocyte volumes, which have attracted recent interest
because of their relevance to perpetuation of atrial fibrillation and selection of sites for ablation. In view of the marked differences in
intimal thickness between RA and LA, a significant difference in electrogram information content for a given myocyte volume is conceivable. Third, the intima likely plays important physical roles during
endocardium-based ablation procedures. Desirable roles would
include prevention of perforation and protection from damage to
contiguous, non-cardiac structures. Undesirable roles would
include prevention of adequate heating/cooling of the myocardial
compartment, physical changes which would result in pulmonary
vein stenosis, and predisposition to thrombus formation. Finally, it
is possible that intimal thickening magnitude carries important prognostic information, akin to that contained in LA volume or contraction vigour.13 Measurement is likely feasible, at least in certain atrial
regions.14
Our data have important limitations. First, the tissues were derived
from patients without known cardiovascular disease (albeit two
adults with moderate coronary atherosclerosis and one with
concentric left ventricular hypertrophy noted at autopsy). This was
intentional, given our interest in understanding changes attributable
to ‘healthy’ ageing which, as previously noted, imply susceptibility
to disease. Inclusion of patients with manifest cardiovascular
disease might have been interesting, but findings would not have
served to substantiate the notion of intima as endoskeleton nor a relationship between the intima and myocardial compartment. Second,
few patients were studied. We base our conclusions as to
age-associated thickening of the LA intima on a single atrial region,
and ultrastructural changes on a single member of each age group.
Third, there is a large gap in age between children and adults, which
limits confidence in our suggestion that intimal thickening is a
gradual lifetime process. Fourth, we studied male patients only.
Gender clearly plays a key role in the blood vessel ageing process,
and is likely to affect the atrium as well.2 Finally, the tissue which
we examined was obtained post-mortem (albeit prior to formalin fixation), which could have affected the measurements performed
herein.
Conflict of interest: none declared.
References
1. Stary HC, Blankenhorn DH, Chandler AB, Glagov S, Insull W, Richardson M et al. A
definition of the intima of human arteries and of its atherosclerosis-prone regions. A
report from the Committee on Vascular Lesions of the Council on Arteriosclerosis,
American Heart Association. Arterioscler Thromb Vasc Biol 1992;12:120 – 34.
2. Lakatta EG. Age-associated cardiovascular changes in health: impact on cardiovascular disease in older persons. Heart Fail Rev 2002;7:29 –49.
3. Lanigan RA, Zaki SA. Ultrastructure of the normal atrial endocardium. Br Heart J
1966;28:785 –95.
4. Sacks MS. Biaxial mechanical evaluation of planar biological materials. J Elasticity 2000;
61:199–246.
5. Parker KJ, Gao L, Lerner RM, Levinson SF. Techniques for elastic imaging. IEEE Eng
Med Biol 1996;15:52–9.
6. Wang M, Monticone RE, Lakatta EG. Arterial aging: a journey into subclinical arterial
disease. Curr Opin Nephrol Hypertens 2010;19:201 –7.
7. Pan N-H, Tsao H-M, Chang N-C, Chen Y-J, Chen S-A. Age dilates atrium and pulmonary veins. Implications for genesis of atrial fibrillation. Chest 2008;133:190 –6.
8. Douglas YL, Jongbloed MRM, Gittenberger-de Groot AC, Evers D, Dion RAE,
Voigt P et al. Histology of vascular myocardial wall of left atrial body after pulmonary
venous incorporation. Am J Cardiol 2006;97:662 – 70.
9. Di Martino ES, Bellini C, Schwartzman D. In vivo porcine left atrial wall stress: computational model. J Biomech 2011;44:2589 –94.
10. Di Martino ES, Bellini C, Schwartzman D. In vivo porcine left atrial wall stress: effect
of ventricular tachypacing on spatial and temporal stress distribution. J Biomech 2011;
44:2755 –60.
11. Hunter RJ, Liu Y, Lu Y, Wang W, Schilling RJ. Left atrial wall stress distribution and its
relationship to electrophysiologic remodeling in persistent atrial fibrillation. Circ
Arrhythm Electrophysiol 2012;5:351 –60.
12. Spencer KT, Mor-Avi V, Gorcsan J, DeMaria AN, Kimball TR, Monaghan MJ et al.
Effects of aging on left atrial reservoir, conduit, and booster pump function: a multiinstitution acoustic quantification study. Heart 2001;85:272–2.
13. Gupta S, Matulevicius SA, Ayers CR, Berry JD, Patel PC, Markham DW et al. Left atrial
structure and function and clinical outcomes in the general population. Eur Heart J
2013;34:278 –85.
14. Lopez-Candales A, Grewal H, Katz W. The importance of increased interatrial septal
thickness in patients with atrial fibrillation: a transesophageal echocardiographic
study. Echocardiography 2005;22:408–14.