|
|
||||||||
J Thorac Cardiovasc Surg 2003;126:1026-1032
© 2003 The American Association for Thoracic Surgery
Surgery for congenital heart disease |
a Department of Pediatric Cardiology and Adult Congenital Heart Disease, Heart Institute (InCor), University of São Paulo Medical School, São Paulo, Brazil
b Laboratory of Pathology, Heart Institute (InCor), University of São Paulo Medical School, São Paulo, Brazil
Received for publication September 23, 2002; revisions received November 13, 2002; accepted for publication December 2, 2002.
* Address for reprints: Maria Angelica Binotto, MD, Heart Institute (InCor), University of São Paulo Medical School, Pediatric Cardiology, Av. Higienopolis, 1048/86, São Paulo 01238-000, Brazil
conangelica{at}incor.usp.br
| Abstract |
|---|
|
|
|---|
METHODS: We examined 32 hearts with tricuspid atresia and 27 normal hearts from control subjects with similar age and sex distribution. Wall thickness, inlet length, and outlet length were obtained from the left ventricle. Immunohistochemical staining for von Willebrand factor was used to label myocardial capillaries. By means of computer-assisted morphometry, the following data were obtained from the inlet, apex, and outlet of the left ventricle: transverse myocyte diameter, myocyte nuclear volume fraction, capillary volume fraction, interstitial fibrous volume fraction, and endocardial thickness.
RESULTS: The wall thickness in affected hearts was not different from that in control hearts. The left ventricular outlet length was significantly greater in the malformed hearts (P = .005). The myocyte diameter did not differ from that in control hearts. The capillary volume fraction was decreased in the malformed hearts (P < .001). The interstitial fibrous deposition was greater in the malformed hearts at all sites analyzed (P < .001). Fibrosis was greater in the inlet and apex (P = .004) and also in the subendocardial half of the ventricular wall than in the subepicardial half. According to a logistic regression model, age was the only variable associated with the probability of occurrence of fibrosis.
CONCLUSIONS: Our findings in hearts with tricuspid atresia possibly represent chronically induced volume overload in the presence of ischemia, rather than a typical model of volume overload. A decreased capillary volume fraction may indicate a greater susceptibility to ischemia. Fibrous deposition probably occurs early in life.
Few studies have addressed ventricular remodeling in hearts with univentricular atrioventricular connections.4-6 The purpose of this study was to analyze several aspects of the ventricular remodeling in hearts with tricuspid atresia, including myocardial hypertrophy, extracellular matrix deposition, and capillary network.
| Methods |
|---|
|
|
|---|
Morphometric analysis
Myocyte transverse diameter
The morphometric measurements were carried out with an interactive computer-assisted image analyzer (Leica Quantimet; Leica Cambridge Ltd, United Kingdom).7 To avoid interobserver variation, a single investigator operated the analyzer.
The myocyte transverse diameter was measured at the level of the nucleus in the mesocardial layer of the ventricular wall, irrespective of the orientation of the section. Final magnification was x400.
In a pilot study, the evolution of the mean values and variance was analyzed to determine the number of measurements required. The mean values of myocyte diameter in 20, 40, 60, 80, 100, 120, 140, 180, and 200 observations were obtained. We opted to measure 60 myocytes per section.
Capillary volume fraction
Immunohistochemical staining for von Willebrand factor was used to label myocardial capillary endothelial cells. The stained inlet LV sections containing circumferentially oriented capillaries were imaged at a final magnification of x1000. For each section 30 random fields 11,456 µm2 in area were analyzed. With computer-assisted morphometry and an interposed grid of 598 points, incident points on myocardial capillaries were counted. The proportion of the points lying over the structure of interest over the number of points in the grid was used as a measure of the volume fraction, or volume density, of myocardium occupied by that structure.8
Nuclear volume fraction
With the same system described previously and an interposed grid of 598 points, incident points on myocyte nuclei were counted in sections of the LV inlet stained with hematoxylin and eosin. For each section 25 random fields were analyzed at the mesocardial layer of the ventricular wall.
Quantification of fibrous tissue
Fibrous content in the interstitial space was estimated by analyzing at least 10 microscopic fields in the subendocardial and subepicardial halves of the ventricular wall at a magnification of x400 (final resolution of 0.48 µm/pixel). Star-shaped scars and concentric perivascular fibrous tissue were excluded from the analysis. The interstitial fibrous volume fraction (IFVF) was expressed as a percentage of the myocardial area. Microscars were described qualitatively.
Endocardial thickness
Linear measurements of the endocardial thickness were obtained in histologic sections stained with hematoxylin-eosin from the inlet, apex, and outlet of the LV. A mean value was obtained from 10 measurements at each site.
Statistical analysis
Values are given as means, SDs, and medians. Unpaired Student t tests or Mann-Whitney tests as appropriate were used to compare values between control and study groups. Comparisons between subendocardial and subepicardial regions were done with the Wilcoxon test. The Friedman test was carried out for multiple comparisons, followed by the Tukey procedure as modified by Nemenyi. Correlation coefficients were obtained by the Spearman method. Logistic regression was used to identify independent predictors of the occurrence of fibrosis. Data were analyzed with SigmaStat 2.0 software (SPSS Inc, Chicago, Ill).
| Results |
|---|
|
|
|---|
|
|
|
Interstitial fibrosis
The IFVF was significantly greater in the malformed hearts at all three sites examined (Table 3 and Figure 1,
A). The LV inlet and apex were more fibrotic than the outlet in the malformed hearts but not in the control group (P = .004 by Friedman test). In addition, the IFVF was greater in the subendocardial halves of the ventricular wall in the inlet and apex of hearts with tricuspid atresia but not the control hearts (P = .002 and P = .01, respectively, by Wilcoxon test Figure 2).
|
|
The multivariate logistic regression model suggested that age was the only variable associated with an increased probability of occurrence of fibrosis, defined as IFVF greater than 3%. The other variables included in the model (type of ventriculoarterial connection, the presence of a restrictive VSD, presence of aortic coarctation or interruption, decreased pulmonary blood flow physiology, and previous surgery) were not independent predictors of occurrence of fibrosis in the multivariate model. In fact, children older than 3 months had an 8.3 times greater probability of occurrence of fibrosis than had a younger population (odds ratio 95% confidence interval 1.47-47.22, sensitivity 71.4%, specificity 76.9%; Figure 3).
|
Endocardial thickness
LV endocardial thickness was greater at the inlet and apex of the malformed hearts than in the control hearts (Table 3).
Type of ventriculoarterial connection
The age at death was similar between the two groups (concordant 3 days to 34 months, median 3.0 months; discordant 6 days to 50 months, median 2.3 months). There was no difference when comparisons were made between hearts with concordant and discordant ventriculoarterial connections to address the following variables: inlet wall thickness (P = .09 by t test), inlet length (P = .5 by t test), transverse myocyte diameter (P = .24 by t test); IFVF (P = .19 by Mann-Whitney test), and endocardial thickness (P = .76 by t test).
| Discussion |
|---|
|
|
|---|
LV hypertrophy
In our study the LVs of the malformed hearts were dilated, without a relative increase in the wall thickness corresponding to eccentric hypertrophy, as would be expected in situations where there is predominantly volume overload. These findings are in accordance with those described by others6 in hearts with pulmonary atresia with intact ventricular septum, suggesting that the ratio of LV mass to end-diastolic volume may be inadequate in univentricular hearts. We did not find specimens in our material with excessive wall thickness, as demonstrated in an autopsy series of patients who died after the Fontan procedure9 and also in a clinical setting.10 This diversity in results is likely to be related to the young age of death of the patients included in our study, most younger than 1 year, and the relative paucity of cases with discordant ventriculoarterial connection in association with a restrictive VSD and subaortic stenosis, in which setting conspicuous ventricular hypertrophy is more frequent.
During the adaptive phase of elevated diastolic wall stress in volume overloads, the ratio of wall thickness to chamber radius remains constant as a result of proportional growth in the transverse and longitudinal dimensions of the myocyte. After the progression toward decompensation, there is a disproportionate lengthening of the myocytes and a dilated, thin-walled ventricular chamber.11 In this study the myocyte transverse diameter in the affected hearts did not differ from that in the control hearts. Although the longitudinal dimension could not be obtained, it is likely that these patients were in the maladaptive phase of eccentric hypertrophy before death.
Experimental data have indicated that the myocyte nuclear volume increases with cellular hypertrophy.12 In addition, previous studies have demonstrated that human myocytes are prevalently mononucleated.13 The myocyte nuclear volume fraction could therefore be viewed as an index of hypertrophy and of cell number in a given area of the histologic section. The decreased nuclear volume fraction found in hearts with tricuspid atresia relative to normal control hearts is in agreement with a predominantly longitudinal growth of the myocytes (eccentric hypertrophy). However, we could not establish a parallel to the hypertrophic process, either because one does not exist or because of the relatively small number of cases analyzed.
Coronary capillaries
The myocardial capillary network expands proportionally during physiologic growth14 but may be inadequate under conditions of hemodynamic overload, increasing the potential for ischemic damage in cardiac hypertrophy. Evidence from animal and human studies indicates that capillary growth in pressure overload hypertrophy does not fully compensate for the increase in ventricular mass,15 although limited capillary growth may occur in young patients16 and during adaptation to a prolonged pressure overload.17 In contrast, several experimental studies in volume overload models suggest that the adaptation of the myocardial capillary network is proportional to the degree of hypertrophy.18 Experimental data on the myocardial capillary supply adaptation in response to hypoxia are controversial.15 Yet hypoxia is considered to be an effective stimulus for growth of endothelial cells.15 The finding of a decreased capillary volume fraction in our cases might indicate a greater vulnerability of these hearts to ischemia.
Extracellular matrix
Hearts with tricuspid atresia showed a higher proportion of fibrous tissue than normal since the first weeks of life, showing a positive correlation with increasing age. The multivariate logistic regression model indicated that infants older than 3 months had an 8.3 times greater probability of occurrence of fibrosis than had younger infants, supporting the view that the myocardium may be inherently abnormal.4,5
It has been well documented that pressure overload cardiac hypertrophy is accompanied by an increased accumulation of collagen in the myocardium.19,20 Fibrosis of the heart accompanies chronic elevations in circulating effector hormones of the renin-angiotensin-aldosterone system, when heart failure or unilateral renal artery stenosis impairs renal perfusion.20 In contrast, most of the published experimental data have indicated that interstitial collagen accumulation does not change in response to induced volume overload.21,22
Much attention is currently focused on the role of apoptosis in the pathogenesis of cardiac disease. Apoptosis has been implicated in a wide range of models, such as ischemia and myocardial infarction, pressure overload hypertrophy, cardiac rejection, and dilated cardiomyopathy.23 It may be a contributing phenomenon in the pathologic remodeling in hearts with tricuspid atresia.
The morphometric method permitted the discrimination of the fibrous deposition across the ventricular wall. Hoyt and colleagues7 compared the quantification of collagen in histologic sections stained with Masson trichrome, performed with a computer-assisted method similar to ours, with those obtained from a hydroxyproline assay. The percentage of collagen from analysis of histologic sections strongly correlated with the hydroxyproline level, with excellent intraobserver and interobserver concordance. Results were unaffected by variations in histochemical staining.
Our study showed a greater deposition of fibrous tissue in the subendocardial half of the LV wall, following the pattern found in concentric hypertrophy,24 dilated cardiomyopathy,25 and ventricular remodeling after myocardial infarction.26 This finding is not in agreement with observations by Ho and colleagues4 in hearts with tricuspid atresia. Experimental studies of physiologic growth have demonstrated a relatively lower capillary luminal volume in the subendocardial layer, indicating a greater vulnerability of this region to ischemic injury.27 In addition, the myocardial tissue pressure and oxygen demands are greater in the subendocardial portion of the wall. Thus the disturbance of oxygen supply and demand may result in chronic tissue injury and subendocardial fibrosis.
This analysis demonstrated regional variations in the fibrous deposition, with the LV inlet and apex being more fibrotic than the outlet in malformed hearts. This finding could explain, at least in part, regional ventricular wall motion abnormalities found in patients with univentricular hearts before and after stages of Fontan reconstruction.28,29
The comparison according to the type of ventriculoarterial connections did not show any significant difference in the content of fibrous tissue. This finding is in conformity with data published by Ho and colleagues,4 possibly reflecting the dominant role of the LV.
The functional consequences of collagen remodeling have been extensively demonstrated.30 It seems clear that both systolic and diastolic properties may be affected. Clinical trials with pharmacologic interventions directed toward collagen remodeling in such malformed hearts should be addressed in future studies.
Conclusion
On the whole, our findings in hearts with tricuspid atresia are more likely to represent a chronically induced volume overload in the presence of ischemia than to represent a typical model of volume overload. Moreover, the pathologic insult occurs in parallel with the expansion of cardiac mass in a growing child. What remains to be clarified is why the capillary network seems not to develop properly in the presence of a strong potential stimulus for angiogenesis (hypoxia).
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Tanoue, H. Kado, T. Maeda, Y. Shiokawa, N. Fusazaki, and S. Ishikawa Left ventricular performance of pulmonary atresia with intact ventricular septum after right heart bypass surgery J. Thorac. Cardiovasc. Surg., November 1, 2004; 128(5): 710 - 717. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |