|
|
||||||||
Ann Thorac Surg 2004;77:36-40
© 2004 The Society of Thoracic Surgeons
a Department of Paediatrics, National Heart and Lung Institute, Imperial College and Royal Brompton and Harefield NHS Trust, London, England, UK
Accepted for publication July 29, 2003.
* Address reprint requests to Dr Ho, Department of Paediatrics, Faculty of Medicine, Imperial College, National Heart and Lung Institute, Dovehouse St, London SW3 6LY, England, UK
e-mail: yen.ho{at}imperial.ac.uk
| Abstract |
|---|
|
|
|---|
METHODS: We selected 23 specimens with hypoplastic left heart syndrome for anatomical study. Using a combination of morphometric analysis and scanning electron microscopy we analyzed the quantity and quality of the collagenous matrix. We compared the results with 16 age-matched controls.
RESULTS: Hearts with hypoplastic left heart syndrome have significantly less collagen matrix than normal. The right ventricle has more collagen than the left and there is significant transmural variation. There was no difference in the ratio of the two main collagen subtypes or in the quality of the matrix.
CONCLUSIONS: We believe this difference in fibrous matrix to be an inherent abnormality intrinsic to the malformation affecting not only the hypoplastic left but also the "normal" right ventricle. This in turn may have significant implications for the expected long-term outcome of reconstructive surgery.
| Introduction |
|---|
|
|
|---|
Despite these observations relatively little attention has been paid to the significance of the collagenous matrix in the clinical setting, particularly in relation to congenital heart disease [17, 18]. As suggested by a recent study on tricuspid atresia [19] an abnormal accumulation of fibrous tissue may be an inherent part of some congenital malformations. Thus far anatomical studies have focused mainly on right-side heart lesions [19, 20]. Because a common surgical repair of hypoplastic left heart syndrome (HLHS) is to convert the right ventricle into the sole pumping chamber, the aim of our study is to investigate quantitatively and qualitatively the collagen matrix found in these hearts for a better understanding of ventricular structure.
| Material and methods |
|---|
|
|
|---|
Morphometric analysis
For total collagen analysis three zones (subepicardial, midmyocardial, subendocardial) were assessed by arbitrarily dividing each transmural section of right and left ventricular free walls into thirds. Evaluation of fibrous tissue was made using a light microscope and the Quantimet 500+ image analyzer system (Leica, Milton Keynes, UK).
To avoid interobserver variation the same operator examined all the sections, while blinded to the morphologic subtype. Sirius red F3Ba stained collagen red (Fig 1, A), transmitting light of a fixed range of wavelength. That was detected and highlighted by the image analyzer, which then calculated the percentage of the total field area occupied by red staining. The mean of six random fields per zone was used in the statistical analysis. Where fields included the dense collagen found in the wall of the intramyocardial blood vessels and the surrounding adventitia or the endocardium affected by endocardial fibroelastosis, these areas were excluded from the analysis.
|
Qualitative analysis
In order to assess the quality of the collagenous matrix 12 of the specimens with HLHS were chosen as representative of the morphologic subgroups. Because there were three hearts from each of the four subgroups, we used only three normally structured hearts of matching age range (2 days to 2 months) as controls. Longitudinal sections through the right and left ventricular free walls, cut at 14 µm, were mounted on glass slides. The slides were processed routinely in preparation for examination by using a scanning electron microscope (Hitachi S4000 Field Emission SEM, Japan) with photographic records.
Statistical methods
The Wald test was employed as it was felt to be the most appropriate test as the data were unbalanced with different number of hearts for each group and stratified with data from both ventricles. The Wald statistic refers to a
2 distribution and is equivalent to the F statistic in analysis of variance (ANOVA), where the data follows an F distribution.
For the data from the study of total collagen using the Quantimet analyzer a mixed model analysis of variance with three factors was used to compare the effects of group (HLHS, normal), ventricle (left ventricle, right ventricle), and zone (subepicardium, mesocardium, subendocardium) on the percentage of collagen in the myocardium. The presence of an interaction between any of these effects was considered. Normality assumptions were checked using the Watson statistic and Bartlett's test was used to check for equal variances. The data were analyzed utilizing the Genstat (5.3) statistics package (NAG Ltd, Oxford, UK).
The data from the analysis of type I and III collagen was analyzed in the same fashion looking at the same three factors of group, ventricle, and zone but two outcome measures were evaluated. These were the total collagen (type I + III) and the percentage of type I collagen (type I / [type I + III] x 100). For all analyses a p value of less than 0.05 was considered significant. The different morphologic subgroups of HLHS were not compared in the analysis as six hearts per category was considered insufficient to have any meaningful value.
| Results |
|---|
|
|
|---|
|
|
Total collagen.
Considering the effect of each factor on the total collagen (sum of types I and III collagen) there is no significant difference in the overall mean collagen between the groups (p = 0.99) or among the three zones of the ventricles (p = 0.67). There was however a difference between the ventricles analyzed (p = 0.006). Independent of the group or zone and concurring with the results of the image analyzer, the right ventricle showed more collagen than the left (2.80%, SD 1.01% versus 2.53% points, SD 0.95% respectively).
Although there was no overall group or zone effect when taken individually, there was an interaction effect between these two factors (p = 0.003). In other words there was a significant difference between the three zones if analyzed within the context of the group. Contrasting the different zones shows that there were significant differences both within the groups and between the groups as follows.
First comparing the differences between groups in contrast to the image analyzer the subendocardium of the normal hearts had significantly lower collagen than the subendocardium of the HLHS hearts (mean, 2.34% versus 0.30%; p = 0.009). On the other hand the subepicardium of the normal hearts had significantly more collagen than the subepicardium of the HLHS hearts (3.04% versus 2.62%; p = 0.026). The same trend was seen for the mesocardium but this did not reach statistical significance (mean, 2.80% versus 2.50%; p = 0.11).
Next considering zonal differences within each group the HLHS hearts showed a decrease in collagen from the subendocardium to the subepicardium to the mesocardium (2.85% to 2.62% to 2.50%). Although not achieving statistical significance (p = 0.12.) it followed the same overall pattern as that seen with the image analyzer. In contrast within the normal hearts although the pattern of the gradient was the opposite of that expected from the Quantimet analyzer, following a decrease from the subepicardium to the subendocardium, it did reach statistical significance (3.04% to 2.80% to 2.34%; subepicardium versus subendocardium; p = 0.002).
Architecture of fibrous matrix
Examination of the normal hearts revealed that the different components of the fibrous matrix were clearly identifiable. Although it was possible to differentiate the endomysial weave from the other components it proved difficult to appreciate its finer details. In all the normal specimens the zonal arrangement of the myocytes and the surrounding matrix could be clearly differentiated into the densely packed subendocardium and the subepicardium in contrast to the much less dense mesocardium. The overall zonal arrangement of the ventricular wall in HLHS was not different from that seen in the normal hearts. The different components of the matrix could be identified but revealed no appreciable qualitative difference compared with normal hearts (Fig 2).
Furthermore, there were no discernable differences between the left and right ventricle in HLHS.
|
| Comment |
|---|
|
|
|---|
Quantitative analysis
Despite several discrepancies between the image analyzer and the point counting method our study has also revealed several highly significant and consistent results. Primarily in comparing hearts with HLHS and normal hearts both methods have revealed that normal hearts have a higher percentage of collagen.
The fixed ratio of types I: III at 2:3 irrespective of ventricle, zone, or group is similar to the ratio found in normal hearts by Marijianowski and coworkers [22] although they used a different method of analysis. There are differences when compared with other analytic methods [13], which applied immunohistochemistry to differentiate and quantify the subtypes, but our study applied polarization microscopy in a quantitative context.
Transmural variation
The main area of discrepancy between the two methods was in the analysis of zonal distribution. The image analyzer revealed a zonal gradient of decreasing collagen from subendocardium to subepicardium to mesocardium irrespective of the group or ventricle analyzed. In contrast point counting revealed that in normal hearts, the gradient of collagen was reversed, decreasing from the subepicardium to subendocardium. There was also significantly less collagen in the subendocardium compared with HLHS. Although the gradient in HLHS followed the same pattern as that found by the image analyzer it did not achieve statistical significance.
The transmural variation in the percentage of collagen per field is in keeping with other studies [8, 16]. The subendocardium in HLHS showing more collagen than any other zone is also in agreement with previous studies [23] and although not fully explained probably represents the inadequacy of the coronary supply to this region. The observation that the right ventricle has more fibrous tissue than the left ventricle in both groups is in keeping with previous studies of adults. However it is in contrast to studies of newborn, which showed that the overall collagen content on biochemical analysis was equal in both ventricles [23]. The discrepancy may arise from differences in methodology [9, 24].
Qualitative analysis
Qualitatively other studies [2, 25] have shown that pressure overloaded ventricles demonstrate an increase in the density of the weave as well as an increase in the diameter of the remaining components. The quality of the different components of the collagenous matrix as seen on our scanning electronmicrographs does not seem to differ between the two groups. It is possible that any small changes in the quality of the matrix were not detectable by us. It proved difficult to appreciate the finer detail of the endomysial weave, which could be explained by inadequate preservation of this more delicate structure.
Conclusions
Hearts with HLHS have significantly less percentage of collagen per field compared with normal. This reduction in collagen has been observed in not only the morphologically abnormal left ventricle but also in the right ventricle. We believe that this reduction in collagen is an intrinsic abnormality and represents a disproportionate quantity of fibrous tissue with possible important implications for the functional outcome of these ventricles over the longer term. Our results raise the question of whether the right ventricle is the appropriate ventricle for assuming the role of the sole pumping chamber following the Norwood procedure.
Limitations of the study
A discrepancy between the two methods of analysis of the subendocardial collagen in the normal hearts could be explained by the size of the samples and the differences in the sensitivities of the two techniques. This could also account for the overall zone and group effects not achieving statistical significance despite showing the same trend.
As for the effects of hypertrophy there is consensus that hemodynamics of a ventricle does influence myocyte growth [13, 26, 27]. Thus one would expect the right ventricle in HLHS to show myocyte hypertrophy. Conversely the left ventricle will be more variable, in cases with such little egress of blood that there may appear to be a degree of atrophy [28]. Assuming myocyte hypertrophy in the right ventricle of HLHS, the effect would be a dilution of the percentage of collagen per field area rather than the increase compared with the left ventricle. Calculation of the myocyte size would have allowed an estimate of the absolute difference in collagen between the two groups. Owing to difficulties in distinguishing cell borders in our tissues we were unable to assess myocyte size.
We were limited by the number of specimens available as the archive material at the Royal Brompton Hospital is also used extensively for education. Nevertheless we selected examples representative of the spectrum of hearts with HLHS. Although our study material had no clinical data for correlations, this we hope would stimulate future research on a well-documented series that includes fetal hearts.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. Davies, Y. d'Udekem, O. C. Ukoumunne, E. M. Algar, D. F. Newgreen, and C. P. Brizard Differences in extra-cellular matrix and myocyte homeostasis between the neonatal right ventricle in hypoplastic left heart syndrome and truncus arteriosus Eur. J. Cardiothorac. Surg., October 1, 2008; 34(4): 738 - 744. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Gambetta, M. K. Al-Ahdab, M. N. Ilbawi, N. Hassaniya, and M. Gupta Transcription repression and blocks in cell cycle progression in hypoplastic left heart syndrome Am J Physiol Heart Circ Physiol, May 1, 2008; 294(5): H2268 - H2275. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Takeuchi, A. Murakami, T. Takaoka, and S. Takamoto Evaluation of valved saphenous vein homograft as right ventricle-pulmonary artery conduit in modified stage I Norwood operation Interactive CardioVascular and Thoracic Surgery, August 1, 2006; 5(4): 345 - 348. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M. Cardis, D. A. Fyfe, and W. T. Mahle Elastic Properties of the Reconstructed Aorta in Hypoplastic Left Heart Syndrome Ann. Thorac. Surg., March 1, 2006; 81(3): 988 - 991. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Takabayashi, H. Kado, Y. Shiokawa, K. Fukae, and T. Nakano Comparison of hemodynamics between Norwood procedure and systemic-to-pulmonary artery shunt for single right ventricle patients Eur. J. Cardiothorac. Surg., June 1, 2005; 27(6): 968 - 974. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Salih, M. N. Sheppard, and S. Y. Ho Morphometry of coronary capillaries in hypoplastic left heart syndrome Ann. Thorac. Surg., March 1, 2004; 77(3): 903 - 907. [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 |