Ann Thorac Surg 2004;77:903-907
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Morphometry of coronary capillaries in hypoplastic left heart syndrome
Caner Salih, MB, ChB, FRCSa,
Mary N. Sheppard, MD, FRCPathb,
Siew Yen Ho, PhD, FRCPatha*
a Departments of Pediatrics and Histopathology, Imperial College, London, United Kingdom
b Royal Brompton and Harefield NHS Trust, London, United Kingdom
Accepted for publication July 10, 2003.
* Address reprint requests to Dr Ho, Department of Paediatrics, Faculty of Medicine, Imperial College, National Heart & Lung Institute, Dovehouse St, London SW3 6LY, UK
e-mail: yen.ho{at}imperial.ac.uk
 |
Abstract
|
|---|
BACKGROUND: Hypoplastic left heart syndrome is a condition characterized by a constellation of morphological malformations affecting the left side of the heart. We studied the capillary network, and quantified the capillarization of the ventricular myocardium, which, if different from normal, may have implications for the success of surgical reconstruction.
METHODS: The capillaries were detected by immunohistochemistry using a monoclonal antibody (von Willebrand's factor) against the endothelium. Hearts with hypoplastic left heart syndrome have higher mean and maximal diffusion distances from any arbitrary point to the nearest capillary than normal hearts.
RESULTS: There was no significant difference in the heterogeneity of capillary distribution between the hearts with hypoplastic left heart syndrome and the control heart. Increase in distance was found in both the right and left ventricles.
CONCLUSIONS: Hearts with hypoplastic left heart syndrome show a reduction in the capillarization of both the right and left ventricles compared with age-matched controls. We believe this may be an inherent abnormality of hypoplastic left heart syndrome that may have implications for ventricular development.
|
Introduction
|
|---|
Hypoplastic left heart syndrome (HLHS) is a malformation associated with a spectrum of restriction to left ventricular inflow, outflow, or both. The left ventricle is usually hypoplastic, as is the ascending aorta [1]. The commonly used reconstructive procedures for HLHS, the Norwood and Fontan procedures, have been implemented successfully only over the last decade or so. There are very little data on the long-term outcome of either operation. Studies of outcome have shown that even after a perfect Fontan procedure, the right ventricle will fail in the long term. The reasons underlying this failure are not clear.
Previous studies [2] on mature hearts with acquired aortic stenosis demonstrated that the capillary density of the left ventricle can decrease with pressure overload hypertrophy. Measuring the intercapillary distance and using standard deviation, Rakusan and associates [2] established that the heterogeneity of spacing showed a general increase, although this did not achieve significance. In contrast, younger hearts with congenital aortic stenosis showed no change in the capillary density or spacing heterogeneity [2]. It has been suggested, therefore, that these hearts maintain the capillary density and heterogeneity of spacing in parallel with angiogenesis.
Several studies [3–7], have focused on the evaluation of coronary arteries in HLHS and the results have been variable. O'Connor and colleagues [6] described nine postmortem specimens with HLHS, all of which demonstrated numerous microscopic coronary-cameral connections and thick-walled coronary arteries. However, these changes had no effect on luminal diameter. In a study of 51 specimens with HLHS, Lloyd and co-workers [4] examined the diameters of the ascending aorta, the right and left coronary arteries, and the respective coronary ostia. They could find no correlation between these diameters and the extensive myocardial necrosis seen in the right ventricular papillary muscle. Sauer and associates [3] reported tortuousity, fibroelastic thickening of the intima, and fragmentation and duplication of the internal elastic lamina. These findings were more common in the coronary arteries of hearts with mitral stenosis and aortic atresia. These authors concluded that this subgroup is less well suited for a palliative procedure because of coronary artery morphology that could impair right ventricular perfusion. In contrast, Baffa and coauthors [5] examined 151 postmortem specimens with HLHS in an attempt to determine if any subgroup showed compromised right ventricular preservation. Although they also found a higher frequency of gross coronary abnormalities in the form of coronary-cameral communications in the left ventricle, they could find no areas of coronary artery stenosis or interruption and concluded that there was no subgroup difference in perfusion of the right ventricle.
Despite the major findings in these studies, there has been very little work on the perfusion of the ventricles at the level of the capillaries, which, ultimately, is the final common link in the perfusion chain and may be one of the factors contributing to ventricular development. The oxygen diffusion distance from capillaries to tissues is critical to aerobic metabolism, with the time needed to attain any specific oxygen tension being directly proportional to the square of the distance [8–10]. Previous studies of capillarization have used either capillary density [11, 12] or intercapillary distance [13] as an index of the average oxygen-diffusion distance. However, these indices are inadequate unless it is known whether the capillary distribution follows a random, ordered, or clustered pattern. Moreover, the maximal diffusion distance is more critical to aerobic metabolism than average distance [14]; again the capillary density and intercapillary distance do not allow calculation of this variable. For this study, therefore, we used the "closest individual method of analysis" [15, 16] as a means of direct measurement and analysis of the complete distribution of distances from arbitrary points in the tissue to the nearest capillary. With this method, we were able to calculate not only the mean and maximal diffusion distance but also the variability in capillary spacing.
|
Material and methods
|
|---|
Specimens and image analysis
Fifteen hearts were chosen from the archives at the Royal Brompton Hospital, London. These represented the whole range of morphological subtypes of HLHS on the basis of the status of inflow and outflow and included the following: mitral stenosis and aortic stenosis (n = 3), aortic atresia and mitral stenosis (n = 3) or atresia (n = 3), aortic stenosis and mitral atresia (n = 2), and normal hearts (n = 4).
All the hearts were from infants less than 2 months of age. None had previous surgical procedures. The 4 normal hearts that served as normal age-matched controls were from infants who died of noncardiac causes. All the specimens had been fixed by immersion in 10% formaldehyde solution. Transmural slices of the left and right ventricular free walls incorporating, where possible, the whole length of the ventricle from the atrioventricular groove to the apex were taken. The slices were processed routinely for histological study and cut 7 µm thick. The sections were stained with antibodies directed against von Willebrand's factor (DAKO Laboratories) to show capillary endothelium using an automated immunohistochemistry processor.
With a magnification of x40, the sections from both the right and left ventricles were viewed under a light microscope. The image was transferred to a monitor by means of a color video camera (JVC TK-1280E) for image analysis using the Quantimet 500+ image analyzer (Leica, Milton Keynes, UK).
A grid of 72 fixed points marked out at regular intervals was overlaid on the screen, and the distance from each of these points to the nearest capillary was measured and recorded (Fig 1). The measurements were repeated for six different frames. These distances were then plotted as a frequency distribution curve from which the maximal and mean diffusion distances (95th percentile and 50th percentile, respectively) can be calculated as well as the standard deviation, which can be used as an index of variation or heterogeneity of spacing.
View larger version (57K):
[in this window]
[in a new window]
|
Fig 1. (A) Photomicrograph of a histological section and (B) corresponding overlay showing the capillaries (brown) and the grid used to analyze the nearest distance (arrows) to each capillary. For simplicity, only the first two rows of points are analyzed here (von Willebrand's factor stain: original magnification x40).
|
|
Statistical analysis
The data from all the ventricles were combined to give a single group for each ventricle from hearts with HLHS and the controls. The data were analyzed using the GenStat 5.3 statistics package. The Wald test, which is restricted maximum likelihood procedure, was used. It was considered the most appropriate test because the data were unbalanced, with a 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, where the data follow an F distribution. Using this estimation method allowed an analysis of variance with two factors to compare the effects of group (HLHS, normal) and side (left ventricle, right ventricle) on the distance outcome measure. Nonnormality of the data was sought using the Watson statistic. Bartlett's test was used to check for equal variances.
|
Results
|
|---|
The analysis of the data revealed greater mean and maximal diffusion distances to the nearest capillary in hearts with HLHS compared with normal hearts (p = 0.02) (Tables 1, 2). There were, however, no significant differences in capillary distribution pattern between the left and right ventricles whether the heart had HLHS or not (p = 0.58) (see Table 2). Although the heterogeneity of spacing did show a tendency to be higher in HLHS hearts (Table 3) compared with normal hearts, this did not achieve significance (p = 0.05) (Table 4).
|
Comment
|
|---|
The cause of the increased diffusion distance is not clear. Obviously, the decrease in capillarization will be either an absolute or an apparent change. An apparent decrease could be due to an increase in the tissue mass separating each capillary, which, in turn, may be the result of an increase in either the fibrous matrix or the myocytes. In our study, we were unable to determine myocyte sizes because of inadequate visualization of the myocyte cell borders. However, in another study [17] we found that the percentage of collagen per field was significantly less in HLHS hearts than normal hearts. Although that study also did not assess the size of the myocytes, it suggested that the reduction in capillarization is unlikely to be due to an increase in the extracellular matrix. The trend toward an increase in variability suggests that HLHS hearts are more likely to demonstrate uneven blood distribution with some areas having more adequate perfusion than others.
There remains a possibility that the decrease in capillarization is secondary to an increase in myocyte size, number, or both. In support of myocyte hypertrophy may be the hemodynamics of the ventricles in HLHS [18]. Both the right and left ventricles would be expected to be under a pressure overload with the right ventricle also being under a volume overload. This would lead to hypertrophy of the ventricular myocytes. However, the precise hemodynamics of hearts in the presence of HLHS is not clear and may vary with not only the precise morphological subtype but also the timing of development of the malformation during fetal life. For instance, it is possible to envisage the contrasting situation in a left ventricle that has very little outflow of blood as a result of mitral and aortic atresia with a degree of myocyte atrophy rather than hypertrophy. Studies of animal models have shown that hypertrophy of the myocardium results in a decrease in capillary reserve [13]. Despite various models involving animal hearts [19–21], further research on capillarization of the myocardium in considerably larger groups of subtypes of HLHS is essential for a better understanding of the natural evolution as well as the development after surgical intervention. We plan to extend this type of investigation into a variety of congenital heart malformations.
Rakusan and colleagues [2] examined the capillary density and the heterogeneity of capillary spacing in left ventricular hypertrophy compared with healthy left ventricles. They found that in congenital aortic stenosis, young hearts were able to adapt to an increased myocardial demand by an increase in angiogenesis to maintain a constant capillary density. Although the heterogeneity of spacing did show an increase, it did not achieve significance. These findings were in contrast to those in older hearts where left ventricular hypertrophy led to a relative decrease in capillary density, attributed to inadequate angiogenesis. Hort and Severidt [22] also found no change in the capillarization of infants and children with a variety of congenital cardiac malformations that lead to a mixture of right and left ventricular pressure and volume overload and cyanosis. In our study, the hearts were all from very young patients (less than 2 months old) and in this respect may represent a different age subgroup from those studied by Rakusan and coauthors [2], whose infants were defined as less than 1 year old. More importantly, it is possible that the hearts would have shown adequate angiogenesis to maintain the capillary distribution as the child grew older.
Studies on tissue oxygenation in relation to capillary distribution or regeneration of the capillary bed after surgical reconstruction would be of value in predicting whether the right ventricle is likely to cope as the systemic ventricle. They would also provide further insight as to whether the left ventricle could be expected to grow after surgical repair that restores both ventricles to the circulation, the option of so-called biventricular repair.
The study has a few limitations in terms of choice and preparation of the specimens and immunohistochemistry. First, we had insufficient numbers to compare the effect of morphological subtypes on capillarization. We overcame any potential differences between subgroups by having a large number of sampling points and using random fields. Second, the method of fixation by immersion in formaldehyde could not be controlled. As highlighted by Rakusan and associates [2], this could have an effect on tissue preservation for studying capillarization, although their experiments with rats showed no significant difference between various fixation techniques using the same capillarization variables.
There is an inherent variability in the sensitivity of immunohistochemistry. Again, the high number of fields studied should have compensated for this. It is also possible that the use of CD34 and CD31 [23] would have been superior in defining the capillaries, but these antibodies did not work in our tissues that were from specimens archived over decades.
|
Conclusions
|
|---|
We have shown that hearts with HLHS have significantly reduced capillarization of both the left and right ventricles compared with normal hearts of a similar age. There is also a tendency for increased heterogeneity of spacing. The right ventricle, which hitherto had been assumed to be "normal," shows the same reduction in capillarization as the "abnormal" left ventricle. In our opinion, the lack of a difference in capillarization between the two ventricles, which are likely to be under different hemodynamic loads, and the young age of the specimens suggest that this may be an inherent abnormality of HLHS. Further investigations with physiological correlates and comparisons with other groups of congenital heart malformations are required to better understand the clinical implications of this morphological observation.
|
References
|
|---|
- Lev M. Pathological anatomy and interrelationship of hypoplasia of the aortic tract complexes. Lab Invest 1952;1:61-70.[Medline]
- Rakusan K., Flanagan M.F., Geva T., Southern J., Van Praagh R. Morphometry of human coronary capillaries during normal growth and the effect of age in left ventricular pressure–overload hypertrophy. Circulation 1992;86:38-46.[Abstract/Free Full Text]
- Sauer U., Gittenberger-de Groot A.C., Geishauser M., Babic R., Buhlmeyer K. Coronary arteries in the hypoplastic left heart syndrome. Histopathologic and histometrical studies and implications for surgery. Circulation 1989;80(Suppl 1):168-176.
- Lloyd T.R., Evans T.C., Marvin W.J. Morphologic determinants of coronary blood flow in the hypoplastic left heart syndrome. Am Heart J 1986;112:666-671.[Medline]
- Baffa J.M., Chen S.-L., Guttenberg M.E., Norwood W.I., Weinberg P.M. Coronary artery abnormalities and right ventricular histology in hypoplastic left heart syndrome. J Am Coll Cardiol 1992;20:350-358.[Abstract]
- O'Connor W.N., Cash J.B., Cottrill C.M., Johnson G.L., Noonan J.A. Ventriculocoronary connections in hypoplastic left hearts: an autopsy microscopic study. Circulation 1982;66:1078-1086.[Free Full Text]
- Freedom R.M., Culham J.A.C., Moes C.A.F., Harrington D.P. Selective root angiography in the hypoplastic left heart syndrome. Eur J Cardiol 1976;4:25-29.[Medline]
- Krogh A. The number and distribution of capillaries in muscles with the circulation of the oxygen pressure head necessary for supplying the tissue. J Physiol 1919;52:409-415.
- Kety S.S. Determinants of tissue oxygen tension. Fed Proc 1957;16:666-670.[Medline]
- Tenney S.M. A theoretical analysis of the relationship between venous blood and mean tissue oxygenation. Respir Physiol 1974;20:283-296.[Medline]
- Rakusan K. Quantitative morphology of capillaries of the heart. Number of capillaries in animal and human hearts under normal and pathological conditions. Methods Achiev Exp Pathol 1971;5:272-286.[Medline]
- Turek Z., Grandtner M., Kreuzer F. Cardiac hypertrophy, capillary and muscle fiber density, muscle fiber diameter, capillary radius and diffusion distance in the myocardium of growing rats adapted to a simulated altitude of 3500 m. Pfluegers Arch 1972;335:19-28.[Medline]
- Henquell L., Honig C.R. Intercapillary distances and capillary reserve in right and left ventricles: significance for control of tissue po2. Microvasc Res 1976;12:35-41.[Medline]
- Turek Z., Rakusan K. Lognormal distribution of intercapillary distance in normal and hypertrophic rat heart as estimated by the method of concentric circles: its effect on tissue oxygenation. Pfluegers Arch 1981;391:17-21.[Medline]
- Kayar S.R., Archer P.G., Lechner A.J., Banchero N. The closest individual method in the analysis of the distribution of capillaries. Microvasc Res 1982;24:326-341.[Medline]
- Greig-Smith P. Quantitative plant ecology. . New York: Academic, 1957.
- Salih C., McCarthy K.P., Yen Ho S. The fibrous matrix of ventricular myocardium in hypoplastic left heart syndrome: a quantitative and qualitative analysis. Ann Thorac Surg 2004;77:36-40.[Abstract/Free Full Text]
- Freedom R.M. Atresia or hypoplasia of the left atrioventricular and/or ventriculoarterial junction. In: Anderson R.H., Macartney F.J., Shinebourne E.A., Tynan M., eds. Paediatric cardiology. Edinburgh: Churchill Livingstone, 1987:737-764.
- Thomas DP, Phillips SJ, Bove AA. Myocardial morphology and blood flow distribution in chronic volume-overload hypertrophy in dogs. Basic Res Cardiol 1984;79:379–8.
- Alyono D., Anderson R.W., Parish D.G., Dai X.Z., Bache R.J. Alteration of myocardial blood flow associated with experimental canine left ventricular hypertrophy secondary to valvular aortic stenosis. Circ Res 2000;58:47-57.
- Tomanek R.J., Palmer P.J., Peiffer G.L., Schreiber K.L., Eastham C.L., Marcus M.L. Morphometry of canine coronary arteries, arterioles, and capillaries during hypertension and left ventricular hypertrophy. Circ Res 1986;58:38-46.[Abstract/Free Full Text]
- Hort W., Severidt H.J. Capillarisation and microscopic changes in the myocardium in congenital heart defects. Virchows Arch Pathol Anat Physiol Klin Med 1966;341:192-203.[Medline]
- Fina L., Molgaard H.V., Robertson D., et al. Expression of the CD34 gene in vascular endothelial cells. Blood 1990;75:2417-2426.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
P. Eghtesady, E. Michelfelder, M. Altaye, E. Ballard, R. Hirsh, and R. H. Beekman III
Revisiting Animal Models of Aortic Stenosis in the Early Gestation Fetus
Ann. Thorac. Surg.,
February 1, 2007;
83(2):
631 - 639.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
