Author + information
- Received April 6, 2009
- Revision received May 19, 2009
- Accepted June 1, 2009
- Published online November 10, 2009.
- Elisabeth Bédard, MD,
- Karen P. McCarthy, BSc,
- Konstantinos Dimopoulos, MD, MSc,
- Georgios Giannakoulas, MD, PhD,
- Michael A. Gatzoulis, MD, PhD and
- Siew Yen Ho, PhD⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Siew Yen Ho, Cardiac Morphology, National Heart and Lung Institute, Imperial College London and Royal Brompton Hospital, Dovehouse Street, London SW3 6LY, United Kingdom
Objectives The purpose of this study was to determine whether intrinsic histological abnormalities of the pulmonary trunk (PT) are present from birth and interact with palliative surgery and/or repair.
Background Little is known about PT histology in patients with tetralogy of Fallot (TOF), especially in the era of surgical intervention in childhood.
Methods We studied 39 formalin-fixed necropsy heart specimens with TOF and compared them with 17 normal control heart specimens. Sections of the PT and aorta were studied by light microscopy using various stains; histological findings were graded according to severity.
Results Among the TOF group (1 fetus, 11 infants, 14 children, and 13 adults), 11 patients had undergone palliative and 10 patients had undergone reparative surgery at a median age of 8 years (range 2.5 to 18 years). Histological changes of grade 2 or higher were present in 59% (medionecrosis), 36% (fibrosis), 56% (cystlike formation), and 56% (abnormal elastic tissue configuration) of TOF patients. Total histology grading scores were higher in TOF hearts (median 6, range 1 to 9) compared with controls (median 1, range 0 to 6; p < 0.0001). Histological abnormalities were present among infants (median score 3.5, range 1 to 9) and after palliative surgery (median score 5, range 2 to 9) or repair (median score 7.5, range 4 to 9).
Conclusions Marked histological abnormalities in the PT of hearts with TOF exist compared with controls. These changes were present from infancy and among patients who had undergone palliative or reparative surgery, although operations in this cohort were performed late. Our data suggest that structural abnormalities of the PT, similar to these recently shown in the aorta, are intrinsic.
Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart defect, accounting for 10% of all congenital heart defects (1,2), with a reported incidence of 3.6 per 10,000 live births (3). Major advances in cardiac surgery over the past 50 years have resulted in a marked increase in the number of patients with TOF reaching adulthood, with a reported survival rate of 85% at 36 years after repair (4). However, repair is not curative, and a significant proportion of patients have important residua and sequelae. Although abnormal histological changes of the ascending aorta (Table 1)were recently reported in as many as 82% of TOF patients late after repair (5), between 6% and 20% of them in clinical series have documented aortic root dilation and secondary aortic regurgitation (6,7). Such histological changes, which are present from birth, also affect the stiffness of the arterial wall, with a potentially adverse effect on the corresponding left ventricle (8,9).
Although very few data exist on pulmonary trunk (PT) histology in patients with TOF, PT histology findings are at least equally relevant because they potentially affect post-repair pulmonary regurgitation (PR) and right ventricular (RV) dysfunction. PR, which has been reported in as many as 73% of TOF patients late after repair (10,11), and RV dysfunction are in turn most commonly closely related with impaired functional capacity, increased risk of ventricular arrhythmias, and sudden cardiac death (11–13). The mechanisms for and timing of RV dysfunction in this setting are not yet fully understood; its severity at times seems to be disproportionate to the extent of PR.
The normal structure of the PT changes substantially from fetal life to adulthood, as shown in Table 2(14). The histological resemblance of the PT to the aorta is striking at birth, when the medial thickness of both vessels is approximately the same. Normally, the PT-to-aortic media thickness ratio (PT/Ao MT ratio) gradually decreases within the first year of life. The elastic tissue configuration (ETC) changes from birth until it achieves the adult pattern by the end of the second year. With aging, the PT becomes wider and its medial thickness slightly increases (15). It remains unknown whether early reports on PT abnormalities in TOF (16) are secondary to abnormal blood flow patterns or intrinsic due to unrecognized genetic defects (5,14,16).
The aim of our study was therefore to examine morphological and histological abnormalities of the PT and of the aorta in patients with TOF of various ages and whether they relate to previous palliative or reparative surgery.
After approval from our hospital's ethics committee, we retrieved 176 formalin-fixed heart specimens with TOF from our cardiac morphology archive. These were TOF with pulmonary stenosis (PS) or pulmonary atresia (PA) with or without palliative surgery and/or reparative surgery. They did not include complex cases associated with congenital heart disease such as atrial isomerism, atrioventricular septal defect, and cases with major systemic-to-pulmonary collateral arteries and absent PT. All TOF specimens had the aortic valve overriding a ventricular septal defect and muscular subpulmonary obstruction due to anterocephalad deviation of the outlet septum. After gross examination, we further excluded 137 specimens because some anatomical structures were missing, thus precluding complete analysis, or the age was not available. Consequently, 39 specimens were suitable for both macroscopic and microscopic study.
Direct measurements of the luminal surface of the PT and aortic circumference 5 mm above the sinutubular junction were performed using a thread (5). The full-thickness of the PT and aorta was measured at the same level using calipers. Measurement of the PT circumference in TOF specimens with previous conduit repair was not attempted when the native PT could not be assessed (only the native PT was measured). The length of the LV and RV cavity was measured from the left and right atrioventricular junction to the left and right ventricular apex, respectively. Because of the age and heart size ranges, the following were indexed: the pulmonary trunk circumference indexed to the length of the left ventricle (PT/LV index) and the aortic circumference indexed to the length of the left ventricle (Ao/LV index). LV and RV free wall thickness was measured at the midpoint between the atrioventricular junction and apex of the corresponding ventricle. Semilunar and atrioventricular valves were inspected to exclude possible incompetence or obstruction that could affect the size of the ventricles. All direct measurements were performed by 1 experienced investigator.
Full-thickness samples of the arterial walls were harvested 5 mm distal to the sinutubular junction of the PT and aorta for histology. Sections 6 μm thick were cut from paraffin blocks of each specimen and stained with hematoxylin and eosin, elastic van Gieson, and Alcian blue. Each section was examined by 2 observers at 2 separate occasions. In instances of discrepancy, the specimens were re-evaluated by both observers and by a third observer, and a consensus was made.
Sections of the aorta were examined under light microscopy for the presence of medionecrosis, fibrosis, cystlike formation (CLF), elastic fragmentation, and number of elastic lamellae. According to the classification proposed by Schlatmann and Becker (17) and adapted from Tan et al. (5), histological changes for each variable were graded from 0 (absent) to 3, depending on the severity of the process (Table 1).
Sections of the PT were also studied by light microscopy for the presence of medionecrosis, fibrosis, and CLF and were graded as described in Table 1. Because elastic lamellae in normal PT are short, overlapping, nonuniform, and somehow fragmented (14), elastic lamellae could not be counted, and elastic fragmentation was not included in the variables studied in PT specimens. The ETC of the PT was classified according to Heath et al. (14) (Table 2). The PT/Ao MT ratio was also calculated.
Overall Histological Grading Score
As we previously described (5), a grading system was used to better classify the severity of histological changes and establish its relationship with the Ao/LV index and the PT/LV index. For the purpose of obtaining an overall histological grading score (HGS), we assigned points for each grade of histological change for each feature (i.e., grade 0 = 0 points, grade 1 = 1 point, grade 2 = 2 points, grade 3 = 3 points; elastic lamellae units with no disruption = 0 points, if units were disrupted = 2 points) and the total scores were added. A similar grading system was adapted for the PT, considering the presence or absence of abnormal ETC (normal ETC = 0 points, abnormal ETC = 2 points), excluding elastic fragmentation, as discussed previously.
Macroscospic and microscopic analyses of the PT and aortic root were similarly performed in 17 normal heart specimens; none had a cardiovascular cause of death.
Analyses were performed using R version 2.6.2 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were expressed as median (range) and categorical variables as number (percentage). Comparisons between groups were performed using the Wilcoxon rank sum test or the Fisher exact test as appropriate. All p values were 2 sided, and a p value <0.05 was pre-specified as indicative of statistical significance.
Baseline characteristics of the TOF patients are presented in Table 3.Among the TOF group (39 specimens), there were 1 fetus, 11 infants, 14 children, and 13 adults. Twenty-two (56%) patients had TOF with PS, 17 (44%) had TOF with PA, and there were no cases of absent pulmonary valve syndrome. Eleven (28%) TOF patients had palliative surgery only, whereas 10 (26%) patients had undergone repair (in 8, repair was preceded by palliative surgery). Seventeen controls with normal hearts and vessels were used for comparison (5 infants, 1 child, 11 adults; median age 38 years, range 1 day to 70 years). Death had occurred between 1975 and 1999 in the TOF group and between 1972 and 2004 in the control group.
The PT/LV index was significantly smaller in TOF specimens than in normal controls (median 0.41, range 0.05 to 1.02 vs. median 0.68, range 0.56 to 1.00; p = 0.002). This was more so in cases with PA (median 0.19, range 0.05 to 0.73) and those with previous palliative surgery only (median 0.19, range 0.05 to 0.74) (Fig. 1A).Among the TOF hearts, the PT/LV index was greater in hearts with previous repair (median 0.67, range 0.30 to 1.02).
Histological Changes and Elastic Tissue Configuration
Histological changes of grade 2 or higher were more frequent in TOF specimens compared with controls, including medionecrosis (59% vs. 12%; p = 0.001), fibrosis (36% vs. 6%; p = 0.02), and CLF (56% vs. 6%; p = 0.0003) (Figs. 2Band 3Dto 3F). Twenty-two (56%) of TOF specimens also had abnormal ETC for their age (Fig. 3A to 3D) compared with none in control hearts (p < 0.0001) (Fig. 2B). Eight (67%) infants with TOF had CLF of grade 2 or higher, and 33% had abnormal ETC; in contrast, none of these abnormalities were present in normal infant hearts (p = 0.03 and p = 0.002, respectively). Furthermore, grade 2 CLF was present in the fetal heart specimen (30 weeks of gestation) with TOF. Hearts with previous repair tended to have a lower prevalence of CLF (30% of specimens vs. 65% in the unrepaired group; p = 0.07), whereas the frequency of medionecrosis (90%), fibrosis (80%), and abnormal ETC (70%) was higher than in other TOF subgroups (Fig. 2D). Only 7.7% of all infants with TOF had normal PT histology, considering that grade 1 abnormality can be normal (1 normal infant control had grade 1 medionecrosis).
Histology Grade Scores
TOF specimens had higher HGS compared with controls (median 6, range 1 to 9 vs. median 1, range 0 to 6; p < 0.0001). As with the aorta (see the following text) even infant hearts with TOF had higher PT HGS compared with normal infant controls (median 3.5, range 1 to 9 vs. median 0, range 0 to 1; p = 0.002). Patients with previous surgery had the highest HGS (median 7.5, range 4 to 9; p < 0.0001). There was no correlation between HGS and PT/LV index (r = 0.11, p = 0.44). Pulmonary HGS showed a good correlation with aortic HGS in TOF patients (p = 0.0005, R = 0.53). There was no difference between the HGS of hearts with PS and those with PA (p = 0.35).
PT/Ao MT Ratio
Hearts with TOF had lower PT/Ao MT ratios compared with normal control hearts (median 0.46, range 0.22 to 1.30 vs. median 0.67, range 0.41 to 1.31; p = 0.003), which was also present among infant heart specimens (median 0.42, range 0.22 to 1.30 vs. median 0.62, range 0.50 to 1.31; p = 0.03). The PT/Ao MT ratio was even lower in TOF hearts with previous repair (median 0.30, range 0.23 to 0.77; p = 0.001). The PT/Ao MT ratio was inversely related to PT HGS (r = 0.43, p = 0.0009) and abnormal ETC (r = 0.30, p = 0.02) (Figs. 4Aand 4B).
The Ao/LV index was greater in the TOF group than in controls (median 0.87, range 0.64 to 1.18 vs. median 0.71, range 0.48 to 1.00; p = 0.003) (Fig. 1B). This difference was still present when comparing infants only (median 0.87, range 0.65 to 1.18 vs. median 0.67, range 0.48 to 0.88; p = 0.04). In the TOF group, hearts that had palliative surgery only had the highest Ao/LV index (median 0.99, range 0.71 to 1.18; p = 0.003), whereas previously repaired hearts did not demonstrate an increased Ao/LV index (median 0.76, range 0.64 to 1.06; p = 0.09). Furthermore, there was no difference in aortic size between patients with PA and PS (median 0.93, range 0.71 to 1.1 vs. median 0.84, range 0.64 to 1.18; p = 0.34) (Fig. 1B).
Histological Changes and Histology Grade Scores
Histological changes of grade 2 or higher were much more common in the TOF hearts compared with controls, including elastic fragmentation (49% vs. 6%; p = 0.002), medionecrosis (54% vs. 6%; p = 0.0008), fibrosis (54% vs. 0%; p < 0.0001), and CLF (74% vs. 6%; p < 0.0001) (Fig. 2A). Infant hearts with TOF also had important histological abnormalities, especially CLF, present in 7 (58%) cases. TOF hearts with previous repair had more histological changes of grade 2 or higher than those without (Fig. 2C). There were more disrupted elastic lamellae in TOF patients compared with controls (31% of specimens vs. 0%; p = 0.01) and a trend toward fewer elastic lamellae units (median 80 U, range 48 to 119 U vs. median 97 U, range 50 to 142 U; p = 0.08). Total repair was associated with higher lamellar count (median 94 U, range 68 to 115 U; p = 0.70), but greater disruption of elastic lamellae (50% of specimens; p = 0.003).
TOF specimens had higher HGS compared with controls (median 7, range 1 to 12 vs. median 1, range 0 to 3; p < 0.0001). HGS were also higher among infant hearts with TOF compared with infant control specimens (median 3.5, range 1 to 9 vs. median 0, range 0 to 2; p = 0.008). Previous surgery was associated with higher HGS (median 9, range 5 to 11). No correlations were found between HGS and sex, Ao/LV index (r = 0.23, p = 0.12), the presence of PA, and right aortic arch.
Our study demonstrates that intrinsic histological abnormalities are present from birth and probably from fetal life, not only in the aortic root, but also in the PT of patients with TOF, and persist even after palliative surgery and/or repair.
Effect of aging on the normal PT histology
The effect of aging on histological changes such as medionecrosis, fibrosis, and cystic medial necrosis in the normal PT has not been fully elucidated. Few reports, however, suggest that histological abnormalities of grade 2 or higher are rare in the normal adult PT (18,19). Our histological analysis of 17 normal hearts of different age groups showed that medionecrosis, fibrosis, and cystic medial necrosis of grade 2 or higher were present in few normal adults but were completely absent in infants (Fig. 2B). Furthermore, as previously described by Heath et al. (14), all normal specimens in our cohort had a normal ETC for their age group.
Histological changes in the PT of patients with TOF
To our knowledge, this is the first study to show that PT histological changes of grade 2 or higher (medionecrosis, fibrosis, and CLF) are common in patients with TOF. This was true even for infants and patients who had previous palliative surgery or repair. Three cases of adults with TOF and grade 3 PT histological changes were previously reported (16), but 2 had absent pulmonary valve syndrome with aneurysmal PT, which may be considered a different clinical entity. Our finding of abnormal PT ETC (Figs. 3A to 3D) in >50% of all TOF specimens (33% of infant specimens) is also consistent with previous observations by Heath et al. (14), who described an abnormal ETC in 58% of TOF specimens.
The effect of repair on PT histology
Although repair of TOF may have a beneficial effect on aortic root dilation (Fig. 1B), it did not seem to relate to the severity of histological abnormalities of the PT and aorta; these abnormalities were even more advanced in the repaired group (Fig. 2). Repair was performed late in this cohort; however, it has been suggested that after a certain age, the reduced content of elastin found in the PT of TOF patients cannot be regenerated (20). It cannot be excluded, however, that worsening of the histological abnormalities after repair reflects the inability of these intrinsically abnormal arteries to adapt to a significant increase in pulmonary blood flow after surgery. Progressive aortic root dilation after reparative surgery has been reported in TOF patients and supports this hypothesis (6).
Potential clinical implications of the histological abnormalities in the PT
Even though histological abnormalities in the aorta have been related to increased risk of aortic dissection and rupture, these complications are relatively uncommon in TOF patients (7). Perhaps more importantly, recent data suggest that the characteristics of the great arteries affect the function of the corresponding ventricle (8). RV dilation and dysfunction after TOF repair represent important causes of morbidity and mortality (11,12). However, the mechanism for this is not fully understood. RV dysfunction is often but not universally related to the degree of PR, and pulmonary valve replacement does not always result in significant recovery of RV function (21). PR itself is known to relate to the type of RV outflow tract reconstruction (22), RV diastolic compliance (23), and the presence and extent of branch pulmonary artery stenosis (24,25). However, the severity and progression of PR vary among individual patients, and, to date, we lack specific predictors of its clinical course and timing.
Histological abnormalities in the PT as described here can affect vascular stiffness, which in turn may affect PR and RV adaptation to volume and/or pressure overload. The high proportion of CLF, elastic fragmentation, and fibrosis found in both great vessels among TOF patients reflects a loss in normal elastic fibers. The lower PT/Ao MT ratio (or thinner PT media), which relates to higher PT HGS (Fig. 4), also suggests reduced elastin content (14,20,26), potentially leading to lower extensibility of pulmonary arteries (27,28).
Similar to Marfan syndrome and to patients with bicuspid aortic valve, aortic stiffness in TOF patients is associated with aortic root dilation (9,29–31) and more recently has been shown to adversely affect LV ejection (8). Reduced aortic elasticity in patients with bicuspid aortic valve has been associated with the severity of aortic regurgitation and the degree of LV hypertrophy (32). We suggest that PT stiffness in TOF patients may similarly predispose to more PR and propagate RV dysfunction. Cardiac magnetic resonance imaging studies are currently being conducted to address these major clinical implications.
Mechanism of histological abnormalities: hemodynamic stress, genetic defect, or apoptosis?
The pathogenetic mechanism responsible for the changes in PT morphology of TOF patients may resemble that of the aorta. Histological changes in the PT may be attributable to low flow or post-stenotic turbulence, even during fetal life. However, there may also be an intrinsic, possibly genetic, origin of the histological abnormalities in the PT of infants and fetuses with TOF. CLF, previously described as cystic medial necrosis (CMN), was a common finding in both great vessels in TOF specimens (Figs. 2and 3E). This abnormality is common in patients with Marfan syndrome (33) and bicuspid aortic valve. In Marfan syndrome, CMN has been attributed to a genetic defect in fibrillin-1, resulting in increased elastolysis by metalloproteinases (33). Fibrillin-1 genetic mutations could also play a role in aortic root dilation found in TOF patients (34). CMN in the bicuspid aortic valve has been related to premature vascular smooth muscle cell apoptosis, which was also recently identified in Marfan syndrome, suggesting a genetic background to premature vascular smooth muscle cell apoptosis and CMN (35–37).
Measurements of the aortic root and PT could not be blinded to the presence of TOF and previous surgical repair because the great vessels had to be retained on the specimens. Furthermore, direct comparison with available normograms of indexed aortic root and PT dimensions could not be made because formalin fixation causes shrinkage of valves and arteries. However, storage of specimens was uniform, and possible changes would have affected similarly both the control hearts and various subgroups of TOF. Morphological/histological analyses were limited to the PT; the pulmonary vascular bed beyond the PT was not available. A very limited amount of clinical data was available in the TOF group; coexisting diseases affecting the aorta and/or PT could not, therefore, be totally excluded, although unlikely in this young population. Hemodynamic information such as pulmonary artery pressures and blood flow, vascular compliance, and cardiac index was not available because of the nature of our study. However, it should form the basis of a larger prospective clinical investigation addressing great vessel stiffness with its impact on PR/aortic regurgitation and RV/LV function, respectively, and on late outcomes after TOF repair.
Patients with TOF have remarkable intrinsic histological abnormalities in both great arteries that seem to be present from birth and even from fetal life. Our data suggest that TOF repair does not seem to improve these changes, although repair was performed late in this cohort. This predisposition to abnormal cellular development of the great arterial media may have a genetic substrate affecting PT stiffness and potentially PR or RV dysfunction, which warrants further investigation.
Dr. Bédard is supported by the Cardiology Institute of Quebec, Laval University, and the Cardiologists Association of the province of Quebec. Dr. Dimopoulos has received support from the European Society of Cardiology. Dr. Giannakoulas has received support from the Hellenic Cardiological Society, the Propondis Foundation, and the Hellenic Heart Foundation. Prof. Gatzoulis and the Royal Brompton Adult Congenital Heart Programme have received support from the British Heart Foundation. The Cardiac Morphology Unit receives funding support from the Royal Brompton and Harefield Hospital Charitable Fund.
- Abbreviations and Acronyms
- Ao/LV index
- aortic circumference indexed to the length of the left ventricle
- cystlike formation
- cystic medial necrosis
- elastic tissue configuration
- histological grading score(s)
- pulmonary atresia
- pulmonary regurgitation
- pulmonary valve stenosis
- pulmonary trunk
- PT/Ao MT ratio
- pulmonary trunk-to-aortic media thickness ratio
- PT/LV index
- pulmonary trunk circumference indexed to the length of the left ventricle
- right ventricular
- tetralogy of Fallot
- Received April 6, 2009.
- Revision received May 19, 2009.
- Accepted June 1, 2009.
- American College of Cardiology Foundation
- Gatzoulis M.A.
- Hoffman J.I.,
- Kaplan S.
- Nollert G.,
- Fischlein T.,
- Bouterwek S.,
- Bohmer C.,
- Klinner W.,
- Reichart B.
- Tan J.L.,
- Davlouros P.A.,
- McCarthy K.P.,
- Gatzoulis M.A.,
- Ho S.Y.
- Niwa K.,
- Siu S.C.,
- Webb G.D.,
- Gatzoulis M.A.
- Senzaki H.,
- Iwamoto Y.,
- Ishido H.,
- et al.
- Cheung Y.F.,
- Ou X.,
- Wong S.J.
- Geva T.,
- Sandweiss B.M.,
- Gauvreau K.,
- Lock J.E.,
- Powell A.J.
- Gatzoulis M.A.,
- Till J.A.,
- Somerville J.,
- Redington A.N.
- Niwa K.,
- Perloff J.K.,
- Bhuta S.M.,
- et al.
- Therrien J.,
- Siu S.C.,
- McLaughlin P.R.,
- Liu P.P.,
- Williams W.G.,
- Webb G.D.
- Davlouros P.A.,
- Kilner P.J.,
- Hornung T.S.,
- et al.
- Gatzoulis M.A.,
- Clark A.L.,
- Cullen S.,
- Newman C.G.,
- Redington A.N.
- Chaturvedi R.R.,
- Kilner P.J.,
- White P.A.,
- Bishop A.,
- Szwarc R.,
- Redington A.N.
- Johnson R.J.,
- Haworth S.G.
- Harris P.,
- Heath D.,
- Apostolopoulos A.
- Nollen G.J.,
- Groenink M.,
- Tijssen J.G.,
- Van Der Wall E.E.,
- Mulder B.J.
- Grotenhuis H.B.,
- Ottenkamp J.,
- Westenberg J.J.,
- Bax J.J.,
- Kroft L.J.,
- de Roos A.
- Bonderman D.,
- Gharehbaghi-Schnell E.,
- Wollenek G.,
- Maurer G.,
- Baumgartner H.,
- Lang I.M.