The Impact of Preexisting Myocardial Remodeling on Ventricular Function Early after Tetralogy of Fallot Repair


Twenty-three patients (median age 23 months) who underwent Fallot’s tetralogy repair were investigated prospectively to detect a possible association between histopathologic myocardial remodeling and echocardiographic findings of systolic or diastolic ventricular dysfunction.


Intraoperatively resected infundibular bands and subendocardial biopsy samples from the right ventricle (RV) and left ventricle were obtained for histopathologic evaluation. Tissue Doppler echocardiographic interrogation of the ventricles was performed before surgery and in the postoperative period.


Histopathologic data revealed hypertrophy of the RV cardiomyocytes and increased interstitial collagen in both ventricles. Mean values of RV isovolumic acceleration decreased significantly at the third evaluation compared with the preoperative values ( P = .006). RV myocardial fibrosis greater than 8.3% was associated with a probability of altered E’ of at least 0.7 (odds ratio = 2.31).


Preoperative histologic myocardial remodeling influenced the postoperative RV function in this group of patients with late repair. Further studies are necessary to evaluate the myocardium in younger patients and to define its influence in the long-term follow-up.

Early repair of tetralogy of Fallot is generally advised to minimize the effects of chronic hypoxia and to reduce long-standing pressure overload to the right ventricle (RV) and its consequences. The surgical correction currently is achieved in children aged less than 1 year, with low mortality and excellent long-term survival in the majority of centers. However, the optimal timing for total correction is still controversial. In Brazil, because of the socioeconomic conditions of the population, some children may reach surgical correction later than patients from developed countries. This late correction could bring additional myocardial remodeling that would affect the postoperative evolution.

It is also well known that alterations in RV function may be present early after surgery, sometimes causing a prolonged, troublesome postoperative recovery, and in the long-term. Moreover, some studies have also shown impaired left ventricle (LV) systolic function after tetralogy of Fallot correction. Tissue Doppler imaging (TDI) associated with dobutamine stress echocardiography was used to evaluate the contractile reserve of the RV after tetralogy of Fallot repair. Impaired ventricular reserve could be predicted by TDI indices at rest.

A few studies have described the histopathology of the myocardium in hearts with tetralogy of Fallot. Findings include myocyte hypertrophy and fiber disarray, fibrosis, edema, and degenerative changes, which become more pronounced in older patients subjected to long-standing cyanosis and pressure overload. Those changes may account for ventricular dysfunction and arrhythmias in the long term. However, to our knowledge, the implications of preexisting myocardial abnormalities in the ventricular dysfunction seen early after repair have not been objectively addressed before.

The objectives of this prospective study were to analyze histomorphometric features of myocardial remodeling in the RV and LV in patients with tetralogy of Fallot, to identify whether any of these features may predispose patients to a higher risk of myocardial dysfunction before surgery and in the early and middle postoperative periods, and to look for possible associations between the histomorphometric aspects and other variables, such as age at operation, systemic arterial oxygen saturation, hematocrit, previous use of propranolol, electrocardiographic findings, surgical technique, and residual lesions.

Materials and Methods

Patient Population

We enrolled 23 consecutive patients with tetralogy of Fallot who were electively admitted for corrective surgery at the Heart Institute (InCor), University of São Paulo Medical School, from March 2005 to April 2007. Patient ages ranged from 12 to 186 months (mean = 39.6 months, median = 23 months), and 14 were male (60.9%). Patients with pulmonary atresia or associated atrioventricular septal defect were excluded from the study. Those presenting small atrial septal defects were included.

Before surgery, 47% of the patients had at least one episode of hypoxemic spell; 65% were receiving daily oral propranolol (mean dosage 1.8 mg/kg/day). Mean room air oxygen saturation was 88.1% ± 5.4%, and mean hematocrit was 40.22% ± 6.24%. Median preoperative Doppler gradient across the RV outflow tract was 74 mm Hg, ranging from 43 to 120 mm Hg. Three patients had undergone palliative Blalock–Taussig shunt previously.

The patients and their parents were informed about the research purpose of the data collection and gave their informed consent. The study was approved by the Ethics and Scientific Committee of the Heart Institute.

Clinical Assessment

All patients had a complete clinical assessment, including evaluation of oxygen saturation measured in room air using finger pulse oximetry. A blood sample was collected preoperatively for determination of hematocrit. All patients had a 12-lead surface electrocardiogram performed using a Hewlett-Packard recorder (Hewlett-Packard, Palo Alto, CA), at a speed of 25 mm/s and 1 mV/cm standardization, and the previous use of propranolol was registered.


A complete two-dimensional and Doppler echocardiographic examination was performed using echocardiography equipment (Philips-Sonus 5500 or Philips-HDI 5000, Andover, MA) with a 3.5- or 5.0-MHz transducer. Pulsed-wave TDI was performed by activating the TDI function in the same unit. Filters were set to exclude high-frequency signals. Gains were minimized to allow a clear tissue signal with minimal background noise. The TDI velocities were obtained from the four-chamber view of each patient. A 2-mm sample volume was placed at the lateral angle of the tricuspid or the mitral valve annulus. The peak myocardial velocities during early and late diastole, systole, and isovolumic contraction were recorded from both ventricles. The myocardial isovolumic acceleration (IVA) was measured by dividing the myocardial peak velocity during isovolumic contraction time to the time from the onset to peak of this wave ( Figure 1 ). The IVA time was measured from the end of the myocardial velocities during late diastole to the beginning of the myocardial velocities during systole.

Figure 1

Typical tissue Doppler tracing obtained at the base of the RV. Wave during IVA precedes the S wave and begins before the peak of the R wave on electrocardiogram. IVA was measured by dividing peak velocity by the time interval from onset of the wave (zero-crossing) during IVA to the time at peak velocity of this wave.

Persistent (residual) pulmonary valvar stenosis was considered present when the maximal systolic gradient at the RV outflow tract was greater than 30 mm Hg. Residual pulmonary insufficiency was considered mild when the regurgitant jet was detected near the valve in the pulmonary artery and holodiastolic in duration, moderate when the regurgitant jet was detected up to the medial portion of the pulmonary trunk, and severe when the jet flow was protodiastolic and present up to the pulmonary branches.

All measurements were performed three times, and the mean value was used for analysis. Echocardiographic studies were performed by the same investigator before surgery, within the first 72 hours after surgery, and 30 to 90 days after surgery.

Histologic and Morphometric Analyses

Tissue samples were obtained for histology from the RV infundibular muscle bands resected as part of the surgical relief of the subpulmonary obstruction (23 cases), as endomyocardial biopsies obtained during surgery from the RV inlet (22 cases) and, when technically feasible, from the free wall of the LV through the ventricular septal defect before its closure (20 cases). Tissue was processed routinely for histology, and 5-μm–thick sections were stained with hematoxylin–eosin and Sirius red for collagen quantification and submitted to immunohistochemistry against factor VIII antigen to label endothelial cells of capillaries.

The morphometric measurements were carried out with the aid of an interactive computer-assisted image analyzer (Leica Quantimet; Leica Cambridge, Cambridge, UK). To avoid interobserver variation, a single investigator operated the analyzer.

Cardiomyocyte Diameter

The smallest transverse cardiomyocyte diameter was measured at the level of the nucleus at a final magnification of 400×. The number of required measurements was determined after analyzing the evolution of the mean values and variance in 20, 40, 60, 80, 100, 120, 140, 180, and 200 observations. Finally, the option was made to measure 60 cardiomyocytes per section, in a minimum of 10 microscopic fields. The results obtained were compared with normal values for age published previously. Mean values of cardiomyocyte diameter greater than the normal value for age plus one standard deviation (SD) were considered as histologic hypertrophy.

Quantification of Interstitial Collagen Content

Collagen content in the interstitial space was estimated by analyzing at least 10 microscopic fields in tissue samples stained with Sirius red at a magnification of 200× (23 samples for the RV infundibulum, 22 samples for the RV inlet, and 11 samples for the LV). The collagen volume fraction was expressed as a percentage of positively stained area relative to the total area of myocardium. Star-shaped microscars (replacement fibrosis) and perivascular fibrous tissue from arterioles more than 30 μm in diameter were excluded from the analysis. The results were compared with values published in the literature.

Capillary Volume Fraction

After immunohistochemical labeling of capillary endothelial cells, at least five microscopic fields were analyzed at a final magnification of 400× for each section, from the three regions sampled (RV infundibulum, inlet, and LV), using a computer-assisted morphometry and an interposed grid of 80 points. Incident points on myocardial capillaries and cardiomyocytes were counted. The capillary volume fraction was expressed as a proportion to the area fraction of cardiomyocytes in the microscopic field.

Statistical Analysis

Values are given as means, SDs, and medians. Student’s t test was used to compare two groups with normally distributed variables; otherwise, a Mann–Whitney U test was performed. Analysis of variance for repeated measures was used for multiple comparisons. Fisher’s exact and chi-square tests were used to assess association between independent and outcome variables. Correlation coefficients were obtained by the Spearman method. Logistic regression analysis was used to test the possible role of morphometric histologic features in the prediction of postoperative echocardiographic outcomes. Statistical significance was inferred at a P value less than .05.


Clinical and Surgical Features

All corrective surgeries were performed using cardiopulmonary bypass and deep hypothermia. Both the transatrial and transpulmonary approaches were applied in five patients. A transannular patch was used in 10 patients. Six patients had patched widening of the RV outflow tract with no need for a transannular patch. One patient required the insertion of a Dacron conduit between the RV and the pulmonary trunk. The mean bypass time (±SD) was 119.3 ± 37.1 minutes, and the mean aortic crossclamping time (±SD) was 87.9 ± 27.5 minutes. The median time for intensive care unit discharge was 72 hours, with need for assisted ventilation for a median of 6 hours (range 1–135 hours) and for inotropic support for a median of 60 hours (range 15–168 hours).

Histologic Findings and Morphometric Analysis

Moderate cellular vacuolization (myocytolysis) was found in subendocardial cardiomyocytes from the RV inlet and infundibulum in 11% and 8.5% of cases, respectively. All samples from the resected infundibular muscular bands showed variable degrees of fibrous endocardial thickening.

The morphometric data are summarized in Table 1 . The transverse cardiomyocyte diameters from the RV samples were increased when compared with normal values ( Figure 2 ) and with the cardiomyocytes from the LV ( P = .004). All but two cases presented a mean cardiomyocyte diameter greater than normal for age plus one SD (cellular hypertrophy). There was a significant positive correlation between the mean cardiomyocyte diameter from both the inlet and infundibular samples of the RV and age ( r = 0.59, P = .003 and r = 0.76, P < .001, respectively).

Table 1

Interstitial collagen (area fraction in %), myocyte diameter, and capillary area fraction in right and left ventricles

N Mean ± SD Median
P value
Interstitial collagen (%) .761
RV inlet 22 5.95 ± 2.71 5.68
RV infundibulum 23 6.05 ± 1.89 6.04
LV free wall 19 5.93 ± 3.37 4.87
Cardiomyocyte diameter (μm) .004
RV inlet 22 11.36 ± 1.45 11.39
RV infundibulum 23 11.57 ± 2.46 11.62
LV free wall 20 10.08 ± 1.54 10.13
Capillarity (%/mm 2 ) .997
RV inlet 19 0.16 ± 0.03 0.16
RV infundibulum 21 0.16 ± 0.06 0.15
LV free wall 19 0.16 ± 0.04 0.16

SD, Standard deviation; N, number of cases available for analysis; RV, right ventricle; LV, left ventricle.

Analysis of variance, RV inlet, and RV infundibulum cardiomyocyte diameters larger than LV cardiomyocyte diameters.

Figure 2

Cardiomyocyte diameters (μm) from the RV infundibulum ( circles ) and RV inlet ( triangles ), plotted against the normal values for age (mean ± SD)

(modified from Nishikawa and colleagues ).

The diameters of cardiomyocytes from the RV inlet and infundibulum were significantly greater ( P = .007 and .018, respectively) in the eight patients not receiving oral propranolol before surgery than in those receiving it continuously for a mean time of 12 months (SD = 20 months) with an average dosage of 1.84 g/kg/d (±0.48).

The interstitial collagen content was increased in both the RV and LV when compared with reference values. There were no differences when comparing samples from the three regions.

There was no significant correlation between the mean interstitial collagen content from both the RV inlet and infundibulum and the age at surgery, or the RV outflow gradient, hematocrit, and oxygen arterial saturation at rest. The capillary volume fraction did not differ among the various regions analyzed. In addition, we found a significant negative correlation between capillary volume fraction in the RV infundibulum and the arterial oxygen saturation at rest ( r = −0.448, P = .042).


Mean QRS duration after surgery was significantly increased compared with preoperative values (100 ± 30 msec vs. 70 ± 30 msec, P = .002; range 40–140 msec in the postoperative evaluation), as well as the QTc interval (460 ± 40 msec vs. 420 ± 30 msec, P = .0002). Mean increase in QRS duration was 28.91 ± 31.26 msec. Forty-eight percent of the patients showed an increase greater than 40 msec. This subset of patients had greater content of interstitial collagen on histology than those with an increase in QRS duration less than 40 msec ( P = .029). The JT interval decreased significantly compared with preoperative values ( P = .003).


Residual Lesions

Residual lesions were common after surgery. All patients had some degree of pulmonary regurgitation early after surgery, being moderate to severe in 60%. Residual pulmonary stenosis (gradient > 30 mm Hg) was less frequent (23%). Mild tricuspid insufficiency was also present in the majority of patients immediately after surgery.

There was a positive association between the use of a transannular patch for enlargement of the RV outflow tract and the presence of moderate or severe pulmonary regurgitation in the postoperative period ( P = .041 and P < .001 in the immediate and medium-term echocardiographic evaluations, respectively).

Right Ventricular Systolic Function

Myocardial IVA, as assessed by TDI, was abnormal in 35% of the patients before surgery. Mean IVA values did not change immediately after surgery, but decreased significantly in the 3-month follow-up assessment ( Table 2 ) and showed a weak negative correlation with the diameter of the RV cardiomyocytes from the inflow tract ( r s = −0.59; P = .06). The RV systolic function available by peak myocardial velocity during systole (S’) decreased in all patients immediately after surgery and remained reduced when compared with preoperative values 3 months after surgery ( Figures 3 and 4 ). A significant correlation was found between the time of use of vasoactive drugs in the early postoperative period and S’ values in the same period ( r s = 0.48; P = .02).

Table 2

Pulsed-wave tissue Doppler imaging velocities on right and left ventricles

Before surgery Immediately after surgery Late after surgery P value
Right ventricle
S’ (cm/s) 10.82 ± 2.74 5.78 ± 1.24 7.32 ± 1.65 <.001
E’ (cm/s) 13.88 ± 3.66 6.36 ± 1.88 9.61 ± 2.63 <.001
A’ (cm/s) 11.42 ± 3.39 6.41 ± 2.78 6.78 ± 2.44 <.0001
IVA (cm/s 2 ) 206.87 ± 83.70 205.58 ± 110.17 130.23 ± 50.91 .006
Left ventricle
S’ (cm/s) 6.90 ± 2.30 6.75 ± 2.02 6.94 ± 1.41 .61
E’ (cm/s) 11.17 ± 3.28 10.25 ± 2.90 12.35 ± 3.09 .09
A’ (cm/s) 11.17 ± 2.32 10.24 ± 2.89 12.35 ± 3.08 .08
IVA (cm/s 2 ) 188.58 ± 92.28 185.76 ± 60.01 127.66 ± 83.37 .047

Only gold members can continue reading. Log In or Register to continue

Jun 16, 2018 | Posted by in CARDIOLOGY | Comments Off on The Impact of Preexisting Myocardial Remodeling on Ventricular Function Early after Tetralogy of Fallot Repair

Full access? Get Clinical Tree

Get Clinical Tree app for offline access