Echocardiographic assessment of discrete subaortic stenosis in childhood

An echocardiographic index for decompensation of the chronically volume-overloaded left ventricle in children

AIMS

The criterions for the timing of surgical intervention in children with rheumatic mitral or aortic valvar regurgitation are not defined. I hypothesized that, in children with chronic mitral or aortic regurgitation, an index for decompensation could be created by using the ratio of the diastolic left ventricular wall thickness to the radius, and that such an index could prove useful in determining the optimal time for surgical intervention.

METHODS

The left ventricular echocardiograms were obtained at the tips of the leaflets of the mitral valve by M-mode echocardiography. The diastolic septal wall thickness was measured between the right and left ventricular endocardial layers, and the posterior wall thickness between the endocardium and the interphase between the epicardium and the myocardium. The left ventricular diastolic dimension was then measured, between the posterior and septal wall endocardial layers, and systolic dimension as the smallest distance detected between these layers. All diastolic measurements were made at the time of the R wave of electrocardiogram, using the leading edge technique. The ratio of wall thickness was measured using the mean of septal and posterior wall thicknesses divided by half the diastolic dimensions, the normalized thickness of the wall previously referred to as the h/r ratio and relative mural thickness.

RESULTS

The ratio of wall thickness to left ventricular radius, and its relation to systolic left ventricular pressure or systolic blood pressure, was found to be linear in 89 normal school children, and 39 children with aortic stenosis. For future predictions, I calculated the 95th percentile limits and the 95th percentile confidence bands for this relation. Using the same data, it proved possible to calculate ratios of wall thickness for various ranges of either systolic blood pressure or left ventricular peak pressure. By using the normal limits of 0.356 plus or minus 0.0316 of the ratio, appropriate for the systolic blood pressure of children with mitral regurgitation, I determined the adequacy of the ratio of wall thickness. Of the children, 51 were in ventricular failure, and these had an inadequate ratio, below two standard deviation. Of the others, 21 had an inadequate ratio to within minus one to minus two standard deviations, and 12 of these were asymptomatic, 8 were symptomatic, but only one was in ventricular failure. For 18 children with aortic regurgitation, using the same limits, one child was within 1 standard deviation and was asymptomatic, 8 fell within minus 1 to minus 2 standard deviations and 2 of these were symptomatic, 5 were in ventricular failure, and 1 was asymptomatic, while the other 9 had ratios falling less than minus 2 standard deviations, and all were in ventricular failure.

CONCLUSION

I conclude that the index of normalized wall thickness defined as the ratio of the left ventricular wall thickness to its radius is adequate, and within normal limits, when there is compensated volume overload, but is inadequate and below normal limits when the volume overloaded left ventricle becomes decompensated. My data suggests that the persistently decreasing ratio of wall thickness below the limits of normality serves as an indicator of ventricular decompensation, and thus can be used as a new criterion for determining the optimal time for surgical intervention.

Use of three-dimensional echocardiography for analysis of outflow obstruction in congenital heart disease

To evaluate the feasibility and accuracy of 3-dimensional (3D) echocardiography in analysis of left and right ventricular outflow tract (LVOT and RVOT) obstruction, 3D echocardiography was performed in 28 patients (age 4 months to 36 years) with outflow tract pathology. Type of lesion and relation to valves were assessed. Length and degree of obstruction were measured. Three-D data sets were adequate for reconstruction in 25 of 28 patients; 47 reconstructions were made. In 13 patients with LVOT obstruction, 3D echocardiography was used to study subvalvular details in 8, valvular in 13, and supravalvular in 1. Four of these 13 patients had complex subaortic obstruction. In 12 patients with RVOT lesions, 3D echocardiography was used to study subvalvular details in 11, valvular in 12, and supravalvular in 2. Three-dimensional reconstructions were suitable for analysis in 100% of subvalvular LVOT, 77% valvular LVOT, 100% supravalvular LVOT, 100% subvalvular RVOT, 50% valvular RVOT, and 50% supravalvular RVOT. Twenty patients underwent operation, and surgical findings served as morphologic control for thirty-four 3D reconstructions (LVOT 17, RVOT 17). Operative findings revealed an accuracy at subvalvular LVOT of 100%, valvular LVOT 90%, supravalvular LVOT 100%, subvalvular RVOT 100%, valvular RVOT 100%, and supravalvular RVOT 100%. Quantitative measurements could adequately be performed. Three-D echocardiography is feasible and accurate for analyzing both outflow tracts of the heart. Particularly, generation of nonconventional horizontal cross sections allows a good definition of extension and severity of lesions.

Echocardiographic assessment of subvalvular aortic stenosis before and after operation

The development of two-dimensional and Doppler echocardiography has provided a noninvasive technique for the diagnosis and serial assessment of patients with subvalvular aortic stenosis. The clinical records and echocardiographic data were reviewed of all patients with subaortic stenosis diagnosed between 1983 and 1991. Of the 77 patients identified (45 male and 32 female), 28 had isolated subaortic stenosis and 49 had associated cardiac lesions. The most frequently encountered associated lesions were ventricular septal defect (n = 19) and coarctation of the aorta/interrupted aortic arch (n = 14). Serial echocardiographic studies, performed in 38 of the 77 patients, documented significant progression of the left ventricular outflow tract gradient in 25 patients (66%) and development of aortic regurgitation in 25 patients (66%). Surgical resection was performed in 36 patients. The preoperative outflow tract peak gradient was 62.9 +/- 31 mm Hg (range 0 to 153), whereas the immediate postoperative gradient was 14.4 +/- 14 mm Hg (range 0 to 67). The two patients with a significant residual gradient (37 and 67 mm Hg, respectively) in the immediate postoperative period had severe subaortic stenosis preoperatively with marked left ventricular hypertrophy and intracavitary gradient. The immediate postoperative echocardiograms demonstrated no worsening of aortic regurgitation in any patient and regression of regurgitation in one patient from mild to none. Intermediate-term follow-up studies were available for review in 13 postoperative patients at a mean of 4 years postoperatively. In 2(15%) of these 13 patients, subaortic stenosis recurred; however, the degree of aortic regurgitation did not increase in any patient.(ABSTRACT TRUNCATED AT 250 WORDS)

Late results after resection of fixed subaortic stenosis

Resection of fixed subaortic stenosis was performed on 44 patients with median age 14 (range 2-61) years. Concomitant aortic valve pathology was present in 14 (32%) cases (congenital stenosis in 2, thick fibrotic cusps in 8 and incompetent cusps in 4) and other congenital cardiovascular malformations in eight (18%). There was no perioperative mortality. Of the six late deaths, three were due to non-cardiac causes. During follow-up (median 6, range 2-21 years), six reoperations were performed for residual or recurrent obstruction and/or aortic incompetence. Aortic valve replacement was required at two primary and four second operations. Actuarial 5-year and 10-year survival rates were 89% and 76%, respectively, and rates with freedom from cardiac death endocarditis and reoperation 83% and 64%. At follow-up evaluation two patients had significant aortic regurgitation and all survivors had a systolic ejection murmur. At Doppler echocardiography in 29 patients without reoperation, the median pressure difference in the left ventricular outflow tract was 10 (range 0-55) mmHg--in three cases greater than or equal to 30 mmHg. Careful follow-up is advisable after resection of fixed subaortic stenosis, because of the risk of residual or recurrent obstruction and of significant aortic valve incompetence.

Discrete sub-aortic stenosis and ventricular septal defect

We present details of 15 children, aged 3 months to 11 years, with discrete sub-aortic stenosis and ventricular septal defect. We emphasise a high index of clinical suspicion and echocardiography as the best means of diagnosing this dangerous combination. Physical signs were those of ventricular septal defect in all patients, with auscultatory evidence of additional sub-aortic stenosis in only one. Five patients had a short early diastolic murmur of mild aortic incompetence. The electrocardiograph showed isolated left ventricular hypertrophy in eight patients. Cardiac catheterisation and angiography identified the ventricular septal defect in all cases but detected the sub-aortic stenosis in only eight. Cross sectional echocardiography showed both lesions in all 11 patients to whom it was available.

Combined discrete subaortic stenosis and ventricular septal defect in infants and children

Eight patients, aged 1 to 8 years, with discrete subaortic stenosis (DSS) and ventricular septal defect (VSD) were studied by 2-dimensional (2-D) and M-mode echocardiography. Initial cardiac catheterization and angiocardiography showed VSD and other associated cardiac lesions, including coarctation of the aorta and patent ductus arteriosus. None had evidence of DSS. Six patients underwent surgical repair of the associated lesions, but none required closure of the VSD. Ultimately, 6 patients had spontaneous closure of VSD, and 2 had a residual small VSD. Subsequent serial echocardiography showed evidence of subaortic membrane, prompting repeat cardiac catheterization, which confirmed moderate to severe peak systolic pressure gradients between the left ventricle and ascending aorta. Surgical resection of the membrane was performed in 5 patients. Thus, in patients with small or spontaneously closed VSDs, DSS may develop. Evaluation of the left ventricular outflow tract area is recommended in patients with small or closed VSD in whom a significant heart murmur or electrocardiographic abnormality remains.

Two-dimensional echocardiography in discrete subaortic stenosis

Thirty-seven patients with discrete subaortic stenosis (DSS) underwent 2-dimensional echocardiography (2-D echo) and cardiac catheterization. The peak systolic pressure gradients ranged from 0 to 150 mm Hg. Thirty-two patients had membranous DSS and 5 had fibromuscular DSS. Of 37 patients with DSS, 2-D echo diagnosed the presence and type in 35; in 2, a membrane was demonstrated by angiography. Of the 35 patients accurately diagnosed by 2-D echo, angiography corroborated the diagnosis in 33, but failed to show the membrane in 2. Subsequent cardiac surgery confirmed the accuracy of the echocardiographic diagnosis in these 2 patients. In all patients with membranous DSS, the anterior insertion of the membrane was demonstrated. In 9 of them the posterior insertion was demonstrated by tilt of the transducer but the anterior insertion disappeared. In 4 patients both insertions were demonstrated simultaneously and in 3 patients the membrane was demonstrated as a continuous line. In 4 of the 5 patients with fibromuscular DSS, both insertions of the lesion were demonstrated simultaneously. However, 2-D echo was unsuccessful in assessing the severity of obstruction. In only 1 patient did demonstration of the whole subaortic membrane as a continuous line below the aortic valve correlate with severe obstruction. Thus, the presence and type of DSS, but not the degree and severity, can be accurately and reliably diagnosed by means of 2-D echo.

Improved prediction of peak left ventricular pressure by echocardiography in children with aortic stenosis

Prediction of peak left ventricular pressure by echocardiography in children with aortic stenosis assumes that wall stress is normal. The recent finding that stress is subnormal in many children with aortic stenosis and elevated ejection performance requires reevaluation of this noninvasive technique. By using M-mode echocardiography, left ventricular end-diastolic dimension and wall thickness and left ventricular shortening fraction were measured in 27 children with aortic stenosis undergoing left ventricular pressure measurement by cardiac catheterization. Similar echocardiographic measurements and systolic blood pressure determinations by the cuff method were obtained from 29 normal children. Peak circumferential wall stress and shortening fraction were calculated from the echocardiographic and pressure data. It was found that stress was inversely proportional to shortening fraction for all patients with aortic stenosis (p less than 0.001, r = -0.86). In a subgroup of patients with a shortening fraction of less than 0.40, stress was 262 +/- 20 mm Hg, similar to 280 +/- 30 mm Hg in the normal group but greater than 205 +/- 27 mm Hg in patients with a shortening fraction of 0.40 or greater (p less than 0.001). In patients with aortic stenosis, the ratio of left ventricular end-diastolic wall thickness to cavity dimension predicted peak left ventricular pressure moderately well (r = 0.83, standard error of the estimate [SEE] = 23). The stress-shortening fraction relation was used to estimate stress and correct this ratio in patients with diminished stress and a shortening fraction greater than or equal to 0.40. This yielded a significantly improved correlation (r = 0.93, SEE = 15, p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical presentation and natural history of mild discrete subaortic stenosis. Follow-up of 1--17 years

We report 21 patients with discrete subaortic stenosis (DSS) causing mild obstruction with a peak systolic left ventricular outflow pressure gradients less than 50 mm Hg. They were followed 1--17 years (mean 6.5 years), and eight were recatheterized before surgery, 2--17 years after the first cardiac catheterization. Three patients (14%) had subacute bacterial endocarditis. Ten (48%) had aortic insufficiency, one of whom had no pressure gradient across the left ventricular outflow tract. In three of the 10 patients, aortic insufficiency was found only at the second catheterization. Nine patients (43%) had hyperactive, asymmetric left ventricular contraction; in three, this finding was present only at the second catheterization. Seven of the eight patients who were recatheterized (33% of the entire group) showed an increase in gradient. The increase was from a mean gradient of 35.2 mm Hg to 76.7 mm Hg. Seventeen patients (81%) had at least one of these four features. In view of these data, we suggest that surgical indications for DSS might be expanded, although definitive recommendations are not possible. All cases of DSS should be carefully followed. Surgery should be performed if signs of progressive complications develop.

Discrete subaortic stenosis

Data concerning 17 consecutive patients with discrete subaortic stenosis are recorded. Twelve patients underwent operative resection of the obstructing lesion. Of these all except one were symptomatic and all had electrocardiographic evidence of left ventricular hypertrophy or left ventricular hypertrophy with strain. They had a peak resting systolic left ventricular outflow tract gradient of greater than 50 mmHg as predicted from the combined cuff measurement of systolic blood pressure and the echocardiographically estimated left ventricular systolic pressure and/or as determined by cardiac catheterisation. The outflow tract gradient as predicted from M-mode echocardiography and peak systolic pressure showed close correlation with that measured at cardiac catheterisation or operation. During the postoperative follow-up from one month to 11 years, of 11 patients, one patient required a further operation for recurrence of the obstruction four years after the initial operation. All patients are now asymptomatic. Five patients have not had an operation. The left ventricular outflow tract gradient as assessed at the time of cardiac catheterisation was greater than 50 mmHg. One patient has been lost to follow-up. The remaining four have been followed from four to eight years and have remained asymptomatic and the electrocardiograms have remained unchanged. Careful follow-up of all patients is essential with continuing clinical assessment, electrocardiograms, M-mode and two-dimensional echocardiograms, and if necessary cardiac catheterisation. Prophylaxis against bacterial endocarditis is also essential.

Noninvasive assessment and differentiation of left ventricular outflow obstruction with Doppler ultrasound

Blood flow velocities in the left ventricle and the ascending aorta were recorded noninvasively with Doppler ultrasound. The ultrasound beam was aligned as much as possible to the direction of velocity, using the frequency shift in the audio signal as a guide to obtain velocities as close as possible to those present. From the maximal velocity recorded by continuous-wave Doppler, a peak pressure drop was calculated in 24 patients with aortic valve stenosis and nine with fixed subaortic stenosis. Fourteen patients with aortic stenosis and three with fixed subaortic stenosis were catheterized. In these patients, the correlation between calculated pressure drops and those obtained by pressure recording was good (r = 0.85). The pressure drop can be underestimated by underestimating velocity, but cannot be overestimated. With pulsed Doppler, the level of obstruction can be determined.

Two-dimensional echocardiographic evaluation of discrete subaortic stenosis from the apical long axis view

M-mode standard two-dimensional (2DE) left parasternal long axis echocardiographic examination of the left ventricular outflow tract (LVOT) was evaluated exclusively with respect to its utility in identifying discrete subaortic stenosis (SUB-AS). Important details of the anatomy of the subaortic area may also be obtained from 2DE apical long axis imaging. According, 18 patients with discrete SUB-AS were prospectively evaluated by M-mode and 2DE. The M-mode findings included narrowing of the LVOT and early systolic closure of the aortic valve. However, these findings were variable and highly dependent upon scan speed, fluid flow dynamics, and beam angulation. 2DE findings varied using the standard long axis view at the left parasternal border, depending upon the type of obstruction present. A discrete membrane produced linear echoes adjacent and parallel to the interventricular septum beneath the aortic valve. Fibromuscular obstruction produced a localized dense ridge of echoes in the LVOT. These findings were not apparent in five patients studied. In these patients, the 2DE apical long axis view was employed to image the subaortic area. From this tomographic corss-section a fibrous membrane was imaged as a linear echo parallel to the aortic valve. The membrane extended across the LVOT from the ventricular septum to the anterior leaflet of the mitral valve. The 2DE apical long axis view therefore provides an additional approach in the evaluation of patients with discrete SUB-AS.