Apical Muscular Ventricular Septal Defects Between the Left Ventricle and the Right Ventricular Infundibulum
Diagnostic and Interventional Considerations
Background Effective transcatheter or surgical closure of apical muscular ventricular septal defects (VSDs) requires accurate delineation of variable and often complex anatomy. These defects have generally been considered as communications between the apexes of both left and right ventricles.
Methods and Results Among 50 consecutive patients with multiple muscular VSDs referred for transcatheter device closure between October 1987 and April 1993, a subset of 10 patients (aged 7 days to 28 years) with apical muscular VSDs shared a unique set of anatomic characteristics: (1) large and often single opening in the left ventricle; (2) multiple right ventricular openings in the anterior aspect of the apical septum; and (3) separation of the right ventricular apical region into which the VSDs open from the rest of the right ventricular inflow and outflow by prominent muscle bundles. Additional analysis of the anatomy by use of echocardiography and cineangiography showed that these muscular defects were between the left ventricular apex and right ventricular infundibular apex. In 6 patients, the transcatheter devices used to create a septum in these hearts were placed in the right ventricle, straddling muscle bundles that separated the apical VSD from the rest of the right ventricular inflow and outflow, resulting in incorporation of a portion of the right ventricular infundibular apex into the physiological left ventricle. Three patients had devices placed between the apexes of the left ventricle and the infundibulum. The defect closed spontaneously within the right ventricle in 1 patient. One patient died after surgery for tetralogy of Fallot in situs inversus. The remaining 9 patients were all clinically well at the time of their most recent follow-up visit (follow-up duration, 32±11 months). This distinct type of apical VSD was identified by echocardiography in 20 of 274 patients who were followed up clinically for muscular VSDs.
Conclusions Left ventricular–infundibular apical VSDs constitute a distinct morphological type of muscular VSD that can be distinguished by echocardiography and cineangiography. In selected cases, the infundibular apex can be separated from the rest of the right ventricular inflow and outflow to eliminate flow across these defects.
The anatomy of apical muscular ventricular septal defects (VSDs) is often complex, with several openings on either side of the septum and an oblique and tortuous intraseptal course.1 2 Although many of these defects are small and close spontaneously,3 the larger defects often persist to cause significant shunt and right ventricular (RV) hypertension. Large apical muscular VSDs complicate management decisions, particularly when they occur in association with other congenital cardiac defects.2 4 The results of surgery for apical muscular VSDs are often suboptimal owing to difficulties in defect visualization, residual shunting, and ventricular dysfunction.2 5 6 Transcatheter device closure has recently emerged as a therapeutic option for patients with muscular VSDs.4 7 Accurate delineation of anatomic details of the apical region, the number and location of defects, their origin and exit on the left ventricular (LV) and RV aspects, and their relationships with neighboring muscle bundles or valvular apparatus are important for effective transcatheter therapy.
Apical VSDs have generally been considered as defects between apexes of both the LV and the RV.1 5 It has not been well appreciated, however, that what is generally regarded as “RV apex” is composed of two distinct apexes that are separated by multiple dense trabeculae coursing from the interventricular septum toward the anterior free wall. Posterior and to the right of this plane of trabeculae is the apex of the RV inflow, and anterior and leftward is the apex of the RV infundibulum. Among the patients who underwent transcatheter device closure for muscular VSDs at Children's Hospital, Boston, Mass, a subset of patients with apical muscular VSDs who had distinct echocardiographic and cineangiographic features was identified. The VSDs resulted from a defect in the septum separating the RV infundibular apex from the LV apex. This report describes the echocardiographic and cineangiographic features of this subset of patients, the specific strategies used for transcatheter device closure, and the incidence of this type of VSD in a consecutive group of patients followed up clinically for muscular VSDs.
The echocardiograms, cineangiograms, and clinical records of all patients undergoing transcatheter device closure for muscular VSDs between October 1987 and April 1993 were reviewed. Patients with apical muscular defects who had the following distinct characteristics were identified: (1) single or multiple openings in the anterior aspect of the RV septum near or at the apex; and (2) presence of prominent RV apical muscle bundles resulting in separation of the RV infundibular apex (into which the VSDs open) from the rest of the RV cavity.
Two-dimensional transthoracic echocardiograms were performed on all patients before and after cardiac catheterization with pulsed and color Doppler according to a previously described protocol.8 Images of the ventricular septum were obtained from subxiphoid, apical, and parasternal views. Particular attention was given to the location of the defect(s) on the RV and LV aspects, the number of defects, their relationship with neighboring structures, and the flow pattern and velocity. In addition, transesophageal echocardiograms were performed in patients with suboptimal transthoracic windows and during transcatheter closure to guide device placement. The methodology for transesophageal echocardiographic guidance during transcatheter device closure of muscular VSDs has been reported previously.9 10
To determine the frequency of this particular type of apical VSD in an unselected population, all echocardiograms of patients diagnosed with muscular septal defect(s) from January 1995 through March 1996 at Children's Hospital (n=274) were reviewed.
Hemodynamic and Cineangiographic Evaluation
Catheterization of the right and left sides of the heart was performed in each patient, and the size, location, and configuration of suspected or previously identified defects were defined angiographically. Multiple angled views profiled the defects, displaying their geometry and relation to surrounding structures.7
Transcatheter device closure was considered if it was believed that the apical defect was hemodynamically significant and operative closure would be difficult in view of the location of the defect or condition of the patient. The decision to attempt a transcatheter closure was made jointly by the patients' cardiologists and the cardiac surgeons. Informed consent was obtained in all cases, and device placements were performed under protocols approved by the US Food and Drug Administration and the Committee on Clinical Investigation at Children's Hospital or were done as emergency procedures. The technique of device placement for apical muscular VSD has been described previously.4 7 11 The decision on where to place the device depended on an arbitrary judgment as to where closure might best be accomplished. Thus, if the musculature separating the infundibular apex from the rest of the RV inflow and outflow was perforated by only one or two discrete openings and the apical septum was highly deficient, the device(s) was placed within the RV straddling the infundibular trabeculae. After device placement, residual shunts were assessed qualitatively by angiography and two-dimensional and color Doppler echocardiography. The ratio of pulmonary to systemic blood flow was not used to quantify residual shunt because of the common occurrence of associated defects. It was believed that accurate and reproducible measurements of the size of the apical VSDs could not be accomplished because the complex three-dimensional anatomy of these defects made it difficult to identify and standardize imaging planes for echocardiographic or angiographic measurements.
Apex of the right ventricular inflow:
the most apical part of the RV inflow cavity (RV sinus) located behind and below the moderator band (Fig 1A⇑).
Infundibular apex (also known as the infundibular apical recess):
Between October 1987 and April 1993, 10 patients (age, 7 days to 28 years; median, 2 years) were identified as having apical muscular VSDs between the LV apex and the infundibular apex. Seven patients had additional congenital cardiac defects requiring surgery (Table 1⇓). These included D-transposition of the great arteries (n=3), tetralogy of Fallot in situs inversus (n=1), and mitral stenosis (n=1). Two patients were identical twins with Holt-Oram syndrome, each of whom had an atrial septal defect and a large conoventricular septal defect. Nine patients had muscular defects in other parts of the ventricular septum in addition to the apical defects (Table 1⇓). Previous palliative operations (pulmonary artery band, n=7; modified Blalock-Taussig shunts, n=1) were performed in 8 patients, and 1 patient had undergone the arterial switch operation for D-transposition of the great arteries. One patient had previously undergone balloon mitral valvuloplasty. In all but 1 patient, transcatheter closure was performed before definitive surgical repair as part of an integrated surgical-transcatheter approach to repair the complex congenital heart disease (Table 2⇓). Because most patients had either undergone a prior pulmonary artery banding or had obstruction to pulmonary blood flow as part of their congenital heart disease, transcatheter closure was chosen primarily on the basis of anatomic and physiological considerations rather than the patient's preoperative clinical symptomatic state.
The anatomic features of the apical defects were best delineated from the subxiphoid and parasternal short-axis and apical views. In addition, six patients underwent transesophageal echocardiography during cardiac catheterization to assist in device placement. The following anatomic features were identified by echocardiography as characteristics of LV-infundibular apical defects:
1. LV aspect: a single opening was seen in six patients. Multiple anterior LV openings were demonstrated in one patient (patient 1) who died after surgery. In three patients (patients 3, 7, and 8), defects were located both in the anterior (infundibular-apical septum) as well as the posterior parts of the apical septum (septum between RV inflow apex and the LV apex).
2. RV aspect: multiple openings were seen in the infundibular apical region in all patients. The moderator band and the more apical trabeculae were hypertrophied, unusually dense, and formed a barrier that separated the infundibular apex and the defect(s) from the rest of the RV cavity. This unusually dense and hypertrophied band created the appearance of a separate “septum” within the RV (Figs 2A⇓ and 3⇓).
Prevalence of LV-Infundibular Apical VSD
Of the 274 consecutive patients with muscular VSD diagnosed by echocardiography, 47 patients (mean age, 1.4 years; range, 0 to 13.1 years) had one or more defects in the apical region. Of these 47 patients, 20 (43%) had an LV-infundibular defect (Fig 2B⇑) and 21 (45%) had a defect between the apexes of the LV and RV sinuses (Fig 2C⇑). In 2 patients, there were defects between the LV apex and both the infundibular apex and the RV sinus apex. In 4 patients, the images were inconclusive. Among the 20 patients with LV-infundibular VSD, the defect was small in 13 patients (65%) and either closed spontaneously or was believed to be hemodynamically insignificant during follow-up. Of the 7 patients with a moderate or large defect, diminution of the size of the defect was found by echocardiography in 5 patients, operative closure in 1 patient, and a modified Fontan operation for single-ventricle physiology in 1 patient.
The cineangiographic features were best outlined through contrast injections in the region of the VSD (using a cutoff pigtail catheter mounted over a wire placed across the VSD). The above-described echocardiographic features were best demonstrated in steep, cranially angled, left anterior oblique views (Fig 3⇓). The precise angles for angiographic delineation of these defects were arrived at by trial and error and differed for each patient. In addition, cineangiograms demonstrated that the region of prominent muscle bundles separating the infundibular apex and defect(s) from the rest of the RV inflow and outflow was often narrower than the defect in the infundibular apical septum.
The results of transcatheter device closure are summarized in Table 2⇑. A single device was used to close the apical muscular defects in four patients (patients 1, 4, 5, and 8); two devices were used in four patients (patients 2, 3, 7, and 10), and three devices were used in one patient (patient 9). Six patients (patients 3, 5, 6, 7, 9, and 10) required device placement in muscular VSDs at other locations (anterior and posterior midmuscular). In six patients (patients 1, 2, 4, 5, 7, and 10), the device was placed in the RV at the narrowest part of the defect straddling the muscle bundles that separated the infundibular apex and the VSD from the rest of the RV inflow and outflow. This resulted in incorporation of the infundibular apex as a part of the physiological left ventricle (Figs 3 and 4⇓⇓). The device was placed between the apexes of the LV and the infundibulum in two patients (patients 3 and 8) (Fig 5⇓). Additional devices were placed within the infundibular apex in four patients (patients 2, 3, 7, and 9). One patient (patient 6) had spontaneous closure of a moderately sized interventricular communication documented by serial cineangiograms (Fig 3C and 3D⇓⇓). The site of spontaneous closure was within the RV at the level of the moderator band, resulting in the physiological incorporation of the RV infundibular apex into the LV.
Immediately after transcatheter device placement, angiographic evidence of a small residual defect was seen in eight patients, and two patients had a moderate residual defect. Eight patients underwent surgery after device placement that included a Rastelli-type repair (n=2), pulmonary arterial band takedown (alone [n=3] or in combination with arterial switch operation [n=1]), or closure of a conoventricular septal defect (n=2). One patient (patient 6) underwent balloon dilation of supravalvar pulmonary stenosis after an arterial switch operation (Table 2⇑).
One patient (patient 1) died after attempted repair of tetralogy of Fallot in situs inversus after device placement. The remaining nine patients were all clinically well at the time of the last follow-up visit (32±11 months). The most recent follow-up echocardiograms showed that the residual interventricular communications at the device margins were small in all patients (Table 2⇑).
Although a significant proportion of muscular VSDs are located in the apical region, little published work is available regarding the variant anatomy of apical muscular VSDs. Despite their complexity, apical muscular VSDs have been considered to be one category of lesions; anatomic heterogeneity within this class of defects is not generally appreciated.1 5 12 Much of the anatomic complexity of the RV apex results from numerous trabeculae that traverse its cavity from the interventricular septum toward the anterior free wall. The moderator band is the largest and the most superior of these trabeculae. These trabeculae are aligned, to a large extent, along a plane that separates the apex of the RV sinus (or inflow) from the apex of the infundibulum (Fig 1⇑). The infundibular apex is located anterosuperior and leftward of the apex of the RV sinus and constitutes the lowermost part of the infundibular chamber (Fig 1A and 1B⇑⇑). Viewed from above, it has the shape of an inverted cone (Fig 1C⇑). Its anterior free wall is continuous with the free wall of the RV. Its posterior surface is formed by the trabeculae of the RV apex (including the moderator band), and its leftward aspect is constituted by the apical interventricular septum. When viewed from the RV aspect, the apical portion of the ventricular septum is constituted by an anterosuperior infundibular apical component and a posteroinferior sinus component (Fig 1⇑). This distinction between the two components of the apical septum, however, is much less obvious when viewed from the LV.
This study identified 10 patients with apical muscular VSDs in whom the defects were located in the infundibular component of the apical septum among the 50 patients referred for transcatheter closure of muscular VSDs. Such defects result in a communication between the apex of the LV and the infundibular apex and have been termed LV-infundibular apical VSDs. The prevalence of LV-infundibular apical VSD among 274 consecutive patients with muscular VSD was 7.3%. This type of defect made up 43% of apical VSDs.
The anatomic features of LV-infundibular apical defects could be clearly demonstrated with the use of echocardiography and cineangiography. The subxiphoid and parasternal short-axis views of the ventricular apexes demonstrated the relationship between the LV apex, the apex of the RV sinus, and the infundibular apex (Fig 2B⇑). The moderator band and the more caudal apical trabeculae formed a barrier separating the apexes of the RV sinus and infundibulum. Defects in the infundibular and sinus components of the apical interventricular septum could be clearly identified in these views by high-resolution imaging and slow sweeps from base to apex and back (Fig 2B⇑). As the imaging sector is swept toward the cardiac apex past the AV valves and chordae tendineae, the moderator band and the more apical trabeculae are seen to traverse between the RV septal surface and anterior free wall. Defects between LV apex and infundibular apex are seen anterior to and leftward of the apical trabeculae (Fig 2B⇑), whereas defects between LV apex and RV sinus apex are seen posterior to and rightward of the apical trabeculae (Fig 2C⇑). The apical four- and two-chamber views complemented the information obtained from the short-axis views by demonstrating the superoinferior extent of the infundibular apex (Fig 2A⇑). In all patients, the apical trabeculae were hypertrophied and created an appearance of a separate septum within the RV apex. The cineangiographic features were outlined by use of steep left anterior oblique views that produce an image similar to the apical views of the echocardiograms. In this projection, the appearance of an additional septum within the RV apex became apparent.
Implications for Transcatheter and Surgical Closure
Identification of this subset of LV-infundibular apical muscular VSDs may allow consideration of specific transcatheter closure strategies and may similarly influence surgical management. The barrier formed by the apical trabeculae that separated the infundibular apex, including the defect in the infundibular-apical septum, from the rest of the RV cavity resulted in several potential sites for device placement. As a result of hypertrophy of the apical trabeculae, the narrowest part of the interventricular communication is often located within the RV at the level of the moderator band. In six patients, partial or complete closure of the interventricular communication was accomplished by placing one or more devices at the level of the moderator band and within the RV infundibular apex. As a result, the infundibular apex became part of the physiological LV (Fig 3A and 3B⇑⇑). Interestingly, this phenomenon occurred naturally in the patient who had spontaneous closure (Fig 3C and 3D⇑⇑). In contrast to LV-infundibular apical defects, anterior muscular defects occur cephalad to the moderator band and are not potentially isolated from the rest of the RV. Transcatheter or surgical closure can also be achieved by closing the plane between the LV septal apex and the infundibular apex. This can potentially be accomplished through a small incision in the infundibular apical free wall that lies to the right of the LV apex, exposing the infundibular septal surface and allowing patch closure of the apical defect.
In conclusion, defects in the interventricular septum separating the infundibular apex from the LV apex seem to constitute a distinct morphological type of muscular VSD. An appreciation of the anatomic basis of these defects and accurate diagnosis by echocardiography and cineangiography allow planning for the most effective closure technique. In patients with small defects, no intervention is necessary, and in selected patients with hemodynamically significant defects, transcatheter device placement is an effective method for “physiological” closure.
We thank Jacqueline Kreutzer, MD, for her useful comments and William McIntosh and Emily Flynn McIntosh for artwork.
- Received July 23, 1996.
- Revision received October 16, 1996.
- Accepted October 28, 1996.
- Copyright © 1997 by American Heart Association
Graham TP, Gutgesell HP. Ventricular septal defects. In: Emmanouilides GC, Allen HP, Reimenschneider TA, Gutgesell HP, eds. Moss and Adam's Heart Disease in Infants, Children, and Adolescents. 5th ed. Baltimore, Md: William & Wilkins; 1995:724-746.
Castaneda AR, Jonas RA, Mayer JE, Hanley FL, eds. Cardiac Surgery in the Neonate and the Infant. Philadelphia, Pa: WB Saunders; 1994:187-201.
Kirklin JW, Barrat-Boyes BG, eds. Cardiac Surgery. 2nd ed. New York, NY: Churchill Livingstone; 1993:749-824.
Lock JE, Block PC, McKay RG, Baim DS, Keane JF. Transcatheter closure of ventricular septal defects. Circulation. 1988;78:361-368.