Read untitled text version

MHBD054-CH57[1077-1090].qxd 11/16/2006 6:05 PM Page 1077 PMAC-131 27A:MHBD054:Chapters:CH-57: Techbooks (PPG-Quark)

CHAPTER

57

VENTRICULAR SEPTAL DEFECTS

David L. S. Morales , Charles D. Fraser, Jr.

DEFINITION AND EPIDEMIOLOGY

A ventricular septal defect (VSD) is a deficiency in the ventricular septum that can vary in size, number, or location on the septum. All three determine its physiology, while the location alone determines its nomenclature. VSD is the most commonly recognized congenital heart defect excluding bicuspid aortic valve. Approximately 20 percent of patients with congenital defects have isolated VSDs and, if one includes VSDs in combination with other defects, VSDs are diagnosed in 50 percent of all patients with congenital heart disease.1 VSDs occur at a

rate of 0.5 per 1000 live births and in 4.5 to 7 of 1000 premature infants,1,2 with a slightly higher prevalence in females (56 percent).3 About 5 percent of VSDs are related to chromosomal syndromes (such as 22q11 deletion and trisomy 21), in which VSD is the most common cardiac defect identified.4

HISTORICAL NOTE

In 1891, Dupren coined the term Maladie de Roger in honor of Henry-Louis Roger, who first described the

KEY CONCEPTS

Epidemiology Ventricular septal defect (VSD) is the most common noncyanotic cardiac anomaly (20 percent of all malformations); it is present in over 50 percent of children with complex congenital heart disease. VSDs occur in 0.5 of 1000 live births, and 5 percent are related to chromosomal syndromes. Morphology VSDs vary in size, number, and location along septum. They are classified as (1) perimembranous (most common), (2) inlet type, (3) nonmuscular outlet, and (4) muscular. The conduction system and aortic valve leaflets are in anatomic proximity and are at risk during repair. Pathophysiology Degree and direction of shunting between the right and left ventricles depend on size (restrictive or nonrestrictive) and balance between the systemic vascular resistance (SVR) and pulmonary vascular resistance (PVR). Aortic regurgitation can ensue as a consequence of leaflet prolapse through the defect.

Clinical features Small, restrictive defects are often asymptomatic. Symptoms of overcirculation are present in unrestrictive VSDs; with development of irreversible pulmonary vascular changes in untreated large shunts, fixed pulmonary hypertension and cyanosis develop (Eisenmenger's syndrome). Diagnosis Echocardiography accurately defines the anatomy of VSD and associated lesions. Cardiac catheterization is rarely required. Treatment Timing of closure is indicated by the degree of shunting and aortic valve involvement and is most typically performed in infancy. Transatrial or transventricular approaches are most commonly utilized, with minimal morbidity and mortality. Percutaneous or periventricular device closure is still experimental, while a staged approach (with pulmonary arterial banding and delayed complete repair) is now considered only in selected cases .

1077

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1078 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1078

PART III CONGENITAL CARDIAC SURGERY

clinical and pathologic findings of a VSD in 1879.5 Eisenmenger then chronicled the natural history of an unrestrictive VSD by his account of the postmortem findings in a cyanotic patient who died at age 32 with a large VSD, a severely hypertrophied right ventricle (RV), pulmonary and tricuspid valve insufficiency, and thickened pulmonary arteries.6 However, the term Eisenmenger's syndrome was not introduced until Abbott delineated the pathophysiology of a VSD in the 1930s.7 Muller first surgically addressed a VSD in 1952 by placement of a pulmonary artery band (PAB).8 Lillehei, using controlled cross circulation, was the first to perform a VSD repair in 1954.9 DuShane reported transventricular repair in 1956, while a transatrial approach was introduced the following year by Stirling.10,11 Truex's description in 1958 of the atrioventricular node (AVN) and the conduction pathway in patients with VSDs is an integral part of all modern surgical techniques of VSD closure.12 Kirklin and associates established in 1961 the ability to repair VSDs in small infants, therefore avoiding the two-staged approach of banding of the pulmonary artery (PA) followed by VSD closure.13

ANATOMY

In order to comprehend the different nomenclatures that are used for VSDs and the approaches and techniques for repairing this anomaly, one must understand

the anatomy of the RV, the conduction system, and the tricuspid valve (TV). The TV has three leaflets: septal, anterior, and posterior (Fig. 57-1). The posterior papillary muscle, which is located on the inferior wall of the RV near the septum, gives rise to the chordae of the septal and posterior leaflets. The anterior papillary muscle gives rise to the chordae of the anterior and posterior leaflets. It is anchored at the acute margin of the RV and fuses with the RV muscle to become the moderator band, which travels inferiorly and becomes the septal band (alternatively termed trabeculum septomarginalis or septomarginal trabeculation), which travels toward the RV outflow tract. The septal band then divides into its posterior and anterior arms. Between these two arms is the infundibular septum, also known as the conal, outlet, or supracristal septum. The medial papillary muscle (the muscle of Lancisi), which gives chordae to the anterior and septal leaflets of the TV, is most prominent and identifiable during infancy and attaches to the septal band or its posterior arm. The parietal band is the continuation of the septal band's posterior arm anteriorly onto the RV free wall. The parietal band and both arms of the septal band join to form the supraventricular crest (crista supraventricularis), the C-shaped nonobstructive entrance into the outlet region (subpulmonary conus) of the RV. This muscle shelf between the tricuspid and pulmonary valve (PV) creates pulmonary­tricuspid valve discontinuity.

Figure 57-1 Anatomy of the right ventricle. (From Netter F. The Netter Collection, vol 5, sec 1. Illustrations. Icon Learning Systems, MediaMedia USA, Inc. All rights reserved. With permission. The drawing has not been modified; however the labeling is that of the authors.)

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1079 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

1079

Figure 57-2 A view of the atrioventricular conduction system from the left ventricle and its relationship to the membranous septum and the aortic valve. AVN atrioventricular node, MS membranous septum, fasc fasciculus, RBB right bundle branch, LBB left bundle branch. (From Titus JL. Normal anatomy of the human cardiac conduction system. Mayo Clin Proc 1973;48:24­30. With permission.)

It is essential to know the location of the AVN and the bundle of His in order to perform safe closure of a VSD. The AVN is located in the triangle of Koch (formed by the tendon of Todaro, the coronary sinus os, and the septal leaflet of the TV). More precisely, it is in the muscular region right below the triangle's apex and directly on the right atrial side of the central fibrous body. From another perspective, the AVN is under the nadir of the noncoronary (posterior) cusp of the aortic valve

(Fig. 57-2). The AV bundle (Bundle of His) arises from the AVN and exits at the apex of the triangle of Koch, passing to the ventricular side through the right aspect of the central fibrous body. At this point, the bundle is on the posteroinferior margin of the membranous septum, which lies just posterior to the commissure of the septal and anterior leaflets of the TV (Fig. 57-3). The bundle penetrates the ventricular septum and continues on the left ventricular (LV) side in 75 percent of patients.

Figure 57-3 Depiction of the atrioventricular node and its course on the inferior border of a perimembranous VSD; viewed from a right atrial surgical approach. (Drawing by Rachid Idriss. Used with the artist's permission, who reserves all rights. The drawing has not been modified, however the labeling is that of the authors.)

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1080 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1080

PART III CONGENITAL CARDIAC SURGERY

Whether on the right or left, it courses along the inferior border of the membranous septum and begins to give off fibers to the left bundle branch over a distance of 1 to 2 cm. When the membranous septum is intact, this area is just below and to the left of the commissure between noncoronary and right coronary cusps of the aortic valve. In considering the bundle from the aortic valve perspective, one should perceive its path as coursing a few millimeters below the area between the right noncoronary (posterior in Fig. 57-2) commissure and the nadir of the right coronary sinus. The remaining fibers of the bundle, which now surface to the anteroinferior border of the membranous septum, become the right bundle branch. The left bundle branch fans out over the septum while the right bundle branch courses as a single radiation (Fig. 57-3). From the anteroinferior border of the membranous septum, the right bundle passes below the muscle of Lancisi and then to the inferior borders of the septal and moderator bands until it reaches the anterior papillary muscle, where it disperses to innervate the RV. Therefore the bundle of His is most often in harm's way during operations on a perimembranous (PM) VSD. The bundle of His runs along a PM VSD's inferior border from where it penetrates the TV annulus to the most inferior papillary muscle on the VSD's muscular rim (the muscle of Lancisi) (Fig. 57-4). The fact that the AVN and bundle of His are specialized myocytes and thus exist only in

muscular tissue and not in fibrous tissue is an important fact to consider in placing sutures for VSD repair. The conduction system can also be at risk at the superior edge of a muscular inlet VSD, where the conduction system runs in the muscle between the VSD and the membranous septum.

NOMENCLATURE AND PATHOLOGIC ANATOMY

There are many classifications of VSDs that have been proposed and used; our preference is a modification of the classic Anderson classification. However, to make our classification useful to the reader, we will compare other terms and classification to this terminology. The ventricular septum can be sectioned into four regions: (1) the inlet septum, which is the area of the septum bounded by the attachments of the TV; (2) the muscular septum, which is the area from apex to the crista supraventricularis outside these attachments; (3) the outlet septum, which is the area from the crista supraventricularis to the PV, and (4) the perimembranous septum, which is quite small and lies under the commissure of the anterior and septal leaflets of the TV. The muscular region is further subdivided into anterior, posterior, apical, and midmuscular regions. The Anderson classification names

A

B

Figure 57-4 Anatomy of the atrioventricular conduction system. A. The atrioventricular bundle and its right branch dissected in the right ventricle. B. The atrioventricular bundle and its left branch dissected in the left ventricle. (From Sobotta. Atlas der Anatomie des Menschen, 16th ed. Munich: Urban & Schwarzenberg, 1963. With permission.)

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1081 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

1081

Figure 57-5 The different types of VSDs as viewed via the right ventricle. (Drawing by Rachid Idriss. Used with the artist's permission, who reserves all rights. The drawing has not been modified, however the labeling is that of the authors.)

VSDs according to the region in which they are located. However, from a surgical perspective, one wants to know not only the location of the defect but also its relation to the conduction system. The latter can be transmitted to the surgeon by first defining whether the defect is PM (an absence of the membranous septum) or not (all other types of defects). This automatically tells the surgeon whether or not the conduction system is remote, and if not remote, where it is (inferior to a PM defect for a dlooped heart and superior to a PM defect for a l-looped heart). The exception to this rule is an inlet VSD, which is discussed further on. One can also describe the type of PM VSDs by the areas defined by Anderson. Therefore a PM VSD can be one with muscular extension, outlet extension, inlet extension, or any combination. Non-PM VSDs can be muscular in type (outlet, inlet, anterior, posterior, apical, midmuscular) or outlet or inlet types bound by valve tissue (Fig. 57-5). The synonyms in other classifications for all the terms used above can be found in Table 57-1. The table lists all the different names by which a VSD within a certain region can be designated. The different names listed under a certain region are sometimes used imprecisely to denote any VSD in the area, but most of the terms are not interchangeable and refer to a very specific type of VSD. For example, some might use the

Table 57-1

Nomenclature utilized for ventricular septal defects

Synonyms Membranous, infracristal, paramembranous Membranous, conoventricular, subaortic Membranous, infracristal, paramembranous Conal, supracristal, subarterial, subaortic, subpulmonary, infundibular, intracristal, doubly committed, conal septal, juxtaarterial Canal type, atrioventricular Canal type Conal, supracristal, subarterial, infundibular, conal septal Central Marginal Inferior

Preferred terminology Perimembranous VSD PM VSD with inlet extension PM VSD with outlet extension PM VSD with muscular extension Non-muscular-outlet VSD

Inlet VSD Muscular Inlet Outlet

Midmuscular VSD Apical muscular VSD Anterior muscular VSD Posterior muscular VSD

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1082 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1082

PART III CONGENITAL CARDIAC SURGERY

terms subaortic and supracristal to refer to the same VSD; however, the former is a general term that refers to any VSD abutting the aortic valve, whereas the latter specifically refers to an outlet VSD abutting a semilunar valve. The PM VSD represents approximately 80 percent of all VSDs; the remaining 20 percent are evenly distributed among inlet, outlet, and muscular VSDs. A notable exception to this distribution is the almost 30 percent overall prevalence of outlet-type VSDs in the Asian population with VSDs.14 Small asymptomatic muscular VSDs, which are probably the most common heart defect (present in about 5 percent of all neonates), are grossly underestimated in these statistics, since the vast majority close in the first months of life without recognition.15 Perimembranous VSDs are intimately related to the tricuspid and aortic valves. In fact, the absence of the PM septum often creates tricuspid­aortic valve fibrous continuity. "Spontaneous closure" of PM VSDs usually occurs from partial or complete occlusion of the VSD by aneurysmal TV tissue. This aneurysmal tissue is formed from the sheer force of the left-to-right (L-R) shunt resulting in creation of fibrotic tissue, "accessory valve tissue," and/or, perhaps, remnants of endocardial cushion tissue. Some have described a defect of just the atrioventricular septum causing a shunt from the left ventricle to the right atrium, known as a "Gerbode defect."16 When seen on echocardiography, this almost always represents a PM VSD with a jet directed through a cleft in the TV or the septal-anterior commissure, giving the echocardiographic appearance of a Gerbode defect. A consequence of this physiologic shunting from the LV to the right atrium can be extensive right atrial enlargement; if an atrial septal defect exists, streaming of the blood can possibly cause right to left atrial (LA) shunting with resulting cyanosis. Malalignment of the infundibular septum is also associated with PM VSDs, usually with outlet extension. If one considers the infundibular septum as being in the coronal plane with the RV outflow tract above and the LV outflow tract below, then comprehension of the malaligned VSDs is more straightforward. In an anteriorly malaligned VSD, the infundibular septum is deviated anteriorly into the RV outflow tract, causing obstruction. The aortic valve also moves anteriorly, so that it overrides the ventricular septum. In a posteriorly malaligned VSD, the infundibular septum deviates posteriorly into the LV outflow tract, causing obstruction. Outlet defects are often differentiated between defects that are completely bound by muscle and those that are bounded on one side by aortic valve tissue (subaortic, juxtaarterial VSDs), by pulmonary valve tissue (subpulmonary VSDs), or both (doubly committed, subarterial VSDs). This differentiation is useful in terms of the pathophysiology. The nonmuscular outlet defects along with PM VSDs with outlet extension can both be associated with aortic valve insufficiency, characterized by lengthening and eventual prolapsing of the right coronary leaflet

(outlet defect) or noncoronary leaflet (PM VSD). This valve tissue can partially or completely close the defect. The conduction system is remote to outlet defects. Inlet defects are beneath the septal leaflet of the TV, just inferior and posterior to the membranous septum. These defects can be an atrioventricular septal defect or a muscular inlet defect. The former is an endocardial cushion defect, characterized by absence of the PM septum, atrioventricular valve abnormalities, and a conduction system traveling along the inferior aspect of the defect. A muscular inlet defect is a VSD that can be remote to the conduction system or have the conduction system border the defect superiorly. Muscular VSDs can be multiple or appear to be so because of the overlying trabeculae in the RV. This topography and multiplicity may cause difficulty in closure of these defects. When these impediments become prohibitive to conventional surgery, the term "Swiss cheese septum" is usually used to describe them. However, it should be noted that some believe that a Swiss cheese septum is actually an entity distinct from multiple VSDs. The morphology of the Swiss cheese septum is believed to originate from septal noncompaction during embryologic development; thus, unlike a group of muscular VSDs, Swiss cheese defects cannot close spontaneously.17,18

PATHOPHYSIOLOGY

The pathophysiology of a VSD is determined by the size of the defect and its location. If the defect is smaller than the aortic annulus (a restrictive VSD), then the shunt's direction and volume is determined by the difference in systolic pressures of the ventricles. If the defect is larger than the aortic valve, then the defect is nonrestrictive and the pressure in the ventricles should equalize. The direction and volume of this shunt is determined by the difference in pulmonary and systemic vascular resistance. Restrictive VSDs can be further categorized as small or moderate in size. A small VSD has a large resistance to flow, resulting in a small L­R shunt, and the pulmonary vasculature is well protected. The small shunt volume does not increase ventricular work or volume, so the LA and LV tend not to dilate. The large pressure gradient across the small VSD favors L-R flow throughout the cardiac cycle, which can produce a continuous murmur. Moderate-sized defects remain restrictive but allow a large enough L-R shunt to cause left-sided volume overload, characterized by LA and LV dilation. The pulmonary vasculature remains protected; however, some pulmonary hypertension can exist, which will cause RV pressure and work to increase. This results in mild RV hypertrophy. Nonrestrictive VSDs allow the RV and LV to have equal pressures, with the determination of flow across the defect resulting from outflow resistance. In the LV,

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1083 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

1083

this is determined by systemic vascular resistance or by LV outflow tract anomalies (subaortic stenosis, aortic stenosis, coarctation, etc.). In the RV, this is determined by PVR or by RV outflow tract anomalies (PV stenosis, tetralogy of Fallot, etc.). PVR is high at birth and decreases the L-R shunt; however, the PVR naturally decreases as the pulmonary vascular bed matures. The decline in PVR occurs mostly over the first few days, but the process is not completed until 2 to 6 weeks after birth.19,20 The thinning of the vascular media, the enlargement and proliferation of the peripheral PAs, and the regression of the PA's perinatal muscularity mark the normal maturation of the pulmonary vessels. However, a large L-R shunt disturbs the growth and remodeling of the pulmonary vasculature, so that medial hypertrophy develops, muscularization of PAs persists, and the size and number of peripheral PAs is reduced. These structural changes are the basis of pulmonary vascular disease.21,22 Any patient with a nonrestrictive VSD and no RV outflow tract obstruction has pulmonary hypertension, but this does not necessarily indicate that the patient has pulmonary vascular disease. The latter is a disease state of the pulmonary vasculature correlating to specific pathologic changes in the PAs, while pulmonary hypertension is simply a hemodynamic state of the PAs at

any given time. A nonrestrictive VSD is characterized by a large L-R shunt, which will cause volume overload of the LA and LV, resulting in significant dilation of these chambers. This shunt will also result in pulmonary hypertension, which in time will cause RV hypertrophy and increasing PVR. As the PVR increases, the L-R shunt decreases. When end-stage pulmonary vascular disease ensues, the shunt can even reverse direction to a right-to-left shunt (i.e. Eisenmenger's syndrome). Reversal of shunting and cyanosis rarely present in children below age 5 with isolated VSDs and is usually seen in adolescents and adults.

CLINICAL FEATURES

The presentation of a patient with a VSD varies according to the size of the defect, the amount of L-R shunting, and PVR (Fig. 57-6). The majority of VSDs are small; these patients are asymptomatic and are diagnosed because of a loud systolic murmur, prompting an echocardiogram. The murmur is often not heard until the first postnatal visit, after PVR has dropped. Many of these defects will close spontaneously by muscle hypertrophy, fibrosis of the defect's margins, or leaflet adherence

Diagnosis of isolated VSD

murmur Echo CXR ECG

Cyanosis

No cyanosis

Medical therapy Cardiac catheterization Nonrestrictive Symptoms LV enlargement Aortic Regurgitation

High PVR No response to pulmonary vasodilators High PVR Responsive to pulmonary vasodilators

Spontaneous closure or Restrictive with no AR

Pulmonary artery band (rare cases)

Heart Transplant Heart Lung Transplant Lung Transplant with VSD closure

Complete repair ± VSD patch or

atrial fenestration

Complete repair

Figure 57-6

Decision-making flowchart for isolated ventricular septal defect.

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1084 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1084

PART III CONGENITAL CARDIAC SURGERY

to the defect. In a study by Turner and colleagues, 290 children with isolated VSDs of all sizes were followed for a mean of 65 months.23 The study revealed that 68 percent of VSDs with complete muscular borders closed spontaneously, and 29 percent of PM VSDs closed. There were no cases of endocarditis in this nonoperative series. The aortic valve can also contribute to the narrowing of defects that have an outlet component, which can cause aortic insufficiency; this is a situation that demands prompt surgical treatment. Despite therapy, the natural history of the damaged aortic valve may no longer be normal. Thus waiting for spontaneous closure of PM VSDs or outlet VSDs, even with close follow-up, is not without risk. Patients with large VSDs present with varying degrees of pulmonary overcirculation. The amount of overcirculation usually determines their age at presentation. Patients with larger defects and significant overcirculation can present in the first months of life, as PVR falls. These patients present with tachypnea, poor feeding, slow weight gain, and diaphoresis with activity. They are at risk for repeated upper respiratory infections, "cardiac asthma," and failure to thrive. Shunt size and the length of exposure to the resulting pulmonary hypertension will determine the time frame for the development of the varying degrees of pulmonary vascular disease. Patients with trisomy 21 have a shorter time course to reach significant pulmonary vascular disease than most children.24 The end stage of pulmonary vascular disease is termed Eisenmenger's syndrome: PVR greater than systemic vascular resistance, reversal of shunt flow from L-R to R-L, cyanosis, and eventually RV failure.25

increases, the third heart sound becomes more prominent, as does a middiastolic rumble, which indicates increased pulmonary blood flow at least double systemic blood flow. The holosystolic nature and the loudness of the murmur begin to dissipate as defects become larger and the pressure gradient between the RV and LV decreases. As pulmonary hypertension increases, the second heart sound has a narrower split and the pulmonary component becomes louder. There are no specific ECG patterns that are pathognomonic for VSDs and most patients' ECGs are normal. There can be ECG evidence of increased left heart volume (left atrial enlargement), but these findings are not specific to VSDs. On CXR, as the LV volume increases, there is a downward and leftward elongation of the cardiac silhouette on the anteroposterior (AP) film, whereas left atrial enlargement is seen on the lateral film or, when severe, by carinal widening on the AP film (Fig. 57-7). The increase in vascular markings denotes the amount of overcirculation. Large VSDs coupled with high PVR allow for minimal shunting. Therefore the LV work is close to normal, resulting in little precordial activity. There will, however, be a notable RV lift. A short or absent VSD murmur is usually heard. Murmurs that may be heard are a result of tricuspid regurgitation (harsh holosystolic murmur) or pulmonary insufficiency (early diastolic murmur). There is usually no third heart sound or diastolic rumble. The pulmonary component of the second heart sound is quite prominent; however, more than 50 percent of the time, the second heart sound is single. ECG may show RV hypertrophy. The CXR on these patients can demonstrate

DIAGNOSIS Physical examination

Small defects usually have minimal physical findings and a normal chest x-ray (CXR) and electrocardiogram (ECG). Precordial activity is typically normal, but a thrill may be palpable on the lower left sternal border. The murmur is a loud, high-frequency holosystolic murmur that includes and sometimes goes beyond the second heart sound and is heard best over the left lower sternal border. The murmur's continuous nature in systole provides evidence of a large pressure gradient between the RV and LV. Small muscular defects can sometimes have short murmurs that cut off at midsystole because of septal contraction.26 Heart sounds are usually normal. Moderate to large defects with minimal elevation of PVR present with similar exams that vary according to the amount of shunting. Precordial activity is accentuated and can span from the left apical area at first, indicating increased LV volume, to the right parasternal area eventually, indicating pulmonary hypertension and RV hypertrophy. As LV volume increases, the left thorax may begin to bulge, especially in younger infants. As the shunt

Figure 57-7 Anteroposterior chest x-ray of a patient with a VSD. Note the leftward elongation of the cardiac silhouette, representing an increase in left ventricular volume, and the plethoric lung fields.

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1085 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

1085

a normal-size heart or RV hypertrophy, both accompanied by marked prominence of the main PA and its immediate branches. There is a paucity of vascular markings on the outer third of the lung fields.

Diagnostic Imaging

Two-dimensional echocardiography with color Doppler flow evaluation is the most widely used imaging to diagnose and characterize a VSD. To assess a VSD completely, one must not only localize it but also define its shape and dimensions, which is accomplished by viewing the defect from multiple imaging planes. Color Doppler allows for small VSDs not seen on two-dimensional echocardiography (usually 2mm) to be identified and, more importantly, provides physiologic information about the VSD. One can measure the peak velocity across the VSD, which, if placed in the modified Bernoulli formula [4 x (peak velocity)2], can yield the interventricular pressure gradient.27 If this velocity is high, one has a restrictive VSD. If this velocity is low, one usually has a nonrestrictive VSD with near equalization of RV and LV pressures. However, at times a low velocity can be seen with a restrictive VSD if there is high RV pressure secondary to RV outflow obstruction or elevated perinatal PVR that has not yet fallen. Therefore a low intraventricular pressure gradient does not necessarily correlate with pulmonary vascular disease, even in the presence of a nonrestrictive VSD. RV pressure may be estimated by measuring the velocity of the tricuspid regurgitant jet [4 (TR jet velocity)2 (central venous pressure RV pressure)]. If a pulmonary insufficiency jet exists, then an estimation of the diastolic PA pressure can be calculated by measuring its velocity [4(PI jet velocity) 2 (RV diastolic pressure)]. One can also get a sense of where the patient is in the spectrum of VSD pathophysiology by assessing the amount of LV and LA dilation as well as RV hypertrophy. The echocardiogram should obviously assess for other cardiac anomalies in particular patent ductus arteriosus, aortic coarctations, and RV or LV outflow tract obstruction. The principal indication for diagnostic cardiac catheterization of a VSD patient is when echocardiography and the clinical assessment indicate the possibility of advanced pulmonary vascular disease. Again, this would be rare for young children. One should keep in mind that the PVR of a child with an isolated VSD would have to be extremely elevated not to attempt a repair in this era of multiple pulmonary vascular bed dilators [inhaled nitric oxide (iNO), prostacyclin, sildenafil milrinone, etc.] Therefore one may question exposure to the risks of catheterization to quantify the echocardiographic findings more precisely when the results will not change the therapeutic decision. Catheterization of a VSD patient is also useful in attempts to define the anatomy of multiple apical VSDs, which magnetic resonance imaging (MRI) and echocardiography sometimes cannot identify.

Catheterization of a patient with an isolated VSD should result in (1) a Qp:Qs that can be estimated by (aortic O2 sat SVC O2 sat) / (pulmonary venous O2 sat PA O2 sat); (2) the calculated PVR (mPAP ­ LA mean pressure) / pulmonary blood flow (Qp) resulting in Wood units ( mmHg/L/min); (3) the PA pressures; (4) if PVR is high, the determination of the pulmonary vasculature reactivity to vasodilators such as 100 percent FIO2 and iNO; and (5) the delineation of any unclear anatomy. MRI has recently been used to provide accurate information about the morphology of VSDs. When surgical referral is being considered, MRI may be recommended in those patients in whom it is difficult to discern ventricular volume overload. The MRI can supply a noninvasive estimate of the Qp:Qs as well as an accurate account of the heart's volumes and of associated anomalies that are sometimes difficult to diagnose by echocardiography (anomalous pulmonary venous return, for example). Also, three-dimensional intracardiac reconstruction MRI may soon be more widely available.

THERAPEUTIC MANAGEMENT Medical therapy

The medical treatment of VSD patients is oriented toward decreasing L-R shunting and the symptoms of overcirculation. This is usually done with a combination of diuretics (i.e., furosemide), afterload-reducing agents (ACE inhibitors), and digoxin. Patients may also require nutritional support such as nasogastric tube feedings. Close surveillance for and aggressive treatment of upper respiratory infections is an important aspect of these children's care. Most children presenting with VSDs have some degree of pulmonary vascular disease; however, the presentation of an older cyanotic child with end-stage pulmonary vascular disease requires catheterization. The treatment and workup of such cases is outside the scope of this chapter. One should exercise caution when a patient requires intubation or inotropic support for a supposedly isolated VSD. Before embarking on surgical therapy, a careful investigation should be performed to make sure that no other cardiac anomaly (for example, left-sided obstruction or an additional source of L-R shunting) or primary pulmonary disease exists.

Invasive Therapy

Surgical Therapy Surgical techniques for VSD closure have progressed tremendously over the past three decades and have allowed for neonatal surgical therapy to move from palliation to repair. At most institutions, all patients with isolated VSDs that require surgical correction undergo single-stage closure. At the authors' institution, all repairs are generally performed with aortic and bicaval cannulation. Once the

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1086 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1086

PART III CONGENITAL CARDIAC SURGERY

heart is arrested with cold cardioplegia, the right atrium is opened and the left heart is vented via the atrial septum. This should produce a bloodless and still intracardiac surgical field in which to work. There are several approaches to VSDs at this point, depending on the location. The majority of VSDs (perimembranous, inlet, and the majority of muscular defects) can be addressed via a right atrial approach (Fig. 57-8). The leaflets of the TV are retracted to visualize the defect. A particular variation in technique to visualize transatrially inlet defects or PM VSDs with outlet extension is the incision of the septal and/or anterior leaflet(s) of the TV via a radial incision or one parallel to the annulus. This will often provide excellent visualization for repair; once this has been completed, the TV is reconstructed. The actual closure of the VSD can be done with a patch secured to the septum with interrupted pledgetted stitches, a running stitch, or a combination of the two. The patch can be made of various materials, including autologous pericardium tanned in glutaraldehyde, Dacron, or GoreTex. Primary (no patch) closure is usually reserved for a small muscular VSD. A key to successful VSD closure is the surgeon's awareness of the conduction system and the aortic valve. The TV annulus, which is the structure to which the VSD patch is often anchored posteriorly and superiorly, is just anterior to and sometimes in fibrous continuity with the aortic valve. Therefore the aortic valve leaflets are at risk for injury when these

Figure 57-8 Transatrial approach to VSD patch closure. In the illustration, a running suture technique is utilized. (From Kouchoukos NT, Blackstone EH, Doty DB, et al (eds). Kirklin/Barratt-Boyes Cardiac Surgery, 3d ed. Philadelphia: Churchill Livingstone, 2003:874. With permission.)

superior sutures are being placed. The conduction system is at risk in stitching the inferior rim of a PM VSD between the muscle of Lancisi (the most inferior papillary muscle bundle on the VSD rim) and the TV annulus (Fig. 57-4). In this area, interrupted sutures can be taken far from the rim or a running suture can be placed very superficially along the rim. It is important to remember that the conduction system exists only in muscular tissue and not in fibrous tissue. Approach through the pulmonary artery is usually reserved for outlet defects. The pure outlet muscular defects can be closed with a simple primary or patch technique. However, the majority of outlet defects abut the PV annulus. These defects are also closed with a patch, but several unique factors must be considered. These particular VSDs have the highest incidence of significant aortic valve involvement. The aortic valve leaflets (usually the right coronary cusp) can prolapse and/or fill the defect. Therefore it is essential to identify the aortic valve and not to include it in the repair. The patch will not only close the defect but also serve to support the aortic valve and eliminate the Venturi effect, which can be the cause of aortic valve prolapse. Some mild forms of aortic insufficiency will improve with repair of the defect alone, but this is unpredictable. If the aortic insufficiency is moderate or greater and/or the aortic leaflet clearly has a pathologic change, then the aortic valve should be addressed. Perimembranous defects can also be associated with aortic insufficiency and usually involve prolapse of the noncoronary cusp. A discussion of aortic valve repair techniques--which include triangular resection, subcommissural stitches, Trussler-type repair (horizontal plication of the redundant leaflet to the aortic wall), and Yacoub-type repair (primary closure of the VSD with vertical plication of the sinus and leaflet)--are beyond the scope of this chapter. There is no superior rim to outlet defects, so interrupted pledgetted sutures are usually placed through the fibrous base of the pulmonary valve leaflets and then through the patch. Interrupted or running sutures can anchor the remainder of the patch, with little concern for the conduction system, which is remote. However, if the posterior limb of the septal band (part of the crista supraventricularis) is not identified, the VSD may be a PM defect with outlet extension. Thus the conduction system would be on the inferior border of the defect. A transaortic approach is usually applied in addressing a VSD that is associated with another left-sided lesion such as aortic valve insufficiency or valvular/subvalvular stenosis. Our practice is still to close these defects from the RA approach to avoid left bundle branch block; however many still use this approach, and discussion of it is helpful in contemplating the relationship of VSDs and the conduction system. An RV approach (Fig. 57-9) for an isolated VSD is unnecessary except in very rare situations in which the VSD is inaccessible from the different approaches; for

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1087 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

1087

authors still attempt closure of these Swiss cheese defects from the right side of the heart.

RESULTS Surgical outcomes

Complications from surgical VSD closure are infrequent; they include the following: 1. Tricuspid valve insufficiency from chordal shortening, leaflet entrapment by the VSD patch, or leaflet distortion if incised for exposure. 2. Possible aortic valve insufficiency if the sutures in the superior aspect of the patch catch or damage the aortic valve leaflets. 3. Complete heart block if the bundle of His is damaged or transected by sutures. Temporary block can occur secondary to edema in the area or trauma caused by careless suctioning or manipulation around the AVN area with instruments. The decision of when to place a pacemaker depends on the individual situation. Factors in one's decision include preoperative AVN function, the nature of the procedure and cardiac anatomy, and the postoperative course. The frequency of this complication should be approximately 1 percent. Right bundle branch block, on the other hand, is a frequently seen rhythm (about 35 percent) change postoperatively.28 4. Residual VSD: Leaving the operating room with a clearly significant residual VSD has been greatly reduced with the use of intraoperative transesophageal echocardiography (TEE). The difficulty arises when one must decide whether a small residual defect is physiologically and clinically significant enough to warrant rearresting the heart and attempting complete closure. A Qp:Qs of less than 1.5:1 is used as a general guideline for a residual VSD shunt that could be observed. However, in practice, the decision to leave a residual VSD is multifactorial and based on such aspects of the case as the difficulty of the VSD closure, one's perception of how the heart will tolerate another arrested period, and the residual shunt that is left. Operative mortality for isolated VSD closure is quite low ranging from 0 to 3 percent.29,30 The most common mode of death is acute cardiac failure, which usually is a result of one of the following: poor intraoperative myocardial protection, pulmonary hypertensive crises, a preoperative viral pulmonary process, or any combination of these with a small malnourished infant in heart failure. The risk factors for hospital death have changed significantly over the past decade. Young age, multiple VSDs, location, aortic insufficiency, PAB, and patient size, which all have been predictive of hospital mortality at various times in the past, are no longer risk factors.28

Figure 57-9 Transatrial approach to VSD closure. In the illustration, an interrupted suture technique is utilized. (From Kouchoukos NT, Blackstone EH, Doty DB, et al (eds). Kirklin/Barratt-Boyes Cardiac Surgery, 3d ed. Philadelphia: Churchill Livingstone, 2003:876. With permission.).

example: (1) When the posterior limb of the septal band is absent in a large outlet VSD, it is difficult to approach its inferior border from the PA or its superior border from the right atrium; thus neither approach suffices. (2) Visualization of a PM VSD with outlet extension can be hindered by hypertrophied infundibular muscle bundles. Resection of such bundles may at times be difficult from the RA or PA approaches; thus exposure is optimized by a small right ventriculotomy. Mapping out the ventricular incision in relation to the coronary anatomy is essential in avoiding significant ischemic complications, such as disruption of an anomalous anterior descending artery coming from the right coronary artery or of a large conal branch responsible for the proximal septal perforators. Again, we emphasize that right ventriculotomy is rarely needed to close an isolated VSD. The LV approach is used even more infrequently than the RV approach and is reserved only for addressing those patients with multiple apical defects. Swiss cheese defects are difficult to approach from the right side because the trabeculations make the identification of distinct VSD borders impossible. However, this sieve-like defect on the right is fairly clear cut on the nontrabeculated smooth septum of the LV. After careful mapping of the coronary arteries, an incision parallel and to the left of the anterior descending artery is started at the apex and extended vertically. These incisions are obviously avoided whenever possible because of pseudoaneurysms, ventricular dysfunction, and ventricular arrhythmias arising from scar tissue. These complications are why many

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1088 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

1088

PART III CONGENITAL CARDIAC SURGERY

In the current era, only major associated cardiac anomalies, especially when associated with multiple VSDs, are risk factors for hospital mortality. Preoperative PA pressures or resistance do not affect hospital mortality but do affect long-term results, which manifest as late deaths secondary to progression of pulmonary vascular disease.31 Late mortality after repair, when pulmonary pressures are low, is less than 2.5 percent; most of these deaths result from ventricular arrhythmias.28 The surgical repair of an isolated VSD before complications begin to arise (usually before 2 years of age, depending on the VSD) can return a patient to a normal life expectancy with full functional activity and normal growth.28

VSD occluder, currently under study for patients weighing more than 8 kg). The largest report in the literature consists of only 25 patients collected from 9 different institutions.33 This study has less than 1-month followup and an average patient age of 14 years and weight of 43 kg. The challenge for this therapeutic approach is that PM VSDs are almost always surgically approachable with negligible mortality and minimal morbidity. Also, the possible complications for devices in this position include not only device embolization, air embolism, perforation, residual shunt, and hemolysis but also are expanded to heart block and aortic insufficiency.

Palliation Device closure

Device closure of VSDs is an emerging field that should be offered as a therapeutic option in selected cases. These devices are usually introduced via percutaneous catheterization techniques but can also be alternatively introduced via a periventricular approach. The periventricular technique, which is performed off bypass and uses TEE guidance, passes an occluding device through the right ventricular free wall and deploys it.32 This technique can be repeated for multiple muscular VSDs. Experience with percutaneous device closure is greatest with muscular defects. The two devices presently offered in the United States for closure of muscular defects is the Cardio-SEAL (NMT, Boston, MA) device (a modified ASD clamshell occluder, FDA-approved for high-risk surgical VSD patients) and the Amplatzer Muscular VSD occluder, a specifically designed nitinol wire with polyester mesh device, which is currently under FDA review. Clinical experience with percutaneous closure of muscular VSDs has reached a point that it can be considered in the management of VSDs that are challenging to address surgically. On the other hand, closure of perimembranous VSDs using percutaneous device closure is still very much experimental, with only one available device (the Amplatzer Membranous The list of clinical and anatomic features that discourage early primary repair of VSDs becomes shorter with every year; thus the frequency of surgical palliation for isolated VSDs using pulmonary artery banding has continued to decline. Size of the patient, which in the past was one of the most frequent reasons for a PAB, is no longer a contraindication for complete repair in major centers, where the expansion of neonatal heart surgery has rendered this consideration almost null. Successful repair in a premature neonate of 700 g has been reported.34 For patients with isolated VSDs, a "Swiss cheese septum" and multiple VSDs are the most frequent diagnoses to be palliated with a PAB. Pulmonary artery banding is performed off bypass, can be approached either via a median sternotomy or a left thoracotomy, and is described in detail in Chap. 54. Although it may be done palliatively, placement of a PAB may not be a simple or low-morbidity procedure. It is not always straightforward to balance the pulmonary and systemic circulations, especially with a reactive and/or high PVR; PA banding can carry a hospital mortality of 8 percent and even higher in neonates.37 It may require early reoperation for PAB adjustment; furthermore, 29 percent of PABs can be inadequate (too loose or tight)37 and after PAB removal, PA reconstruction is often required at the time of complete intracardiac repair.

References

1. Wells WJ, Lindesmith GG. Ventricular septal defect. In: Arciniegas E (ed). Pediatric Cardiac Surgery. Chicago: Year Book, 1985. 2. Moe DG, Guntheroth WG. Spontaneous closure of uncomplicated ventricular septal defect. Am J Cardiol 1987;60:674­678. 3. Hoffman JLE, Rudolph AM. The natural history of ventricular septal defects in infancy. Am J Cardiol 1965; 16:634­653. 4. Nora JJ, Fraser FLC. Medical Genetics. Philadelphia: Lea & Febiger, 1974:334. 5. Roger H. Recherches cliniques sur la communication congenitale des duex coeurs, par inocclusion du septum interventriculaire. Bull Acad Nat Med (Paris) 1879;8:1074­1075. 6. Wood P. The Eisenmenger syndrome or pulmonary hypertension with reversed central shunt. Br Med J 1958;46: 755­762. 7. Abbott ME. Congenital heart disease. In: Nelson's Loose-Leaf Medicine, vol 5. New York: Thomas Nelson, 1932:207. 8. Muller WH Jr, Dammann JF. The treatment of certain congenital malformations of the heart by the creation of pulmonic stenosis to reduce pulmonary hypertension and excessive pulmonary blood flow. Surg Gynecol Obstet 1952;95:213­219. 9. Lillehei CW, Cohen M, Warden HE, et al. The results of direct vision closure of ventricular septal defects in eight patients by means of controlled cross circulation. Surg Gynecol Obstet 1955;101:446­466.

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1089 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Chapter 57 Ventricular Septal Defects

10. DuShane JW, Kirklin JW, Patrick RT, et al. Ventricular septal defects with pulmonary hypertension: Surgical treatment by means of a mechanical pump-oxygenator. JAMA 1956;160:950­953. 11. Stirling GR, Stanley PH, Lillehei CW. Effect of cardiac bypass and ventriculotomy upon right ventricular function. Surg Forum 1957;8:433­438. 12. Truex RC, Bishof JK. Conduction system in human hearts with interventricular septal defects. J Thorac Surg 1958;35: 421­439. 13. Kirklin JW, DuShane JW. Repair of ventricular septal defect in infancy. Pediatrics 1961;27:61­66. 14. Tatsuno K, Ando M, Takan A, et al. Diagnostic importance of aortography in conal ventricular-septal defect. Am Heart J 1975;89:171­177. 15. Johnson WH Jr, Moller JH. Pediatric Cardiology. Philadelphia: Lippincott Williams & Wilkins, 2001:99. 16. Velebit V, Schoneberger A, Ciaroni S, et al. "Acquired" left ventricular-to-right atrial shunt (gerbode defect) after bacterial endocarditis. Tex Heart Inst J 1995;22:100­102. 17. Seddio F, Reddy VM, McElhinney DB, et al. Multiple ventricular septal defects: How and when should they be repaired? J Thorac Cardiovasc Surg 1999;117:134­139. 18. Agmon Y, Connolly HM, Olson LJ, et al. Noncompaction of the ventricular myocardium. J Am Soc Echocardiogr 1999;12:859­863. 19. Krovetz LJ, Goldbloom S. Normal standards for cardiovascular data. II. Pressure and vascular resistance. John Hopkins Med J 1972;130:187­195. 20. Rudolph AM, Auld PAM, Golinko RJ, et al. Pulmonary vascular adjustments in the neonatal period. Pediatrics 1961;28:28­34. 21. Wennik ACG, Oppenheimer-Dekker A, Moulaert AJ. Muscular ventricular septal defects: A reappraisal of the anatomy. Am J Cardiol 1979;43:259­264. 22. Van Praagh R, McNamara JJ. Anatomic types of ventricular septal with aortic insufficiency. Am Heart J 1968;75: 604­619. 23. Turner SW, Hornung T, Hunter S. Closure of ventricular septal defects: A study of factors influencing spontaneous and surgical closure. Cardiol Young 2002;12:357­363. 24. Hasegawa N, Oshima M, Kawakami H, et al. Changes in pulmonary tissue of patients with congenital heart disease and Down syndrome: A morphological and histochemical study. Acta Paediatr Jpn 1990;32:60­66.

1089

25. Lucas RV Jr, Adams P Jr, Anderson RC, et al. The natural history of isolated ventricular septal defect: A serial physiologic study. Circulation 1961;24:1372­1387. 26. Moss AJ, Adams FH. In: Emmanouilides GC, Allen HD, Riemenschneider TA, Gutgesell HP (eds). Heart Disease in Infants, Children, and Adolescents, Including the Fetus and Young Adult, 5th ed. Baltimore: Williams & Wilkins, 1995:734. 27. Murphy DJ, Ludomirsky A, Huhta JC. Continuous wave Doppler in children with ventricular septal defect: Noninvasive estimation of pressure gradient. Am J Cardiol 1986;57:428­432. 28. Kouchoukos NT, Blackstone EH, Doty DB, et al (eds). Kirklin/Barratt-Boyes Cardiac Surgery, 3d ed. Philadelphia: Churchill Livingstone, 2003:880. 29. Richardson JV, Schieken RM, Lauer RM, et al. Repair of large ventricular septal defects in infants and small children. Ann Surg 1982;195:318­322. 30. Backer CL, Winters RC, Zales VR, et al. The restrictive ventricular septal defect: How small is too small to close? Ann Thorac Surg 1993;56:1014­1018. 31 Blackstone EH, Kirklin JW, Bradley EL, et al. Optimal age and results in repair of large ventricular septal defects. J Thorac Cardiovasc Surg 1976;72:661­679. 32. Bacha EA, Cao QL, Starr JP, et al. Perventricular device closure of muscular ventricular septal defect. J Thorac Cardiovasc Surg 126:1718, 2003. 33. Bass JL, Kalra GS, Arora R, et al. Initial human experience with the Amplatzer perimembranous ventricular septal occluder device. Cath Cardiovasc Intervent 2003;58: 238­239. 34. Reddy VM, McElhinney DB, Sagrado T, et al. Results of 102 cases of complete repair of congenital heart defects in patients weighing 700 to 2500 grams. J Thorac Cardiovasc Surg 1999;117:324­331. 35. Trusler GA, Mustard WT. A method of banding the pulmonary artery for large isolated ventricular septal defect with and without transposition of the great arteries. Ann Thorac Surg 1972;13:351­355. 36. Albus RA, Trusler GA, Izukawa T, et al. Pulmonary artery banding. J Thorac Cardiovasc Surg 1984;88: 645­653. 37. Pinho P, Von Oppell UO, Brink J, et al. Pulmonary artery banding: Adequacy and long-term outcome. Eur J Cardiothorac Surg 1997;11:105­111.

MHBD054-CH57[1077-1090].qxd 11/10/06 12:05 Page 1090 p-mac292 27A:MHBD054:Chapters:CH-57: TechBooks

Information

untitled

14 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

29472


You might also be interested in

BETA
BLUK111c01.tex
GATZOULS.PM5
untitled