| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;59:933-940
© 1995 The Society of Thoracic Surgeons
Sections of Pediatric Cardiology and Cardiovascular Surgery, and the Congenital Heart Registry, University of Chicago, Chicago, Illinois, and Children's Hospital of San Diego, San Diego, California
Accepted for publication November 21, 1994.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Despite improved surgical techniques and better understanding of morbid anatomy, pulmonary atresia with intact ventricular septum (PAcIVS) continues to have high mortality [1–4]. Current controversy centers around the right heart: (1) when is the right ventricle (RV) adequate to accept all of the systemic venous return, and (2) what should be done if there are right ventricle-to-coronary artery connections (RV-CACs)?
In 1984 [5], we reported surgical closure of the tricuspid valve (TVC) in a patient with PAcIVS and large RV-CACs. The present report assesses our experience with TVC in 10 children with PAcIVS and large RV-CACs.
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Our criteria to separate clinically ``important'' [6] from unimportant RV-CACs were qualitative rather than quantitative. Angiographic features on a combination of RV contrast injection and coronary angiography included (1) anatomically large RV-CAC with no point in the system smaller than the largest coronary artery caliber (Fig 1D
), (2) filling the entire length of one or more main coronary arteries (Fig 1E), (3) pronounced retrograde filling of the aorta (see Fig 1D), and (4) visualization of the coronary sinus (Fig 1F) on recirculation of contrast. All patients met one or more of the above criteria.
|
), the RV-CACs communicated with both coronary arteries in 7 and only the left coronary artery in 3. None had angiographic evidence of coronary artery stenosis in any branch. All had hypoplasia of the tricuspid valve with a median Z value of -3.7 (range, -2.6 to 5.2). Four patients had associated problems: chronic lung disease after birth at 28 weeks' gestation (patient 6), postoperative discontinuous pulmonary arteries (patient 10), and hypoplasia of one or both branch pulmonary arteries (patients 4 and 7). Five of the 10 patients had Rashkind balloon atrial septostomy performed before TVC; 5 had surgical atrial septectomy concurrent with TVC.
|
), and echocardiography was used to assure cessation of flow across the tricuspid valve. Left ventricular end-diastolic pressure was assessed and an increase in right ventricular oxygen saturation was confirmed. Aortography or selective coronary angiography was performed while the tricuspid valve was held closed.
|
), the patch was sewn to the base of the septal leaflet of the tricuspid valve; care was taken to place the sutures quite shallowly in the endocardium to avoid damage to the penetrating bundle of His and the right coronary artery.
|
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
). No patient had heart block postoperatively. Three patients died perioperatively. Patient 3 was a 2-day-old infant who had obliteration of the RV cavity and RV-CACs concurrent with TVC and atrial septectomy. She died in the operating room; permission for autopsy examination was denied. Patient 9 was a neonate who had TVC, imbrication of the RV, atrial septectomy, and systemic--pulmonary shunt; she died with low cardiac output 12 hours after operation. Autopsy revealed multiple foci of myofiber degeneration and necrosis in the left ventricular myocardium. Although more concentrated in the subendocardial regions, these changes were diffuse and included the conduction system. Marked dilation of intramyocardial venous and lymphatic channels also was noted. It is possible that extensive intrauterine damage to the myocardium may have made this patient's condition ``inoperable''.
In patient 10, TVC was recommended for several years. When she went into severe congestive heart failure at 11 years of age with a left ventricular end-diastolic pressure of 31 mm Hg, the family agreed to operation; she too died on the operating table and the family refused autopsy permission. It is probable that chronic ischemia coupled with volume overload from systemic--pulmonary shunts caused irretrievable myocardial failure.
Four of 10 patients had ``ischemic changes'' on ECG. Three survived operation and in each the ECG changes resolved. Long-term follow-up of the 7 survivors of TVC reveals that 3 have died, none related to myocardial dysfunction (Table 2). The original patient (patient 1) died of a perioperative junctional ectopic tachycardia after a Fontan operation 3 years after TVC. The child with chronic lung disease (patient 6) died at home at age 16 months (see Fig 3B); postmortem examination of the lungs revealed diffuse lymphangiectasia, areas of chronic focal inflammation, and extensive bilateral pleural fibrosis. Patient 7 died at 13 months of age after multiple attempts to enlarge bilaterally hypoplastic branch pulmonary arteries. Two survivors have had a successful Fontan operation and 1 awaits this procedure with a functioning cavopulmonary anastomosis. Follow-up echocardiographic studies of the right ventricle show no change in cavity size-estimated at 2 to 4 mL up to 7 years after TVC.
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Multiple different terms have been used to describe the channels between RV and coronary arteries, including fistula [3] and sinusoid [15]. A ``fistula'' connects a hollow organ to another hollow organ or to the surface whereas a ``sinusoid'' is a terminal vessel [16]; neither of these is correct if applied to the cavity-to-coronary artery communications in pulmonary atresia with intact ventricular septum. We prefer the more ambiguous but accurate word connection.
Despite numerous reports about the RV-CACs in PAcIVS, information is incomplete and sometimes contradictory. The frequency of RV-CACs in PAcIVS has been reported as 17% to 75% [1, 2, 11, 17]; it was 56% in our series. Right ventricle-to-coronary artery connections can be found as early as 23 weeks' gestation [18], suggesting that a neonate could be delivered with an already ischemic myocardium. Both Grant [19]-in the original paper on RV-CACs-and Freedom and Harrington [15] have shown that anomalous channels can connect the RV with both coronary arteries and coronary veins. Some authors do not distinguish small from large RV-CACs.
Our current state of knowledge is quite limited in assessing the hemodynamic importance of RV-CACs. These connections have been linked statistically to left ventricular dyskinesia [20] and seem to be an incremental risk factor for mortality [1]. For the individual patient, the only criteria currently used to assess the physiologic effect of RV-CACs are the ECG and the angiographic appearance. We believe that the effect of temporary TVC at diagnostic cardiac catheterization is an important additional datum. If there is RV-dependent myocardial perfusion, balloon closure of the tricuspid valve will unmask it, as this maneuver stops RV-to-anomalous channel-to-coronary artery flow; one would expect acute development of ischemic changes in the ECG. Conversely, because temporary TVC also stops flow from native coronary artery to the RV, any ``steal'' phenomenon also would cease; this can manifest as improvement in the ECG (see Figs 2B, 2C).
At present, one cannot predict future behavior of RV-CACs. Both spontaneous regression [15] and enlargement [10] have been reported. Operation on the RV has a variable effect on RV-CACs [10, 17, 21]; sudden death has been reported after both RV decompression [22] and RV exclusion [4].
Although the determination of which RV-CACs are important may be subjective or semiquantitative, there was little difficulty in our experience distinguishing small channels from large, unrestrictive connections (30% in our series). Angiography in the aorta or the right atrium is often inadequate to define RV-CACs (Figs 4A, 4B). Contrast medium should be delivered exclusively into the RV. Because the RV cavity is usually very small, a catheter with injection ports limited to the end of the catheter (eg, Goodall-Lubin) (see Fig 1A) is preferable to regular angiographic catheters (eg, Berman) as the latter may have proximal injection ports in the right atrium when the catheter tip and distal holes are in the RV.
|
).
Our current reasoning and therapeutic planning in children with PAcIVS and important RV-CACs is described below and in Figure 5.
|
Flow Through RV-CACs Is Detrimental in Both Antegrade and Retrograde Directions
In systole, deoxygenated blood is delivered to the coronary artery system. During diastole, the RV-CACs provide a lower resistance circuit than the myocardium itself; thereby aortic-derived (oxygenated) blood is ``stolen'' away from the tissues. It is further possible that this reciprocal flow pattern may promote the development of coronary artery stenoses, which usually are located in immediate proximity to the insertion of the anomalous channel into the native coronary artery.
Preservation of Left Ventricular Myocardium Is Paramount
We believe that attention should be directed more to the left ventricle than the right. Early, severe left ventricular myocardial abnormalities have been documented in PAcIVS even without RV-CACs [23]. Left ventricular dysfunction is a known risk factor for Fontan-type operations [13, 14], and dyskinesia has been shown to be associated both with PAcIVS in general [24] and specifically when RV-CACs are present [9, 20].
In a recent article from Boston, Gentles and associates [25] studied the effect of RV decompression on left ventricular wall motion in patients with PAcIVS and RV-CACs. They found that reducing RV pressure did not cause perioperative mortality unless bilateral coronary artery stenoses were present; therefore, the Boston group recommends RV decompression unless there are bilateral stenoses of the native coronary arteries. However, Gentles and associates also reported exacerbation of left ventricular wall dyskinesia after RV decompression. We prefer RV exclusion (TVC), noting that anything harmful to left ventricular function ultimately reduces the long-term outlook for the patient. (It is better to have one healthy ventricle than two sick ones.)
When There Are Important RV-CACs, Growth of the Right Ventricle Is Neither Practical Nor Desirable
First, children with PAcIVS and important RV-CACs invariably have a small (``prune-pit'') RV cavity making an ultimate biventricular repair unlikely. Second, the tricuspid valve is always hypoplastic [2], which in turn prevents adequate inflow to encourage RV growth. Most important, the act of venting or decompressing the RV to encourage growth also encourages greater coronary-to-RV flow, thereby increasing left ventricular ischemia and dysfunction [12, 22, 25].
Decompression of the Right Ventricle in Patients With Important RV-CACs Risks Sudden Death by Acute Ischemia
Whether by tricuspid valve avulsion or excision [17, 21] or by opening the RV outflow tract, reduction in RV pressure is likely to increase coronary artery-to-anomalous channel-to-RV diastolic flow when the channels are large; this in turn acutely (and chronicly) reduces myocardial perfusion. Without a method to quantify precisely such flow, RV decompression has an unpredictable effect on left ventricular myocardial perfusion, which can be lethal [22].
Balloon Septostomy Should Be Done at the Initial Neonatal Catheterization
Some might recommend avoiding a Rashkind procedure as decompression of the right atrium reduces the likelihood of RV growth. However, because patients with important RV-CACs are destined for a ``three-chambered'', Fontan-type repair, RV growth is unnecessary. These patients need right atrial decompression and usually benefit from the augmentation of systemic output achieved by balloon atrial septostomy.
Early, Detailed Delineation of Both the RV-CACs and the Native Coronary Arteries Must Be Accomplished
Contrast injection limited to the RV cavity is critical to delineating the RV-CACs. This usually requires use of a bird's eye catheter rather than an angiocatheter. Right ventricular injection should be done at neonatal catheterization for comparison with a second study at 4 to 6 months of age. Although an ascending aortogram at first catheterization may be adequate to visualize the origins of native coronary arteries, at follow-up study, selective coronary arteriography is advised.
A Systemic--Pulmonary Shunt Is Recommended as the Sole Neonatal Surgical Procedure
We prefer a 4-mm H-type subclavian-to-pulmonary artery Gore-Tex graft to minimize transmission of pressure into a pulmonary artery system that ultimately will be part of a Fontan procedure. Additionally, in those in whom spontaneous discontinuity of the pulmonary arteries might develop (Waldman JD, unpublished data) a shunt larger than 4 mm could be highly detrimental.
Repeat Cardiac Catheterization Is Recommended at 4 to 6 Months of Age
This delay is chosen to give some chance for spontaneous involution of the RV-CACs and to allow normal regression of pulmonary vascular resistance. Specific purposes of this second study include (1) subselective coronary arteriography to assure absence of stenoses, (2) therapeutic trial of TVC to assess ECG effects, and (3) confirmation of low pulmonary artery pressure. Coronary arteriography is done during temporary balloon occlusion of the tricuspid valve to enhance visualization of the native coronary arteries.
Tricuspid Valve Closure Is Recommended by 6 Months of Age
Right ventricle-to-coronary artery connection flow promotes left ventricular ischemia and may be a factor in the progression of abnormalities of the native coronary arteries. Therefore, the sooner this flow is interrupted, the better for both the coronary arteries and the myocardium. Concurrent procedures with this operation include atrial septectomy (if needed), take-down of the systemic--pulmonary shunt, and cavopulmonary anastomosis. Although a small RV cavity remains, ectopic dysrhythmias have not been observed in a follow-up as long as 7 years.
Although one hesitates to recommend a procedure such as TVC that had a 30% perioperative mortality, careful review suggests that at least two of our three deaths were in children whose condition was ``inoperable'' because of loss of left ventricular myocardium before operation [23]. Furthermore, the late deaths (see Table 2) were unrelated to the RV. Therefore, when compared with the very poor outcome of other approaches [3], TVC seems promising.
Fontan Operation Should Be Done Using the Usual Criteria Developed for This Approach
This is true with respect to timing, patient selection, and handling of the interatrial communication. Those who had TVC should be grouped with those who had tricuspid atresia. Patients with PAcIVS and RV-dependent coronary circulation-who are not candidates for TVC-should be considered for a Fontan operation modified [4] to baffle caval blood to the pulmonary arteries but allowing pulmonary venous blood to enter both mitral and tricuspid valves.
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Address reprint requests to Dr Waldman, Section of Pediatric Cardiology, The University of Chicago, Wyler Children's Hospital, 5841 S Maryland Ave, Chicago, IL 60637.
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  J.-H. Yang, T.-G. Jun, P. W. Park, K. Sung, W. S. Kim, Y. T. Lee, J. Huh, and I-S. Kang Exclusion of the non-functioning right ventricle in children with pulmonary atresia and intact ventricular septum Eur. J. Cardiothorac. Surg., February 1, 2008; 33(2): 251 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Guleserian, L. B. Armsby, R. R. Thiagarajan, P. J. del Nido, and J. E. Mayer Jr Natural History of Pulmonary Atresia With Intact Ventricular Septum and Right-Ventricle-Dependent Coronary Circulation Managed by the Single-Ventricle Approach Ann. Thorac. Surg., June 1, 2006; 81(6): 2250 - 2258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ono, S. Otake, N. Fukushima, Y. Sawa, H. Ichikawa, K. Kagisaki, and H. Matsuda Huge right ventricle-right coronary artery fistula compromising right ventricular function in a patient with pulmonary atresia and intact ventricular septum: A case report J. Thorac. Cardiovasc. Surg., November 1, 2001; 122(5): 1030 - 1032. [Full Text] [PDF]
|
|
|
|
|
|
T. Asou, K. Matsuzaki, K. Matsui, T. R. Karl, and R. B.B. Mee Veno-venous bypass to prevent myocardial ischemia during right heart bypass operation in PA, IVS, and RV dependent coronary circulation Ann. Thorac. Surg., March 1, 2000; 69(3): 955 - 956. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. W. McCrindle Commentary J. Thorac. Cardiovasc. Surg., December 1, 1999; 118(6): 1052 - 1055. [Full Text] [PDF]
|
|
|
|
|
|
H. K. Najm, W. G. Williams, J. G. Coles, I. M. Rebeyka, and R. M. Freedom Pulmonary Atresia With Intact Ventricular Septum: Results of the Fontan Procedure Ann. Thorac. Surg., March 1, 1997; 63(3): 669 - 675. [Abstract] [Full Text]
|
|
|
|
|
|
S. Sivasubramanian and J. D. Waldman Pulmonary Atresia, IVS, and RV-CAC: Can CABG Alter Choice of Repair? Ann. Thorac. Surg., September 1, 1996; 62(3): 949 - 950. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
