The Cavopulmonary Anastomosis: Evolution of a Palliative Technique

Introduction

For many years pulmonary blood flow was considered to be exclusively dependent on the contractile function of the right ventricle (RV). Then experimental and clinical evidence demonstrated that under certain conditions it was possible to have adequate pulmonary blood flow even if the pumping action of the RV was severely impaired or even absent. ¹ This observation led clinicians at the University of Padova in Italy to develop a new procedure. This procedure was proposed to palliate congenital cardiac anomalies associated with decreased pulmonary blood flow by delivering venous blood directly into the pulmonary circulation, bypassing the RV. ² In this setting, direct connection of the superior vena cava to the right pulmonary artery, the Cavopulmonary anastomosis (CPA), became one of the most popular palliative procedures for many years after its clinical introduction. This article reviews the history of this milestone operation focusing on its birth, widespread adoption, subsequent decline, and more recent revival as a bidirectional CPA.

Background

The studies on the pulmonary circulation carried out by William Harvey almost 400 years ago supported the dual concept of the pump function of both the RV and the lungs during respiration. Indeed, as Harvey stated “The continual rising and falling of the lungs in respiration must open and close the vessels and porosities as in a sponge or a thing of similar structure, when it is compressed and allowed to expand.” ² Such studies were certainly instrumental for the subsequent development of the Cavopulmonary anastomosis.

The feasibility of bypassing the RV was first demonstrated experimentally by Rodbard and Wagner in 1949, by anastomosing the right atrial appendage to the pulmonary artery after ligating the main pulmonary artery and bypassing the RV in both acute and chronic preparations in dogs. ³ The acute experiments demonstrated that “the venous pressure may operate as a vis a tergo (latin - a force from behind) to pump at least some of the venous return through the lungs without benefit of right ventricular action,” confirming the low resistance of the pulmonary circuit. In the chronic experiments “with production of the atrio-pulmonary shunt, a functional tricuspid regurgitation is produced because of the flow of some of the output of the right ventricle via the pulmonary artery anastomosis to the right atrium.” This, however, did not produce any deleterious effect at a two-month observation on such dogs. ³ Interestingly, the concept of a direct anastomosis between the right atrial appendage and the right ventricular chamber would not be successfully applied clinically until 30 years later by Björk et al to correct tricuspid atresia. ⁴ Others had previously adopted modifications of the same principle by interposing a pulmonary homograft between the right auricle and the pulmonary trunk ⁵ or to a pulmonary artery branch as part of a more complex procedure. ⁶

Birth of the Cavopulmonary Anastomosis(CPA)

The CPA is a palliative procedure aimed to increase the pulmonary blood flow in congenital cardiac malformations characterized by an impaired or obstructed pulmonary circulation. The idea of the CPA was first developed by Carlo A. Carlon (Figure 1), together with two colleagues, Paolo G. Mondini and Renato De Marchi, in a series of experiments performed on a canine model at the University of Padua in Italy. The initial results of such experiments were presented at a meeting of the International College of Surgeons held in May, 1950. ⁷ According to the original drawings describing the technique, creation of a CPA was performed using these four steps (Figure 2). The first step included isolation, ligation and division of the azygos vein (Figure 2A). This was followed by isolation, ligation, and division of the right pulmonary artery as the second step (Figure 2B). In the third step, an anastomosis between the azygos vein stump and the right pulmonary artery was started (Figure 2C). In the last step the suture line was completed, the vascular clamps were removed, and the superior vena cava was ligated just before entering the right atrium (Figure 2D). At the end of the procedure, the blood from the superior vena cava flows directly into the right lung through the azygos vein into the right pulmonary artery. This procedure was based on the belief that “it was possible and much more convenient to increase this blood flow with venous blood instead of arteriovenous blood as used by Blalock.” ⁸ The experiments, performed on a series of dogs, were quite troublesome and plagued by many failures. Indeed, initially an end-to-end anastomosis between the superior vena cava and the right pulmonary artery was attempted but it was unsuccessful, most likely because dogs did not tolerate sudden closure of the superior vena cava. Other causes of failure were thrombosis and size mismatch between the two vessels. Finally, the first successful operations were obtained when the azygos vein was sequentially connected to the right pulmonary artery using the steps noted above and the dogs were anticoagulated. ⁸

OPEN IN VIEWER

Figure 1.

Carlo Adolfo Carlon (1919-1991).

OPEN IN VIEWER

Figure 2.

Schematic drawing of the original cavopulmonary anastomosis as described by Carlon et al. ⁷ (Reproduced with permission by Edizioni Idelson Gnocchi 1908 Srl, Naples, Italy). See text for explanation.

Carlo A Carlon, at the time of such experiments, was associated with the Department of Surgery, at the University of Padua, Italy. ⁹ He was basically a general surgeon particularly famous for his skills in performing gastric resections and vagotomies for the treatment of peptic ulcer. In the late 1940s, he went to the United States becoming interested in the management of valvular heart disease and particularly in mitral stenosis. After his return to Padua, he performed over 200 mitral commissurotomies reporting gratifying results. ¹⁰

The concept of the CPA was also utilized by Glenn et al a few years later for similar experiments. ¹¹ As the authors described “An anastomosis was then performed between the stump of the central end of the azygos vein and the distal end of the divided right main pulmonary artery. The cardiac end of the right pulmonary artery was ligated. Following the completion of the anastomosis, the superior vena cava was occluded as it entered the right heart by a double ligature of heavy silk.” ¹¹ The same authors performed a CPA in the clinical setting in a seven-year-old cyanotic boy with a functional single left ventricle, transposition of the great arteries, and severe pulmonary stenosis. ¹² The authors later reported their results with a series of 38 children with various congenital cardiac anomalies. ¹³ Such reports contributed to the popularity of the CPA which became known as the “Glenn anastomosis” without recognizing Carlon's pioneering work. ⁹ Others had performed similar experiments or reported clinical results with the CPA, also providing important steps in the evolution of the CPA.^14–18 In particular, the first application to a patient was performed in Russia by Eugenii Meshalkin on April 3,1956, ¹⁴ while Glenn's first patient received the CPA on February 25, 1958, at Yale University. ¹² This indicates that the CPA was developed experimentally and introduced into clinical practice by many surgeons, working independently, and most likely unaware of each other's attempts. ¹⁹ Indeed, many milestone contributions by others in this specific field have been recognized and previously underlined. ⁹

Decline of the CPA

The Italian contribution to the CPA has been for many years overlooked, most likely because it was published in scientific journals with limited circulation rarely available to researchers from other countries and cardiac surgeons.^7,8 Nevertheless, it was eventually recognized by Robicsek and others^2,19 who gave adequate credit to Carlon's work more than 30 years after his first publication. ⁷

The CPA gained rapid acceptance among other surgical procedures used to palliate children with certain congenital cardiac anomalies. The initial reports showed excellent early results following CPA but there was little medium- or long-term data. ² Nevertheless, some recent reports indicate that the CPA can provide adequate palliation also in a long-term follow-up period.^20–22 It must be noted, however, that in some of these patients a subsequent systemic artery-pulmonary artery shunt had to be created despite initial satisfactory results due to gradual clinical deterioration. ² In this respect, it is noteworthy that that there was a 40% survival at 40 years after the Glenn anastomosis, without a subsequent Fontan procedure, in patients with single-ventricle physiology, which demonstrates the effectiveness of the CPA as an isolated procedure.^23,24

Despite these seemingly satisfactory results, interest in the CPA as described by Cordon steadily declined. This procedure lost its popularity for a number of reasons, well underlined by Robicsek in his historical review of the CPA. ² He recognized that the CPA was considered by some surgeons to be an irreversible procedure and one that provided poor results in the small infant. ² But, more importantly , there are two main reasons explaining why the CPA was progressively abandoned and replaced by the bidirectional CPA: (1) demonstration that even complex congenital anomalies could be managed with complete single-stage repair (even in newborns); thus, eliminating the need for palliative procedures, and (2) development of more radical procedures to correct specific cardiac malformations for which the CPA was previously considered to be a procedure of choice.^4–6 Needless to say, when Carlon, Glenn, Robicsek, and others^7,8,11–17 were experimenting with such an innovative procedure, none of them were aware of the importance for the first pass into the lungs of the inferior vena cava blood that carried an important angiogenesis inhibitor factor which is responsible for preventing pulmonary arteriovenous malformations. ²⁵

Nevertheless, although somewhat different from the original description, a revival of the CPA was achieved when Fontan and Baudet applied a direct CPA as part of a procedure for children with tricuspid atresia. ⁶ The pioneers of the original CPA (Unidirectional SVC to RPA)could not foresee such a brilliant future of the Bidirectional CPA which has nowadays become an extremely common surgical procedure for staging children in preparation for a Fontan-Kreutzer operation.^26,27

Conclusions

Cavopulmonary anastomosis is an historical milestone in the treatment of congenital heart malformations. Carlo A. Carlon and his associates must definitely receive credit for being the first to demonstrate experimentally the feasibility of this procedure although they never applied it clinically. ²⁸ Such experiments suggested the possibility that in the animal laboratory complete long-term circulatory exclusion of the right heart is possible. Seventy years ago, it was demonstrated that the CPA was also an effective palliative operation in the clinical setting. We believe that Carlo A. Carlon deserves a special mention in the cardiac surgeons’ hall of fame, particularly remembering that the University of Padua, of which he had been one of the most prestigious graduates, has just celebrated the 800th anniversary of its foundation, which occurred in 1222.