Creation of brachial artery-basilic vein fistula

J THORAC CARDIOVASC SURG 1989;98:214-6

Creation of brachial artery-basilic vein fistula A supplement to the cavopulmonary shunt Since 1984 five patients who had previously had cavopulmonary shunts for cyanotic congenital heart disease have returned to the Birmingham Childrens' Hospital with increasing breathlessness andcyanosis. Two had had a classic Glenn (unidirectional) cavopulmonary shunt and three, a bidirectional shunt; aU shunts had been performed 5 to 18 years earlier. Each patient was reinvestigated and underwent creation of a brachial artery-basilic vein fistula. This procedure is known to have good patency rates when used in other situations (for example, for hemodialysis), but it has not been previously described for use in this context. Of the five patients, four had a considerable and sustained symptomatic improvement on follow-up of 12 to 49 months (mean 31 months), but the condition of one continued to deteriorate; that child has subsequently undergone a Fontan procedure, also with no improvement. Only one patient complained of coldness of the arm, and there were no other complications. We believe a brachial artery-basilic vein fistula has considerable advantages in terms of ease of operation and postoperative complications when compared with an axillary arteriovenous fistula, used as a supplement to an inadequate cavopulmonary shunt.

I. M. Mitchell, FRCS, D. W. Goh, FRCS, and L. D. Abrams, FRCS, Birmingham. England

h e concept of diversion of blood from the superior vena cava to the pulmonary circulation was first realized experimentally by Carlon, Mondini, and de Marchi I in 1951. The first clinical experience was recorded by Glenn- in 1958. He described the use of a superior vena cava-right pulmonary artery shunt, involving transection of the right pulmonary artery with end-to-side anastomosis to the superior vena cava, ligation of the azygos vein, and ligation of the superior vena cava distal to the anastomosis. Since the first report, several other series have been published3.4 showing good results, with a mortality rate of less than 10%. Since 1967 we have abandoned the classic Glenn operation in favor of a modification allowing bidirectional shunting of venous blood. This is achieved by side-to-side anastomosis of the proximal right pulmonary artery and the superior vena cava with preservation of pulmonary artery continuity and ligation of the From the Department of Cardiothoracic Surgery, Birmingham Childrens' Hospital, Ladywood, Birmingham, England. Received for publication July 14, 1988. Accepted for publication Dec. I, 1988. Address for reprints: I. M. Mitchell, FRCS, Department of Thoracic Surgery, East Birmingham Hospital, Bordesley Green East, Birmingham B9 SST, England.

214

azygos vein and superior vena cava distal to the anastomosis. Despite the success of both these operations in palliating certain types of cyanotic congenital heart disease, it is our experience, as elsewhere,' that in some patients there is a gradual reduction in oxygenation of the blood at around 5 to 10 years after the operation, as evidenced by increasing cyanosis and polycythemia. Oxygenation can be improved by creating an axillary arteriovenous fistula to increase flow through the cavopulmonary shunt." Brachial artery-basilic vein anastomosis is a widely recognized procedure to establish venous access for hemodialysis," S but to our knowledge it has not been previously described for use as a supplement to an inadequate cavopulmonary shunt. We believe it offers significant advantages over axillary arteriovenous fistula creation, and we report our experience in five cases. Patients and methods Clinical features of the five patients (three male and two female patients) are summarized in Table I. Ages ranged from 7 years 2 months to 27 years 6 months (mean 17 years I month). Two had classic Glenn (unidirectional) cavopulmonary shunts, in one case supplemented at 9 years by a systemic-pulmonary artery shunt with considerable improvement in symptoms for a further 9 years. In all patients brachial artery-basilic vein

Volume 98 Number 2 August 1989

Brachial artery-basilic vein fistula

2 15

Table I. Clinical features offive patients with brachial artery-basilic vein fistula

Patient

Type of CP shunt

Subsequent operation

Time interval from CP shunt to fistula creation

Date of operation

Age

Sex

Diagnosis

9 yr II mo

M

Tricuspid atresia

Bidirectional

8 yr I mo

3/14/84

2

26 yr

M

Unidirectional

18 yr

5/18/84

3

7 yr 2 mo

M

TGA, ASD. VSD Single ventricle, mitral stenosis

5 yr 4 mo

5/23/84

4

15 yr

F

Tricuspid atresia

Bidirectional

14 yr 3 mo

4/22/87

5

27 yr 6 mo

F

Tricuspid atresia, pulmonary atresia, TGA

Unidirectional

18 yr

4/30/87

Bidirectional

lnfundibulectomy

Systemic-LPA shunt 9 yr after Glenn operation

Outcome Minimal cyanosis and dyspnea, 4 yr 1 mo No symptoms, 3 yr 11 mo Fistula patent, but remains breathless and cyanosed despite subsequent Fontan No symptoms at I yr No breathlessness or cyanosis at I yr, but cold arm with mild claudication

CPo Cavopulrnonary; TGA. transposition of the great arteries; ASD. atrial septal defect; VSD. ventricular septal defect; LPA. left pulmonary artery.

anastomosis was performed because of the recurrence of breathlessness, cyanosis, and polycythemia 5 to 18 years (mean 12 years 9 months) after the initial operation. In each case the operation was performed with the patient under general anesthesia. A small incision was made along the medial border of the biceps, just below the anterior axillary fold. The brachial artery and basilic vein were identified and a I em side-to-side anastomosis created with continuous 7-0 Prolene suture (Ethicon, Inc., Somerville, N.J.). The vein was then doubly ligated distal to the fistula and venesection performed as necessary before closure. Postoperative recovery was uneventful in all patients.

Results

In all five patients the brachial artery-basilic vein fistula has remained patent on follow-up of 1 year to 4 years I month (mean 2 years 7 months), as evidenced by clinical examination and in one case by angiography. Four patients (80%) have reported considerable lessening of their symptoms and improvement in exercise tolerance, with an average fall in hemoglobin value of 5 gmjdl. Only one has had any adverse effects on the arm (coldness and claudication); however, these symptoms are minor and are far outweighed by the benefits conferred by the fistula. In no patient has swelling of the arm been a problem. In the one patient whose symptoms did not improve postoperatively, a Fontan procedure was subsequently performed, but also with little success. Deteriorating

symptoms, increasing pulmonary vascular resistance, and persistent right ventricular failure now make transplantation the only option. Discussion

The gradual deterioration in blood oxygenation that is known to occur 5 to 10 years after cavopulmonary anastomosis' is in part a consequence of diminished flow through the cavopulmonary shunt. Flow diminishes because of two factors: (1) increased pulmonary vascular resistance, either primary or as a result of increasing viscosity of the blood (polycythemia in response to hypoxia); (2) expansion of venous collaterals between the two cavae or, possibly, recurrence of a connection between the superior vena cava and the right atrium. The fall in blood oxygenation may also be due to intravascular pulmonary thrombosis and diminished flow through the main pulmonary artery resulting from increasing obstruction of the pulmonary outflow tract. Furthermore, with the classic Glenn (undirectionai) shunt, expansion of precapillary arteriovenous fistula in time, may allow more blood to bypass the alveolar capillaries, particularly in the right lower lobe, to which most of the shunted blood gravitates." Although arteriovenous fistulas are not prevalent, 10. II any reduction in flow through the shunted lung is significant, as at best only 33% of the systemic venous return passes to the

2 16

The Journal of Thoracic and Cardiovascular Surgery

Mitchell. Goh. Abrams

right lung, which accounts for 60% of the pulmonary vascular bed. Several options exist for improving oxygenation. 1. Ligation or embolization of collateral venous channels or pulmonary arteriovenous fistulas. 1J. 14 2. Systemic-pulmonary artery shunting of the contralateral lung after unidirectional (Glenn) shunts. This procedure can give good palliation but depends on the presence of a left pulmonary artery of adequate size. 3. Definitive correction of the underlying cardiac anomaly. 4. Axillary or brachial arteriovenous fistula creation.

Elevated pulmonary vascular resistance, with its concomitant decrease in flow and increase in superior vena caval hypertension, has been clearly shown to fall when the blood flow to the lungs is converted from nonpulsatile to pulsatile." Since flow through a Glenn shunt is nonpulsatile, the creation of an axillary or brachial arteriovenous fistula improves pulmonary flow and leads to a fall in pulmonary vascular resistance. With a bidirectional cavopulmonary shunt, flow is already pulsatile to varying extents (depending on the patency of the right ventricular outflow tract); however, creation of an arteriovenous fistula enhances this pulsatility and therefore may increase flow more than might be expected. One further advantage of an arteriovenous fistula is that it increases in size over time, and there is therefore no sudden increase in load on the pulmonary circulation. Blood oxygenation can be expected to continue improving for some months after the operation, and this in tum substantially reduces plasma viscosity. Furthermore, the creation of an arteriovenous fistula is associated with few short-term or long-term complications, and in the event of failure and the return of symptoms at a later date, the shunt will not prevent or hinder subsequent definitive operations to correct the underlying cardiac anomaly, for example, a Fontan procedure. The advantage of a brachial artery-basilic vein anastomosis over an axillary arteriovenous fistula is threefold. First, it is an easier operation to perform. It avoids a dissection of the axillary artery in the region of the brachial plexus and can be done under local anesthesia if necessary. Second, it is a more cosmetic procedure, since it requires a small and unobtrusive incision on the inner aspect of the upper arm and does not leave a scar on the front of the chest. Third, as the distal part of the basilic vein can be safely ligated, it may produce less venous hypertension in the arm. In summary, we believe that brachial artery-basilie vein fistula creation is a useful method of improving and

sustaining improvement in blood oxygenation in situations in which a cavopulmonary shunt is becoming inadequate. We think it has significant advantages over the previously described technique of axillary arteryvein anastomosis. REFERENCES 1. Carlon CA, Mondini PG, de Marchi R. Surgical treatment of some cardiovascular diseases (new vascular anastomosis). J Internat Coli Surg 1951;16:1-11. 2. Glenn WWL. Circulatory bypass of the right side of the heart. IV. Shunt between superior vena cava and distal right pulmonary artery: report of clinical application. N Engl J Med 1958;259:117-20. 3. Glenn WWL, Browne M, Whittemore R. Circulatory bypass of the right side of the heart. Cavopulmonary artery shunt: indications and results (report of a collected series of 537 cases). In: The heart and circulation in the newborn and infant. New York: Grune & Stratton, 1966;345-57. 4. Glenn WWL, Hellenbrand WE, Henisz A, et al. Superior vena cava to right pulmonary artery anastomosis: present status. In: Obstructive lesions of the right heart. Baltimore: University Park Press, 1984:121-34. 5. Glenn WWL. Superior vena cava-pulmonary artery anastomosis. Ann Thorac Surg 1983;37:9-11. 6. Glenn WWL, Fenn JE. Axillary arteriovenous fistula: a means of supplementing blood flow through a cavapulmonary artery shunt. Circulation 1972;54:1013-7. 7. Dagher FJ. The upper arm AV haemoaccess: long-term follow-up. J Cardiovasc Surg 1986;27:447-9. 8. Breza J. Construction of fistula between a. brachialis and v. basilica transposed into subcutaneous tissue of the arm as a preparation for haemodialysis treatment. J Czech Med 1986;9:38-43. 9. Martinez-Catinchi FL, Johnson CD. Development of rib notching after a cavopulmonary anastomosis. Chest 1976; 70:777-9. 10. di Carlo 0, Williams WG, Freedom RM, Trusler GA, Rowe RD. The role of cavopulmonary (Glenn) anastomosis in the palliative treatment of congenital heart disease. J THORAC CARDIOVASC SURG 1982;83:437-42. 11. Pennington DG, Nouri S, Ho J, et al. Glenn shunt: long-term results and current role in congenital heart operations. Ann Thorac Surg 1981;31:532-9. 12. Furuse A, Brawley RK, Gott VL. Pulsatile cavo-pulmonary artery shunt: surgical technique and haemodynamic characteristics. J THORAC CARDIOVASC SURG 1972; 63:495-500. 13. Robicsek F. An epitaph for cavopulmonary anastomosis. Ann Thorac Surg 1982;34:208-20. 14. Gomes AS, Benson L, George B, Laks H. Management of pulmonary arteriovenous fistulas after superior vena cava-right pulmonary artery (Glenn) anastomosis. J THORAC CARDIOVASC SURG 1984;87:636-9.