Central aorta-pulmonary artery shunts in neonates with complex cyanotic congenital heart disease

J THORAC CARDIOVASC SURG 1987;93:767-74

Central aorta-pulmonary artery shunts in neonates with complex cyanotic congenital heart disease Methods of palliating critical pulmonary oligemia in neonates with complex cyanotic congenital heart disease continue to evolve. Pulmonaryartery distortionand other complications of the IR of native vessels toincrease pulmonary blood flow hasled to the morefrequentuse of polytetrafluoroethylene shunts either in a centralposition or as a modified Blalock-Taussig shunt. Central aorta-pulmonary artery shunts have largely fallen into disfavor becauseof previously reported unacceptably high incidences of complications such as shunt thrombosis, congestive heart failure, and pulmonary artery distortion. This report details our experience palliating 23 neonates with pulmonary atresia or severe pulmonary stenosis by placing central aorta-pulmonary artery shunts utilizing a short segment «1 cm) of polytetrafluoroethylene. Although three of the 23 died postoperatively, noneof the 23 patients had evidence of shunt thrombosis. Congestive heart failure, a potentialcomplication of any pulmonary artery shunt, was present in eight of the 20 survivors but did not require shunt takedown and was readily controlled by digoxin. Repeat catheterization was performed in 12 patients; pulmonary angiography showed good growth of hoth pulmonary arteries and there was no evidence of pulmonary artery hypertension. Although minor pulmonary artery distortion was present in two patients, this distortion was centrally located and easily remedied at the time of total correction. Thus we havefound the central aorta-pulmonary artery shunt to be an extremely effective and reliable means of palliating pulmonary artery hypoplasia as a result of pulmonary atresia or severe pulmonary stenosis in neonates.

Thomas P. Barragry, M.D., W. Steves Ring, M.D., James W. Blatchford, M.D., and John E. Foker, M.D., Ph.D., Minneapolis. Minn.

Since the introduction of the Blalock-Taussig procedure in 1945,1 the systemic-pulmonaryartery shunt has been a mainstay of surgical therapy for many children with cyanotic congenital heart disease. Subsequently central-type aorta-pulmonary artery shunts were also devised, first by Potts, Smith, and Gibson,' then by Waterston,' and later by Cooley and Hallman.' The Potts anastomosis is associated with the development of pulmonary vasculardiseasebecause of excessive pulmonary blood flow and is difficult to dismantle at the time of total correction." The Waterston shunt gained wide acceptance because of its ease of construction in small infants. However, there have been numerous reports':" documenting the complications associated with aortaFrom the Department of Cardiovascular Surgery, The University of Minnesota Hospitals, Minneapolis, Minn. Received for publication April 14, 1986. Accepted for publication May 23, 1986. Address for reprints: Thomas P. Barragry, M.D., Box 35, University of Minnesota Hospitals and Clinics, 420 Delaware St. S.E., Minneapolis, Minn. 55455.

right pulmonary artery anastomosis. Shunt failure has been reported in 3% to 14% of patientsv"-" and congestive heart failure has been a problem in some series.l" II, 14, 15,18 Additionally, kinking and distortion of the right pulmonary artery at the site of the anastomosis can cause preferential flow to the right lung, narrowing and obliteration of the right pulmonary artery proximal to the anastomosis, and lack of growth of the main and left pulmonary arteries.v" 17, 19,20 The distortion is probably the result of two factors: attachment of the right pulmonary artery high on the aorta and growth and realignment of the great vessels.21 Limitations of the Blalock-Taussig shunt in newborn infants include kinking of the subclavian artery at its origin, pulmonary artery tenting, ischemic sequelae to the ipsilateral upper limb,22.23 and frequent early occlusion necessitating early revision. 24,25 For these reasons, polytetrafluoroethylene (PTFE) prostheses have been employed either as modified Blalock-Taussig shunts or centrally between the ascending aorta and the pulmonary artery. Acceptance of the centrally placed ascending aorta-pulmonary artery shunt, however, has been 767

768

Barragry et al.

The Journal of Thoracic and Cardiovascular Surgery

Sutures Marking Site of Anastomosis Fig. 1. Diagram of the typical relationship between the aorta and pulmonary artery (PA). The proposed insertion sites are marked with sutures so that they can be accurately determined after placement of partially occluding clamps. In position, the graft directly bridges the gap between the vessels and is consistently less than I em in length. (From Kulik TJ, Foker lE, Lucas RV, Anderson RW, Lock lE. Postoperative Hemodynamics in Children With Polytetrafluoroethylene Shunts. Circulation 1981;64(Pt 2):II123-30, with permission of the American Heart Association, Inc.)

limited by the supposed high incidence of shunt failure and congestive heart failure induced by the shunt. -The following report details our clinical experience in 23 cyanotic neonates in whom PTFE was used to construct aorta-pulmonary artery shunts.

Patients and methods Twenty-three neonates underwent isolated central aorta-pulmonary artery shunts at the University of Minnesota Hospitals from December 1977 to November 1985. Mean patient age was 4.4 days (range I to 30 days) with 17 of the 23 being less than 1 week of age. The mean patient weight was 3.2 kg (range 1.7 to 4.0 kg). Eleven of the 23 were female and 12 were male. Underlying congenital cardiac lesions included pulmonary atresia (N = 19) or severe pulmonary stenosis (N = 4) associated with tetralogy of Fallot (N = 8), transposition complexes (N = 5), single ventricle (N = 5), intact ventricular septum (N = 3), and Ebstein's anomaly (N = 2). Initial arterial oxygen tension (with variable inspired oxygen concentrations) ranged from 22 to 47 mm Hg (mean 30 mm Hg). Preoperatively, all patients were treated with continuous intravenous infusion of prostaglandin E) (0.05 to 0.10 JLg/kg/min) to maintain ductal patency and improve pulmonary blood flow.

Fig. 2. The typical position of a shunt to the main pulmonary artery. The pulmonary artery is not distorted by the shunt. (From Kulik 'rr, Foker rs, Lucas RV, Anderson RW, Lock lE. Postoperative Hemodynamics in Children With Polytetraf1uoroethylene Shunts. Circulation 1981;64(Pt 2):II123-30, 1981 with permission of the American Heart Association, Inc.)

Operative technique. All operations were performed on an urgent basis through a median sternotomy. After the pericardium was opened and the main pulmonary artery was dissected, a decision was made to anastomose the aorta to the main pulmonary artery (N = 17) or, if the main pulmonary artery was too small, to the proximal portion of the right pulmonary artery (N = 4) or the left pulmonary artery (N = 2). The anastomotic sites on the vessels were chosen and marked with sutures (Fig. 1) so that the incision point could be accurately identified after placement of partially occluding clamps. A Castaneda clamp was placed on the pulmonary artery, although frequently, since only the distal end of the main pulmonary artery needed to be occluded, a Heifitz neurosurgical aneurysm clip provided unobtrusive control. A short segment of PTFE graft was anastomosed end to side to the pulmonary artery with running 7-0 polypropylene suture. In 16 cases, a 4 mm diameter PTFE graft was used; in the remaining seven cases, a 3.7 mm diameter PTFE graft was used. After completion of the pulmonary anastomosis, the partiallj occluding Castaneda clamp was placed on the aorta fOJ creation of the aortic anastomosis. After a longitudina aortotomy, running 7-0 polypropylene suture wa:

Volume 93

Central aorta-pulmonary artery shunts

Number 5

May 1987

70

769

~ 21i

cr 2.0 J,.tll

/

~ 1.2

60

c(

g' 1.1

~

Ol

11.0

50

~

1:

E E

40

>.

2

0 '"

.{

o,

O; .;:

>.

30

lij

~

<

.9

.8 .8

.5

:;

.4

1:

.3

c,

Ol

a:

10

0 Before Shunt

After Shunt

Fig. 3. Improvement in arterial oxygenation after construction of central aorta-pulmonary artery shunt (N = 23).

employed in the aortic anastomosis. The graft bridged the distance between the vesselswithout excessivelength or distortion of the pulmonary artery (Fig. 2). Before completion of the anastomosis, the graft was suctioned and irrigated with heparinized saline to remove any possible thrombus within the graft. Eleven of the 23 patients also had ligation of a patent ductus arteriosus. Statistical analysis was performed with Student's paired t test.

Results The postoperative arterial oxygen tension (mean 59 mm Hg, range 49 to 84 mm Hg) with the patient breathing room air was higher (p < 0.(05) than the preoperative level (Fig. 3). Early graft thrombosis did notoccur. Three of the 23 (13.3%) neonates died within 24hours of operation. At postmortem examination, two of these three patients were found to have tricuspid insufficiency resulting from tricuspid valve dysplasia. Accompanying findings included pulmonary atresia with intact ventricular septum and marked right atrial and right ventricular dilatation in one patient, whereas the other of the two had pulmonary atresia with right atrial and right ventricular hypertrophy: The third operative death occurred in a premature neonate weighing 1,700 gm who had pulmonary atresia, severe right ventricular hypoplasia, and Ebstein's anomaly. Significantly, patent aorta-pulmonary artery shunts were

.6

<: 0

E

20

.7

{

e I

I

Before Shunt

After Shunt

Fig. 4. Changes in the ratio of the right pulmonary artery diameter to the descending aorta diameter in 12 patients undergoing follow-up catheterization. Solid circles (e) represent patients with 4.0 mm grafts and open circles (0) represent patients with 3.7 mm grafts.

found at postmorten examination in each of these three patients. These deaths occurred fairly early in our experience with central shunting in neonates. The most recent of these was in January 1982; thus the last 16 consecutive shunts in neonates have been performed without operative mortality. Two late deaths occurred. One patient with tetralogy of Fallot, pulmonary atresia, and diGeorge syndrome died of sepsis 5 months postoperatively, and a second patient with atrioventricular canal, pulmonary atresia, transposition of the great vessels, and tracheomalacia died suddenly of probable arrhythmia 32 days after tracheopexy and 2 months after central shunt creation. Both patients had patent shunts at autopsy. All longterm survivors were followed up from 6 to 49 months postoperatively. Digoxin was used in patients with clinical evidence of congestive heart failure and maintained for variable periods of time. Eight of the 20 survivors were maintained on a regimen of digoxin for periods of time ranging from 6 to 27 months on an outpatient basis. Four children continued to receive digoxin until the time of their total repair. Although postoperative congestive heart failure was frequent (40%), the failure was easily controlled by digoxin, and no patient required shunt takedown or readmission for treatment of congestive heart failure. Other postoperative complications involved the respiratory tract and included reintubation in two patients, one for atelectasis and the other for episodes of apnea and bradycardia.

The Journal of

770

Thoracic and Cardiovascular

Barragry et al.

Surgery

1.5 .Q

1.4


1.3

t:

'"

1.2

Ol

1.1

0

«

c '0 c

Ql

o

(/)

Ql

0

1.0

.9

.9

8

ti;

.7

>.

~

>.

.6

c 0 E

.5

a.

.4

.0;

.3

~

"3

c

~

e

1

I Before Shunt

I

After Shunt

Fig. 5. Changes in the ratio of the main pulmonary artery diameter to the descending aorta diameter in 12 patients undergoing follow-up catheterization. Solid circles (e) represent patients with 4.0 mm grafts and open circles (0) represent patients with 3.7 mm grafts.

Another patient required ventilatory support until the seventh postoperative day. One patient had intravenous catheter-related sepsis, which resolved with antibiotics and removal of intravenous catheters. Eight of the 20 patients received platelet-inhibitory agents after discharge. Aspirin was usually administered for at least 1 year, if not longer, and dipyridamole, when used, was usually maintained for less than 3 months. Presently all children are discharged on a regimen of aspirin and dipyridamole and continue to receive aspirin (one-half to one baby tablet per day) alone indefinitely. Angiocardiography. In general, indications for repeat catheterization were failure to thrive (especially in patients less than 6 months of age), progressively increasing hemoglobin (usually greater than 20 gmjdl), progressive exercise intolerance, or a combination of these indications. The interval between shunt placement and catheterization ranged from 2 to 47 months (mean 20 months) (Table I). Graft patency was demonstrated in all 12 patients who were catheterized, and no patient had elevated pulmonary artery pressure or pulmonary vascular resistance. Minor pulmonary artery distortion was present in the right pulmonary artery in a patient with an aorta-right pulmonary artery shunt 4 years after its creation. A second patient had stenosis of the main pulmonary artery 22 months after shunt place-

ment between the main pulmonary artery and the aorta. The mean ratio of the right pulmonary artery diameter to descending aorta diameter as seen on cineangiograms increased from 0.50 ± 0.03* to 1.05 ± 0.12 (p < 0.(03) (Fig. 4). The mean ratio of the main pulmonary artery diameter to descending aorta diameter increased from 0.61 ± 0.07 to 0.98 ± 0.08 (p < 0.003) (Fig. 5). The mean ratio of the left pulmonary artery diameter to descending aorta diameter increased from 0.63 ± 0.04 to 0.99 ± 0.10 (p < 0.007) (Fig. 6). The increase of these ratios in patients with 4.0 mm grafts was not statistically different from the increase noted in patients with 3.7 mm grafts. Ten of the 12 patients recatheterized had shunts from the aorta to the main pulmonary artery. The other two patients had shunts to a branch pulmonary artery, one to the right and the other to the left pulmonary artery. Although 11 of the 12 patients catheterized showed a substantial rise in the pulmonary artery-descending aorta ratio, the one patient (Patient I) with a shunt to the right pulmonary artery had no increase in the ratio for the main and right pulmonary arteries and a decrease in the ratio for the left pulmonary artery at catheterization 47 months after central shunt placement. Discontinuous pulmonary arteries were detected, for which he was treated with a left Blalock-Taussig shunt. In addition to pulmonary artery discontinuity, the right pulmonary artery was observed to be stenotic at the PTFE shunt insertion site; however, this patient has not required repair and remains well with a hemoglobin concentration of 15.5 gmjdl 6Y2 years after central shunt placement. Typical increases in pulmonary artery-descending aorta ratio were Seen in Patient 3, who underwent a shunt to the left pulmonary artery. Reoperation. Of the 18 long-term survivors, 12 have been reoperated on: Four underwent reshunting 4 to 47 months after initial shunting, and eight underwent repair 2 to 50 months after initial shunting. No patient was reoperated on for central aorta-pulmonary shunt thrombosis. None of the four patients who required operation within 1 year of shunt placement had reoperation because of inadequate pulmonary blood flow through the central shunt. Patient 2 (Table I) was admitted for respiratory distress 2 months after central shunt placement. A tracheopexy was performed to relieve severe compression of the distal trachea and area of the bifurcation that was seen on an angiogram. Thirty-three days after tracheopexy, this patient died suddenly, probably of an arrhythmia. His discharge had

*All ratios reported

as mean ± standard error of the mean.

Volume 93 Number 5 May 1987

been planned for the following week. Patient 3 had outflow tract patching performed at 3 months of age as a staged procedure to treat pulmonary atresia with intact ventricular septum. Our current policy for treatment of patients with pulmonary atresia and intact ventricular septum has since evolved such that outflow tract patching is performed in the neonatal period. Patient 4 with pulmonary atresia, intact ventricular septum,and Ebstein's anomaly had persistent congestive heart failure after central shunt placement, which necessitated the repair of his Ebstein's anomaly at 3 months of age. The central shunt was found to provide excellent pulmonary blood flow at catheterization before his Ebstein's repair. Patient 5 with pulmonary atresia and intact ventricular septum underwent placement of an aorta-pulmonary artery shunt into a 2 mm main pulmonary artery at two days of age. The branch pulmonary arteries were severely hypoplastic. Despite a patent shunt demonstrated angiographically, she had persistent cyanosis and was discharged receiving oral prostaglandin E! with an arterial oxygen tension between 30 and 40 mm Hg. Cyanosis progressed over the next 3 months as indicated by a rise in the hemoglobin concentration to greater than 20 gm/dl, An arteriogram showed good growth of the right pulmonary artery and moderate left pulmonary artery stenosis distal to the central shunt that was believed not to be related to the shunt. She underwent a left BlalockTaussig shunt to further improve left pulmonary artery blood flow and has done well, with a hemoglobin value of 18.6 gm/dl 2 years after reshunting. After these four patientsare eliminated from the 18 long-term survivors, 14 of 14 (100%) patients 1 year after central shunt placement had adequate function of their central

shunt. Three of the 14 patients have been followed up for less than 2 years. One patient (Patient 8) had tetralogy of Fallot repair at 18 months because of poor physical growth despite excellent pulmonary artery growth, excellent pulmonary artery blood flow through the central shunt, and normal-sized pulmonary arteries and right ventricle. Eliminating these four patients from the 14 who had adequate shunt function at 1 year leaves seven of 10 (70%) patients with adequate central shunt function 2 years after shunt placement. Three patients (Patients 6, 7, and 9) were believed to have outgrown their shunts and required repair or reshunting before 2 years of age. Thus, we believe that a central aortapulmonary artery shunt provides about 2 years of palliation. The six children with a central shunt who have not had repair or reshunting have been examined regularly

Central aorta-pulmonary artery shunts

1.7

•

1.6 .Q

1.5

cr

1.4

OJ


5

« l?

i5

e

'" 0

77 1

1.3 1.2 1.1

(/)

c'"

E >. a;

1.0 .9

«

t:

.8

~

.7

~

.t::

'"

....J

.6

.5 .4

3

{ Before Shunt

After Shunt

Fig. 6. Changes in the ratio of the left pulmonary artery diameter to the descending aorta diameter in 12 patients undergoing follow-up catheterization. Solid circles (e) represent patients with 4.0 mm grafts and open circles (0) represent patients with 3.7 mm grafts.

and are believed to have functioning shunts on the basis of shunt murmurs. Their hemoglobin concentrations range from 16.5 to 19.5 gmjdl (median 18.0 gm/dl) at follow-up examination 12 months to 3V2 years (median 15 months) after shunting. Discussion Although early reports regarding central PTFE shunts were encouraging." 27 subsequent reports have described a significant incidence of early and late shunt closure. Lamberti and associates" performed aortapulmonary artery shunts with long (2 to 6 em) segments of PTFE via a thoracotomy on 20 neonates and reported one early and one late shunt failure caused by thrombosis. Donahoo and co-workers" noted three of five early shunt failures requiring revision in infants with aortamain pulmonary artery shunts. In a more recent report, Lamberti and colleagues" described another case of early shunt thrombosis as well as late partial thrombosis in a second patient out of a total of five who had ascending aorta-main pulmonary artery shunts. They therefore suggested that the central shunt is not suitable for routine use. The modified Blalock-Taussig shunt has been reported to have a much lower, but not inconsequential, incidence of thrombosis. McKay and associates" reported two occluded modified Blalock-Taussig

The Journal of Thoracic and Cardiovascular Surgery

Barragry et al.

772

Table I. Angiographic results and eventual procedures on 12 patients recatheterized

Patient I 2

3 4

5 6 7 8 9 10 II 12

Diagnoses PA, SV, TGA PS, SV, TGA, AV canal, T APVC, tracheomalacia PA,IVS PA, IVS, Ebstein's anomaly PA, IVS, Ebstein's anomaly PA, ToF, AV canal PA, ToF PA, ToF Asplenia, PA, TGA, AV canal PA, ToF, IPA PA, ToF, partial CT PA, TGA, VSD

Shunt diameter (mm)

Pulmonary artery anastomosed

Interval between shunting and catheterization (mo)

Pulmonary artery distortion present

Yes

4 4

RPA MPA

47

2

No

4 3.7 4 3.7 4 4 3.7

LPA MPA MPA MPA MPA MPA MPA

3 3 4

22

No No No No No No No

4 4 4

MPA MPA MPA

33 33 46

10 13

18

No

Yes No

Legend: MPA, Main pulmonary artery. RPA, Right pulmonary artery. LPA, Left pulmonary artery. PA, Pulmonary atresia. SV, Single ventricle. TGA, Transposition of the great arteries. PS, Pulmonary stenosis. A V canal, Atrioventricular canal. TAPVC, Total anomalous pulmonary venous connection. lVS, Intact ventricular septum. ToF, Tetralogy of Fallot, IPA, Interrupted pulmonary artery. CT, Cor triatriatum. VSD, Ventricular septal defect.

shunts in 36 patients studied angiographically after shunt placement. Ilbawi and associates" described a 3.3% early shunt occlusion rate in newborn infants receiving modified Blalock-Taussig shunts. Moulton and colleagues" found four instances of shunt thrombosis (two early and two late) out of 35 neonates and infants receiving modified Blalock-Taussig shunts. The present series, on the other hand, differs from these reports in that no patient had shunt thrombosis or kinking. We believe that the very short length of PTFE material «1 em) employed in the construction of these shunts and not the use of antiplatelet agents was primarily responsible for the uniformly reliable patency, since antiplatelet agents were employed in only a minority of these patients. Another objection to the use of a direct graft between the aorta and the main pulmonary artery has been a high incidence of congestive heart failure. However, all arterial-pulmonary artery shunts have been associated with congestive heart failure, even the modified BlalockTaussig shunt. McKay and associates" reported that 21% of patients receiving modified Blalock-Taussig shunts had congestive heart failure necessitating treatment with digoxin and diuretics. Ilbawi and colleagues" found a 7% incidence of moderately severe congestive heart failure in patients after modified Blalock-Taussig shunts. One such patient required banding of the subclavian artery proximal to the shunt to control failure. However, Moulton and colleagues" reported that only 6% of their patients receiving modified Blalock-Taussig shunts had mild congestive failure.

Although the 40% incidence of congestive failure found in the present series is higher than reported after modified Blalock-Taussig shunts, the failure produced by the central shunts in our patients was mild and readily controlled by digoxin. The contribution of the aorta-pulmonary artery shunt to the growth of the pulmonary arterial tree is an important parameter of its function and can be assessed by measuring the ratio between the diameter of the branch pulmonary arteries and the descending aorta as seen on cineangiogram. The increase in the right pulmonary artery-descending aorta ratio was 118% whereas the increase in the left pulmonary arterydescending aorta ratio was 57% at a mean interval between shunt placement and catheterization of 19.5 months. These increases compare favorably with those reported in three previous series of shunts in cyanotic infants. Two of these series involved patients with modified Blalock-Taussig shunts and the third employed classical Blalock-Taussig shunts. In restudying six patients with PTFE shunts placed during the neonatal period, McKay and associates" reported a mean increase in the pulmonary artery-descending aorta ratio of 28% for the right pulmonary artery and 32% for the left pulmonary artery at a mean interval between shunt placement and catheterization of 15.6 months. Ilbawi and others" reported an increase in this ratio for the right pulmonary artery of 50% and for the pulmonary valve anulus of 52% at a mean interval of 14.7 months. Employing classical Blalock-Taussig shunts in cyanotic infants less than 1 year old, Guyton and co-workers"

Volume 93

Central aorta-pulmonary artery shunts

Number 5 May 1987

Change in pulmonary artery to descending aorta ratio MPA

I

RPA

I

LPA

0.71~0.81

0.57~0.59

0.71~.65

0.52~0.92

0.62~1.l6

0.88~1.06

0.31~1.25

0.42~0.88

0.38~1.l6

0.52~1.06

0.43~0.56

0.48~0.63

0.63~0.82

0.47~1.08

0.58~0.73

0.70~0.84

0.70~1.l2

0.80~1.00

0.80~0.95

0.52~.92

0.48~1.00

0.8

0.52~1.00

0.76~0.94

0.53~0.97

0.39-- 1.11 0.53--0.68

0.58~1.09

0.39--0.91

0.59~0.82

0.56~0.92

0.50~1.27

0.94~1.25

1.13~1.51

0.35~2.05

0.67~1.68

0.46~

noted a 36% increase in the ipsilateral pulmonary artery-descending aorta ratio and a 35% increase in the contralateral ratio at a mean interval between shunt placement and catheterization of 18.3 months. Criticism has been levied against the WaterstonCooley shunt because of its propensity to produce pulmonary vascular disease and distort the pulmonary arteries. No patient in the present series had pulmonary artery hypertension at the time of recatheterization. Two of the 12 patients recatheterized in the present series had pulmonary artery distortion. The distortion of the pulmonary artery in a central location, however, lends itselfto ready correction at the time of total repair, unlike the rather distal pulmonary artery distortion that occurs in a Blalock-Taussig shunt. The incidence of this distortion after placement of a Blalock-Taussig shunt is variable. Classic Blalock-Taussig shunts have been associated with an incidence of pulmonary artery deformity as high as 21%.34 This complication is apparently much less frequent with modified Blalock-Taussig shunts in that none of the 5631.32.35 patients recatheterized and reported on after modified Blalock-Taussig shunts have been found to have pulmonary artery distortion. In summary, the advantages of a short central aorta-pulmonary artery shunt are as follows: 1. They are technically feasible despite extremely complex anatomy and can be reliably placed in neonates with very hypoplastic pulmonary arteries. 2. These grafts remain reliably patent. 3. Any congestive heart failure produced is readily controlled medically. 4. They apparently have no tendency to produce elevated pulmonary vascular resistance.

Eventual procedure

773

Age at procedure (mo)

Left Blalock-Taussig shunt Tracheopexy, pulmonary valvectomy

47 2

Outflow tract patching Ebstein's repair Left Blalock-Taussig shunt Tetralogy repair Right Blalock-Taussig shunt Tetralogy repair Left Blalock-Taussig shunt

3 3 4 15 IS 18 22

Tetralogy repair Tetralogy repair Conduit placement

33 33 50

5. They allow good growth of the entire pulmonary tree. 6. At the time of total correction, these shunts are easily taken down and any pulmonary artery distortion present can be readily corrected because of its central location. For these reasons, we have preferred the central aorta-pulmonary artery shunt over other shunts to palliate severe pulmonary oligemia when present in cyanotic neonates with complex congenital heart disease. Wewish toexpress ourgratitude to Nina Van DeLinde and Patricia Longworth for assistance in preparation of this manuscript. REFERENCES 1. Blalock A, Taussig HB. Surgical treatment of malformations of the heart in which there is pulmonary stenosis or atresia. JAMA 1945;128:189-202. 2. Potts WJ, Smith S, Gibson S. Anastomosis of the aortato a pulmonary artery. JAMA 1946;132:627-31. 3. Waterston DJ. Treatment of Fallot's tetralogy in children under one year of age. Rozhl Chir 1962;41: 181-3. 4. Cooley DA, Hallman GL. Intrapericardial aortic-right pulmonary arterial anastomosis. Surg Gynecol Obstet 1966;122:1084-6.

5. Cole RB, Meester AJ, Fixler DE, Paul MH. Long-term results of aortopulmonary anastomosis for tetralogy of Fallot. Circulation 1971;43:263-71. 6. Kaplan S, Helmsworth JA, Ahearn EN, Benzing G III, Daored G, Schwartz DC. Results of palliative procedures for tetralogy of Fallot in infants and young children. Ann Thorac Surg 1968;5:489-97.

774

Barragry et al.

7. Yacoub M, Somerville J, Ross DN. Waterston's operation (aorta to right pulmonary artery anastomosis) for cyanotic heart disease. Br Heart J 1968;30:423. 8. Somerville J, Yacoub M, Ross DN, Ross K. Aorta to right pulmonary artery anastomosis (Waterston's operation) for cyanotic heart disease. Circulation 1969;39:593602. 9. Bernard WF, Jones JE, Friedberg DZ, Litwin SB. Ascending aorta-right pulmonary artery shunts in infants with certain types of cyanotic congenital heart disease. Circulation 1971;43:580-4. 10. Ebert PA, Gay WA, Oldham HN. Management of aorta-right pulmonary artery anastomosis during total correction of tetralogy of Fallot. Surgery 1972;71:231-4. II. Semb GS, Hall KY. Aorto-pulmonary shunt in congenital cyanotic heart disease in infants under two years of age. Scand J Thorac Surg 1972;6:43-51. 12. Gay WA, Ebert PA. Aorta-to-right pulmonary artery anastomosis causing obstruction of the right pulmonary artery. Ann Thorac Surg 1973;16:402-10. 13. Tay DJ, Engle MA, Ehleis KH, Levin AR. Early results and late developments of the Waterston anastomosis. Circulation 1974;50:220-9. 14. Fruccone NJ, Bowman FO, Maim JR, Gersony WM. Systemic-pulmonary arterial shunts in the first year of life. Circulation 1974;49:508-11. 15. Rectman MJ, Galioto FM, EI-Said G, Cooley DA, Hallman GL, McNamara DG. Ascending aorta to right pulmonary artery anastomosis. Circulation 1974;49: 952-7. 16. Idriss FS, Cavallo CA, Nikaidoh H, Paul MH, Koopot R, Muster AJ. Ascending aorta-right pulmonary artery shunt. J THORAC CARDIOVASC SURG 1976;71:49-57. 17. Stewart S, Harris P, Manning J. Current results with constriction and interruption of the Waterston anastomosis. Ann Thorac Surg 1978;25:431-7. 18. Norberg WJ, Tadavarthy M, Knight L, Nicoloff DM. Late hemodynamic and angiographic findings after ascending aorta-pulmonary artery anastomosis. J THORAC CARDlOVASC SURG 1978;76:345-52. 19. Yetter RL, Rashkind WJ, Waldhausen JA. Ascending aorta-right pulmonary artery anastomosis: long-term results in 137 patients with cyanotic congenital heart disease. J THORAC CARDIOVASC SURG 1978;76:115-25. 20. Wilson JM, Mach JW, Turley K, Ebert PA. Persistent stenosis and deformity of the right pulmonary artery after correction of the Waterston anastomosis. J THORAC CARDlOVASC SURG 1981;82:169-75. 21. Rao PS, Ellison RG. The cause of kinking of the right pulmonary artery in the Waterston anastomosis. J THORAC CARDlOVASC SURG 1978;76:126-9.

The Journal of Thoracic and Cardiovascular Surgery

22. Mearns AJ, Deverall DB, Kester RC: Revascularization of an arm for incipient gangrene after the Blalock-Taussig shunt. Br J Surg 1978;65:467-8. 23. Skovranek J, Goentzova A, Sahavek M. Changes in muscle blood flow and development of the arm following the Blalock-Taussig anastomosis. Cardiology 1975;6: 131-7. 24. Laks H, Fagan L, Barner HB, Willman YL. The Blalock-Taussig shunt in the neonate. Ann Thorac Surg 1978;25:220-4. 25. Edmunds H, Stephenson LW, Gadzik JP: The BlalockTaussig anastomosis in infants younger than one week of age. Circulation 1980;62:597-603. 26. Gazzaniga AB, Lamberti JJ, Siewers RD, et al. Arterial prosthesis of microporous expanded polytetrafluoroethylene for construction of aorta-pulmonary shunts. J THORAC CARDlOVASC SURG 1976;72:357-63. 27. Jennings RB, Innes BJ, Brickman RD. Use of microporous expanded polytetrafluoroethylene grafts for aortapulmonary shunts in infants with complex cyanotic heart disease. J THORAC CARDlOVASC SURG 1978;76:489-94. 28. Lamberti JJ, Campbell C, Replogle RL, et al. The prosthetic (Teflon) central aortopulmonary shunt for cyanotic infants less than three weeks old: results and long-term followup. Ann Thorac Surg 1979;28:568-77. 29. Donahoo JS, Gardner TJ, Zahka K, Kidd BSL. Systemicpulmonary shunts in neonates and infants using microporous expanded polytetrafluoroethylene: immediate and late results. Ann Thorac Surg 1980;30:146-50. 30. Lamberti JJ, Carlisle J, Waldman JD, et al. Systemicpulmonary shunts in infants and children. J THORAC CARDIOVASC SURG 1984;88:76-81. 31. McKay R, de Leval MR, Rees P, Taylor JFN, Macartney FJ, Stark J. Postoperative angiographic measurement of modified Blalock-Taussig shunts using expanded polytetrafluoroethylene (Gore-Tex). Ann Thorac Surg 1980;30:137-45. 32. Ilbawi MN, Grieco J, De Leon SY, et al. Modified Blalock-Taussig shunt in newborn infants. J THORAC CARDlOVASC SURG 1984;88:770-5. 33. Moulton AL, Brenner 11, Ringel R, et al. Classic versus modified Blalock-Taussig shunts in neonates and infants. Circulation 1985;72 (Pt 2):1135-44. 34. Guyton RA, Owens JE, Waumett JD, Dooley KJ, Hatcher CR, Williams WHo The Blalock-Taussig shunt. J THORAC CARDlOVASC SURG 1983;85:917-22. 35. Bove EL, Sondheimer HM, Kavey REW, Byram CJ, Blackman MS, Parker FB Jr. Subclavian-pulmonary artery shunts with polytetrafluoroethylene interposition grafts. Ann Thorac Surg 1984;37:88-91.