Archive of SID
other prosthesis, in which case patient size is a critical
factor. For still others, intracardiac correction may carry
a significantly higher mortality than that for preliminary
shunting and secondary repair. The modified Blalock
shunt (MBTS), used by the Great Ormond Street Group
(1,2), and subsequently popularized by several others
(3,4), interposes a prosthetic tube between the sub-
clavian and pulmonary arteries and its use requires less
dissection and guarantees a proximal and distal an-
*Corresponding Author: Farzad Kakaei
Department of General Surgery and Emergency Medicine, Imam Reza’ s Hospital, Shahid Madani Heart Center, Tabriz University of Medical Sci-
ences, Tehran, Iran, Daneshgah St, Tabriz, Iran
Tel: +98 9153118095, Fax: +98 411 3341317, E-mail: firstname.lastname@example.org
Outcome of Systemic-to-Pulmonary Shunts in Cyanotic Congenital
Heart Disease- A 9-year Experience
Khosro Hashemzadeh1, Shahryar Hashemzadeh2, and Farzad Kakaei3
1Department of cardiovascular Surgery, Shahid Madani Heart Center, Tabriz University of Medical Sciences, Tabriz, Iran
2Department of Thoracic Surgery, Shahid Madani Heart Center, Tabriz University of Medical Sciences, Tabriz, Iran
3Department of General Surgery and Emergency Medicine, Shahid Madani Heart Center, Tabriz University of
Medical Sciences, Tehran, Iran
Received: 9 Jan. 2007; Received in revised form: 8 Jun. 2007; Accepted: 8 Sep. 2007
Abstract- The aim of this study is to evaluate early and late results of surgery in children with congenital
heart disease and decreased pulmonary blood flow, who underwent a palliative systemic-to-pulmonary shunt.
During the past 9 years, 157 systemic-to-pulmonary artery shunts were performed in 130 patients (69 males,
61 females) with ages from 1 day to 31 years old. They had been evaluated for their clinical effectiveness, the
need for a repeat operation and the mortality and morbidity. There were 101 (77.7%) modified Blalock-
Tausig (BT) shunts, 19 (14.6%) modified Waterston shunts, 8 (6.2%) Central shunts, 1 Waterston shunt, and
1 Glenn shunt created throughout the study. Tetralogy of Fallot comprised the majority of cases (113; 86.9%)
while the remaining 17 (13.1%) included transposition of the great arteries with pulmonary stenosis, tricuspid
atresia, pulmonary atresia, atrioventricular septal defect (canal) with pulmonary stenosis, and univentricular
heart complex. Early mortality was 12.3% (16 patients). Second shunts were created in 25 (19.2%) patients.
Forty patients (30.8%) have undergone subsequent intracardiac repair 37.7 ± 17.8 months after original shunt
procedure. There were four (3%) late deaths. Follow-up was achieved in 105 of 114 early survivors for a pe-
riod of 3 to 117 months (mean 31.7 ± 19.4 months). Modified BT shunt was performed most frequently in our
service; it was associated with less closure and mortality than other types of shunt and had excellent clinical
palliation and patency rates.
© 2009 Tehran University of Medical Sciences. All rights reserved.
Acta Medica Iranica 2009; 47(2): 109-114.
Key word: Congenital heart disease, cardiovascular surgery, shunts
Although primary definitive intracardiac repair is now
possible and often preferable in many patients with con-
genital anomalies, there is still a need for palliative sys-
temic-to-pulmonary shunts in some cyanotic patients.
For some, definitive repair may not be presently avail-
For others, repair may involve placement of conduit or
astomosis of large diameter, with the maximal amount
of flow being regulated by the diameter of the patient’ s
own uninterrupted subclavian artery, thus allowing for
possible increase in shunt flow with growth. Somerville
and associations (5) reported a series of patients with
tetralogy malformations and Waterston shunts, who, at
the time of repair, had a closure of the shunt from within
the aorta. In this series of patients, there was a high mor-
tality rate as a consequence of the residual stenosis at the
level of the shunt in combination with the underdevel-
opment of the left pulmonary artery.
This reviews the early and late results in a series of 130
consecutive patients who underwent systemic-to-
pulmonary artery anastomosis during a 9 year period.
Our center is the leading cardiothoracic surgery hospital
in north-western states of Iran and over 19 percent of
those who underwent surgery in this center are congeni-
Archive of SID
Outcome of systemic to pulmonary
110 Acta Medica Iranica, Vol. 47, No. 2 (2009)
Patients and Methods
During the 9 year period from March 1998, through
March 2005, 157 systemic-to-pulmonary artery shunts
were performed in 130 patients [69 (53.1%) male, 61
(46.9%) female]. Ages of the patients ranged from one
day to 31 years. Age distribution of patients underwent
130 procedures was shown in Figure 1.
Twenty (15.4%) patients were infants (<1 year,
mean weight of 6.5 ± 1.3), two neonates (<1 month old),
seven more than 10 years and the remaining 101
(77.7%) between 1 and 10 years old. Preoperative clini-
cal data are depicted in Table 1.
The cardiac anatomy was tetralogy of Fallot in 113
(86.9%), the remaining 17 (13.1%) included transposi-
tion of the great arteries with pulmonary stenosis (6),
tricuspid atresia (4), pulmonary atresia (2), atrioven-
tricular septal defect (canal) with pulmonary stenosis
(3), and univentricular heart complex (2). All patients
had cardiac catheterization and angiocardiography prior
to operation. All in-patient and out-patient records, car-
diac catheterization data, and finding at subsequent in-
tracardiac repair were reviewed. The patients were
grouped according to age, type of shunt, anatomic diag-
nosis. Clinical and fallow-up data were then tabulated.
The patients who had two or three shunt procedures are
classified according to the initial procedure. There were
101 (right = 74, left = 27) modified Blalock-Tausig
shunts (MBTSs), and 19 (right = 18, left = 1) modified
Waterston shunts (MWSs), and 8 central shunts (CSs),
and 1 Waterston shunt (WS), and 1 Glenn shunt (GS)
created (Figure 2). In two patients, MBTSs was per-
formed by cardiopulmonary bypass because of hypoten-
sion. Of 130 patients, 128 (98.5%) received prosthetic
graft (polytetrafluoroethylene (PTFE) or Gore-Tex)
shunts (Table 2).
<1 mo 1-mo to 1-
1 to 10 y10 y<
nomber of patients
Figure 1. Age distribution of patients undergoing 130 shunt
MBTS MWSCS WSGS
type of sunts
Nomber of patients
Figure 2. Distribution of patients according to the type of
shunts applied to them. MBTS indicates Modified Blalock-
Taussig shunt; MWS, Modified Waterston shunt; CS, Central
shunt: WS, Waterston shunt; GS, Glenn shunt.
Table 1. Preoperative clinical data
Dyspnea on exertion
Failure to thrive
shunt, a 4 mm graft in 25 ( 19.4%), a 5 mm graft was
used in 74 (57.8%), and a 6 mm graft in 29 (22.7%) pa-
tients. 128 (98.5%) patients had received perioperative
heparin and 116 (89.2%) aspirin, postoperatively.
those receiving systemic-to-pulmonary
Archive of SID
Causes of early death*
Congestive heart failure 1
Cardiac arrest 2
Respiratory failure 2
Mediastinitis (septic shock) 0
*MOF indicates Multiorgan Failure; MBTS, Modified Blalock-Taussig Shunt; MWS, Modified Waterston Shunt; WS, Waterston
Shunt; CS, Central Shunt.
Kh. Hashemzadeh, et al.
Acta Medica Iranica, Vol. 47, No. 2 (2009) 111
Table 2. Distribution of shunts according to diameter of shunts
Diameter of graft ( Gore-Tex)
Type of shunt* Number(percent)
* MBTS indicates Modified Blalock Tausig Shunt; MWS, Modified Waterston Shunt; CS, Central Shunt
There was no intraoperative death. In this series of 130
patients, there were sixteen deaths within 30 days of
surgery or during the same hospitalization for an overall
early mortality of 12.3%. There were 6 deaths among
101 patients receiving MBTS, for a mortality of 5.9%.
An attempt was made to group these patients by the type
of shunt procedure performed and the cause of early
death (Table 3).
There were eight thrombosed shunts postoperatively
that underwent subsequent shunt procedure (6) or
thrombectomy (2). Four of these patients died after op-
eration. Three patients had poor function shunt post-
operatively that underwent subsequent shunt procedure
and one of them died after operation. The 22 patients
who were less than 1 year of age at the time of operation
composed only 16.9 percent of the total number of pa-
tients but contributed 50 percent of the deaths (8 of the
16 deaths). The 101 patients who were more than 1 year
and less than 10 years old composed 77.7 percent of the
total series but accounted for 50 percent of the total mor-
tality rate (8 of the 16 deaths). Of 16 early mortalities,
11 (68.8%) patients had tetralogy of Fallot, 3 (18.8%)
tricuspid atresia, 1 (6.3%) pulmonary atresia, and 1
(6.3%) transposition of great artery with pulmonary ste-
Early postoperative complications
Two patients had intraoperative cardiac arrest that
responded to cardiopulmonary resuscitation. Postopera-
tiove complications included thrombosed shunt (8,
6.2%) rebleeding (7, 5.3%), acute heart failure (7,
5.3%), respiratory failure (5, 3.8%), and acute renal fail-
ure (3, 2.3%). (Table 4) One patient (MBTS) had ipsi-
lateral pulmonary edema and dense opacification of the
ipsilateral lung postoperatively that immediately re-
turned to the operating room and narrowed the proximal
anastomosis by stitch.
Early and late shunt failures
There were eleven patients (8.5%) whom we consid-
ered early shunt failures, seven in the MBTS, 3 in the
MWS, and 1 in the CS. Of eleven patients, 8 had throm-
bosed shunts and 3 poor function shunts. Eight throm-
bosed shunt underwent subsequent shunt procedures6
and thrombectomy of the shunt (2). One of the throm-
bectomy shunt was poor function that second shunt was
created. Four of these patients discharged from hospital.
Early shunt failure was seen in 8 tetralogy of Fallot,
1 atrioventricular defect (canal) with PS, and 2 transpo-
sition of great artery with pulmonary stenosis. There
were 14 (10.8%) late shunt failures. All late shunt fail-
ures were seen in patients with tetralogy of Fallot. Ele-
ven patients had MBTS, two MWS, and one central
Table 3. Causes of early mortality
Type of shunt
Archive of SID
TGA + P.S 2 2
P.A 1 0
U.H.C 1 1
Number 63 17 Total
Percent 70 18.9
*TOF indicates Tetralogy of Fallot; TA, Tricuspid Atresia; AVSD, Atrioventricular Septal Defect; PS, Pulmonary Ste-
nosis; TGA, Transposition of Great Arteries; PA, Pulmonary Atresia; UHC, Univentricular Heart Complex; MBTS,
Modified Blalock Taussig Shunt; MWS, Modified Waterston Shunt; CS, Central Shunt; WS, Waterston Shunt; GS,
Outcome of systemic to pulmonary
112 Acta Medica Iranica, Vol. 47, No. 2 (2009)
Table 4. Early postoperative complications
Acute heart failure
Acute renal failure
Ipsilateral pulmonary edema
Subsequent shunt procedures
Second shunts were created in 25 (19.2%) patients.
In the eleven early shunt failures a second shunt was
created 1 to 7 (3 ± 2.86) days after original procedures
because of thrombosed shunts (6) and poor functions
shunts (5). In remaining fourteen late shunt failures a
second shunt was created 3 to 59 (28.6 ± 14.8) months
after original shunt because of poor function (10) and
none function (4) shunts. The second shunt procedures
were MBTSs (16), MWS (5), CS (3) and WS (1). There
were 5 (20%), 16 (21%), and 4 (13%) shunt failures, in
4, 5, and 6 mm PTFE grafts consequently. Postoperative
complications in these patients were bleeding (3), car-
diac arrest, and respiratory failure (1). There was no
death. Two patients (TOF) received three separate
shunts 2 and 66 months after original shunt and one of
them died because of cardiac arrest.
Subsequent intracardiac repair
Forty patients (30.8%) have undergone subsequent
intracardiac repair 37.7 ± 17.8 months (range 10 to 85
months) after original shunt procedure. These patients
had tetralogy of Fallot (39) and transposition of great
artery with PS (1). These patients had received MBTS
(38, 95%), and MWS (2, 5 %). There was no operative
and late mortality in these patients. In all this patients,
cardiac catheterization was performed before total cor-
rection and there was lack of distortion of pulmonary
Patients without further procedures
There are 90(69.2%) patients who remain well and
free of further surgery after receiving their palliative
shunts. The diagnoses and type of shunt of patients are
shown in table 5.
Of 90 patients, there were four (3%) late deaths, 3 to
63 months after the original procedure, because of mul-
tiorgan failure (2), and cardiac arrest (2). Nine patients
lost to follow up. Follow-up was achieved in 105 of 114
early survivors for a period of 3 to 117 months (mean;
31.7 ± 19.4 months). All surviving patients had subjec-
tive and objective improvement.
Table 5. Diagnosis and type of shunt of patients without procedures
Type of shunt
A.V.S.D + P.S
Archive of SID
previously reported poor success with the Blalock–
Taussig anastomosis in this age group (8,9,11,14-16).
The results of the present study indicate that the overall
survival rate (94.1%) with modified Blalock shunt is
better than that with the other shunt procedures, even in
very young infants. The early mortality in the MBTS
was 6 (5.9%) that this conclusion closely parallel those
in the multicenter study by Woolf et al. (21-23). In this
series, the modified Waterston shunt (MWS) had high
mortality (6/19, 31.5%) versus the other study (10.7 %)
Kh. Hashemzadeh, et al.
Acta Medica Iranica, Vol. 47, No. 2 (2009) 113
For patients with lesions with complex anomalies or in
critically ill cyanotic newborns, definitive intracardiac
repair may not be possible. For others a two-stage cor-
rection may still be preferable (6,7). It is widely ac-
cepted that the Blalock-Taussig shunt is the shunt of
choice for children with pulmonary oligemia who older
than 1 year (8-13). The incidence of congestive heart
failure and pulmonary microvascular disease that fol-
lows the Blalock-Taussig shunt is low when compared
with that of lateral shunts (8-11,13-16). Chopra and as-
sociates17 proposed that the Blalock-Taussig shunt was
the shunt of choice for infants of any age with severe
tetralogy of Fallot. However, the procedure requires
extensive surgical dissection and sacrifice of the sub-
clavian artery, and the maximal anastomotic diameter is
limited by the size of the subclavian artery. In addition,
the anastomosis tiny neonate does not keep pace with
the rest of the child's growth.
Use of the modified shunt obviates some of these
problems, but carries risks of it is own. It is technically
simple operation, requiring less dissection and guaran-
teeing a wide proximal and distal anastomosis. Also,
distal subclavian flow can be preserved during construc-
tion of this shunt, and it provides the potential for in-
creased shunt flow with growth as the subclavian diame-
ter increases. Furthermore, it allows easy creation of a
shunt on the side of the aortic arch when it is necessary
or preferable. Despite the fixed length of the prosthetic
tube, we have not observed significant pulmonary arte-
rial distortion with growth. Further follow-up is of
course necessary, but the longer experience of other
groups (1-4,18-20) shows a similar absence of distor-
tion. At the time of subsequent surgery, the previous
shunt had been simply ligated or clipped.
In most centers Blalock-Taussig shunt is the pallia-
tive procedure of choice in patients over 1 year of age
(8,9,11,14). In infants under 1 year of age. However, the
Waterston or Potts anastomosis has been recommended
because of the small size of the subclavian artery and the
(24). In the MWS, congestive heart failure was causes of
early death in 4 (21%) patients which in the MBTS it
was 1 (≈ 1%). On the other hand, early shunt patency for
the MBTS and MWS groups were 93.1% (95/101) and
84.3% (3/19) consequently. Therefore, our results indi-
cate that MBTS had lower early mortality and higher
early shunt patency rate than MWS. Systemic heparini-
zation (98.5%) was routinely used postoperatively.
However, there was 8 thrombosed shunts after original
shunts and heparinization resulted in a reoperation rate
of 5.4%. There were eleven (8.5%) early and fourteen
(10.8%) late shunt failures that second shunt was created
3.0 ± 2.86 days and 28.6 ± 14.8 months after original
shunt, consequently. We observed no significant differ-
ence in the shunt patency rate between 4mm (80%) and
5 mm (79%) grafts for a period 3 to 117 months. How-
ever, the work of Woolf et al. (21) demonstrated a rapid
decrease in patency as a result of pseudointimal prolif-
eration after 41 months in a group of patients receiving
PTFE (Polytetrafluoroethylene) shunts, and the work of
de Leval et al(1), Di Benedetto et al (25), and Miyamuto
et al (3) clearly suggests that this problem is less fre-
quent with larger size grafts.
In summary, our experience suggests that systemic to
pulmonary shunt (MBTS, MWS, WS, CS, etc.) can be
safely and reproducibly constructed. This series shows
an early mortality rate comparable to that of other stud-
ies, with excellent clinical palliation and patency rate.
On the other hand, in our series, there was lack of distor-
tion of pulmonary arterial anatomy in forty patients with
total correction. Therefore, PTFE shunts are preferred
for the management of severe pulmonary oligemia in
infants and children.
1. de Leval MR, McKay R, Jones M, Stark J, Macartney FJ:
Modified Blalock– Taussig shunt: Use of subclavian artery
orifice as flow regulator in prosthetic systemic pulmonary
artery shunts. J Thorac Cardiovasc Surg 1981; 81:112.
2. McKay R, de Leval MR, Rees P, Taylor JFN, Macartney
FJ, Stark J: Postoperative angiographic assessment of modi-
fied Blalock– Taussig shunts using expanded polytetra-
fluoroethylene (Gore-Tex). Ann Thorac Surg 1980; 30:137.
3. Miyamuto K, Zavanella C, Lewin AN, Subramanian S: Aor-
topulmonary artery shunts with expanded polytetrafluoro-
ethylene (PTFE) tube. Ann Thorac Surg 1979; 27:413.
4. Bove EL, Sondheimer HM, Kavey REW, Byram CJ,
Blackman MS, Paker FB Jr: Subclavian– pulmonary artery
shunts with polytetrafluoroethylene interposition grafts.
Ann Thorac Surg 1984; 37:88.
Archive of SID Download full-text
Outcome of systemic to pulmonary
114 Acta Medica Iranica, Vol. 47, No. 2 (2009)
5. Somerville J, Barbosa R, Ross D, Oslen E: Problems with
radical corrective surgery after ascending aorta to right
pulmonary artery shunt (Waterston’s anastomotic) for cya-
notic congenital heart disease. Br Heart J 1975; 37:1105.
6. Kirklin JW, Blackstone EH, Pacifico AD, et al: Routine
primary repair versus two-stage repair of tetralogy of Fallot.
Circulation 1979; 60:373.
7. Kirklin JW, Blackstone EH, Kirklin JR, Pacifico AD, Ara-
medi J, Bargeron LM Jr: Surgical results and protocols in
the spectrum of tetralogy of Fallot. Ann Surg 1983;
8. Cole RB, Muster AJ, Fixler DE, Paul MH: Long term re-
sults of aorto-pulmonary anastomosis for tetralogy of Fal-
lot. Circulation 1971; 43:263.
9. Harris AM, Segel N, Bishop JM: Blalock– Taussig anasto-
mosis for tetralogy of Fallot. Br Heart J 1964; 26: 266.
10. Neches WH, Naifeh JG, Park SC, Lenox CC, Zuberbuhler
JR, Siewers RO, Pontius RG, Bahnson HT: Systemic- pul-
monary artery anastomosis in infancy. J Thorac Cardiovasc
Surg 1975; 70:921.
11. Bernhard WF, Jones LE, Friedbarge DZ, Litwin SB: As-
cending aorta– right pulmonary artery shunt in infants and
older patients with certain types of cyanotic congenital
heart disease. Circulation 1971; 43:580.
12. Laks H, Marco JD, Willman VL: The Blalock-Taussig
shunt in the first six months of life. J Thorac Cardiovasc
Surg 1975; 70:687.
13. Wood WC, McCue CM, Lower RR: Blalock– Taussig
shunts in infants. Ann Thorac Surg 1973; 16:454.
14. Reitman MJ, Galioto FM Jr, El-Said GM, Cooley DA,
Hallman Gl, McNamara DG: Ascending aorta to right pul-
monary artery anastomosis. Circulation 1974; 49:952.
15. Somerville J, Yacomb M, Ross DN, Ross K: Aorta to right
pulmonary artery anastomosis (Waterston’ s operation) for
cyanotic heart disease. Circulation 1969; 39:593.
16. Truccone NJ, Bowman FO Jr, Malm JR, Gersony WM:
Systemic- pulmonary arterial shunts in the first year of life.
Circulation 1974; 49:508.
17. Chopra PS, Levy JM, Dacumas GC Gr, Berkoff HA, Lor-
ing LL, Kahn DR: The Blalock– Taussig operation: The
procedure of choice in the hypoxic infant with tetralogy of
Fallot. Ann Thorac Surg 1976; 22:235.
18. Donahoo JS, Gardner TJ, Zahka K, Kidd BSL: Systemic-
pulmonary shunts in neonates and infants using micropor-
ous expanded polytetrafluoroethylene: immediate and late
results. Ann Thorac Surg 1980; 30:146.
19. Kay PH, Capuani A, Franks R, Lincoln C: Experience with
the modified Blalock– Taussig operation using polytetra-
fluoroethylene (Impra) graft. Br Heart 1983; J 49: 359.
20. Lamberti JJ, Carlisle J, Waldman JD, Lodge FA, Kirkpa-
trick SE, George L, Matthewson JW, Turner SW, Pappel-
baum ST: Systemic– pulmonary shunts in infants and chil-
dren: Early and late results. J Thorac Cardiovasc Surg
21. Woolf PK, Stephenson LW, Meijboom E, Bavinck JH,
Gardner TJ, Donahoo JS, Edie RN, Edmunds LH Jr: A
comparison of Blalock– Taussig, Waterstone and polytetra-
fluoroethylene shunts in children less than two weeks of
age. Ann Thorac Surg 1984; 38:26.
22. Wright J, Albercht H, Beveridge J: Palliation in cyanotic
congenital Heart Disease. Fifteen years, experience of vari-
ous shunt procedures. Med J Aust 1986; 144 (4):178-182.
23. Maghur HA, Ben-Musa AA, Salim ME, Abuzakhar SS:
The modified Blalock-Taussig shunt: a 6-year experience
from a developing country.Pediatric Cardiology 2002;
24. Fenchel G, Steil E, Seboldt H, Quintenz R, Alpitz J, Hoff-
meister HE: Early and late results of the modified Wa-
terston shunt with PTFE grafts for palliation of compolex
congenital cyanotic heart disease in neonates. Thorac Car-
diovasc Surg. 1991; 39(5): 268-272.
25. DiBenedetto G, Tiraboschi R, Vanini V, Annecduno P,
Aiazzi L, Capuoli C, Parenzan L: Systemic-pulmonary ar-
tery shunt PTFE prosthesis (Gore-Tex): early results and
long-term follow-up on 105 consecutive cases. Thorac Car-
diovasc Surgeon 1981; 29:143.