European Journal of Cardio-thoracic Surgery 17 (2000) 631±636

www.elsevier.com/locate/ejcts

Tetralogy of Fallot: what operation, at which age q

Marco Pozzi*, Dipesh B. Trivedi, Denise Kitchiner, Robert A. Arnold

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

Cardiac Unit, Royal Liverpool Children's Hospital, AlderHey, Eaton Road, Liverpool L12 2AP, Liverpool, UK
Received 6 September 1999; received in revised form 28 January 2000; accepted 21 February 2000

Abstract
Background: The optimal management of tetralogy of Fallot is still under debate, particularly with respect to surgical approach and the
age of operation. In recent times a transatrial-transpulmonary approach and primary repair in younger patients is favoured. The purpose of the
present study was to analyze the result of our current surgical management by assessing the perioperative and intermediate term follow up in
order to de®ne the optimal strategy and timing of operation for our institution. Methods: One hundred and thirty two patients with tetralogy
of Fallot who underwent de®nitive repair between May 1993 and December 1998 were analyzed by reviewing their medical records and
follow-up. Median age was 15.5 (2.3±68.6) months and median weight was 8.8 (5±16) kg. Ten (7.57%) patients were under 6 months, 38
(28.78%) were between 6 and 12 months, 36 (27.27%) were between 12 and 18 months, 23 (17.42%) were between 18 and 24 months and 25
(18.93%) were more than 24 months age. During the study period there was a move to earlier surgery and differing methods of repair
depending on the anatomy observed. Follow up was conducted by the referring cardiologist. Median follow up was 35.48 (8.07±74.93)
months. Results: Forty-two (31.8%) patients required a palliative procedure before total correction due to unfavourable anatomy. Subpul-
monary infundibular obstruction with a ®brous component increased signi®cantly with age (P , 0:05). Operations were entirely transatrial in
14 (10.6%), transatrial and transpulmonary in 69 (52.2%), transatrial and transventricularly in 42 (31.8%) and a homograft conduit was used
in seven (5.3%) patients. Younger patients had narrower pulmonary valves and required a transannular patch more frequently. All patients
were in sinus rhythm, 28 (21.1%) showing right bundle branch block. Median hospital stay was 8 (5±54) days. No patient required
reintervention during follow up and there was no early or late mortality. Conclusion: Correction of tetralogy of Fallot at younger age
does not increase morbidity or mortality and has potential advantages. A surgical technique adapted to the anatomy of the right ventricular
out¯ow tract, achieves the best results. q 2000 Elsevier Science B.V. All rights reserved.
Keywords: Tetralogy of Fallot; Young age; Surgical approach; Anatomical

1. Introduction repair in younger patients including neonates [9±12]. This

move to earlier surgery was justi®ed as the operative risk
The medical and surgical management of tetralogy of had fallen signi®cantly. Studies of correction in infants have
Fallot (TOF) has continued to evolve and many thousands also con®rmed the low morbidity and mortality in this age
of patients have excellent long-term survival following group [9±12]. Many centres, however, continue to favour a
correction. However, the optimal management is still staged approach for patients requiring intervention in the
under debate, particularly with respect to the surgical neonatal period or infancy [13,14].
approach and the age of operation. Documentation of right We have developed a policy of detailed investigation
ventricular dysfunction [1,2], pulmonary insuf®ciency and including cardiac catheterization at the time of presentation
ventricular arrhythmias [3] after transventricular repair, has so that the management is planned before the development
led many centres to adopt to transatrial/transpulmonary of signi®cant cyanosis or cyanotic spells. Patients with
approach and excellent results have been demonstrated favourable morphology proceed to early primary correction
[4±8]. irrespective of the age but those with poorly developed
There has also been a trend from two stage to primary pulmonary arteries or coronary anomalies undergo a staged
approach. The purpose of the present study was to analyze
the results of our current surgical management by assessing
q
Presented at the 13 th Annual meeting of the European Association for the peri-operative and intermediate period of follow up, in
Cardio-thoracic Surgery, Glasgow, Scotland, UK, September 5±8, 1999. order to de®ne the optimal strategy and timing of operation.
* Corresponding author. Tel.: 144-151-252-5635; fax: 144-151-252-
5643.
E-mail address: mpozzi@liverpedcard.u-net.com (M. Pozzi).

1010-7940/00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
PII: S 1010-794 0(00)00415-2
632 M. Pozzi et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 631±636

Table 1
Nature and level of right ventricular out¯ow tract obstruction a

RVOTO ,6 months 6±12 months 12±18 months 18±24 months .24 months Total

Infundibular 10 (100) 38 (100) 36 (100) 23 (100) 25 (100) 132 (100)

PV annulus 9 (90) 31 (81.5) 25 (69.4) 15 (65.2) 14 (56) 94 (71.2)
Pulmonary valve 9 (90) 31 (81.5) 25 (69.4) 21 (91.3) 21 (84) 107 (81.1)
MPA 7 (70) 26 (68.4) 30 (83.3) 21 (91.3) 20 (80) 104 (78.7)
LPA 1 (10) 17 (44.7) 13 (36.1) 8 (34.7) 10 (40) 49 (37.1)
RPA 1 (10) 4 (10.5) 6 (16.6) 3 (13.04) 1 (4) 15 (11.3)
Peripheral PA 2 (20) 3 (7.8) 5 (13.8) 2 (8.6) 3 (12) 15 (11.3)
a
Values in parentheses are percentage of patients in that group. RVOTO, right ventricular out¯ow tract obstruction; PV annulus, pulmonary valve annulus;

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

MPA, main pulmonary artery; LPA, left pulmonary artery; RPA, right pulmonary artery; peripheral PA, peripheral pulmonary artery.

2. Patients and methods patient with absent pulmonary valve syndrome were
included. Patients with pulmonary atresia and a ventricular
From May 1993 to December 1998, 132 cases of TOF septal defect were excluded. The nature and level of right
who underwent de®nitive repair by one surgeon (M.P.) were ventricular out¯ow tract obstruction are shown in Table 1.
retrospectively assessed. The medical records were Subpulmonary infundibular stenosis was present in all the
reviewed to obtain information on the pre and postoperative patients. The ®brous component of the infundibular obstruc-
condition. tion increased signi®cantly with age (P , 0:04) (Fig. 1).
The inter-atrial septum was intact in 80 (60.6%), patent The median age at the time of repair was 15.3 (2.3±68.6)
foramen ovale was present in 37 (28.03%) and atrial septal months. The median weight was 8.81 (5±16) kg and median
defect in 15 (11.3%) of patients. Seventy-eight (59.09%) body surface area was 0.42 (0.26±0.98) m 2. Ten patients
patients had less than 50% aortic override and 54 (40.9%) (7.5%) were less than 6 months, 38 (28.78%) were between
patients had more than 50% aortic override. Four patients 6 and 12 months, 36 (27.27%) were between 12 and 18
with TOF and an atrioventricular septal defect, and one months, 23 (17.42%) were between 18 and 24 months and

Fig. 1. Incidence of ®brous obstruction in RVOT. Obstruction by age of patient.

 M. Pozzi et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 631±636 633

25 (18.93%) were more than 24 months of age. The age bands was undertaken through the pulmonary valve before
distribution by year of operation is shown in Fig. 2. assessing the adequacy of the right ventricular out¯ow tract
Forty-two patients (31.8%) required a palliative proce- with Hegar's dilators introduced through the tricuspid valve
dure before total correction. Twenty-two patients and the pulmonary valve. Residual narrowing could either
(16.66%) had right or left modi®ed Blalock±Taussig be due to a restrictive pulmonary valve or hypoplasia of the
shunt, 20 patients (15.1%) had balloon dilatation of right right ventricular out¯ow tract. If the pulmonary valve was
ventricular out¯ow tract and three patients (2.2%) had both stenotic but the infundibular obstruction had been success-
shunt and balloon dilatation. Unfavourable anatomy which fully relieved, we incised the annulus without ventriculot-
consisted of either major coronary arteries crossing right omy or by limiting the ventriculotomy to 3±5 mm, just
ventricular out¯ow tract, diminutive pulmonary arteries or enough to place a patch to effectively enlarge the annulus.
prematurity with worsening cyanosis were our indications In cases where the out¯ow tract was still obstructed (due to

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

for palliative intervention before total repair. The median hypoplasia rather than hypertrophy of the out¯ow tract), we
age at the time of balloon dilatation of right ventricular felt that a right ventriculotomy was necessary and should be
out¯ow tract or pulmonary valve was 6.83 months and extended at least a few millimetres beyond the length of the
complete repair was performed after a median period of infundibular septum. The left and right pulmonary arteries
12.13 months. The median age at insertion of systemic to were also measured and if necessary the incision in the main
pulmonary artery shunt (right or left modi®ed Blalock± pulmonary artery was extended into either pulmonary
Taussig shunt) was 2.93 months and complete repair was artery. The pulmonary artery and right ventricular out¯ow
performed after a median period of 20.23 months. tract were reconstructed using untreated autologous pericar-
The operative technique was uniform during the study dium. In the presence of severely hypoplastic peripheral
period. Access was obtained through a right atriotomy and pulmonary arteries, a monocusp homograft patch was
the tricuspid valve. The parietal band was excised and used. Fifteen patients (11.3%) had right ventricular out¯ow
dissection was carried upwards as far as possible. Obstruct- tract reconstruction using monocusp homograft patch.
ing muscle bands were divided or excised and ®brous tissue The surgical approach was transatrial in 14 (10.6%)
was excised. The main pulmonary artery was incised long- patients, transatrial- transpulmonary in 69 (52.2%) patients
itudinally and valvotomy was performed through the exist- and transatrial-transventricular in 42 (31.8%) patients. In
ing commissures. Further division and excision of muscle seven (5.3%) patients, one with absent pulmonary valve

Fig. 2. Age distribution of patients by year of operation.

634 M. Pozzi et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 631±636

Table 2
Surgical approach a

Approach ,6 months 6±12 months 12±18 months 18±24 months .24 months Total

TA 0 (0.0) 2 (5.26) 3 (8.33) 5 (21.7) 4 (16) 14 (10.6)

TATP 5 (50) 25 (65.7) 21 (58.3) 8 (34.7) 10 (25) 69 (52.2)
TATV 5 (50) 10 (26.3) 12 (33.3) 9 (39.1) 6 (24) 42 (31.8)
Homograft 0 (0) 1 (2.6) 0 (0) 1 (4.3) 5 (20) 7 (5.3)
a
Values in parentheses are the percentage of patients in that group. TA, transatrial; TATP, transatrial transpulmonary; TATV, transatrial transventricular.

syndrome and six with abnormal coronary arteries crossing (P ˆ 0:08). This may be due to uneven number of patients

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

the right ventricular out¯ow, a homograft valve conduit was in different age groups. However higher incidence of trans-
inserted (Table 2). The ventricular septal defect was closed annular patch in younger patients did not require larger right
transatrially. Other additional procedures such as closure of ventriculotomy to relieve the obstruction. One hundred and
the arterial duct, repair of atrioventricular canal defect were twenty (90.9%) patients required main pulmonary artery
carried out as indicated. reconstruction. Fifty-eight (43.9%) and 16 (12.1%) patients
Follow-up by the referring cardiologist included clinical required left pulmonary artery and right pulmonary artery
evaluation, electrocardiogram and echocardiography. reconstruction, respectively, with no difference between
Median follow up was 35.48 (8.07±74.93) months and various age groups.
was complete to 100%. Data collection was by case-note During this study period there was no early or late mortal-
review. ity. Median hospital stay was 8 (5±54) days. Postoperative
course was uneventful in 102 (77.27%) of patients. Minor
complications in the form of pleural effusion, fever, respira-
3. Results tory infection were present in 20 (15.15%). Amongst the
major complications, eight (6.06%) patients required
Younger patients had narrower pulmonary annuli and prolonged ventilation for more than 5 days, ®ve (3.7%)
required transannular patch more frequently compared to patients had ventricular dysfunction which required
older patients (Fig. 3). This was not statistically signi®cant prolonged ionotropic support and later digoxin. Four

Fig. 3. Incidence of transannular patch by age of patients.

 M. Pozzi et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 631±636 635

(3.03%) patients had junctional ectopic tachycardia which arrhythmias and encourage the development of more normal
were treated by systemic cooling to 358C and overdrive pulmonary vasculature. Over the period of this study we
atrial pacing. One patient had disseminated intravascular have achieved a progressive reduction in the age at opera-
clotting and recovered from it completely. tion. At present the limiting factor to earlier operation is the
At the time of discharge all 132 (100%) patients were in timing of presentation or referral. A common attitude is not
sinus rhythm with 28 (21.2%) showing right bundle branch to intervene in asymptomatic, pink patients. We believe this
block, 56 (42.42%) were on digoxin and 90 (68.18%) were is wrong and that this is probably the group of patients that
on diuretics. There was no difference among different age could bene®t most from an early repair.
groups with respect to major or minor complications, iono- We have opted for primary repair whenever possible.
trope and ventilatory requirement, intensive care or hospital However, we still exercise a certain degree of selection.
stay. We prefer not to use conduits in neonates and infants and

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

At the last follow up 128 (96.96%) patients were in consequently we prefer a shunt in those patients with anom-
NYHA class I and four (3.03%) patients were in NYHA alous coronary arteries. We also opt for a shunt in neonates
class II. Eighteen (23.67%) patients were on digoxin and with multiple, severe, peripheral pulmonary artery stenosis
24 (31.68%) patient were on diuretics. Follow up echocar- or in the presence of hypoplastic pulmonary arteries. As a
diogram showed good ventricular function in 125 (94.6%) consequence of this policy 16.6% of our patients had a
patients. Tricuspid regurgitation was seen in 39 (29.5%) modi®ed Blalock±Taussig shunt at a median age of 2.93
patients with none having severe tricuspid regurgitation. months.
Fifty-one (38.6%) patients had variable degree of pulmon- This selective approach has allowed us to produce good
ary regurgitation with 28 (21.21%) having moderate early and medium term results in all age groups. The
pulmonary regurgitation and three (2.27%) patients having remaining question is whether or not these good results
severe pulmonary regurgitation. Small residual ventricular will also be maintained in the long term. We are also not
septal defect was noted on follow up echocardiogram in in the position to comment on whether or not the same
eight patients (6.05%). No patient required re-intervention results could be produced with a policy of neonatal or infant
for pulmonary regurgitation, residual right ventricular correction in patients without any form of selection.
out¯ow tract obstruction or residual ventricular septal defect The other important issue is the surgical technique. A
during the follow up period. transventricular approach has been used for many years
and is still in use with good results by some surgeons
[16]. However, avoiding a ventriculotomy is very appealing
4. Discussion and comments and many authors have proposed a transatrial or transatrial/
transpulmonary approach.
Tetralogy of Fallot is progressive with an unfavourable We have certainly aimed at avoiding a ventriculotomy
natural history. Progressive hypoxia, cyanotic spells, cere- but have come to the conclusion that this is not always
bral infarction or abscess and endocarditis are major causes possible.
of morbidity and mortality [15] and the risk is not entirely Certainly, all the ventricular septal defects can be closed
removed by palliation. The early and late results of correc- through a right atriotomy. This is our normal practice as we
tive surgery have steadily improved and most centres now believe that in this way we can minimize the size of ventri-
report low morbidity and mortality [5,7,11±14,16], as culotomy or avoid it altogether.
con®rmed in this study. Over the years the surgical approach With regard to the right ventricular out¯ow tract, we have
to this condition has changed. There has been a move from a recognized two distinct situations. In some patients the
staged approach in favour of a primary repair, a progressive obstruction is due to hypertrophied muscular bands while
lowering the age at repair and a surgical technique that in others it is due to an hypoplastic out¯ow tract. Between
avoids or reduces the ventriculotomy is preferred. these two extremes there is a whole spectrum of intermedi-
There is no doubt that primary repair, without prior ate patterns. In the ®rst case we believe that either a right
palliation, is preferable if this can be achieved safely. The atrial or right atrial/pulmonary artery approach with resec-
immediate consequence of this approach is that patients will tion of hypertrophied muscular bands and ®brous tissue can
need to undergo operation at an earlier age. Several authors successfully deal with the obstruction without any need for a
have demonstrated that primary repair can be performed in ventriculotomy. In the second case hypoplasia is the
infancy and in the neonatal period without increased mortal- problem and we ®nd very little muscular tissue to remove
ity or morbidity [5,16±18]. It is a common experience, or divide. In addition these patient have very prominent
however, that repair at younger age is associated to a higher anterior displacement of the infundibular septum. In our
incidence of transanular patches. This does not appear to opinion the only way to deal with this type of right ventri-
have negative effects at least in the short and medium term. cular out¯ow obstruction is to perform a ventriculotomy
On the other hand early repair has certainly several advan- which should be at least a few millimetres longer than the
tages. It can prevent or reduce the development of severe anteriorly displaced infundibular septum.
right ventricular hypertrophy and ®brosis, reduce the risk of Although it is dif®cult to prove, we believe that an exces-
636 M. Pozzi et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 631±636

sively aggressive muscle resection is more damaging than a favourable outcome of nonneonatal transatrial, transpulmonary
ventriculotomy. Hence the best way to deal with this repair. Ann Thorac Surg 1992;54:903±907.
[6] Kawashima Y, Kitamura S, Nakano S, Yagihara T. Corrective surgery
obstruction is to adjust the technique to the individual anat- for tetralogy of Fallot without or with minimal right ventriculotomy
omy. To ®nd the balance between muscle resection and and with repair of the pulmonary valve. Circulation 1981;64(Suppl
ventriculotomy is of paramount importance. 2):147±153.
In our series of patients we have described a 31.8% inci- [7] Touati GD, Vouhe PR, Amodeo A, Pouard P, Mauriat P, Leca F,
dence of ventriculotomy, but it is worth emphasizing that Neveux JY. Primary repair of tetralogy of Fallot in infancy. J Thorac
Cardiovasc Surg 1990;99:396±403.
ventriculotomy and transannular patch are often used syno- [8] Paci®co AD, Sand ME, Bargeren LM, Colvin EC. Transatrial ± trans-
nymously. Without trying to be semantically too pedantic, pulmonary repair of tetralogy of Fallot. J Thorac Cardiovasc Surg
we would stress that this is not always the case. There are a 1987;93:919±924.
number of patients in our series in whom the pulmonary [9] Walsh EP, Rockenmnher S, Keane JF, Hougen TJ, Lock JE, Casta-

Downloaded from https://academic.oup.com/ejcts/article-abstract/17/6/631/385490 by guest on 15 September 2018

valve ring was stenotic. In these patients the obstruction neda AR. Late results in patients with tetralogy of Fallot repaired
during infancy. Circulation 1988;77:1062±1067.
could be relieved without ventriculotomy, by dividing [10] Martin R, Khaghani A, Radley-Smith R, Yacoub M. Patient status 10
pulmonary valve ring and enlarging it with a patch without or more years after primary total correction of tetralogy of Fallot
extension of the incision into the out¯ow tract. As a conse- under the age of two years. Br Heart J 1985;53:666±667.
quence these patients will have a `tranannular patch' albeit [11] Castaneda AR, Freed MD, Williams RG, Norwood WI. Repair of
not a ventriculotomy. tetralogy of Fallot in infancy: early and late results. J Thorac Cardi-
ovasc Surg 1977;74:372±381.
In conclusion, we believe that it is possible to operate on [12] Gustafson RA, Murray GF, Warden HE, Hill FC, Edward Rozar Jr. G.
patients with Fallot's tetralogy with very low mortality and Early primary repair of tetralogy of Fallot. Ann Thorac Surg
morbidity but in order to achieve the best results, it is neces- 1988;45:235±241.
sary to adapt the surgical approach to the individual patient. [13] Starnes VA, Luciani GB, Latter DA, Grif®n MC. Current surgical
Palliation is sometimes necessary, but primary repair at management of tetralogy of Fallot. Ann Thorac Surg 1994;58:211±
215.
presentation remains our aim. [14] Barrat-Boyes BG, Neutz JM. Primary repair of tetralogy of Fallot in
infancy using profound hypothermia with circulatory arrest and
limited cardiopulmonary bypass: a comparison with conventional
References two stage management. Ann Surg 1973;178:406±411.
[15] Hornereffer PJ, Zahka KG, Rowe SA, Manolio TA, Gott VL, Reitz
[1] Fuster V, McGoon DC, Kennedy MA, Ritter DG, Kirklin JW. Long BA, Gardner TJ. Long term results of total repair of tetralogy of Fallot
term evaluation [12±22 years] of open heart surgery for tetralogy of in childhood. Ann Thorac Surg 1990;50:179±185.
Fallot. Am J Cardiol 1980;46:635±642. [16] Uva MS, Lacour-Gayet F, Komiya T, Serraf A, Bruniaux J, Touchot
[2] Katz NM, Blackstone EH, Kirklin JW, Paci®co AD, Bargeron Jr LM. A, Roux D, Petit J, Planche C. Surgery for tetralogy of Fallot at less
Late survival and symptoms after repair of tetralogy of Fallot. Circu- than six months of age. J Thorac Cardiovasc Surg 1994;107:1291±
lation 1982;65:403±410. 1300.
[3] Horowitz LN, Vetter VL, Harken AH, Josephson ME. Electrophysio- [17] Caspi J, Zalstein E, Zucker N, Applebaum A, Harrison LH, Munfach
logic characteristics of sustained ventricular tachycardia occurring NA, Heck HA, Ferguson TB, Stopa A, White M, Fontenot EE. Surgi-
after repair of tetralogy of Fallot. Am J Cardiol 1980;46:446±452. cal management of Tetralogy of Fallot in the ®rst year of life. Ann
[4] Dietl CA, Torres AR, Cazzaniga ME, Favaloro RG. Right atrial Thorac Surg 1999;68:1344±1349.
approach for surgical correction of tetralogy of Fallot. Ann Thorac [18] Hennein HA, Mosca RS, Urcelay G, Crowley DC, Bove EL. Inter-
Surg 1989;47:546±552. mediate results after complete repair of Tetralogy of Fallot in
[5] Karl TR, Sano S, Pornviliwan S, Mee RBB. Tetralogy of Fallot: neonates. J Thorac Cardiovasc Surg 1995;109:332±344.