- Email: [email protected]
Recurrent obstruction after subclavian flap repair of coarctation of the aorta in infants
J
THoRAc CARDIOVASC SURG
91:738-746, 1986
Recurrent obstruction after subclavian flap repair of coarctation of the aorta in infants Can it be predicted or prevented? Recoarctation is a problem in some patients after subclavian flap aortoplasty. To investigate the reason for recoarctation, we reviewed the records of 26 infants who underwent subclavian flap repair for symptomatic coarctation of the aorta at less than 3 months of age between June, 1979, and December, 1983. Age at repair ranged from 2 to 65 days (median 16 days) and weight from 2.1 to 4.9 kg (median 3.4 kg), In 14 patients the coarctation was associated with significant intracardiac defects (complex in six). There were two intraoperative deaths and one early death (surgical mortality 12 %). The survivors were followed from 6 weeks to 66 months (median 12 months). Five survivors(22 %~ all operated on at less than 14 days of age, developed severe recoarctation 6 weeks to 6 months (median 5 months) after repair. The obstruction appeared to be due to lumen obliteration by shelf-life posterior wall tissue. Morphometric analysis of preoperative angiograms showed no correlation between recoarctation and (1) distance between the left subclavian artery and the site of coarctation, (2) length of the isthmus, (3)diameter of the isthmus, (4) combined cross-sectional area of the left subclavian artery and isthmus, or (5) the ratio of the combined cross-sectional area of the left subclavian artery and isthmus to the cross-sectional area of the descending thoracic aorta. Recoarctation did not correlate with weight at operation, but it correlated significantly with age at aortoplasty (p = 0.02). The results suggest that intrinsic abnormalities of the periductal aortic wall are responsible for recoarctation after subclavian flap aortoplasty. Particular attention to this abnormal tissue at repair may prevent early recurrence in young infants.
Guillermo R. Sanchez, M.D., Rohinton K. Balsara, M.D., Jeffrey M. Dunn, M.D., Ashok V. Mehta, M.D., and Anna C. O'Riordan, M.D., Philadelphia, Pa.
Subclavian flap aortoplasty is considered by some to be the operation of choice in infants with symptomatic coarctation of the thoracic aorta.':' Recurrent obstruction remains a problem in some patients, however, for reasons that are not known. We have reviewed our experience with subclavian flap aortoplasty in infants less than 3 months of age, attempting in particular to identify the causes of recoarctation.
From the Sections of Cardiology and Cardiothoracic Surgery, St. Christopher's Hospital for Children, and the Departments of Pediatrics and Surgery, Temple University School of Medicine, Philadelphia, Pa. Received for publication April 15, 1985. Accepted for publication June 21, 1985. Address for reprints: Guillermo R. Sanchez, M.D., Section of Cardiology, St. Christopher's Hospital for Children, 5th St. and Lehigh Ave., Philadelphia, Pa. 19133.
738
Patients and methods
Between June, 1979, and December, 1983, 26 infants less than 3 months of age underwent subclavian flap aortoplasty for repair of coarctation of the aorta. Twenty-three infants had congestive heart failure, one had signs of right-to-left shunting at the ductus level, one had cyanosis, and one had hypertension and a cerebrovascular accident at the time of presentation. In eight patients the coarctation was associated with a significant ventricular septal defect (VSD), and in six patients complex intracardiac anomalies were present (univentricular heart in three; dextrocardia with atrioventricular defect, right-sided aortic arch, and bilateral patent ductus arteriosi in one; tricuspid atresia and d-transposition of the great arteries in one; and mitral and aortic valve stenosis in one patient). Age at aortoplasty ranged from 2 to 65 days (median 16 days) and weight from 2.1 to 4.9 kg (median 3.4 kg).
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
7 39
May. 1986
Subclavian flap aortoplasty was performed through a lateral thoracotomy at the fourth intercostal space. The subclavian artery was mobilized as far distally as possible and its branches were ligated and divided. The ductus or ligamentum arteriosum was ligated (also transected in one patient). An intraluminal coarctation shelf, if present, was excised. The repair was accomplished with 6-0 continuous nonabsorbable monofilament sutures. In each case the distal end of the subclavian flap appeared to extend well beyond the coarctation site. Relief of coarctation was assessed by arm/leg blood pressure determinations in the postoperative period. Outcome was described as good, fair, or poor depending on the arm/leg pressure gradient on followup (good, pressure gradient ~ 10 mm Hg; fair, 11 to 20 mm Hg; poor, >20 mm Hg). Recoarctation was suspected when previously normal femoral pulses became feeble or absent. Postoperative cardiac catheterization was performed in 12 patients either in preparation for recoarctectomy (five patients) or as part of the evaluation of associated intracardiac defects (seven patients). Preoperative angiograms were reviewed and the following measurements were performed: distance between the origin of the left subclavian artery and the site of coarctation, length of the isthmus, diameter of the left subclavian artery, diameter of the isthmus, and diameter of the descending thoracic aorta (Fig. 1). Magnification was corrected using known catheter diameters. On the basis of angiographic measurements, the following dimensions were calculated: cross-sectional area of the left subclavian artery, cross-sectional area of the isthmus, total proximal cross-sectional area (i.e., subclavian artery plus isthmus), cross-sectional area of the descending thoracic aorta, and the ratio of total proximal cross-sectional area to cross-sectional area of the descending thoracic aorta. Statistical analysis was performed by paired t testing and chi square testing. Results Early deaths. There were three early deaths, all in patients with serious associated intracardiac abnormalities and some variant of the hypoplastic left heart syndrome (Table I, A). Patient 1 had a VSD and aortic valve stenosis, and ventricular fibrillation developed during pericardiotomy for pulmonary artery banding. Patient 2 had a VSD, and at operation (after completion of the flap aortoplasty) associated hypoplasia of the transverse aortic arch was noted. Ventricular fibrillation developed intraoperatively in this patient also, during isolation of the ascending aorta for insertion of an
Fig. 1. Aortogram indicating points of measurement for morphometric analysis. a, Distance between origin of left subclavian artery and site of coarctation. b, Length of isthmus. C, Diameter of left subclavian artery. d, Diameter of the isthmus. e, Diameter of descending aorta.
ascending aorta-descending aorta conduit. Patient 3, who had tricuspid atresia and transposition of the great arteries, died during atrial septectomy 3 weeks after the operation. Thus, the surgical mortality was 12% (intraoperative mortality 8%). Late deaths. Six patients died during the follow-up period (Table I, B). Patient 4, who had aortic and mitral valve stenosis, died during the initial hospitalization 3 months postoperatively, after a low-output course complicated by central nervous system dysfunction and necrotizing enterocolitis. Autopsy showed no sign of recoarctation, and this patient is included in the group with a good aortoplasty result. Patient 5, with a VSD and other major, noncardiac congenital anomalies, had undergone subclavian flap aortoplasty at 11 days of age and VSD repair at 2Vz months of age. Follow-up blood pressures had shown no sign of recoarctation. He died at 5V2 months of age of suspected sepsis. At autopsy there was no sign of recoarctation. Patient 6, with a large
The Journal of
7 4 0 Sanchez et al.
Thoracic and Cardiovascular Surgery
Table I. Hospital and late deaths Associated Patient
defect
Age at operalion (days)
at operalion
19 33 12
3.1
Weight (kg)
Cause of death
A. Hospital deaths 1 2 3
YSD,AS YSD, AAH TA,TGA
B. Late deaths 4 AS, MS 5 YSD 6 YSD,ACA 7 UYH 8 YSD, MS 9 UYH
37
3.2
4.2
8
3.0 2.1 3.6 4.3 3.2
7
3.5
10 21
65
YF at pulmonary artery banding Y F at dissection of the aorta YF at atrial septectomy 3 wk postop. Low output, CNS, NEC, chronic CHF, 4 mo postop. Suspected sepsis 5 mo after YSD repair Respiratory infection at 6 mo of age; ACA at autopsy Suspected arrhythmia at 12 mo of age Intraop. death at YSD repair at 14 mo of age Intraop. death at open palliative operation at 15 mo of age
Legend: AAH, Hypoplastic tranverse arch. ACA, Anomalous left coronary artery. AS, Valvular aortic stenosis. CHF, Congestive heart failure. CNS, Central nervous system. MS, Mitral stenosis. NEC, Necrotizing enterocolitis. TA, Tricuspid atresia. TGA, d-Transposition of the great arteries. UVH, Univentricular heart. VF, Ventricular fibrillation. VSD, Ventricular septal defect.
VSD and chronic congestive heart failure, died 5 months after subclavian flap aortoplasty during a respiratory infection. Blood pressure measurements 3 weeks after repair had shown no sign of recoarctation (pressure gradient 0 mm Hg). Subsequent blood pressure measurements were not performed, but pulses remained of equal amplitude in the right arm and legs. Autopsy showed an anomalous origin of the left coronary artery from the main pulmonary artery, in addition to the previously known defects. No myocardial infarct was present. The area of coarctation repair was patent but moderately narrow. The outcome of aortoplasty in this patient was characterized as fair, with signs of recoarctation. Patient 7, who had a univentricular heart and an outlet chamber supporting the aorta, had had subclavian flap repair at 2 months of age and enlargement of the bulboventricular foramen at 4 months of age. Postoperatively there was intermittent atrioventricular block. Blood pressure measurements at 11 months of age showed a good result (pressure gradient 10 mm Hg). He died suddenly during a respiratory illness 1 month later, presumably of an arrhythmia. Autopsy showed a good coarctation repair. Patient 8, who had a VSD and mitral valve stenosis, had undergone recoarctectomy at 6 months of age. He died during intracardiac repair of the VSD at 14 months of age. Patient 9, with a univentricular heart, died at 15 months of age, 1 day after an open palliative operation. Blood pressure measurements on this last admission and autopsy showed no sign of recurrent coarctation. Deaths in patients with associated VSD. In eight of our patients coarctation was associated with a significant VSD. Two of these patients had a variant of hypoplastic left heart syndrome, and died at initial
repair (Table I, A). Three of the other six patients with significant VSDs died late. However, one of these died of unrelated causes 5 months after VSD repair (Patient 5, Table I, B), one had a variant of hypoplastic left heart syndrome (Patient 8, Table I, B), and the third had an associated anomalous left coronary artery (Patient 6, Table I, B). Thus, of the five patients with a significant VSD without hypoplasia of the left heart who survived the subclavian flap procedure, only one has eventually died of cardiac-related causes and three were alive and well when evaluated 12, 21, and 31 months after aortoplasty (one required open repair of the defect at 7 months of age). Follow-up of surgical survivors. The period of follow-up was taken to be the interval from subclavian flap repair to recognition of recoarctation necessitating operation, the time of in-hospital late death, or the time of the last follow-up visit. Thus, in the 23 surgical survivors, the follow-up period ranged from 6 weeks to 66 months (median 12 months). During this period 15 of the surgical survivors continued to show a good result (12 patients with pressure gradient 0 mm Hg or less). In two patients signs of moderate recoarctation (pressure gradient 20 mm Hg) appeared 10 months after repair. Five patients became hypertensive and developeda large arm/leg systolic pressure gradient (>60 mm Hg), all within 6 months of aortoplasty (range 6 weeks to 6 months, median 5 months). One patient (Patient 6), as described in the previous section, had signs of moderate recoarctation at autopsy at 6 months of age. Thus, in the group of surgical survivors, recoarctation was always recognized within 1 year of subclavian flap aortoplasty. No sign of recoarctation was observed in 12 patients who were followed up for 12months or longer.
Volume 91
Subclavian flap repair of aortic coarctation 7 4 1
Number 5 May, 1986
40
A
30 20 10
8
• I
t
--
B
•
70 60 50
30
10
II
o
• •
Q
30
10
20
U
o o
••
4
2 GOOD
FAIR
I
•
• c • •
•
POOR
OUTCOME Fig. 2. Relationship between outcome in patients who survived initial repair and anatomic dimensions of distal aortic arch, A. Distance between origin of left subclavian artery and site of coarctation. B, Length of isthmus. C. Diameter of the isthmus. LSCA, Left subclavian artery. DAo. Descending aorta. Filled circles represent those patients operated on at 3 weeks of age or less.
1.0
.9
o
.8
o o
.7 .6
.5
o
9
10
20
6
80 E E
40
40
8
N
•
•• • •
•
• E
•
o
•
o
•a
I
o
•
.4
• • •
.3
.2 .1 GOOD
FAIR
POOR
OUTCOME
Fig. 2. Cont'd. D. Total proximal cross-sectional area. E. Morphometric analysis of the distal aortic arch and developmentof recoarctation. For purposes of this analysis, the three patients with signs of moderate recoarctation were combined with the group having severe recoarctation. In 19 of 23 surgical survivors, preoperative angiograms of satisfactory quality were available that allowed measurements of the distal aortic arch and periductal region. These measurements were then correlated with the surgical outcome. Recoarctation wasnot significantly related to the distance between the origin of the left subclavian artery and the site of coarctation (Fig. 2, A), to the length of the aortic isthmus (Fig. 2, B), to the diameter of the isthmus (Fig. 2, C), to the total proximal cross-sectional area (Fig. 2, D), or to the ratio of the proximal cross-sectional area to
Ratio of total proximal cross-sectional area to cross-sectional area of descending aorta. Filled circlesrepresent those patients operated on at 3 weeks of age or less.
the cross-sectional area of the descending aorta (Fig. 2, E). Weight and age at operation and development of recoarctation. Recoarctation was not related significantly to the dimensions of the distal aortic arch and periductal region. Thus, the four surgical survivors without satisfactory preoperative angiograms were included In the study of the relationship between recoarctation and the weight and age at subclavian flap aortoplasty. No significant relationship was observed
The Journal of
7 4 2 Sanchez et al.
-a= C) ~
~
~
Surgery
A
6.0 5.0
w C) a= 4.0 ~ en 3.0 ~ ~
Thoracic and Cardiovascular
2.0
. o
88
••
SO
• •
•L
C)
w 1.0 ~
en
-sa= ~
70 60 50
B
8 o o
~
w 40 a= 30 ~ en ~ 20 w 10 C) C)
«
0 0
9
•
0 0 00
0 0
80
GOOD
FAIR
POOR
OUTCOME Fig. 3. Relationship between outcome in patients who survived initial repair and (A) weight and (B) age at subclavian flap aortoplasty. Filled circles represent those patients operated on at 3 weeks of age or less.
between recoarctation and weight at operation (Fig. 3, A). On the other hand, a significant relationship was observed between recoarctation and the age at subclavian flap aortoplasty, with recoarctation being observed only in patients who were 21 days of age or less at operation (Fig. 3, B) (p = 0.02). Outcome in patients 3 weeks of age or less at aortoplasty. Because recoarctation did not develop in some patients operated on in the first 3 weeks of age, the group of infants in this age group who had recoarctation was compared to the group who did not (Table 11). No significant difference was found between the two groups with respect to the age and weight at operation or to the dimensions of the distal aortic arch and periductal
region. The incidence of recoarctation was higher in girls operated on in the first 3 weeks of age than in boys operated on at the same age, but this difference was not statistically significant. There were more patients with a patent ductus arteriosus at aortoplasty in the group that did not have recoarctation than in the group that did, but this difference also was not statistically significant. Perioperative infusion of prostaglandin E I had no effect on outcome. Recoarctation was not related to the presence of complex congenital heart disease or to a hemodynamically significant YSD. Recoarctation: Surgical findings. Six patients underwent reoperation because of recoarctation. In one of these, Patient 14, a 20 mm Hg systolic pressure gradient between the right arm and the legs was recognized 10 months after subclavian flap aortoplasty, and it was elected to repair the recurrent coarctation during insertion of a permanent epicardial pacemaker (Table III). At reoperation, the striking finding in all six patients was the obliteration of the lumen by local tissue growth that appeared to originate from the posterior aortic wall. In four of the six patients the tissue growth resembled a shelf. One of these four patients was a patient in whom specific mention had been made during the subclavian flap aortoplasty that no shelf was present. In the other two patients the obstruction was less discrete and was at or near the distal anastomosis. External inspection of the aorta at reoperation in one of these two patients also showed the local aorta to be constricted, apparently because of traction by the remnant of the ductus that had been ligated (but not divided) during aortoplasty. Reoperation in all six patients consisted of polytetrafluoroethylene patch angioplasty, which relieved the obstruction without complications. Discussion Subclavian flap aortoplasty has been recommended as the procedure of choice in infants with coarctation of the aorta because of technical reasons, its potential to allow long-term growth of the periductal aorta, and the excellent results that it has produced.':" Recoarctation has occurred after subclavian flap aortoplasty in a significant number of patients, 1,2,8-10.13 however, and for this reason, some have recommended other methods of repair of coarctation in infancy.":" The reason for recoarctation is not fully understood but has been attributed to tension at the anastomotic site," a short flap," 9. II poor growth of the aorta at the site of repair," inadequate shelf resection," proliferation of residual abnormal tissue," and constriction of residual ductal tissue." Our results show, as those of others have,
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
743
May, 1986
Table II. Comparison of clinical aspects and morphometric results in patients 3 weeks of age or less at subclavian flap aortoplasty No. of patients Age at aortoplasty (days) Weight at aortoplasty (kg) Males/females Patent ductus arteriosus Prostaglandin E 1 Distance from left subclavian artery to coarctation (ern) Length of isthmus (em) Diameter of isthmus (mm') Total proximal cross-sectional area (mm') Ratio of total proximal cross-sectional area to cross-sectional area of descending aorta Complex intracardiac anomaly Ventricular septal defect
No recoarctation
Recoarctation
p Value
7 9.2 ± 5.1 3.2 ± 0.7 5/2 6
8 13.8 ± 6.1 3.2 ± OJ 2/6 4 2 1.0 ± 0.3 0.6 ± 0.2 4.1 ± 1.3 30 ± 19 0.68 ± 0.23
NS NS NS NS NS NS NS NS NS NS
I 2
NS NS
2 1.0 0.8 4.0 28 0.54
± 0.3
± 0.2 ± 0.8 ± 5
± 0.04 2 3
Legend: Values are expressed as mean ± standard deviation. NS. Not significant.
Table m. Descriptive data of patients who underwent recoarctectomy Patient
10
Age at SFA (days) 5
8·
8
II 12 13
II 12 13
14
20
Findings at SFA
Age at recoarctation
Discrete coarctation Isthmus hypoplasia Discrete coarctation Discrete coarctation Isthmus hypoplasia, "no shelf' Discrete coarctation
6 3 5 6 5
Findings at recoarctectomy
wk mo mo mo mo
N arrow distal anastomosis Narrow distal anastomosis with ligament traction Shelf Hypoplastic back wall with shelf-like tissue Shelf
t
Shelf
Legend: SF A, Subclavian flap aortoplasty. 'Same patient as in Table I. tArm/leg pressure gradient of 20 mm Hg 10 months after subclavian flap aortoplasty. Elective patch aortoplasty at time of pacemaker insertion at 11 months of age.
that the subclavian flap procedure provides excellent immediate relief of coarctation of the aorta in very young infants. Thus, the length of the dissected subclavian artery is usually sufficient to bridge over the stenotic area. We have also observed, as others have, that the subclavian flap aortoplasty is complicated by a significant incidence of early recurrence. The results of our morphometric analysis of the distal aortic arch in the group as a whole as well as in the high-risk group indicate that recoarctation after subclavian flap aortoplasty is not dependent on the length of the aortic isthmus or on the distance between the origin of the left subclavian artery and the site of coarctation. We believe itisunlikely, therefore, that recoarctation is the result of early tension at the distal anastomosis. The early relief of coarctation in these patients demonstrates that the subclavian artery and isthmus, when brought together, provide a vascular lumen satisfactory to flow, regardless of the dimensions of the distal
aortic arch and periductal region. This observation agrees with the conclusions drawn from an anatomopathologic study that the subclavian flap procedure was the one best suited to achieve early success in this lesion." The long-term success of the procedure, however, requires satisfactory growth of the new lumen. In an animal model, growth of the aortic area was satisfactory for 6 months after subclavian flap aortoplasty, regardless of whether continuous or interrupted nonabsorbable sutures were used in the repair." Clinical studies in infants, on the other hand, support the view that the use of interrupted sutures or continuous, absorbable sutures may prevent recoarctation." If recoarctation after subclavian flap aortoplasty in our patients were due to poor growth of the new lumen, then recoarctation should manifest itself first in those patients with the narrowest isthmus or the smallest total proximal cross-sectional area (either in absolute terms or in proportion to the diameter of the descending aorta). This pattern was not
7 4 4 Sanchez et al.
observed, nor was a significant relationship observed between recoarctation and the duration of follow-up. Therefore, we do not believe that recoarctation in our patients is due to poor growth of the anastomosis. The lack of correlation between the weight of the infant at subclavian flap aortoplasty and the development of recoarctation argues against technical factors being primarily responsible for recoarctation. The finding of a significant relationship between age at operation and development of recoarctation was unexpected, but is in agreement with the recently published observations in a larger group of patients." That this relationship had not been generally appreciated may have been due to the relatively small number of infants who have undergone the subclavian flap procedure at less than 1 month of age. Review of the literature, however, tends to support this observation. In a series of nine infants less than 6 months of age at subclavian flap repair (average age at repair 4.4 weeks), reported by Kamau and associates,'? the two infants who subsequently had "significant" (40 mm Hg) arm/leg gradients were 3 days and 9 days of age at repair. Pierce and associates' reported on 12 infants, the two youngest of whom were operated on at 2 weeks of age. One had a 10 mm Hg arm/leg gradient 4 months after the operation. The other was followed for only 1 month. In a series presented by Thibault, Sperling, and Gazzaniga," three patients were 14 days of age or younger at operation and two of these survived the procedure. One of these two survivors later had a 30 mm Hg arm/leg gradient (proximal arch hypoplasia was not ruled out) and the other had a 12 mm Hg gradient. None of the other survivors, operated on at ages over 2 weeks of age, had significant gradients. In Penkoske's series," the two infants who required reoperation were 17 and 22 days of age at subclavian aortoplasty. Other, larger series do not lend themselves to evaluation because of the age breakdown that is provided,' absence of information about the age at operation of those infants who had recoarctation,':" I I or absence of information concerning blood pressure measurements- II or arm/leg pressure gradients on follow-up." 5, I I If recoarctation is not due to poor growth of the anastomosis, the development of obstruction after repair must be due to another process that eventually compromises the lumen. In patients with coarctation of the aorta, ductus tissue extends in circumferential fashion into the aorta to a greater degree than normal." In addition, encirclement of the aortic isthmus by ductal tissue has been observed in young infants with coarctation.'? Furthermore, in young infants the coarctation shelf is usually preductal in position, whereas in older
The Journal of Thoracic and Cardiovascular Surgery
patients it is most often postductal," suggesting that all coarctations develop preductally with gradual distal migration of the shelf as the child grows. 18 Since subclavian flap aortoplasty does not involve resection of the coarctation segment, repair by this method, even when any shelf that is present is excised, is likely to leave behind ductal tissue. The latter in the young infant may not have completely involuted and would therefore have the potential to cause local distortion of the aorta. The presence of residual ductal tissue may explain the relationship that we and others" have observed between the age at repair and the development of recoarctation. Local distortion, scarring, and intimal proliferation, complicated by growth and distal migration of residual/ recurrent shelf tissue, would lead to obliteration of the lumen and would explain the development of recoarctation in these patients, as well as the unexpected proximity between the restenosing tissue and the distal end of the flap observed at reoperation in two of our patients. Our series clearly shows that recoarctation is not necessarily the rule in infants operated on in the first 3 weeks of life. This is also shown by the results reported in infants less than 1 month" and less than 1 week of age' at subclavian flap aortoplasty. Thus, it is possible that the problems presented by the abnormal local tissue can be surmounted by meticulous attention to excision of local coarctation tissue in the very young infant. However, such technique might require extensive, "deep" shelf resection and be complicated by late aneurysm formation. It is clearly important that the length of the divided subclavian artery be sufficient to satisfactorily bridge over the area of coarctation. In our patients the degree of dissection of the subclavian artery was similar to that described by others.v" The observation that in our patients subclavian flap aortoplasty relieved the obstruction indicates that the divided subclavian artery was long enough at the initial repair. It is possible, however, that in very young infants the growth of the distal aortic arch and periductal region is such that obstruction may redevelop if the flap is not sufficiently long to allow for local scarring and retraction. One way to assess this would have been to determine the length of the "surgically available" subclavian artery (i.e., distance between the origin of the subclavian and the origin of the vertebral and/or internal mammary arteries) and to correlate this length with the distance between the subclavian artery and the site of coarctation. Unfortunately, the angiograms in our patients did not permit measurement of the length of the subclavian artery. Nevertheless, the need to achieve maximum length of
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
745
May, 1986
the subclavian artery before it is divided cannot be overemphasized. The observationof local traction by the ligamentum arteriosum in one patient with recoarctation (Patient 8, Table III) is noteworthy but not sufficient to conclude that division of the ductus or ligamentum arteriosum at subclavian aortoplasty is beneficial. It is difficult to conclude from our findings that subclavian flap aortoplasty should be delayed until at least 3 weeks of age, particularly because infants with associated intracardiac defects require emergency coarctectomy' and because no deaths occurred in conjunction with recoarctation. The results suggest, however, that other methods of repair should be considered in infants younger than 3 weeksof age. Resection with end-to-end anastomosis has been associated with a higher incidence of recoarctation.>" even when extensive resection has been performed." However, better results have been observed when extensive resection has been combined with improved suture material. 13, 23, 24 Prosthetic patch aortoplasty has also relieved the coarctation in infants. 14, 25, 26 It may be that the ability to enlarge the circumference of the coarctation area with a patch considerably larger in width than the opened subclavian artery may prevent critical compromise of the new lumen by local tissue scarring and growth. Our results in infants older than 3 weeks of age at aortoplasty support the conclusions of others that subclavianflap aortoplasty produces excellentresults despite anatomic variations of the aortic arch. Long-term evaluation of the patients at rest and with physical exercise remains essential to completely evaluate the outcome of subclavian patch repair. We gratefully acknowledge the assistance of Lydia Garcia inpreparation of this manuscript.
REFERENCES I Midgley FM, Scott LP, Perry LW, Shapiro SR, McClenathan JE: Subclavian flap aortoplasty for treatment of coarctation in early infancy. J Pediatr Surg 13:265-268, 1978 2 Hamilton 01, Di Eusanio G, Sandrasagra FA, Donnelly RJ: Early and late results of aortoplasty with a left subclavian flap for coarctation of the aorta in infancy. J THORAC CARDIOVASC SURG 75:699-704, 1978 3 Pierce WS, Waldhausen JA, Berman W Jr, Whitman V: Late results of the subclavian patch procedure in infants with coarctation of the thoracic aorta. Circulation 58:Suppl 1:78-82, 1978 4 Waldhausen JA, Whitman V, Werner JC, Pierce WS: Surgical intervention in infants with coarctation of the aorta. J THoRAc CARDIOVASC SURG 81:323-325, 1981
5 Bergdahl LAL, Blackstone EH, Kirklin JW, Pacifico AD, Bargeron LM: Determinants of early success in repair of aortic coarctation in infants. J THORAC CARDIOVASC SURG 83:736-742,1982 6 Moulton AL, Brenner JI, Roberts G, Tavares S, Ali S, Nordenberg A, Burns JE, Ringel R, Berman MA: Subclavian flap repair of coarctation of the aorta in neonates. Realization of growth potential? J THORAC CARDIOVASC SURG 87:220-235, 1984 7 Penkoske PA, Williams WG, Olley PM, LeBlanc J, Trusler GA, Moes CAF, Judakin R, Rowe RD: Subclavian arterioplasty repair of coarctation of the aorta in the first year of life. J THoRAc CARDIOVASC SURG 87:894-900, 1984 8 Campbell DB, Waldhausen JA, Pierce WS, Fripp R, Whitman V: Should elective repair of coarctation of the aorta be done in infancy? J THORAC CARDIOVASC SURG 88:929-938, 1984 9 Thibault WN, Sperling DR, Gazzaniga AB: Subclavian artery patch angioplasty. Treatment of infants and young children with aorta coarctation. Arch Surg 110: 10951098, 1975 10 Kamau P, Miles V, Toews W, Kelminson L, Friesen R, Lockhart C, Butterfield J, Hernandez J, Hawes CR, Pappas G: Surgical repair of coarctation of the aorta in infants less than six months of age. Including the question of pulmonary artery banding. J THORAC CARDIOV ASC SURG 81:171-179, 1981 11 Nair UR, Jones 0, Walker DR: Surgical management of severe coarctation of the aorta in the first month of life. Review of 48 consecutive cases. J THoRAc CARDIOVASC SURG 86:587-590, 1983 12 Allen RG, Marin-Garcia J, Nayek G: Methods of management and results following surgery for coarctation of the aorta in infancy. J Pediatr Surg 15:953-960, 1980 13 Cobanoglu A, Teply JF, Grunkemeier GL, Sunderland CO, Starr A: Coarctation of the aorta in patients younger than three months. A critique of the subclavian flap operation. J THoRAc CARDIOVASC SURG 89:128-135, 1985 14 Sade RM, Taylor AB, Chariker EP: Aortoplasty compared with resection for coarctation of the aorta in young children. Ann Thorac Surg 28:346-353, 1979 15 Metzdorff MT, Cobanoglu A, Grunkemeier GL, Sunderland CO, Starr A: Influence of age at operation on late results with subclavian flap aortoplasty. J THORAC CARDIOVASC SURG 89:235-241, 1985 16 Pellegrino A, Deverall PB, Anderson RH, Smith A, Wilkinson JL, Russo P, Girod DA, Tynan M: Aortic coarctation in the first three months of life. An anatomopathological study with respect to treatment. J THoRAc CARDIOVASC SURG 89:121-127,1985 17 Pae WE Jr, Myers JL, Waldhausen JA, Prophet GA, Pierce WS: Subclavian flap angioplasty. Experimental study in growing piglets. J Thor Cardiovasc Surg 82:922927, 1981 18 Elzenga NJ, Gittenberger-de Groot AC: Localised coarc-
7 4 6 Sanchez et al.
19
20
21
22
tation of the aorta. An age dependent spectrum. Br Heart J 49:317-323, 1983 Ho SY, Anderson RH: Coarctation of the aorta, Paediatric Cardiology, vol 2, Heart Disease in the Newborn, MS Godman, RM Marquis, OOs., Edinburgh, 1979 Churchill Livingstone, pp 173-186 Khoury GH, Hawes CR: Recurrent coarctation of the aorta in infancy and childhood. J Pediatr 72:801-806, 1968 Hartmann AF Jr, Goldring D, Hernandez A, Behrer MR, Schad N, Ferguson T, Burford T: Recurrent coarctation of the aorta after successful repair in infancy. Am J Cardiol 25:405-415, 1970 Williams WG, Shindo G, Trusler GA, Dische MR, Olley PM: Results of repair of coarctation of the aorta during infancy. J THORAC CARDIOVASC SURG 79:603-608, 1980
The Journal of Thoracic and Cardiovascular Surgery
23 Harlan JL, Doty DB, Brandt B III, Ehrenhaft JL: Coarctation of the aorta in infants. J THoRAc CARDIOVASC SURG 88:1012-1019,1984 24 Korfer R, Meyer H, Kleikamp G, Bircks W: Early and late results after resection and end-to-end anastomosis of coarctation 'of the thoracic aorta in early infancy. J THoRAc CARDIOVASC SURG 89:616-622, 1985 25 Fleming WH, Sarafian LB, Clark EB, Dooley KJ, Hofschire PJ, Hopeman AR, Ruckman RN, Mooring PK: Critical aortic coarctation. Patch aortoplasty in infants less than age 3 months. Am J Cardiol 44:687-690, 1979 26 Sade RM, Crawford FA, Hohn AR, Riopel DA, Taylor AB: Growth of the aorta after prosthetic patch aortoplasty for coarctation in infants. Ann Thorac Surg 38:21-25, 1984
THoRAc CARDIOVASC SURG
91:738-746, 1986
Recurrent obstruction after subclavian flap repair of coarctation of the aorta in infants Can it be predicted or prevented? Recoarctation is a problem in some patients after subclavian flap aortoplasty. To investigate the reason for recoarctation, we reviewed the records of 26 infants who underwent subclavian flap repair for symptomatic coarctation of the aorta at less than 3 months of age between June, 1979, and December, 1983. Age at repair ranged from 2 to 65 days (median 16 days) and weight from 2.1 to 4.9 kg (median 3.4 kg), In 14 patients the coarctation was associated with significant intracardiac defects (complex in six). There were two intraoperative deaths and one early death (surgical mortality 12 %). The survivors were followed from 6 weeks to 66 months (median 12 months). Five survivors(22 %~ all operated on at less than 14 days of age, developed severe recoarctation 6 weeks to 6 months (median 5 months) after repair. The obstruction appeared to be due to lumen obliteration by shelf-life posterior wall tissue. Morphometric analysis of preoperative angiograms showed no correlation between recoarctation and (1) distance between the left subclavian artery and the site of coarctation, (2) length of the isthmus, (3)diameter of the isthmus, (4) combined cross-sectional area of the left subclavian artery and isthmus, or (5) the ratio of the combined cross-sectional area of the left subclavian artery and isthmus to the cross-sectional area of the descending thoracic aorta. Recoarctation did not correlate with weight at operation, but it correlated significantly with age at aortoplasty (p = 0.02). The results suggest that intrinsic abnormalities of the periductal aortic wall are responsible for recoarctation after subclavian flap aortoplasty. Particular attention to this abnormal tissue at repair may prevent early recurrence in young infants.
Guillermo R. Sanchez, M.D., Rohinton K. Balsara, M.D., Jeffrey M. Dunn, M.D., Ashok V. Mehta, M.D., and Anna C. O'Riordan, M.D., Philadelphia, Pa.
Subclavian flap aortoplasty is considered by some to be the operation of choice in infants with symptomatic coarctation of the thoracic aorta.':' Recurrent obstruction remains a problem in some patients, however, for reasons that are not known. We have reviewed our experience with subclavian flap aortoplasty in infants less than 3 months of age, attempting in particular to identify the causes of recoarctation.
From the Sections of Cardiology and Cardiothoracic Surgery, St. Christopher's Hospital for Children, and the Departments of Pediatrics and Surgery, Temple University School of Medicine, Philadelphia, Pa. Received for publication April 15, 1985. Accepted for publication June 21, 1985. Address for reprints: Guillermo R. Sanchez, M.D., Section of Cardiology, St. Christopher's Hospital for Children, 5th St. and Lehigh Ave., Philadelphia, Pa. 19133.
738
Patients and methods
Between June, 1979, and December, 1983, 26 infants less than 3 months of age underwent subclavian flap aortoplasty for repair of coarctation of the aorta. Twenty-three infants had congestive heart failure, one had signs of right-to-left shunting at the ductus level, one had cyanosis, and one had hypertension and a cerebrovascular accident at the time of presentation. In eight patients the coarctation was associated with a significant ventricular septal defect (VSD), and in six patients complex intracardiac anomalies were present (univentricular heart in three; dextrocardia with atrioventricular defect, right-sided aortic arch, and bilateral patent ductus arteriosi in one; tricuspid atresia and d-transposition of the great arteries in one; and mitral and aortic valve stenosis in one patient). Age at aortoplasty ranged from 2 to 65 days (median 16 days) and weight from 2.1 to 4.9 kg (median 3.4 kg).
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
7 39
May. 1986
Subclavian flap aortoplasty was performed through a lateral thoracotomy at the fourth intercostal space. The subclavian artery was mobilized as far distally as possible and its branches were ligated and divided. The ductus or ligamentum arteriosum was ligated (also transected in one patient). An intraluminal coarctation shelf, if present, was excised. The repair was accomplished with 6-0 continuous nonabsorbable monofilament sutures. In each case the distal end of the subclavian flap appeared to extend well beyond the coarctation site. Relief of coarctation was assessed by arm/leg blood pressure determinations in the postoperative period. Outcome was described as good, fair, or poor depending on the arm/leg pressure gradient on followup (good, pressure gradient ~ 10 mm Hg; fair, 11 to 20 mm Hg; poor, >20 mm Hg). Recoarctation was suspected when previously normal femoral pulses became feeble or absent. Postoperative cardiac catheterization was performed in 12 patients either in preparation for recoarctectomy (five patients) or as part of the evaluation of associated intracardiac defects (seven patients). Preoperative angiograms were reviewed and the following measurements were performed: distance between the origin of the left subclavian artery and the site of coarctation, length of the isthmus, diameter of the left subclavian artery, diameter of the isthmus, and diameter of the descending thoracic aorta (Fig. 1). Magnification was corrected using known catheter diameters. On the basis of angiographic measurements, the following dimensions were calculated: cross-sectional area of the left subclavian artery, cross-sectional area of the isthmus, total proximal cross-sectional area (i.e., subclavian artery plus isthmus), cross-sectional area of the descending thoracic aorta, and the ratio of total proximal cross-sectional area to cross-sectional area of the descending thoracic aorta. Statistical analysis was performed by paired t testing and chi square testing. Results Early deaths. There were three early deaths, all in patients with serious associated intracardiac abnormalities and some variant of the hypoplastic left heart syndrome (Table I, A). Patient 1 had a VSD and aortic valve stenosis, and ventricular fibrillation developed during pericardiotomy for pulmonary artery banding. Patient 2 had a VSD, and at operation (after completion of the flap aortoplasty) associated hypoplasia of the transverse aortic arch was noted. Ventricular fibrillation developed intraoperatively in this patient also, during isolation of the ascending aorta for insertion of an
Fig. 1. Aortogram indicating points of measurement for morphometric analysis. a, Distance between origin of left subclavian artery and site of coarctation. b, Length of isthmus. C, Diameter of left subclavian artery. d, Diameter of the isthmus. e, Diameter of descending aorta.
ascending aorta-descending aorta conduit. Patient 3, who had tricuspid atresia and transposition of the great arteries, died during atrial septectomy 3 weeks after the operation. Thus, the surgical mortality was 12% (intraoperative mortality 8%). Late deaths. Six patients died during the follow-up period (Table I, B). Patient 4, who had aortic and mitral valve stenosis, died during the initial hospitalization 3 months postoperatively, after a low-output course complicated by central nervous system dysfunction and necrotizing enterocolitis. Autopsy showed no sign of recoarctation, and this patient is included in the group with a good aortoplasty result. Patient 5, with a VSD and other major, noncardiac congenital anomalies, had undergone subclavian flap aortoplasty at 11 days of age and VSD repair at 2Vz months of age. Follow-up blood pressures had shown no sign of recoarctation. He died at 5V2 months of age of suspected sepsis. At autopsy there was no sign of recoarctation. Patient 6, with a large
The Journal of
7 4 0 Sanchez et al.
Thoracic and Cardiovascular Surgery
Table I. Hospital and late deaths Associated Patient
defect
Age at operalion (days)
at operalion
19 33 12
3.1
Weight (kg)
Cause of death
A. Hospital deaths 1 2 3
YSD,AS YSD, AAH TA,TGA
B. Late deaths 4 AS, MS 5 YSD 6 YSD,ACA 7 UYH 8 YSD, MS 9 UYH
37
3.2
4.2
8
3.0 2.1 3.6 4.3 3.2
7
3.5
10 21
65
YF at pulmonary artery banding Y F at dissection of the aorta YF at atrial septectomy 3 wk postop. Low output, CNS, NEC, chronic CHF, 4 mo postop. Suspected sepsis 5 mo after YSD repair Respiratory infection at 6 mo of age; ACA at autopsy Suspected arrhythmia at 12 mo of age Intraop. death at YSD repair at 14 mo of age Intraop. death at open palliative operation at 15 mo of age
Legend: AAH, Hypoplastic tranverse arch. ACA, Anomalous left coronary artery. AS, Valvular aortic stenosis. CHF, Congestive heart failure. CNS, Central nervous system. MS, Mitral stenosis. NEC, Necrotizing enterocolitis. TA, Tricuspid atresia. TGA, d-Transposition of the great arteries. UVH, Univentricular heart. VF, Ventricular fibrillation. VSD, Ventricular septal defect.
VSD and chronic congestive heart failure, died 5 months after subclavian flap aortoplasty during a respiratory infection. Blood pressure measurements 3 weeks after repair had shown no sign of recoarctation (pressure gradient 0 mm Hg). Subsequent blood pressure measurements were not performed, but pulses remained of equal amplitude in the right arm and legs. Autopsy showed an anomalous origin of the left coronary artery from the main pulmonary artery, in addition to the previously known defects. No myocardial infarct was present. The area of coarctation repair was patent but moderately narrow. The outcome of aortoplasty in this patient was characterized as fair, with signs of recoarctation. Patient 7, who had a univentricular heart and an outlet chamber supporting the aorta, had had subclavian flap repair at 2 months of age and enlargement of the bulboventricular foramen at 4 months of age. Postoperatively there was intermittent atrioventricular block. Blood pressure measurements at 11 months of age showed a good result (pressure gradient 10 mm Hg). He died suddenly during a respiratory illness 1 month later, presumably of an arrhythmia. Autopsy showed a good coarctation repair. Patient 8, who had a VSD and mitral valve stenosis, had undergone recoarctectomy at 6 months of age. He died during intracardiac repair of the VSD at 14 months of age. Patient 9, with a univentricular heart, died at 15 months of age, 1 day after an open palliative operation. Blood pressure measurements on this last admission and autopsy showed no sign of recurrent coarctation. Deaths in patients with associated VSD. In eight of our patients coarctation was associated with a significant VSD. Two of these patients had a variant of hypoplastic left heart syndrome, and died at initial
repair (Table I, A). Three of the other six patients with significant VSDs died late. However, one of these died of unrelated causes 5 months after VSD repair (Patient 5, Table I, B), one had a variant of hypoplastic left heart syndrome (Patient 8, Table I, B), and the third had an associated anomalous left coronary artery (Patient 6, Table I, B). Thus, of the five patients with a significant VSD without hypoplasia of the left heart who survived the subclavian flap procedure, only one has eventually died of cardiac-related causes and three were alive and well when evaluated 12, 21, and 31 months after aortoplasty (one required open repair of the defect at 7 months of age). Follow-up of surgical survivors. The period of follow-up was taken to be the interval from subclavian flap repair to recognition of recoarctation necessitating operation, the time of in-hospital late death, or the time of the last follow-up visit. Thus, in the 23 surgical survivors, the follow-up period ranged from 6 weeks to 66 months (median 12 months). During this period 15 of the surgical survivors continued to show a good result (12 patients with pressure gradient 0 mm Hg or less). In two patients signs of moderate recoarctation (pressure gradient 20 mm Hg) appeared 10 months after repair. Five patients became hypertensive and developeda large arm/leg systolic pressure gradient (>60 mm Hg), all within 6 months of aortoplasty (range 6 weeks to 6 months, median 5 months). One patient (Patient 6), as described in the previous section, had signs of moderate recoarctation at autopsy at 6 months of age. Thus, in the group of surgical survivors, recoarctation was always recognized within 1 year of subclavian flap aortoplasty. No sign of recoarctation was observed in 12 patients who were followed up for 12months or longer.
Volume 91
Subclavian flap repair of aortic coarctation 7 4 1
Number 5 May, 1986
40
A
30 20 10
8
• I
t
--
B
•
70 60 50
30
10
II
o
• •
Q
30
10
20
U
o o
••
4
2 GOOD
FAIR
I
•
• c • •
•
POOR
OUTCOME Fig. 2. Relationship between outcome in patients who survived initial repair and anatomic dimensions of distal aortic arch, A. Distance between origin of left subclavian artery and site of coarctation. B, Length of isthmus. C. Diameter of the isthmus. LSCA, Left subclavian artery. DAo. Descending aorta. Filled circles represent those patients operated on at 3 weeks of age or less.
1.0
.9
o
.8
o o
.7 .6
.5
o
9
10
20
6
80 E E
40
40
8
N
•
•• • •
•
• E
•
o
•
o
•a
I
o
•
.4
• • •
.3
.2 .1 GOOD
FAIR
POOR
OUTCOME
Fig. 2. Cont'd. D. Total proximal cross-sectional area. E. Morphometric analysis of the distal aortic arch and developmentof recoarctation. For purposes of this analysis, the three patients with signs of moderate recoarctation were combined with the group having severe recoarctation. In 19 of 23 surgical survivors, preoperative angiograms of satisfactory quality were available that allowed measurements of the distal aortic arch and periductal region. These measurements were then correlated with the surgical outcome. Recoarctation wasnot significantly related to the distance between the origin of the left subclavian artery and the site of coarctation (Fig. 2, A), to the length of the aortic isthmus (Fig. 2, B), to the diameter of the isthmus (Fig. 2, C), to the total proximal cross-sectional area (Fig. 2, D), or to the ratio of the proximal cross-sectional area to
Ratio of total proximal cross-sectional area to cross-sectional area of descending aorta. Filled circlesrepresent those patients operated on at 3 weeks of age or less.
the cross-sectional area of the descending aorta (Fig. 2, E). Weight and age at operation and development of recoarctation. Recoarctation was not related significantly to the dimensions of the distal aortic arch and periductal region. Thus, the four surgical survivors without satisfactory preoperative angiograms were included In the study of the relationship between recoarctation and the weight and age at subclavian flap aortoplasty. No significant relationship was observed
The Journal of
7 4 2 Sanchez et al.
-a= C) ~
~
~
Surgery
A
6.0 5.0
w C) a= 4.0 ~ en 3.0 ~ ~
Thoracic and Cardiovascular
2.0
. o
88
••
SO
• •
•L
C)
w 1.0 ~
en
-sa= ~
70 60 50
B
8 o o
~
w 40 a= 30 ~ en ~ 20 w 10 C) C)
«
0 0
9
•
0 0 00
0 0
80
GOOD
FAIR
POOR
OUTCOME Fig. 3. Relationship between outcome in patients who survived initial repair and (A) weight and (B) age at subclavian flap aortoplasty. Filled circles represent those patients operated on at 3 weeks of age or less.
between recoarctation and weight at operation (Fig. 3, A). On the other hand, a significant relationship was observed between recoarctation and the age at subclavian flap aortoplasty, with recoarctation being observed only in patients who were 21 days of age or less at operation (Fig. 3, B) (p = 0.02). Outcome in patients 3 weeks of age or less at aortoplasty. Because recoarctation did not develop in some patients operated on in the first 3 weeks of age, the group of infants in this age group who had recoarctation was compared to the group who did not (Table 11). No significant difference was found between the two groups with respect to the age and weight at operation or to the dimensions of the distal aortic arch and periductal
region. The incidence of recoarctation was higher in girls operated on in the first 3 weeks of age than in boys operated on at the same age, but this difference was not statistically significant. There were more patients with a patent ductus arteriosus at aortoplasty in the group that did not have recoarctation than in the group that did, but this difference also was not statistically significant. Perioperative infusion of prostaglandin E I had no effect on outcome. Recoarctation was not related to the presence of complex congenital heart disease or to a hemodynamically significant YSD. Recoarctation: Surgical findings. Six patients underwent reoperation because of recoarctation. In one of these, Patient 14, a 20 mm Hg systolic pressure gradient between the right arm and the legs was recognized 10 months after subclavian flap aortoplasty, and it was elected to repair the recurrent coarctation during insertion of a permanent epicardial pacemaker (Table III). At reoperation, the striking finding in all six patients was the obliteration of the lumen by local tissue growth that appeared to originate from the posterior aortic wall. In four of the six patients the tissue growth resembled a shelf. One of these four patients was a patient in whom specific mention had been made during the subclavian flap aortoplasty that no shelf was present. In the other two patients the obstruction was less discrete and was at or near the distal anastomosis. External inspection of the aorta at reoperation in one of these two patients also showed the local aorta to be constricted, apparently because of traction by the remnant of the ductus that had been ligated (but not divided) during aortoplasty. Reoperation in all six patients consisted of polytetrafluoroethylene patch angioplasty, which relieved the obstruction without complications. Discussion Subclavian flap aortoplasty has been recommended as the procedure of choice in infants with coarctation of the aorta because of technical reasons, its potential to allow long-term growth of the periductal aorta, and the excellent results that it has produced.':" Recoarctation has occurred after subclavian flap aortoplasty in a significant number of patients, 1,2,8-10.13 however, and for this reason, some have recommended other methods of repair of coarctation in infancy.":" The reason for recoarctation is not fully understood but has been attributed to tension at the anastomotic site," a short flap," 9. II poor growth of the aorta at the site of repair," inadequate shelf resection," proliferation of residual abnormal tissue," and constriction of residual ductal tissue." Our results show, as those of others have,
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
743
May, 1986
Table II. Comparison of clinical aspects and morphometric results in patients 3 weeks of age or less at subclavian flap aortoplasty No. of patients Age at aortoplasty (days) Weight at aortoplasty (kg) Males/females Patent ductus arteriosus Prostaglandin E 1 Distance from left subclavian artery to coarctation (ern) Length of isthmus (em) Diameter of isthmus (mm') Total proximal cross-sectional area (mm') Ratio of total proximal cross-sectional area to cross-sectional area of descending aorta Complex intracardiac anomaly Ventricular septal defect
No recoarctation
Recoarctation
p Value
7 9.2 ± 5.1 3.2 ± 0.7 5/2 6
8 13.8 ± 6.1 3.2 ± OJ 2/6 4 2 1.0 ± 0.3 0.6 ± 0.2 4.1 ± 1.3 30 ± 19 0.68 ± 0.23
NS NS NS NS NS NS NS NS NS NS
I 2
NS NS
2 1.0 0.8 4.0 28 0.54
± 0.3
± 0.2 ± 0.8 ± 5
± 0.04 2 3
Legend: Values are expressed as mean ± standard deviation. NS. Not significant.
Table m. Descriptive data of patients who underwent recoarctectomy Patient
10
Age at SFA (days) 5
8·
8
II 12 13
II 12 13
14
20
Findings at SFA
Age at recoarctation
Discrete coarctation Isthmus hypoplasia Discrete coarctation Discrete coarctation Isthmus hypoplasia, "no shelf' Discrete coarctation
6 3 5 6 5
Findings at recoarctectomy
wk mo mo mo mo
N arrow distal anastomosis Narrow distal anastomosis with ligament traction Shelf Hypoplastic back wall with shelf-like tissue Shelf
t
Shelf
Legend: SF A, Subclavian flap aortoplasty. 'Same patient as in Table I. tArm/leg pressure gradient of 20 mm Hg 10 months after subclavian flap aortoplasty. Elective patch aortoplasty at time of pacemaker insertion at 11 months of age.
that the subclavian flap procedure provides excellent immediate relief of coarctation of the aorta in very young infants. Thus, the length of the dissected subclavian artery is usually sufficient to bridge over the stenotic area. We have also observed, as others have, that the subclavian flap aortoplasty is complicated by a significant incidence of early recurrence. The results of our morphometric analysis of the distal aortic arch in the group as a whole as well as in the high-risk group indicate that recoarctation after subclavian flap aortoplasty is not dependent on the length of the aortic isthmus or on the distance between the origin of the left subclavian artery and the site of coarctation. We believe itisunlikely, therefore, that recoarctation is the result of early tension at the distal anastomosis. The early relief of coarctation in these patients demonstrates that the subclavian artery and isthmus, when brought together, provide a vascular lumen satisfactory to flow, regardless of the dimensions of the distal
aortic arch and periductal region. This observation agrees with the conclusions drawn from an anatomopathologic study that the subclavian flap procedure was the one best suited to achieve early success in this lesion." The long-term success of the procedure, however, requires satisfactory growth of the new lumen. In an animal model, growth of the aortic area was satisfactory for 6 months after subclavian flap aortoplasty, regardless of whether continuous or interrupted nonabsorbable sutures were used in the repair." Clinical studies in infants, on the other hand, support the view that the use of interrupted sutures or continuous, absorbable sutures may prevent recoarctation." If recoarctation after subclavian flap aortoplasty in our patients were due to poor growth of the new lumen, then recoarctation should manifest itself first in those patients with the narrowest isthmus or the smallest total proximal cross-sectional area (either in absolute terms or in proportion to the diameter of the descending aorta). This pattern was not
7 4 4 Sanchez et al.
observed, nor was a significant relationship observed between recoarctation and the duration of follow-up. Therefore, we do not believe that recoarctation in our patients is due to poor growth of the anastomosis. The lack of correlation between the weight of the infant at subclavian flap aortoplasty and the development of recoarctation argues against technical factors being primarily responsible for recoarctation. The finding of a significant relationship between age at operation and development of recoarctation was unexpected, but is in agreement with the recently published observations in a larger group of patients." That this relationship had not been generally appreciated may have been due to the relatively small number of infants who have undergone the subclavian flap procedure at less than 1 month of age. Review of the literature, however, tends to support this observation. In a series of nine infants less than 6 months of age at subclavian flap repair (average age at repair 4.4 weeks), reported by Kamau and associates,'? the two infants who subsequently had "significant" (40 mm Hg) arm/leg gradients were 3 days and 9 days of age at repair. Pierce and associates' reported on 12 infants, the two youngest of whom were operated on at 2 weeks of age. One had a 10 mm Hg arm/leg gradient 4 months after the operation. The other was followed for only 1 month. In a series presented by Thibault, Sperling, and Gazzaniga," three patients were 14 days of age or younger at operation and two of these survived the procedure. One of these two survivors later had a 30 mm Hg arm/leg gradient (proximal arch hypoplasia was not ruled out) and the other had a 12 mm Hg gradient. None of the other survivors, operated on at ages over 2 weeks of age, had significant gradients. In Penkoske's series," the two infants who required reoperation were 17 and 22 days of age at subclavian aortoplasty. Other, larger series do not lend themselves to evaluation because of the age breakdown that is provided,' absence of information about the age at operation of those infants who had recoarctation,':" I I or absence of information concerning blood pressure measurements- II or arm/leg pressure gradients on follow-up." 5, I I If recoarctation is not due to poor growth of the anastomosis, the development of obstruction after repair must be due to another process that eventually compromises the lumen. In patients with coarctation of the aorta, ductus tissue extends in circumferential fashion into the aorta to a greater degree than normal." In addition, encirclement of the aortic isthmus by ductal tissue has been observed in young infants with coarctation.'? Furthermore, in young infants the coarctation shelf is usually preductal in position, whereas in older
The Journal of Thoracic and Cardiovascular Surgery
patients it is most often postductal," suggesting that all coarctations develop preductally with gradual distal migration of the shelf as the child grows. 18 Since subclavian flap aortoplasty does not involve resection of the coarctation segment, repair by this method, even when any shelf that is present is excised, is likely to leave behind ductal tissue. The latter in the young infant may not have completely involuted and would therefore have the potential to cause local distortion of the aorta. The presence of residual ductal tissue may explain the relationship that we and others" have observed between the age at repair and the development of recoarctation. Local distortion, scarring, and intimal proliferation, complicated by growth and distal migration of residual/ recurrent shelf tissue, would lead to obliteration of the lumen and would explain the development of recoarctation in these patients, as well as the unexpected proximity between the restenosing tissue and the distal end of the flap observed at reoperation in two of our patients. Our series clearly shows that recoarctation is not necessarily the rule in infants operated on in the first 3 weeks of life. This is also shown by the results reported in infants less than 1 month" and less than 1 week of age' at subclavian flap aortoplasty. Thus, it is possible that the problems presented by the abnormal local tissue can be surmounted by meticulous attention to excision of local coarctation tissue in the very young infant. However, such technique might require extensive, "deep" shelf resection and be complicated by late aneurysm formation. It is clearly important that the length of the divided subclavian artery be sufficient to satisfactorily bridge over the area of coarctation. In our patients the degree of dissection of the subclavian artery was similar to that described by others.v" The observation that in our patients subclavian flap aortoplasty relieved the obstruction indicates that the divided subclavian artery was long enough at the initial repair. It is possible, however, that in very young infants the growth of the distal aortic arch and periductal region is such that obstruction may redevelop if the flap is not sufficiently long to allow for local scarring and retraction. One way to assess this would have been to determine the length of the "surgically available" subclavian artery (i.e., distance between the origin of the subclavian and the origin of the vertebral and/or internal mammary arteries) and to correlate this length with the distance between the subclavian artery and the site of coarctation. Unfortunately, the angiograms in our patients did not permit measurement of the length of the subclavian artery. Nevertheless, the need to achieve maximum length of
Volume 91 Number 5
Subclavian flap repair of aortic coarctation
745
May, 1986
the subclavian artery before it is divided cannot be overemphasized. The observationof local traction by the ligamentum arteriosum in one patient with recoarctation (Patient 8, Table III) is noteworthy but not sufficient to conclude that division of the ductus or ligamentum arteriosum at subclavian aortoplasty is beneficial. It is difficult to conclude from our findings that subclavian flap aortoplasty should be delayed until at least 3 weeks of age, particularly because infants with associated intracardiac defects require emergency coarctectomy' and because no deaths occurred in conjunction with recoarctation. The results suggest, however, that other methods of repair should be considered in infants younger than 3 weeksof age. Resection with end-to-end anastomosis has been associated with a higher incidence of recoarctation.>" even when extensive resection has been performed." However, better results have been observed when extensive resection has been combined with improved suture material. 13, 23, 24 Prosthetic patch aortoplasty has also relieved the coarctation in infants. 14, 25, 26 It may be that the ability to enlarge the circumference of the coarctation area with a patch considerably larger in width than the opened subclavian artery may prevent critical compromise of the new lumen by local tissue scarring and growth. Our results in infants older than 3 weeks of age at aortoplasty support the conclusions of others that subclavianflap aortoplasty produces excellentresults despite anatomic variations of the aortic arch. Long-term evaluation of the patients at rest and with physical exercise remains essential to completely evaluate the outcome of subclavian patch repair. We gratefully acknowledge the assistance of Lydia Garcia inpreparation of this manuscript.
REFERENCES I Midgley FM, Scott LP, Perry LW, Shapiro SR, McClenathan JE: Subclavian flap aortoplasty for treatment of coarctation in early infancy. J Pediatr Surg 13:265-268, 1978 2 Hamilton 01, Di Eusanio G, Sandrasagra FA, Donnelly RJ: Early and late results of aortoplasty with a left subclavian flap for coarctation of the aorta in infancy. J THORAC CARDIOVASC SURG 75:699-704, 1978 3 Pierce WS, Waldhausen JA, Berman W Jr, Whitman V: Late results of the subclavian patch procedure in infants with coarctation of the thoracic aorta. Circulation 58:Suppl 1:78-82, 1978 4 Waldhausen JA, Whitman V, Werner JC, Pierce WS: Surgical intervention in infants with coarctation of the aorta. J THoRAc CARDIOVASC SURG 81:323-325, 1981
5 Bergdahl LAL, Blackstone EH, Kirklin JW, Pacifico AD, Bargeron LM: Determinants of early success in repair of aortic coarctation in infants. J THORAC CARDIOVASC SURG 83:736-742,1982 6 Moulton AL, Brenner JI, Roberts G, Tavares S, Ali S, Nordenberg A, Burns JE, Ringel R, Berman MA: Subclavian flap repair of coarctation of the aorta in neonates. Realization of growth potential? J THORAC CARDIOVASC SURG 87:220-235, 1984 7 Penkoske PA, Williams WG, Olley PM, LeBlanc J, Trusler GA, Moes CAF, Judakin R, Rowe RD: Subclavian arterioplasty repair of coarctation of the aorta in the first year of life. J THoRAc CARDIOVASC SURG 87:894-900, 1984 8 Campbell DB, Waldhausen JA, Pierce WS, Fripp R, Whitman V: Should elective repair of coarctation of the aorta be done in infancy? J THORAC CARDIOVASC SURG 88:929-938, 1984 9 Thibault WN, Sperling DR, Gazzaniga AB: Subclavian artery patch angioplasty. Treatment of infants and young children with aorta coarctation. Arch Surg 110: 10951098, 1975 10 Kamau P, Miles V, Toews W, Kelminson L, Friesen R, Lockhart C, Butterfield J, Hernandez J, Hawes CR, Pappas G: Surgical repair of coarctation of the aorta in infants less than six months of age. Including the question of pulmonary artery banding. J THORAC CARDIOV ASC SURG 81:171-179, 1981 11 Nair UR, Jones 0, Walker DR: Surgical management of severe coarctation of the aorta in the first month of life. Review of 48 consecutive cases. J THoRAc CARDIOVASC SURG 86:587-590, 1983 12 Allen RG, Marin-Garcia J, Nayek G: Methods of management and results following surgery for coarctation of the aorta in infancy. J Pediatr Surg 15:953-960, 1980 13 Cobanoglu A, Teply JF, Grunkemeier GL, Sunderland CO, Starr A: Coarctation of the aorta in patients younger than three months. A critique of the subclavian flap operation. J THoRAc CARDIOVASC SURG 89:128-135, 1985 14 Sade RM, Taylor AB, Chariker EP: Aortoplasty compared with resection for coarctation of the aorta in young children. Ann Thorac Surg 28:346-353, 1979 15 Metzdorff MT, Cobanoglu A, Grunkemeier GL, Sunderland CO, Starr A: Influence of age at operation on late results with subclavian flap aortoplasty. J THORAC CARDIOVASC SURG 89:235-241, 1985 16 Pellegrino A, Deverall PB, Anderson RH, Smith A, Wilkinson JL, Russo P, Girod DA, Tynan M: Aortic coarctation in the first three months of life. An anatomopathological study with respect to treatment. J THoRAc CARDIOVASC SURG 89:121-127,1985 17 Pae WE Jr, Myers JL, Waldhausen JA, Prophet GA, Pierce WS: Subclavian flap angioplasty. Experimental study in growing piglets. J Thor Cardiovasc Surg 82:922927, 1981 18 Elzenga NJ, Gittenberger-de Groot AC: Localised coarc-
7 4 6 Sanchez et al.
19
20
21
22
tation of the aorta. An age dependent spectrum. Br Heart J 49:317-323, 1983 Ho SY, Anderson RH: Coarctation of the aorta, Paediatric Cardiology, vol 2, Heart Disease in the Newborn, MS Godman, RM Marquis, OOs., Edinburgh, 1979 Churchill Livingstone, pp 173-186 Khoury GH, Hawes CR: Recurrent coarctation of the aorta in infancy and childhood. J Pediatr 72:801-806, 1968 Hartmann AF Jr, Goldring D, Hernandez A, Behrer MR, Schad N, Ferguson T, Burford T: Recurrent coarctation of the aorta after successful repair in infancy. Am J Cardiol 25:405-415, 1970 Williams WG, Shindo G, Trusler GA, Dische MR, Olley PM: Results of repair of coarctation of the aorta during infancy. J THORAC CARDIOVASC SURG 79:603-608, 1980
The Journal of Thoracic and Cardiovascular Surgery
23 Harlan JL, Doty DB, Brandt B III, Ehrenhaft JL: Coarctation of the aorta in infants. J THoRAc CARDIOVASC SURG 88:1012-1019,1984 24 Korfer R, Meyer H, Kleikamp G, Bircks W: Early and late results after resection and end-to-end anastomosis of coarctation 'of the thoracic aorta in early infancy. J THoRAc CARDIOVASC SURG 89:616-622, 1985 25 Fleming WH, Sarafian LB, Clark EB, Dooley KJ, Hofschire PJ, Hopeman AR, Ruckman RN, Mooring PK: Critical aortic coarctation. Patch aortoplasty in infants less than age 3 months. Am J Cardiol 44:687-690, 1979 26 Sade RM, Crawford FA, Hohn AR, Riopel DA, Taylor AB: Growth of the aorta after prosthetic patch aortoplasty for coarctation in infants. Ann Thorac Surg 38:21-25, 1984