- Email: [email protected]
The Blalock-Taussig Shunt in Infants: Standard versus Modified
The Blalock-Taussi Shunt in Infants: Standar versus Modified
%
Rebecca L. Ullom, M.S., Robert M. Sade, M.D., Fred A. Crawford, Jr., M.D., Bertrand A. Ross, M.D., and Frank Spinale, M.S. ABSTRACT In recent years, the modified Blalock-Taussig shunt-a polytetrafluoroethylene graft from the subclavian artery to the pulmonary artery-has been preferred over the standard shunt by some surgeons because (1)it requires less dissection and (2)length of native vessels is not critical. From January, 1979,to June, 1985, we operated on 51 infants less than 1 year of age, including 26 less than 1 week of age, to palliate severe complex cyanotic congenital cardiac malformations. Twenty-four modified Blalock-Taussig shunts and 29 standard Blalock-Taussig shunts were created. The groups were concurrent. We reviewed all available cineangiogramsand measured branch pulmonary and subclavian arteries. Pulmonary artery index was not different preoperatively in patients given a modified versus a standard BlalockTaussig shunt (144 f 118 and 118 f 59 mm2/m2,respectively), but it was greater postoperatively in patients with a modified shunt (431 f 188 and 189 f 106 mmz/m2)( p = 0.07). Distortion of the pulmonary artery occurred less often after a modified Blalock-Taussig shunt (all) than a standard Blalock-Taussig shunt (6/8) ( p = 0.06), though none of the distortions was severe. Early and late shunt failure occurred less often with a modified shunt (5/24) than with a standard shunt (15/29)( p < 0.05). The modified Blalock-Taussig shunt had advantages over the standard Blalock-Taussigshunt in our series: pulmonary artery growth was greater, distortion of pulmonary arteries was less commonly seen, and shunt failure occurred less often. Thus, in infants, we believe the modified Blalock-Taussig shunt should be considered a reasonable alternative to the standard Blalock-Taussig shunt.
has become associated with higher patency and survival than were previously possible [3]. In addition, closure during future corrective procedures may be simpler with the Blalock-Taussig shunt than with central systemicpulmonary shunts [5]. Despite these advantages, the Blalock-Taussig procedure has several disadvantages: a lengthy dissection time, differential pulmonary artery perfusion [6], a reduction in upper extremity growth on the side of the anastomosis [7], a relatively high risk of phrenic nerve injury [8], the possibility of upper extremity gangrene [9], and pulmonary artery distortion. Also, some infants do not have a suitable anatomy for a Blalock-Taussig shunt because of an unusually narrow or short subclavian artery. With the introduction of polytetrafluoroethylene (PTFE) as a biomaterial, the modified Blalock-Taussig shunt, in which a PTFE interposition graft is placed between the subclavian artery and the pulmonary artery, has become increasingly popular as an alternative to the standard Blalock-Taussigshunt [101. The modified shunt has the advantage of requiring less dissection than the standard shunt, while retaining accurate limitation of pulmonary blood flow by the diameter of the subclavian artery [ll]. The modified Blalock-Taussig shunt has been used in some patients in whom the standard shunt is an anatomically undesirable choice [12]. The purpose of this study was to evaluate and compare the early and late results with modified and standard Blalock-Taussig shunts in similar groups of infants. Changes in arterial oxygen saturation (Sa02), shunt patency and longevity, mortality, incidence of diaphragmatic paralysis, growth of pulmonary arteries, and vascular distortion were evaluated and compared.
Since its introduction in 1945, the Blalock-Taussig procedure [l] has become the palliative procedure of choice at many institutions for infants with cyanotic congenital heart malformations [2,3]. It has a lower associated incidence of congestive heart failure and other postoperative complications than central systemic-pulmonary shunts [4]. With the advent of better microvascular surgical techniques and improvements in intraoperative and postoperative patient management, this procedure
Material and Methods
From the Divisions of Cardiothoracic Surgery and Pediatric Cardiology, Medical University of South Carolina, Charleston, SC. Accepted for publication June 30, 1987. Address correspondence to Dr. Sade, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 29425.
539
Ann Thorac Surg 44:539-543,Nov 1987
Between July 1, 1979, and June 30, 1985, 51 infants less than 1 year of age underwent a Blalock-Taussig shunt (2 patients had 2 shunts, a total of 53 shunts) for palliation of cyanotic congenital heart disease at the Medical University of South Carolina. Twenty-four modified Blalock-Taussigshunts were created (11of these patients were less than 1 week old) and 29 standard BlalockTaussig shunts (15 patients were less than 1week old). One additional child received both a modified and a standard shunt at the same operation; however, followup information could not be distinguished between the two types of procedures, and the child was excluded from this study except when stated otherwise. The patients were not prospectively randomized. The decision on which type of shunt to construct was made by the
540 The Annals of Thoracic Surgery Vol 44 No 5 November 1987
surgeon, but each of the two main implanting surgeons chose each type about equally. The two groups were concurrent: time from operation to June 30, 1985, was 36.4 15.6 months (mean f standard deviation) for the group with a modified shunt and 36.1 f 19.4 months for the group with a standard shunt (not significant [NS]).
*
Preoperative and Postoperative Studies For each patient, preoperative and postoperative Sa02 values were recorded, and congestive heart failure, defined as need of treatment with digoxin, was noted. Shunt failure was defined as shunt thrombosis or occlusion, or late rise in hemoglobin levels greater than 18 @lo0 ml, which was taken to indicate shunt inadequacy. Hospital failures were those occurring before hospital discharge, and late failures were those occurring after hospital discharge. Postoperative diaphragmatic paralysis was noted. When available, preoperative cineangiograms were reviewed (18 patients with modified Blalock-Taussig shunts and 19 with standard Blalock-Taussig shunts), and measurements were recorded for the diameter of each pulmonary artery near its origin, each subclavian artery just proximal to its first branch, and the cardiac catheter. All measurements were taken during ventricular systole and, when possible, from the same angiographic frame. In addition, the body surface area at the time of cardiac catheterization and all catheter sizes were recorded. Magnification factors were determined by comparing the measured diameter of the cardiac catheter on the cineangiogram with its known diameter, and from this, actual values of subclavian and pulmonary artery diameters were determined. The pulmonary artery index (PAI) as described by Nakata and co-workers [13] was calculated; this index is the ratio of the total cross-sectional area of the branch pulmonary arteries to the body surface area. When postoperative cineangiograms were available (11 patients with modified Blalocl-Taussig shunts and 8 with standard Blalock-Taussig shunts), they also were reviewed, the same dimensions were recorded, and the same index was derived. Finally, any distortion of the subclavian artery or the pulmonary artery on follow-up catheterization films was noted and graded (mild, moderate, severe). Operative Technique All shunts were constructed through a standard lateral thoracotomy. The subclavian artery and pulmonary artery were identified. When it was elected to perform a modified Blalock-Taussig shunt, a segment of Gore-Tex vascular graft 5 to 6 mm in diameter was selected according to the size of the child and the diameter of the subclavian artery, and was anastornosed to a longitudinal incision in the side of the subclavian artery. The graft was then cut to a length that allowed it to lie without tension at the pulmonary artery, and it was sewn to a longitudinal incision in that artery. When a standard Blalock-Taussig shunt was chosen, the subclavian artery
was dissected free from the surrounding tissue, divided proximal to its first branch, ligated distally, and anastomosed end-to-end to the pulmonary artery. To help ensure shunt patency, systemic blood pressure was maintained at 80 mm Hg with Neo-Synephrine (phenylephrine hydrochloride), if needed, and the patient was heparinized (100 units per kilogram of body weight) intravenously every 4 hours for 12 to 24 hours after operation. Statistical Analysis Results were analyzed using Pearson’s chi-square test, Fischer’s exact test, or the Wilcoxon test, where appropriate. Statistical significance was established at alpha equal to 0.05.
Results
Preoperative Status The mean age and weight of patients given a modified Blalock-Taussig shunt were 2.4 2 3.4 months and 4.2 & 1.8 kg, respectively. Far patients given a standard Blalock-Taussig shunt, the mean age was 1.4 f 2.5 months and the mean weight, 3.5 & 1.2 kg. The anomalies in these patients were as follows: pulmonary stenosis or atresia and intact ventricular septum, 2 patients in the modified shunt group and 5 in the standard shunt group; pulmonary stenosis or atresia, intact ventricular septum, and transposition of the great arteries, 3 in the standard shunt group; pulmonary stenosis or atresia and ventricular septa1 defect (includes double-outlet right ventricle, transposition of the great arteries, atrioventricular canal, and single ventricle), 11 in the modified and 16 in the standard shunt group; and tricuspid atresia, 11 in the modified and 5 in the standard shunt group. With the exception of tricuspid atresia, which was statistically more prevalent in the group with a modified Blalock-Taussig shunt (p < 0.05), there were no significant differences between the two groups preoperatively. Arterial Oxygen Saturation and Congestive Heart Failure Average preoperative Sa02 values were statistically the same for the two groups (55 f 21% for the modified shunt group and 53 2 29% for the standard shunt group). However, average postoperative values were significantly different (p = 0.01): 84 ? 9% for the modified shunt group and 74 f 19% for the standard shunt group. Although these average postoperative values were different, the patient-specific changes in Sa02 from before to after operation were not different between the groups (p > 0.05): Sa02 = +32 5 21% for the modified shunt group and + 20 & 35% for the standard shunt group. The apparent discrepancy is due to the effect of patients in whom the Sa02decreased rather than increased after the shunt, primarily patients with early shunt failure. No instances of congestive heart failure were identified.
541 Ullom et al: Standard versus Modified Blalock-Taussig Shunt in Infants
a standard shunt was included in this comparison; it is of interest to note that the diaphragmatic paralysis occurred on the side with the standard shunt.
Shunt Failure and Mortality with the Blalock-Taussig ShunP Variable Total deaths
Early Late Total failures Early Late
Modified Shunt (N = 24)
9 (38)
4 5 (21) 5 (21) 1 (4Ib 4 (17)
Standard Shunt
(N = 29) 10 (35) 8
(Wb
2 (7) 15 (52) 4 (14)b 11 (38)
p Value
Cineangiography
NS NS
ACTUAL PULMONARY A N D SUBCLAVIAN ARTERY SIZE.
NS 0.05 0.08 0.03
"Numbersin parentheses are percentages. patient who died had both a modified and a standard shunt. Each shunt failed in the hospital, and the patient is counted as a death and a failure in both shunt categories. NS
=
not significant.
Deaths and Shunt Failures Nineteen patients died (9 with a modified and 10 with a standard shunt) (Table). The cause of death was unexplained in 4 patients (2 modified and 2 standard shunts). It was due to sepsis in 5 patients (2 modified and 3 standard shunts), intraoperative myocardial ischemia in 1patient (modified shunt), associated malformations in 3 (1modified and 2 standard shunts), reoperation in 4 (3 modified shunts and 1 standard shunt), and shunt thrombosis in 2 (standard shunts). Death was related to early shunt failure (all 4 of these patients died, 3 early and 1 late), but not to late failure (only 2 of 15 died). Shunt failure occurred in 20 patients, significantly more often in the group with a standard Blalock-Taussig shunt ( p < 0.05) (see Table). Actuarial analysis also demonstrated a lower risk of failure over time with this type of shunt (p < 0.05). The 1early failure of a modified Blalock-Taussig shunt occurred in the patient with a contralateral standard Blalock-Taussig shunt, which also failed in the hospital. The patient died of postoperative sepsis. The remaining three instances of early failure (all standard shunts) involved shunt thrombosis associated with early death. The 4 patients with late failure of a modified BlalockTaussig shunt were reoperated on for lesion repair (3 patients) or creation of a second shunt (1 patient). Three patients are living, and 1 patient died of postoperative sepsis following the corrective procedure. Among the patients with late failure of a standard Blalock-Taussig shunt, 1 died at home with a hemoglobin count of 20.8 @lo0 ml and 10 are alive after a second operation for lesion repair (7 patients) or construction of a second shunt (3 patients). Diaphragmatic Paralysis Postoperative diaphragmatic paralysis requiring subsequent operation for plication developed in 4 patients: 1 (4%)of 25 modified Blalock-Taussig shunts and 3 (10%) of 30 standard Blalock-Taussig shunts (NS). The child who underwent concurrent placement of a modified and
Pd-
monary artery diameters on the side of the shunt ranged from 2.7 to 7.4 mm preoperatively. Four (22%) of 18 patients with a modified shunt and 7 (37%)of 19 with a standard shunt for whom measurements were available had a shuntside pulmonary artery diameter of 3.0 mm or less (NS). Among these patients with small vessels, none of the modified shunts failed but 4 of the standard shunts failed, 2 in the hospital and 2 late (0 of 4 modified shunts versus 4 of 7 standard shunts [NS]). Comparing all patients with small vessels with patients with larger vessels, 2 early failures (18%) occurred in 11 patients with a shuntside pulmonary artery diameter of 3.0 mm or less, and 2 early failures (8%) also occurred in 26 patients with a pulmonary artery diameter greater than 3.0 mm (NS). Subclavian artery diameter on the side of the shunt ranged from 1.6 to 4.9 mm preoperatively. A total of 10 patients (5 with each type of shunt) had a shuntside subclavian artery measuring less than 2.5 mm. Three of these shunts failed (0 modified shunts versus 3 standard shunt [NS]). RELATIVE PULMONARY ARTERY SIZE. Eighteen patients (7 with a modified and 11 with a standard shunt) had a shuntside PA1 of 50 mm2/m2or less (normal PA1 for one side = 165 mm2/m2and for both sides, 330 mm2/m2[13]). Among the 7 patients with a modified shunt, 1 early failure and 2 late failures occurred, and among the 11 with a standard shunt, 1 early failure and 5 late failures occurred (NS). Comparing all patients having small vessels with those having larger vessels, 2 early failures (11%)occurred in 18 patients with a shuntside PA1 of 50 mm2/m2or less, and 1 failure (6%) occurred in 17 patients with a shuntside PA1 of more than 50 mm2/m2 (NS). COMPARATIVE PULMONARY ARTERY (PAI) GROWTH. Growth in pulmonary arteries following palliative procedures was evaluated by comparing the preoperative total PA1 with the postoperative total PAI. The average preoperative total PA1 values of the modified and the standard Blalock-Taussig shunts were not statistically different, but the average postoperative total PAI was greater with the modified than the standard shunt (p = 0.07). For the modified shunt group, the average PAI increased from 144 118 mmz/m2 (N = 16) preoperatively to 431 188 mm2/m2(N = 6) ( p = 0.03) at an average follow-up of 29 months. For the standard shunt group, the average PA1 increased from 118 59 mm2/m2 (N = 18) preoperatively to 189 f 106 mm2/m2(N = 6) (p = 0.05) at an average follow-up of 18 months. Twelve patients (6 from each group) had both preoperative and postoperative angiography, thereby allowing comparison of paired data. From preoperative to postoperative measurements, the total PA1 increased by
*
*
*
542 The Annals of Thoracic Surgery Vol 44 No 5 November 1937
296 2 273 mm2/m2at an average time of 29 months in the modified shunt group and by 129 a 117 mm2/m2at an average time of 20 months in the standard shunt group. However, this difference was not significant because of the large variability. PULMONARY AND SUBCLAVIAN ARTERY DISTORTION. Distortion of the pulmonary artery and subclavian artery on follow-up catheterization was also noted. Of the 11 patients with a modified Blalock-Taussig shunt who were reviewed, only 4 (36%)had distorted pulmonary arteries (all mild) on the shunt side and of the 8 with a standard shunt evaluated, 6 (75%) had distortion of the pulmonary artery on the shunt side, 4 mild and 2 moderate ( p = 0.06). Six (55%) of the 11 patients with a modified shunt had a distorted subclavian artery on the shunt side on follow-up catheterization films, 4 had severe and 2 had mild distortion. The artery was patent in every instance, but some decrease in distal flow was appreciated in the 4 with severe subclavian artery distortion. Because of the nature of the operation, all patients with a standard shunt had interrupted subclavian arteries with no direct flow to the distal part of the artery. NEWBORNS LESS THAN 1 WEK OLD. Newborns less than 1 week of age exhibited the same trends as the older infants, but because of the small number of patients, differences did not reach statistical significance.
Comment
Arterial Oxygen Saturation Group differences in postoperative SaOzvalues were not significant when considering only those patients whose shunt did not fail in the hospital. This is expected, since flows in both modified and standard shunts are limited by the diameter of the subclavian artery, and improvement in SaOZis a direct function of added pulmonary blood flow resulting from the constructed shunt. Deaths and Shunt Failures Overall mortality was not different between the two groups. Shunt failure, however, occurred significantly less often after creation of a modified shunt than a standard shunt. These results concur with those reported by Moulton and co-workers [14], although their study was not restricted to infants. In the current study, early failures always resulted in death, regardless of which type of shunt had been constructed. Late failures, however, could usually be treated with further operation and were seldom fatal. Both early and late failures occurred less frequently with a modified shunt than with a standard shunt. Diaphragmatic Paralysis Although the incidence of diaphragmatic paralysis in this study was low and statistically significant differences between the two groups were not found, it is of interest to note that diaphragmatic paralysis occurred less frequently in the modified shunt group than in the
standard shunt group. This might be expected, since more dissection is required for a standard shunt than for a modified shunt.
Cineangiography The modified Blalock-Taussig shunt procedure is associated with more pulmonary artery growth than the standard Blalock-Taussig shunt procedure, as suggested by the larger increase in PA1 at similar follow-up times, although both procedures resulted in a significant PA1 increase. This difference is important because adequate pulmonary artery size is a critical determinant of survival and late results after definitive repair of cyanotic heart malformations [13]. Although the reason for this difference is uncertain, it may be related to higher flow over longer periods through the modified shunt, as evidenced by a higher average Sa02postoperatively and a lower rate of shunt failure both early and late. Preservation of blood flow in the subclavian artery distal to the shunt was excellent in most patients with a modified Blalock-Taussig shunt, but was compromised by severe distortion of the subclavian artery in 4 (36%)of the 11 patients studied. The creation of a modified Blalock-Taussig shunt appears to preserve pulmonary artery anatomy well, little or no distortion being found after that procedure, compared with a high rate of distortion after a standard Blalock-Taussig shunt procedure. It is also of interest that we observed no instance of congestive heart failure despite the use of 5-mm and 6mm grafts in these infants. This problem may have been alleviated by our placement of the proximal anastomosis well onto the subclavian artery, thereby avoiding inclusion of any part of the aortic arch or innominate artery. Overall, the modified Blalock-Taussig shunt had advantages over the standard Blalock-Taussig shunt in our series of infants: shunt failure rates were lower, pulmonary artery growth was greater, and distortion of the pulmonary arteries occurred less often. With a modified shunt, the rise in Sa02 was greater and the incidence of diaphragmatic paralysis was lower than with a standard shunt, although these differences were not statistically significant. A modified Blalock-Taussig shunt should be considered a reasonable choice in infants requiring a palliative procedure for cyanotic congenital heart disease.
References 1. Blalock A, Taussig HB: The surgical treatment of malformations of the heart in which there is pulmonary stenosis or pulmonary atresia. JAMA 128;189, 1945 2. Jennings RB Jr, Innes BJ, Brickman R D Use of microporous expanded polytetrafluoroethylene grafts for aortapulmonary shunts in infants with complex cyanotic heart disease. J Thorac Cardiovasc Surg 76489, 1978 3. Neches WH, Naifeh JG, Park SC, et a1 Systemic-pulmonary artery anastomoses in infancy. J Thorac Cardiovasc Surg 70:921, 1975 4. Arciniegas E, Farooki ZQ, Hakimi M, Green EW: Results of
543 Ullom et ai: Standard versus Modified Blalock-Taussig Shunt in Infants
5. 6. 7. 8. 9.
two-stage surgical treatment of tetralogy of Fallot. J Thorac Cardiovasc Surg 79:876, 1980 Laks H, Fagan L, Bamer HB, Willman VL: The BlalockTaussig shunt in the neonate. Ann Thorac Surg 25:220,1978 Fort L 111, Morrow AG, Pierce GE, et al: The distribution of pulmonary blood flow after subclavian-pulmonary anastomosis. J Thorac Cardiovasc Surg 50:671, 1965 Currarino G, Engle MA: The effects of ligation of the subclavian artery on the bones and soft tissues of the arms. J Pediatr 67:808, 1965 Mickell JJ, Oh KS, Siewers RD, et al: Clinical implications of postoperative unilateral phrenic nerve paralysis. J Thorac Cardiovasc Surg 76:297, 1978 Geiss D, Williams WG, Lindsay WK, Rowe RD: Upper extremity gangrene: a complication of subclavian artery division. Ann Thorac Surg 30487, 1980
10. Karpawich PP, Bush CP, Antillon JR, et al: Modified Blalock-Taussig shunt in infants and young children. J Thorac Cardiovasc Surg 89:275, 1985 11. de Leva1 MR, McKay R, Jones M, et al: Modified BlalockTaussig shunt. J Thorac Cardiovasc Surg 81:112, 1981 12. Ilbawi MN, Grieco J, DeLeon SY, et al: Modified BlalockTaussig shunt in newborn infants. J Thorac Cardiovasc Surg 88:770, 1984 13. Nakata S, Imai Y, Takanashi Y, et al: A new method for the quantitative standardization of cross-sectional area of the pulmonary arteries in congenital heart diseases with decreased pulmonary blood flow. J Thorac Cardiovasc Surg 88:610, 1984 14. Moulton AL, Brenner JI, Ringel R, et al: Classic versus modified Blalock-Taussig shunts in neonates and infants. Circulation 72:Suppl 2:35, 1985
Notice from the American Board of Thoracic Surgery The American Board of Thoracic Surgery began its recertification process in 1984. Diplomates interested in participating in this examination should maintain a documented list of the cardiothoracic operations they performed during the year prior to application for recertification. They should also keep a record of their attendance at thoracic surgical meetings, and other continuing medical education activities pertaining to thoracic surgery and thoracic disease, for the two years prior to application. A minimum of 100 hours of approved CME activity is required. In place of a cognitive examination, candidates for recertification will be required to complete both the general thoracic and cardiac portions of the SESATS I11 syllabus (Self-EducatiodSelf-Assessmentin Thoracic Surgery). It is not necessary for candidates to purchase
SESATS I11 booklets prior to applying for recertification. SESATS I11 booklets will be forwarded to candidates after their applications have been accepted. Diplomates whose 10-year certificates will expire in 1990 may begin the recertification process in 1988. This new certificate will be dated 10 years from the time of expiration of the original certificate. Recertification is also open to any Diplomate with an unlimited certificate and will in no way affect the validity of the orignal certificate. The deadline for submission of applications is July 1, 1988. A recertification brochure outlining the rules and requirements for recertification in thoracic surgery is available upon request from the American Board of Thoracic Surgery, One American Plaza, Suite 803, Evanston, IL 60201.
%
Rebecca L. Ullom, M.S., Robert M. Sade, M.D., Fred A. Crawford, Jr., M.D., Bertrand A. Ross, M.D., and Frank Spinale, M.S. ABSTRACT In recent years, the modified Blalock-Taussig shunt-a polytetrafluoroethylene graft from the subclavian artery to the pulmonary artery-has been preferred over the standard shunt by some surgeons because (1)it requires less dissection and (2)length of native vessels is not critical. From January, 1979,to June, 1985, we operated on 51 infants less than 1 year of age, including 26 less than 1 week of age, to palliate severe complex cyanotic congenital cardiac malformations. Twenty-four modified Blalock-Taussig shunts and 29 standard Blalock-Taussig shunts were created. The groups were concurrent. We reviewed all available cineangiogramsand measured branch pulmonary and subclavian arteries. Pulmonary artery index was not different preoperatively in patients given a modified versus a standard BlalockTaussig shunt (144 f 118 and 118 f 59 mm2/m2,respectively), but it was greater postoperatively in patients with a modified shunt (431 f 188 and 189 f 106 mmz/m2)( p = 0.07). Distortion of the pulmonary artery occurred less often after a modified Blalock-Taussig shunt (all) than a standard Blalock-Taussig shunt (6/8) ( p = 0.06), though none of the distortions was severe. Early and late shunt failure occurred less often with a modified shunt (5/24) than with a standard shunt (15/29)( p < 0.05). The modified Blalock-Taussig shunt had advantages over the standard Blalock-Taussigshunt in our series: pulmonary artery growth was greater, distortion of pulmonary arteries was less commonly seen, and shunt failure occurred less often. Thus, in infants, we believe the modified Blalock-Taussig shunt should be considered a reasonable alternative to the standard Blalock-Taussig shunt.
has become associated with higher patency and survival than were previously possible [3]. In addition, closure during future corrective procedures may be simpler with the Blalock-Taussig shunt than with central systemicpulmonary shunts [5]. Despite these advantages, the Blalock-Taussig procedure has several disadvantages: a lengthy dissection time, differential pulmonary artery perfusion [6], a reduction in upper extremity growth on the side of the anastomosis [7], a relatively high risk of phrenic nerve injury [8], the possibility of upper extremity gangrene [9], and pulmonary artery distortion. Also, some infants do not have a suitable anatomy for a Blalock-Taussig shunt because of an unusually narrow or short subclavian artery. With the introduction of polytetrafluoroethylene (PTFE) as a biomaterial, the modified Blalock-Taussig shunt, in which a PTFE interposition graft is placed between the subclavian artery and the pulmonary artery, has become increasingly popular as an alternative to the standard Blalock-Taussigshunt [101. The modified shunt has the advantage of requiring less dissection than the standard shunt, while retaining accurate limitation of pulmonary blood flow by the diameter of the subclavian artery [ll]. The modified Blalock-Taussig shunt has been used in some patients in whom the standard shunt is an anatomically undesirable choice [12]. The purpose of this study was to evaluate and compare the early and late results with modified and standard Blalock-Taussig shunts in similar groups of infants. Changes in arterial oxygen saturation (Sa02), shunt patency and longevity, mortality, incidence of diaphragmatic paralysis, growth of pulmonary arteries, and vascular distortion were evaluated and compared.
Since its introduction in 1945, the Blalock-Taussig procedure [l] has become the palliative procedure of choice at many institutions for infants with cyanotic congenital heart malformations [2,3]. It has a lower associated incidence of congestive heart failure and other postoperative complications than central systemic-pulmonary shunts [4]. With the advent of better microvascular surgical techniques and improvements in intraoperative and postoperative patient management, this procedure
Material and Methods
From the Divisions of Cardiothoracic Surgery and Pediatric Cardiology, Medical University of South Carolina, Charleston, SC. Accepted for publication June 30, 1987. Address correspondence to Dr. Sade, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 29425.
539
Ann Thorac Surg 44:539-543,Nov 1987
Between July 1, 1979, and June 30, 1985, 51 infants less than 1 year of age underwent a Blalock-Taussig shunt (2 patients had 2 shunts, a total of 53 shunts) for palliation of cyanotic congenital heart disease at the Medical University of South Carolina. Twenty-four modified Blalock-Taussigshunts were created (11of these patients were less than 1 week old) and 29 standard BlalockTaussig shunts (15 patients were less than 1week old). One additional child received both a modified and a standard shunt at the same operation; however, followup information could not be distinguished between the two types of procedures, and the child was excluded from this study except when stated otherwise. The patients were not prospectively randomized. The decision on which type of shunt to construct was made by the
540 The Annals of Thoracic Surgery Vol 44 No 5 November 1987
surgeon, but each of the two main implanting surgeons chose each type about equally. The two groups were concurrent: time from operation to June 30, 1985, was 36.4 15.6 months (mean f standard deviation) for the group with a modified shunt and 36.1 f 19.4 months for the group with a standard shunt (not significant [NS]).
*
Preoperative and Postoperative Studies For each patient, preoperative and postoperative Sa02 values were recorded, and congestive heart failure, defined as need of treatment with digoxin, was noted. Shunt failure was defined as shunt thrombosis or occlusion, or late rise in hemoglobin levels greater than 18 @lo0 ml, which was taken to indicate shunt inadequacy. Hospital failures were those occurring before hospital discharge, and late failures were those occurring after hospital discharge. Postoperative diaphragmatic paralysis was noted. When available, preoperative cineangiograms were reviewed (18 patients with modified Blalock-Taussig shunts and 19 with standard Blalock-Taussig shunts), and measurements were recorded for the diameter of each pulmonary artery near its origin, each subclavian artery just proximal to its first branch, and the cardiac catheter. All measurements were taken during ventricular systole and, when possible, from the same angiographic frame. In addition, the body surface area at the time of cardiac catheterization and all catheter sizes were recorded. Magnification factors were determined by comparing the measured diameter of the cardiac catheter on the cineangiogram with its known diameter, and from this, actual values of subclavian and pulmonary artery diameters were determined. The pulmonary artery index (PAI) as described by Nakata and co-workers [13] was calculated; this index is the ratio of the total cross-sectional area of the branch pulmonary arteries to the body surface area. When postoperative cineangiograms were available (11 patients with modified Blalocl-Taussig shunts and 8 with standard Blalock-Taussig shunts), they also were reviewed, the same dimensions were recorded, and the same index was derived. Finally, any distortion of the subclavian artery or the pulmonary artery on follow-up catheterization films was noted and graded (mild, moderate, severe). Operative Technique All shunts were constructed through a standard lateral thoracotomy. The subclavian artery and pulmonary artery were identified. When it was elected to perform a modified Blalock-Taussig shunt, a segment of Gore-Tex vascular graft 5 to 6 mm in diameter was selected according to the size of the child and the diameter of the subclavian artery, and was anastornosed to a longitudinal incision in the side of the subclavian artery. The graft was then cut to a length that allowed it to lie without tension at the pulmonary artery, and it was sewn to a longitudinal incision in that artery. When a standard Blalock-Taussig shunt was chosen, the subclavian artery
was dissected free from the surrounding tissue, divided proximal to its first branch, ligated distally, and anastomosed end-to-end to the pulmonary artery. To help ensure shunt patency, systemic blood pressure was maintained at 80 mm Hg with Neo-Synephrine (phenylephrine hydrochloride), if needed, and the patient was heparinized (100 units per kilogram of body weight) intravenously every 4 hours for 12 to 24 hours after operation. Statistical Analysis Results were analyzed using Pearson’s chi-square test, Fischer’s exact test, or the Wilcoxon test, where appropriate. Statistical significance was established at alpha equal to 0.05.
Results
Preoperative Status The mean age and weight of patients given a modified Blalock-Taussig shunt were 2.4 2 3.4 months and 4.2 & 1.8 kg, respectively. Far patients given a standard Blalock-Taussig shunt, the mean age was 1.4 f 2.5 months and the mean weight, 3.5 & 1.2 kg. The anomalies in these patients were as follows: pulmonary stenosis or atresia and intact ventricular septum, 2 patients in the modified shunt group and 5 in the standard shunt group; pulmonary stenosis or atresia, intact ventricular septum, and transposition of the great arteries, 3 in the standard shunt group; pulmonary stenosis or atresia and ventricular septa1 defect (includes double-outlet right ventricle, transposition of the great arteries, atrioventricular canal, and single ventricle), 11 in the modified and 16 in the standard shunt group; and tricuspid atresia, 11 in the modified and 5 in the standard shunt group. With the exception of tricuspid atresia, which was statistically more prevalent in the group with a modified Blalock-Taussig shunt (p < 0.05), there were no significant differences between the two groups preoperatively. Arterial Oxygen Saturation and Congestive Heart Failure Average preoperative Sa02 values were statistically the same for the two groups (55 f 21% for the modified shunt group and 53 2 29% for the standard shunt group). However, average postoperative values were significantly different (p = 0.01): 84 ? 9% for the modified shunt group and 74 f 19% for the standard shunt group. Although these average postoperative values were different, the patient-specific changes in Sa02 from before to after operation were not different between the groups (p > 0.05): Sa02 = +32 5 21% for the modified shunt group and + 20 & 35% for the standard shunt group. The apparent discrepancy is due to the effect of patients in whom the Sa02decreased rather than increased after the shunt, primarily patients with early shunt failure. No instances of congestive heart failure were identified.
541 Ullom et al: Standard versus Modified Blalock-Taussig Shunt in Infants
a standard shunt was included in this comparison; it is of interest to note that the diaphragmatic paralysis occurred on the side with the standard shunt.
Shunt Failure and Mortality with the Blalock-Taussig ShunP Variable Total deaths
Early Late Total failures Early Late
Modified Shunt (N = 24)
9 (38)
4 5 (21) 5 (21) 1 (4Ib 4 (17)
Standard Shunt
(N = 29) 10 (35) 8
(Wb
2 (7) 15 (52) 4 (14)b 11 (38)
p Value
Cineangiography
NS NS
ACTUAL PULMONARY A N D SUBCLAVIAN ARTERY SIZE.
NS 0.05 0.08 0.03
"Numbersin parentheses are percentages. patient who died had both a modified and a standard shunt. Each shunt failed in the hospital, and the patient is counted as a death and a failure in both shunt categories. NS
=
not significant.
Deaths and Shunt Failures Nineteen patients died (9 with a modified and 10 with a standard shunt) (Table). The cause of death was unexplained in 4 patients (2 modified and 2 standard shunts). It was due to sepsis in 5 patients (2 modified and 3 standard shunts), intraoperative myocardial ischemia in 1patient (modified shunt), associated malformations in 3 (1modified and 2 standard shunts), reoperation in 4 (3 modified shunts and 1 standard shunt), and shunt thrombosis in 2 (standard shunts). Death was related to early shunt failure (all 4 of these patients died, 3 early and 1 late), but not to late failure (only 2 of 15 died). Shunt failure occurred in 20 patients, significantly more often in the group with a standard Blalock-Taussig shunt ( p < 0.05) (see Table). Actuarial analysis also demonstrated a lower risk of failure over time with this type of shunt (p < 0.05). The 1early failure of a modified Blalock-Taussig shunt occurred in the patient with a contralateral standard Blalock-Taussig shunt, which also failed in the hospital. The patient died of postoperative sepsis. The remaining three instances of early failure (all standard shunts) involved shunt thrombosis associated with early death. The 4 patients with late failure of a modified BlalockTaussig shunt were reoperated on for lesion repair (3 patients) or creation of a second shunt (1 patient). Three patients are living, and 1 patient died of postoperative sepsis following the corrective procedure. Among the patients with late failure of a standard Blalock-Taussig shunt, 1 died at home with a hemoglobin count of 20.8 @lo0 ml and 10 are alive after a second operation for lesion repair (7 patients) or construction of a second shunt (3 patients). Diaphragmatic Paralysis Postoperative diaphragmatic paralysis requiring subsequent operation for plication developed in 4 patients: 1 (4%)of 25 modified Blalock-Taussig shunts and 3 (10%) of 30 standard Blalock-Taussig shunts (NS). The child who underwent concurrent placement of a modified and
Pd-
monary artery diameters on the side of the shunt ranged from 2.7 to 7.4 mm preoperatively. Four (22%) of 18 patients with a modified shunt and 7 (37%)of 19 with a standard shunt for whom measurements were available had a shuntside pulmonary artery diameter of 3.0 mm or less (NS). Among these patients with small vessels, none of the modified shunts failed but 4 of the standard shunts failed, 2 in the hospital and 2 late (0 of 4 modified shunts versus 4 of 7 standard shunts [NS]). Comparing all patients with small vessels with patients with larger vessels, 2 early failures (18%) occurred in 11 patients with a shuntside pulmonary artery diameter of 3.0 mm or less, and 2 early failures (8%) also occurred in 26 patients with a pulmonary artery diameter greater than 3.0 mm (NS). Subclavian artery diameter on the side of the shunt ranged from 1.6 to 4.9 mm preoperatively. A total of 10 patients (5 with each type of shunt) had a shuntside subclavian artery measuring less than 2.5 mm. Three of these shunts failed (0 modified shunts versus 3 standard shunt [NS]). RELATIVE PULMONARY ARTERY SIZE. Eighteen patients (7 with a modified and 11 with a standard shunt) had a shuntside PA1 of 50 mm2/m2or less (normal PA1 for one side = 165 mm2/m2and for both sides, 330 mm2/m2[13]). Among the 7 patients with a modified shunt, 1 early failure and 2 late failures occurred, and among the 11 with a standard shunt, 1 early failure and 5 late failures occurred (NS). Comparing all patients having small vessels with those having larger vessels, 2 early failures (11%)occurred in 18 patients with a shuntside PA1 of 50 mm2/m2or less, and 1 failure (6%) occurred in 17 patients with a shuntside PA1 of more than 50 mm2/m2 (NS). COMPARATIVE PULMONARY ARTERY (PAI) GROWTH. Growth in pulmonary arteries following palliative procedures was evaluated by comparing the preoperative total PA1 with the postoperative total PAI. The average preoperative total PA1 values of the modified and the standard Blalock-Taussig shunts were not statistically different, but the average postoperative total PAI was greater with the modified than the standard shunt (p = 0.07). For the modified shunt group, the average PAI increased from 144 118 mmz/m2 (N = 16) preoperatively to 431 188 mm2/m2(N = 6) ( p = 0.03) at an average follow-up of 29 months. For the standard shunt group, the average PA1 increased from 118 59 mm2/m2 (N = 18) preoperatively to 189 f 106 mm2/m2(N = 6) (p = 0.05) at an average follow-up of 18 months. Twelve patients (6 from each group) had both preoperative and postoperative angiography, thereby allowing comparison of paired data. From preoperative to postoperative measurements, the total PA1 increased by
*
*
*
542 The Annals of Thoracic Surgery Vol 44 No 5 November 1937
296 2 273 mm2/m2at an average time of 29 months in the modified shunt group and by 129 a 117 mm2/m2at an average time of 20 months in the standard shunt group. However, this difference was not significant because of the large variability. PULMONARY AND SUBCLAVIAN ARTERY DISTORTION. Distortion of the pulmonary artery and subclavian artery on follow-up catheterization was also noted. Of the 11 patients with a modified Blalock-Taussig shunt who were reviewed, only 4 (36%)had distorted pulmonary arteries (all mild) on the shunt side and of the 8 with a standard shunt evaluated, 6 (75%) had distortion of the pulmonary artery on the shunt side, 4 mild and 2 moderate ( p = 0.06). Six (55%) of the 11 patients with a modified shunt had a distorted subclavian artery on the shunt side on follow-up catheterization films, 4 had severe and 2 had mild distortion. The artery was patent in every instance, but some decrease in distal flow was appreciated in the 4 with severe subclavian artery distortion. Because of the nature of the operation, all patients with a standard shunt had interrupted subclavian arteries with no direct flow to the distal part of the artery. NEWBORNS LESS THAN 1 WEK OLD. Newborns less than 1 week of age exhibited the same trends as the older infants, but because of the small number of patients, differences did not reach statistical significance.
Comment
Arterial Oxygen Saturation Group differences in postoperative SaOzvalues were not significant when considering only those patients whose shunt did not fail in the hospital. This is expected, since flows in both modified and standard shunts are limited by the diameter of the subclavian artery, and improvement in SaOZis a direct function of added pulmonary blood flow resulting from the constructed shunt. Deaths and Shunt Failures Overall mortality was not different between the two groups. Shunt failure, however, occurred significantly less often after creation of a modified shunt than a standard shunt. These results concur with those reported by Moulton and co-workers [14], although their study was not restricted to infants. In the current study, early failures always resulted in death, regardless of which type of shunt had been constructed. Late failures, however, could usually be treated with further operation and were seldom fatal. Both early and late failures occurred less frequently with a modified shunt than with a standard shunt. Diaphragmatic Paralysis Although the incidence of diaphragmatic paralysis in this study was low and statistically significant differences between the two groups were not found, it is of interest to note that diaphragmatic paralysis occurred less frequently in the modified shunt group than in the
standard shunt group. This might be expected, since more dissection is required for a standard shunt than for a modified shunt.
Cineangiography The modified Blalock-Taussig shunt procedure is associated with more pulmonary artery growth than the standard Blalock-Taussig shunt procedure, as suggested by the larger increase in PA1 at similar follow-up times, although both procedures resulted in a significant PA1 increase. This difference is important because adequate pulmonary artery size is a critical determinant of survival and late results after definitive repair of cyanotic heart malformations [13]. Although the reason for this difference is uncertain, it may be related to higher flow over longer periods through the modified shunt, as evidenced by a higher average Sa02postoperatively and a lower rate of shunt failure both early and late. Preservation of blood flow in the subclavian artery distal to the shunt was excellent in most patients with a modified Blalock-Taussig shunt, but was compromised by severe distortion of the subclavian artery in 4 (36%)of the 11 patients studied. The creation of a modified Blalock-Taussig shunt appears to preserve pulmonary artery anatomy well, little or no distortion being found after that procedure, compared with a high rate of distortion after a standard Blalock-Taussig shunt procedure. It is also of interest that we observed no instance of congestive heart failure despite the use of 5-mm and 6mm grafts in these infants. This problem may have been alleviated by our placement of the proximal anastomosis well onto the subclavian artery, thereby avoiding inclusion of any part of the aortic arch or innominate artery. Overall, the modified Blalock-Taussig shunt had advantages over the standard Blalock-Taussig shunt in our series of infants: shunt failure rates were lower, pulmonary artery growth was greater, and distortion of the pulmonary arteries occurred less often. With a modified shunt, the rise in Sa02 was greater and the incidence of diaphragmatic paralysis was lower than with a standard shunt, although these differences were not statistically significant. A modified Blalock-Taussig shunt should be considered a reasonable choice in infants requiring a palliative procedure for cyanotic congenital heart disease.
References 1. Blalock A, Taussig HB: The surgical treatment of malformations of the heart in which there is pulmonary stenosis or pulmonary atresia. JAMA 128;189, 1945 2. Jennings RB Jr, Innes BJ, Brickman R D Use of microporous expanded polytetrafluoroethylene grafts for aortapulmonary shunts in infants with complex cyanotic heart disease. J Thorac Cardiovasc Surg 76489, 1978 3. Neches WH, Naifeh JG, Park SC, et a1 Systemic-pulmonary artery anastomoses in infancy. J Thorac Cardiovasc Surg 70:921, 1975 4. Arciniegas E, Farooki ZQ, Hakimi M, Green EW: Results of
543 Ullom et ai: Standard versus Modified Blalock-Taussig Shunt in Infants
5. 6. 7. 8. 9.
two-stage surgical treatment of tetralogy of Fallot. J Thorac Cardiovasc Surg 79:876, 1980 Laks H, Fagan L, Bamer HB, Willman VL: The BlalockTaussig shunt in the neonate. Ann Thorac Surg 25:220,1978 Fort L 111, Morrow AG, Pierce GE, et al: The distribution of pulmonary blood flow after subclavian-pulmonary anastomosis. J Thorac Cardiovasc Surg 50:671, 1965 Currarino G, Engle MA: The effects of ligation of the subclavian artery on the bones and soft tissues of the arms. J Pediatr 67:808, 1965 Mickell JJ, Oh KS, Siewers RD, et al: Clinical implications of postoperative unilateral phrenic nerve paralysis. J Thorac Cardiovasc Surg 76:297, 1978 Geiss D, Williams WG, Lindsay WK, Rowe RD: Upper extremity gangrene: a complication of subclavian artery division. Ann Thorac Surg 30487, 1980
10. Karpawich PP, Bush CP, Antillon JR, et al: Modified Blalock-Taussig shunt in infants and young children. J Thorac Cardiovasc Surg 89:275, 1985 11. de Leva1 MR, McKay R, Jones M, et al: Modified BlalockTaussig shunt. J Thorac Cardiovasc Surg 81:112, 1981 12. Ilbawi MN, Grieco J, DeLeon SY, et al: Modified BlalockTaussig shunt in newborn infants. J Thorac Cardiovasc Surg 88:770, 1984 13. Nakata S, Imai Y, Takanashi Y, et al: A new method for the quantitative standardization of cross-sectional area of the pulmonary arteries in congenital heart diseases with decreased pulmonary blood flow. J Thorac Cardiovasc Surg 88:610, 1984 14. Moulton AL, Brenner JI, Ringel R, et al: Classic versus modified Blalock-Taussig shunts in neonates and infants. Circulation 72:Suppl 2:35, 1985
Notice from the American Board of Thoracic Surgery The American Board of Thoracic Surgery began its recertification process in 1984. Diplomates interested in participating in this examination should maintain a documented list of the cardiothoracic operations they performed during the year prior to application for recertification. They should also keep a record of their attendance at thoracic surgical meetings, and other continuing medical education activities pertaining to thoracic surgery and thoracic disease, for the two years prior to application. A minimum of 100 hours of approved CME activity is required. In place of a cognitive examination, candidates for recertification will be required to complete both the general thoracic and cardiac portions of the SESATS I11 syllabus (Self-EducatiodSelf-Assessmentin Thoracic Surgery). It is not necessary for candidates to purchase
SESATS I11 booklets prior to applying for recertification. SESATS I11 booklets will be forwarded to candidates after their applications have been accepted. Diplomates whose 10-year certificates will expire in 1990 may begin the recertification process in 1988. This new certificate will be dated 10 years from the time of expiration of the original certificate. Recertification is also open to any Diplomate with an unlimited certificate and will in no way affect the validity of the orignal certificate. The deadline for submission of applications is July 1, 1988. A recertification brochure outlining the rules and requirements for recertification in thoracic surgery is available upon request from the American Board of Thoracic Surgery, One American Plaza, Suite 803, Evanston, IL 60201.