I undergone elective repair once they are older than 18

Transcription

1 Survival Analysis of Infants Under Age 18 Months Presenting With Tetralogy of Fallot Suzanne J. Vobecky, MD, William G. Williams, MD, George A. Trusler, MD, John G. oles, MD, Ivan M. Rebeyka, MD, Jeffrey Smallhorn, MD, Patricia Burrows, MD, Robert Gow, MD, and Robert M. Freedom, MD Departments of ardiovascular Surgery, ardiology, and Radiology, The Hospital for Sick hildren, Toronto, Ontario, anada The trend to perform early primary repair of tetralogy of Fallot prompted us to review our experience in the current era with the traditional protocol consisting of palliation during infancy, if required, and repair after infancy. During a 1-year period, 7 infants with tetralogy of Fallot presented aged less than 18 months. Thirteen infants (4.8%) had major noncardiac lesions that precluded definitive care for their congenital heart disease. Twenty infants (7.4%) had major associated cardiac lesions (atrioventricular septal defect or absent pulmonary valve syndrome, or both). Survival in this group was poor, with only 58% 1% reaching the age of 1 years. Four of the seven deaths occurred before intracardiac repair was performed. The remaining 37 infants presented with isolated tetralogy of Fallot. Eighty-nine percent.3% survived to age 1 years. Sixty percent of these infants required palliation, and survival in these infants did not differ from that in those who never required palliation. However, 19 infants (8%) required palliation in the first month of life. In these children, survival to age 1 years was significantly lower (77%), secondary palliation was frequently required (n = ll), and a transannular patch or conduit at the time of repair (1 of 14 patients) was more likely needed than it was in children who had not undergone a palliative procedure during the neonatal period. The survival in infants with tetralogy of Fallot is unlikely to be different, regardless of whether primary repair or a staged repair is carried out. The quality of survival, including the exercise capability and absence of arrhythmias, must be assessed to determine which protocol is superior. (Ann Thorac Surg 1993;56:944-5) nfants born with tetralogy of Fallot have traditionally I undergone elective repair once they are older than 18 months, but, if symptomatic in the first 18 months of life, a preliminary palliative shunt is created. There is now a trend toward performing earlier primary repairs [l-61 in symptomatic (and perhaps even in asymptomatic) infants, and the results from these procedures should be compared with the results achieved by the traditional protocol. This report details our findings concerning the survival of infants with tetralogy seen in the first 18 months of life and managed by a protocol consisting of palliation, if required under the age 18 months, and repair after the age 18 months. Methods All infants with tetralogy of Fallot who were under 18 months of age when diagnosed by cardiologists at the Hospital for Sick hildren between January 1, 1978, and December 31, 1988, are included in this review. The diagnosis was confirmed by echocardiography and cardiac catheterization using standard definitions of atrioventricular and ventriculoarterial concordance with a Accepted for publication Dec 3, 199. Address reprint requests to Dr Williams, Department of ardiovascular Surgery, The Hospital For Sick hildren, University Ave, Toronto, Ont M5G 1x8, anada. perimembranous ventricular septal defect and anterior deviation of the infundibular septum causing right ventricular outflow tract obstruction. Infants with pulmonary valve atresia (previously reported on [7]) or a doubleoutlet right ventricle were excluded. The diagnosis of a double-outlet right ventricle, although somewhat arbitrary, was made when more than 5% of each great artery arose from the right ventricle. A total of 7 infants were identified from hospital records. Thirteen (4.8%) had additional noncardiac lesions that precluded care for their cardiac disease. These concomitant lesions included severe cerebral anomalies (n = 5), trisomy 13 (n = 3), trisomy 18 (n = l), Koffin-Siris syndrome (n = l), ectopic cordis (n = l), and other multiple severe noncardiac anomalies (n = ). These 13 infants were excluded from the analysis and all died of their associated anomalies. Major associated cardiac lesions were present in infants, and consisted of a complete atrioventricular septal defect in 11 and absent pulmonary valve syndrome in 1, with 1 child having both atrioventricular septal defect and an absent pulmonary valve. A brief analysis of this group is included, although, the major focus of this review is on the remaining 37 infants who had isolated tetralogy of Fallot. The hospital inpatient and outpatient records for each infant were reviewed and the follow-up status of all children established in 199 or 1991, either by outpatient 1993 by The Society of Thoracic Surgeons /93/$6.

2 Ann Thorac Surg 1993;569&5 VOBEKY ET A 945 Table 1. Twenty Infants Aged 18 Months With Major Associated esions Variable No. Mortality Associated lesiona omplete AVSD Absent pulmonary valve Timing of death Before palliation At palliation Secondary to palliation After palliationhefore repair At repair After repair Total (35%) a One infant has both AVSD and an absent pulmonaly valve. AVSD = atrioventricular septa1 defect. clinic visit or by telephone contact with the referring cardiologist. Results Tetralogy With Major Associated ardiac esions Seven of the infants with major associated cardiac lesions died during the follow-up period. The timing of death is summarized in Table 1. A preliminary palliative operation was required in 11 infants, with one death. No infants required a second palliative operation. Sixteen underwent intracardiac repair, 8 who had undergone previous palliation and 8 who had not, and there was one death in each group. The operative mortality associated with repair is 1.5%. The actuarial survival to age 1 years is 58% (*1%) (Fig 1). Isolated Tetralogy of Fallot Twenty-two of 37 children (9.%) with isolated tetralogy of Fallot died during the follow-up period of from to $.6 Birth Fig I. The actuarial survival rate to age 1 years of infants with tetralogy of Fallot and major associated lesions is 58% * 1% i E.1 c n = 37 O I I I I I I I I I I I Birth Fig. In 37 infants presenting with isolated tetralogy of Fallot, the survival rate to age 1 years was 89.5% f.3%. years (mean, 6. * 3.1 years). The actuarial survival to age 1 years was 89.5% k.3% (Fig ). The timing of death with respect to the treatment protocol is listed in Table. Fourteen of these patients (64%) died before cardiac repair was performed. At present, only 8 surviving children have not yet undergone repair. Their average age is 6.5 years and ranges between.6 and 1.9 years. Repair has been postponed in 4 of these 8 children because of hypoplastic pulmonary arteries in, a small left ventricle in 1, and severe mental retardation in 1. INITIA PAIATION. One hundred forty-one infants (59.5% of the total) underwent surgical palliation because of cyanotic spells or a rising hemoglobin level and falling oxygen saturation detected during cardiologic follow-up. The timing of intervention is illustrated in Figure 3. Although 95 infants (4% of all the infants with isolated tetralogy) were first seen in the first month of life, only 19 (8%) required palliation during this time. The actuarial survival of the 141 infants requiring palliation is compared to that in the 96 infants who did not undergo palliation (Fig 4). The survival in both groups is similar, with no statistically significant difference. Survival to age 1 years in the 19 infants requiring palliation in the first month of life was 77%, that in the 1 Table. Infants With Isolated Tetraloay (n = 37) Timing of Death None Prepalliation At palliation Secondary palliation After palliationbefore repair At repair No. 18 (9%) After repair 1 Total deaths za (9.3%) a Fourteen deaths occurred before repair.

3 946 VOBEKY ET A Ann Thorac Surg 1993;5694&5 c Q).- > - t;;.4 3 f. n " < AGE (Months) First Seen (n = 37) Palliated (n = 141) Fig 3. The ages at which the 141 infants requiring palliation were first seen and their ages at initial palliation are shown. Forty percent of these infants were initially seen in the neonatal period, but only 8% required neonatal palliation. who underwent palliation after the age of 1 month was 89% ( p <.5; Fig 5), and that in the 96 who never underwent palliation was 9%. The size of the pulmonary artery noted at the time of the initial palliation is an important determinant of survival (Table 3). In 5 infants with a pulmonary artery size of less than 3 mm (external diameter), there were three deaths compared with 13 deaths in the 11 infants with a pulmonary artery size greater than 3 mm. In 15 infants, the pulmonary artery size could not be discerned from the operative records. Initial palliation failed in 36 patients (6%). Failure due to death occurred at initial palliation in and in the interval before definitive repair was performed in 8, including one death after subsequent palliation. The need to perform a subsequent palliative operation was the source of failure in the remaining 6. SUBSEQUENT PAIATION. Data regarding the 6 infants who required secondary palliation are presented in Table 1- - I).g > 3.6- cn - S.O.4- r - P g.- n - 77% 89% O J I I (, I, , 1, Birth Fig 5. Survival in the 19 infants requiring neonatal palliation in the first month of life (77% survived to age 1 years) is significantly worse than that in infants requiring palliation after the age of 1 month (89% survived to age 1 years; p <.5.) 4. In 3 infants, two further palliative procedures were required before proceeding to intracardiac repair. The indications for secondary palliation, as opposed to definitive repair, were age less than 18 months (n = 14), pulmonary arteries judged too small for definitive repair (n = 6), unfavorable coronary artery anatomy (n = ), associated cardiac anomalies such as a small left ventricle (n = *), or associated noncardiac disease, specifically severe mental retardation (n = ). Two children died subsequent to secondary palliation. One child, who required initial palliation at the age of 4 days, underwent revision of a right Blalock-Taussig shunt 3 days later and was considered an early death. The second death occurred 1 month after secondary palliation when a left Blalock-Taussig shunt was required to supplement a previously inadequate right Blalock-Taussig shunt created 3% months earlier at the age of 6 weeks. * One of these children has undergone repair since submission of the manuscript and died at operation. Table 3. Pulmonary Artery Size at Initial Palliation Non-Palliated Infants (n = 96) mi;.4 Palliated Infants (n = 141). n Birth Fig 4. Survival in the 141 infants with isolated tetralogy of Fallot requiring palliation is not significantly different from that seen in the 96 who never underwent palliation. Artery Size" (mm) No. Mortality (%) < 1 l(l) -3 4 (5) (11) (4) 5-6 (1) (15) (1) l%9 6 1 (17) (6) >1 7 Unknown 15 Total (11.3) a External diameter.

4 Ann Thorac Surg 1993; VOBEKY ET A 947 Table 4. Repair of Isolated Tetralogy Previous Palliation Variables None Initial Only Secondary Total No. of patients Mortality 3 (3.3%) 4 (3.8%) 7 (3.%) P u 1 m o n a ry 16 (15%) 1 (5%) 6 (1%) arterioplasty Transannular 49 (54%) 5 (48%) (1%) 11 (47%) patch onduit 4 (4%) 8 (4%) 1 (6%) Subsequent 6 (7%) 9 (8.6%) 1(5%) 16 (7%) operation The actuarial survival to age 1 years in the 6 infants requiring secondary palliation was 9%. Twenty of the 4 survivors underwent subsequent intracardiac repair, with no early or late deaths. Ten of the required pulmonary arterioplasty at repair; bilateral arterioplasty was necessary in 3 children. In contrast, pulmonary arterioplasty was required in 15% of the infants who had only one palliative shunt (see Table 4). The average age of these patients at repair was 5 years and the average interval after secondary palliation was 3. years. INTERVA DEATHS. Eight interval deaths occurred after palliation and before intracardiac repair. The average age at death was.3 years (range, 6 months to 4.8 years). These children died an average of 17.7 months after palliation (range, 4. to 43.3 months). Five died less than 1 year after initial palliation. None of these patients presented before the age of 1 month and their age at palliation averaged 7.6 months (range, 1.1 to 15.3 months). Four of the patients were from distant remote areas. Four deaths were due to infections, including cerebral abscess in and meningitis in 1. The cause of death in the other 4 infants is unknown. In addition, 3 infants died who had undergone no surgical intervention at any time, and were considered interval deaths. One was admitted in extremis as a newborn and another died at 5 months and 7 months, 1 of pneumonia and the other of acute leukemia. Otherwise, the cardiologists decision to defer repair in asymptomatic infants until after the age of 18 months, based on the findings obtained at three monthly outpatient visits, which included review of the child s clinical state, hemoglobin level, and peripheral oxygen saturation, was entirely successful in managing these patients safely without mortality. INTRAARDIA REPAIR OF ISOATED TETRAOGY OF FAOT. Two hundred sixteen (91%) of the 37 infants have undergone intracardiac repair (see Table 4). The overall operative mortality at intracardiac repair consists of 7 infants (3.%), including three deaths in 17 patients who underwent repair elsewhere. In 8 infants, the protocol of only carrying out palliation under 18 months of age was not followed and primary repair was performed as the initial intervention. The indication for such early repair was a restrictive ventricu- lar septa1 defect in 3 infants and tetralogy in the presence of significant left-to-right shunting in. Two of the children underwent early repair at another institution, and 1 child was close to the elective age of 17 months at the time of repair. One death occurred in these 8 patients (1.5%), a 15-month-old child who underwent repair elsewhere. Previous Palliation. The operative risk associated with repair in children who have previously undergone palliation (4 of 15) versus that in those who have not (3 of 91) did not differ. However, the age at intracardiac repair was significantly lower in the patients who had had previous palliation (mean, 37.1 months) than that in those who had not (mean, 48.9 months; p <.1). Infants who had undergone a previous palliative operation were more likely to require a valved conduit at the time of repair (1 of 15 versus of 91; see Table 4). The indication for a valved conduit in 5 patients was coronary artery anomalies, and, in 7, small pulmonary arteries. Type of Repair. A transannular patch was used in 11 children (47%). The various types of repair are listed in Table 4. Whether a transannular patch was placed at repair had no important influence on the operative risk. The details of the procedure performed in 7 of the 17 children who underwent repair elsewhere are unknown. Either a transannular patch or a conduit was required in 3 of the 95 infants who were first seen in the first month of life, compared with 41 of the 11 infants seen after the first month of life, but this difference did not attain statistical significance. However, of the 19 infants who underwent palliation in the first month of life, 4 required a transannular patch and 6 a valved conduit. onduits appear to be more frequently needed in patients requiring secondary palliation before repair; 8 of such patients in this study required a valved conduit. Pulmonary Arterioplasty. Patch arterioplasty of one or more of the branch pulmonary arteries was required in 6 infants at the time of total repair. Sixteen patients (15.%) had had a single previous palliative shunt created, whereas 1 of the patients with two previous shunts required an arterioplasty on at least one side. Three of these 1 children required bilateral arterioplasty, which was not necessary in any of the children with one previous shunt. The need for a pulmonary arterioplasty was not always directly related to the shunt, and, in 5 patients, was performed on the side opposite the palliative shunt. None of the 91 children who had not had previous palliative operations required pulmonary arterioplasty on either branch pulmonary artery. Whether a pulmonary arterioplasty was performed did not affect mortality. oronary Anomalies. oronary anomalies were noted in angiograms obtained after repair. Twenty-one of these angiograms were available for review, and were examined by one of us (P.B.). The following findings were noted:

5 948 VOBEKY ET A Ann Thorac Surg 1993;56: > n = 9 Re-Operation = 15 Average Follow-up 3. yrs. c -i;.4 g. n Repair Follow-up Post Repair(Years) Fig 6. Two hundred nine children were discharged from the hospital after intracardiac repair of isolated tetralogy. During their subsequent follow-up (average, 3. years per patient), the proportion surviving free of reoperation for the treatment of residual defects is 93% (*1.8%) 5 years after repair. eft anterior descending coronary artery 5" from right coronary artery Accessory left anterior descending coronary 1 artery from right coronary artery arge right conus branch Single coronary artery 1 Normal coronary anatomy 3 Five of the 1 valved conduits used at repair were for the management of coronary artery anomalies. There were no deaths in these patients. Of importance, of 4 infants diagnosed as having a left anterior descending coronary artery from the right coronary artery on initial angiograms (n = 31) proved to have normal coronary artery anatomy according to their prerepair angiograms. ATE RESUTS. Two hundred nine infants were discharged from the hospital after successful intracardiac repair, and, of these, there has been one late death. During the follow-up period since intracardiac repair, which averages 3. *.3 years, there have been 15 reoperations for the treatment of residual intracardiac defects, including: a residual ventricular septa1 defect (n = 4), right ventricular aneurysm (n = 5), pulmonary valve rep1ace:ment for right ventricle failure (n = ), aortic valve repair (n = l), conduit revision (n = ), and pacemaker insertion (n = 1). Portrayed actuarially (Fig 6), 93% 1.8% of the children had been free from the need for further surgcal intervention as of 5 years after repair. At the most recent follow-up, 9% of the children are in New York Heart Association class I and 7% are in class 11. * Four required a valved conduit at repair. The 8 children who have not yet undergone repair, including 7 who have received palliative treatment, continue to be observed and at recent follow-up were found to be well. omment We reviewed the "unnatural" history of 7 infants who were initially seen with tetralogy of Fallot in the first 18 months of life over a 1-year period, and who were managed by a protocol consisting of a palliative shunt procedure under the age of 18 months, if required, and intracardiac repair after the age of 18 months. Many published series have demonstrated excellent survival of infants with tetralogy repaired primarily [1-6]. However, it is seldom clear what population these surgical cases come from. We have included in our series all infants diagnosed as having tetralogy under the age of 18 months in an attempt to view the entire spectrum of the disease, including those children who died before operation could be performed. Most of these infants (37 of 7; 88%) had isolated tetralogy of Fallot, and their survival to age 1 years is 89.5% f.3%. Sixty percent of the infants with isolated tetralogy required palliation, and survival was unaffected by the need for palliation. However, palliation carried out in the first month of life (n = 19) was associated with a poorer survival of only 77% to age 1 years and a higher incidence of the need for a transannular patch, conduit, or both at repair. Palliation using a modified Blalock-Taussig shunt was also unsatisfactory in infants with very small (less than 3 mm in diameter) branch pulmonary arteries (3 of 5 such infants died). For this small group of infants, an alternative to palliative modified Blalock-Taussig shunt is required. Although survival in the patients in our study was unaffected by the need for palliation, there is a major failure rate associated with initial palliation (6%), which includes 8 infants who died after discharge from the hospital following palliation but before intracardiac repair could be undertaken. Four of the 8 deaths occurred in children from remote areas in whom the routine outpatient checkups at every three months were impractical. Distortion of the pulmonary arteries by the palliative shunt required arterioplasty at repair in 15% of the infants who had one palliative shunt overall and in 5% of the patients who required more than one palliative operation. It is important to note that 4% of the infants first seen when less than 18 months old were asymptomatic and were not referred for surgical intervention until the time of intracardiac repair performed after they had reached the age of 18 months. These asymptomatic patients fared well, with almost no mortality, the two exceptions being a death from leukemia and another from pneumonia. We excluded from analysis infants with both tetralogy and pulmonary valve atresia, as this group has been previously reported on [7]. In children with pulmonary atresia, the dominant problem is hypoplasia, nonconfluence, and arborization anomalies, problems that are rare

6 Ann Thorac Surg 1993;56:9445 VOBEKY ET A 949 in the setting of tetralogy and have a major impact on survival. The lesions associated with complete atrioventricular septal defect and absent pulmonary valve syndrome have a major unfavorable impact on outcome, with only a 58% survival in such patients to age 1 years. Whether a policy of early primary repair in these infants would improve survival is unknown. It may be of epidemiologic interest that 13 infants with tetralogy (4.8% of the total) had such severe associated noncardiac lesions [8] as to preclude anything but terminal supportive care. esions of this severity are uncommon in most other forms of congenital heart disease. The 89% survival to age 1 years in these 37 patients presenting in the first 18 months of life is unlikely to differ from that reported for early primary repair [1-6]. To demonstrate a statistically important improvement in survival for either approach to the management in such patients would require a cohort of over 7 infants. learly, survival alone is not an issue when deciding whether primary repair or a two-stage approach is superior. Early primary repair may protect late neurologic function [9] and ventricular function [lo], including less frequent ventricular arrhythmias [ll], a complication occurring in 3% of the patients in Garson and colleagues' [1] series who underwent late two-stage repair. However, repair at a younger age is associated with a higher incidence of the need for transannular patching, which may compromise late right ventricular function and result in a higher risk of reoperation [13, 141 than we have witnessed in our patients to date (see Fig 6). We have not examined the financial implications of the two-staged repair required in 6% of the patients in our series, and it may be an appropriate consideration in deciding upon a treatment protocol for infants presenting with tetralogy of Fallot. Therefore, specific assessment of the well-being of these children, their ability to exercise, their neurologic development, and the frequency of late-onset arrhythmias and ventricular dysfunction is needed to determine the most appropriate timing of repair. For the high-risk subgroup of symptomatic neonates (8% of the total) and infants with very small pulmonary arteries (% of the total) alternatives to the policy of initial palliation using a modified Blalock-Taussig shunt should be sought. References 1. Kirklin JW, Blackstone EH, olvin EV, Mconnell ME. Early primarv correction of tetralogy of Fallot. Ann Thorac SurE - i988;45: Gustafson RA. Murrav GF. Warden HE. Hill R. Rozar GE Jr. Early primary repiir of tetralogy of'fallot. Ann Thorac Surg 1988;45: Walsh EP, Rockenmacher S, Keane JF, Hougen TJ, ock JE, astaneda AR. ate results in patients with tetralogy of Fallot repaired during infancy. irculation 1988;77: Touati GD, Vouhe PR, Amodeo A, et al. Primary repair of tetralogy of Fallot in infancy. J Thorac ardiovasc Surg 199;99: DiDonato RM, Jonas RA, ang P, Rome JJ, Mayer JE, astaneda AR. Neonatal repair of tetralogy of Fallot with and without pulmonary atresia. J Thorac ardiovasc Surg 1991; 11:1& Groh MA, Meliones JN, Bove E, et al. Repair of tetralogy of Fallot in infancy. Effect of pulmonary artery size on outcome. irculation 1991;84(suppl 3): Hofbeck M, Sunnegardh JT, Burrows PE, et al. Analysis of survival in patients with pulmonic valve atresia and ventricular septal defect. Am J ardiol 1991; Kappetein AP, Gittenberger-de Groot A, Zwinderman AH, Rohmer J, Poelmann RE, Huysmans HA. The neural crest as a possible pathogenetic factor in coarctation of the aorta and bicuspid aortic valve. J Thorac ardiovasc Surg 1991;1: Newburger JW, Silbert AR, Buckley P, Fyler D. ognitive function and age at repair of transposition of the great arteries in children. New Engl J Med 1984;3: Graham TP Jr. Ventricular performance in congenital heart disease. irculation 1991;84: Deanfield JE, McKenna WJ, Presbitero P, England D, Graham GR, Hallidie-Smith K. Ventricular arrhythmia in unrepaired and repaired tetralogy of Fallot. Relation of age, timing of repair, and haemodynamic status. Br Heart J 1984;5: Garson A, Porter B, Gillette P, McNamara DT. Induction of ventricular tachycardia during electrophysiologic studies after repair of tetralogy of Fallot. J Am oll ardiol 1983;l: Kirklin JK, Kirklin JW, Blackstone EH, Milano A, Pacific AD. Effect of transannular patching on outcome after repair of tetralogy of Fallot. Ann Thorac Surg 1989;48: Zhao HY, Miller D, Reitz BA, Shumway NE. Surgical repair of tetralogy of Fallot. ong term follow-up with particular emphasis on late death and reoperation. J Thorac ardiovasc Surg 1985;89:4-. INVITED OMMENTARY Doctor Vobecky and her associates from The Hospital for Sick hildren in Toronto have reviewed here the experience of this internationally renowned pediatric cardiac unit with the treatment of tetralogy of Fallot (TOF) in infants during January 1, 1978, through December 31, Of the 7 consecutive infants with TOF identified during that decade, Vobecky and associates appropriately excluded from analysis 13 infants with fatal noncardiac lesions as well as infants with major associated cardiac defects, such as complete atrioventricular canal and absent pulmonary valve syndrome. ikewise, patients with TOF and pulmonary valve atresia were also excluded. The exclusion of this subset of patients is more debatable, because the nature of the pathophysiology and the surgical palliation or repair are similar to those of the isolated or simple form of TOF. In any event, of the 37 patients