Improved Results in Newborns Undergoing Coarctation Repair

Improved Results in Newborns Undergoing Coarctation Repair James Campbell, M.D., Raymond Delorenzi, M.S., John Brown, M.D., Donald Girod, M.D., Roger Hurwitz, M.D., Randall Caldwell, M.D., and Harold King, M.D. ABSTRACT Among 172 children who underwent repair of coarctation of the thoracic aorta from June 1,1967, to June 1, 1979, there were 50 who were less than 6 weeks old. A review was undertaken to assess mortality and postoperative residual hypertension in this newborn age group. Forty-five (90%) had one or more associated congenital cardiac lesions, which were repaired or palliated at the time of coarctation repair. The three most common associated lesions were patent ductus arteriosus (74%), atrial septal defect (74%), and ventricular septal defect (62%). Thirty-one newborns underwent resection with primary end-to-end repair with 8 deaths (26%), and 18 underwent Dacron patch angioplasty and 1 underwent subclavian angioplasty with 4 deaths (21%). Of the 38 survivors, 36 (95%) had good pedal pulses on postoperative clinical examination 3 months to eleven years postoperatively. Seventeen (45%) had upper and lower extremity blood pressure measured; 2 (5%) showed residual upper extremity hypertension (greater than 2 standard deviations above the mean and a gradient less than 20 mm Hg). One of these 2 patients has undergone repeat repair. Based on this low mortality and infrequent need for repeat repair, we advocate aggressive early operation and judicious use of palliative procedures in newborns, when indicated, for associated defects. Beginning with the reports of Gross [61 and Crafoord and Nylin [21, the results of surgical treatment of coarctation of the aorta have progressively improved. Today the outcome of such treatment in children and young adults is gratifying. The same has not been true for the From the Indiana University Medical Center, Indianapolis, IN. Presented at the Sixteenth Annual Meeting of The Society of Thoracic Surgeons, Jan 21-23, 1980, Atlanta, GA. Address reprint requests to Dr. Brown, Department of Surgery, Indiana University Medical Center, 1100 W Michigan St, Indianapolis, IN 46223. infant age group. Of the 172 patients who underwent coarctation repair at the Indiana University Medical Center between June 1, 1967, and June 1, 1979, 12 of the 15 deaths occurred in patients less than 6 weeks old (Fig 1). The uniform presence of congestive heart failure and high incidence of associated congenital cardiac anomalies combined with the technical difficulty of operation and increased incidence of residual or recurrent coarctation make this age group unique and challenging. This report describes our experience with the surgical management and follow-up evaluation of 50 consecutive infants who were seen during the first 6 weeks of life. Materials and Methods Between June 1, 1967, and June 1, 1979, 172 children underwent repair of coarctation of the aorta at the Indiana University Medical Center. Of these, 50 patients were less than 6 weeks old at the time of operation. This young subgroup is considered in detail. There were 36 male and 14 female infants with a mean age at operation of 2.56 weeks. The primary indication for operation in all patients was intractable cardiac failure. Forty-five patients (90 /0) were seen with one or more associated congenital cardiac anomalies. All patients underwent preoperative cardiac catheterization, which included intracardiac pressure determinations, oximetry measurements, and aortography. The site of coarctation, the presence of aortic arch hypoplasia, and the nature of the associated cardiac defects were defined. All 45 patients with associated defects had at least one palliative or reparative procedure performed in addition to coarctation repair. The postoperative follow-up period extended from 3 months to eleven years. Information was obtained by review of hospital and outpatient 273 0003-4975/80/090273-08$01.25 @ 1980 by The Society of Thoracic Surgeons

274 The Annals of Thoracic Surgery Vol 30 No 3 September 1980 60 50 50 z w 40 I- d LL 30 0 z w m 5 20 z MORTALITY 34 18 2o 10 0 Wna h, 6/52 3/12 6/12 1 yr 3 r 6 r 3 12 15 18+ AGE AT OPERATION Fig 1. Age at operation of 172 patients who underwent repair of coarctation of the aorta at Indiana University during a twelve-year period. records. Data consisted of systemic blood pressure evaluation, clinical examination, and comparison of upper and lower extremity pulses and blood pressure. Operative Technique Thirty-one patients (62%) underwent resection with end-to-end anastomosis, 18 patients (36%) undewent Dacron patch aortoplasty, and 1 patient (2%) was treated with subclavian patch angioplasty. There was no systematic assignment of patients to one group or another. Prior to 1978, resection with end-to-end anastomosis was the preferred procedure. Since that time, Dacron patch aortoplasty has been utilized nearly exclusively. We are attracted to this procedure because of its relative ease, the shorter operating time, the potential for growth of the remaining aortic wall, and recent reports such as that by Freed and co-workers [41 concerning the lowered postoperative exercise gradients in patients repaired using this technique. All operations were performed through a left posterolateral thoracotomy. The pleural space is entered through the fourth intercostal space or the bed of the unresected fourth rib. Regardless of the proposed procedure, the distal transverse aortic arch, the left subclavian artery, the first two or three pairs of intercostal arteries, the ductus arteriosus, and the distal aorta are dissected free and controlled. If resection is to be performed, the involved segment must be circumferentially freed and dissection carried distally to the level of the fifth or sixth intercostal arteries or in some cases even further to ensure minimal anastomotic tension. When possible, the coarctation should be widely excised, including the segment in front of the stenosis, and.the anastomosis carried out between the base of the arch and the descending thoracic aorta (Fig 2). To accomplish this, the descending thoracic aorta must be dissected free to allow it to be moved up to the arch. Dissection distal to the coarctation site is limited to the level of the first two or three pairs of intercostal vessels if patch aortoplasty is done. The management of patent ductus arteriosus is determined by the technique of coarctation repair to be used. When resection is elected, the ductus is divided between fine vascular clamps

275 Campbell et al: Newborns Undergoing Coarctation Repair Fig 2. Coarctation repair using resection and end-to-end anastomosis. and the divided ends are oversewn with continuous fine Dacron or polypropylene suture. If aortoplasty is done, the ductus is ligated in continuity with a No. 5 silk tie or a large hemoclip. Fine microvascular instruments are used. If resection is chosen, the distal arch, left subclavian artery, and distal aorta are clamped. Intercostal vessels are controlled with doubly looped, temporary ligatures or microvascular clips. All involved aorta is resected. Repair is performed with a posterior continuous simple suture of 5-0 polypropylene and anterior interrupted simple sutures. When Dacron patch aortoplasty is elected, proximal control is obtained with a single, fine, curved vascular clamp, which incorporates both the distal arch and the left subclavian artery (Fig 3). The aorta is opened longitudinally through the coarctation site with the incision carried proximally 2 to 4 mm into the origin of the left subclavian artery. The coarctation diaphragm is ex- Fig 3. Coarctation repair using Dacron patch aortoplasty. cised as completely as possible. A rectangular knitted Dacron patch of a width equal to the circumference of the descending aorta is selected, trimmed to a length of 1 to 2 cm longer than the aortic incision, and preclotted. The proximal end is tailored in elliptical fashion. The graft is sutured in place with a continuous 6-0 polypropylene or Dacron suture. Graft tailoring is completed when the proximal half of the suture line is in place. The remainder of the repair is then completed. Air is vented before the suture is tied. The clamps are slowly removed to prevent sudden hypotension. Hemorrhage is controlled with packs, and all blood loss is immediately replaced with warmed whole blood. In selected patients, pulmonary artery banding is performed if preoperative catheterization showed a large ventricular septal defect with pulmonary artery pressure greater than twothirds of systemic pressure. Results Twelve of the 50 patients in this series who were operated on when they were 6 weeks old or less died-an overall mortality of 24%. There was 1 late death due to congestive heart failure in a child with complex congenital heart disease, two and one-half years following coarctation repair. The remaining 11 deaths took place during the perioperative period, for a 22% operative mortality. Two of these 11 deaths occurred intraoperatively and were attributable to hemorrhage or intractable cardiac rhythm abnormalities. In 2 patients, death was due to underlying metabolic derangement and cardiac arrhythmia. In the remaining 7 patients, death was directly related to ongoing cardiac failure secondary to the presence of one or more associated congenital cardiac anomalies. Forty-five (90%) of the 50 infants had one or more associated congenital cardiac anomalies (Fig 4). The most frequently found lesions were patent ductus arteriosus (74%), atrial septal defect (74'/0), and ventricular septal defect (62%). Twenty-one patients (42%) had varying degrees of aortic arch hypoplasia. Seven of these infants died. Multiple lesions were present in all of the patients who died. The follow-up period for the 38 survivors ranged from 3 months to eleven years with a

276 The Annals of Thoracic Surgery Vol 30 No 3 September 1980 Fig 4. Associated congenital anomalies (frequency relative to each other and within the population). (PDA = patent ductus arteriosus; ASD = atrial septal defect; VSD = ventricular septal defect; TAPVD = total anomalous pulmonary venous drainage; SV = single ventricle; TGV = transposition of the great vessels.) mean of 48.38 months. Data on clinical examination revealed good pedal pulses in 36 (95%) of the patients. In 17 patients (&YO), upper and lower extremity blood pressure measurements were recorded. Two of the 38 (5%) had substantial upper extremity hypertension and an arm to leg pressure gradient in excess of 20 mm Hg. Both were female patients. A gradient of 30 mm Hg was noted in 1 when she was 10 weeks old. She had undergone repair when she was 4 weeks old. The patient was doing well clinically and was not recommended for reoperation. The other patient was repaired at 1 week of age. Fifteen months postoperatively, a gradient of 52 mm Hg was documented on examination. Due to the presence of increasing congestive heart failure, aortography was performed and showed stenosis at the site of repair. She underwent operative revision with good results and was the only patient to undergo reoperation during the follow-up period (2.6%). Both patients were seen early in the series and were originally repaired by resection and end-to-end anastomosis. Comment Considerable progress has been made in the surgical management of patients with coarctation since 1945 and 1946 when reports of suc- cessful repair were published [2, 61. In 1960, Glass and associates [51 reported a 41% mortality among 34 patients operated on in the first year of life. Thirty-three of the repairs were done utilizing resection and primary end-toend anastomosis. In the same series, 7 deaths (70%) occurred in 10 patients who had coarctation repair at less than 4 weeks of age. At follow-up, 11 patients (55%) had no resting postoperative hypertension or arm to leg gradient, while a gradient or hypertension or both were found in 9 (45%). In 1967, Hallman and associates [71 cited a 21% operative mortality among 58 patients less than 1 year old when resection and end-to-end repair were used. Four additional patients died during the follow-up period of up to eleven years. Five of 42 patients available for followlup had unsatisfactory results with 2 undergoing reoperation for recurrent coarctation at five and eight years after the original repair. Tawes and co-workers 1131 reported in 1969 the largest series of coarctation repairs in infants. One hundred seventy-nine of 333 patients were less than l year old at the time of operation. Operative mortality in this group was 45%. In a high-risk subgroup of 75 patients less than 6 weeks old, the operative mortality was 67%. Two hundred nine of the 333 patients were available for follow-up. Twenty (9.6%) had evidence of recoarctation. Nine (36%) of 25 survivors less than 6 weeks old were thought to have recoarctation. Ten patients out of the 209 survivors eventually underwent reoperation. Macmanus and colleagues [91 in 1977 reported their experience with 44 infants less than 90 days old. They cited 14 operative and 8 late deaths, an overall mortality of 50%. Of 17 survivors with adequate follow-up data, 8 were considered to have excellent results, 4 to have fair, and 5 to have poor results. Four patients underwent early anastomotic revision, with 3 deaths. Reoperation was anticipated for 4 late survivors. Herrmann and co-workers [8] in 1978 published a series of 25 patients less than 1 year old undergoing coarctation repair. There were 8 (32%) operative and 3 (12%) late deaths, for an overall mortality of 44%. The operative mortality in 18 patients less-than 3 months old was

277 Campbell et al: Newborns Undergoing Coarctation Repair 44%. Three of 17 early survivors had residual hypertension. Mortality in older age groups was minimal in all foregoing reports. A review of our own experience was undertaken to evaluate our results with the highest risk group, those patients 6 weeks old and less. Among 50 neonates less than 6 weeks old, we had an operative mortality of 22% and 1 late death, an overall mortality of 24%. The incidence of residual arm to leg gradient or hypertension or both conditions, was 5% and 1 patient underwent operative revision during the follow-up period. We have used the definition of hypertension of Macmanus and associates [91 as being greater than 2 standard deviations above the mean pressure for age. These results compare very favorably with other series in the literature [5, 7-9, 131. This fact has led us to evaluate our method of management. The advisability of operation in these patients is rarely open to question. Mortensen and associates [lo] reported a 64% mortality among 134 infants treated medically. Waldhausen and colleagues [151 reported a 75 to 100% medical mortality, depending on the presence or absence of associated intracardiac defects. Comparison of these findings with the operative result of this series and those previously quoted indicate that surgical treatment will be beneficial to this group of patients. We advocate early catheterization in patients who are seen with congestive heart failure and whose clinical examination suggests coarctation with or without associated congenital heart disease. Operation is performed immediately if the clinical impression is verified at catheterization. The neonatologists, cardiologists, and cardiovascular surgeons all participate in the rapid preoperative preparation of these patients. Intraoperative management depends largely on an open, ongoing dialogue between the surgeon and anesthesiologist to "prevent" situations in which profound volume changes, hypoxemia, or hypotension may occur. If these situations do arise, immediate measures are taken rapidly to correct the underlying problem. Postoperative care also is approached with a team concept and is usually delivered in a newborn intensive care setting where monitoring of the cardiac, respiratory, and metabolic status of each patient is most effectively carried out. We believe that each of these things has contributed importantly to the improved results indicated in our survival data. The contribution of associated congenital cardiac anomalies to mortality and morbidity in these clinical situations is unmistakable. Hallman and colleagues [7] reported a 93% incidence of associated anomalies, with patent ductus arteriosus (93%) and ventricular septal defect (41%) being the most frequent. Glass and associates [51 similarly reported a 49% incidence of anomalies other than patent ductus arteriosus, including ventricular septal defect (33%), transposition of the great vessels (loo/~), and atrial septal defect (6%). Our own series is comparable in this regard with an overall incidence of 90%. Patent ductus arteriosus (74%), atrial septal defect (74'/0), and ventricular septal defect (62%) were the most commonly found lesions (see Fig 4). The number and severity of these lesions are major determining factors in the early onset and intractability of congestive heart failure and of mortality. The 31 patients with coarctation and associated ventricular septal defect have presented a great management challenge due to the severity of congestive heart failure that prevails in this group. Nine (75%) of the deaths in our series occurred in this subgroup. It has been our practice to perform pulmonary artery banding in those patients who demonstrate a pulmonary artery pressure greater than two-thirds of systemic pressure at preoperative catheterization. Seventeen patients underwent coarctation repair in conjunction with pulmonary artery banding; there were 4 deaths, an operative mortality of 23%. Other groups report operative mortalities of 25 to 62% for this combined procedure [9,14]. Presently, support is growing for concomitant or subsequent early open repair of ventricular septal defect in this setting [8]. While we agree that improved results in this area are needed, we have not been dissatisfied enough with our own experience to proceed with open repair of ventricular septal defect in these patients. We have used Dacron patch graft aortoplasty since 1978. Reports by Reul [ll], Fleming [3],

278 The Annals of Thoracic Surgery Vol 30 No 3 September 1980 Sade [12], and their associates utilizing this technique have indicated favorable results. We have been impressed by the relative ease of performance and utility of the technique, particularly in situations involving hypoplasia of the aortic isthmus. There were no complications referable to the use of this procedure in our patient group. Comparison of the group repaired using resection and end-to-end anastomosis with the group having Dacron patch aortoplasty reveals no concrete evidence that would recommend one procedure over the other. Overall mortality was 26% for resection versus 22% for aortoplasty. The incidence of associated anomalies was essentially the same in each group. The follow-up period for our aortoplasty group is not yet long enough to allow us to draw a meaningful comparison concerning recurrence. It is our impression, however, that patch aortoplasty will reduce the incidence of a gradient across the repair if patients are exercised. There are several technical points concerning the technique of patch aortoplasty that deserve emphasis. In performing the aortotomy, the incision must be carried well into the subclavian artery and well beyond the site of coarctation on the descending aorta. These maneuvers will place the widest portion of the carefully tailored patch graft at the outlet of the distal transverse aortic arch and the coarctation segment, avoiding narrowing in both of these critical areas. Attention to this aspect of technique in the performance of aortoplasty can abolish pressure gradients even in instances of severe isthmus hypoplasia. The choice of prosthetic material is another point of controversy. We regard Gore-Tex as an attractive prosthetic but have been concerned about potential aneurysm formation, as reported by Campbell and colleagues [l] in 1976. We have chosen knitted Dacron for use in these patients because of its longer period of demonstrated safety and its ease of handling. Knitted Dacron is soft and pliable and conforms well to the neonate s delicate aorta. The method of subclavian angioplasty as reported by Waldhausen and Nahrwold [161 at our institution has been used sparingly for several reasons. It has been our impression that the vessel is often of inadequate width and length to allow proper repair. Also, we have been hesitant to use this technique when alternative techniques have proved to be adequate and we can avoid the possible complications of subclavian division, namely, negative effects on arm growth, possible subclavian steal syndrome, and Horner s syndrome. In addition, the procedure involves added operating time in situations where this can be ill-afforded. The lower incidence of residual blood pressure gradients (5%) and need for reoperation (2.6%) presented here compare very favorably with other published reports [5, 7-91. We think these results are based primarily on strict adherence to details of operative technique in performing either end-to-end or patch aortoplasty repair. Only further follow-up will determine whether these numbers hold true in the long term, particularly in the very young patients considered here. In summary, we have presented our experience with coarctation repair in the newborn less than 6 weeks old, a highly complicated patient group. Our results in terms of mortality and lowered recurrence have been gratifying. We believe that a concerted team approach to patient management, early operation, and strict adherence to careful operative technique are the keys to success with these infants. References 1. Campbell CD, Brooks DH, Webster MW, et al: Aneurysm formation in expanded polytetrafluoroethylene prosthesis. Surgery 79:491, 1976 2. Crafoord C, Nylin G: Congenital coarctation of the aorta and its surgical treatment. J Thorac Surg 14:347, 1945 3. Fleming WH, Sarafian LB, Clark EB, et al: Critical aortic coarctation: patch aortoplasty in infants less than age 3 months. Am J Cardiol44:687,1979 4. Freed M, Rocchini A, Rosenthal A, et al: Pathogenesis of exercise induced hypertension after coarctectomy. Circulation 55, 56:Suppl 3:173, 1977 5. Glass IH, Mustard WT, Keith JD: Coarctation of the aorta in infants: a review of twelve years experience. Pediatrics 26: 109, 1960 6. Gross RE: Technical considerations in surgical therapy for coarctation of the aorta. Surgery 20:1, 1946 7. Hallman GL, Yashar JJ, Bloodwell RD, Cooley DA: Surgical correction of coarctation of the aorta

279 Campbell et al: Newborns Undergoing Coarctation Repair in the first year of life. Ann Thorac Surg 4:106, 1967 8. Herrmann Vh4, Laks H, Fagan L, et al: Repair of aortic coarctation in the first year of life. Ann Thorac Surg 25:57, 1978 9. Macmanus Q, Starr A, Lambert LE, Grunkemeier G: Correction of aortic coarctation in neonates: mortality and late results. Ann Thorac Surg 24~544, 1977 10. Mortensen JD, Cutler PR, Rumel WR, et al: Management of coarctation of the aorta in infancy. J Thorac Cardiovasc Surg 37:502, 1969 11. Reul GJ, Kabbani SS, Sandiford FM, et al: Repair of coarctation of the thoracic aorta by patch graft aortoplasty. J Thorac Cardiovasc Surg 68:696, 1974 12. Sade RM, Taylor AB, Chariker EP: Aortoplasty compared with resection for coarctation of the aorta in young children. Ann Thorac Surg 28:346, 1979 13. Tawes RL, Aberdeen E, Waterston DJ, et al: Coarctation of the aorta in infants and children: a review of 333 operative cases, including 179 infants. Circulation 39, 4O:Suppl 1:173, 1969 14. Tiraboschi R, Alfieri 0, Carpentier A, Parenzan L: One stage correction of coarctation of the aorta associated with intracardiac defects in infancy. J Cardiovasc Surg (Torino) 19:11, 1978 15. Waldhausen JA, King H, Nahrwold DL, et al: Management of coarctation in infancy. JAMA 187:268, 1964 16. Waldhausen JA, Nahnvold DL: Repair of coarctation of the aorta with a subclavian flap. J Thorac Cardiovasc Surg 51:532, 1966 Discussion DR. ROBERT M. SADE (Charleston, SC): I congratulate the authors on a well-analyzed and well-presented series, as well as on their fine results in this difficult group of patients. I fundamentally agree with their techniques and with their conclusions, but do have a few technical disagreements. First, I don t think it is necessary to dissect the intercostal arteries. Doing so prolongs the operation a bit and increases the risk of injury to those vessels. Therefore, I don t dissect any intercostal arteries. We use Gore-Tex rather than Dacron for patching in small infants because Gore-Tex has a very thin neointima. In the small vessels of infants, this kind of neointima may contribute to maintenance of patency and less gradient late postoperatively. A microscopic view of a Gore-Tex patch recovered a year after implantation showed the thin neointima that can be measured in microns rather than millimeters, as is often seen with Dacron. The major problem with this paper was the methodology of follow-up, as I think the authors recognize. In all the patients, blood pressure was measured in the upper extremity and pedal pulses were palpated, but in only 45% was there measurement of both upper and lower extremity pressures. Furthermore, none of the patients had either cardiac catheterization or exercise testing to measure gradients. On the basis of these rather thin follow-up data, the authors were correct in saying that they cannot conclude that one kind of operation is better than another. We presented to this Society last year data that indicate clearly that patch aortoplasty, whether subclavian flap or Gore-Tex or Dacron, is superior to endto-end anastomosis. This was suggested, though not proved, by the data we just heard. For future follow-up of patients undergoing this repair, it will be inadequate simply to measure blood pressure or even to measure pressure differences between upper and lower extremities. We are going to have to do exercise testing, as suggested by Michael Freed at Boston Children s Hospital a couple of years ago. We have done exercise testing of our patients who underwent coarctation repair in the past couple of years and have found that many who have no resting pressure gradient and no hypertension, on exercise testing have very significant pressure gradients, sometimes in the range of 50 to 60 mm Hg. Significant hypertension also develops, indicating that they have residual narrowing at the coarctation site. This narrowing is simply not detectable by hemodynamic measurements taken during the resting state. I ask the authors whether they have any data gathered during exercise in their patients that allows comparison between patch grafting and end-to-end anastomosis. DR. WILLIAM H. FLEMING (Omaha, NE): I, too, congratulate the authors on their very excellent results and on a careful analysis. We have been using this technique for about five years, and it works. Out of 40 patients, 2 have died. One died of digitalis toxicity 5 days postoperatively and 1, of bowel perforation 6 weeks postoperatively. I want to point out several things we have found on longer follow-up. First, most of our patients have been recatheterized now, and they do not have gradients at the repair site. A few of them will have gradients on the hypoplastic isthmic portion of the aorta. Second, in infants with intact ventricular septum and critical coarctation, there may be absolutely dreadful left ventricular function. Don t be deterred by it. Talk to the anesthesiologist, go carefully, and keep following the ejection fraction with echo. It is amazing the recovery those ventricles will show in just a few days. By the time the child is ready to go home, they will be back to normal ejection fraction. I was delightfully startled by that. Third, 2 patients now have calcified patches. These patients were catheterized a second time and have

280 The Annals of Thoracic Surgery Vol 30 No 3 September 1980 perfectly normal flow. There is no problem from the calcification that we know of, but the patches are calcified on plain roentgenogram. I don t know where that leads us. Finally, when it is time to close the ventricular septal defects in the patients with this accompanying condition, be very sure to have a good definition of the anatomy; at least a quarter of the children will have an unusual ventricular septal defect, for example, muscular or multiple. DR. BROWN: I thank the discussants. Also, I will update our data by reporting that 10 more neonates have undergone repair without mortality at our institution since we submitted the abstract. In response to Dr. Sade, we do not sacrifice intercostal arteries but we do control the first pair in order to expose a very large segment of aorta. We are aware of Dr. Sade s clamp and look forward to seeing it and possibly using it in the near future. We also are attracted to Gore-Tex as a graft material but await others experience before we undertake its use as our routine prosthetic material. We are aware of the current inadequacies of our follow-up data in newborns following coarctation repair. currently, we and our cardiologists are engaged in an extensive evaluation of this series and are doing exercise testing on all of these patients. We are encouraged by Dr. Fleming s long-term follow-up of Dacron patch angioplasty. This has stimulated us to continue our current patching technique. We have not yet seen calcification of Dacron patches, but, again, our follow-up is relatively short. Our experience with closure of ventricular septal defect in patients less than 6 weeks old has been very limited. Published results of closure in this very young age group deter us from undertaking primary closure as a routine practice at the present time.