- Email: [email protected]
Ease of correction in early infancy
J
THORAC CARDIOVASC SURG
85:388-395, 1983
Mitral repair in complete atrioventricular canal Ease of correction in early infancy During the past 5 years all patients with complete atrioventricular (AV) canal were subjected to surgical correction regardless of age. Thus we were able to assess the ease and reliability of mitral repair in early infancy. Key features of operation include division of bridging leaflets when indicated, selection of a patch sufficiently small to prevent postoperative annular dilatation, attachment of the leaflets to the patch with continuous nonpledget-supported sutures at a level determined by the chordal structure, and, based on the size of the mural leaflet, construction of a bicuspid or tricuspid mitral valve. Operative mortality and postoperative morbidity were no greater in patients under 6 months of age than in older patients. Reoperation for mitral regurgitation was required in three patients. In all of them, the primary reason for the failure of repair was that the mitral valve had been left tricuspid; in two of them, the mitral anulus was dilated, as well. Repair was uniformly accomplished by bicuspidization, combined in two instances with a Wooler-type annuloplasty and shortening of the free margin of the aortic leaflet. The absence of secondary scarring of the leaflets in the patients under 6 months of age facilitated repair in this age group. Early repair is possible, and preferable, provided that meticulous attention is paid to the technical features of the repair and a proper decision is made with regard to treatment of the mitral" cleft."
Pietro A. Abbruzzese, M.D., Joanie Livermore, B.S., Cecille O. Sunderland, M.D., David L. Nunley, M.D., Henry Issenberg, M.D., Siavosh Khonsari, M.D., Carlos E. Garcia, M.D., and Albert Starr, M.D., Portland, Ore.
Results with corrective operations for complete atrioventricular (AV) canal malformations have steadily improved during the past decade, and increasingly younger children have undergone successful correction. Long-term follow-up has shown late mitral regurgitation to be a major problem.l" We have reviewed the experience of The Oregon Health Sciences University in an attempt to assess the ease, safety, and reliability of our valvular repair. Patients
Since 1977, 34 consecutive patients with complete AV canal and without other major intracardiac or extracardiac anomalies underwent total correction. All deFrom the Divisions of Cardiac Surgery, Pediatric Cardiology, and Medical Graphics, Oregon Health Sciences University, Portland, Ore. 97201. Read at the Eighth Annual Meeting of The Samson Thoracic Surgical Society, San Diego, Calif., June 23-27, 1982. Address for reprints: Albert Starr, M.D., Division of Cardiopulmonary Surgery, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, Ore. 97201.
388
Fig. 1. Cold saline is injected under pressure into the ventricles for valvular assessment. fects were of the balanced form," i.e., without hypoplasia of either ventricle. There were 13 boys and 21 girls, with ages ranging from 2 to 69 months (median 7 months, mean 13 months) and weights ranging from 3.5 to 19.8 kg (me-
Volume 85
Mitral repair in complete AV canal
Number 3 March,1983
389
t/2" ~\.V. Fig. 2. The anterior and posterior leaflets are divided. Fig. 3. The base of the mitral' 'cleft" is reapproximated to allow further assessment of the valvular geometry.
Fig. 4. A single Dacron patch is carefully tailored to tightly fit the septal defect. Its width at the level of resuspension of the leaflets is carefully assessed (see text). Fig. 5. The patch is sutured to the ventricular component of the defect with fine continuous monofilament material. Shallow bites are taken on the right side of the posterior septum to avoid injury to the conduction system (see text). Fig. 6. Leaflet reconstruction is started. The left and right sides of the divided common leaflets are resuspended to the patch at a level indicated by the chordal attachment (when present) and by the previous assessment (Fig. 1).
dian 4.8 kg, mean 6.2 kg). Thirty-one patients had trisomy 21, one had 47,XXX syndrome, and two had nq chromosomal abnormalities. One patient, the oldest in the series, had undergone a previous pulmonary artery banding at 5 months of age. Associated minor cardiac anomalies are as follows: secundum atrial septal defect, nine patients; patent ductus arteriosus, six patients; and a left superior vena cava draining into the coronary sinus, one patient. Severe congestive heart failure was present in 13
patients and was refractory to medical management in seven. Seven patients had failure to thrive and frequent upper respiratory tract infections. Twelve patients were operated upon an elective basis, but all had evidence of a large left-to-right shunt. Preoperative evaluation included left and right heart catheterization at this institution within 1 month before the operation. All patients, except the one with the pulmonary artery banding, had systemic or near systemic pulmonary artery pressure. The hemodynamic
3 9 0 Abbruzzese et al.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 7. Fine, continuous, horizontal braided polyester mattress sutures, taken well away from the cut edges with a stapling technique (see text) are used for resuspension of the leaflets.
Fig. 8. The mitral "cleft" is completely closed in those patients in whom the mural leaflet inserts to at least one third of the circumference of the anulus. Fine interrupted braided polyester sutures are used.
calculations were based on an assumed oxygen consumption. The pulmonary vascular resistance (PVR) ranged from 1.4 to 10 WU . m2 * (mean 3.8 ± 2.1 WU . m-). Mitral regurgitation of a significant degree was present in nine patients. Profound hypothermia with surface cooling, partial cardiopulmonary bypass, and circulatory arrest was used in 28 operations. Six patients, who weighed more than 8.0 kg, underwent conventional cardiopulmonary bypass with moderate hypothermia. Cold hyperkalernic cardioplegia was used in all patients. Details of the operative technique" are illustrated in Figs. 1 to 10 and described in their legends. *Wood units = (mean pulmonary artery pressure - mean left atrial pressure)/cardiac output.
Assessment of the anatomy and physiology of the valvular apparatus with cold saline injected under pressure into the ventricles was a first, important step in the procedure (Fig. 1). Attention was given to the width of the patch at the level of resuspension of the leaflets, to provide an adequate diameter of the newly constructed mitral anulus (Fig. 4). The conduction system was avoided by taking shallow bites in the posterior aspect of the ventricular component of the defect (Fig. 5) and in the area of the coronary sinus (Fig. 10). The cut edges of the divided common leaflets were resuspended to the patch without the use of pledgets by placement of large horizontal mattress sutures very close together on the mitral side, well away from the edges, to provide secure coaptation to the patch (Fig. 7).
Volume 85 Number 3
Mitral repair in complete A V canal
March, 1983
39 I
Fig. 9. The "cleft" is left partially or completely open in those patients in whom the mural leaflet inserts to less than one third of the circumference of the anulus.
15
14 Pts. 14 1±91 % 11 Pts. 181±12)%
TOTAL 34 Patients 6 Deaths 18 (±7%)
10 Number of Patients
5 Pts. 40 1±221 %
5
4 ·Pts. 0%
2-5
6-12 13-24 AGE (months)
>24
Fig. 11. Operative mortality by age at operation. Fig. 10. The atrial component of the defect is closed with the same fine continuous monofilament sutures. The coronary sinus is left in the right atrium and injury to the atrioventricular node is carefully avoided by taking shallow bites in that area.
The mitral "cleft" was managed according to the size of the mural leaflet. It was completely closed in 23 patients in whom the mural leaflet inserted to at least one third of the circumference of the anulus (Fig. 8). In the remaining 11 patients the mural leaflet was less developed, with an insertion line that comprised less than one third of the circumference of the anulus. In these patients the "cleft" was left partially open (six cases) or completely (five cases) open (Fig. 9). In all cases the mitral valve was checked for competence by saline injection after the repair. After closure of the atrium, pressure lines were left in the pulmonary artery and in the left atrium for postoperative monitoring.
Results Early results. There were six early deaths (18%), two of which occurred in patients less than 6 months of age (Fig. 11 and Table I). No statistical difference in mortality was found among the various age groups. Death was attributed to intractable right ventricular failure in three patients who had a high PVR preoperatively (5.7,6.3, and 9.0 WU . m-) and in one patient who had an abrupt elevation of the PVR postoperatively. Three of them underwent autopsy, which showed an intact valvular repair and Heath-Edwards" Grade II to IV changes in the pulmonary vascular bed. Both of the remaining nonsurvivors died of massive hemorrhage, which occurred after removal of a pulmonary artery line in one and through a tom aortic cannulation site in the other. Five major, nonfatal complications occurred as shown in Table II. Both cases of complete A V block occurred in the older age groups. Detachment of the
The Journal of
3 9 2 Abbruzzese et al.
Thoracic and Cardiovascular Surgery
Table I. Operative mortality and postoperative morbidity according to age range
1------.----Age ranges
No. of patients Operative deaths Complete A V block Residual VSD Mitral regurgitation
2-5 rna
6-69 rna
14
20 4
2
o o I
2
o 2
Legend: A V, Atrioventricular. VSD. Ventricular septal defect.
Table II. Postoperative complications Complete A V block Mediastinitis Transient mechanical hemolysis Early repair dehiscence
2 I I I
Legend: AV. Atrioventricular.
newly constructed anterior mitral leaflet occurred in an l S-month-old child on postoperative day 8. Reoperation was performed immediately with reattachment of the mitral leaflet to the patch, again by means of our standard technique. Recovery was uneventful. Late results (Table III). Two late deaths occurred, both unrelated to the repair. A 34-month-old girl, who had the highest PVR in our series, died suddenly at home 1.3 months after the operation. Her postoperative course had been clinically uneventful, with the pulmonary artery pressure being higher than systemic levels. An autopsy showed intact intracardiac repair and Grade IV pulmonary vascular changes. An 18-month-old child died of pneumonia at home 4 months after the operation. Clinical follow-up of from 3 to 43 months (mean 17.5 months, median 13 months) is available in all 26 long-term survivors. Twenty-one are considered to have good or excellent results, as defined by normal growth, no significant murmur, and normal or significantly decreased heart size on chest roentgenogram. The result in one patient is considered fair, with evidence of at least moderate mitral regurgitation and persistence of cardiac enlargement. The child is awaiting catheterization. The remaining four patients underwent repeat cardiac catheterization 3, 18, 18, and 26 months postoperatively. One patient who was failing to thrive and had a loud holosytolic murmur was found to have no significant hemodynamic residua and is included in the group with "hemodynamically satisfactory results." The other three underwent reoperation for late mitral
Table III. Late results (28 patients) Late deaths Hemodynamically unsatisfactory results Reoperations for mitral regurgitation Awaiting catheterization Hemodynamically satisfactory results
2 4
3 I
22
regurgitation 18 to 26 months after the original operation, as described below and shown in Figs. 12 to 14. Reoperations. In all three cases standard cardiopulmonary bypass, moderate hypothermia and cold crystalloid cardioplegia were used. The left atrium was uniformly opened parallel to the interatrial groove. The first patient had been operated upon at the age of 6 months and the mitral valve had been left trileaflet. No mitral regurgitation was present preoperatively. At reoperation the open mitral "cleft" was found to be the cause of the mitral regurgitation. The mural leaflet appeared to insert to at least one third of the circumference of the anu1us. Closure of the "cleft" produced an excellent result, as shown by the dramatic decrease in heart size on chest films and the absence of a murmur 10 months postoperatively (Fig. 12). The second patient had undergone his first operation at the age of 8 months, and the mitral "cleft" had been reported to be completely closed. Moderate mitral regurgitation was present preoperatively. At reoperation the "cleft" was found to be only partially closed, and the anterior mitral leaflet was redundant and had elongated chordae. The residual "cleft" was closed, the free margin of the leafiet was plicated, and the chordae were shortened. Persistent mitral regurgitation led to a new operation 7 days later. The mitral anu1us was found to be enlarged and a Wooler-type annuloplasty was added to the repair. A murmur persists 7 months postoperatively, and the heart is smaller but still enlarged on the chest film (Fig. 13). The third patient had been operated upon at the age of 5 months, and the "cleft" had been left partially open. No mitral regurgitation was present preoperatively. At reoperation the mitral anulus was moderately enlarged. The residual "cleft" was closed and a Wooler-type annuloplasty was performed. Three months postoperatively a murmur persists, and the heart is smaller but still enlarged on chest roentgenography (Fig. 14).
Discussion Rastelli and associates? first described an effective operation for total repair of complete AV canal malformations in 1968. Since then, increasingly lower mor-
Volume 85 Number 3
Mitral repair in complete AV canal
March,1983
Fig. 12. Patient 1. View from left atriotomy. Left. Intraoperative anatomy: The "cleft" is completely open. The mural leaflet insertsto approximately one third of the annular circumference. Right, Repair. The "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. tality rates have been reported in all age groups.v 8-20 However, hemodynamically satisfactory repair remains a formidable challenge because of the variations in the anatomy and geometry of the valvular apparatus. 4.21-23 Extensive anatomic studies 24- 27 have led some groups to a new type of repair, 27 the primary purpose of which is to decrease the incidence of repair dehiscence and to avoid bicuspidization in an effort to improve mitral function. Our group has employed a modification of the Mayo Clinic techniqueP- 7 with special emphasis on the following technical aspects: The use of a single patch, which implies division of the common leaflets, when needed, allows excellent exposure of the ventricular portion of the septal defect and may account for the absence of residual shunts in our series. Adequate sizing of the patch is essential. Too large a patch can cause annular dilatation, as might have occurred in two of our three cases of late mitral regurgitation. On the other hand, too small a patch may cause redundancy of the reconstructed anterior leaflet of the mitral valve, and may necessitate plication of its free margin. Use of fine continuous horizontal mattress sutures to reattach the left and right sides of the common leaflets to the patch provides a simple and reliable repair. Closely spaced bites on the mitral side, well away from the cut edges;" resulted in only one case of repair dehiscence (3%), which occurred early in our experience. We did not find it necessary to use pledgers" 10 or a double patch technique 19, 20, 27 to ensure adequate strength to the suture line. Especially in infants, pledgets may interfere with valvular function. Moreover, even in infants, we always had enough valvular tissue left to reconstruct the AV valves adequately. 20 Our decision whether to construct a bileaflet or trileaflet mitral valve is based on the size of the mural leaflet. The trileaflet repair is reserved for those cases in
39 3
Fig. 13. Patient 2. View from left atriotomy. Left, Intraoperative anatomy: The "cleft" is partially open. The anterior mitral leaflet is redundant and has elongated chordae. The anulus is enlarged. Right, Final repair (see text): The residual "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. The free margin of the anterior leaflet has been plicatedand the chordae have been shortened. One 3-0 figure-of-eight braided polyester suture has been used at the level of each commissure for the Wooler-type annuloplasty.
.•
-.-
~ .F,':
:": -
tt.,
'!Ii
"
""
Fig. 14. Patient 3. View from left atriotomy. Left, Intraoperative anatomy: The "cleft" is partially open. The mural leaflet inserts to one third of the circumference of the anulus. The anulus is enlarged. Right, Repair: The residual "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. One 3-0 figure-of-eight braided polyester suture has been used at the level of each commissure for the Wooler-type annuloplasty. which the insertion of the mural leaflet at the anulus is less than one third of the annular circumference, in which case closure of the "cleft" would restrict the motion of the "anterior" leaflet. 5, 7 The uniform finding of regurgitation through the open "cleft" in all three late reoperations might be related to our lack of recognition that the "cleft" is a commissure'" and should be treated as such. Therefore, invagination of the base of the commissure may provide better coaptation of the two leaflets and further decrease the incidence of late mitral regurgitation. Early correction was frequently performed in our series, and we have reached the following conclusions: 1. Repair is as safe in infants less than 6 months of age as it is in older age groups. This observation justifies early intervention to alter the poor natural history
The Journal of
394 Abbruzzese et al.
of the malformation, which shows a high early mortality'" and rapidly occurring irreversible pulmonary vascular changes.i" In fact, in our series PVR was again''? found to be a determinant of surgical results. If deaths resulting from technical (bleeding) or unrelated (pneumonia) causes are excluded, four of nine patients (45%) with a PVR greater than 5 WU . m2 and one of 22 patients (4.5%) with a PVR less than 5 WU . m2 died after the operation (p = 0.017 by Fisher's exact test). 2. Repair is easy in the youngest age group. Profound hypothermia with partial cardiopulmonary bypass and circulatory arrest, cold crystalloid cardioplegia, and the absence of venous cannulas allowed perfect exposure and satisfactory assessment of the anatomy. The lack of acquired valvular changes, such as scarring of the leaflets and annular dilatation, facilitated the repair and may also be responsible for the low incidence (26%) or preoperative mitral regurgitation. 3. Repair of the valvular structures is reliable in early infancy. There was only one case (7%) of severe late mitral regurgitation which necessitated reoperation in the youngest age group, as compared to a similar percentage (10%) in the older age groups.
2
3 4 5
6
7
8
9
REFERENCES Espulgas E, Olley PM, Kidd BSL, Trusler GA: Hemodynamic results following repair of complete atrioventricular canal defects, The Child With Congenital Heart Disease After Surgery, BSL Kidd, RD Rowe, eds., Mount Kisco, 1976, Futura Publishing Co., Inc., pp 25-34 McCabe rc, Engle MA, Gay WA lr, Ebert PA: Surgical treatment of endocardial cushion defects. Am 1 Cardiol 39: 72-77, 1977 McGoon DC, Puga Fl: Atrioventricular canal. Cardiovasc Clin 11:31 I-322, 1981 Bharati S, Lev M: The spectrum of common atrioventricular orifice (canal). Am Heart 1 86:553-561, 1973 Harlan Bl, Starr A, Harwin FM: Complete atrioventricular canal, Manual of Cardiac Surgery, Vol II, New York, 1981, Springer-Verlag, pp 269-275 Heath D, Edwards IE: The pathology of hypertensive pulmonary vascular disease. Circulation 18:533-547, 1958 Rastelli GC, Ongley PA, Kirklin lW, McGoon DC: Surgical repair of the complete form of persistent common atrioventricular canal. 1 THoRAc CARDIOVASC SURG 55:299-308, 1968 Alfieri 0, Subramanian S: Successful repair of complete atrioventricular canal with undivided anterior common leafiet in a 6 month old infant. Ann Thorac Surg 19:9296, 1975 Hardesty RL, Zuberbuhler lR, Bahnson HT: Surgical
Thoracic and Cardiovascular Surgery
10
11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
treatment of atrioventricular canal defect. Arch Surg 110:1391-1396, 1975 Mills NL, Ochsner Jl., King TD: Correction of type C complete atrioventricular canal. Surgical considerations. 1 THoRAc CARDIOVASC SURG 71:20-28, 1976 Kahn DR, Levy 1, France NE, Chung Kl, Dacumos GC: Recent results after repair of atrioventricular canal. 1 THoRAc CARDIOVASC SURG 73:413-415, 1977 Mair DD, McGoon DC: Surgical correction of atrioventricular canal during the first year of life. Am 1 Cardiol 40:66-69, 1977 Berger TJ, Kirklin JW, Blackstone EH, Pacifico AD, Kouchoukos NT: Primary repair of complete atrioventricular canal in patients less than 2 years old. Am 1 CardioI41:906-913, 1978 Berger TJ, Blackstone EH, Kirklin lW, Bargeron LM, Hazelrig IB, Turner ME: Survival and probability of cure without and with operation in complete atrioventricular canal. Ann Thorac Surg 27: 104-111, 1979 Foglia RP, Mulder DG: Surgical correction of complete atrioventricular canal. Am 1 Surg 136: 121-127, 1978 Culpepper W, Kolff 1, Lin CW, Vitullo D, Lamberti 1, Arcilla RA, Replogle R: Complete common atrioventricular canal in infancy. Surgical repair and postoperative hemodynamics. Circulation 58:550-558, 1978 Stewart S, Harris P, Manning 1: Complete endocardial cushion defect. Operative technique and results. 1 THoRAc CARDIOVASC SURG 78:914-919, 1979 Midgley M, Galioto FM, Shapiro SR, Perry LW, Scott LP: Experience with repair of complete atrioventricular canal. Ann Thorac Surg 30: 151-159, 1980 Kirklin JW, Chung G, Pacifico AD, Blackstone EH, Bargeron M: Repair of AV canal defects, Congenital Heart Disease in the First Three Months of Life: Medical and Surgical Aspects, L Parenzan, G Crupi, G Graham, eds., Bologna, 1981, Casa Editrice Patron SAS, pp 431437 Mavroudis C, Weinstein G, Turley K, Ebert PA: Surgical management of complete atrioventricular canal. 1 THoRAc CARDIOVASC SURG 83:670-679, 1982 Rastelli GC, Kirklin lW, Titus lL: Anatomic observations on complete form of persistent common atrioventricular canal with special reference to atrioventricular valves. Mayo Clin Proc 41:296-308, 1966 Goor D, Lillehei CW, Edwards IE: Further observations on the pathology of the atrioventricular canal malformation. Arch Surg 97:954-962, 1968 Ebert PA, Goor DA: Complete atrioventricular canal malformation: Further clarification of the anatomy of the common leafiet and its relationship to the VSD in surgical correction. Ann Thorac Surg 25: 134-143, 1978 Ugarte M, Enriquez de Salamanca F, Quero M: Endocardial cushion defects. An anatomical study of 54 specimens. Br Heart 1 38:674-682, 1976 Piccoli GP, Gerlis LM, Wilkinson lL, Lozsadi K, Macartney Fl, Anderson RH: Morphology and classifica-
Volume 85 Number 3 March,1983
26
27
28
29
tion of atrioventricular defects, Br Heart J 42:621-632, 1979 Piccoli GP, Wilkinson JL, Macartney FJ, Gerlis LM, Anderson RH: Morphology and classification of complete atrioventricular defects, Br Heart J 42:633-639, 1979 Carpentier A: Surgical anatomy and management of the mitral component of atrioventricular canal defects, Paediatric Cardiology 1977, RH Anderson, EA Shinebourne, eds., Edinburgh, Churchill Livingstone Katz NM, Blackstone EH, Kirklin JW, Bradley EL, Lemons JE: Suture techniques for atrioventricular valves. Experimental study. J THoRAc CARDIOVASC SURG 81:528536, 1981 Newfeld EA, Sher M, Paul MH, Nikaidoh H: Pulmonary vascular disease in complete atrioventricular canal defect. Am J Cardiol 39:721-726, 1977
Discussion DR. LEONARD L. BAILEY Loma Linda. Calif.
The technical features for repair of A V canal in infancy have been appropriately stressed in this report. The Oregon group's attention to detail both during and after operation is obvious. I was particularly pleased with their honest assessment of the mitral valve apparatus as a trileaflet structure. I have not encountered a case in which I felt biscuspidization of the mitral apparatus would produce stenosis. For comparison, I have reviewed our smaller series of 11 consecutive patients ranging in age from 7 days to 24 months who have had repair of complete AV canal since mid-1976. Seventy-three percent of the patients had the Down syndrome. All had severe persistent cardiac failure and marked growth delay. Weight ranged from 2.2 to 10 kg. Five (45%) patients were under 3 months of age including 2 neonates. The two newborn infants were free of genetic disorders but had complex associated intracardiac malforrnations-doubleoutlet right ventricle (DORV) and pulmonary venous obstruction in one and dextro-transposition of the great arteries
Mitral repair in complete AV canal
395
(d-TGA) and tracheoesophageal fistula (TEF) in the other. Both required preoperative ventilatory support and relatively urgent surgical intervention. Complete repair was accomplished in both. The baby with DOR V continues to do well three years postoperatively. The neonate with d-TGA and TEF died of respiratory insufficiency produced by an untreated tracheobronchial mucus cast. Complete repair was accomplished in all 11 infants in our series. Repairs were of the Mayo Clinic-type: The common leaflets were divided when indicated (type C) and a twoleaflet mitral valve apparatus was created. Pledgets were not used to mount valve tissue to the prosthetic patch. Pericardial pledgets and simple running 7-0 monofilament suture were used to repair the mitral valve cleft. There have been three operative deaths in our series for an overall mortality of 27%. However, there has been only one death (the neonate with d-TGA and TEF) in the last seven consecutive infants treated since early 1977. There have been no late cardiac deaths and no instances of heart block or residual intracardiac shunts. Significant mitral valve regurgitation exists in only one patient who, 3 months postoperatively, still requires cardiac medication although his growth and activity level have dramatically increased. The authors have presented data which, when combined with other recent reports, appear to establish the validity of complete correction of AV canal as a reasonable alternative to pulmonary artery banding or protracted medical management in symptomatic infants of any age or weight. I am convinced, however, that this type of operation should be limited to centers quite competent with corrective cardiac procedures in early infancy. I have no disagreement with the data or techniques presented by the authors, and I thank them for clarifying the issue of a trileaflet mitral valve. DR. A B B R U Z Z ESE (Closing) I thank Dr. Bailey for participating in the discussion, and I congratulate him for his excellent results, especially his late results.
THORAC CARDIOVASC SURG
85:388-395, 1983
Mitral repair in complete atrioventricular canal Ease of correction in early infancy During the past 5 years all patients with complete atrioventricular (AV) canal were subjected to surgical correction regardless of age. Thus we were able to assess the ease and reliability of mitral repair in early infancy. Key features of operation include division of bridging leaflets when indicated, selection of a patch sufficiently small to prevent postoperative annular dilatation, attachment of the leaflets to the patch with continuous nonpledget-supported sutures at a level determined by the chordal structure, and, based on the size of the mural leaflet, construction of a bicuspid or tricuspid mitral valve. Operative mortality and postoperative morbidity were no greater in patients under 6 months of age than in older patients. Reoperation for mitral regurgitation was required in three patients. In all of them, the primary reason for the failure of repair was that the mitral valve had been left tricuspid; in two of them, the mitral anulus was dilated, as well. Repair was uniformly accomplished by bicuspidization, combined in two instances with a Wooler-type annuloplasty and shortening of the free margin of the aortic leaflet. The absence of secondary scarring of the leaflets in the patients under 6 months of age facilitated repair in this age group. Early repair is possible, and preferable, provided that meticulous attention is paid to the technical features of the repair and a proper decision is made with regard to treatment of the mitral" cleft."
Pietro A. Abbruzzese, M.D., Joanie Livermore, B.S., Cecille O. Sunderland, M.D., David L. Nunley, M.D., Henry Issenberg, M.D., Siavosh Khonsari, M.D., Carlos E. Garcia, M.D., and Albert Starr, M.D., Portland, Ore.
Results with corrective operations for complete atrioventricular (AV) canal malformations have steadily improved during the past decade, and increasingly younger children have undergone successful correction. Long-term follow-up has shown late mitral regurgitation to be a major problem.l" We have reviewed the experience of The Oregon Health Sciences University in an attempt to assess the ease, safety, and reliability of our valvular repair. Patients
Since 1977, 34 consecutive patients with complete AV canal and without other major intracardiac or extracardiac anomalies underwent total correction. All deFrom the Divisions of Cardiac Surgery, Pediatric Cardiology, and Medical Graphics, Oregon Health Sciences University, Portland, Ore. 97201. Read at the Eighth Annual Meeting of The Samson Thoracic Surgical Society, San Diego, Calif., June 23-27, 1982. Address for reprints: Albert Starr, M.D., Division of Cardiopulmonary Surgery, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, Ore. 97201.
388
Fig. 1. Cold saline is injected under pressure into the ventricles for valvular assessment. fects were of the balanced form," i.e., without hypoplasia of either ventricle. There were 13 boys and 21 girls, with ages ranging from 2 to 69 months (median 7 months, mean 13 months) and weights ranging from 3.5 to 19.8 kg (me-
Volume 85
Mitral repair in complete AV canal
Number 3 March,1983
389
t/2" ~\.V. Fig. 2. The anterior and posterior leaflets are divided. Fig. 3. The base of the mitral' 'cleft" is reapproximated to allow further assessment of the valvular geometry.
Fig. 4. A single Dacron patch is carefully tailored to tightly fit the septal defect. Its width at the level of resuspension of the leaflets is carefully assessed (see text). Fig. 5. The patch is sutured to the ventricular component of the defect with fine continuous monofilament material. Shallow bites are taken on the right side of the posterior septum to avoid injury to the conduction system (see text). Fig. 6. Leaflet reconstruction is started. The left and right sides of the divided common leaflets are resuspended to the patch at a level indicated by the chordal attachment (when present) and by the previous assessment (Fig. 1).
dian 4.8 kg, mean 6.2 kg). Thirty-one patients had trisomy 21, one had 47,XXX syndrome, and two had nq chromosomal abnormalities. One patient, the oldest in the series, had undergone a previous pulmonary artery banding at 5 months of age. Associated minor cardiac anomalies are as follows: secundum atrial septal defect, nine patients; patent ductus arteriosus, six patients; and a left superior vena cava draining into the coronary sinus, one patient. Severe congestive heart failure was present in 13
patients and was refractory to medical management in seven. Seven patients had failure to thrive and frequent upper respiratory tract infections. Twelve patients were operated upon an elective basis, but all had evidence of a large left-to-right shunt. Preoperative evaluation included left and right heart catheterization at this institution within 1 month before the operation. All patients, except the one with the pulmonary artery banding, had systemic or near systemic pulmonary artery pressure. The hemodynamic
3 9 0 Abbruzzese et al.
The Journal of Thoracic and Cardiovascular Surgery
Fig. 7. Fine, continuous, horizontal braided polyester mattress sutures, taken well away from the cut edges with a stapling technique (see text) are used for resuspension of the leaflets.
Fig. 8. The mitral "cleft" is completely closed in those patients in whom the mural leaflet inserts to at least one third of the circumference of the anulus. Fine interrupted braided polyester sutures are used.
calculations were based on an assumed oxygen consumption. The pulmonary vascular resistance (PVR) ranged from 1.4 to 10 WU . m2 * (mean 3.8 ± 2.1 WU . m-). Mitral regurgitation of a significant degree was present in nine patients. Profound hypothermia with surface cooling, partial cardiopulmonary bypass, and circulatory arrest was used in 28 operations. Six patients, who weighed more than 8.0 kg, underwent conventional cardiopulmonary bypass with moderate hypothermia. Cold hyperkalernic cardioplegia was used in all patients. Details of the operative technique" are illustrated in Figs. 1 to 10 and described in their legends. *Wood units = (mean pulmonary artery pressure - mean left atrial pressure)/cardiac output.
Assessment of the anatomy and physiology of the valvular apparatus with cold saline injected under pressure into the ventricles was a first, important step in the procedure (Fig. 1). Attention was given to the width of the patch at the level of resuspension of the leaflets, to provide an adequate diameter of the newly constructed mitral anulus (Fig. 4). The conduction system was avoided by taking shallow bites in the posterior aspect of the ventricular component of the defect (Fig. 5) and in the area of the coronary sinus (Fig. 10). The cut edges of the divided common leaflets were resuspended to the patch without the use of pledgets by placement of large horizontal mattress sutures very close together on the mitral side, well away from the edges, to provide secure coaptation to the patch (Fig. 7).
Volume 85 Number 3
Mitral repair in complete A V canal
March, 1983
39 I
Fig. 9. The "cleft" is left partially or completely open in those patients in whom the mural leaflet inserts to less than one third of the circumference of the anulus.
15
14 Pts. 14 1±91 % 11 Pts. 181±12)%
TOTAL 34 Patients 6 Deaths 18 (±7%)
10 Number of Patients
5 Pts. 40 1±221 %
5
4 ·Pts. 0%
2-5
6-12 13-24 AGE (months)
>24
Fig. 11. Operative mortality by age at operation. Fig. 10. The atrial component of the defect is closed with the same fine continuous monofilament sutures. The coronary sinus is left in the right atrium and injury to the atrioventricular node is carefully avoided by taking shallow bites in that area.
The mitral "cleft" was managed according to the size of the mural leaflet. It was completely closed in 23 patients in whom the mural leaflet inserted to at least one third of the circumference of the anulus (Fig. 8). In the remaining 11 patients the mural leaflet was less developed, with an insertion line that comprised less than one third of the circumference of the anulus. In these patients the "cleft" was left partially open (six cases) or completely (five cases) open (Fig. 9). In all cases the mitral valve was checked for competence by saline injection after the repair. After closure of the atrium, pressure lines were left in the pulmonary artery and in the left atrium for postoperative monitoring.
Results Early results. There were six early deaths (18%), two of which occurred in patients less than 6 months of age (Fig. 11 and Table I). No statistical difference in mortality was found among the various age groups. Death was attributed to intractable right ventricular failure in three patients who had a high PVR preoperatively (5.7,6.3, and 9.0 WU . m-) and in one patient who had an abrupt elevation of the PVR postoperatively. Three of them underwent autopsy, which showed an intact valvular repair and Heath-Edwards" Grade II to IV changes in the pulmonary vascular bed. Both of the remaining nonsurvivors died of massive hemorrhage, which occurred after removal of a pulmonary artery line in one and through a tom aortic cannulation site in the other. Five major, nonfatal complications occurred as shown in Table II. Both cases of complete A V block occurred in the older age groups. Detachment of the
The Journal of
3 9 2 Abbruzzese et al.
Thoracic and Cardiovascular Surgery
Table I. Operative mortality and postoperative morbidity according to age range
1------.----Age ranges
No. of patients Operative deaths Complete A V block Residual VSD Mitral regurgitation
2-5 rna
6-69 rna
14
20 4
2
o o I
2
o 2
Legend: A V, Atrioventricular. VSD. Ventricular septal defect.
Table II. Postoperative complications Complete A V block Mediastinitis Transient mechanical hemolysis Early repair dehiscence
2 I I I
Legend: AV. Atrioventricular.
newly constructed anterior mitral leaflet occurred in an l S-month-old child on postoperative day 8. Reoperation was performed immediately with reattachment of the mitral leaflet to the patch, again by means of our standard technique. Recovery was uneventful. Late results (Table III). Two late deaths occurred, both unrelated to the repair. A 34-month-old girl, who had the highest PVR in our series, died suddenly at home 1.3 months after the operation. Her postoperative course had been clinically uneventful, with the pulmonary artery pressure being higher than systemic levels. An autopsy showed intact intracardiac repair and Grade IV pulmonary vascular changes. An 18-month-old child died of pneumonia at home 4 months after the operation. Clinical follow-up of from 3 to 43 months (mean 17.5 months, median 13 months) is available in all 26 long-term survivors. Twenty-one are considered to have good or excellent results, as defined by normal growth, no significant murmur, and normal or significantly decreased heart size on chest roentgenogram. The result in one patient is considered fair, with evidence of at least moderate mitral regurgitation and persistence of cardiac enlargement. The child is awaiting catheterization. The remaining four patients underwent repeat cardiac catheterization 3, 18, 18, and 26 months postoperatively. One patient who was failing to thrive and had a loud holosytolic murmur was found to have no significant hemodynamic residua and is included in the group with "hemodynamically satisfactory results." The other three underwent reoperation for late mitral
Table III. Late results (28 patients) Late deaths Hemodynamically unsatisfactory results Reoperations for mitral regurgitation Awaiting catheterization Hemodynamically satisfactory results
2 4
3 I
22
regurgitation 18 to 26 months after the original operation, as described below and shown in Figs. 12 to 14. Reoperations. In all three cases standard cardiopulmonary bypass, moderate hypothermia and cold crystalloid cardioplegia were used. The left atrium was uniformly opened parallel to the interatrial groove. The first patient had been operated upon at the age of 6 months and the mitral valve had been left trileaflet. No mitral regurgitation was present preoperatively. At reoperation the open mitral "cleft" was found to be the cause of the mitral regurgitation. The mural leaflet appeared to insert to at least one third of the circumference of the anu1us. Closure of the "cleft" produced an excellent result, as shown by the dramatic decrease in heart size on chest films and the absence of a murmur 10 months postoperatively (Fig. 12). The second patient had undergone his first operation at the age of 8 months, and the mitral "cleft" had been reported to be completely closed. Moderate mitral regurgitation was present preoperatively. At reoperation the "cleft" was found to be only partially closed, and the anterior mitral leaflet was redundant and had elongated chordae. The residual "cleft" was closed, the free margin of the leafiet was plicated, and the chordae were shortened. Persistent mitral regurgitation led to a new operation 7 days later. The mitral anu1us was found to be enlarged and a Wooler-type annuloplasty was added to the repair. A murmur persists 7 months postoperatively, and the heart is smaller but still enlarged on the chest film (Fig. 13). The third patient had been operated upon at the age of 5 months, and the "cleft" had been left partially open. No mitral regurgitation was present preoperatively. At reoperation the mitral anulus was moderately enlarged. The residual "cleft" was closed and a Wooler-type annuloplasty was performed. Three months postoperatively a murmur persists, and the heart is smaller but still enlarged on chest roentgenography (Fig. 14).
Discussion Rastelli and associates? first described an effective operation for total repair of complete AV canal malformations in 1968. Since then, increasingly lower mor-
Volume 85 Number 3
Mitral repair in complete AV canal
March,1983
Fig. 12. Patient 1. View from left atriotomy. Left. Intraoperative anatomy: The "cleft" is completely open. The mural leaflet insertsto approximately one third of the annular circumference. Right, Repair. The "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. tality rates have been reported in all age groups.v 8-20 However, hemodynamically satisfactory repair remains a formidable challenge because of the variations in the anatomy and geometry of the valvular apparatus. 4.21-23 Extensive anatomic studies 24- 27 have led some groups to a new type of repair, 27 the primary purpose of which is to decrease the incidence of repair dehiscence and to avoid bicuspidization in an effort to improve mitral function. Our group has employed a modification of the Mayo Clinic techniqueP- 7 with special emphasis on the following technical aspects: The use of a single patch, which implies division of the common leaflets, when needed, allows excellent exposure of the ventricular portion of the septal defect and may account for the absence of residual shunts in our series. Adequate sizing of the patch is essential. Too large a patch can cause annular dilatation, as might have occurred in two of our three cases of late mitral regurgitation. On the other hand, too small a patch may cause redundancy of the reconstructed anterior leaflet of the mitral valve, and may necessitate plication of its free margin. Use of fine continuous horizontal mattress sutures to reattach the left and right sides of the common leaflets to the patch provides a simple and reliable repair. Closely spaced bites on the mitral side, well away from the cut edges;" resulted in only one case of repair dehiscence (3%), which occurred early in our experience. We did not find it necessary to use pledgers" 10 or a double patch technique 19, 20, 27 to ensure adequate strength to the suture line. Especially in infants, pledgets may interfere with valvular function. Moreover, even in infants, we always had enough valvular tissue left to reconstruct the AV valves adequately. 20 Our decision whether to construct a bileaflet or trileaflet mitral valve is based on the size of the mural leaflet. The trileaflet repair is reserved for those cases in
39 3
Fig. 13. Patient 2. View from left atriotomy. Left, Intraoperative anatomy: The "cleft" is partially open. The anterior mitral leaflet is redundant and has elongated chordae. The anulus is enlarged. Right, Final repair (see text): The residual "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. The free margin of the anterior leaflet has been plicatedand the chordae have been shortened. One 3-0 figure-of-eight braided polyester suture has been used at the level of each commissure for the Wooler-type annuloplasty.
.•
-.-
~ .F,':
:": -
tt.,
'!Ii
"
""
Fig. 14. Patient 3. View from left atriotomy. Left, Intraoperative anatomy: The "cleft" is partially open. The mural leaflet inserts to one third of the circumference of the anulus. The anulus is enlarged. Right, Repair: The residual "cleft" has been closed with fine (5-0) interrupted braided polyester sutures. One 3-0 figure-of-eight braided polyester suture has been used at the level of each commissure for the Wooler-type annuloplasty. which the insertion of the mural leaflet at the anulus is less than one third of the annular circumference, in which case closure of the "cleft" would restrict the motion of the "anterior" leaflet. 5, 7 The uniform finding of regurgitation through the open "cleft" in all three late reoperations might be related to our lack of recognition that the "cleft" is a commissure'" and should be treated as such. Therefore, invagination of the base of the commissure may provide better coaptation of the two leaflets and further decrease the incidence of late mitral regurgitation. Early correction was frequently performed in our series, and we have reached the following conclusions: 1. Repair is as safe in infants less than 6 months of age as it is in older age groups. This observation justifies early intervention to alter the poor natural history
The Journal of
394 Abbruzzese et al.
of the malformation, which shows a high early mortality'" and rapidly occurring irreversible pulmonary vascular changes.i" In fact, in our series PVR was again''? found to be a determinant of surgical results. If deaths resulting from technical (bleeding) or unrelated (pneumonia) causes are excluded, four of nine patients (45%) with a PVR greater than 5 WU . m2 and one of 22 patients (4.5%) with a PVR less than 5 WU . m2 died after the operation (p = 0.017 by Fisher's exact test). 2. Repair is easy in the youngest age group. Profound hypothermia with partial cardiopulmonary bypass and circulatory arrest, cold crystalloid cardioplegia, and the absence of venous cannulas allowed perfect exposure and satisfactory assessment of the anatomy. The lack of acquired valvular changes, such as scarring of the leaflets and annular dilatation, facilitated the repair and may also be responsible for the low incidence (26%) or preoperative mitral regurgitation. 3. Repair of the valvular structures is reliable in early infancy. There was only one case (7%) of severe late mitral regurgitation which necessitated reoperation in the youngest age group, as compared to a similar percentage (10%) in the older age groups.
2
3 4 5
6
7
8
9
REFERENCES Espulgas E, Olley PM, Kidd BSL, Trusler GA: Hemodynamic results following repair of complete atrioventricular canal defects, The Child With Congenital Heart Disease After Surgery, BSL Kidd, RD Rowe, eds., Mount Kisco, 1976, Futura Publishing Co., Inc., pp 25-34 McCabe rc, Engle MA, Gay WA lr, Ebert PA: Surgical treatment of endocardial cushion defects. Am 1 Cardiol 39: 72-77, 1977 McGoon DC, Puga Fl: Atrioventricular canal. Cardiovasc Clin 11:31 I-322, 1981 Bharati S, Lev M: The spectrum of common atrioventricular orifice (canal). Am Heart 1 86:553-561, 1973 Harlan Bl, Starr A, Harwin FM: Complete atrioventricular canal, Manual of Cardiac Surgery, Vol II, New York, 1981, Springer-Verlag, pp 269-275 Heath D, Edwards IE: The pathology of hypertensive pulmonary vascular disease. Circulation 18:533-547, 1958 Rastelli GC, Ongley PA, Kirklin lW, McGoon DC: Surgical repair of the complete form of persistent common atrioventricular canal. 1 THoRAc CARDIOVASC SURG 55:299-308, 1968 Alfieri 0, Subramanian S: Successful repair of complete atrioventricular canal with undivided anterior common leafiet in a 6 month old infant. Ann Thorac Surg 19:9296, 1975 Hardesty RL, Zuberbuhler lR, Bahnson HT: Surgical
Thoracic and Cardiovascular Surgery
10
11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
treatment of atrioventricular canal defect. Arch Surg 110:1391-1396, 1975 Mills NL, Ochsner Jl., King TD: Correction of type C complete atrioventricular canal. Surgical considerations. 1 THoRAc CARDIOVASC SURG 71:20-28, 1976 Kahn DR, Levy 1, France NE, Chung Kl, Dacumos GC: Recent results after repair of atrioventricular canal. 1 THoRAc CARDIOVASC SURG 73:413-415, 1977 Mair DD, McGoon DC: Surgical correction of atrioventricular canal during the first year of life. Am 1 Cardiol 40:66-69, 1977 Berger TJ, Kirklin JW, Blackstone EH, Pacifico AD, Kouchoukos NT: Primary repair of complete atrioventricular canal in patients less than 2 years old. Am 1 CardioI41:906-913, 1978 Berger TJ, Blackstone EH, Kirklin lW, Bargeron LM, Hazelrig IB, Turner ME: Survival and probability of cure without and with operation in complete atrioventricular canal. Ann Thorac Surg 27: 104-111, 1979 Foglia RP, Mulder DG: Surgical correction of complete atrioventricular canal. Am 1 Surg 136: 121-127, 1978 Culpepper W, Kolff 1, Lin CW, Vitullo D, Lamberti 1, Arcilla RA, Replogle R: Complete common atrioventricular canal in infancy. Surgical repair and postoperative hemodynamics. Circulation 58:550-558, 1978 Stewart S, Harris P, Manning 1: Complete endocardial cushion defect. Operative technique and results. 1 THoRAc CARDIOVASC SURG 78:914-919, 1979 Midgley M, Galioto FM, Shapiro SR, Perry LW, Scott LP: Experience with repair of complete atrioventricular canal. Ann Thorac Surg 30: 151-159, 1980 Kirklin JW, Chung G, Pacifico AD, Blackstone EH, Bargeron M: Repair of AV canal defects, Congenital Heart Disease in the First Three Months of Life: Medical and Surgical Aspects, L Parenzan, G Crupi, G Graham, eds., Bologna, 1981, Casa Editrice Patron SAS, pp 431437 Mavroudis C, Weinstein G, Turley K, Ebert PA: Surgical management of complete atrioventricular canal. 1 THoRAc CARDIOVASC SURG 83:670-679, 1982 Rastelli GC, Kirklin lW, Titus lL: Anatomic observations on complete form of persistent common atrioventricular canal with special reference to atrioventricular valves. Mayo Clin Proc 41:296-308, 1966 Goor D, Lillehei CW, Edwards IE: Further observations on the pathology of the atrioventricular canal malformation. Arch Surg 97:954-962, 1968 Ebert PA, Goor DA: Complete atrioventricular canal malformation: Further clarification of the anatomy of the common leafiet and its relationship to the VSD in surgical correction. Ann Thorac Surg 25: 134-143, 1978 Ugarte M, Enriquez de Salamanca F, Quero M: Endocardial cushion defects. An anatomical study of 54 specimens. Br Heart 1 38:674-682, 1976 Piccoli GP, Gerlis LM, Wilkinson lL, Lozsadi K, Macartney Fl, Anderson RH: Morphology and classifica-
Volume 85 Number 3 March,1983
26
27
28
29
tion of atrioventricular defects, Br Heart J 42:621-632, 1979 Piccoli GP, Wilkinson JL, Macartney FJ, Gerlis LM, Anderson RH: Morphology and classification of complete atrioventricular defects, Br Heart J 42:633-639, 1979 Carpentier A: Surgical anatomy and management of the mitral component of atrioventricular canal defects, Paediatric Cardiology 1977, RH Anderson, EA Shinebourne, eds., Edinburgh, Churchill Livingstone Katz NM, Blackstone EH, Kirklin JW, Bradley EL, Lemons JE: Suture techniques for atrioventricular valves. Experimental study. J THoRAc CARDIOVASC SURG 81:528536, 1981 Newfeld EA, Sher M, Paul MH, Nikaidoh H: Pulmonary vascular disease in complete atrioventricular canal defect. Am J Cardiol 39:721-726, 1977
Discussion DR. LEONARD L. BAILEY Loma Linda. Calif.
The technical features for repair of A V canal in infancy have been appropriately stressed in this report. The Oregon group's attention to detail both during and after operation is obvious. I was particularly pleased with their honest assessment of the mitral valve apparatus as a trileaflet structure. I have not encountered a case in which I felt biscuspidization of the mitral apparatus would produce stenosis. For comparison, I have reviewed our smaller series of 11 consecutive patients ranging in age from 7 days to 24 months who have had repair of complete AV canal since mid-1976. Seventy-three percent of the patients had the Down syndrome. All had severe persistent cardiac failure and marked growth delay. Weight ranged from 2.2 to 10 kg. Five (45%) patients were under 3 months of age including 2 neonates. The two newborn infants were free of genetic disorders but had complex associated intracardiac malforrnations-doubleoutlet right ventricle (DORV) and pulmonary venous obstruction in one and dextro-transposition of the great arteries
Mitral repair in complete AV canal
395
(d-TGA) and tracheoesophageal fistula (TEF) in the other. Both required preoperative ventilatory support and relatively urgent surgical intervention. Complete repair was accomplished in both. The baby with DOR V continues to do well three years postoperatively. The neonate with d-TGA and TEF died of respiratory insufficiency produced by an untreated tracheobronchial mucus cast. Complete repair was accomplished in all 11 infants in our series. Repairs were of the Mayo Clinic-type: The common leaflets were divided when indicated (type C) and a twoleaflet mitral valve apparatus was created. Pledgets were not used to mount valve tissue to the prosthetic patch. Pericardial pledgets and simple running 7-0 monofilament suture were used to repair the mitral valve cleft. There have been three operative deaths in our series for an overall mortality of 27%. However, there has been only one death (the neonate with d-TGA and TEF) in the last seven consecutive infants treated since early 1977. There have been no late cardiac deaths and no instances of heart block or residual intracardiac shunts. Significant mitral valve regurgitation exists in only one patient who, 3 months postoperatively, still requires cardiac medication although his growth and activity level have dramatically increased. The authors have presented data which, when combined with other recent reports, appear to establish the validity of complete correction of AV canal as a reasonable alternative to pulmonary artery banding or protracted medical management in symptomatic infants of any age or weight. I am convinced, however, that this type of operation should be limited to centers quite competent with corrective cardiac procedures in early infancy. I have no disagreement with the data or techniques presented by the authors, and I thank them for clarifying the issue of a trileaflet mitral valve. DR. A B B R U Z Z ESE (Closing) I thank Dr. Bailey for participating in the discussion, and I congratulate him for his excellent results, especially his late results.