- Email: [email protected]
Long-term results of repair of incomplete persistent atrioventricular canal
J THoRAc CARDIOVASC SURG 82:669-673, 1981
Long-term results of repair of incomplete persistent atrioventricular canal We evaluated the late results following repair of otherwise anatomically uncomplicated incomplete persistent atrioventricular canal in 39 consecutive operative survivors who underwent operation at our institution prior to 1976. Average follow-up was 12 years. Postoperative cardiac catheterization was performed in 35 patients (90%) at an average of 11 months after operation. Seven (20%) had residual mitral regurgitation with elevated mean pulmonary arterial wedge or left atrial pressures with abnormal v waves. Regurgitation was mild to moderate (pulmonary artery wedge or left atrial pressure 12 to 15 mm Hg) in five and severe (pulmonary artery wedge pressure over 20 mm Hg) in two patients. Clinically significant arrhythmias including complete heart block, sudden death, nodal rhythm, and chronic atrial fibrillation occurred in seven patients (/8%). Two patients have required reoperationfor mitral regurgitation. Five have clinically recognizable mild-to-moderate mitral regurgitation controlled with medical management; 25 patients are asymptomatic at current evaluation. Estimated actuarial survival rate at· 13 years is 88% ± 6%, with an actuarial survival free of reoperation of 82% ± 6%. However, actuarial survival free of any late complication including late death, reoperation, serious arrhythmia, or mitral regurgitation is only 52% ± 10% at 13 years.
Daniel M. Goldfaden, M.D. (by invitation), Michael Jones, M.D. (by invitation), and Andrew G. Morrow, M.D., Bethesda, Md.
During the twenty-six years since the first reported successful repair of incomplete persistent atrioventricular canal by Lillehei and associates, 1 there have been several studies showing that operative repair provides substantial hemodynamic improvement with low rates of operative death or complication.v? Late clinical results have been reported by others, but those series had incomplete follow-up on many patients and included many patients with other serious associated cardiac detects.>> To define better the late results after repair of uncomplicated incomplete persistent atrioventricular canal, we reviewed our results in all patients who underwent repair at the National Heart Institute prior to 1976.
ventricular canal between 1956 and 1976 were reviewed. Patients with other major cardiac anomalies were excluded. Most (90%) of the operative survivors underwent routine postoperative evaluation including cardiac catheterization 6 to 12 months after operation. Late follow-up of at least 4 years or until death was obtained in 87%. In most instances the latest follow-up was within the past year. Late clinical evaluation included physical examination, chest roentgenogram, and electrocardiogram in each. M-mode echocardiograms were obtained in 15. Actuarial analysis was used to estimate survival, survival free of reoperation, and survival free of late complication. 6
Methods
The median age of the 44 patients who were operated upon was 12 years (range 1 to 68 years). There were 22 female and 22 male patients. Thirty-nine patients survived operation (89%). Average follow-up was 10.8 years with the median age at the time of latest followup being 25 years. Operative findings. All 44 patients had a typical low-lying atrial septal defect of the ostium primum type which varied in size from 0.5 cm to complete absence of the interatrial septum. The anterior mitral leaflet showed a complete cleft extending to the base of
Records of the 44 patients who underwent operative repair of uncomplicated incomplete persistent atrioFrom the Surgery Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. Read at the Sixty-first Annual Meeting of The American Association for Thoracic Surgery, Washington, D. C., May 11-I3, 1981. Address for reprints: Dr. Andrew G. Morrow, Clinic of Surgery, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. 20205.
Results
669
The Journal of
670
Goldfaden, Jones. Morrow
Thoracic and Cardiovascular Surgery
% 100
80
o
Asymptomatic
~
Recurrent cardiac symptoms
60 40 20
<
12 mmHg
12-15 mmHg
15-20 mmHg
>20 mmHg
Fig. 1. Mean pulmonary arterial wedge or left atrial pressure at time of postoperative catheterization.
80
%
60
f
S.E.M.
40 20,
, , , oIL...-_----'--_----'--_-'--_-'--_-::'-:-_--"-4
!
I
8
12
years
16
20
24
Fig. 2. Actuarial survival. Operative mortality not included.
the valve in 37 patients, while in seven the cleft was incomplete, forming a notch or partial cleft in the free margin of the leaflet. Repair included patch closure of the septal defect and direct suture closure of the mitral cleft in all patients. Three patients were believed at the time of operation to have significant tricuspid regurgitation, and annuloplasty or suture closure of an abnormal cleft in the tricuspid valve was performed. Most of the other patients were noted to have some deficiency of the septal leaflet of the tricuspid valve but not of a degree sufficient to produce a clinically significant incompetent valve. Other simultaneous procedures in these patients included ligation of an anomalous left superior vena cava in three and suture closure of a small membranous ventricular septal defect in three. Because the patients with ventricular septal defects had two complete atrioventricular valve rings without continuity between the atrial and ventricular defects, they were still considered as incomplete atrioventricular canal defects. Early results. There were five operative deaths, all occurring prior to 1965. Cause of death was low car-
diac output in two, technical error in one, sudden death on postoperative day 12 in one, and aplastic anemia related to chloramphenicol usage in one. Many of the patients had transient atrial arrhythmias or temporary conduction disturbances, but only three (8%) had significant arrhythmias at the time of discharge from the hospital. One patient remained in atrial fibrillation, one had nodal rhythm but was asymptomatic with a heart rate of 60 to 80 beats/min, and the third had complete heart block necessitating implantation of a permanent pacemaker. Postoperative cardiac catheterization was performed in 35 patients (90%) at an average of II months after operation. The majority (80%) had normal mean pulmonary arterial wedge or left atrial pressures (less than 12 mm Hg), indicating that there was no hemodynamically significant mitral valve abnormality (Fig. I). Mild mitral regurgitation was suggested in five patients (14%) by elevated mean pulmonary arterial wedge or left atrial pressure (12 to 15 mm Hg) with elevated v waves. Two patients appeared to have serious mitral regurgitation, with mean pulmonary arterial wedge pressures greater than 20 mm Hg. Dye curves were normal in all but one patient, who was found to have recurrent left-to-right shunting at the atrial level. At reoperation he was found to have partial dehiscence of the atrial patch. Prior to operation 22 patients (67%) were symptomatic. At the time of postoperative catheterization only six (17%) had residual symptoms. There was good correlation between residual symptoms and catheterization findings, with five of six symptomatic patients having abnormal catheterization findings (Fig. I). Four were found to have evidence of residual mitral regurgitation, and one had recurrent atrial shunting. Even though the majority of the patients had no ab-
Volume 82
Incomplete atrioventricular canal
Number 5 November, 1981
100
65
80
60
60
55
%
40 20 0
S.E.M.
4
•
50
!
45 12
8
16
20
years
24
•• ••
67 I
r::»INormal I;;:;:;jRange
•
• • •
Fig. 3. Actuarial survival free of reoperation.
normalities on cardiac catheterization and no symptoms, there was an apical systolic murmur of Grade 1/6 to 4/6 in 33 of the 35 patients. Neither the presence of, nor the intensity of, the murmur correlated with catheterization evidence of mitral regurgitation. Persistence of a significant apical systolic murmur, despite a mitral valve that shows no regurgitation on the angiogram, has been demonstrated previously by Griffiths and associates 7 in 1969. Late deaths. There have been three late deaths among the 39 operative survivors; two died of cardiac and one of noncardiac causes. The noncardiac death was from an intestinal malignant tumor in the oldest patient in the series (aged 72). One of the two cardiac deaths occurred in a patient with severe persistent mitral regurgitation that was uncorrected despite a second attempted valvuloplasty. The other cardiac death was due to sudden, unexpected cardiac arrest in a previously asymptomatic patient 13 years after operation. Actuarial survival rate at 13 years after operation was 88% ± 6% of operative survivors (Fig. 2). Late reoperation. Three patients have required late reoperation, one for dehiscence of the atrial patch and two for mitral regurgitation. One patient was found to have a severely scarred, fixed-orifice valve at initial operation in 1959. A second attempted repair of the valve was also unsuccessful, and she died in congestive heart failure a few months after the second operation. Undoubtedly, this patient today would have undergone mitral replacement at the initial operation. The other patient with mitral regurgitation had mitral regurgitation noted at early postoperative catheterization, but was initially asymptomatic. Her symptoms worsened at the age of 17 years; 10 years after the initial repair the mitral valve was replaced. Fig. 3 shows the estimated survival free of late reoperation. The risks of both late death and reoperation are included. The predicted survival without reoperation at 13 years is 82% ± 6% (Fig. 2). Late arrhythmias. Significant arrhythmias or con-
ASYMPTOMATIC
SYMPTOMATIC
Fig. 4. Left atrial size by echocardiogram at latest clinical evaluation in 15 patients. The left atrial (LA) size is expressed in mm millimeters (anteroposterior dimension).
duction disturbances have occurred late after operation in four patients. One patient with moderate persistent mitral regurgitation has chronic atrial fibrillation. Two patients have complete block 5 and 9 years after operation. Both of these patients have had permanent pacemaker implantation. One patient who was free of symptoms had a sudden cardiac arrest and died 13 years after operation. Including the three patients who were discharged from the hospital with arrhythmias or conduction defects, seven of the initial 39 operative survivors have significant arrhythmias. Thus the cumulative incidence of arrhythmias during the average follow-up of 10.8 years is 18%. Late clinical evaluation. From the initial 39 operative survivors, there were 30 late survivors who had not undergone mitral reoperation and were available for late clinical evaluation. Average follow-up in these 30 patients was 13 years. Twenty-five of these 30 patients are currently asymptomatic and have no clinical evidence of significant mitral regurgitation. The other five have mild cardiac symptoms (New York Heart Association Class II) and, on the basis of clinical examination, are believed to have mild-to-moderate mitral regurgitation. Two of the patients take digoxin for control of atrial fibrillation. Otherwise, none of the patients takes any cardiac medication.
672
The Journal of Thoracic and Cardiovascular Surgery
Goldfaden, Jones, Morrow
100
60 40
S.E.M.
!
20
o
4
8
years
12
16
20
24
Fig. 5. Actuarial survival free of complication (late death, reoperation, arrhythmia, or mitral regurgitation). Since most of the patients have persistent apical systolic murmurs, the presence of a murmur alone was not considered as evidence of significant mitral regurgitation. In evaluating the severity of mitral regurgitation, we believed the presence and character of murmurs to be of lesser importance than radiologic, physical, or echocardiographic evidence of cardiac enlargement' or pulmonary congestion. On the basis of the latter findings, clinically significant mitral regurgitation has developed in seven (18%) of the operative survivors during the period of follow-up, although only two have come to reoperation and the other five have mild symptoms. Echocardiograms were obtained in 15 patients at the time of late clinical evaluation. No specific abnormalities of mitral valve motion were seen, other than multiple echoes from the anterior mitral leaflet suggestive of some thickening. Each of the symptomatic patients was noted to have an enlarged left atrium, as were four of the nine asymptomatic patients (Fig. 4). Actuarial analysis of the late results, including the risk of development of any late complication (late death, reoperation, significant arrhythmia, or mitral regurgitation) is shown in Fig. 5. The estimated survival, free from any of these late complications, is 52% ± 10% at 13 years after operation. Discussion Patients with incomplete persistent atrioventricular canal who do not undergo operative repair usually exhibit serious clinical deterioration by the fourth decade of life, with a 54% incidence of death or severe symptoms being reported by Somerville" in patients over 30 years of age. In her series of 122 patients reported in 1965, those who had severe symptoms or died usually did so because of serious arrhythmias (56%) or severe mitral regurgitation (22%). In evaluating the late results after operative repair, we attempted to determine
whether these late serious complications were prevented or made less likely to occur. The overall incidence of serious arrhythmia during the period of follow-up was 18%, with over half of the arrhythmias occurring in the late postoperative period. This incidence actually exceeds the 12% overall incidence reported by Somerville," although a direct comparison is not possible since the median age in our series was 25 years at the time of follow-up as opposed to a median age of 15 years in her series. Still, it seems that operative repair may not significantly reduce the risk of serious arrhythmias. The most alarming arrhythmias were sudden death and late spontaneous complete heart block, which occurred in three of the operative survivors (8%). This incidence is very similar to the 5% incidence of heart block or sudden death reported by Somerville. Atrial fibrillation following repair was less common in our experience, with only two patients (5%) having this arrhythmia. In both instances there was associated mitral regurgitation. Satisfactory function of the repaired mitral valve was demonstrated in the report by Braunwald and Morrow" from this institution in 1966. Most of the patients whose early postoperative hemodynamic results were reported in that paper are included in the current review of late postoperative results. As in that earlier study, we found a low incidence (6%) of severe mitral regurgitation at the time of early postoperative cardiac catheterization, although mild mitral regurgitation persisted in 14%. Similar good hemodynamic results were documented by Griffiths and associates? in 1969. They performed cardiac catheterization and left ventricular angiography postoperatively in eight patients. Despite persistent apical systolic murmurs in all, they found only one patient to have residual mitral regurgitation. The possibility exists that in many of these patients the systolic murmur is due to turbulence across the abnormal left ventricular outflow tract in incomplete persistent atrioventricular canal. Although seven (18%) of the patients in our series have symptomatic mitral regurgitation, only two (5%) have required operation. The others have mild symptoms which are easily controlled. Of these seven patients, five had evidence of residual mitral regurgitation on early postoperative catheterization. Only two of the 28 patients with normal postoperative catheterization have subsequently had clinically significant mitral regurgitation. This observation suggests that if the initial repair produces a hemodynamically competent valve, the risk of late deterioration is small. The operative mortality of 11% in our series of patients is comparable to results reported by several
Volume 82
Incomplete atrioventricular canal
Number 5
673
November, 1981
others-5.6% to 16%.3-;' As in our series, several of those authors found that operative mortality has fallen substantially in more recent years. 3. 4 Estimated survival using actuarial tables was reported by Mclvlullan," and Losay," and their associates. They found an estimated 15 year survival rate of 90% to 94%. This is similar to our actuarial estimate of an 88% survival rate at 13 years. By using actuarial analysis to evaluate the risk of any late complication developing, we found that only about half of the patients who survive operative repair will survive 13 years free of any late complication. This substantial risk certainly demands continued follow-up of patients operated upon for incomplete persistent atrioventricular canal. It is gratifying to see, however, that with appropriate medical and surgical management of these late complications, 25 (83%) of the 30 late survivors available for evaluation were free of cardiac symptoms.
REFERENCES Lillehei CW, Cohen M, Warden HE, Yarco RL: The direct-vision intracardiac correction of congenital anomalies by controlled cross circulation. Results in 32 patients with ventricular septal defects, tetralogy of Fallot, and
2
3
4
5
6
7
8
atrioventricularis communis defects. Surgery 38:11-29, 1955 Braunwald NS, Morrow AG: Incomplete persistent atrioventricular canal. Operative methods and the results of preand postoperative hemodynamic assessments. J THORAC CARDIOVASC SURG 51:71-80, 1966 McMullan MH, McGoon DC, Wallace R, Danielson GK, Weidman WH: Surgical treatment of partial atrioventricular canal. Arch Surg 107:705-710, 1973 Losay J, Rosenthal A, Castenada AR, Bernhard WH, Nadas AS: Repair of atrial septal defect primum. Results, course, and prognosis. J THoRAc CARDIOVASC SURG 75:248-254, 1978 Levy S, Blondeau P, Dubost C: Long-term follow-up after surgical correction of the partial form of atrioventricular canal (ostium primum). J THORAC CARDIOVASC SURG 67:353-363, 1974 Anderson RP, Bonchek LI, Grunkemeier GL, Lanhert LE, Starr A: The analysis and presentation of surgical results by actuarial methods. J Surg Res 16:224-230, 1974 Griffiths SP, Ellis K, Burris 10, Blumenthal S, Bowman FO, Maim JR: Postoperative evaluation of mitral valve function in ostium primum defect with cleft mitral valve (partial form of atrioventricular canal). Circulation 40:2129, 1969 Somerville J: Ostium primum defect. Factors causing deterioration in the natural history. Br Heart J 27:413-419, 1968
Long-term results of repair of incomplete persistent atrioventricular canal We evaluated the late results following repair of otherwise anatomically uncomplicated incomplete persistent atrioventricular canal in 39 consecutive operative survivors who underwent operation at our institution prior to 1976. Average follow-up was 12 years. Postoperative cardiac catheterization was performed in 35 patients (90%) at an average of 11 months after operation. Seven (20%) had residual mitral regurgitation with elevated mean pulmonary arterial wedge or left atrial pressures with abnormal v waves. Regurgitation was mild to moderate (pulmonary artery wedge or left atrial pressure 12 to 15 mm Hg) in five and severe (pulmonary artery wedge pressure over 20 mm Hg) in two patients. Clinically significant arrhythmias including complete heart block, sudden death, nodal rhythm, and chronic atrial fibrillation occurred in seven patients (/8%). Two patients have required reoperationfor mitral regurgitation. Five have clinically recognizable mild-to-moderate mitral regurgitation controlled with medical management; 25 patients are asymptomatic at current evaluation. Estimated actuarial survival rate at· 13 years is 88% ± 6%, with an actuarial survival free of reoperation of 82% ± 6%. However, actuarial survival free of any late complication including late death, reoperation, serious arrhythmia, or mitral regurgitation is only 52% ± 10% at 13 years.
Daniel M. Goldfaden, M.D. (by invitation), Michael Jones, M.D. (by invitation), and Andrew G. Morrow, M.D., Bethesda, Md.
During the twenty-six years since the first reported successful repair of incomplete persistent atrioventricular canal by Lillehei and associates, 1 there have been several studies showing that operative repair provides substantial hemodynamic improvement with low rates of operative death or complication.v? Late clinical results have been reported by others, but those series had incomplete follow-up on many patients and included many patients with other serious associated cardiac detects.>> To define better the late results after repair of uncomplicated incomplete persistent atrioventricular canal, we reviewed our results in all patients who underwent repair at the National Heart Institute prior to 1976.
ventricular canal between 1956 and 1976 were reviewed. Patients with other major cardiac anomalies were excluded. Most (90%) of the operative survivors underwent routine postoperative evaluation including cardiac catheterization 6 to 12 months after operation. Late follow-up of at least 4 years or until death was obtained in 87%. In most instances the latest follow-up was within the past year. Late clinical evaluation included physical examination, chest roentgenogram, and electrocardiogram in each. M-mode echocardiograms were obtained in 15. Actuarial analysis was used to estimate survival, survival free of reoperation, and survival free of late complication. 6
Methods
The median age of the 44 patients who were operated upon was 12 years (range 1 to 68 years). There were 22 female and 22 male patients. Thirty-nine patients survived operation (89%). Average follow-up was 10.8 years with the median age at the time of latest followup being 25 years. Operative findings. All 44 patients had a typical low-lying atrial septal defect of the ostium primum type which varied in size from 0.5 cm to complete absence of the interatrial septum. The anterior mitral leaflet showed a complete cleft extending to the base of
Records of the 44 patients who underwent operative repair of uncomplicated incomplete persistent atrioFrom the Surgery Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. Read at the Sixty-first Annual Meeting of The American Association for Thoracic Surgery, Washington, D. C., May 11-I3, 1981. Address for reprints: Dr. Andrew G. Morrow, Clinic of Surgery, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. 20205.
Results
669
The Journal of
670
Goldfaden, Jones. Morrow
Thoracic and Cardiovascular Surgery
% 100
80
o
Asymptomatic
~
Recurrent cardiac symptoms
60 40 20
<
12 mmHg
12-15 mmHg
15-20 mmHg
>20 mmHg
Fig. 1. Mean pulmonary arterial wedge or left atrial pressure at time of postoperative catheterization.
80
%
60
f
S.E.M.
40 20,
, , , oIL...-_----'--_----'--_-'--_-'--_-::'-:-_--"-4
!
I
8
12
years
16
20
24
Fig. 2. Actuarial survival. Operative mortality not included.
the valve in 37 patients, while in seven the cleft was incomplete, forming a notch or partial cleft in the free margin of the leaflet. Repair included patch closure of the septal defect and direct suture closure of the mitral cleft in all patients. Three patients were believed at the time of operation to have significant tricuspid regurgitation, and annuloplasty or suture closure of an abnormal cleft in the tricuspid valve was performed. Most of the other patients were noted to have some deficiency of the septal leaflet of the tricuspid valve but not of a degree sufficient to produce a clinically significant incompetent valve. Other simultaneous procedures in these patients included ligation of an anomalous left superior vena cava in three and suture closure of a small membranous ventricular septal defect in three. Because the patients with ventricular septal defects had two complete atrioventricular valve rings without continuity between the atrial and ventricular defects, they were still considered as incomplete atrioventricular canal defects. Early results. There were five operative deaths, all occurring prior to 1965. Cause of death was low car-
diac output in two, technical error in one, sudden death on postoperative day 12 in one, and aplastic anemia related to chloramphenicol usage in one. Many of the patients had transient atrial arrhythmias or temporary conduction disturbances, but only three (8%) had significant arrhythmias at the time of discharge from the hospital. One patient remained in atrial fibrillation, one had nodal rhythm but was asymptomatic with a heart rate of 60 to 80 beats/min, and the third had complete heart block necessitating implantation of a permanent pacemaker. Postoperative cardiac catheterization was performed in 35 patients (90%) at an average of II months after operation. The majority (80%) had normal mean pulmonary arterial wedge or left atrial pressures (less than 12 mm Hg), indicating that there was no hemodynamically significant mitral valve abnormality (Fig. I). Mild mitral regurgitation was suggested in five patients (14%) by elevated mean pulmonary arterial wedge or left atrial pressure (12 to 15 mm Hg) with elevated v waves. Two patients appeared to have serious mitral regurgitation, with mean pulmonary arterial wedge pressures greater than 20 mm Hg. Dye curves were normal in all but one patient, who was found to have recurrent left-to-right shunting at the atrial level. At reoperation he was found to have partial dehiscence of the atrial patch. Prior to operation 22 patients (67%) were symptomatic. At the time of postoperative catheterization only six (17%) had residual symptoms. There was good correlation between residual symptoms and catheterization findings, with five of six symptomatic patients having abnormal catheterization findings (Fig. I). Four were found to have evidence of residual mitral regurgitation, and one had recurrent atrial shunting. Even though the majority of the patients had no ab-
Volume 82
Incomplete atrioventricular canal
Number 5 November, 1981
100
65
80
60
60
55
%
40 20 0
S.E.M.
4
•
50
!
45 12
8
16
20
years
24
•• ••
67 I
r::»INormal I;;:;:;jRange
•
• • •
Fig. 3. Actuarial survival free of reoperation.
normalities on cardiac catheterization and no symptoms, there was an apical systolic murmur of Grade 1/6 to 4/6 in 33 of the 35 patients. Neither the presence of, nor the intensity of, the murmur correlated with catheterization evidence of mitral regurgitation. Persistence of a significant apical systolic murmur, despite a mitral valve that shows no regurgitation on the angiogram, has been demonstrated previously by Griffiths and associates 7 in 1969. Late deaths. There have been three late deaths among the 39 operative survivors; two died of cardiac and one of noncardiac causes. The noncardiac death was from an intestinal malignant tumor in the oldest patient in the series (aged 72). One of the two cardiac deaths occurred in a patient with severe persistent mitral regurgitation that was uncorrected despite a second attempted valvuloplasty. The other cardiac death was due to sudden, unexpected cardiac arrest in a previously asymptomatic patient 13 years after operation. Actuarial survival rate at 13 years after operation was 88% ± 6% of operative survivors (Fig. 2). Late reoperation. Three patients have required late reoperation, one for dehiscence of the atrial patch and two for mitral regurgitation. One patient was found to have a severely scarred, fixed-orifice valve at initial operation in 1959. A second attempted repair of the valve was also unsuccessful, and she died in congestive heart failure a few months after the second operation. Undoubtedly, this patient today would have undergone mitral replacement at the initial operation. The other patient with mitral regurgitation had mitral regurgitation noted at early postoperative catheterization, but was initially asymptomatic. Her symptoms worsened at the age of 17 years; 10 years after the initial repair the mitral valve was replaced. Fig. 3 shows the estimated survival free of late reoperation. The risks of both late death and reoperation are included. The predicted survival without reoperation at 13 years is 82% ± 6% (Fig. 2). Late arrhythmias. Significant arrhythmias or con-
ASYMPTOMATIC
SYMPTOMATIC
Fig. 4. Left atrial size by echocardiogram at latest clinical evaluation in 15 patients. The left atrial (LA) size is expressed in mm millimeters (anteroposterior dimension).
duction disturbances have occurred late after operation in four patients. One patient with moderate persistent mitral regurgitation has chronic atrial fibrillation. Two patients have complete block 5 and 9 years after operation. Both of these patients have had permanent pacemaker implantation. One patient who was free of symptoms had a sudden cardiac arrest and died 13 years after operation. Including the three patients who were discharged from the hospital with arrhythmias or conduction defects, seven of the initial 39 operative survivors have significant arrhythmias. Thus the cumulative incidence of arrhythmias during the average follow-up of 10.8 years is 18%. Late clinical evaluation. From the initial 39 operative survivors, there were 30 late survivors who had not undergone mitral reoperation and were available for late clinical evaluation. Average follow-up in these 30 patients was 13 years. Twenty-five of these 30 patients are currently asymptomatic and have no clinical evidence of significant mitral regurgitation. The other five have mild cardiac symptoms (New York Heart Association Class II) and, on the basis of clinical examination, are believed to have mild-to-moderate mitral regurgitation. Two of the patients take digoxin for control of atrial fibrillation. Otherwise, none of the patients takes any cardiac medication.
672
The Journal of Thoracic and Cardiovascular Surgery
Goldfaden, Jones, Morrow
100
60 40
S.E.M.
!
20
o
4
8
years
12
16
20
24
Fig. 5. Actuarial survival free of complication (late death, reoperation, arrhythmia, or mitral regurgitation). Since most of the patients have persistent apical systolic murmurs, the presence of a murmur alone was not considered as evidence of significant mitral regurgitation. In evaluating the severity of mitral regurgitation, we believed the presence and character of murmurs to be of lesser importance than radiologic, physical, or echocardiographic evidence of cardiac enlargement' or pulmonary congestion. On the basis of the latter findings, clinically significant mitral regurgitation has developed in seven (18%) of the operative survivors during the period of follow-up, although only two have come to reoperation and the other five have mild symptoms. Echocardiograms were obtained in 15 patients at the time of late clinical evaluation. No specific abnormalities of mitral valve motion were seen, other than multiple echoes from the anterior mitral leaflet suggestive of some thickening. Each of the symptomatic patients was noted to have an enlarged left atrium, as were four of the nine asymptomatic patients (Fig. 4). Actuarial analysis of the late results, including the risk of development of any late complication (late death, reoperation, significant arrhythmia, or mitral regurgitation) is shown in Fig. 5. The estimated survival, free from any of these late complications, is 52% ± 10% at 13 years after operation. Discussion Patients with incomplete persistent atrioventricular canal who do not undergo operative repair usually exhibit serious clinical deterioration by the fourth decade of life, with a 54% incidence of death or severe symptoms being reported by Somerville" in patients over 30 years of age. In her series of 122 patients reported in 1965, those who had severe symptoms or died usually did so because of serious arrhythmias (56%) or severe mitral regurgitation (22%). In evaluating the late results after operative repair, we attempted to determine
whether these late serious complications were prevented or made less likely to occur. The overall incidence of serious arrhythmia during the period of follow-up was 18%, with over half of the arrhythmias occurring in the late postoperative period. This incidence actually exceeds the 12% overall incidence reported by Somerville," although a direct comparison is not possible since the median age in our series was 25 years at the time of follow-up as opposed to a median age of 15 years in her series. Still, it seems that operative repair may not significantly reduce the risk of serious arrhythmias. The most alarming arrhythmias were sudden death and late spontaneous complete heart block, which occurred in three of the operative survivors (8%). This incidence is very similar to the 5% incidence of heart block or sudden death reported by Somerville. Atrial fibrillation following repair was less common in our experience, with only two patients (5%) having this arrhythmia. In both instances there was associated mitral regurgitation. Satisfactory function of the repaired mitral valve was demonstrated in the report by Braunwald and Morrow" from this institution in 1966. Most of the patients whose early postoperative hemodynamic results were reported in that paper are included in the current review of late postoperative results. As in that earlier study, we found a low incidence (6%) of severe mitral regurgitation at the time of early postoperative cardiac catheterization, although mild mitral regurgitation persisted in 14%. Similar good hemodynamic results were documented by Griffiths and associates? in 1969. They performed cardiac catheterization and left ventricular angiography postoperatively in eight patients. Despite persistent apical systolic murmurs in all, they found only one patient to have residual mitral regurgitation. The possibility exists that in many of these patients the systolic murmur is due to turbulence across the abnormal left ventricular outflow tract in incomplete persistent atrioventricular canal. Although seven (18%) of the patients in our series have symptomatic mitral regurgitation, only two (5%) have required operation. The others have mild symptoms which are easily controlled. Of these seven patients, five had evidence of residual mitral regurgitation on early postoperative catheterization. Only two of the 28 patients with normal postoperative catheterization have subsequently had clinically significant mitral regurgitation. This observation suggests that if the initial repair produces a hemodynamically competent valve, the risk of late deterioration is small. The operative mortality of 11% in our series of patients is comparable to results reported by several
Volume 82
Incomplete atrioventricular canal
Number 5
673
November, 1981
others-5.6% to 16%.3-;' As in our series, several of those authors found that operative mortality has fallen substantially in more recent years. 3. 4 Estimated survival using actuarial tables was reported by Mclvlullan," and Losay," and their associates. They found an estimated 15 year survival rate of 90% to 94%. This is similar to our actuarial estimate of an 88% survival rate at 13 years. By using actuarial analysis to evaluate the risk of any late complication developing, we found that only about half of the patients who survive operative repair will survive 13 years free of any late complication. This substantial risk certainly demands continued follow-up of patients operated upon for incomplete persistent atrioventricular canal. It is gratifying to see, however, that with appropriate medical and surgical management of these late complications, 25 (83%) of the 30 late survivors available for evaluation were free of cardiac symptoms.
REFERENCES Lillehei CW, Cohen M, Warden HE, Yarco RL: The direct-vision intracardiac correction of congenital anomalies by controlled cross circulation. Results in 32 patients with ventricular septal defects, tetralogy of Fallot, and
2
3
4
5
6
7
8
atrioventricularis communis defects. Surgery 38:11-29, 1955 Braunwald NS, Morrow AG: Incomplete persistent atrioventricular canal. Operative methods and the results of preand postoperative hemodynamic assessments. J THORAC CARDIOVASC SURG 51:71-80, 1966 McMullan MH, McGoon DC, Wallace R, Danielson GK, Weidman WH: Surgical treatment of partial atrioventricular canal. Arch Surg 107:705-710, 1973 Losay J, Rosenthal A, Castenada AR, Bernhard WH, Nadas AS: Repair of atrial septal defect primum. Results, course, and prognosis. J THoRAc CARDIOVASC SURG 75:248-254, 1978 Levy S, Blondeau P, Dubost C: Long-term follow-up after surgical correction of the partial form of atrioventricular canal (ostium primum). J THORAC CARDIOVASC SURG 67:353-363, 1974 Anderson RP, Bonchek LI, Grunkemeier GL, Lanhert LE, Starr A: The analysis and presentation of surgical results by actuarial methods. J Surg Res 16:224-230, 1974 Griffiths SP, Ellis K, Burris 10, Blumenthal S, Bowman FO, Maim JR: Postoperative evaluation of mitral valve function in ostium primum defect with cleft mitral valve (partial form of atrioventricular canal). Circulation 40:2129, 1969 Somerville J: Ostium primum defect. Factors causing deterioration in the natural history. Br Heart J 27:413-419, 1968