- Email: [email protected]
Persistent low cardiac output after the Fontan operation
J
THORAC CARDIOVASC SURG
92:402-405, 1986
Persistent low cardiac output after the Fontan operation Should takedown be considered? Four of 44 patients who bad undergone the Fontan operation bad persistent low cardiac output necessitating takedown of the shunt 6 to 65 hours (average 23 hours) postoperatively. All four were in a group of 22 patients with complexlesions other than tricuspidatresia withventriculoarterial concordance. The development of postoperative right atrial hypertension (average 24 torr), hepatomegaly, marked ascites, and decreasinglung compliance led to severe systemichypotension with systolic arterial pressure ranging from 55 to 82 torr (average 68 torr), persistent metabolic acidosis, and oliguria despite massive colloid and crystalloid infusions (11,000 m1/m2/24 hr) and inotropic support. At reoperation the atriopulmonaryanastomosis, which wasfoundto be wide open,was taken downand an atrial septal defect was created in all patients. Three patients were left with a Glennshunt and an aortopulmonary shunt to the left lung. One patient bad bilateral aortopulmonary shunts.Two patients whosurvived reoperationbad immediatepostoperative improvement in systolicarterial and mean right atrial pressure(average 100 torr and 11.5 torr, respectively). Both are well 5 monthsand 4 years later. Repeat Fontan operation remains a possibility with acceptable risks because of the presence of the Glennshunt in both patients. We believe that takedown should be considered in patients with persistent low cardiac output after the Fontan operation.
Serafin Y. DeLeon, M.D., Michel N. Ilbawi, M.D., Farouk S. Idriss, M.D., Alexander J. Muster, M.D., Samuel S. Gidding, M.D., Teresa E. Berry, M.D., and Milton H. Paul, M.D., Chicago. Ill.
Since the application of the fontan operation for tricuspid atresia was clearly established,'? its use has been extended to other complex congenital heart defects."? The morbidity and mortality, however, remain high (15% to 27%) in patients undergoing the Fontan operation who have complex lesions other than tricuspid atresia with ventriculoarterial concordance. 1, 5-8 Right atrial hypertension, severe loss of fluid into the intersti-
From the Divisions of Cardiovascular-Thoracic Surgery and Cardiology, The Children's Memorial Hospital, and the Departments of Surgery and Pediatrics, Northwestern University Medical School, Chicago, Ill. Supported in part by the A. C. Buehler Foundation, Park Ridge,
Ill. Received for publication Oct. 29, 1985. Accepted for publication Nov. 19, 1985. Address for reprints: Serafin Y. DeLeon, M.D., Division of Cardiovascular-Thoracic Surgery, 2300 Children's Plaza, Chicago, Ill. 60614.
402
tial space with development of pleural and pericardial effusion and ascites, and decreasing lung compliance are fairly common postoperative management problems. In four of 44 patients undergoing the Fontan operation, low cardiac output developed postoperatively, associated with right atrial hypertension, significant ascites, and decreasing lung compliance. Because the low cardiac output persisted in all patients despite massive fluid infusion and inotropic support, the Fontan shunt was taken down. To determine whether takedown of the Fontan shunt was worthwhile and should be considered under these circumstances, we reviewed our experience.
Patients Four of 44 patients who underwent the Fontan operation at the Children's Memorial Hospital in Chicago in the past 8 years had persistent low cardiac output in the immediate postoperative period necessitating takedown of the Fontan shunt. All four patients were
Volume 92 Number 3, Part 1
Fontan operation
September, 1986
403
Table I. Clinical summary of patients undergoing Fontan takedown Patient (DOB)
Diagnosis
Previous operation
1(9/11/74)
SV (S,L,L)t closing bulboventricular foramen (80 torr gradient)
PAB
2 ( 3/2/76)
Tricuspid atresia type II-C,t restricted VSD (80 torr gradient)
PAB, ligation PDA
3 ( 8/3/77)
Tricuspid atresia type II-A, bilateral SVC
Wa terston-Cooley shunt, left Blalock-Taussig shunt
Tricuspid atresia type II-C, restricted VSD (35 torr gradient), COA
COA repair, PAB, rebanding of PA, creation of APW
4 (10/8/81)
Fontan (date)*
9/2/81: Closure of TV, direct RA-PA anastomosis with 25 mm porcine valve, proximal MPA-AO anastomosis, Glenn shunt 12/2/80: Direct RA-PA anastomosis, proximal MPA-AO anastomosis, Glenn shunt 8/3/83: Direct RA-PA anastomosis, takedown of shunts 5/8/85: Direct RA-PA anastomosis
Takedown (date)
Outcome
9/5/81: Takedown RA-PA anastomosis, removal porcine valve, creation ASD, AO-LPA shunt
Well at 4 yr
12/3/80: Takedown RA-PA anastomosis, creation ASD, AO-LPA shunt
Died intraoperatively
8/4/83: Takedown of RA-PA anastomosis, creation ASD, bilateral AO-PA shunts 5/8/85: Takedown of RA-PA anastomosis, creation ASD, AO-LPA shunt, Glenn shunt
Died 8 hr
postoperatively
Well at 5 mo
Legend:AO-LPA, Aorta-left pulmonary artery. AO-PA, Aorta-pulmonary artery. APW, Aortopulmonary window. ASD, Atrial septal defect. COA, Coarctation of the aorta. DOB, Date of birth. MPA-AO, Main pulmonary artery-aorta. PAB, Pulmonary artery banding. RA-PA, Atriopulmonary anastomosis. SV, Single ventricle. SVC, Superior vena cava. TV, Tricuspid valve. VSD, Ventricular septal defect. 'The coronary sinus drainage was to the right atrium in all patients. tBased on Van Praagh's classification of malposition of the heart, Heart Disease in Infants, Children, and Adolescents, ed 3, FH Adams, GC Emmanouilides, eds., Baltimore, 1983, The Williams & Wilkins Company, pp 422-458. tBased on classification by Edwards and Burchell (Edwards lE, Burchell HB: Congenital tricuspid atresia. A classification. Med Clin North Am 33:1177-1196, 1949).
from a group of 22 patients who had the Fontan operation for complex cardiac defects other than tricuspid atresia with ventriculoarterial concordance, The clinical summary is outlined in Table I. The ages ranged from 4 to 7 years. All had factors (pulmonary artery banding in three, varied forms of subaortic stenosis in three, and coarctation in one patient) that increase the risks associated with the Fontan operation.6, 9, lo The subaortic pressure gradient ranged from 35 to 80 torr. The average calculated pulmonary vascular resistance in three patients with subaortic obstructions was 1.9 U . m-. The left ventricular ejection fraction (range 0.67 to 0.73) and left ventricular end-diastolic pressure (range 3 to 12 torr) were within normal limits. One patient (Case 3) had relatively small pulmonary arteries despite two aortopulmonary shunts, mild pulmonary venous obstruction, and left bronchial stenosis.
All patients showed evidence of low cardiac output in the immediate postoperative period. Administration of inotropic agents and colloid and crystalloid fluids was started in the operating room and continued postoperatively. The arterial systolic pressure ranged from 55 to 82 torr (average 68 torr). The mean right atrial pressure ranged from 21 to 31 torr (average 24 torr). The urine output ranged from 0.1 to 5 rnI/kg/hr (average 1.7 ml) despite colloid and crystalloid infusions that ranged from 6,500 to 15,000 rnI/m 2/24 hr (average 11,000). Large doses of furosemide and mannitol were given without appreciable effect. All patients had significant pleural effusion and ascites necessitating increased tidal volumes resulting in higher airway pressure. Chest tubes were inserted for pleural effusion. Initial tidal volume and ventilating pressures were 10 nil/kg and 27 em H 20 , which were increased to 13 ml/kg and 37 em H 20 , respectively, for more effective ventilation just before
The Journal of
4 0 4 DeLeon et al.
takedown of the Fontan operation. All patients remained acidotic, requiring repeated bicarbonate administrations. Atelectasis of the left lung developed in one patient (Case 3) and persisted despite vigorous irrigation and suction of the tracheobronchial tree. Reoperations were performed 6 to 65 hours (average 23 hours) after the Fontan operation. At takedown, no mechanical obstruction was seen at the atriopulmonary anastomosis. In the two patients (Cases 1 and 2) who had a proximal main pulmonary artery-ascending aorta anastomosis to bypass the subaortic stenosis and simultaneous Glenn shunt with the Fontan procedure, takedown was accomplished by creating an atrial septal defect and aortopulmonary shunt to the left lung. In another patient (Case 4) who had a previous proximal main pulmonary artery-ascending aorta anastomosis, but in whom a Glenn shunt was not performed with the Fontan procedure, a Glenn shunt was performed in addition to creating an atrial septal defect and aortopulmonary shunt to the left lung. In the fourth patient, who at takedown of the Fontan operation had creation of an atrial septal defect and bilateral aortopulmonary shunts, a Glenn shunt could not be performed because of bilateral superior venae cavae. Two patients died after takedown of the Fontan shunt. One (Case 2) died intraoperatively and the other (Case 3) 8 hours postoperatively despite immediate improvement after takedown of the Fontan shunt. The systolic arterial pressure averaged 90 torr the first 4 hours and the right atrial pressure dropped to 14 torr. The left lung remained atelectatic, the patient's condition slowly deteriorated, and the child died. The two patients who survived the reoperation (Cases 1 and 4) had immediate improvement in systolic blood pressure (average 100 torr) and the right atrial pressure dropped to 11 and 12 torr, respectively. The ascites, urine output, and ventilatory parameters improved and both patients were extubated 1 and 3 days later, respectively. Both surviving patients are doing well 5 months and 4 years postoperatively. Discussion The morbidity and mortality remain high in patients undergoing the Fontan operation who have complex cardiac defects other than tricuspid atresia with ventriculoarterial concordance. I, 5-8 The majority of these patients are not ideal candidates for the Fontan operation by criteria outlined by Choussat and colleagues." In such patients several factors are usually present that result in higher than optimal pulmonary vascular resistance for the Fontan operation. A significant number of patients will have right atrial hypertension accompanied
Thoracic and Cardiovascular Surgery
by the shift of intravascular fluid into the interstitial space after the Fontan operation. Adequate cardiac output can be maintained only by infusion of large amounts of colloid and crystalloid fluid. Significant pleural and pericardial effusions and ascites usually develop, which will decrease lung compliance. In some patients the right atrial hypertension, development of marked ascites with elevation of the diaphragm, and decreasing lung compliance create a vicious cycle, resulting in persistent low cardiac output that makes survival unlikely unless the Fontan shunt is taken down. Such a cycle was present in our patients, and this along with persistent severe metabolic acidosis led us to believe that takedown of the Fontan shunt was the only way to save these patients. It is possible that the use of assisted circulation by phasic extemallower body compression as described by Heck and Doty" and also used by Laks and colleagues" could have been helpful. Although the mean right atrial pressures of their patients were lower (17 and 15 torr, respectively) than those of our patients (24 torr), the venous assist device could still have had a role in the management of our patients. Although the Glenn shunt was included in the first successful application of total right heart bypass," it is presently not considered essential for the Fontan operation. 2, 6, 14, 15 However, since the Glenn shunt preceded the Fontan operation by several years, numerous patients who had the Glenn shunt for certain complex cyanotic heart defects have subsequently undergone the Fontan operation. We" found that the morbidity and mortality of the Fontan operation in patients with established Glenn shunt is low. Pennington and colleagues" had similar experience. Because the majority of patients undergoing either the Glenn shunt or the Fontan operation have transient systemic venous hypertension, performing these two operations simultaneously offers no distinct advantage." Hopkins and colleagues" found that simultaneous bidirectional cavopulmonary shunt and Fontan operation carried an early mortality of 25%. We lO have proposed previously that a Glenn shunt should be considered before the Fontan operation in patients whose hemodynamics or anatomy is not ideal for the primary Fontan operation. However, as in other institutions, we have performed the Fontan operation without a prior Glenn shunt in certain patients from this group in an attempt to avoid a separate operation and the late sequelae associated with the Glenn shunt. All four of our patients who required takedown of the Fontan shunt were not ideal candidates for an orthoterminal procedure. Three patients had subaortic obstruction and pulmonary banding, and one patient had two previous shunts with relatively small and distorted
Volume 92 Number 3, Part 1
Fontan operation 405
September, 1986
pulmonary arteries.s? 10 Two of the four patients died, the only early deaths (4.5%) in our series of 44 patients undergoing the Fontan operation. Both patients who died (9%) were from a group of 22 patients undergoing the Fontan operation for complex defects other than tricuspid atresia with ventriculoarterial concordance. Because patients with an established Glenn shunt can undergo the Fontan operation with lower morbidity and mortality, we believe that the two surviving patients still can be considered for correction in the future. In summary, we believe that patients who have persistent low cardiac output after the Fontan operation should be considered for takedown of orthoterminal connections. Additionally, at takedown a Glenn shunt should be performed if possible so as not to preclude the possibility of a later Fontan operation.
2
3
4
5
6
REFERENCES Fontan F, Deville C, Quaegebeur J, Ottenkamp J, Sourdille N, Choussat A, Brom GA: Repair of tricuspid atresia in 100 patients. J THORAC CARDIOVASC SURG 85:647-660, 1983 Gale AW, Danielson GK, McGoon DC, Wallace RB, Mair DD: Fontan procedure for tricuspid atresia. Circulation 62:91-96, 1980 de Brux J-L, Zannini L, Binet J-P, Neveux J-Y, Langlois J, Hazan E, Planche C, Leca F, Marchand M: Tricuspid atresia. Results of treatment in 115 children. J THORAC CARDIOVASC SURG 85:440-446, 1983 Laks H, Williams WG, Hellenbrand WE, Freedom RM, Talner NS, Rowe RD, Trusler GA: Results of right atrial to right ventricular and right atrial to pulmonary artery conduits for complex congenital heart disease. Ann Surg 192:382-389, 1980 Kreutzer GO, Vargas FJ, Schlichter AD, Laura JP, Suarez JG, Coronel AR, Kreutzer EA: Atriopulmonary anastomosis. J THORAC CARDIOVASC SURG 83:427-436, 1982 Gale AW, Danielson GK, McGoon DC, Mair DD: Modified Fontan operation for univentricular heart and
7
8 9
10
II
12
13 14
15
16
17
complicated congenital lesions. J THORAC CARDIOVASC SURG 78:838-838, 1979 Marcelletti C, Mazzera E, Olthof H, Sebel PS, Duren DR, Losekoot TG, Becker AE: Fontan's operation. An expanded horizon. J THORAC CARDIOVASC SURG 80:764769, 1980 Danielson GK: Discussion of Fontan et al' Choussat A, Fontan F, Besse P, Vallot F, Chauve A, Bricaud H: Selection criteria for Fontan's procedure, Pediatric Cardiology, RH Anderson, EA Shinebourne, eds. Edinburgh, 1978, Churchill Livingstone, pp 559-566 DeLeon SY, Idriss FS, Ilbawi MN, Muster AJ, Paul MH, Cole RB, Riggs TW, Berry TE: The role of the Glenn shunt in patients undergoing the Fontan operation. J THORAC CARDIOVASC SURG 85:669-677, 1983 Heck HA Jr, Doty DB: Assisted circulation by phasic external lower body compression. Circulation 64Suppl 2:118-122, 1981 Laks H, Milliken JC, Perloff JK, Hellenbrand WE, George BL, Chin A, DiSessa TG, Williams RG: Experience with the Fontan procedure. J THORAC CARDIOVASC SURG 88:939-951, 1984 Fontan F, Baudet E: Surgical repair of tricuspid atresia. Thorax 26:240-248, 1971 Bjork VO, Olin CL, Bjarke BB, Thoren CA: Right atrial-right ventricular anastomosis for correction of tricuspid atresia. J THORAC CARDIOVASC SURG 77:452-458, 1979 Doty DB, Marvin WJ, Lauer RM: Modified Fontan procedure. Methods to achieve direct anastomosis of right atrium to pulmonary artery. J THORAC CARDIOVASC SURG 81:470-475,1981 Pennington DG, Nouri S, Ho J, Seeker-Walker R, Patel B, Sivakoff M, Willman VL: Glenn shunt. Long-term results and current role in congenital heart operations. Ann Thorac Surg 31:532-539, 1981 Hopkins RA, Armstrong BE, Serwer GA, Peterson RJ, Oldham HW Jr: Physiological rationale for a bidirection cavopulmonary shunt. A versatile complement to the Fontan Principle. J THORAC CARDIOVASC SURG 90:391398, 1985
THORAC CARDIOVASC SURG
92:402-405, 1986
Persistent low cardiac output after the Fontan operation Should takedown be considered? Four of 44 patients who bad undergone the Fontan operation bad persistent low cardiac output necessitating takedown of the shunt 6 to 65 hours (average 23 hours) postoperatively. All four were in a group of 22 patients with complexlesions other than tricuspidatresia withventriculoarterial concordance. The development of postoperative right atrial hypertension (average 24 torr), hepatomegaly, marked ascites, and decreasinglung compliance led to severe systemichypotension with systolic arterial pressure ranging from 55 to 82 torr (average 68 torr), persistent metabolic acidosis, and oliguria despite massive colloid and crystalloid infusions (11,000 m1/m2/24 hr) and inotropic support. At reoperation the atriopulmonaryanastomosis, which wasfoundto be wide open,was taken downand an atrial septal defect was created in all patients. Three patients were left with a Glennshunt and an aortopulmonary shunt to the left lung. One patient bad bilateral aortopulmonary shunts.Two patients whosurvived reoperationbad immediatepostoperative improvement in systolicarterial and mean right atrial pressure(average 100 torr and 11.5 torr, respectively). Both are well 5 monthsand 4 years later. Repeat Fontan operation remains a possibility with acceptable risks because of the presence of the Glennshunt in both patients. We believe that takedown should be considered in patients with persistent low cardiac output after the Fontan operation.
Serafin Y. DeLeon, M.D., Michel N. Ilbawi, M.D., Farouk S. Idriss, M.D., Alexander J. Muster, M.D., Samuel S. Gidding, M.D., Teresa E. Berry, M.D., and Milton H. Paul, M.D., Chicago. Ill.
Since the application of the fontan operation for tricuspid atresia was clearly established,'? its use has been extended to other complex congenital heart defects."? The morbidity and mortality, however, remain high (15% to 27%) in patients undergoing the Fontan operation who have complex lesions other than tricuspid atresia with ventriculoarterial concordance. 1, 5-8 Right atrial hypertension, severe loss of fluid into the intersti-
From the Divisions of Cardiovascular-Thoracic Surgery and Cardiology, The Children's Memorial Hospital, and the Departments of Surgery and Pediatrics, Northwestern University Medical School, Chicago, Ill. Supported in part by the A. C. Buehler Foundation, Park Ridge,
Ill. Received for publication Oct. 29, 1985. Accepted for publication Nov. 19, 1985. Address for reprints: Serafin Y. DeLeon, M.D., Division of Cardiovascular-Thoracic Surgery, 2300 Children's Plaza, Chicago, Ill. 60614.
402
tial space with development of pleural and pericardial effusion and ascites, and decreasing lung compliance are fairly common postoperative management problems. In four of 44 patients undergoing the Fontan operation, low cardiac output developed postoperatively, associated with right atrial hypertension, significant ascites, and decreasing lung compliance. Because the low cardiac output persisted in all patients despite massive fluid infusion and inotropic support, the Fontan shunt was taken down. To determine whether takedown of the Fontan shunt was worthwhile and should be considered under these circumstances, we reviewed our experience.
Patients Four of 44 patients who underwent the Fontan operation at the Children's Memorial Hospital in Chicago in the past 8 years had persistent low cardiac output in the immediate postoperative period necessitating takedown of the Fontan shunt. All four patients were
Volume 92 Number 3, Part 1
Fontan operation
September, 1986
403
Table I. Clinical summary of patients undergoing Fontan takedown Patient (DOB)
Diagnosis
Previous operation
1(9/11/74)
SV (S,L,L)t closing bulboventricular foramen (80 torr gradient)
PAB
2 ( 3/2/76)
Tricuspid atresia type II-C,t restricted VSD (80 torr gradient)
PAB, ligation PDA
3 ( 8/3/77)
Tricuspid atresia type II-A, bilateral SVC
Wa terston-Cooley shunt, left Blalock-Taussig shunt
Tricuspid atresia type II-C, restricted VSD (35 torr gradient), COA
COA repair, PAB, rebanding of PA, creation of APW
4 (10/8/81)
Fontan (date)*
9/2/81: Closure of TV, direct RA-PA anastomosis with 25 mm porcine valve, proximal MPA-AO anastomosis, Glenn shunt 12/2/80: Direct RA-PA anastomosis, proximal MPA-AO anastomosis, Glenn shunt 8/3/83: Direct RA-PA anastomosis, takedown of shunts 5/8/85: Direct RA-PA anastomosis
Takedown (date)
Outcome
9/5/81: Takedown RA-PA anastomosis, removal porcine valve, creation ASD, AO-LPA shunt
Well at 4 yr
12/3/80: Takedown RA-PA anastomosis, creation ASD, AO-LPA shunt
Died intraoperatively
8/4/83: Takedown of RA-PA anastomosis, creation ASD, bilateral AO-PA shunts 5/8/85: Takedown of RA-PA anastomosis, creation ASD, AO-LPA shunt, Glenn shunt
Died 8 hr
postoperatively
Well at 5 mo
Legend:AO-LPA, Aorta-left pulmonary artery. AO-PA, Aorta-pulmonary artery. APW, Aortopulmonary window. ASD, Atrial septal defect. COA, Coarctation of the aorta. DOB, Date of birth. MPA-AO, Main pulmonary artery-aorta. PAB, Pulmonary artery banding. RA-PA, Atriopulmonary anastomosis. SV, Single ventricle. SVC, Superior vena cava. TV, Tricuspid valve. VSD, Ventricular septal defect. 'The coronary sinus drainage was to the right atrium in all patients. tBased on Van Praagh's classification of malposition of the heart, Heart Disease in Infants, Children, and Adolescents, ed 3, FH Adams, GC Emmanouilides, eds., Baltimore, 1983, The Williams & Wilkins Company, pp 422-458. tBased on classification by Edwards and Burchell (Edwards lE, Burchell HB: Congenital tricuspid atresia. A classification. Med Clin North Am 33:1177-1196, 1949).
from a group of 22 patients who had the Fontan operation for complex cardiac defects other than tricuspid atresia with ventriculoarterial concordance, The clinical summary is outlined in Table I. The ages ranged from 4 to 7 years. All had factors (pulmonary artery banding in three, varied forms of subaortic stenosis in three, and coarctation in one patient) that increase the risks associated with the Fontan operation.6, 9, lo The subaortic pressure gradient ranged from 35 to 80 torr. The average calculated pulmonary vascular resistance in three patients with subaortic obstructions was 1.9 U . m-. The left ventricular ejection fraction (range 0.67 to 0.73) and left ventricular end-diastolic pressure (range 3 to 12 torr) were within normal limits. One patient (Case 3) had relatively small pulmonary arteries despite two aortopulmonary shunts, mild pulmonary venous obstruction, and left bronchial stenosis.
All patients showed evidence of low cardiac output in the immediate postoperative period. Administration of inotropic agents and colloid and crystalloid fluids was started in the operating room and continued postoperatively. The arterial systolic pressure ranged from 55 to 82 torr (average 68 torr). The mean right atrial pressure ranged from 21 to 31 torr (average 24 torr). The urine output ranged from 0.1 to 5 rnI/kg/hr (average 1.7 ml) despite colloid and crystalloid infusions that ranged from 6,500 to 15,000 rnI/m 2/24 hr (average 11,000). Large doses of furosemide and mannitol were given without appreciable effect. All patients had significant pleural effusion and ascites necessitating increased tidal volumes resulting in higher airway pressure. Chest tubes were inserted for pleural effusion. Initial tidal volume and ventilating pressures were 10 nil/kg and 27 em H 20 , which were increased to 13 ml/kg and 37 em H 20 , respectively, for more effective ventilation just before
The Journal of
4 0 4 DeLeon et al.
takedown of the Fontan operation. All patients remained acidotic, requiring repeated bicarbonate administrations. Atelectasis of the left lung developed in one patient (Case 3) and persisted despite vigorous irrigation and suction of the tracheobronchial tree. Reoperations were performed 6 to 65 hours (average 23 hours) after the Fontan operation. At takedown, no mechanical obstruction was seen at the atriopulmonary anastomosis. In the two patients (Cases 1 and 2) who had a proximal main pulmonary artery-ascending aorta anastomosis to bypass the subaortic stenosis and simultaneous Glenn shunt with the Fontan procedure, takedown was accomplished by creating an atrial septal defect and aortopulmonary shunt to the left lung. In another patient (Case 4) who had a previous proximal main pulmonary artery-ascending aorta anastomosis, but in whom a Glenn shunt was not performed with the Fontan procedure, a Glenn shunt was performed in addition to creating an atrial septal defect and aortopulmonary shunt to the left lung. In the fourth patient, who at takedown of the Fontan operation had creation of an atrial septal defect and bilateral aortopulmonary shunts, a Glenn shunt could not be performed because of bilateral superior venae cavae. Two patients died after takedown of the Fontan shunt. One (Case 2) died intraoperatively and the other (Case 3) 8 hours postoperatively despite immediate improvement after takedown of the Fontan shunt. The systolic arterial pressure averaged 90 torr the first 4 hours and the right atrial pressure dropped to 14 torr. The left lung remained atelectatic, the patient's condition slowly deteriorated, and the child died. The two patients who survived the reoperation (Cases 1 and 4) had immediate improvement in systolic blood pressure (average 100 torr) and the right atrial pressure dropped to 11 and 12 torr, respectively. The ascites, urine output, and ventilatory parameters improved and both patients were extubated 1 and 3 days later, respectively. Both surviving patients are doing well 5 months and 4 years postoperatively. Discussion The morbidity and mortality remain high in patients undergoing the Fontan operation who have complex cardiac defects other than tricuspid atresia with ventriculoarterial concordance. I, 5-8 The majority of these patients are not ideal candidates for the Fontan operation by criteria outlined by Choussat and colleagues." In such patients several factors are usually present that result in higher than optimal pulmonary vascular resistance for the Fontan operation. A significant number of patients will have right atrial hypertension accompanied
Thoracic and Cardiovascular Surgery
by the shift of intravascular fluid into the interstitial space after the Fontan operation. Adequate cardiac output can be maintained only by infusion of large amounts of colloid and crystalloid fluid. Significant pleural and pericardial effusions and ascites usually develop, which will decrease lung compliance. In some patients the right atrial hypertension, development of marked ascites with elevation of the diaphragm, and decreasing lung compliance create a vicious cycle, resulting in persistent low cardiac output that makes survival unlikely unless the Fontan shunt is taken down. Such a cycle was present in our patients, and this along with persistent severe metabolic acidosis led us to believe that takedown of the Fontan shunt was the only way to save these patients. It is possible that the use of assisted circulation by phasic extemallower body compression as described by Heck and Doty" and also used by Laks and colleagues" could have been helpful. Although the mean right atrial pressures of their patients were lower (17 and 15 torr, respectively) than those of our patients (24 torr), the venous assist device could still have had a role in the management of our patients. Although the Glenn shunt was included in the first successful application of total right heart bypass," it is presently not considered essential for the Fontan operation. 2, 6, 14, 15 However, since the Glenn shunt preceded the Fontan operation by several years, numerous patients who had the Glenn shunt for certain complex cyanotic heart defects have subsequently undergone the Fontan operation. We" found that the morbidity and mortality of the Fontan operation in patients with established Glenn shunt is low. Pennington and colleagues" had similar experience. Because the majority of patients undergoing either the Glenn shunt or the Fontan operation have transient systemic venous hypertension, performing these two operations simultaneously offers no distinct advantage." Hopkins and colleagues" found that simultaneous bidirectional cavopulmonary shunt and Fontan operation carried an early mortality of 25%. We lO have proposed previously that a Glenn shunt should be considered before the Fontan operation in patients whose hemodynamics or anatomy is not ideal for the primary Fontan operation. However, as in other institutions, we have performed the Fontan operation without a prior Glenn shunt in certain patients from this group in an attempt to avoid a separate operation and the late sequelae associated with the Glenn shunt. All four of our patients who required takedown of the Fontan shunt were not ideal candidates for an orthoterminal procedure. Three patients had subaortic obstruction and pulmonary banding, and one patient had two previous shunts with relatively small and distorted
Volume 92 Number 3, Part 1
Fontan operation 405
September, 1986
pulmonary arteries.s? 10 Two of the four patients died, the only early deaths (4.5%) in our series of 44 patients undergoing the Fontan operation. Both patients who died (9%) were from a group of 22 patients undergoing the Fontan operation for complex defects other than tricuspid atresia with ventriculoarterial concordance. Because patients with an established Glenn shunt can undergo the Fontan operation with lower morbidity and mortality, we believe that the two surviving patients still can be considered for correction in the future. In summary, we believe that patients who have persistent low cardiac output after the Fontan operation should be considered for takedown of orthoterminal connections. Additionally, at takedown a Glenn shunt should be performed if possible so as not to preclude the possibility of a later Fontan operation.
2
3
4
5
6
REFERENCES Fontan F, Deville C, Quaegebeur J, Ottenkamp J, Sourdille N, Choussat A, Brom GA: Repair of tricuspid atresia in 100 patients. J THORAC CARDIOVASC SURG 85:647-660, 1983 Gale AW, Danielson GK, McGoon DC, Wallace RB, Mair DD: Fontan procedure for tricuspid atresia. Circulation 62:91-96, 1980 de Brux J-L, Zannini L, Binet J-P, Neveux J-Y, Langlois J, Hazan E, Planche C, Leca F, Marchand M: Tricuspid atresia. Results of treatment in 115 children. J THORAC CARDIOVASC SURG 85:440-446, 1983 Laks H, Williams WG, Hellenbrand WE, Freedom RM, Talner NS, Rowe RD, Trusler GA: Results of right atrial to right ventricular and right atrial to pulmonary artery conduits for complex congenital heart disease. Ann Surg 192:382-389, 1980 Kreutzer GO, Vargas FJ, Schlichter AD, Laura JP, Suarez JG, Coronel AR, Kreutzer EA: Atriopulmonary anastomosis. J THORAC CARDIOVASC SURG 83:427-436, 1982 Gale AW, Danielson GK, McGoon DC, Mair DD: Modified Fontan operation for univentricular heart and
7
8 9
10
II
12
13 14
15
16
17
complicated congenital lesions. J THORAC CARDIOVASC SURG 78:838-838, 1979 Marcelletti C, Mazzera E, Olthof H, Sebel PS, Duren DR, Losekoot TG, Becker AE: Fontan's operation. An expanded horizon. J THORAC CARDIOVASC SURG 80:764769, 1980 Danielson GK: Discussion of Fontan et al' Choussat A, Fontan F, Besse P, Vallot F, Chauve A, Bricaud H: Selection criteria for Fontan's procedure, Pediatric Cardiology, RH Anderson, EA Shinebourne, eds. Edinburgh, 1978, Churchill Livingstone, pp 559-566 DeLeon SY, Idriss FS, Ilbawi MN, Muster AJ, Paul MH, Cole RB, Riggs TW, Berry TE: The role of the Glenn shunt in patients undergoing the Fontan operation. J THORAC CARDIOVASC SURG 85:669-677, 1983 Heck HA Jr, Doty DB: Assisted circulation by phasic external lower body compression. Circulation 64Suppl 2:118-122, 1981 Laks H, Milliken JC, Perloff JK, Hellenbrand WE, George BL, Chin A, DiSessa TG, Williams RG: Experience with the Fontan procedure. J THORAC CARDIOVASC SURG 88:939-951, 1984 Fontan F, Baudet E: Surgical repair of tricuspid atresia. Thorax 26:240-248, 1971 Bjork VO, Olin CL, Bjarke BB, Thoren CA: Right atrial-right ventricular anastomosis for correction of tricuspid atresia. J THORAC CARDIOVASC SURG 77:452-458, 1979 Doty DB, Marvin WJ, Lauer RM: Modified Fontan procedure. Methods to achieve direct anastomosis of right atrium to pulmonary artery. J THORAC CARDIOVASC SURG 81:470-475,1981 Pennington DG, Nouri S, Ho J, Seeker-Walker R, Patel B, Sivakoff M, Willman VL: Glenn shunt. Long-term results and current role in congenital heart operations. Ann Thorac Surg 31:532-539, 1981 Hopkins RA, Armstrong BE, Serwer GA, Peterson RJ, Oldham HW Jr: Physiological rationale for a bidirection cavopulmonary shunt. A versatile complement to the Fontan Principle. J THORAC CARDIOVASC SURG 90:391398, 1985