Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Peri-operative comparison of different transient external shunt techniques in bidirectional cavo-pulmonary shunt

Peri-operative comparison of different transient external shunt techniques in bidirectional... Abstract Objective: In patients with functional single ventricular physiology, the avoidance of cardiopulmonary bypass offers many advantages including earlier extubation, decreased necessity of inotropic support, improved hemodynamical status and reduced likehood of post-operative prolonged pleural effusion. We believe that the bidirectional cavopulmonary anastomosis operations may be performed with transient external shunt techniques. The purpose of this prospective study is the peri- and post-operative comparison of different transient external shunt methods used in bidirectional cavopulmonary shunt operations. Methods: Between years 1997 and 2000, 30 patients have undergone bidirectional cavo-pulmonary shunt operation by using three different types of external shunt. The mean patient age was 13 months (range, 3 months–3 years). Previous operations had been performed in ten patients (33%). All patients were divided into three groups according to type of external shunt used. In group A (ten patients), the transient external shunt was constructed between superior vena cava and right atrium by uniting two standard venous cannulas with a Y-connector. In group B (ten patients), the external shunt was performed with a single short venous cannula constructed between superior vena cava and right atrium. In group C (ten patients), the external shunt was constructed between superior vena cava and left pulmonary artery by using a single short venous cannula. During operation, central venous pressure (CVP), arterial O2 saturation and mean arterial blood pressure were recorded continuously. Results: All operations are completed without the establishment of cardiopulmonary bypass. Hospital mortality was 3.3%. One patient in group A died because of low cardiac output at the end of postoperative day 2. All patients were extubated within 4 h. In groups A–C mean superior vena caval pressures were measured 28, 24 and 21 mmHg, respectively during superior vena cava-right pulmonary artery anastomosis. In both groups A and B patients, arterial O2 saturation decreased to a minimum 53±2 and 53±2%, respectively during the operation. In the group C, minimum arterial O2 saturation was measured 82±2%. Although mean arterial pressure decreased in all groups during clampage; in group C patients, this drop is not significant. Conclusion: Based on the study presented here, bidirectional cavo-pulmonary anastomosis can be carried out by using different types of transient external shunt. The best hemodynamical condition and arterial O2 levels were achieved with the shunt constructed between superior vena cava and left pulmonary artery. Bidirectional cavopulmonary shunt, Transient external shunts 1 Introduction Bidirectional cavo-pulmonary shunt operation is one of the palliative procedures in patients with functional single ventricular physiology. With increased knowledge about the detrimental effects of cardiopulmonary bypass on patients with functional single ventricle, many surgeons had begun to search for new and safer methods. Since the first suggestion made by Lambarti et al. [1], transient external shunt techniques used for caval drainage had become increasingly popular. However, criteria for performing the bidirectional cavo-pulmonary shunt without cardiopulmonary bypass are not well established. Some surgeons have employed this approach in patients at particular risk for cardiopulmonary bypass [2]. Another controversy is about the type of external shunt used for caval drainage. In this prospective study, we wanted to compare the peri- and post-operative outcome of three different types of external shunt used in 30 patients who had undergone bidirectional cavo-pulmonary shunt operation. This analysis focuses on the technical aspects of the procedure and reviews the short-term morbidity and mortality. 2 Materials and methods Between February 1997 and December 2000, 30 patients have undergone bidirectional cavo-pulmonary anastomosis operation with transient external shunts at Medical Faculty of Istanbul University, Department of Cardiovascular Surgery. The mean patient age was 13 months (range, 3 months–3 years). Seventeen patients were under 1 year of age. Weights ranged from 6 to 28–kg. Primary diagnosis and previous operations were listed in Table 1 . Previous operations had been performed in ten patients (33→7% patients systemic to pulmonary artery shunt, two patients pulmonary arterial banding and one patient pulmonary arterial banding and one patient pulmonary arterial banding+repair of aortic coarctation and arcus hypoplasia). All operations were performed without cardiopulmonary bypass. The patients requiring any additional intracardiac intervention were not included in the study. Indications for performing the bidirectional cavo-pulmonary shunt operation without cardiopulmonary bypass are the presence of an unrestrictive atrial septal defect, the absence of atrioventricular valve regurgitation and any other additional intracardiac defects requiring correction. Bidirectional cavo-pulmonary shunt was used as a first stage of complete Fontan operation for all patients. Table 1 Open in new tabDownload slide Clinical characteristics and previous operations Table 1 Open in new tabDownload slide Clinical characteristics and previous operations In this prospective study, 30 patients were divided into three groups according to the type of external shunt used for caval drainage (Table 2 ): Table 2 Open in new tabDownload slide Type and size of external shunts Table 2 Open in new tabDownload slide Type and size of external shunts Group A (ten patients): external shunt was established between superior vena cava and right atrium by using two standard venous cannulas connected with a Y connector. Group B (ten patients): external shunt was established between superior vena cava and right atrium by using a single short standard venous cannula. Group C (ten patients): external shunt was established between superior vena cava and left pulmonary artery by using a standard venous cannula. Our strategy in systemic to pulmonary artery shunts is the placement of polytetrafluoroethylene (PTFE) graft from the innominate artery or ascending aorta to the main pulmonary artery. In group C, purse string sutures were placed on the main pulmonary artery and venous cannula was moved into the left pulmonary artery. Since group C patients require the presence of intact and undistorted native pulmonary arteries, we made a decision to create this group among the patients without the history of previous pulmonary arterial banding or systemic to pulmonary artery shunt operations. 2.1 Surgical technique Through a median sternotomy and peri-cardiotomy, superior venae cava (SVC) and pulmonary arteries were dissected free. Azygos vein was divided to allow mobilization. Patent ductus arteriosus (PDA) and if present, previous modified Blalock Taussig (BT) shunt were dissected, controlled; but, not ligated to improve the arterial oxygenation during the reconstruction of anastomosis. In cases with right-sided PDA (two patients in group A), ductus had to be ligated before the anastomosis. Fortunately, we did not notice any deterioration in both patients following the ligation of PDA. Systemic heparin (1 mg/kg) was administered intravenously before the procedure. In group A, purse-string sutures were placed in the right atrium and superior vena cava and a transient external shunt was created between SVC and right atrium by using two standard venous cannulas connected with a Y connector (Fig. 1a ). In group B, external shunt was established between distal part of SVC and right atrium with a single short straight venous cannula (Fig. 1b). In group C, purse-string sutures were placed longitudinally in the distal part of SVC and main pulmonary artery (MPA). The transient shunt was created between SVC and left pulmonary artery by pushing the pulmonary arterial end of short venous cannula into the left pulmonary artery (Fig. 1c). The short straight venous cannulas utilized in groups B and C were created by cutting one end of the cannula obliquely to facilitate the placement of the cannula into right atrium and left pulmonary artery. The size of the venous cannulas was given in Table 2. To prevent the possible bias caused by the length of venous cannulas, we used the venous cannulas with the same length (8 cm) in all patients belonging to groups B and C. Venous cannulas were filled with blood to allow deairing and placed parallel to the patient to overcome the negative effects of gravity. Superior vena cava was clamped and divided from cavoatrial junction with taking care to avoid sinus node. The atrial end was oversewn. During clampage, superior vena caval pressures above the clamp and arterial O2 saturations were recorded in all patients. Right pulmonary artery (RPA) was clamped and arteriotomy was performed on the superior aspect. The distal end of the SVC was anastomosed to the RPA with end-to-side technique. Two separate 7-0 polypropylene sutures were used to prevent the purse-string effect. After completion of the anastomosis, external shunt was disconnected and blood in the cannula was drained into the SVC, right atrium and PA. To prevent the future stenosis, the purse-string sutures placed on the SVC and main pulmonary artery were not tied. The defect on the SVC was repaired primarily with a 7-0 polypropylene suture. In seven patients from group C with an adequate sized main pulmonary artery, a decision was made to close the defect primarily. However, for the remaining three patients with a younger age (<9 months) and small sized main pulmonary artery, we used a pericardial patch to repair the defect. During the entire procedure, the patients' heads were elevated. Fig. 1 Open in new tabDownload slide (a) External shunt between SVC and right atrium by using 2 cannulas connected with a Y connector in Group A. (b) External shunt between SVC and right atrium by using a short straight venous cannula. (c) External shunt between SVC and left pulmonary artery by using a standard venous cannula. Fig. 1 Open in new tabDownload slide (a) External shunt between SVC and right atrium by using 2 cannulas connected with a Y connector in Group A. (b) External shunt between SVC and right atrium by using a short straight venous cannula. (c) External shunt between SVC and left pulmonary artery by using a standard venous cannula. Before operations, basal superior vena caval pressure (SVCP0A,B,C), arterial O2 saturation level (S0A,B,C) , mean arterial pressure (P0A,B,C) were recorded in all groups. During the clampage of superior vena cava, superior vena caval pressures were recorded continouosly and first (SVCP1A,B,C) , fifth (SVCP2A,B,C) and tenth (SVCP3A,B,C) minute pressures were compared. During clampage, minimum arterial O2 saturation (S1A,B,C) levels and arterial blood pressures (P1A,B,C) were also recorded in all groups. 2.2 Statistical analysis Statistical analysis was performed with analysis of variance (ANOVA) test. P values of <0.1 were considered significant. In situations with significant P value, post-test was performed with Tukey. Comparisons between two parameters were performed with student t-test. 3 Results Hospital mortality was 3.3%. All patients tolerated the procedure well and their hemodynamical statuses were not deteriorated during the reconstruction of anastomosis. Cardiopulmonary bypass (CPB) was not used in any step of operation. Dopamine or α-agonist agents were not required to increase blood pressure. The duration of operations ranged from 85 to 164 min (mean 122 min) in group A, 83 to 159 min (mean 120 min) in group B and 89 to 168 min (mean 124 min) in group C. Duration of anastomosis between SVC and RPA ranged from 10 to 17 min (mean 13 min) in all patients. The predictors of caval drainage during the reconstruction of the anatomosis are the pressure difference between SVC and right atrium or pulmonary artery, the central gravity, the radius and length of venous cannulas that were used. We used the venous cannulas with the same length (8 cm) in all patients belonging to groups B and C to prevent the possible bias that may be caused by the length of venous cannulas. According to Poseuilles formula, the most important determinant of flow is the size of the cannula because of the strong inverse relationship between flow and fourth power of radius. The size of the cannula depends on the weight and age of the patients. We generally tried to use the same sized (18 F) venous cannula in all patients (eight patients in group A, eight patients in group B and eight patients in group C). Superior vena caval pressures increased in all three groups; but, the most effective caval drainage was performed with the shunt constructed between superior vena cava and left pulmonary artery (P<0.001 for SVCP1A versus SVCP1C and P<0.01 for SVCP1B and SVCP1C). In group A patients, SVCP1A, SVCP2A and VCP3A levels were measured 28±2 mmHg. In group B patients, SVCP1B, SVCP2B and SVCP3B levels were measured 24±2 mmHg. In group C patients, SVCP1C, SVCP2C and SVCP3C levels were measured 21±1, 21±1, 22±1 mmHg, respectively. Superior vena caval pressure changes during clampage were given in Fig. 2 . We also compared the superior vena caval pressures of the 24 patients in whom we used the same sized cannula and this is shown in Fig. 3 . Basal arterial O2 saturations were 64±2% for Groups A–C. There was no significant difference in terms of basal arterial O2 levels among three groups. However, arterial O2 saturations were increased from 64±2 (S0C) to 82±2% (S1C) in group C patients during clampage. This elevation is significant when compared to groups A and B (P<0.001 for group A versus group C and P<0.001 for groups B and C) (Fig. 4 ). In groups A and B patients, basal arterial blood pressures were recorded 66±5 (POA) and 67±3 mmHg (P0B), respectively before the procedure. During clampage, minimum arterial blood pressures were recorded 56±3 mmHg (P1A) for group A patients and 57±2 mmHg (P1B) for group B patients (P<0.0001 for both groups). In group C, basal arterial blood pressure was 68±2 mmHg (P0C). During clampage, minimum arterial blood pressure was measured 66±3 mmHg for group C patients (P1C). This decrease is not significant. No blood transfusions were required during operation and post-operative period. One patient (in group A) diagnosed with double outlet right ventricle, ventricular septal defect, subaortic conus, mitral atresia and previous pulmonary banding operation developed low cardiac output and died at the end of post-operative day 2. Other patients were extubated in the first 4 h. Hemodynamical studies revealed a mean post-operative central venous pressure of 13±2 mmHg. Arterial oxygen saturation increased from 65±5 to 91±4% post-operatively. Average durations of intensive care unit and hospital stays were 36 h and 8 days, respectively. Two patients suffered from prolonged pleural effusion post-operatively. Post-operative echocardiography and angiography demonstrated a functional cavopulmonary anastomosis in all patients. Fig. 2 Open in new tabDownload slide Recorded superior vena caval pressure changes during anastomosis in all three groups (0, 1, 5, 10 minutes). Fig. 2 Open in new tabDownload slide Recorded superior vena caval pressure changes during anastomosis in all three groups (0, 1, 5, 10 minutes). Fig. 3 Open in new tabDownload slide The comparison of patients operated by using the same size of venous cannula (Group A, eight patients; Group B, eight patients; Group C, eight patients. In SVCP0 the difference is not significant. In SVCP1 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP2 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP3 P<0.001 for groups A versus C; P<0.01 for groups B and C). Fig. 3 Open in new tabDownload slide The comparison of patients operated by using the same size of venous cannula (Group A, eight patients; Group B, eight patients; Group C, eight patients. In SVCP0 the difference is not significant. In SVCP1 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP2 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP3 P<0.001 for groups A versus C; P<0.01 for groups B and C). Fig. 4 Open in new tabDownload slide Arterial O2 saturations during anastomosis. Fig. 4 Open in new tabDownload slide Arterial O2 saturations during anastomosis. The ineffective vena caval drainage during the clampage may lead an acute increase in intracranial pressure. Following extubation, a full neurological examination implicating motor and sensory systems was performed by the same physician. With increasing knowledge and experience, we learned that the most important and earlier sign of increased intracranial pressure is papilledema. The degree of venous engorgement was determined with ophtalmologic evaluation. Two patients with significant papilledema were medically treated (dexamethasone 0.2 mg/kg/IV/6 h, furosemide 1.5 mg/kg /IV/6 h and mannitol 0.2 g/kg/IV/6 h) and followed with periodic neurological examination and computed tomographic (CT) scanning until discharge. Full recovery was detected in all patients (range 6–11 days). We had three patients with seizure activity during hospitalization period. Two of them (one from group A and one from group B) were febrile in nature and improved with anti-epileptic therapy. However, the activity of the remaining patient from group A became permanent and this patient is still following by our pediatric neurology department. 4 Conclusion Bidirectional cavo-pulmonary shunt operation is a universally accepted palliative procedure in patients with functional single ventricle [3]. It preserves ventricular and atrioventricular valve function by decreasing volume overload on single ventricle. It does not cause any distortion in pulmonary arteries and possess no risk of precipitating pulmonary vascular obstructive disease. It is generally considered as a second line operation after some other forms of palliative procedures (either systemic to PA shunt or pulmonary banding procedure) performed during neonatal or early infancy period [1]. Bidirectional cavo-pulmonary shunt operation was generally performed by using cardiopulmonary bypass. However, Jahangiri et al. reported seven bidirectional Glenn shunt procedures performed without cardiopulmonary bypass through a thoracotomy. They clamped directly superior vena cava and did not use any transient external shunt [4]. In their series, mean right internal jugular venous pressure was measured 26 mmHg and dopamine or α-agonist agents were required only in three patients to increase blood pressure. They encountered no neurological complications and reported that establishing an intraoperative shunt was unnecessary. With increasing knowledge about the detrimental effects of cardiopulmonary bypass on infants, Lamberti and et al. [1] searched for a new approach and performed their operations by a right thoracotomy. In their experience of seven patients, they could decrease caval pressure about 15 mmHg by the help of an external shunt. Subsequent studies also demonstrated that bidirectional cavo-pulmonary shunt and even extracardiac Fontan operations might be executed without cardiopulmonary bypass [2,5,8]. Cherian et al. performed bidirectional Glenn shunt operation by using different types of transient external shunt [5]. In their series of five patients, they demonstrated that the avoidance of cardiopulmonary bypass and aortic cross-clamping has the advantages including earlier extubation, lesser usage of blood products, decreased necessity and duration of inotropic support. Mohan Reddy et al. demonstrated a post-operative increase in pulmonary vascular resistance and hypoxia after cardiopulmonary bypass and reported that transient external shunt approach must be indicated on a wider basis [6]. Some other authors demonstrated that aortopulmonary shunts might occur after establishing cardiopulmonary bypass and this might cause prolonged pleural effusions [7,9]. In our department, we perform bidirectional Glenn and extracardiac Fontan operations without cardiopulmonary bypass in patients without additional intracardiac intervention. In our series of 30 patients, during post-operative follow-up period, hypoxia and increase in central venous and pulmonary artery pressures were not detected. All patients were weaned off mechanical ventilator during early post-operative hours; except one patient who died because of low cardiac output. We tried to decrease mean superior vena caval pressures by external shunts to prevent neurological damage during clampage. Although uppermost level of safe central venous pressure is not still clear, we demonstrated that lowest venous pressures were detected in groups B and C. We conclude that short venous cannulas decrease the resistance against blood flow in groups B and C. Length of cannulas in groups B and C were half of the ones in group A. Some authors demonstrated reduction in diastolic and mean peak systolic flow velocities more than 50% in middle cerebral artery by transcranial Doppler ultrasonography and detected mild electrocortical alterations in evoked potentials. They reported that such deteriorations did not occur in patients operated with cardiopulmonary bypass [10,11]. In subsequent studies, it was demonstrated that electrocortical alterations improved by decreasing superior vena caval pressure with an external shunt. In our series, we also decreased superior vena caval pressure by using an external shunt. Transcranial Doppler ultrasonography or electroencephalography was not used intra-operatively. Some authors reported that cavoatrial shunt techniques might compromise the quality of cavo-pulmonary anastomosis and restrict the cerebral blood flow in cyanotic children [12]. In group C patients, after establishing a short venous cannula between superior vena cava and left pulmonary artery, we observed an increase in arterial oxygen saturations. We consider that this technique may increase cerebral blood flow. As a conclusion, bidirectional cavo-pulmonary shunt operation may be performed with different external shunt techniques. Among different types, external shunt constructed between superior vena cava and pulmonary artery provides better hemodynamics and reduces surgical stress by increasing the O2 saturation levels during anastomosis. References [1] Lamberti J.J. , Spicer R.L. , Waldman J.D. , Grehl T.M. , Thomson D. , George L. , Kirkpatrick S.E. , Mathewson J.W. . The bidirectional cavopulmonary shunt , J Thorac Cardiovasc Surg , 1990 , vol. 100 (pg. 22 - 29 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat [2] Burke R.P. , Jacobs J.P. , Ashraf M.H. , Aldousany A. , Chang A.C. . Extracardiac Fontan operation without cardiopulmonary bypass , Ann Thorac Surg , 1997 , vol. 63 (pg. 1175 - 1177 ) Google Scholar Crossref Search ADS PubMed WorldCat [3] Xie Bin M.D. , Zhang Jin Fang M.D. , Devi Prasad Shetty M.S. . Bidirectional Glenn shunt: 170 cases , Asian Cardiovasc Thorac Ann , 2001 , vol. 9 (pg. 196 - 199 ) Google Scholar Crossref Search ADS WorldCat [4] Jahangiri M. , Keogh B. , Shinebourne E.A. , Lincoln C. . Should the bidirectional Glenn procedure be performed through a thoracotomy without cardiopulmonary bypass? , J Thorac Cardiovasc Surg , 1999 , vol. 118 (pg. 367 - 368 ) Google Scholar Crossref Search ADS PubMed WorldCat [5] Murthy K.S. , Coelho R. , Naik S.K. , Punnoose A. , Thomas W. , Cherian K.M. . Novel techniques of bidirectional Glenn shunt without cardiopulmonary bypass , Ann Thorac Surg , 1999 , vol. 67 (pg. 1771 - 1774 ) Google Scholar Crossref Search ADS PubMed WorldCat [6] Reddy V.M. , Liddicoat J.R. , Hanley F.L. . Primary bidirectional superior cavopulmonary shunt in infants between 1 and 4 months of age , Ann Thorac Surg , 1995 , vol. 59 (pg. 1120 - 1125 ) Google Scholar Crossref Search ADS PubMed WorldCat [7] van de Wal H.J.C.M. , Ouknine R. , Tamisier D. , Lévy M. , Vouhè P.R. , Leca F. . Bi-directional cavopulmonary shunt: is accessory pulsatile flow, good or bad? , Eur J Cardiothorac Surg , 1999 , vol. 16 (pg. 104 - 110 ) Google Scholar Crossref Search ADS PubMed WorldCat [8] Uemura H. , Yagihara T. , Yamashita K. , Ishizaka T. , Yoshizumi K. , Kawahira Y. . Establishment of total cavopulmonary connection without use of cardiopulmonary bypass , Eur J Cardiothorac Surg , 1998 , vol. 13 (pg. 504 - 507 ) Google Scholar Crossref Search ADS PubMed WorldCat [9] Kim S.J. , Bae E.J. , Cho D.J. , Park I.S. , Kim Y.M. , Kim W.-H. , Kim S.H. . Development of pulmonary arteriovenous fistulas after bidirectional cavopulmonary shunt , Ann Thorac Surg , 2000 , vol. 70 (pg. 1918 - 1922 ) Google Scholar Crossref Search ADS PubMed WorldCat [10] Zimmerman A.A. , Burrows F.A. , Jonas R.A. , Hickey P.R. . The limits of detectable cerebral perfusion by transcranial doppler sonography in neonates undergoing deep hypothermic low-flow cardiopulmonary bypass , J Thorac Cardiovasc Surg , 1997 , vol. 114 (pg. 594 - 600 ) Google Scholar Crossref Search ADS PubMed WorldCat [11] Rodriguez R.A. , Cornel G. , Semelhago L. , Spinter W.M. , Weerasena N.A. . Cerebral effects in superior vena caval cannula obstruction: the role of brain monitoring , Ann Thorac Surg , 1997 , vol. 64 (pg. 1820 - 1822 ) Google Scholar Crossref Search ADS PubMed WorldCat [12] Stamm C. , Friehs I. , Mayer J.E. Jr , Zurakowski D. , Triedman J.K. , Moran A.M. , Walsh E.P. , Lock J.E. , Jonas R.A. , Del Nido P.J. . Long-term results of the lateral tunnel Fontan operation , J Thorac Cardiovasc Surg , 2001 , vol. 121 (pg. 28 - 41 ) Google Scholar Crossref Search ADS PubMed WorldCat © 2003 Elsevier Science B.V. Elsevier Science B.V. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Peri-operative comparison of different transient external shunt techniques in bidirectional cavo-pulmonary shunt

Download PDF
7 pages

Loading next page...
 
/lp/oxford-university-press/peri-operative-comparison-of-different-transient-external-shunt-KdYf9RYT0M

References (12)

Publisher
Oxford University Press
Copyright
© 2003 Elsevier Science B.V.
Subject
Original Articles
ISSN
1010-7940
eISSN
1873-734X
DOI
10.1016/S1010-7940(03)00012-5
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective: In patients with functional single ventricular physiology, the avoidance of cardiopulmonary bypass offers many advantages including earlier extubation, decreased necessity of inotropic support, improved hemodynamical status and reduced likehood of post-operative prolonged pleural effusion. We believe that the bidirectional cavopulmonary anastomosis operations may be performed with transient external shunt techniques. The purpose of this prospective study is the peri- and post-operative comparison of different transient external shunt methods used in bidirectional cavopulmonary shunt operations. Methods: Between years 1997 and 2000, 30 patients have undergone bidirectional cavo-pulmonary shunt operation by using three different types of external shunt. The mean patient age was 13 months (range, 3 months–3 years). Previous operations had been performed in ten patients (33%). All patients were divided into three groups according to type of external shunt used. In group A (ten patients), the transient external shunt was constructed between superior vena cava and right atrium by uniting two standard venous cannulas with a Y-connector. In group B (ten patients), the external shunt was performed with a single short venous cannula constructed between superior vena cava and right atrium. In group C (ten patients), the external shunt was constructed between superior vena cava and left pulmonary artery by using a single short venous cannula. During operation, central venous pressure (CVP), arterial O2 saturation and mean arterial blood pressure were recorded continuously. Results: All operations are completed without the establishment of cardiopulmonary bypass. Hospital mortality was 3.3%. One patient in group A died because of low cardiac output at the end of postoperative day 2. All patients were extubated within 4 h. In groups A–C mean superior vena caval pressures were measured 28, 24 and 21 mmHg, respectively during superior vena cava-right pulmonary artery anastomosis. In both groups A and B patients, arterial O2 saturation decreased to a minimum 53±2 and 53±2%, respectively during the operation. In the group C, minimum arterial O2 saturation was measured 82±2%. Although mean arterial pressure decreased in all groups during clampage; in group C patients, this drop is not significant. Conclusion: Based on the study presented here, bidirectional cavo-pulmonary anastomosis can be carried out by using different types of transient external shunt. The best hemodynamical condition and arterial O2 levels were achieved with the shunt constructed between superior vena cava and left pulmonary artery. Bidirectional cavopulmonary shunt, Transient external shunts 1 Introduction Bidirectional cavo-pulmonary shunt operation is one of the palliative procedures in patients with functional single ventricular physiology. With increased knowledge about the detrimental effects of cardiopulmonary bypass on patients with functional single ventricle, many surgeons had begun to search for new and safer methods. Since the first suggestion made by Lambarti et al. [1], transient external shunt techniques used for caval drainage had become increasingly popular. However, criteria for performing the bidirectional cavo-pulmonary shunt without cardiopulmonary bypass are not well established. Some surgeons have employed this approach in patients at particular risk for cardiopulmonary bypass [2]. Another controversy is about the type of external shunt used for caval drainage. In this prospective study, we wanted to compare the peri- and post-operative outcome of three different types of external shunt used in 30 patients who had undergone bidirectional cavo-pulmonary shunt operation. This analysis focuses on the technical aspects of the procedure and reviews the short-term morbidity and mortality. 2 Materials and methods Between February 1997 and December 2000, 30 patients have undergone bidirectional cavo-pulmonary anastomosis operation with transient external shunts at Medical Faculty of Istanbul University, Department of Cardiovascular Surgery. The mean patient age was 13 months (range, 3 months–3 years). Seventeen patients were under 1 year of age. Weights ranged from 6 to 28–kg. Primary diagnosis and previous operations were listed in Table 1 . Previous operations had been performed in ten patients (33→7% patients systemic to pulmonary artery shunt, two patients pulmonary arterial banding and one patient pulmonary arterial banding and one patient pulmonary arterial banding+repair of aortic coarctation and arcus hypoplasia). All operations were performed without cardiopulmonary bypass. The patients requiring any additional intracardiac intervention were not included in the study. Indications for performing the bidirectional cavo-pulmonary shunt operation without cardiopulmonary bypass are the presence of an unrestrictive atrial septal defect, the absence of atrioventricular valve regurgitation and any other additional intracardiac defects requiring correction. Bidirectional cavo-pulmonary shunt was used as a first stage of complete Fontan operation for all patients. Table 1 Open in new tabDownload slide Clinical characteristics and previous operations Table 1 Open in new tabDownload slide Clinical characteristics and previous operations In this prospective study, 30 patients were divided into three groups according to the type of external shunt used for caval drainage (Table 2 ): Table 2 Open in new tabDownload slide Type and size of external shunts Table 2 Open in new tabDownload slide Type and size of external shunts Group A (ten patients): external shunt was established between superior vena cava and right atrium by using two standard venous cannulas connected with a Y connector. Group B (ten patients): external shunt was established between superior vena cava and right atrium by using a single short standard venous cannula. Group C (ten patients): external shunt was established between superior vena cava and left pulmonary artery by using a standard venous cannula. Our strategy in systemic to pulmonary artery shunts is the placement of polytetrafluoroethylene (PTFE) graft from the innominate artery or ascending aorta to the main pulmonary artery. In group C, purse string sutures were placed on the main pulmonary artery and venous cannula was moved into the left pulmonary artery. Since group C patients require the presence of intact and undistorted native pulmonary arteries, we made a decision to create this group among the patients without the history of previous pulmonary arterial banding or systemic to pulmonary artery shunt operations. 2.1 Surgical technique Through a median sternotomy and peri-cardiotomy, superior venae cava (SVC) and pulmonary arteries were dissected free. Azygos vein was divided to allow mobilization. Patent ductus arteriosus (PDA) and if present, previous modified Blalock Taussig (BT) shunt were dissected, controlled; but, not ligated to improve the arterial oxygenation during the reconstruction of anastomosis. In cases with right-sided PDA (two patients in group A), ductus had to be ligated before the anastomosis. Fortunately, we did not notice any deterioration in both patients following the ligation of PDA. Systemic heparin (1 mg/kg) was administered intravenously before the procedure. In group A, purse-string sutures were placed in the right atrium and superior vena cava and a transient external shunt was created between SVC and right atrium by using two standard venous cannulas connected with a Y connector (Fig. 1a ). In group B, external shunt was established between distal part of SVC and right atrium with a single short straight venous cannula (Fig. 1b). In group C, purse-string sutures were placed longitudinally in the distal part of SVC and main pulmonary artery (MPA). The transient shunt was created between SVC and left pulmonary artery by pushing the pulmonary arterial end of short venous cannula into the left pulmonary artery (Fig. 1c). The short straight venous cannulas utilized in groups B and C were created by cutting one end of the cannula obliquely to facilitate the placement of the cannula into right atrium and left pulmonary artery. The size of the venous cannulas was given in Table 2. To prevent the possible bias caused by the length of venous cannulas, we used the venous cannulas with the same length (8 cm) in all patients belonging to groups B and C. Venous cannulas were filled with blood to allow deairing and placed parallel to the patient to overcome the negative effects of gravity. Superior vena cava was clamped and divided from cavoatrial junction with taking care to avoid sinus node. The atrial end was oversewn. During clampage, superior vena caval pressures above the clamp and arterial O2 saturations were recorded in all patients. Right pulmonary artery (RPA) was clamped and arteriotomy was performed on the superior aspect. The distal end of the SVC was anastomosed to the RPA with end-to-side technique. Two separate 7-0 polypropylene sutures were used to prevent the purse-string effect. After completion of the anastomosis, external shunt was disconnected and blood in the cannula was drained into the SVC, right atrium and PA. To prevent the future stenosis, the purse-string sutures placed on the SVC and main pulmonary artery were not tied. The defect on the SVC was repaired primarily with a 7-0 polypropylene suture. In seven patients from group C with an adequate sized main pulmonary artery, a decision was made to close the defect primarily. However, for the remaining three patients with a younger age (<9 months) and small sized main pulmonary artery, we used a pericardial patch to repair the defect. During the entire procedure, the patients' heads were elevated. Fig. 1 Open in new tabDownload slide (a) External shunt between SVC and right atrium by using 2 cannulas connected with a Y connector in Group A. (b) External shunt between SVC and right atrium by using a short straight venous cannula. (c) External shunt between SVC and left pulmonary artery by using a standard venous cannula. Fig. 1 Open in new tabDownload slide (a) External shunt between SVC and right atrium by using 2 cannulas connected with a Y connector in Group A. (b) External shunt between SVC and right atrium by using a short straight venous cannula. (c) External shunt between SVC and left pulmonary artery by using a standard venous cannula. Before operations, basal superior vena caval pressure (SVCP0A,B,C), arterial O2 saturation level (S0A,B,C) , mean arterial pressure (P0A,B,C) were recorded in all groups. During the clampage of superior vena cava, superior vena caval pressures were recorded continouosly and first (SVCP1A,B,C) , fifth (SVCP2A,B,C) and tenth (SVCP3A,B,C) minute pressures were compared. During clampage, minimum arterial O2 saturation (S1A,B,C) levels and arterial blood pressures (P1A,B,C) were also recorded in all groups. 2.2 Statistical analysis Statistical analysis was performed with analysis of variance (ANOVA) test. P values of <0.1 were considered significant. In situations with significant P value, post-test was performed with Tukey. Comparisons between two parameters were performed with student t-test. 3 Results Hospital mortality was 3.3%. All patients tolerated the procedure well and their hemodynamical statuses were not deteriorated during the reconstruction of anastomosis. Cardiopulmonary bypass (CPB) was not used in any step of operation. Dopamine or α-agonist agents were not required to increase blood pressure. The duration of operations ranged from 85 to 164 min (mean 122 min) in group A, 83 to 159 min (mean 120 min) in group B and 89 to 168 min (mean 124 min) in group C. Duration of anastomosis between SVC and RPA ranged from 10 to 17 min (mean 13 min) in all patients. The predictors of caval drainage during the reconstruction of the anatomosis are the pressure difference between SVC and right atrium or pulmonary artery, the central gravity, the radius and length of venous cannulas that were used. We used the venous cannulas with the same length (8 cm) in all patients belonging to groups B and C to prevent the possible bias that may be caused by the length of venous cannulas. According to Poseuilles formula, the most important determinant of flow is the size of the cannula because of the strong inverse relationship between flow and fourth power of radius. The size of the cannula depends on the weight and age of the patients. We generally tried to use the same sized (18 F) venous cannula in all patients (eight patients in group A, eight patients in group B and eight patients in group C). Superior vena caval pressures increased in all three groups; but, the most effective caval drainage was performed with the shunt constructed between superior vena cava and left pulmonary artery (P<0.001 for SVCP1A versus SVCP1C and P<0.01 for SVCP1B and SVCP1C). In group A patients, SVCP1A, SVCP2A and VCP3A levels were measured 28±2 mmHg. In group B patients, SVCP1B, SVCP2B and SVCP3B levels were measured 24±2 mmHg. In group C patients, SVCP1C, SVCP2C and SVCP3C levels were measured 21±1, 21±1, 22±1 mmHg, respectively. Superior vena caval pressure changes during clampage were given in Fig. 2 . We also compared the superior vena caval pressures of the 24 patients in whom we used the same sized cannula and this is shown in Fig. 3 . Basal arterial O2 saturations were 64±2% for Groups A–C. There was no significant difference in terms of basal arterial O2 levels among three groups. However, arterial O2 saturations were increased from 64±2 (S0C) to 82±2% (S1C) in group C patients during clampage. This elevation is significant when compared to groups A and B (P<0.001 for group A versus group C and P<0.001 for groups B and C) (Fig. 4 ). In groups A and B patients, basal arterial blood pressures were recorded 66±5 (POA) and 67±3 mmHg (P0B), respectively before the procedure. During clampage, minimum arterial blood pressures were recorded 56±3 mmHg (P1A) for group A patients and 57±2 mmHg (P1B) for group B patients (P<0.0001 for both groups). In group C, basal arterial blood pressure was 68±2 mmHg (P0C). During clampage, minimum arterial blood pressure was measured 66±3 mmHg for group C patients (P1C). This decrease is not significant. No blood transfusions were required during operation and post-operative period. One patient (in group A) diagnosed with double outlet right ventricle, ventricular septal defect, subaortic conus, mitral atresia and previous pulmonary banding operation developed low cardiac output and died at the end of post-operative day 2. Other patients were extubated in the first 4 h. Hemodynamical studies revealed a mean post-operative central venous pressure of 13±2 mmHg. Arterial oxygen saturation increased from 65±5 to 91±4% post-operatively. Average durations of intensive care unit and hospital stays were 36 h and 8 days, respectively. Two patients suffered from prolonged pleural effusion post-operatively. Post-operative echocardiography and angiography demonstrated a functional cavopulmonary anastomosis in all patients. Fig. 2 Open in new tabDownload slide Recorded superior vena caval pressure changes during anastomosis in all three groups (0, 1, 5, 10 minutes). Fig. 2 Open in new tabDownload slide Recorded superior vena caval pressure changes during anastomosis in all three groups (0, 1, 5, 10 minutes). Fig. 3 Open in new tabDownload slide The comparison of patients operated by using the same size of venous cannula (Group A, eight patients; Group B, eight patients; Group C, eight patients. In SVCP0 the difference is not significant. In SVCP1 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP2 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP3 P<0.001 for groups A versus C; P<0.01 for groups B and C). Fig. 3 Open in new tabDownload slide The comparison of patients operated by using the same size of venous cannula (Group A, eight patients; Group B, eight patients; Group C, eight patients. In SVCP0 the difference is not significant. In SVCP1 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP2 P<0.001 for groups A versus C; P<0.01 for groups B and C. In SVCP3 P<0.001 for groups A versus C; P<0.01 for groups B and C). Fig. 4 Open in new tabDownload slide Arterial O2 saturations during anastomosis. Fig. 4 Open in new tabDownload slide Arterial O2 saturations during anastomosis. The ineffective vena caval drainage during the clampage may lead an acute increase in intracranial pressure. Following extubation, a full neurological examination implicating motor and sensory systems was performed by the same physician. With increasing knowledge and experience, we learned that the most important and earlier sign of increased intracranial pressure is papilledema. The degree of venous engorgement was determined with ophtalmologic evaluation. Two patients with significant papilledema were medically treated (dexamethasone 0.2 mg/kg/IV/6 h, furosemide 1.5 mg/kg /IV/6 h and mannitol 0.2 g/kg/IV/6 h) and followed with periodic neurological examination and computed tomographic (CT) scanning until discharge. Full recovery was detected in all patients (range 6–11 days). We had three patients with seizure activity during hospitalization period. Two of them (one from group A and one from group B) were febrile in nature and improved with anti-epileptic therapy. However, the activity of the remaining patient from group A became permanent and this patient is still following by our pediatric neurology department. 4 Conclusion Bidirectional cavo-pulmonary shunt operation is a universally accepted palliative procedure in patients with functional single ventricle [3]. It preserves ventricular and atrioventricular valve function by decreasing volume overload on single ventricle. It does not cause any distortion in pulmonary arteries and possess no risk of precipitating pulmonary vascular obstructive disease. It is generally considered as a second line operation after some other forms of palliative procedures (either systemic to PA shunt or pulmonary banding procedure) performed during neonatal or early infancy period [1]. Bidirectional cavo-pulmonary shunt operation was generally performed by using cardiopulmonary bypass. However, Jahangiri et al. reported seven bidirectional Glenn shunt procedures performed without cardiopulmonary bypass through a thoracotomy. They clamped directly superior vena cava and did not use any transient external shunt [4]. In their series, mean right internal jugular venous pressure was measured 26 mmHg and dopamine or α-agonist agents were required only in three patients to increase blood pressure. They encountered no neurological complications and reported that establishing an intraoperative shunt was unnecessary. With increasing knowledge about the detrimental effects of cardiopulmonary bypass on infants, Lamberti and et al. [1] searched for a new approach and performed their operations by a right thoracotomy. In their experience of seven patients, they could decrease caval pressure about 15 mmHg by the help of an external shunt. Subsequent studies also demonstrated that bidirectional cavo-pulmonary shunt and even extracardiac Fontan operations might be executed without cardiopulmonary bypass [2,5,8]. Cherian et al. performed bidirectional Glenn shunt operation by using different types of transient external shunt [5]. In their series of five patients, they demonstrated that the avoidance of cardiopulmonary bypass and aortic cross-clamping has the advantages including earlier extubation, lesser usage of blood products, decreased necessity and duration of inotropic support. Mohan Reddy et al. demonstrated a post-operative increase in pulmonary vascular resistance and hypoxia after cardiopulmonary bypass and reported that transient external shunt approach must be indicated on a wider basis [6]. Some other authors demonstrated that aortopulmonary shunts might occur after establishing cardiopulmonary bypass and this might cause prolonged pleural effusions [7,9]. In our department, we perform bidirectional Glenn and extracardiac Fontan operations without cardiopulmonary bypass in patients without additional intracardiac intervention. In our series of 30 patients, during post-operative follow-up period, hypoxia and increase in central venous and pulmonary artery pressures were not detected. All patients were weaned off mechanical ventilator during early post-operative hours; except one patient who died because of low cardiac output. We tried to decrease mean superior vena caval pressures by external shunts to prevent neurological damage during clampage. Although uppermost level of safe central venous pressure is not still clear, we demonstrated that lowest venous pressures were detected in groups B and C. We conclude that short venous cannulas decrease the resistance against blood flow in groups B and C. Length of cannulas in groups B and C were half of the ones in group A. Some authors demonstrated reduction in diastolic and mean peak systolic flow velocities more than 50% in middle cerebral artery by transcranial Doppler ultrasonography and detected mild electrocortical alterations in evoked potentials. They reported that such deteriorations did not occur in patients operated with cardiopulmonary bypass [10,11]. In subsequent studies, it was demonstrated that electrocortical alterations improved by decreasing superior vena caval pressure with an external shunt. In our series, we also decreased superior vena caval pressure by using an external shunt. Transcranial Doppler ultrasonography or electroencephalography was not used intra-operatively. Some authors reported that cavoatrial shunt techniques might compromise the quality of cavo-pulmonary anastomosis and restrict the cerebral blood flow in cyanotic children [12]. In group C patients, after establishing a short venous cannula between superior vena cava and left pulmonary artery, we observed an increase in arterial oxygen saturations. We consider that this technique may increase cerebral blood flow. As a conclusion, bidirectional cavo-pulmonary shunt operation may be performed with different external shunt techniques. Among different types, external shunt constructed between superior vena cava and pulmonary artery provides better hemodynamics and reduces surgical stress by increasing the O2 saturation levels during anastomosis. References [1] Lamberti J.J. , Spicer R.L. , Waldman J.D. , Grehl T.M. , Thomson D. , George L. , Kirkpatrick S.E. , Mathewson J.W. . The bidirectional cavopulmonary shunt , J Thorac Cardiovasc Surg , 1990 , vol. 100 (pg. 22 - 29 ) Google Scholar PubMed OpenURL Placeholder Text WorldCat [2] Burke R.P. , Jacobs J.P. , Ashraf M.H. , Aldousany A. , Chang A.C. . Extracardiac Fontan operation without cardiopulmonary bypass , Ann Thorac Surg , 1997 , vol. 63 (pg. 1175 - 1177 ) Google Scholar Crossref Search ADS PubMed WorldCat [3] Xie Bin M.D. , Zhang Jin Fang M.D. , Devi Prasad Shetty M.S. . Bidirectional Glenn shunt: 170 cases , Asian Cardiovasc Thorac Ann , 2001 , vol. 9 (pg. 196 - 199 ) Google Scholar Crossref Search ADS WorldCat [4] Jahangiri M. , Keogh B. , Shinebourne E.A. , Lincoln C. . Should the bidirectional Glenn procedure be performed through a thoracotomy without cardiopulmonary bypass? , J Thorac Cardiovasc Surg , 1999 , vol. 118 (pg. 367 - 368 ) Google Scholar Crossref Search ADS PubMed WorldCat [5] Murthy K.S. , Coelho R. , Naik S.K. , Punnoose A. , Thomas W. , Cherian K.M. . Novel techniques of bidirectional Glenn shunt without cardiopulmonary bypass , Ann Thorac Surg , 1999 , vol. 67 (pg. 1771 - 1774 ) Google Scholar Crossref Search ADS PubMed WorldCat [6] Reddy V.M. , Liddicoat J.R. , Hanley F.L. . Primary bidirectional superior cavopulmonary shunt in infants between 1 and 4 months of age , Ann Thorac Surg , 1995 , vol. 59 (pg. 1120 - 1125 ) Google Scholar Crossref Search ADS PubMed WorldCat [7] van de Wal H.J.C.M. , Ouknine R. , Tamisier D. , Lévy M. , Vouhè P.R. , Leca F. . Bi-directional cavopulmonary shunt: is accessory pulsatile flow, good or bad? , Eur J Cardiothorac Surg , 1999 , vol. 16 (pg. 104 - 110 ) Google Scholar Crossref Search ADS PubMed WorldCat [8] Uemura H. , Yagihara T. , Yamashita K. , Ishizaka T. , Yoshizumi K. , Kawahira Y. . Establishment of total cavopulmonary connection without use of cardiopulmonary bypass , Eur J Cardiothorac Surg , 1998 , vol. 13 (pg. 504 - 507 ) Google Scholar Crossref Search ADS PubMed WorldCat [9] Kim S.J. , Bae E.J. , Cho D.J. , Park I.S. , Kim Y.M. , Kim W.-H. , Kim S.H. . Development of pulmonary arteriovenous fistulas after bidirectional cavopulmonary shunt , Ann Thorac Surg , 2000 , vol. 70 (pg. 1918 - 1922 ) Google Scholar Crossref Search ADS PubMed WorldCat [10] Zimmerman A.A. , Burrows F.A. , Jonas R.A. , Hickey P.R. . The limits of detectable cerebral perfusion by transcranial doppler sonography in neonates undergoing deep hypothermic low-flow cardiopulmonary bypass , J Thorac Cardiovasc Surg , 1997 , vol. 114 (pg. 594 - 600 ) Google Scholar Crossref Search ADS PubMed WorldCat [11] Rodriguez R.A. , Cornel G. , Semelhago L. , Spinter W.M. , Weerasena N.A. . Cerebral effects in superior vena caval cannula obstruction: the role of brain monitoring , Ann Thorac Surg , 1997 , vol. 64 (pg. 1820 - 1822 ) Google Scholar Crossref Search ADS PubMed WorldCat [12] Stamm C. , Friehs I. , Mayer J.E. Jr , Zurakowski D. , Triedman J.K. , Moran A.M. , Walsh E.P. , Lock J.E. , Jonas R.A. , Del Nido P.J. . Long-term results of the lateral tunnel Fontan operation , J Thorac Cardiovasc Surg , 2001 , vol. 121 (pg. 28 - 41 ) Google Scholar Crossref Search ADS PubMed WorldCat © 2003 Elsevier Science B.V. Elsevier Science B.V.

Journal

European Journal of Cardio-Thoracic Surgery – Oxford University Press

Published: Apr 1, 2003

Keywords: Bidirectional cavopulmonary shunt Transient external shunts

There are no references for this article.