The Utility of Aortic Blood Flow Measurements in the Prediction of Pulmonary Artery Banding Outcome Yasuhiro Kotani, MD, PhD, Michael Coles, Nimesh D. Desai, MD, PhD, Osami Honjo, MD, PhD, Christopher A. Caldarone, MD, John G. Coles, MD, and Glen S. Van Arsdell, MD Division of Cardiovascular Surgery, The Labatt Family Heart Centre, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
Background. The purpose of this study was to evaluate the clinical utility of intraoperative aortic blood flow measurements on clinical outcome in patients undergoing pulmonary artery banding (PAB). Methods. We reviewed 12 patients who underwent a PAB between September 2008 and March 2013 who also had intraoperative aortic blood flow measurements. Diagnosis included biventricular physiology in 6, singleventricle physiology in 4, and inadequate systemic ventricle in 2 patients. Aortic blood flow was measured at the time of surgery by Transonic flow probe (Transonic Systems Inc, Ithaca, NY). Aortic flow, intraoperative hemodynamics, and clinical outcomes were analyzed to determine the potential predictive utility of intraoperative variables on postoperative outcomes. Results. The aortic flow increased after PAB from 1.56 ± 0.73 to 2.20 ± 1.10 L $ minL1 $ mL2 (41.0% increase; p [ 0.001). The efficacy of the PAB procedure was found
to be directly related to the percentage increase in aortic blood flow measured intraoperatively. Three patients with less than 20% increase in aortic blood flow died, required re-PAB, or developed ventricular dysfunction, while patients with successful PAB had more than 40% increase in aortic blood flow. The percentage increase in aortic blood flow was not predictable based on preband or post-band absolute aortic blood flow measurements. The percentage increase in aortic blood flow was inversely correlated to the tightness of the PAB as defined relative to that predicted by the Trusler formula (r [ 0.67; p [ 0.01). Conclusions. This study identifies the change in the aortic blood flow as a new, physiologically based parameter to help predict PAB outcome.
P
physiologically appropriate PAB circumference remains elusive in the presence of single-ventricle circulation. Currently, there is no firm consensus among surgeons regarding optimal band tightness based on morphology and body weight [4–6]. Some surgeons adjust the tightness based on the intraoperative pulmonary blood flow ratio to the systemic blood flow (Qp/Qs) [4] and others use the arterial oxygen saturation (SaO2) as a target [6]. Other indicators include increase in systemic blood pressure, the pressure gradient across the band, and diagnosis-based decision (ie, single-ventricle vs biventricular physiology) [3, 5, 6]. Furthermore, variability in the material of the band used for PAB makes it difficult to identify a uniform formula for determination of appropriate tightness of PAB. In this study we examined the utility of measuring the aortic blood flow; ie, systemic blood flow, at the time of PAB as a predictive indicator of postoperative physiologic and clinical outcomes. In general, we have relied on the Trusler formula to determine initial band circumference with adjustment in some cases based on SaO2 and systolic pressure ratios. The circumference of the band was adjusted in millimeters based on the following rule: an infant with VSD–20mm plus the
ulmonary artery banding (PAB) remains as a standard procedure for patients with high pulmonary blood flow awaiting subsequent single-ventricle palliation or definitive biventricular repair [1–3]. The PAB is designed to provide adequate systemic flow by restricting pulmonary blood flow and allows the patients to grow and improve the safety profile of subsequent definitive or palliative procedure. Trusler and Mustard [2] reported a formula to determine the appropriate tightness of PAB based on the patients’ body weight. Determination of appropriate tightness of PAB in the operating room, however, remains a challenge due to the heterogeneity of individual patient physiology such that additional fine adjustment of band circumference is frequently required. Furthermore, the Trusler formula was originally devised for patients with ventricular septal defect (VSD) and transposition physiology, so that determination of
Accepted for publication Jan 27, 2015. Address correspondence to Dr Van Arsdell, Division of Cardiovascular Surgery, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada, M5G 1X8; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.01.066
2
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
infant’s weight in kilograms; an infant with VSD with transposition of the great arteries–24mm plus the infant’s weight in kilograms [2].
Patients and Methods We performed chart reviews on 12 patients who had aortic blood flow measurements recorded during PAB between September 2008 and March 2013 at The Hospital for Sick Children, Toronto. Approval for the study was granted by The Hospital for Sick Children Research Ethics Board, which waived the requirement for individual patient consent. Pre-PAB demographics are shown in Table 1. Median age at surgery was 69 (range, 3 to 605) days and median body weight was 3.2 kg (range, 2.4 to 9.3 kg). Diagnosis includes biventricular physiology in 6, single-ventricle physiology in 4, and biventricular morphology with presumed inadequate systemic ventricle in 2 patients. No patients underwent a concomitant procedure.
Surgical Technique A PAB was performed with median sternotomy in all patients. A 4-mm wide silastic-coated umbilical tape (made in house) was used. The circumferential length of the band was estimated by the Trusler formula [2] and marked with fine silk sutures. After stabilization of ventilation support with fraction of inspired oxygen (FiO2) of less than 0.4, the band was tightened. Systemic blood pressure, SaO2, and heart rate were monitored during the procedure. The main pulmonary artery pressure was also measured just before and after PAB by means of direct needle measurement. The tightness of the band was then adjusted based on the following criteria: (1) increase in systemic blood pressure; (2) decreased pulmonary artery pressure to less than 50% of systemic blood pressure; and (3) adequate SaO2 level (>85% for biventricular physiology and >75% for single-ventricle physiology). The band was then secured with 5-0 or 6-0
Ann Thorac Surg 2015;-:-–-
polypropylene sutures (Ethicon, Inc, Somerville, NJ) to prevent its migration.
Measurement of Ascending Aortic and PAB Flow and Hemodynamics After mobilization of the ascending aorta and PA, a baseline measurement of aortic and PA flow was obtained directly using an 8- or 10-mm Transonic flow probe (model No. A2; Transonic Systems Inc, Ithaca, NY) placed on the vessels and connected to an ultrasonic blood flow meter (model T201; Transonic Systems Inc). A repeat aortic flow measurement was made after PAB was performed. During measurements, FiO2 was maintained less than 0.4. Milrinone or epinephrine was used only when the patient developed significant bradycardia or hypotension after PAB. The percentage increase in aortic flow after PAB was calculated. The blood sample was taken from the superior vena cava, pulmonary artery, and the aorta and the oxymetric saturations (O2%) were measured. The oxymetric saturation of the pulmonary vein was considered as 98%. The Qp/Qs was then determined based on the oxymetric saturations (O2%) using the following formula: Qp Aortic O2% Superior Vena Cava O2% ¼ Qs Pulmonary Venous O2% Pulmonary Arterial O2%
Statistical Analysis Data are presented as means standard deviation. The level of statistical significance was set at p ¼ 0.05. Differences between the groups were analyzed by the Student t test. Pearson correlation was used to determine potential correlations among variables.
Results Intraoperative Data The PAB circumference data are shown in Table 1. There were 5 patients who required a larger band circumference
Table 1. Preoperative, Intraoperative, and Postoperative Data No. 1 2 3 4 5 6 7 8 9 10 11 12
Diagnosis
Body Weight (kg)
Circumference of PAB (mm)
Deviation From the Trusler Formula (mm)
DORV DORV AVSD ccTGA ccTGA, VSD VSD DORV, MA TA DILV, VSD DORV, hypo LV Unbalanced AVSD Unbalanced AVSD
2.8 3.4 2.6 9.3 3.8 3.0 3.1 3.2 6.4 3.0 2.4 3.7
25 24 25 30 27 23 24 25 27 26 23 23
þ2 þ1 þ2 þ1 1 0 3 2 3 1 þ1 1
Re-PAB No No No No No No No Yes No No No No
Latest Status s/p BiV repair s/p BiV repair Awaiting BiV repair Awaiting BiV repair Awaiting BiV repair Awaiting BiV repair Died s/p BCPS Assessed for HTx Died Awaiting repair Awaiting repair
AVSD ¼ atrioventricular septal defect; BCPS ¼ bidirectional cavopulmonary shunt; BiV ¼ biventricular; ccTGA ¼ congenitally corrected transposition of the great arteries; DILV ¼ double-inlet left ventricle; DORV ¼ double-outlet right ventricle; HTx ¼ heart transplantation; LV ¼ left ventricle; MA ¼ mitral atresia; TA ¼ tricuspid atresia; PAB ¼ pulmonary artery banding; s/p ¼ status post; VSD ¼ ventricular septal defect.
Ann Thorac Surg 2015;-:-–-
3
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
than indicated by the Trusler formula, whereas 6 patients required further PAB tightening. Deviation from the Trusler formula ranged from 3 to þ2 mm and was unrelated to diagnosis and body weight. One patient did not need any adjustment from the Trusler formula.
Hemodynamic During Pulmonary Artery Banding The pre-PAB absolute and indexed aortic blood flow was 0.42 0.33 (range, 0.18 to 1.40) L/minute and 1.56 0.73 (range, 0.53 to 3.18) L $ min1 $ m2, respectively (Table 2). The aortic blood flow increased after PAB from 1.56 0.73 to 2.20 1.10 L $ min1 $ m2 (41.0% increase; p ¼ 0.001) in all but 2 patients. The post-PAB absolute and indexed flow was 0.58 0.48 (range, 0.26 to 2.00) L/minute and 2.20 1.10 (range, 0.53 to 4.55) L $ min1 $ m2, respectively. One patient (patient 5), with congenitally corrected transposition of the great arteries (ccTGA) and VSD had a relatively loose band intentionally for planned subsequent double switch operation, and the aortic flow was decreased by 12% after PAB in this patient. The Qp/Qs decreased after PAB from 4.4 2.9 to 1.6 0.6 (p ¼ 0.047) (Table 2). The peak pressure gradient across the band determined by echocardiography was 41.2 15.0 mm Hg. The post-PAB systolic systemic blood pressure, systolic pulmonary artery pressure, and systolic systemic blood pressure ratio to pulmonary artery pressure by direct needle measurement was 69.5 9.7 mm Hg, 37.0 10.2 mm Hg, and 0.53 0.11, respectively.
Relationship Between Aortic Flow and Hemodynamics During Pulmonary Artery Banding There was a strong positive correlation between pre-PAB Qp/Qs and % increase in aortic flow (r ¼ 0.760, p ¼ 0.004) (Fig 1). The pre-PAB aortic flow did not correlate with % increase in aortic flow (r ¼ 0.038, p ¼ 0.906) (Fig 2). The circumference of the band corrected by body weight did not correlate with % increase in aortic flow (r ¼ 0.035,
p ¼ 0.915) (Fig 3). However, when the tightness of PAB is considered as deviation from the Trusler formula (final circumference; circumference calculated by the Trusler formula), there was a positive correlation between deviation from the Trusler formula and % increase in aortic flow (r ¼ 0.667, p ¼ 0.018) (Fig 4).
Clinical Outcomes During a follow-up period of 18 (4 to 58) months, there were 2 deaths. One patient (patient 7) died from pulmonary hypertension and another patient (patient 10), who was diagnosed as 22q11 microdeletion, died from noncardiac causes. One patient (patient 8) required rePAB for persistent pulmonary hypertension 5 months after initial PAB. One patient (patient 9) developed severe ventricular dysfunction and assessment for heart transplantation is in progress. There was a significant variation in post-PAB indexed aortic blood flow from 0.5 to 4.5 L $ min1 $ m2, suggesting a wide range of physiologic responses to the stress associated with PAB. Importantly, 3 patients with less than 20% increase in aortic blood flow died, required re-PAB, or developed ventricular dysfunction. One patient (patient 5), who had a decrease in aortic flow after PAB (12%), had ccTGA and VSD and the PAB was intentionally made loose in preparation for future definitive repair. Three patients who died required re-PAB, and ventricular dysfunction did not increase aortic flow even with tighter PAB than indicated using the Trusler formula, while survivors and patients with subsequent successful repair exhibited a higher percentage aortic flow increase with relatively loose PAB relative to the Trusler formula (Fig 4).
Comment Pulmonary artery banding was originally introduced for patients with large VSD or single ventricle to control the pulmonary blood flow by reducing the pulmonary
Table 2. Hemodynamics During Pulmonary Artery Banding (PAB) Aortic Flow (L/min/m2)
Qp/Qs
No.
Pre
Post
% Increase in Aortic Flow
1 2 3 4 5 6 7 8 9 10 11 12
0.95 0.87 1.18 3.18 1.62 2.24 0.53 1.35 1.69 1.76 2.18 1.13
2.11 1.30 1.79 4.55 1.43 3.38 0.53 1.48 2.02 2.95 3.06 1.83
222 150 151 143 88 151 100 110 119 168 141 162
Pre
Post
11.61 3.40 0.91 3.00 5.50 1.65 5.10 2.36 2.78 3.40 5.60
2.50
1.75 1.86 1.90 0.93
Systolic BP (mm Hg) Post
Systolic PAP (mm Hg) Post
Pp/Ps Post
70 58 64 80 77 84 63 58 71 56 72 81
35 28 31 40 47 45 46 27 25 26 37 57
0.50 0.48 0.48 0.50 0.61 0.54 0.73 0.47 0.35 0.46 0.51 0.70
Pressure Gradient Across PAB (mm Hg)
BP ¼ blood pressure; PAP ¼ pulmonary artery pressure; Pp/Ps ¼ pulmonary artery pressure/systemic blood pressure; blood flow ratio to the systemic blood flow; SaO2 ¼ arterial oxygen saturation.
60 58 32 55 22 17 51 42 42 32 29
SaO2 (%) Post 92 75 84 80 90 100 77 80 90 89 81 86
Qp/Qs ¼ pulmonary
4
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
Fig 1. Pre-pulmonary artery banding (PAB) pulmonary blood flow ratio to the systemic blood flow (Qp/Qs) and change in aortic (Ao) blood flow. Open circle shows patients with good outcome. Closed circle shows patients with worse outcome.
artery size by approximately 70% [1]. This initial experience of 28 operations resulted in 9 operative deaths and 1 late death due to sudden hemodynamic deterioration in 5 patients, indicating the difficulty in achieving optimal PAB tightness [7]. Trusler and Mustard [2] reported a formula to determine the appropriate tightness of PAB based on the patients’ body weight. This idea to adjust the tightness of PAB based on patients’ body weight, stratified by ventricular morphology, remains widely used with modifications as reported [4–6]. These reports suggest that physiologic assessment of desired pulmonary blood flow based on changes in hemodynamics, echocardiography, and Qp/Qs measurement may be useful additional parameters to determine optimal tightness of PAB.
Fig 2. Pre-pulmonary artery banding (PAB) aortic (Ao) blood flow and its change. Pre-PAB aortic blood flow did not correlate to % increase in the aortic blood flow. Patients with death, requirement of PAB, and ventricular dysfunction had less than 20% increase in the aortic blood flow, while more than 40% increase in the aortic flow was seen in successful PAB. Open circle shows patients with good outcome. Closed circle shows patients with worse outcome. (ccTGA ¼ congenitally corrected transposition of the great arteries; VSD ¼ ventricular septal defect.)
Ann Thorac Surg 2015;-:-–-
Fig 3. Pre-pulmonary artery banding (PAB). Band tightness and change in aortic blood flow. Circumference of the band (anatomic determinant) did not correlate to increase in the aortic blood flow (physiologic factor). Open circle shows patients with good outcome. Closed circle shows patients with worse outcome. (ccTGA ¼ congenitally corrected transposition of the great arteries; VSD ¼ ventricular septal defect.)
The use of aortic blood flow, however, has not been investigated as a potential parameter and predictor of subsequent circulatory physiology and clinical outcome. This is the first study to evaluate the utility of direct intraoperative measurement of the aortic blood flow in predicting clinical outcome after PAB. Low percentage increase (
Background. The purpose of this study was to evaluate the clinical utility of intraoperative aortic blood flow measurements on clinical outcome in patients undergoing pulmonary artery banding (PAB). Methods. We reviewed 12 patients who underwent a PAB between September 2008 and March 2013 who also had intraoperative aortic blood flow measurements. Diagnosis included biventricular physiology in 6, singleventricle physiology in 4, and inadequate systemic ventricle in 2 patients. Aortic blood flow was measured at the time of surgery by Transonic flow probe (Transonic Systems Inc, Ithaca, NY). Aortic flow, intraoperative hemodynamics, and clinical outcomes were analyzed to determine the potential predictive utility of intraoperative variables on postoperative outcomes. Results. The aortic flow increased after PAB from 1.56 ± 0.73 to 2.20 ± 1.10 L $ minL1 $ mL2 (41.0% increase; p [ 0.001). The efficacy of the PAB procedure was found
to be directly related to the percentage increase in aortic blood flow measured intraoperatively. Three patients with less than 20% increase in aortic blood flow died, required re-PAB, or developed ventricular dysfunction, while patients with successful PAB had more than 40% increase in aortic blood flow. The percentage increase in aortic blood flow was not predictable based on preband or post-band absolute aortic blood flow measurements. The percentage increase in aortic blood flow was inversely correlated to the tightness of the PAB as defined relative to that predicted by the Trusler formula (r [ 0.67; p [ 0.01). Conclusions. This study identifies the change in the aortic blood flow as a new, physiologically based parameter to help predict PAB outcome.
P
physiologically appropriate PAB circumference remains elusive in the presence of single-ventricle circulation. Currently, there is no firm consensus among surgeons regarding optimal band tightness based on morphology and body weight [4–6]. Some surgeons adjust the tightness based on the intraoperative pulmonary blood flow ratio to the systemic blood flow (Qp/Qs) [4] and others use the arterial oxygen saturation (SaO2) as a target [6]. Other indicators include increase in systemic blood pressure, the pressure gradient across the band, and diagnosis-based decision (ie, single-ventricle vs biventricular physiology) [3, 5, 6]. Furthermore, variability in the material of the band used for PAB makes it difficult to identify a uniform formula for determination of appropriate tightness of PAB. In this study we examined the utility of measuring the aortic blood flow; ie, systemic blood flow, at the time of PAB as a predictive indicator of postoperative physiologic and clinical outcomes. In general, we have relied on the Trusler formula to determine initial band circumference with adjustment in some cases based on SaO2 and systolic pressure ratios. The circumference of the band was adjusted in millimeters based on the following rule: an infant with VSD–20mm plus the
ulmonary artery banding (PAB) remains as a standard procedure for patients with high pulmonary blood flow awaiting subsequent single-ventricle palliation or definitive biventricular repair [1–3]. The PAB is designed to provide adequate systemic flow by restricting pulmonary blood flow and allows the patients to grow and improve the safety profile of subsequent definitive or palliative procedure. Trusler and Mustard [2] reported a formula to determine the appropriate tightness of PAB based on the patients’ body weight. Determination of appropriate tightness of PAB in the operating room, however, remains a challenge due to the heterogeneity of individual patient physiology such that additional fine adjustment of band circumference is frequently required. Furthermore, the Trusler formula was originally devised for patients with ventricular septal defect (VSD) and transposition physiology, so that determination of
Accepted for publication Jan 27, 2015. Address correspondence to Dr Van Arsdell, Division of Cardiovascular Surgery, The Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada, M5G 1X8; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.01.066
2
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
infant’s weight in kilograms; an infant with VSD with transposition of the great arteries–24mm plus the infant’s weight in kilograms [2].
Patients and Methods We performed chart reviews on 12 patients who had aortic blood flow measurements recorded during PAB between September 2008 and March 2013 at The Hospital for Sick Children, Toronto. Approval for the study was granted by The Hospital for Sick Children Research Ethics Board, which waived the requirement for individual patient consent. Pre-PAB demographics are shown in Table 1. Median age at surgery was 69 (range, 3 to 605) days and median body weight was 3.2 kg (range, 2.4 to 9.3 kg). Diagnosis includes biventricular physiology in 6, single-ventricle physiology in 4, and biventricular morphology with presumed inadequate systemic ventricle in 2 patients. No patients underwent a concomitant procedure.
Surgical Technique A PAB was performed with median sternotomy in all patients. A 4-mm wide silastic-coated umbilical tape (made in house) was used. The circumferential length of the band was estimated by the Trusler formula [2] and marked with fine silk sutures. After stabilization of ventilation support with fraction of inspired oxygen (FiO2) of less than 0.4, the band was tightened. Systemic blood pressure, SaO2, and heart rate were monitored during the procedure. The main pulmonary artery pressure was also measured just before and after PAB by means of direct needle measurement. The tightness of the band was then adjusted based on the following criteria: (1) increase in systemic blood pressure; (2) decreased pulmonary artery pressure to less than 50% of systemic blood pressure; and (3) adequate SaO2 level (>85% for biventricular physiology and >75% for single-ventricle physiology). The band was then secured with 5-0 or 6-0
Ann Thorac Surg 2015;-:-–-
polypropylene sutures (Ethicon, Inc, Somerville, NJ) to prevent its migration.
Measurement of Ascending Aortic and PAB Flow and Hemodynamics After mobilization of the ascending aorta and PA, a baseline measurement of aortic and PA flow was obtained directly using an 8- or 10-mm Transonic flow probe (model No. A2; Transonic Systems Inc, Ithaca, NY) placed on the vessels and connected to an ultrasonic blood flow meter (model T201; Transonic Systems Inc). A repeat aortic flow measurement was made after PAB was performed. During measurements, FiO2 was maintained less than 0.4. Milrinone or epinephrine was used only when the patient developed significant bradycardia or hypotension after PAB. The percentage increase in aortic flow after PAB was calculated. The blood sample was taken from the superior vena cava, pulmonary artery, and the aorta and the oxymetric saturations (O2%) were measured. The oxymetric saturation of the pulmonary vein was considered as 98%. The Qp/Qs was then determined based on the oxymetric saturations (O2%) using the following formula: Qp Aortic O2% Superior Vena Cava O2% ¼ Qs Pulmonary Venous O2% Pulmonary Arterial O2%
Statistical Analysis Data are presented as means standard deviation. The level of statistical significance was set at p ¼ 0.05. Differences between the groups were analyzed by the Student t test. Pearson correlation was used to determine potential correlations among variables.
Results Intraoperative Data The PAB circumference data are shown in Table 1. There were 5 patients who required a larger band circumference
Table 1. Preoperative, Intraoperative, and Postoperative Data No. 1 2 3 4 5 6 7 8 9 10 11 12
Diagnosis
Body Weight (kg)
Circumference of PAB (mm)
Deviation From the Trusler Formula (mm)
DORV DORV AVSD ccTGA ccTGA, VSD VSD DORV, MA TA DILV, VSD DORV, hypo LV Unbalanced AVSD Unbalanced AVSD
2.8 3.4 2.6 9.3 3.8 3.0 3.1 3.2 6.4 3.0 2.4 3.7
25 24 25 30 27 23 24 25 27 26 23 23
þ2 þ1 þ2 þ1 1 0 3 2 3 1 þ1 1
Re-PAB No No No No No No No Yes No No No No
Latest Status s/p BiV repair s/p BiV repair Awaiting BiV repair Awaiting BiV repair Awaiting BiV repair Awaiting BiV repair Died s/p BCPS Assessed for HTx Died Awaiting repair Awaiting repair
AVSD ¼ atrioventricular septal defect; BCPS ¼ bidirectional cavopulmonary shunt; BiV ¼ biventricular; ccTGA ¼ congenitally corrected transposition of the great arteries; DILV ¼ double-inlet left ventricle; DORV ¼ double-outlet right ventricle; HTx ¼ heart transplantation; LV ¼ left ventricle; MA ¼ mitral atresia; TA ¼ tricuspid atresia; PAB ¼ pulmonary artery banding; s/p ¼ status post; VSD ¼ ventricular septal defect.
Ann Thorac Surg 2015;-:-–-
3
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
than indicated by the Trusler formula, whereas 6 patients required further PAB tightening. Deviation from the Trusler formula ranged from 3 to þ2 mm and was unrelated to diagnosis and body weight. One patient did not need any adjustment from the Trusler formula.
Hemodynamic During Pulmonary Artery Banding The pre-PAB absolute and indexed aortic blood flow was 0.42 0.33 (range, 0.18 to 1.40) L/minute and 1.56 0.73 (range, 0.53 to 3.18) L $ min1 $ m2, respectively (Table 2). The aortic blood flow increased after PAB from 1.56 0.73 to 2.20 1.10 L $ min1 $ m2 (41.0% increase; p ¼ 0.001) in all but 2 patients. The post-PAB absolute and indexed flow was 0.58 0.48 (range, 0.26 to 2.00) L/minute and 2.20 1.10 (range, 0.53 to 4.55) L $ min1 $ m2, respectively. One patient (patient 5), with congenitally corrected transposition of the great arteries (ccTGA) and VSD had a relatively loose band intentionally for planned subsequent double switch operation, and the aortic flow was decreased by 12% after PAB in this patient. The Qp/Qs decreased after PAB from 4.4 2.9 to 1.6 0.6 (p ¼ 0.047) (Table 2). The peak pressure gradient across the band determined by echocardiography was 41.2 15.0 mm Hg. The post-PAB systolic systemic blood pressure, systolic pulmonary artery pressure, and systolic systemic blood pressure ratio to pulmonary artery pressure by direct needle measurement was 69.5 9.7 mm Hg, 37.0 10.2 mm Hg, and 0.53 0.11, respectively.
Relationship Between Aortic Flow and Hemodynamics During Pulmonary Artery Banding There was a strong positive correlation between pre-PAB Qp/Qs and % increase in aortic flow (r ¼ 0.760, p ¼ 0.004) (Fig 1). The pre-PAB aortic flow did not correlate with % increase in aortic flow (r ¼ 0.038, p ¼ 0.906) (Fig 2). The circumference of the band corrected by body weight did not correlate with % increase in aortic flow (r ¼ 0.035,
p ¼ 0.915) (Fig 3). However, when the tightness of PAB is considered as deviation from the Trusler formula (final circumference; circumference calculated by the Trusler formula), there was a positive correlation between deviation from the Trusler formula and % increase in aortic flow (r ¼ 0.667, p ¼ 0.018) (Fig 4).
Clinical Outcomes During a follow-up period of 18 (4 to 58) months, there were 2 deaths. One patient (patient 7) died from pulmonary hypertension and another patient (patient 10), who was diagnosed as 22q11 microdeletion, died from noncardiac causes. One patient (patient 8) required rePAB for persistent pulmonary hypertension 5 months after initial PAB. One patient (patient 9) developed severe ventricular dysfunction and assessment for heart transplantation is in progress. There was a significant variation in post-PAB indexed aortic blood flow from 0.5 to 4.5 L $ min1 $ m2, suggesting a wide range of physiologic responses to the stress associated with PAB. Importantly, 3 patients with less than 20% increase in aortic blood flow died, required re-PAB, or developed ventricular dysfunction. One patient (patient 5), who had a decrease in aortic flow after PAB (12%), had ccTGA and VSD and the PAB was intentionally made loose in preparation for future definitive repair. Three patients who died required re-PAB, and ventricular dysfunction did not increase aortic flow even with tighter PAB than indicated using the Trusler formula, while survivors and patients with subsequent successful repair exhibited a higher percentage aortic flow increase with relatively loose PAB relative to the Trusler formula (Fig 4).
Comment Pulmonary artery banding was originally introduced for patients with large VSD or single ventricle to control the pulmonary blood flow by reducing the pulmonary
Table 2. Hemodynamics During Pulmonary Artery Banding (PAB) Aortic Flow (L/min/m2)
Qp/Qs
No.
Pre
Post
% Increase in Aortic Flow
1 2 3 4 5 6 7 8 9 10 11 12
0.95 0.87 1.18 3.18 1.62 2.24 0.53 1.35 1.69 1.76 2.18 1.13
2.11 1.30 1.79 4.55 1.43 3.38 0.53 1.48 2.02 2.95 3.06 1.83
222 150 151 143 88 151 100 110 119 168 141 162
Pre
Post
11.61 3.40 0.91 3.00 5.50 1.65 5.10 2.36 2.78 3.40 5.60
2.50
1.75 1.86 1.90 0.93
Systolic BP (mm Hg) Post
Systolic PAP (mm Hg) Post
Pp/Ps Post
70 58 64 80 77 84 63 58 71 56 72 81
35 28 31 40 47 45 46 27 25 26 37 57
0.50 0.48 0.48 0.50 0.61 0.54 0.73 0.47 0.35 0.46 0.51 0.70
Pressure Gradient Across PAB (mm Hg)
BP ¼ blood pressure; PAP ¼ pulmonary artery pressure; Pp/Ps ¼ pulmonary artery pressure/systemic blood pressure; blood flow ratio to the systemic blood flow; SaO2 ¼ arterial oxygen saturation.
60 58 32 55 22 17 51 42 42 32 29
SaO2 (%) Post 92 75 84 80 90 100 77 80 90 89 81 86
Qp/Qs ¼ pulmonary
4
KOTANI ET AL AORTIC BLOOD FLOW IN PULMONARY ARTERY BANDING
Fig 1. Pre-pulmonary artery banding (PAB) pulmonary blood flow ratio to the systemic blood flow (Qp/Qs) and change in aortic (Ao) blood flow. Open circle shows patients with good outcome. Closed circle shows patients with worse outcome.
artery size by approximately 70% [1]. This initial experience of 28 operations resulted in 9 operative deaths and 1 late death due to sudden hemodynamic deterioration in 5 patients, indicating the difficulty in achieving optimal PAB tightness [7]. Trusler and Mustard [2] reported a formula to determine the appropriate tightness of PAB based on the patients’ body weight. This idea to adjust the tightness of PAB based on patients’ body weight, stratified by ventricular morphology, remains widely used with modifications as reported [4–6]. These reports suggest that physiologic assessment of desired pulmonary blood flow based on changes in hemodynamics, echocardiography, and Qp/Qs measurement may be useful additional parameters to determine optimal tightness of PAB.
Fig 2. Pre-pulmonary artery banding (PAB) aortic (Ao) blood flow and its change. Pre-PAB aortic blood flow did not correlate to % increase in the aortic blood flow. Patients with death, requirement of PAB, and ventricular dysfunction had less than 20% increase in the aortic blood flow, while more than 40% increase in the aortic flow was seen in successful PAB. Open circle shows patients with good outcome. Closed circle shows patients with worse outcome. (ccTGA ¼ congenitally corrected transposition of the great arteries; VSD ¼ ventricular septal defect.)
Ann Thorac Surg 2015;-:-–-
Fig 3. Pre-pulmonary artery banding (PAB). Band tightness and change in aortic blood flow. Circumference of the band (anatomic determinant) did not correlate to increase in the aortic blood flow (physiologic factor). Open circle shows patients with good outcome. Closed circle shows patients with worse outcome. (ccTGA ¼ congenitally corrected transposition of the great arteries; VSD ¼ ventricular septal defect.)
The use of aortic blood flow, however, has not been investigated as a potential parameter and predictor of subsequent circulatory physiology and clinical outcome. This is the first study to evaluate the utility of direct intraoperative measurement of the aortic blood flow in predicting clinical outcome after PAB. Low percentage increase (
