- Email: [email protected]
Pulse oximetry during pulmonary artery banding
Pulse Oximetry During Pulmonary Artery Banding Pierre A~ Casthely, MD, Vladimir Redko, MD, John Dluzneski, MD, Kenneth Goodman, MD, Thil Yoganathan, MD, and Joseph I. Simpson, MD In children w i t h a ventricular septal defect and congestive heart failure, banding of the pulmonary artery (PA} causes equalization of right and left ventricular pressures, reduces the volume of the left-to-right shunt, and diminishes the w o r k of the left ventricle and the engorgement of the pulmonary vessels. However, banding the PA too tightly usually produces hypoxemia by reversing the left-to-right shunt and causes severe hemodynamic changes. A series of 14 infants is reported who underwent PA banding during which a pulse oximeter was used as an early indicator of excessively tight PA banding. Significant hemodynamic changes occurred in eight infants in whom the PA banding was too tight. This . I cons=sted of hypotension and bradycardia three to four minutes after the banding. The eight patients also showed significant desaturation of the blood
after application of the band, w i t h the arterial hemoglobin saturation (SaO2) dropping from a preband value of 98 • 6% to a postband value of 80 • 2%. The decrease in SaOz preceded the hypotenslon and bradycardia by two to three minutes in all cases. W h e n the band was removed, the hemodynamic and SaO~ changes returned toward baseline. Subsequently, a less tight band was applied; this was associated with a smaller decrease in SaO 2, an elevation of blood pressure, and no bradycardia. This was considered to be acceptable banding. The right ventricle/PA pressure gradient significantly decreased after acceptable banding, and a gradient higher than 45 mmHg was usually associated w i t h hypoxemia. 9 1 9 8 7 b y G r une & S t r a t t o n , Inc.
METHODS
N EARLY CHILDHOOD a ventricular Sep-
tai defect (VSD) is usually associated with Ipulmonary hypertension, severe circulatory in-
sufficiency, poor physical development, and recurrent pneumonia; this results in a poor prognosis. The high mortality during the first year of life combined with lack of effective pharmacologic therapy makes an early surgical treatment mandatory. In many institutions, total correction is performed during the first year of life, while in others a palliative surgical procedure is used to delay definitive repair to an older age. Pulmonary artery (PA) banding is the most widely used palliative operation, serving as the first stage in the two-stage surgical treatment of this defect. TM Constriction of the PA causes equalization of right and left ventricular pressures and reduces the volume of the left-to-right shunt, thus diminishing the work of the left ventricle (LV) and engorgement of the pulmonary vessels. Banding the PA too tightly usually produces deoxygenation of blood by converting the left-to-right shunt to a right-to-left shunt, and produces severe hemodynamic changes. Pulse oximetry is a noninvasive monitoring technique that has been increasingly used to detect hypoxemia during surgery in infants and newborns. 5"9 This clinical report demonstrates the use of the pulse oximeter to rapidly detect excessive banding of the PA.
This report includes 14 infants with a VSD whose ages yaried from 1 to 4 months, and who weighed from 2.5 to 3.5 kg. All had failure to thrive, signs and symptoms of congestive heart failure, and were digitalized. They were divided into two groups: Group A had the PA band applied only once, while group B ha d it reapplied because it was too tight. All of the patients were premedicated intramuscularly (IM) with pentobarbital, 2 mg/kg, and atropine, 0.15 rag, one hour before induction. An inhalation induction with oxygen and N20 plus halothane (1 to 2 MAC) preceded nasotracheal intubation. Pancuronium, 0.1 mg/kg, was administered prior to intubation. Anesthesia was maintained with 50% Oz in N20 and fentanyl, I to 5 t~g/kg, as needed. Radial arterial and femoral venous pressure catheters were inserted via cutdowns. An electrocardiogram (ECG), rectal temperature probe, and esophageal stethoscope were also used. Continuous monitoring of arterial oxygen saturation (SaO,) was performed using a Nellcor pulse oximeter. A digit oxisensor (D-20) was positioned on a great toe. Oxygen saturation was monitored throughout each operation along with heart rate (HR), blood pressure (BP), ECG, temperature, and arterial blood gases. The right ventricle/pulmonary artery (RV/PA) pressure gradient was measured by the surgeon.
Front the Departments of Cardiac Anesthesia and Anesthesiology, St Joseph's Hospital and Medical Center, Paterson, N J, and State University of New York Health Science Center, Brooklyn, NY. Address reprint requests to Pierre A. Casthely, MD. Department of Cardiac Anesthesia. St Joseph's Hospital and Medical Center, Paterson. NJ 07503. 9 1987 by Grune & Stratton, Inc. 0888-6296/87/0104-0005503.00/0
Journal of Cardiothoracic Anesthesia, Vol 1, No 4 (August), 1987" pp 297-299
297
298
CASTHELY ET AL
The surgical procedure was performed in the right lateral decubitus position through a left thoracotomy. Surgery consisted of encircling the PA with an umbilical tape to produce a gradual narrowingof the PA, using a Hegar dilator as a guide. The objective was to reduce the intraarterial pressure distal to the band to a level approximately one half of that proximalto the band. To achievethis, the diameter of the PA was narrowed 30% to 50% by tightening the band. When the banding was adequate a second ligature was applied for additional security. The band was thought to be too tight when the SaO2 dropped below 90% or the patient deteriorated hemodynamieally.It was then released until the patient recovered,and reapplied in a looserfashion. Statistical analysis within each group was performed using a Student's t-test for paired data, and between the two groups a multivariant analysis was used. A P value of less than .01 was considered significant.
RESULTS
There were statistically significant changes in the BP, HR, SaO2, pulmonary artery pressure, and R V / P A pressure gradient in the patients. During adequate PA banding (Group A; N -- 8), mean blood pressure increased from 60 _+0.7 to 75 .4_ 0.6 m m H g (P < .005) (Table 1). Most of the increase in mean blood pressure was due to an increase in systolic blood pressure. Oxygen saturation and H R did not change after the banding. Pulmonary artery pressure decreased from 42 • 6 to 28 _+ 0.5 m m H g (P < .01) after banding. The R V / P A gradient increased significantly from 16 _+ 0.5 before to 25 _+ 0.3 after banding (P < .01). When the band was too tight (Group B; N = 6), SaO2 and H R significantly decreased from 98 + 6% to 80 • 2% (P < .005) and from 112 _+ 11 beats/min to 50 _+ 2 beats/min (P <
.005), respectively. Mean blood pressure decreased from 58 _+ 0.5 to 30 _+ 0.5 m m H g after the banding (P < .01), and increased to 73 _+ 3 m m H g (P < .01) after the band was reapplied. Pulmonary artery pressure increased from 41 _+ 9 to 60 _+ 5 m m H g (P < .01) after the banding, and decreased to 25 _+ 0.2 m m H g after the band was reapplied. The R V / P A gradient increased to 58 _+ 0.6 m m H g with the initial banding. Changes in the SaO2 preceded hemodynamic changes by two to three minutes in all group B patients. Blood pressure, HR, and SaO~ returned to near baseline values when the band was cut. A less tight band was reapplied, which was associated with a smaller drop in SaO2 (92% to 96%) and an elevation of BP without bradycardia. This was considered to be acceptable banding. The R V / P A gradient significantly decreased after rebanding, and a gradient higher ~.:.than 45 m m H g was usually associated with an "increase in oxygen desaturation. .
. .
After Banding
Mean blood pressure(mmHg) Heartrate (beats/min) Ozsaturation (%) Pulmonary artery pressure (mrnHg)
A 60 • 0.7 B 58 • 0.5 A 110-+ 10 B 112-+ 11 A 98 • 6 B 98• A 42 • 6 B 41 • 9
75 _+ 0.6* 30 • 0.5* 1 1 5 • 10 50• 95-+ 7 80-+2" 28 • 0.5* 60 • 5*
RV/PA gradient (mmHg)
A B
16 • 0.5 20 • 0.5
25 • 0.3* 58 _+ 0.6*
After Rebanding
73 • 3* 125 • 13" 95•
..
9
Table 1. HemodynamicChanges During Pulmonary A r t e r y Banding Prebanding
.
7*
25 -+ 0.2* 27 -+ 0.2*
Abbreviations: A, adequate banding group (N - B); B, rebanding - , ~ group (N - 6). *P < .01 comparedwith previousstage.
DISCUSSION
Continuous monitoring of arterial oxygen saturation can be a valuable tool in the anesthetic management of patients. Arterial blood gas analyses provide accurate but intermittent information. They are not immediately or continuously available, which limits their usefulness in the moment-to-moment anesthetic management of a critically ill patient. Another technique for measuring oxygenation is transcutaneous bxygen analysis. This method provides continuous information on the adequacy of oxygenation, but its use during prolonged surgical procedures is fraught with technical difficulties. The pulse oximeter has gained popularity as an anesthetic monitoring device because it is noninvasive, hazard-free, and easy to use. It provides a continuous assessment of SaO2, using spectrophotoelectric techniques, s~ It is based on the measurement of the optical density of hemoglobin by discrete wavelengths of light. Any pulsating vascular bed, such as an earlobe or fingertip, can be placed between a two-wavelength light source and a detecting sensor. Based on this group of patients, intraoperative criteria for successful PA banding include the following: ( I ) oxygen saturation should not decrease below 90%, (2) the R V / P A pressure gradient should not be greater than 45 m m H g ,
PULSE OXIMETRY AND PULMONARY ARTERY BANDING
(3) a slight elevation of BP by 15 to 25 m m H g p o s t b a n d i n g should occur, and (4) b r a d y c a r d i a should not occur. In conclusion, pulse o x i m e t r y can be used as an i m p o r t a n t e a r l y i n d i c a t o r o f the a d e q u a c y o f the p u l m o n a r y circulation a f t e r P A b a n d i n g . It was found to be an excellent m e t h o d for e a r l y detection of i n t r a o p e r a t i v e h y p o x e m i a following P A b a n d i n g . T h e h y p o x e m i a is d u e to a P A b a n d
299
which is too tight, causing a reversal of a leftto-right shunt to a right-to-left shunt. This is followed in several m i n u t e s by b r a d y c a r d i a , hypotension, and elevation o f the R V / P A pressure gradient. ACKNOWLEDGEMENT
We thank Ellen L. Jackson for manuscript preparation.
REFERENCES
I. Stewart S, Harris P: Pulmonary artery banding. J Thorac Cardiovasc Surg 80:431-436, 1980 2. Doty DB: Pulmonary artery banding, in Glenn WWL (ed): Thoracic and Cardiovascular Surgery (ed 4). Norwalk, CT, Appleton-Century-Crofts, 1983, pp 681-687 3. Oldham HN, Kakos GS, Jarmakani MM: Pulmonary arte/y band ng in infants with complex congenital heart defects. Ann Thorac Surg 13:342-350, 1972 4. Burakobsky VI, Businola LA: Experience with surgical treatment of congenital heart disease with increased pulmonary blood flow in small children. USA-USSR Joint Symposiu m on Congenital Heart Disease, Washington, DC, ~ 1973
5. Friesen RH: Pulse oximetry during pulmonary artery surgery. Anesth Analg 64:376, 1985 6. Monaco F, Feaster WW, McQuitly JC: Continuous noninvasive oxygen saturation monitoring in sick newborns. Respir Care 28:i362, 1983 7. Whitcher C, New W Jr: Perianesthetic oxygen saturation vs the skill of the anesthetist. Anesthesiology 57:A172, 1962 (abstr) 8. Taishley RS, Kline CE: Effect on oxygen saturation by different anesthetic induction agents in cyanotic congenital heart disease. Anesthesiology 63:A3, 1985 (abstr) 9. Yelderman MV, New W Jr: Evaluation of pulse oximetry. Anesthesiology 59:349-352, 1983
after application of the band, w i t h the arterial hemoglobin saturation (SaO2) dropping from a preband value of 98 • 6% to a postband value of 80 • 2%. The decrease in SaOz preceded the hypotenslon and bradycardia by two to three minutes in all cases. W h e n the band was removed, the hemodynamic and SaO~ changes returned toward baseline. Subsequently, a less tight band was applied; this was associated with a smaller decrease in SaO 2, an elevation of blood pressure, and no bradycardia. This was considered to be acceptable banding. The right ventricle/PA pressure gradient significantly decreased after acceptable banding, and a gradient higher than 45 mmHg was usually associated w i t h hypoxemia. 9 1 9 8 7 b y G r une & S t r a t t o n , Inc.
METHODS
N EARLY CHILDHOOD a ventricular Sep-
tai defect (VSD) is usually associated with Ipulmonary hypertension, severe circulatory in-
sufficiency, poor physical development, and recurrent pneumonia; this results in a poor prognosis. The high mortality during the first year of life combined with lack of effective pharmacologic therapy makes an early surgical treatment mandatory. In many institutions, total correction is performed during the first year of life, while in others a palliative surgical procedure is used to delay definitive repair to an older age. Pulmonary artery (PA) banding is the most widely used palliative operation, serving as the first stage in the two-stage surgical treatment of this defect. TM Constriction of the PA causes equalization of right and left ventricular pressures and reduces the volume of the left-to-right shunt, thus diminishing the work of the left ventricle (LV) and engorgement of the pulmonary vessels. Banding the PA too tightly usually produces deoxygenation of blood by converting the left-to-right shunt to a right-to-left shunt, and produces severe hemodynamic changes. Pulse oximetry is a noninvasive monitoring technique that has been increasingly used to detect hypoxemia during surgery in infants and newborns. 5"9 This clinical report demonstrates the use of the pulse oximeter to rapidly detect excessive banding of the PA.
This report includes 14 infants with a VSD whose ages yaried from 1 to 4 months, and who weighed from 2.5 to 3.5 kg. All had failure to thrive, signs and symptoms of congestive heart failure, and were digitalized. They were divided into two groups: Group A had the PA band applied only once, while group B ha d it reapplied because it was too tight. All of the patients were premedicated intramuscularly (IM) with pentobarbital, 2 mg/kg, and atropine, 0.15 rag, one hour before induction. An inhalation induction with oxygen and N20 plus halothane (1 to 2 MAC) preceded nasotracheal intubation. Pancuronium, 0.1 mg/kg, was administered prior to intubation. Anesthesia was maintained with 50% Oz in N20 and fentanyl, I to 5 t~g/kg, as needed. Radial arterial and femoral venous pressure catheters were inserted via cutdowns. An electrocardiogram (ECG), rectal temperature probe, and esophageal stethoscope were also used. Continuous monitoring of arterial oxygen saturation (SaO,) was performed using a Nellcor pulse oximeter. A digit oxisensor (D-20) was positioned on a great toe. Oxygen saturation was monitored throughout each operation along with heart rate (HR), blood pressure (BP), ECG, temperature, and arterial blood gases. The right ventricle/pulmonary artery (RV/PA) pressure gradient was measured by the surgeon.
Front the Departments of Cardiac Anesthesia and Anesthesiology, St Joseph's Hospital and Medical Center, Paterson, N J, and State University of New York Health Science Center, Brooklyn, NY. Address reprint requests to Pierre A. Casthely, MD. Department of Cardiac Anesthesia. St Joseph's Hospital and Medical Center, Paterson. NJ 07503. 9 1987 by Grune & Stratton, Inc. 0888-6296/87/0104-0005503.00/0
Journal of Cardiothoracic Anesthesia, Vol 1, No 4 (August), 1987" pp 297-299
297
298
CASTHELY ET AL
The surgical procedure was performed in the right lateral decubitus position through a left thoracotomy. Surgery consisted of encircling the PA with an umbilical tape to produce a gradual narrowingof the PA, using a Hegar dilator as a guide. The objective was to reduce the intraarterial pressure distal to the band to a level approximately one half of that proximalto the band. To achievethis, the diameter of the PA was narrowed 30% to 50% by tightening the band. When the banding was adequate a second ligature was applied for additional security. The band was thought to be too tight when the SaO2 dropped below 90% or the patient deteriorated hemodynamieally.It was then released until the patient recovered,and reapplied in a looserfashion. Statistical analysis within each group was performed using a Student's t-test for paired data, and between the two groups a multivariant analysis was used. A P value of less than .01 was considered significant.
RESULTS
There were statistically significant changes in the BP, HR, SaO2, pulmonary artery pressure, and R V / P A pressure gradient in the patients. During adequate PA banding (Group A; N -- 8), mean blood pressure increased from 60 _+0.7 to 75 .4_ 0.6 m m H g (P < .005) (Table 1). Most of the increase in mean blood pressure was due to an increase in systolic blood pressure. Oxygen saturation and H R did not change after the banding. Pulmonary artery pressure decreased from 42 • 6 to 28 _+ 0.5 m m H g (P < .01) after banding. The R V / P A gradient increased significantly from 16 _+ 0.5 before to 25 _+ 0.3 after banding (P < .01). When the band was too tight (Group B; N = 6), SaO2 and H R significantly decreased from 98 + 6% to 80 • 2% (P < .005) and from 112 _+ 11 beats/min to 50 _+ 2 beats/min (P <
.005), respectively. Mean blood pressure decreased from 58 _+ 0.5 to 30 _+ 0.5 m m H g after the banding (P < .01), and increased to 73 _+ 3 m m H g (P < .01) after the band was reapplied. Pulmonary artery pressure increased from 41 _+ 9 to 60 _+ 5 m m H g (P < .01) after the banding, and decreased to 25 _+ 0.2 m m H g after the band was reapplied. The R V / P A gradient increased to 58 _+ 0.6 m m H g with the initial banding. Changes in the SaO2 preceded hemodynamic changes by two to three minutes in all group B patients. Blood pressure, HR, and SaO~ returned to near baseline values when the band was cut. A less tight band was reapplied, which was associated with a smaller drop in SaO2 (92% to 96%) and an elevation of BP without bradycardia. This was considered to be acceptable banding. The R V / P A gradient significantly decreased after rebanding, and a gradient higher ~.:.than 45 m m H g was usually associated with an "increase in oxygen desaturation. .
. .
After Banding
Mean blood pressure(mmHg) Heartrate (beats/min) Ozsaturation (%) Pulmonary artery pressure (mrnHg)
A 60 • 0.7 B 58 • 0.5 A 110-+ 10 B 112-+ 11 A 98 • 6 B 98• A 42 • 6 B 41 • 9
75 _+ 0.6* 30 • 0.5* 1 1 5 • 10 50• 95-+ 7 80-+2" 28 • 0.5* 60 • 5*
RV/PA gradient (mmHg)
A B
16 • 0.5 20 • 0.5
25 • 0.3* 58 _+ 0.6*
After Rebanding
73 • 3* 125 • 13" 95•
..
9
Table 1. HemodynamicChanges During Pulmonary A r t e r y Banding Prebanding
.
7*
25 -+ 0.2* 27 -+ 0.2*
Abbreviations: A, adequate banding group (N - B); B, rebanding - , ~ group (N - 6). *P < .01 comparedwith previousstage.
DISCUSSION
Continuous monitoring of arterial oxygen saturation can be a valuable tool in the anesthetic management of patients. Arterial blood gas analyses provide accurate but intermittent information. They are not immediately or continuously available, which limits their usefulness in the moment-to-moment anesthetic management of a critically ill patient. Another technique for measuring oxygenation is transcutaneous bxygen analysis. This method provides continuous information on the adequacy of oxygenation, but its use during prolonged surgical procedures is fraught with technical difficulties. The pulse oximeter has gained popularity as an anesthetic monitoring device because it is noninvasive, hazard-free, and easy to use. It provides a continuous assessment of SaO2, using spectrophotoelectric techniques, s~ It is based on the measurement of the optical density of hemoglobin by discrete wavelengths of light. Any pulsating vascular bed, such as an earlobe or fingertip, can be placed between a two-wavelength light source and a detecting sensor. Based on this group of patients, intraoperative criteria for successful PA banding include the following: ( I ) oxygen saturation should not decrease below 90%, (2) the R V / P A pressure gradient should not be greater than 45 m m H g ,
PULSE OXIMETRY AND PULMONARY ARTERY BANDING
(3) a slight elevation of BP by 15 to 25 m m H g p o s t b a n d i n g should occur, and (4) b r a d y c a r d i a should not occur. In conclusion, pulse o x i m e t r y can be used as an i m p o r t a n t e a r l y i n d i c a t o r o f the a d e q u a c y o f the p u l m o n a r y circulation a f t e r P A b a n d i n g . It was found to be an excellent m e t h o d for e a r l y detection of i n t r a o p e r a t i v e h y p o x e m i a following P A b a n d i n g . T h e h y p o x e m i a is d u e to a P A b a n d
299
which is too tight, causing a reversal of a leftto-right shunt to a right-to-left shunt. This is followed in several m i n u t e s by b r a d y c a r d i a , hypotension, and elevation o f the R V / P A pressure gradient. ACKNOWLEDGEMENT
We thank Ellen L. Jackson for manuscript preparation.
REFERENCES
I. Stewart S, Harris P: Pulmonary artery banding. J Thorac Cardiovasc Surg 80:431-436, 1980 2. Doty DB: Pulmonary artery banding, in Glenn WWL (ed): Thoracic and Cardiovascular Surgery (ed 4). Norwalk, CT, Appleton-Century-Crofts, 1983, pp 681-687 3. Oldham HN, Kakos GS, Jarmakani MM: Pulmonary arte/y band ng in infants with complex congenital heart defects. Ann Thorac Surg 13:342-350, 1972 4. Burakobsky VI, Businola LA: Experience with surgical treatment of congenital heart disease with increased pulmonary blood flow in small children. USA-USSR Joint Symposiu m on Congenital Heart Disease, Washington, DC, ~ 1973
5. Friesen RH: Pulse oximetry during pulmonary artery surgery. Anesth Analg 64:376, 1985 6. Monaco F, Feaster WW, McQuitly JC: Continuous noninvasive oxygen saturation monitoring in sick newborns. Respir Care 28:i362, 1983 7. Whitcher C, New W Jr: Perianesthetic oxygen saturation vs the skill of the anesthetist. Anesthesiology 57:A172, 1962 (abstr) 8. Taishley RS, Kline CE: Effect on oxygen saturation by different anesthetic induction agents in cyanotic congenital heart disease. Anesthesiology 63:A3, 1985 (abstr) 9. Yelderman MV, New W Jr: Evaluation of pulse oximetry. Anesthesiology 59:349-352, 1983