- Email: [email protected]
Continuous transcutaneous oxygen monitoring during an intraoperative cardiac arrest
Resuscitation, 10 (1983) 213-218 Eisevier Scientific Publishers Ireland Ltd.
213
CONTINUOUS TRANSCUTANEOUS OXYGEN AN INTRAOPERATIVE CARDIAC ARREST
LARRY S. NICI-ITERa**, CAROL A, BRYANTb, E. WILSONa
MONITORING
DURING
KEVIN K. TREMPERC and SAMUEL
aDeportment of Surgery, Box 25, UCLA Medical Center, Los Angeles, 1000 West Carson Street, Torrance, CA 90509, bDepartment of Emergency Medicine, Santa Monica Emergency Center, Santa Monica, CA and CDepartment of Anesthesiology, University of California Irvine Medical Center, Orange, CA (U.S.A.) (Received
November
lst, 1982)
SUMMARY
A transcutaneous oxygen sensor was used continuously during surgical management of a ruptured abdominal aortic aneurysm. Closed chest compression initiated for intraoperative cardiac arrest gave an inadequate cardiac output on the basis of falling Ptco, despite transmitted femoral pulses and an excellent Pa02. This discordance provided a rationale for open cardiac massage, which increased the cardiac output and tissue perfusion (PtcO,) needed for successful resuscitation. The &cOz sensor provides immediate, non-invasive, and continuous information regarding tissue oxygenation. It reflects the Pao, in hemodynamically stable patients as well as providing a sensitive indicator for inadequate cardiac output during shock. In patients undergoing cardiopulmonary resuscitation, a falling &CO, with an acceptable Pao, indicates poor tissue perfusion and, in select circumstances, may warrant open cardiac massage.
INTRODUCTION
Transcutaneous oxygen sensors, extensively used in monitoring of the critically ill neonate, provide data on the adequacy of peripheral oxygen delivery and tissue perfusion. fn th’is report we describe intraoperative use of this technique during resection of a ruptured abdominal aortic aneurysm. Measurement of tissue oxygen transcutaneously afforded recognition of inadequate peripheral perfusion and facilitated successful resuscitation. *Address all correspondence and reprint requests to: Larry S. Nicbter, Department of Surgery, Box 25, UCLA Medical Center, Los Angeles, 1000 West Carson Street, Torrance, CA 90509, U.S.A. 0300-9572/83/0000-0000/$03.00 Printed and Published in Ireland
o 1983 Eisevier Scientific
Publishers Ireland Ltd.
214 CASEREPORT A 60-year-old white male was hospitalized because of syncope 6 h previously and chronic low back pain. The patient was in his usual state of good health until the day of admission, when his wife found him seated on the toilet, unconscious and leaning against the wall. He resumed consciousness after being placed supine for approximately 15 min. There was no past history of syncope, seizures, stroke, transient ischemic attacks or head trauma. The patient complained of severe headache over the past several days but denied abdominal and chest pain, exacerbation of lumbar back pain, or exercise intolerance. He took no medications and had not had a medical examination in the past 25 years. On physical examination, an obese mentally alert male was seen resting comfortably. Vital signs were: heart rate 68 per min, blood pressure 150/ 100 mmHg and respirations 20 per min. The conjunctivae were pale; the heart and lungs were normal to auscultation. Abdominal examination showed a pulsatile, non-tender abdominal aortic aneurysm of 10 cm diameter. No sign of retroperitoneal bleeding was evident. Lower extremity peripheral pulses were full and symmetrical. Initial laboratory data were: hematocrit = 29 vol.%, amylase 133 units (nl = 85), WBC 13 700 cells/mm j, bilirubin 1.1 mg/dl, creatinine 3 mg/dl, BUN 40 mg/dl and CPK 200 units. The chest radiograph was normal. The electrocardiogram showed ST-T wave changes indicative of ischemia. The initial diagnosis on transfer was syncope secondary to myocardial infarction, but hypovolemia secondary to a leaking abdominal aneurysm was soon suspected. The patient was directly admitted to the Intensive Care Unit where a urinary catheter was inserted and a central venous catheter was placed for continuous monitoring, Two units of packed red blood cells were infused and the patient appeared comfortable with stable vital signs. Approximately 90 min after admission the patient suddenly became hypotensive with tachycardia. This was reversed with rapid infusion of intravenous fluids, MAST suit insufflation, and awake endotracheal intubation. The patient was transported to the operating suite as an emergency and anesthetized with pancuronium bromide, fentanyl and diazepam. A radial artery catheter was placed and a transcutaneous oxygen monitor (Normetrix Medical Systems, Wallingfast, CT) was applied to the left shoulder. Laparotomy confirmed the presence of a ruptured abdominal aortic aneurysm confined to the retroperitoneum. Rapid proximal control at the neck of the infrarenal aneurysm was obtained with aortic cross clamping. The aneurysm was replaced with a 22-mm woven Dacron tube graft without difficulty. During the procedure the patient remained normotensive with transcutaneous oxygen tensions (Ptco2)in the 140-150 torr range while ventilated with 50% inspired oxygen. The blood volume was restored before gradual release of the aortic cross clamp over 10 min. The total operative time was by then 90 min.
215
Approximately 5 min after restoration of blood flow to the extremities, the PtcO, value dropped to 90 torr while the blood pressure dropped from to fall and approximately 116/70 to 106/68. The Ptco, value continued 2 min later cardiac arrest ensued with a ventricular fibrillation pattern on EKG. Immediate aortic reclamping, closed cardiac massage, and intravenous administration of bicarbonate and epinephrine preceded unsuccessful attempts at defibrillation by direct countershock (200 and 400 J). The Ptco2 value fell to zero and remained between zero and 8 torr during the resuscitation by closed chest cardiac massage. Because of a continued idioventricular cardiac rhythm and the low &CO* values it was decided to perform open cardiac massage. Through a left anterolateral thoracotomy open cardiac massage was initiated at approximately 15 min after the cardiac arrest. A rise in PtcO, values was evident within 15 s and the values reached 35 torr over the next 3 min (Fig, 1). When the Ptco, value reached 40 torr fibrillatory activity was noted, and direct cardiac counter shock (10 W-s) was successful in restoring normal sinus rhythm and normotension 29 min after the initial event. The aortic clamp was again slowly removed and the &CO, value remained stable at about 290 torr with a corresponding Pao, value of 328 torr. The systemic blood pressure progressively rose to a maximum of 220/140 mmHg and an infusion of sodium nitroprusside was
INDUCTION FENTANYL PAVULON OXYGEN
OPEN CHEST CPR (5O%F102) PnOn = 326 PTCOZ = 290 NITROPRLISSIDE
Q) AORTA UNCLAMPED,
5
CARDIAC
1
P,,O,
-----
MAP (Mean Arterial
I
‘_“_“_“,“:
(Transcutaneous PoJ Pressure)
I
I
I
3
2
-
PA0263 Prc 0266
a=*.... NSR
40
Heart Rate Normal Sinus Rhythm
Fig. 1. Changes in transcutaneous PO z, mean arterial pressure and heart rate before during and after cardiac arrest.
216
started. Blood pressure was easily controlled at 120/70 mmHg with an infusion rate of less than 1 pg/kg/min sodium nitroprusside but Ptco, decreased from nearly 300 torr to 65 torr over the next 2-3 min. An arterial blood gas revealed a substantial pulmonary shunt causing a drop in Pao, from 328 torr to 83 torr. Throughout this period the patient was well ventilated with 100% oxygen. During the remainder of the operation the Ptco, values paralleled the changes in Pao2. The patient was taken to the surgical intensive care unit in stable condition. DISCUSSION
Heated cutaneous oxygen sensors developed in 1972 (Eberhard, Hammache, and Mindt, 1972; Huch, Huch, and Meinzer, 1972) are now used routinely for neonatal intensive care monitoring (Eberhard, Mindt, Jann and Hammache, 1975; Huch, Huch, Albani, Gabriel, Schulte, Wolf, Rupprath, Emmrich, Stechele and DUC, 1976). This convenient monitoring device has not found common usage in adults due to reports of poor correlation between &co2 and Pa0 2 in hemodynamically compromised patients (Rooth, Hedstrand, Tyden and Ogren, 1976). Fortunately therein lies its value. Recently measurement of transcutaneous oxygen tension has been shown to .equate with adequacy of cardiac output during shock without hypoxia, and to correlate with PaOz during hypoxia without shock (Tremper, Waxman and Shoemaker, 1979, Tremper, Waxman, Bowman and Shoemaker, 1980; Tremper and Shoemaker, 1981a,b). Actually Ptco, values are a type of peripheral tissue oxygen tension. Specifically the Ptco, shows good correlation with the cardiac index (r = 0.90) and oxygen delivery during low flow shock. Moreover it is a reliable non-invasive trend monitor of Pao, in adult patients with adequate cardiac output (cardiac index > 2.2 l/min X m2) (Tremper and Shoemaker, 1981b). In our patient the &CO, was a sensitive indicator of the inadequacy of a closed cardiopulmonary resuscitation attempt: poor tissue perfusion persisted despite the presence of weakly transmitted femoral pulses and excellent radial PaO,. This provided the basis for abandoning closed chest cardiopulmonary resuscitation in favor of open manual cardiac massage which resulted in marked improvement in tissue perfusion and successful resuscitation. The two functional phases of the Ptcoz sensor were demonstrated in this case report. First during the cardiac arrest and resuscitation, the PtcO, “real time” evaluation of the oxygen delivery values gave a continuous (Tremper et al., 1980; Tremper and Shoemaker, 1981a,b). Secondly, the Ptco 2/Pa0 2 gradient during shock disclosed the peripheral oxygen delivery deficit. The ratio of central to peripheral 0, tension may be termed the transcutaneous PO, index (PtcO, index = Ptco2/Pao2) (Tremper and Shoemaker, 1981b). In hemodynamically stable patients the PtcO, index is normally around 0.8 (averaged 0.83 in this case) while it falls to approxi-
217
mately 0.5 during low flow states (cardiac index approximately 2 l/min X m2 ) and to approximately 0.1 during severe shock (cardiac index approx. 1.0 l/min mZ) (Tremper and Shoemaker, 1981b). Once our patient was resuscitated the Ptco, values closely followed the PaO, values. A linear regression analysis produced a correlation coefficient of r = 0.99 with a slope and intercept of 1.1 and 10 torr, respectively. When sodium nitroprusside was infused to control the hypertension a marked pulmonary shunt developed. Vasodilating drugs (especially nitroprusside) block the reflex hypoxic pulmonary vasoconstriction which is the normal compensatory response the lung uses to divert pulmonary blood flow from poorly oxygenated portions of the lung (Benumof, 1979). This patient had obvious pulmonary damage after 29 min of cardiopulmonary resuscitation and the SNP dilated the constricted pulmonary capillaries in the damaged hypoxic portion of lung causing the large shunt and low PaO, values (83 torr with FIo2 = 1.0). The advantages of transcutaneous oxygen monitoring routinely in potentially unstable patients include simplicity of application, immediate availability, relatively low cost, and provision of a non-invasive continuous monitoring system which reflects peripheral oxygen delivery. As an early indicator of cardiorespiratory embarrassment it may be used as an adjunct to or as an indication for more invasive monitoring. Even accurate Paoz measurements may be obtained within 1 min by simply placing a drop of arterial blood on the surface of the PtcO, electrode (Applebaum, Bowman, Tremper and Shoemaker, 1981). Looking to the future, non-invasive continuous monitoring devices such as the transcutaneous oxygen electrode represent a major advance in the care of all critically ill patients. From a mainstay in well controlled environments such as the surgical theater, intensive care unit or emergency room, we anticipate extension of their use to paramedics in field situations. REFERENCES Applebaum, R., Bowman, R., Tremper, K.K. and Shoemaker, W.C. (1981) Emergency Pao, estimates in one minute with a transcutaneous oxygen sensor. Crit. Care Med., 9,742-743. Benumof, J.L. (1979) Hypoxic pulmonary vasoconstriction and sodium nitroprusside infusion. Anesthesiology, 50,481-483. Eberhard, P., Hammache, K. and Mindt, W. (1972) Perkutane messung des sauerstaffpartialdruckes: methodik und anwendungen. Stuttg. Proc. Med.-Tech., 26. Eberhard, P., Mindt, W., Jann, F. and Hammache, K. (1975) Continuous Po? monitoring in the neonate by skin electrodes. Med. Biol. Eng., 13, 436-442. Huch, R., Huch, A., Albani, M., Gabriel, M., Schulte, F.J., Wolf, H., Rupprath, G., Emmrich, P., Stechele, U. and Due. G. (1976) Transcutaneous PO, monitoring in routine management of infants and children with cardiorespiratory problems. Pediatrics, 57, 681-690. Huch, A., Huch, R. and Meinzer, K., (1972) Eine schnelle, Behizte ptoberflachenelektrode zur kontinuirt lichen unberwach ling des PO, beim menschen. Eleklrodenaufbau und eigenschafte. Methodik und anwendungen. Stuttg. Proc. Med.-Tech., 26.
218 Rooth, G., Hedstrand, U.L.F., Tyden, H. and Ogren, C. (1976) The validity of the transcutaneous oxygen tension method in adults. Grit. Care Med., 4, 162-165. Tremper, K.K. and Shoemaker, W.C. (1981a) Continuous CPR monitoring with transcutaneous oxygen and carbondioxide sensors. Crit. Care Med., 9,417-418. Tremper, K.K. and Shoemaker, W.C. (1981b) Transcutaneous oxygen monitoring of critically ill adults with and without low flow shock. Crit. Care Med., 9, 706-709. Tremper, K.K., Waxman, K., Bowman and Shoemaker, W.C. (1980) Continuous trancutaneous oxygen monitoring during respiratory failure, cardiac decompensation, cardiac arrest, and CPR. Crit. Care Med., 8, 377-381. Tremper, K.K., Waxman, K. and Shoemaker, W.C. (1979) Effects of hypoxia and shock on transcutaneousPo z values in dogs. Crit. Care Med., 7,526-531.
213
CONTINUOUS TRANSCUTANEOUS OXYGEN AN INTRAOPERATIVE CARDIAC ARREST
LARRY S. NICI-ITERa**, CAROL A, BRYANTb, E. WILSONa
MONITORING
DURING
KEVIN K. TREMPERC and SAMUEL
aDeportment of Surgery, Box 25, UCLA Medical Center, Los Angeles, 1000 West Carson Street, Torrance, CA 90509, bDepartment of Emergency Medicine, Santa Monica Emergency Center, Santa Monica, CA and CDepartment of Anesthesiology, University of California Irvine Medical Center, Orange, CA (U.S.A.) (Received
November
lst, 1982)
SUMMARY
A transcutaneous oxygen sensor was used continuously during surgical management of a ruptured abdominal aortic aneurysm. Closed chest compression initiated for intraoperative cardiac arrest gave an inadequate cardiac output on the basis of falling Ptco, despite transmitted femoral pulses and an excellent Pa02. This discordance provided a rationale for open cardiac massage, which increased the cardiac output and tissue perfusion (PtcO,) needed for successful resuscitation. The &cOz sensor provides immediate, non-invasive, and continuous information regarding tissue oxygenation. It reflects the Pao, in hemodynamically stable patients as well as providing a sensitive indicator for inadequate cardiac output during shock. In patients undergoing cardiopulmonary resuscitation, a falling &CO, with an acceptable Pao, indicates poor tissue perfusion and, in select circumstances, may warrant open cardiac massage.
INTRODUCTION
Transcutaneous oxygen sensors, extensively used in monitoring of the critically ill neonate, provide data on the adequacy of peripheral oxygen delivery and tissue perfusion. fn th’is report we describe intraoperative use of this technique during resection of a ruptured abdominal aortic aneurysm. Measurement of tissue oxygen transcutaneously afforded recognition of inadequate peripheral perfusion and facilitated successful resuscitation. *Address all correspondence and reprint requests to: Larry S. Nicbter, Department of Surgery, Box 25, UCLA Medical Center, Los Angeles, 1000 West Carson Street, Torrance, CA 90509, U.S.A. 0300-9572/83/0000-0000/$03.00 Printed and Published in Ireland
o 1983 Eisevier Scientific
Publishers Ireland Ltd.
214 CASEREPORT A 60-year-old white male was hospitalized because of syncope 6 h previously and chronic low back pain. The patient was in his usual state of good health until the day of admission, when his wife found him seated on the toilet, unconscious and leaning against the wall. He resumed consciousness after being placed supine for approximately 15 min. There was no past history of syncope, seizures, stroke, transient ischemic attacks or head trauma. The patient complained of severe headache over the past several days but denied abdominal and chest pain, exacerbation of lumbar back pain, or exercise intolerance. He took no medications and had not had a medical examination in the past 25 years. On physical examination, an obese mentally alert male was seen resting comfortably. Vital signs were: heart rate 68 per min, blood pressure 150/ 100 mmHg and respirations 20 per min. The conjunctivae were pale; the heart and lungs were normal to auscultation. Abdominal examination showed a pulsatile, non-tender abdominal aortic aneurysm of 10 cm diameter. No sign of retroperitoneal bleeding was evident. Lower extremity peripheral pulses were full and symmetrical. Initial laboratory data were: hematocrit = 29 vol.%, amylase 133 units (nl = 85), WBC 13 700 cells/mm j, bilirubin 1.1 mg/dl, creatinine 3 mg/dl, BUN 40 mg/dl and CPK 200 units. The chest radiograph was normal. The electrocardiogram showed ST-T wave changes indicative of ischemia. The initial diagnosis on transfer was syncope secondary to myocardial infarction, but hypovolemia secondary to a leaking abdominal aneurysm was soon suspected. The patient was directly admitted to the Intensive Care Unit where a urinary catheter was inserted and a central venous catheter was placed for continuous monitoring, Two units of packed red blood cells were infused and the patient appeared comfortable with stable vital signs. Approximately 90 min after admission the patient suddenly became hypotensive with tachycardia. This was reversed with rapid infusion of intravenous fluids, MAST suit insufflation, and awake endotracheal intubation. The patient was transported to the operating suite as an emergency and anesthetized with pancuronium bromide, fentanyl and diazepam. A radial artery catheter was placed and a transcutaneous oxygen monitor (Normetrix Medical Systems, Wallingfast, CT) was applied to the left shoulder. Laparotomy confirmed the presence of a ruptured abdominal aortic aneurysm confined to the retroperitoneum. Rapid proximal control at the neck of the infrarenal aneurysm was obtained with aortic cross clamping. The aneurysm was replaced with a 22-mm woven Dacron tube graft without difficulty. During the procedure the patient remained normotensive with transcutaneous oxygen tensions (Ptco2)in the 140-150 torr range while ventilated with 50% inspired oxygen. The blood volume was restored before gradual release of the aortic cross clamp over 10 min. The total operative time was by then 90 min.
215
Approximately 5 min after restoration of blood flow to the extremities, the PtcO, value dropped to 90 torr while the blood pressure dropped from to fall and approximately 116/70 to 106/68. The Ptco, value continued 2 min later cardiac arrest ensued with a ventricular fibrillation pattern on EKG. Immediate aortic reclamping, closed cardiac massage, and intravenous administration of bicarbonate and epinephrine preceded unsuccessful attempts at defibrillation by direct countershock (200 and 400 J). The Ptco2 value fell to zero and remained between zero and 8 torr during the resuscitation by closed chest cardiac massage. Because of a continued idioventricular cardiac rhythm and the low &CO* values it was decided to perform open cardiac massage. Through a left anterolateral thoracotomy open cardiac massage was initiated at approximately 15 min after the cardiac arrest. A rise in PtcO, values was evident within 15 s and the values reached 35 torr over the next 3 min (Fig, 1). When the Ptco, value reached 40 torr fibrillatory activity was noted, and direct cardiac counter shock (10 W-s) was successful in restoring normal sinus rhythm and normotension 29 min after the initial event. The aortic clamp was again slowly removed and the &CO, value remained stable at about 290 torr with a corresponding Pao, value of 328 torr. The systemic blood pressure progressively rose to a maximum of 220/140 mmHg and an infusion of sodium nitroprusside was
INDUCTION FENTANYL PAVULON OXYGEN
OPEN CHEST CPR (5O%F102) PnOn = 326 PTCOZ = 290 NITROPRLISSIDE
Q) AORTA UNCLAMPED,
5
CARDIAC
1
P,,O,
-----
MAP (Mean Arterial
I
‘_“_“_“,“:
(Transcutaneous PoJ Pressure)
I
I
I
3
2
-
PA0263 Prc 0266
a=*.... NSR
40
Heart Rate Normal Sinus Rhythm
Fig. 1. Changes in transcutaneous PO z, mean arterial pressure and heart rate before during and after cardiac arrest.
216
started. Blood pressure was easily controlled at 120/70 mmHg with an infusion rate of less than 1 pg/kg/min sodium nitroprusside but Ptco, decreased from nearly 300 torr to 65 torr over the next 2-3 min. An arterial blood gas revealed a substantial pulmonary shunt causing a drop in Pao, from 328 torr to 83 torr. Throughout this period the patient was well ventilated with 100% oxygen. During the remainder of the operation the Ptco, values paralleled the changes in Pao2. The patient was taken to the surgical intensive care unit in stable condition. DISCUSSION
Heated cutaneous oxygen sensors developed in 1972 (Eberhard, Hammache, and Mindt, 1972; Huch, Huch, and Meinzer, 1972) are now used routinely for neonatal intensive care monitoring (Eberhard, Mindt, Jann and Hammache, 1975; Huch, Huch, Albani, Gabriel, Schulte, Wolf, Rupprath, Emmrich, Stechele and DUC, 1976). This convenient monitoring device has not found common usage in adults due to reports of poor correlation between &co2 and Pa0 2 in hemodynamically compromised patients (Rooth, Hedstrand, Tyden and Ogren, 1976). Fortunately therein lies its value. Recently measurement of transcutaneous oxygen tension has been shown to .equate with adequacy of cardiac output during shock without hypoxia, and to correlate with PaOz during hypoxia without shock (Tremper, Waxman and Shoemaker, 1979, Tremper, Waxman, Bowman and Shoemaker, 1980; Tremper and Shoemaker, 1981a,b). Actually Ptco, values are a type of peripheral tissue oxygen tension. Specifically the Ptco, shows good correlation with the cardiac index (r = 0.90) and oxygen delivery during low flow shock. Moreover it is a reliable non-invasive trend monitor of Pao, in adult patients with adequate cardiac output (cardiac index > 2.2 l/min X m2) (Tremper and Shoemaker, 1981b). In our patient the &CO, was a sensitive indicator of the inadequacy of a closed cardiopulmonary resuscitation attempt: poor tissue perfusion persisted despite the presence of weakly transmitted femoral pulses and excellent radial PaO,. This provided the basis for abandoning closed chest cardiopulmonary resuscitation in favor of open manual cardiac massage which resulted in marked improvement in tissue perfusion and successful resuscitation. The two functional phases of the Ptcoz sensor were demonstrated in this case report. First during the cardiac arrest and resuscitation, the PtcO, “real time” evaluation of the oxygen delivery values gave a continuous (Tremper et al., 1980; Tremper and Shoemaker, 1981a,b). Secondly, the Ptco 2/Pa0 2 gradient during shock disclosed the peripheral oxygen delivery deficit. The ratio of central to peripheral 0, tension may be termed the transcutaneous PO, index (PtcO, index = Ptco2/Pao2) (Tremper and Shoemaker, 1981b). In hemodynamically stable patients the PtcO, index is normally around 0.8 (averaged 0.83 in this case) while it falls to approxi-
217
mately 0.5 during low flow states (cardiac index approximately 2 l/min X m2 ) and to approximately 0.1 during severe shock (cardiac index approx. 1.0 l/min mZ) (Tremper and Shoemaker, 1981b). Once our patient was resuscitated the Ptco, values closely followed the PaO, values. A linear regression analysis produced a correlation coefficient of r = 0.99 with a slope and intercept of 1.1 and 10 torr, respectively. When sodium nitroprusside was infused to control the hypertension a marked pulmonary shunt developed. Vasodilating drugs (especially nitroprusside) block the reflex hypoxic pulmonary vasoconstriction which is the normal compensatory response the lung uses to divert pulmonary blood flow from poorly oxygenated portions of the lung (Benumof, 1979). This patient had obvious pulmonary damage after 29 min of cardiopulmonary resuscitation and the SNP dilated the constricted pulmonary capillaries in the damaged hypoxic portion of lung causing the large shunt and low PaO, values (83 torr with FIo2 = 1.0). The advantages of transcutaneous oxygen monitoring routinely in potentially unstable patients include simplicity of application, immediate availability, relatively low cost, and provision of a non-invasive continuous monitoring system which reflects peripheral oxygen delivery. As an early indicator of cardiorespiratory embarrassment it may be used as an adjunct to or as an indication for more invasive monitoring. Even accurate Paoz measurements may be obtained within 1 min by simply placing a drop of arterial blood on the surface of the PtcO, electrode (Applebaum, Bowman, Tremper and Shoemaker, 1981). Looking to the future, non-invasive continuous monitoring devices such as the transcutaneous oxygen electrode represent a major advance in the care of all critically ill patients. From a mainstay in well controlled environments such as the surgical theater, intensive care unit or emergency room, we anticipate extension of their use to paramedics in field situations. REFERENCES Applebaum, R., Bowman, R., Tremper, K.K. and Shoemaker, W.C. (1981) Emergency Pao, estimates in one minute with a transcutaneous oxygen sensor. Crit. Care Med., 9,742-743. Benumof, J.L. (1979) Hypoxic pulmonary vasoconstriction and sodium nitroprusside infusion. Anesthesiology, 50,481-483. Eberhard, P., Hammache, K. and Mindt, W. (1972) Perkutane messung des sauerstaffpartialdruckes: methodik und anwendungen. Stuttg. Proc. Med.-Tech., 26. Eberhard, P., Mindt, W., Jann, F. and Hammache, K. (1975) Continuous Po? monitoring in the neonate by skin electrodes. Med. Biol. Eng., 13, 436-442. Huch, R., Huch, A., Albani, M., Gabriel, M., Schulte, F.J., Wolf, H., Rupprath, G., Emmrich, P., Stechele, U. and Due. G. (1976) Transcutaneous PO, monitoring in routine management of infants and children with cardiorespiratory problems. Pediatrics, 57, 681-690. Huch, A., Huch, R. and Meinzer, K., (1972) Eine schnelle, Behizte ptoberflachenelektrode zur kontinuirt lichen unberwach ling des PO, beim menschen. Eleklrodenaufbau und eigenschafte. Methodik und anwendungen. Stuttg. Proc. Med.-Tech., 26.
218 Rooth, G., Hedstrand, U.L.F., Tyden, H. and Ogren, C. (1976) The validity of the transcutaneous oxygen tension method in adults. Grit. Care Med., 4, 162-165. Tremper, K.K. and Shoemaker, W.C. (1981a) Continuous CPR monitoring with transcutaneous oxygen and carbondioxide sensors. Crit. Care Med., 9,417-418. Tremper, K.K. and Shoemaker, W.C. (1981b) Transcutaneous oxygen monitoring of critically ill adults with and without low flow shock. Crit. Care Med., 9, 706-709. Tremper, K.K., Waxman, K., Bowman and Shoemaker, W.C. (1980) Continuous trancutaneous oxygen monitoring during respiratory failure, cardiac decompensation, cardiac arrest, and CPR. Crit. Care Med., 8, 377-381. Tremper, K.K., Waxman, K. and Shoemaker, W.C. (1979) Effects of hypoxia and shock on transcutaneousPo z values in dogs. Crit. Care Med., 7,526-531.