- Email: [email protected]
Arch versus femoral artery perfusion during cardiopulmonary bypass
Arch versus femoral artery perfusion during cardiopulmonary bypass Aortic arch perfusion is favored by most cardiac surgeons. Perfusion via the femoral artery is still used sporadically at reoperations, for aneurysms of the ascending aorta, or for the institution of partial pump support in very sick patients prior to opening of the chest. Our over-all experience indicates that surgical complications occurred primarily in the group of patients perfused via the femoral artery. On the other hand, serious disturbance in cerebral perfusion, as determined by electroencephalogram (EEG) monitoring, occurred in 7 percent of the patients perfused via the arch and 3 percent of those perfused via the femoral artery, a difference that was not statistically significant. We continue to advocate aortic arch cannulation and EEG monitoring during cardiopulmonary bypass procedures.
Tomas A. Salerno, M.D., F.R.C.S.(C), Dario P. Lince, M.D., F.R.C.S.(C), Denis N. White, M . D . , F.R.C.P.,* R. Beverley Lynn, M . D . , F.R.C.S.(C), and Edward J. P. Charrette, M.D., F.R.C.S.(C), Kingston, Ontario, Canada
M ost cardiac surgeons favor aortic arch cannulation during cardiopulmonary bypass (CPB) because of technical simplicity, superior antegrade perfusion, and elimination of the risk of retrograde dissection. 1 Abnormal perfusion of the carotids, 2, 3 hemorrhage, 1 and development of aneurysm at the cannulation site 4 are potential hazards of aortic arch perfusion. Currently, femoral artery cannulation is used only in selected cases, such as in reoperations and ascending aorta and arch aneurysms, because there are the potential hazards of limb ischemia and retrograde aortic dissection in the atherosclerotic vessels, 1, 5 " 6 infection, lymph fistula, muscle necrosis, arterial thrombosis, bleeding, and stenosis at the arteriotomy site. 7 The purpose of this study was to determine the incidence of surgical complications and of abnormalities in cerebral function in comparable groups of patients undergoing elective CPB procedures utilizing these two modes of perfusion and to compare this prospective analysis with a retrospective over-all experience at Queen's University. From the Departments of Surgery and Neurology, Queen's University, Kingston, Ontario, Canada. Supported by the Ontario Heart Foundation Grant 2-4. Received for publication April 25, 1978. Accepted for publication July 31, 1978. Address for reprints: Dr. T. A. Salerno, Department of Surgery, Queen's University, Kingston, Ontario, Canada. *Department of Neurology.
Patients and methods From January to June, 1977, 94 patients undergoing CPB procedures were perfused via the femoral artery. The mean age for the group was 53 years (range, 9 to 60 years). Procedures performed are shown in Table I. From July, 1977, to March, 1978, 84 patients underwent similar procedures (Table II) and were perfused via the aortic arch. Mean age for this group was 51 years (range, 8 to 72 years). All patients were premedicated with morphine and Valium and anesthesia was induced with morphine, Pentothal sodium, and succinylcholine. Anesthesia was maintained with nitrous oxide-oxygen mixture prior to CPB and with morphine and d-tubocurarine during CPB. All operations were performed with the disposable Bentley bubble oxygenator.* Arterial gases, hematocrit value, pump flow, arterial pressure, and body temperature were monitored routinely. Body temperature was 27 to 30° C. in all 94 patients perfused via the femoral artery; of patients perfused via the arch 36 were normothermic (37° C.) and 48 were hypothermic (28° C ) . Prior to caval cannulation and after systemic heparinization, an 8 mm. Morris aortic arch cannulat was inserted into the ascending aorta through a small stab wound. The cannula was held in place by double purse*Bentley Laboratories, Inc., Santa Ana, Calif. tSarns Laboratories, Inc., Ann Arbor, Mich.
0022-5223/78/110681+04$00.40/0 © 1978 The C. V. Mosby Co.
681
The Journal of Thoracic and Cardiovascular Surgery
6 8 2 Salerno et al.
Table I. Procedures performed using femoral artery perfusion (body temperature, 27 to 30° C.)
Table II. Procedures performed using aortic arch perfusion No. of patients
No. of patients Coronary bypass Aortic valve replacement Mitral valve replacement Atrial septal defect Miscellaneous
57 12 12 7 6
Coronary bypass Aortic valve replacement Mitral valve replacement Double valve Miscellaneous
49 11 11 4 9
Total
94
Total
84
string sutures, buttressed with Teflon felt pledgets. Femoral artery cannulation was carried out by performing a transverse arteriotomy in the superficial femoral artery and introducing a size 18 Fr. arterial catheter* into the artery. Operative field flooding with carbon dioxide and the usual measures to remove air from the left heart were used in all cases in which the heart was opened. The electroencephalogram (EEG) was monitored in all cases by means of eight recording channels from bipolar intradermal electrodes placed at the beginning of the operation. A baseline EEG was recorded prior to the commencement of the surgical procedure. When EEG abnormalities occurred during CPB an effort was made to identify and correct the problem. Our over-all retrospective experience with EEG monitoring during CPB and the surgical complications resulting from femoral and arch cannulation were then reviewed. It was hoped that such an analysis might shed some light on the relative safety of these techniques of perfusion. Results Prospective analysis. There were no serious complications associated with arch cannulation. The only minor problem was bleeding at the site of the arterial puncture at the end of CPB in two patients, necessitating extra sutures. Two patients with femoral artery cannulation required exploration for bleeding at the arteriotomy site. Vascular complications developed in three patients and consisted of claudication (one), absent pedal pulses (one), and arterial damage requiring angioplasty (one). There were three deaths unrelated to the cannulation technique in the group of patients perfused via the femoral artery: one due to a myocardial infarction and two due to low-output state. Four deaths occurred in the group perfused via the *USCI International, Inc., Glens Falls, N. Y.
Body temperature 37 28 28 28 28
arch, all unrelated to the technique in use: Three patients died owing to low-output state (one aortic valve replacement, one aorta-coronary bypass, and one mitral valve with aorta-coronary bypass); one patient who underwent aortic valve replacement died after 30 days because of a stroke secondary to a calcium embolus. The EEG was normal during CPB in 76 of 94 patients (81 percent) undergoing femoral artery perfusion. Transient, slow wave activity occurred in the EEG's of 15 of 94 patients (16 percent) at the start of CPB. Major EEG abnormalities occurred in three of 84 patients (3 percent) and consisted of wave suppression owing to low systemic pressure in all three (Table III). The EEG returned to normal when the pump flow was increased. Of the 84 patients undergoing arch perfusion the EEG's were normal in 64 patients (76 percent): 29 normothermic and 35 hypothermic patients. The EEG's showed minor abnormalities during the first minutes of CPB in 14 patients (17 percent): six normothermic and eight hypothermic patients. Major EEG abnormalities occurred in six patients (7 percent): three normothermic and three hypothermic patients. The EEG's in five of six patients returned to normal after identification and correction of the problem (inadequate perfusion in two and malposition of the arch cannula in three). One patient had a cardiovascular accident prior to surgery which accounted for the abnormal EEG (Table IV). There was no significant difference between EEG abnormalities during arch and femoral perfusions (chi square analysis). Retrospective analysis. The records of 300 patients undergoing cardiopulmonary bypass procedures in which the EEG was monitored were reviewed over a period beginning in 1964. There were 220 male and 80 female patients ranging in age from 3 to 67 years (mean, 53 years). Procedures performed are shown in Table V. Femoral artery cannulation and hypothermia (30° C.) were used in all patients. Analysis of the EEG recordings revealed normal EEG's throughout CPB in 186
Volume 76 Number 5 November, 1978
Arch
Table I I I . EEG abnormalities femoral perfusion No. of patients
Normal Minor EEG abnormality Major EEG abnormality
76 15 3
81 16 3
Totals
94
100
Percent
artery perfusion
during
6 8 3
CPB
Table V. Procedures performed in 300 CPB operations with EEG monitoring
during
EEG
vs. femoral
Procedures
No. of patients
Coronary surgery Valve surgery Congenital heart surgery Left ventricular aneurysm resection Others
114 89 70 15 12
Total
300
Body temperature (°C.) 28 28 28
Table IV. EEG abnormality in arch perfusion EEG Normal
No. of patients 64
Percent 76
Minor EEG abnormality Major EEG abnormality
14
17
6
7
Totals
84
100
No. of patients/ body temperature 29Normothermic37°C. 35 Hypothermic 28° C. 6 Normothermic 8 Hypothermic 3 Normothermic 3 Hypothermic
patients (62 percent); minor EEG changes (slow wave activity and decreased voltage at the start of CPB) occurred temporarily in 73 patients (24 percent). In 41 patients (14 percent) major EEG changes occurred (wave suppression either localized to one hemisphere or diffuse). These changes were due to hypotension in 33 patients, cannula problems in five patients, hypoxia in one patient, airembolus in one patient, and unknown cause in one patient. To study the incidence of surgical complications caused by the perfusion techniques the records of 867 patients who have undergone CPB procedures since 1958 (702 femoral artery and 165 aortic arch cannulations) were reviewed. Five patients with minor bleeding at the site of aortic arch cannulation required a simple suture at time of operation. No other complication resulted from this modality of perfusion. The complications resulting from femoral artery cannulation are presented in Table VI. As noted, in 15 patients the femoral artery could not be used owing to severe arteriosclerosis. There was no retrograde aortic dissection and no loss of limbs. No death in the series could be attributed to femoral artery cannulation. Therefore, in both the prospective and the retrospective series there were more complications due to femoral artery perfusion than to aortic arch cannulation. Discussion Most cardiac surgeons favor cannulation of the aortic arch during routine CPB procedures because the
Table VI. Complications due to femoral artery cannulation in 702 patients Complications Iliac artery perforation Blocked femoral artery Asymptomatic Claudication Transient ischemia (spasm) Bleeding femoral artery requiring operation Difficult arterial repair post perfusion* Minor wound infection Totals
No. of patients 1
Percent 0.14
2 3 3 5
0.71
10
1.42
3
0.43
27
3.84
1.14
*One patient required interposition of a femoral artery graft.
technique is relatively simple and the complication rate is low. 4,8 ' 9 A groin incision is not required, damage to the femoral artery is eliminated, and atheromatous debris and thrombi cannot be dislodged from the iliac arteries.10 Cannulation of the arch, however, is not without hazard. Imperfect perfusion of the arch vessels may occur during this type of cannulation,11, 12 causing unexplained neurologic complications. Poor perfusion of the carotid arteries may occur despite high pump flow rates because of the Coanda effect.3' 13, 14 Without proper monitoring of cerebral perfusion the resultant neurologic deficits may wrongly be ascribed to other problems during perfusion.2,15 The EEG is a very sensitive, noninvasive means of determining the adequacy of cerebral blood flow.15,16 Changes in the EEG closely parallel the changes in cerebral perfusion.17, 18 The prospective analysis demonstrates that the incidence of major EEG abnormalities, i.e., changes indicating EEG wave suppression secondary to poor cerebral circulation, was not significantly different between arch and femoral perfusion. Most of these abnormalities were identified and corrected by increasing the
684
The Journal of Thoracic and Cardiovascular Surgery
Salerno et al.
pump flow rate in two patients with unsuspected carotid artery disease and by cannula repositioning in three patients perfused via the arch. In the three patients perfused via the femoral artery major EEG wave abnormalities were corrected by increasing the pump flow rate. Minor EEG abnormalities occurred equally in both groups of patients and reverted to normal immediately after full CPB was instituted without requiring manipulation by the surgeon. The retrospective analysis demonstrated that surgical complications occurred primarily in the group of patients perfused via the femoral artery. However, there was no retrograde dissection, limb loss, or death that could be attributed to this modality of perfusion. Because more patients have been perfused via the femoral artery and the experience with this technique is more extensive, follow-up of our patients perfused via the arch will be necessary in order to complete the information provided herein. Major EEG changes were more common in the retrospective series (14 percent) than in the prospective series (3 percent), probably reflecting low pump flows and higher incidence of hypotension and low-output state in our early experience. At present we continue to advocate the use of arch perfusion for most CPB procedures. We feel, however, that it is of great importance to monitor cerebral function when extracorporeal circulation is used. Clinical assessment during CPB may be misleading and cerebral perfusion may be abnormal because of unsuspected carotid disease, flow disturbance, or the Coanda effect occurring with arch perfusion. EEG monitoring allows greater safety for the patient undergoing open-heart surgery, irrespective of the type of perfusion. We would like to acknowledge the technical assistance of Mrs. B. Marsh and Miss N. Drew for EEG monitoring and Miss D. Judson for typing this manuscript.
REFERENCES 1 Taylor PC, Groves LK, Loop FD, Effler DB: Cannulation of the ascending aorta for cardiopulmonary bypass. J THORAC CARDIOVASC SURG 71:225, 1975
2 Krous HF, Mansfield PB, Sauvage LR: Carotid artery
hyperperfusion during open-heart surgery. J THORAC CARDIOVASC SURG 66:118, 1973
3 Magilligan DJ Jr, Eastland MW, Lell WA, DeWeese JA, Mahoney EB: Decreased carotid flow with ascending aortic cannulation. Circulation 45, 46:Suppl 1:130, 1972 4 Flick WF, Hallermann FJ, Feldt RH, Danielson GK: Aneurysm of aortic cannulation site. Successful repair by means of peripheral cannulation, profound hypothermia and circulatory arrest. J THORAC CARDIOVASC SURG
61:419, 1971 5 Benedict JS, Buhl TL, Henney RP: Acute aortic dissection during cardiopulmonary bypass. Arch Surg 108:810, 1974 6 Reinke T, Harris RD, Klein AJ, Daily PO: Aorto iliac dissection due to aortic cannulation. Ann Thorac Surg 18:295, 1974 7 Daily PO, Fogarty TJ, Shumway NE: Cannulation of the ascending aorta. Ann Thorac Surg 12:85, 1971 8 Roe BB, Kelly PB: Perfusion through the ascending aorta: Experience with 410 cases. Ann Thorac Surg 7:238,1969 9 Salama FD, Blesovsky A: Complications of cannulation of the ascending aorta for open heart surgery. Thorax 25:604, 1970 10 Kay JR, Dykstra PC, Tsuji HK: Retrograde ilioaortic dissection: A complication of common femoral artery perfusion during open heart surgery. Am J Surg 111:464, 1966 11 Magner JB: Complications of aortic cannulation for open-heart surgery. Thorax 26:172, 1971 12 Parker R: Aortic cannulation. Thorax 24:742, 1969 13 Angrist SW: Fluid control devices. Sci Am 211:80, 1964 14 Cutler SS, Burbank B: Fluid control devices, "rheostasis" and the human circulation. J THORAC CARDIOVASC SURG 56:558, 1968
15 Salerno TA, Lince DP, White DN, Lynn RB, Charrette EJP: E.E.G. monitoring during open-heart surgery: A prospective analysis of 118 cases. J THORAC CARDIOVASC SURG 76:97, 1978
16 Laxenaire M, Tridon P, Mathieu P, Picard L, Wever M: E.E.G. features during open heart surgery under extracorporeal circulation. Personal experience of 30 cases. Electroencephalogr Clin Neurophysiol 23:381, 1967 17 Sulg I, Ingvar DH: Regional cerebral blood flow and E.E.G. frequency content. Electroencephalogr Clin Neurophysiol 23:395, 1967 18 Freeman J, Invar DH: Influence of tissue hypoxia upon the E.E.G.—cerebral blood flow relationship. Electroencephalogr Clin Neurophysiol 23:395, 1967
Tomas A. Salerno, M.D., F.R.C.S.(C), Dario P. Lince, M.D., F.R.C.S.(C), Denis N. White, M . D . , F.R.C.P.,* R. Beverley Lynn, M . D . , F.R.C.S.(C), and Edward J. P. Charrette, M.D., F.R.C.S.(C), Kingston, Ontario, Canada
M ost cardiac surgeons favor aortic arch cannulation during cardiopulmonary bypass (CPB) because of technical simplicity, superior antegrade perfusion, and elimination of the risk of retrograde dissection. 1 Abnormal perfusion of the carotids, 2, 3 hemorrhage, 1 and development of aneurysm at the cannulation site 4 are potential hazards of aortic arch perfusion. Currently, femoral artery cannulation is used only in selected cases, such as in reoperations and ascending aorta and arch aneurysms, because there are the potential hazards of limb ischemia and retrograde aortic dissection in the atherosclerotic vessels, 1, 5 " 6 infection, lymph fistula, muscle necrosis, arterial thrombosis, bleeding, and stenosis at the arteriotomy site. 7 The purpose of this study was to determine the incidence of surgical complications and of abnormalities in cerebral function in comparable groups of patients undergoing elective CPB procedures utilizing these two modes of perfusion and to compare this prospective analysis with a retrospective over-all experience at Queen's University. From the Departments of Surgery and Neurology, Queen's University, Kingston, Ontario, Canada. Supported by the Ontario Heart Foundation Grant 2-4. Received for publication April 25, 1978. Accepted for publication July 31, 1978. Address for reprints: Dr. T. A. Salerno, Department of Surgery, Queen's University, Kingston, Ontario, Canada. *Department of Neurology.
Patients and methods From January to June, 1977, 94 patients undergoing CPB procedures were perfused via the femoral artery. The mean age for the group was 53 years (range, 9 to 60 years). Procedures performed are shown in Table I. From July, 1977, to March, 1978, 84 patients underwent similar procedures (Table II) and were perfused via the aortic arch. Mean age for this group was 51 years (range, 8 to 72 years). All patients were premedicated with morphine and Valium and anesthesia was induced with morphine, Pentothal sodium, and succinylcholine. Anesthesia was maintained with nitrous oxide-oxygen mixture prior to CPB and with morphine and d-tubocurarine during CPB. All operations were performed with the disposable Bentley bubble oxygenator.* Arterial gases, hematocrit value, pump flow, arterial pressure, and body temperature were monitored routinely. Body temperature was 27 to 30° C. in all 94 patients perfused via the femoral artery; of patients perfused via the arch 36 were normothermic (37° C.) and 48 were hypothermic (28° C ) . Prior to caval cannulation and after systemic heparinization, an 8 mm. Morris aortic arch cannulat was inserted into the ascending aorta through a small stab wound. The cannula was held in place by double purse*Bentley Laboratories, Inc., Santa Ana, Calif. tSarns Laboratories, Inc., Ann Arbor, Mich.
0022-5223/78/110681+04$00.40/0 © 1978 The C. V. Mosby Co.
681
The Journal of Thoracic and Cardiovascular Surgery
6 8 2 Salerno et al.
Table I. Procedures performed using femoral artery perfusion (body temperature, 27 to 30° C.)
Table II. Procedures performed using aortic arch perfusion No. of patients
No. of patients Coronary bypass Aortic valve replacement Mitral valve replacement Atrial septal defect Miscellaneous
57 12 12 7 6
Coronary bypass Aortic valve replacement Mitral valve replacement Double valve Miscellaneous
49 11 11 4 9
Total
94
Total
84
string sutures, buttressed with Teflon felt pledgets. Femoral artery cannulation was carried out by performing a transverse arteriotomy in the superficial femoral artery and introducing a size 18 Fr. arterial catheter* into the artery. Operative field flooding with carbon dioxide and the usual measures to remove air from the left heart were used in all cases in which the heart was opened. The electroencephalogram (EEG) was monitored in all cases by means of eight recording channels from bipolar intradermal electrodes placed at the beginning of the operation. A baseline EEG was recorded prior to the commencement of the surgical procedure. When EEG abnormalities occurred during CPB an effort was made to identify and correct the problem. Our over-all retrospective experience with EEG monitoring during CPB and the surgical complications resulting from femoral and arch cannulation were then reviewed. It was hoped that such an analysis might shed some light on the relative safety of these techniques of perfusion. Results Prospective analysis. There were no serious complications associated with arch cannulation. The only minor problem was bleeding at the site of the arterial puncture at the end of CPB in two patients, necessitating extra sutures. Two patients with femoral artery cannulation required exploration for bleeding at the arteriotomy site. Vascular complications developed in three patients and consisted of claudication (one), absent pedal pulses (one), and arterial damage requiring angioplasty (one). There were three deaths unrelated to the cannulation technique in the group of patients perfused via the femoral artery: one due to a myocardial infarction and two due to low-output state. Four deaths occurred in the group perfused via the *USCI International, Inc., Glens Falls, N. Y.
Body temperature 37 28 28 28 28
arch, all unrelated to the technique in use: Three patients died owing to low-output state (one aortic valve replacement, one aorta-coronary bypass, and one mitral valve with aorta-coronary bypass); one patient who underwent aortic valve replacement died after 30 days because of a stroke secondary to a calcium embolus. The EEG was normal during CPB in 76 of 94 patients (81 percent) undergoing femoral artery perfusion. Transient, slow wave activity occurred in the EEG's of 15 of 94 patients (16 percent) at the start of CPB. Major EEG abnormalities occurred in three of 84 patients (3 percent) and consisted of wave suppression owing to low systemic pressure in all three (Table III). The EEG returned to normal when the pump flow was increased. Of the 84 patients undergoing arch perfusion the EEG's were normal in 64 patients (76 percent): 29 normothermic and 35 hypothermic patients. The EEG's showed minor abnormalities during the first minutes of CPB in 14 patients (17 percent): six normothermic and eight hypothermic patients. Major EEG abnormalities occurred in six patients (7 percent): three normothermic and three hypothermic patients. The EEG's in five of six patients returned to normal after identification and correction of the problem (inadequate perfusion in two and malposition of the arch cannula in three). One patient had a cardiovascular accident prior to surgery which accounted for the abnormal EEG (Table IV). There was no significant difference between EEG abnormalities during arch and femoral perfusions (chi square analysis). Retrospective analysis. The records of 300 patients undergoing cardiopulmonary bypass procedures in which the EEG was monitored were reviewed over a period beginning in 1964. There were 220 male and 80 female patients ranging in age from 3 to 67 years (mean, 53 years). Procedures performed are shown in Table V. Femoral artery cannulation and hypothermia (30° C.) were used in all patients. Analysis of the EEG recordings revealed normal EEG's throughout CPB in 186
Volume 76 Number 5 November, 1978
Arch
Table I I I . EEG abnormalities femoral perfusion No. of patients
Normal Minor EEG abnormality Major EEG abnormality
76 15 3
81 16 3
Totals
94
100
Percent
artery perfusion
during
6 8 3
CPB
Table V. Procedures performed in 300 CPB operations with EEG monitoring
during
EEG
vs. femoral
Procedures
No. of patients
Coronary surgery Valve surgery Congenital heart surgery Left ventricular aneurysm resection Others
114 89 70 15 12
Total
300
Body temperature (°C.) 28 28 28
Table IV. EEG abnormality in arch perfusion EEG Normal
No. of patients 64
Percent 76
Minor EEG abnormality Major EEG abnormality
14
17
6
7
Totals
84
100
No. of patients/ body temperature 29Normothermic37°C. 35 Hypothermic 28° C. 6 Normothermic 8 Hypothermic 3 Normothermic 3 Hypothermic
patients (62 percent); minor EEG changes (slow wave activity and decreased voltage at the start of CPB) occurred temporarily in 73 patients (24 percent). In 41 patients (14 percent) major EEG changes occurred (wave suppression either localized to one hemisphere or diffuse). These changes were due to hypotension in 33 patients, cannula problems in five patients, hypoxia in one patient, airembolus in one patient, and unknown cause in one patient. To study the incidence of surgical complications caused by the perfusion techniques the records of 867 patients who have undergone CPB procedures since 1958 (702 femoral artery and 165 aortic arch cannulations) were reviewed. Five patients with minor bleeding at the site of aortic arch cannulation required a simple suture at time of operation. No other complication resulted from this modality of perfusion. The complications resulting from femoral artery cannulation are presented in Table VI. As noted, in 15 patients the femoral artery could not be used owing to severe arteriosclerosis. There was no retrograde aortic dissection and no loss of limbs. No death in the series could be attributed to femoral artery cannulation. Therefore, in both the prospective and the retrospective series there were more complications due to femoral artery perfusion than to aortic arch cannulation. Discussion Most cardiac surgeons favor cannulation of the aortic arch during routine CPB procedures because the
Table VI. Complications due to femoral artery cannulation in 702 patients Complications Iliac artery perforation Blocked femoral artery Asymptomatic Claudication Transient ischemia (spasm) Bleeding femoral artery requiring operation Difficult arterial repair post perfusion* Minor wound infection Totals
No. of patients 1
Percent 0.14
2 3 3 5
0.71
10
1.42
3
0.43
27
3.84
1.14
*One patient required interposition of a femoral artery graft.
technique is relatively simple and the complication rate is low. 4,8 ' 9 A groin incision is not required, damage to the femoral artery is eliminated, and atheromatous debris and thrombi cannot be dislodged from the iliac arteries.10 Cannulation of the arch, however, is not without hazard. Imperfect perfusion of the arch vessels may occur during this type of cannulation,11, 12 causing unexplained neurologic complications. Poor perfusion of the carotid arteries may occur despite high pump flow rates because of the Coanda effect.3' 13, 14 Without proper monitoring of cerebral perfusion the resultant neurologic deficits may wrongly be ascribed to other problems during perfusion.2,15 The EEG is a very sensitive, noninvasive means of determining the adequacy of cerebral blood flow.15,16 Changes in the EEG closely parallel the changes in cerebral perfusion.17, 18 The prospective analysis demonstrates that the incidence of major EEG abnormalities, i.e., changes indicating EEG wave suppression secondary to poor cerebral circulation, was not significantly different between arch and femoral perfusion. Most of these abnormalities were identified and corrected by increasing the
684
The Journal of Thoracic and Cardiovascular Surgery
Salerno et al.
pump flow rate in two patients with unsuspected carotid artery disease and by cannula repositioning in three patients perfused via the arch. In the three patients perfused via the femoral artery major EEG wave abnormalities were corrected by increasing the pump flow rate. Minor EEG abnormalities occurred equally in both groups of patients and reverted to normal immediately after full CPB was instituted without requiring manipulation by the surgeon. The retrospective analysis demonstrated that surgical complications occurred primarily in the group of patients perfused via the femoral artery. However, there was no retrograde dissection, limb loss, or death that could be attributed to this modality of perfusion. Because more patients have been perfused via the femoral artery and the experience with this technique is more extensive, follow-up of our patients perfused via the arch will be necessary in order to complete the information provided herein. Major EEG changes were more common in the retrospective series (14 percent) than in the prospective series (3 percent), probably reflecting low pump flows and higher incidence of hypotension and low-output state in our early experience. At present we continue to advocate the use of arch perfusion for most CPB procedures. We feel, however, that it is of great importance to monitor cerebral function when extracorporeal circulation is used. Clinical assessment during CPB may be misleading and cerebral perfusion may be abnormal because of unsuspected carotid disease, flow disturbance, or the Coanda effect occurring with arch perfusion. EEG monitoring allows greater safety for the patient undergoing open-heart surgery, irrespective of the type of perfusion. We would like to acknowledge the technical assistance of Mrs. B. Marsh and Miss N. Drew for EEG monitoring and Miss D. Judson for typing this manuscript.
REFERENCES 1 Taylor PC, Groves LK, Loop FD, Effler DB: Cannulation of the ascending aorta for cardiopulmonary bypass. J THORAC CARDIOVASC SURG 71:225, 1975
2 Krous HF, Mansfield PB, Sauvage LR: Carotid artery
hyperperfusion during open-heart surgery. J THORAC CARDIOVASC SURG 66:118, 1973
3 Magilligan DJ Jr, Eastland MW, Lell WA, DeWeese JA, Mahoney EB: Decreased carotid flow with ascending aortic cannulation. Circulation 45, 46:Suppl 1:130, 1972 4 Flick WF, Hallermann FJ, Feldt RH, Danielson GK: Aneurysm of aortic cannulation site. Successful repair by means of peripheral cannulation, profound hypothermia and circulatory arrest. J THORAC CARDIOVASC SURG
61:419, 1971 5 Benedict JS, Buhl TL, Henney RP: Acute aortic dissection during cardiopulmonary bypass. Arch Surg 108:810, 1974 6 Reinke T, Harris RD, Klein AJ, Daily PO: Aorto iliac dissection due to aortic cannulation. Ann Thorac Surg 18:295, 1974 7 Daily PO, Fogarty TJ, Shumway NE: Cannulation of the ascending aorta. Ann Thorac Surg 12:85, 1971 8 Roe BB, Kelly PB: Perfusion through the ascending aorta: Experience with 410 cases. Ann Thorac Surg 7:238,1969 9 Salama FD, Blesovsky A: Complications of cannulation of the ascending aorta for open heart surgery. Thorax 25:604, 1970 10 Kay JR, Dykstra PC, Tsuji HK: Retrograde ilioaortic dissection: A complication of common femoral artery perfusion during open heart surgery. Am J Surg 111:464, 1966 11 Magner JB: Complications of aortic cannulation for open-heart surgery. Thorax 26:172, 1971 12 Parker R: Aortic cannulation. Thorax 24:742, 1969 13 Angrist SW: Fluid control devices. Sci Am 211:80, 1964 14 Cutler SS, Burbank B: Fluid control devices, "rheostasis" and the human circulation. J THORAC CARDIOVASC SURG 56:558, 1968
15 Salerno TA, Lince DP, White DN, Lynn RB, Charrette EJP: E.E.G. monitoring during open-heart surgery: A prospective analysis of 118 cases. J THORAC CARDIOVASC SURG 76:97, 1978
16 Laxenaire M, Tridon P, Mathieu P, Picard L, Wever M: E.E.G. features during open heart surgery under extracorporeal circulation. Personal experience of 30 cases. Electroencephalogr Clin Neurophysiol 23:381, 1967 17 Sulg I, Ingvar DH: Regional cerebral blood flow and E.E.G. frequency content. Electroencephalogr Clin Neurophysiol 23:395, 1967 18 Freeman J, Invar DH: Influence of tissue hypoxia upon the E.E.G.—cerebral blood flow relationship. Electroencephalogr Clin Neurophysiol 23:395, 1967