- Email: [email protected]
Inaccuracy of radial artery pressure measurement after cardiac operations
J
THoRAc CARDIOVASC SURG
1987;94:286-90
Inaccuracy of radial artery pressure measurement after cardiac operations The phenomenon of a pressure gradient between central and radial arteries was evaluated in 48 patients immediately after coronary artery bypass operations. All were in stable hemodynamic condition, none receiving catecholamine support. In eight patients (Group A) mean femoral pressure was significantly higher than mean radial pressure (range 10 to 30 mm Hg), In the remaining 40 (Group B) radial and femoral pressures were equal. Mean cardiac index (thermodilution) was 3.3 ± 0.68 versus 2.1 ± 0.4 Ljminjm2, systemic vascular resistance 1,181 ± 218.4 versus 2,049 ± 501 dynes/sec/em:", toe temperature 23.8 0 ± 1.2 0 C versus 24.02 0 ± 0.9 0 C, core temperature 33.9 0 ± 0.5 0 C versus 34.1 0 ± 0.6 0 C, mixed venous oxygen saturation 78 % ± 3 % versus 62 % ± 5 %, and peak radial dP j dt 1,485 ± 366 versus 2,028 ± 392 in Groups A and B, respectively. These data indicate, first, that the low radial pressures measured in Group A patients did not represent the true central aortic pressures; that is, they were faIse. Second, these low pressures had nothing to do with compromised cardiac function; rather, they were due to peripheral constriction and volume factors andalso probably to proximal shunting. It is therefore recommended that while the chest is still open, if a discrepancy exists between a low radial artery pressure, a high palpable aortic pressure, and a satisfactory cardiac contraction, a femoral cannula for pressure measurement should be inserted. Treatment is by blood infusion until the femoral-radial gradient has been abolished.
Rephael Mohr, M.D., Jacob Lavee, M.D., and Daniel A. Goor, M.D.,
Tel Hashomer and Tel Aviv, Israel
Intra-arterial radial pressure is the most frequently chosen mode for systemic blood pressure monitoring after cardiac operations. Usually this measurement is considered a true representation of the central aortic blood pressure.l-? However, for some years we and others' have observed that occasionally there is a discrepancy between radial and central (aortic) pressure recordings. This discrepancy may be of critical importance. The purpose of this report is to describe the hemodynamic basis of the discrepancy and the therapeutic approach. Material and methods Forty-eight consecutive patients undergoing coronary artery bypass grafting were studied. Patients who From the Department of Cardiac Surgery, The Chaim Sheba Medical Center, Tel Hashomer, and the Sadder School of Medicine, Tel Aviv University, Tel Aviv, Israel. Received for publication July 15, 1986. Accepted for publication Aug. 22, 1986. Address for reprints: Daniel A. Goor, M.D., Department of Cardiac Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel.
286
required balloon counterpulsation or inotropic support of any kind at termination of cardiopulmonary bypass were excluded. The pertinent clinical, catheterization, and intraoperative data of all patients are summarized in Table I. All patients were operated on by the same surgeon (DA.G.) and the operative technique was identical in all.' The pump was primed with 2,000 ml of Hartman's solution. All urine output was replaced during pump time by crystalloid solution. In addition, each patient received between 500 and 1,000 mlof crystalloid solution during the period of rewarming. At the time of arterial decannulation the amount of blood in the oxygenator ranged between 500 and 1,000 ml. The blood was collected in transfusion bags and infused to the patients. Mean radial pressure, mean femoral pressure, and cardiac output were measured in all patients before operation, after termination of cardiopulmonary bypass, and upon arrival in the intensive care unit. The data for the present study are the preoperative data and those that were recorded in the intensive care unit. Mean radial and femoral pressures in each patient were recorded consecutively within a few seconds of
Volume 94 Number 2 August 1987
Radial artery pressure measurements 2 8 7
each other with the same pressure transducer (Gould P23 ID, Gould Inc. Cardiovascular Products, Oxnard, Calif.), connected to a blood pressure monitor (Mennen Medical 741, MG Electronics Ltd., Rehovot, Israel), givingan electrical digital reading of the mean pressure. The radial artery was cannulated percutaneously with a Teflon cannula 45 mm long and 1.1 mm in outer diameter (Medicut, Argyle, Sherwood Medical Industries, Crawley, Sussex, England). The femoral artery was cannulated percutaneously, with a 20 em long catheter (Bard-I-Cath, C. R. Bard International Ltd., Sunderland, England), which was introduced via a 17 gauge needle. Both the cannula and catheter were connected to the transducer by a 4 foot rigid pressure monitoring line (Cobe Laboratories, Inc., Lakewood, Colo.). In the last 20 patients femoral and radial pressures were also compared to aortic pressure. This was done just before closure of the sternum. Aortic pressure was measured through an 18 gauge needle. The same measurement system was used for any given patient in the operating room and in the intensive care unit. There were identical readouts of the femoral and aortic pressures. Hence in this report the terms femoral and aortic pressures are interchangeable. The midaxillary line was used as the zero reference point for the mean arterial pressure. The frequency response of the pressure monitoring system with the Steriflo II continuous flush device' (Cobe Laboratories, Inc., Lakewood, Colo.) was 23 Hz or more. The damping coefficients ranged between 0.12 and 0.2. The same values were obtained via the two types of cannulas (Medicut and Bard) that were used for the radial and femoral pressures. Distortions and attenuations (or socalled "damping") of the pulse wave shape during its propagationto the more peripherally located radial artery were evaluated by measuring the rate of pressure rise (peakdP j dt).This measurement was added here because it has been shown previously that attenuation in the amplitude and slope of the pressure wave in its travel to the arteriolar level occurs before reduction of the mean blood pressure," We therefore compared the femoral and radial peak dPjdt to quantitate arterial pulse wave distortion during its propagation. The frequency response of the differentiation was flat to 100Hz. Cardiac output was measured by thermodilution with a cardiac output computer (Model 601, Instrumentation Laboratories, Inc., Boston, Mass.). Cardiac index (CI) was calculated from the cardiac output (CO) and body surface area (BSA) by the formula: CI
= (CO/BSA)
Table I. Clinical. catheterization, and operative data
Male/female Age (yr) Stable/unstable angina Previous myocardial infarction Hypertension Diabetes Ejection fraction (%) LVEDP (mm Hg) No. of grafts per patient Aortic cross-clamp time (min) Bypass time (min)
Group B (n = 40)
Group A (n = 8)
6/34
7/1 53.8 ± 9 3/5
54.2 ± 8
7/33 28
4
18 5 62.8 ± 12.3 15.1 ± 6.4 3.22
2 2 54.0 ± 10 19.5 ± 13 3.8
41.7 ± 14.5
48.8 ± 5.7
104.3 ± 26
108.0 ± 25
Legend: LVEDP, Left ventricular end-diastolic pressure.
Systemic vascular resistance (SVR) was calculated by the formula SVR = (MFP - MRAP)/CO
where MFP is mean femoral pressure and MRAP is mean right atrial pressure. All results were compared by the Student's t test. Results The two groups were comparable regarding preoperative and operative parameters (Table I). Preoperatively, mean femoral and mean radial pressures were equal in all patients (Table II, Fig. 1). In eight patients (Group A, Fig. 1) there was a femoral-radial mean pressure gradient of 10 mm Hg or more. Mean femoral pressure was 80.2 ± 19.34, and mean radial pressure was 67.2 ± 20.28 mm Hg (p < 0.001). Mean blood pressure differences of up to 30 mm Hg were detected, and the difference in systolic pressure reached 60 mm Hg (Fig. 2). Cardiac index was 3.3 Ljminjm 2, SVR was 1,181 dynesjsecjcm- S (Table II, Fig. 3), and mixed oxygen saturation was 78% ± 3%. A significant difference was observed between toe (23.8° ± 1.20 C) and core (33.8° ± 0.6° C) temperatures. This was indicative of peripheral vasoconstriction in the extremities. In addition, there was a significant postoperative decrease in the radial peak dP j dt (1,485 ± 366) compared to the femoral peak dPjdt (1,995 ± 398) (Table II). In the remaining 40 patients (Group B) there was no significant postoperative femoral-radial gradient (Table 11). Average SVR was higher in Group A than in group B
288
The Journal of Thoracic and Cardiovascular
Mohr, Lavee, Goor
Surgery
PRE OP
POST OP
mean pressure
mm Hg
100 90
80 7
60 50
40 30 20 10
A
B
B
A
Fig. 1. Mean radial pressure (MRP) and meal femoral pressure (MFP) before and after cardiac procedures. p < 0.001 (paired t test).
Table
n. Hemodynamic parameters of the two groups before and after cardiac operations Group B (MFP= MRP)
Group A (MFP>MRPj Preoperative
MFP (mm Hg) MRP (mm Hg) Mixed venous 0, sat. (%) Core temp. (0C) Toe temp. (DC) SVR (dynes/sec/cm") Peak dPjdt Radial Femoral Cardiac index (Ljminjm')
I
88.6 ± 17.6 85.3 ± 13.5
Postoperative
1,062.2 ± 396.5
80.0 ± 19.34 67.2 ± 20.28t 78 ± 3 33.9 ± 0.5 23.8 ± 1.2 1,181.8 ± 218.4
2,040 ± 471 1,966 ± 503 3.2 ± 0.13
1,485 ± 366 1,995 ± 398t 3.3 ± 0.68
Preoperative 91.2 ± 11.88 90.8 ± 12.53
1,264 ± 258 1,955 1,765 3.1
::!: 443 ::!: 463 ::!: 0.43
i
Postoperative 93.7 :!: 13.62 92.9 :!: 12.9 62 :!: 5 34.1 :!: 0.6 24.0:!: 0.9 2,049 :!: 551* 2,028 1,836 2.1
::!: 392* ::!: ::!:
529 0.4*
Legend: MFP, Mean femoral pressure. MRP, Mean radial pressure. SVR, Systemic vascular resistance. dP/dt, Rate of rise of left ventricular pressure.
*p < 0.001, A versus B. tp
< 0.001,
radial versus femoral.
patients (2,028 ± 392 versus 1,181.8 ± 218.4), cardiac index was lower (2.1 ± 0.4 versus 3,3 ± 0,68), and mixed venous oxygen saturation was lower (62% ± 5% versus 78% ± 3%). Core-to-toe temperature gradient was 34.1 0 ± 0.6 0 C to 24.0 0 ± 0.9 0 C in Group B, and it was not significantly different from that of Group A. Peak dP /dt in the radial artery in Group B was higher than in Group A (2,028 ± 329 versus 1,485 ± 366) and peak dP /dt in the femoral artery was equal in the two groups. After volume replacement the femoral-radial gradient in Group A disappeared (Table III), and this was associated with a decrease in cardiac index (Table III)
(from 3.3 ± 0.68 to 2.6 ± 0.51), an increase in SVR (from 1,181 ± 8.218 to 1,660 ± 172), and a decrease in mixed venous blood oxygen saturation (from 78% ± 3% to 64% ± 3%). Discussion
An observation made by others' is confirmed here, namely, that a significant number (15% in the present study) of patients have a significant pressure gradient between the femoral (or aortic) and radial arteries after coronary operations. A systolic pressure gradient as high as 60 mm Hg may be observed. To exclude artifacts of the pressure readouts in the present study, a simple
Volume 94 Number 2 August 1987
Radial artery pressure measurements
2 89
Table m. Effect of volume replacement on patients in Group A Before volume replacement
CVP (mm Hg) MFP (mm Hg) MRP (mm Hg)
CI (L/min/m') SVR (dynes/sec/em:")
6.9 80.0 67.2 3.3 1,181.8
± 4.3 ± 19.34 ± 20.28 ± 0.68 ± 218.4
After volume replacement 8.2 90.1 91.2 2.6 1,660.7
± 1.8 ± 15.9 ± 13 ± 0.51 ± 172
Legend: CVP, Central venous pressure. MFP, Mean femoral pressure. MRP, Mean radial pressure. CI, Cardiac index. SVR, Systemic vascular resistance.
system with adequate dynamic response (natural frequency >23 Hz and damping coefficient 0,12 to 0.2) was set up. In addition, the same monitor, transducer, and connecting tubings were used for both radial and femoral pressure measurements. It is also shown here that the frequency response and damping coefficient (so-called "dynamic response'") were riot affected by using the two different cannulas (Medicut and Bard) that were inserted for the radial and femoral pressure measurements. The high fidelity of the measurement system is of importance for the validity of the results when peak systolic pressures' or the arterial dP/dt are monitored, Therefore, it may be stated that the dP/dt measurements, which are more sensitive to changes than mean pressure," were affected by the true hemodynamic events, and not by artifacts. However, the present study is primarily based on the mean blood pressures picked at different locations. Ordinarily, the mean blood pressure is a low-frequency phenomenon and is therefore not distorted by artifacts of the monitoring system.' Consequently, the attenuation of the radial wave form, as was evident from the reduced mean pressure and dP / dt, is not due to artifact but to true distortion of the pulse wave during its propagation to the periphery. Several vascular mechanisms described by O'Rourke" for the changes in arterial pulse wave shape may explain the radial-femoral pressure differences after cardiopulmonary bypass. First, there is damping or attenuation of the pulse wave in its travel to the periphery of the arterial tree. Second, there is reduction or amplification of components of the pulse wave by reflection waves." Third, there is a progressive stiffness of peripheral arteries, which may be related to the cold temperature." Stem and associates' showed that both SVR and forearm resistance ate low in patients with aortic-radial pressure differences. They concluded that the lower radial pressure at the wrist results from diversion of flow
200 150 100 50 0 ..........- - - - - - - - - - - ' - - - - - - - -
200 150 100
50
O..J------------------
Fig. 2. A, One hour after termination of cardiopulmonary bypass. Radial pressure (left side): 92/48 (mean 58 mrn Hg). Femoral pressure (right side): 156/62 (mean 88 mrn Hg). E, Twelve hours after cardiopulmonary bypass. Radial pressure (left side): 144/70 (mean 94 mrn Hg). Femoral pressure (right side): 137/69 (mean 99 mrn Hg).
to the more proximally located vasodilated vascular bed of the forearm. The present study endorses the observation of Stem and colleagues' that the SVR is low in patients with a femoral-radial gradient. This work also supports their thesis that proximal arterial dilatation accounts for the attenuation of the radial pulse wave whenever there is a femoral-radial gradient. Further support for the early postoperative proximal shunting is obtained in the present study from the high oxygen saturation of the mixed venous blood in Group A patients (78% ± 3%). This suspected proximal shunting, if present, may be at the splanchnic level, but this theory remains to be proved. Analysis of the hemodynamic data of Group A patients reveals that in all of them the constricted arterial tree in the extremities was not a response to a low output syndrome or to any other shock mechanism. On the contrary, cardiac output was higher and the calculated total body SVR was lower than in the Group B patients (Table II). In Group A, when volume load abolished the femoral-
290
The Journal of Thoracic and Cardiovascular Surgery
Mohr, Lavee, Goor
A.
B.
CI
SVR
lit/min/m 2
dyn/sec/cm- 5
4
2500 2000
3
1500
2 1000 500
PRE OP
POST OP
PRE
OP
POST OP
Fig. 3. A, Effect of cardiac procedures on cardiac index (Cl]. B, Effect of cardiac procedures on systemic vascular resistance (SVR).
radial pressure gradient, the cardiac output declined (Table III), SVR increased (Table III), and hemodynamics became similar to that of Group B. Group B patients, on the other hand, exhibited the typical reported postoperative hemodynamic behavior'? "; that is, cardiac index immediately postoperatively is lower and SVR is higher than the preoperative values. Postoperative core (~34° C) and toe temperatures (~24° C) were equal in Groups A and B, which indicates similar constriction, and probably resistance, of the vascular bed in the extremities. A possible explanation for the great discrepancy between the cold constricted extremities and the normal calculated SVR 0,181.8 ± 218.4) in Group A may be the presence of a central shunt. In addition, hypovolemia, cold temperature, and central shunting can explain inadequate filling of the low peripheral arterial tree and the observed low radial pressures. Not all patients who have a femoral-radial gradient belong to the type of Group A patient of the present series. A femoral-radial gradient may also be present in patients with poor cardiac function and low central arterial pressure. Patients with low femoral arterial pressure required catecholamine support and were not included in the study. In conclusion, the radial artery pressure measurement is frequently misleading. In every situation in which, on being weaned from cardiopulmonary bypass, the heart is beating well, the palpable aortic pressure is good, but the radial pressure is unexpectedly low, a femoral line should be inserted. Correction of the femoral-radial gradient will be achieved by volume load in most instances.
REFERENCES 1. Kroeker EJ, Wood EH. Comparison of simultaneously recorded central and peripheral arterial pressure pulses during rest, exercise and tilted position in man. Circ Res 1955;3:623-32. 2. Pascarelli EF, Bertrand CA. Comparison of blood pressures in the arm and legs. N Engl J Med 1964;270:6938. 3. Stern DH, Gerson JI, Allen FB, Parker FB. Can we trust the direct radial artery pressure immediately following cardiopulmonary bypass? Anesthesiology 1985;62:55761. 4. Goor DA, Lavee J, Mohr R. Enhanced protection of myocardial function by systemic deep hypothermia during cardioplegic arrest in multiple coronary bypass grafting. J THORAC CARDIOVASC SURG 1982;84:237-42. 5. Gardner RM. Direct blood pressure 'measurement: dynamic response requirements. Anesthesiology 1981; 54:227-36. 6. O'Rourke MF. Pressure and flow waves in systemic arteries and the anatomical design of the arterial system. J Appl Physiol 1967;23:139-49. 7. Bruner JMR: Handbook of blood pressure monitoring. Littleton: PSG Publishing, 1978:63-7. 8. O'Rourke MF. Arterial function in health and disease. New York: Churchill Livingstone, 1982:134-8. 9. Hamilton WF, Dow P. An experimental study of the standing waves in the pulse propagated through the aorta. Am J Physiol 1939;125:48-59. 10. Roberts AJ, Spies SM, Sanders JH, et al. Serial assessment of left ventricular performance following coronary artery bypass grafting. J THORAC CARDIOVASC SURG 1981;81:69-84. 11. Czer L, Hamer A, Murphy F, et al. Transient hemodynamic dysfunction after myocardial revascularization. J THORAC CARDIOVASC SURG 1983;86:226-34.
THoRAc CARDIOVASC SURG
1987;94:286-90
Inaccuracy of radial artery pressure measurement after cardiac operations The phenomenon of a pressure gradient between central and radial arteries was evaluated in 48 patients immediately after coronary artery bypass operations. All were in stable hemodynamic condition, none receiving catecholamine support. In eight patients (Group A) mean femoral pressure was significantly higher than mean radial pressure (range 10 to 30 mm Hg), In the remaining 40 (Group B) radial and femoral pressures were equal. Mean cardiac index (thermodilution) was 3.3 ± 0.68 versus 2.1 ± 0.4 Ljminjm2, systemic vascular resistance 1,181 ± 218.4 versus 2,049 ± 501 dynes/sec/em:", toe temperature 23.8 0 ± 1.2 0 C versus 24.02 0 ± 0.9 0 C, core temperature 33.9 0 ± 0.5 0 C versus 34.1 0 ± 0.6 0 C, mixed venous oxygen saturation 78 % ± 3 % versus 62 % ± 5 %, and peak radial dP j dt 1,485 ± 366 versus 2,028 ± 392 in Groups A and B, respectively. These data indicate, first, that the low radial pressures measured in Group A patients did not represent the true central aortic pressures; that is, they were faIse. Second, these low pressures had nothing to do with compromised cardiac function; rather, they were due to peripheral constriction and volume factors andalso probably to proximal shunting. It is therefore recommended that while the chest is still open, if a discrepancy exists between a low radial artery pressure, a high palpable aortic pressure, and a satisfactory cardiac contraction, a femoral cannula for pressure measurement should be inserted. Treatment is by blood infusion until the femoral-radial gradient has been abolished.
Rephael Mohr, M.D., Jacob Lavee, M.D., and Daniel A. Goor, M.D.,
Tel Hashomer and Tel Aviv, Israel
Intra-arterial radial pressure is the most frequently chosen mode for systemic blood pressure monitoring after cardiac operations. Usually this measurement is considered a true representation of the central aortic blood pressure.l-? However, for some years we and others' have observed that occasionally there is a discrepancy between radial and central (aortic) pressure recordings. This discrepancy may be of critical importance. The purpose of this report is to describe the hemodynamic basis of the discrepancy and the therapeutic approach. Material and methods Forty-eight consecutive patients undergoing coronary artery bypass grafting were studied. Patients who From the Department of Cardiac Surgery, The Chaim Sheba Medical Center, Tel Hashomer, and the Sadder School of Medicine, Tel Aviv University, Tel Aviv, Israel. Received for publication July 15, 1986. Accepted for publication Aug. 22, 1986. Address for reprints: Daniel A. Goor, M.D., Department of Cardiac Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel.
286
required balloon counterpulsation or inotropic support of any kind at termination of cardiopulmonary bypass were excluded. The pertinent clinical, catheterization, and intraoperative data of all patients are summarized in Table I. All patients were operated on by the same surgeon (DA.G.) and the operative technique was identical in all.' The pump was primed with 2,000 ml of Hartman's solution. All urine output was replaced during pump time by crystalloid solution. In addition, each patient received between 500 and 1,000 mlof crystalloid solution during the period of rewarming. At the time of arterial decannulation the amount of blood in the oxygenator ranged between 500 and 1,000 ml. The blood was collected in transfusion bags and infused to the patients. Mean radial pressure, mean femoral pressure, and cardiac output were measured in all patients before operation, after termination of cardiopulmonary bypass, and upon arrival in the intensive care unit. The data for the present study are the preoperative data and those that were recorded in the intensive care unit. Mean radial and femoral pressures in each patient were recorded consecutively within a few seconds of
Volume 94 Number 2 August 1987
Radial artery pressure measurements 2 8 7
each other with the same pressure transducer (Gould P23 ID, Gould Inc. Cardiovascular Products, Oxnard, Calif.), connected to a blood pressure monitor (Mennen Medical 741, MG Electronics Ltd., Rehovot, Israel), givingan electrical digital reading of the mean pressure. The radial artery was cannulated percutaneously with a Teflon cannula 45 mm long and 1.1 mm in outer diameter (Medicut, Argyle, Sherwood Medical Industries, Crawley, Sussex, England). The femoral artery was cannulated percutaneously, with a 20 em long catheter (Bard-I-Cath, C. R. Bard International Ltd., Sunderland, England), which was introduced via a 17 gauge needle. Both the cannula and catheter were connected to the transducer by a 4 foot rigid pressure monitoring line (Cobe Laboratories, Inc., Lakewood, Colo.). In the last 20 patients femoral and radial pressures were also compared to aortic pressure. This was done just before closure of the sternum. Aortic pressure was measured through an 18 gauge needle. The same measurement system was used for any given patient in the operating room and in the intensive care unit. There were identical readouts of the femoral and aortic pressures. Hence in this report the terms femoral and aortic pressures are interchangeable. The midaxillary line was used as the zero reference point for the mean arterial pressure. The frequency response of the pressure monitoring system with the Steriflo II continuous flush device' (Cobe Laboratories, Inc., Lakewood, Colo.) was 23 Hz or more. The damping coefficients ranged between 0.12 and 0.2. The same values were obtained via the two types of cannulas (Medicut and Bard) that were used for the radial and femoral pressures. Distortions and attenuations (or socalled "damping") of the pulse wave shape during its propagationto the more peripherally located radial artery were evaluated by measuring the rate of pressure rise (peakdP j dt).This measurement was added here because it has been shown previously that attenuation in the amplitude and slope of the pressure wave in its travel to the arteriolar level occurs before reduction of the mean blood pressure," We therefore compared the femoral and radial peak dPjdt to quantitate arterial pulse wave distortion during its propagation. The frequency response of the differentiation was flat to 100Hz. Cardiac output was measured by thermodilution with a cardiac output computer (Model 601, Instrumentation Laboratories, Inc., Boston, Mass.). Cardiac index (CI) was calculated from the cardiac output (CO) and body surface area (BSA) by the formula: CI
= (CO/BSA)
Table I. Clinical. catheterization, and operative data
Male/female Age (yr) Stable/unstable angina Previous myocardial infarction Hypertension Diabetes Ejection fraction (%) LVEDP (mm Hg) No. of grafts per patient Aortic cross-clamp time (min) Bypass time (min)
Group B (n = 40)
Group A (n = 8)
6/34
7/1 53.8 ± 9 3/5
54.2 ± 8
7/33 28
4
18 5 62.8 ± 12.3 15.1 ± 6.4 3.22
2 2 54.0 ± 10 19.5 ± 13 3.8
41.7 ± 14.5
48.8 ± 5.7
104.3 ± 26
108.0 ± 25
Legend: LVEDP, Left ventricular end-diastolic pressure.
Systemic vascular resistance (SVR) was calculated by the formula SVR = (MFP - MRAP)/CO
where MFP is mean femoral pressure and MRAP is mean right atrial pressure. All results were compared by the Student's t test. Results The two groups were comparable regarding preoperative and operative parameters (Table I). Preoperatively, mean femoral and mean radial pressures were equal in all patients (Table II, Fig. 1). In eight patients (Group A, Fig. 1) there was a femoral-radial mean pressure gradient of 10 mm Hg or more. Mean femoral pressure was 80.2 ± 19.34, and mean radial pressure was 67.2 ± 20.28 mm Hg (p < 0.001). Mean blood pressure differences of up to 30 mm Hg were detected, and the difference in systolic pressure reached 60 mm Hg (Fig. 2). Cardiac index was 3.3 Ljminjm 2, SVR was 1,181 dynesjsecjcm- S (Table II, Fig. 3), and mixed oxygen saturation was 78% ± 3%. A significant difference was observed between toe (23.8° ± 1.20 C) and core (33.8° ± 0.6° C) temperatures. This was indicative of peripheral vasoconstriction in the extremities. In addition, there was a significant postoperative decrease in the radial peak dP j dt (1,485 ± 366) compared to the femoral peak dPjdt (1,995 ± 398) (Table II). In the remaining 40 patients (Group B) there was no significant postoperative femoral-radial gradient (Table 11). Average SVR was higher in Group A than in group B
288
The Journal of Thoracic and Cardiovascular
Mohr, Lavee, Goor
Surgery
PRE OP
POST OP
mean pressure
mm Hg
100 90
80 7
60 50
40 30 20 10
A
B
B
A
Fig. 1. Mean radial pressure (MRP) and meal femoral pressure (MFP) before and after cardiac procedures. p < 0.001 (paired t test).
Table
n. Hemodynamic parameters of the two groups before and after cardiac operations Group B (MFP= MRP)
Group A (MFP>MRPj Preoperative
MFP (mm Hg) MRP (mm Hg) Mixed venous 0, sat. (%) Core temp. (0C) Toe temp. (DC) SVR (dynes/sec/cm") Peak dPjdt Radial Femoral Cardiac index (Ljminjm')
I
88.6 ± 17.6 85.3 ± 13.5
Postoperative
1,062.2 ± 396.5
80.0 ± 19.34 67.2 ± 20.28t 78 ± 3 33.9 ± 0.5 23.8 ± 1.2 1,181.8 ± 218.4
2,040 ± 471 1,966 ± 503 3.2 ± 0.13
1,485 ± 366 1,995 ± 398t 3.3 ± 0.68
Preoperative 91.2 ± 11.88 90.8 ± 12.53
1,264 ± 258 1,955 1,765 3.1
::!: 443 ::!: 463 ::!: 0.43
i
Postoperative 93.7 :!: 13.62 92.9 :!: 12.9 62 :!: 5 34.1 :!: 0.6 24.0:!: 0.9 2,049 :!: 551* 2,028 1,836 2.1
::!: 392* ::!: ::!:
529 0.4*
Legend: MFP, Mean femoral pressure. MRP, Mean radial pressure. SVR, Systemic vascular resistance. dP/dt, Rate of rise of left ventricular pressure.
*p < 0.001, A versus B. tp
< 0.001,
radial versus femoral.
patients (2,028 ± 392 versus 1,181.8 ± 218.4), cardiac index was lower (2.1 ± 0.4 versus 3,3 ± 0,68), and mixed venous oxygen saturation was lower (62% ± 5% versus 78% ± 3%). Core-to-toe temperature gradient was 34.1 0 ± 0.6 0 C to 24.0 0 ± 0.9 0 C in Group B, and it was not significantly different from that of Group A. Peak dP /dt in the radial artery in Group B was higher than in Group A (2,028 ± 329 versus 1,485 ± 366) and peak dP /dt in the femoral artery was equal in the two groups. After volume replacement the femoral-radial gradient in Group A disappeared (Table III), and this was associated with a decrease in cardiac index (Table III)
(from 3.3 ± 0.68 to 2.6 ± 0.51), an increase in SVR (from 1,181 ± 8.218 to 1,660 ± 172), and a decrease in mixed venous blood oxygen saturation (from 78% ± 3% to 64% ± 3%). Discussion
An observation made by others' is confirmed here, namely, that a significant number (15% in the present study) of patients have a significant pressure gradient between the femoral (or aortic) and radial arteries after coronary operations. A systolic pressure gradient as high as 60 mm Hg may be observed. To exclude artifacts of the pressure readouts in the present study, a simple
Volume 94 Number 2 August 1987
Radial artery pressure measurements
2 89
Table m. Effect of volume replacement on patients in Group A Before volume replacement
CVP (mm Hg) MFP (mm Hg) MRP (mm Hg)
CI (L/min/m') SVR (dynes/sec/em:")
6.9 80.0 67.2 3.3 1,181.8
± 4.3 ± 19.34 ± 20.28 ± 0.68 ± 218.4
After volume replacement 8.2 90.1 91.2 2.6 1,660.7
± 1.8 ± 15.9 ± 13 ± 0.51 ± 172
Legend: CVP, Central venous pressure. MFP, Mean femoral pressure. MRP, Mean radial pressure. CI, Cardiac index. SVR, Systemic vascular resistance.
system with adequate dynamic response (natural frequency >23 Hz and damping coefficient 0,12 to 0.2) was set up. In addition, the same monitor, transducer, and connecting tubings were used for both radial and femoral pressure measurements. It is also shown here that the frequency response and damping coefficient (so-called "dynamic response'") were riot affected by using the two different cannulas (Medicut and Bard) that were inserted for the radial and femoral pressure measurements. The high fidelity of the measurement system is of importance for the validity of the results when peak systolic pressures' or the arterial dP/dt are monitored, Therefore, it may be stated that the dP/dt measurements, which are more sensitive to changes than mean pressure," were affected by the true hemodynamic events, and not by artifacts. However, the present study is primarily based on the mean blood pressures picked at different locations. Ordinarily, the mean blood pressure is a low-frequency phenomenon and is therefore not distorted by artifacts of the monitoring system.' Consequently, the attenuation of the radial wave form, as was evident from the reduced mean pressure and dP / dt, is not due to artifact but to true distortion of the pulse wave during its propagation to the periphery. Several vascular mechanisms described by O'Rourke" for the changes in arterial pulse wave shape may explain the radial-femoral pressure differences after cardiopulmonary bypass. First, there is damping or attenuation of the pulse wave in its travel to the periphery of the arterial tree. Second, there is reduction or amplification of components of the pulse wave by reflection waves." Third, there is a progressive stiffness of peripheral arteries, which may be related to the cold temperature." Stem and associates' showed that both SVR and forearm resistance ate low in patients with aortic-radial pressure differences. They concluded that the lower radial pressure at the wrist results from diversion of flow
200 150 100 50 0 ..........- - - - - - - - - - - ' - - - - - - - -
200 150 100
50
O..J------------------
Fig. 2. A, One hour after termination of cardiopulmonary bypass. Radial pressure (left side): 92/48 (mean 58 mrn Hg). Femoral pressure (right side): 156/62 (mean 88 mrn Hg). E, Twelve hours after cardiopulmonary bypass. Radial pressure (left side): 144/70 (mean 94 mrn Hg). Femoral pressure (right side): 137/69 (mean 99 mrn Hg).
to the more proximally located vasodilated vascular bed of the forearm. The present study endorses the observation of Stem and colleagues' that the SVR is low in patients with a femoral-radial gradient. This work also supports their thesis that proximal arterial dilatation accounts for the attenuation of the radial pulse wave whenever there is a femoral-radial gradient. Further support for the early postoperative proximal shunting is obtained in the present study from the high oxygen saturation of the mixed venous blood in Group A patients (78% ± 3%). This suspected proximal shunting, if present, may be at the splanchnic level, but this theory remains to be proved. Analysis of the hemodynamic data of Group A patients reveals that in all of them the constricted arterial tree in the extremities was not a response to a low output syndrome or to any other shock mechanism. On the contrary, cardiac output was higher and the calculated total body SVR was lower than in the Group B patients (Table II). In Group A, when volume load abolished the femoral-
290
The Journal of Thoracic and Cardiovascular Surgery
Mohr, Lavee, Goor
A.
B.
CI
SVR
lit/min/m 2
dyn/sec/cm- 5
4
2500 2000
3
1500
2 1000 500
PRE OP
POST OP
PRE
OP
POST OP
Fig. 3. A, Effect of cardiac procedures on cardiac index (Cl]. B, Effect of cardiac procedures on systemic vascular resistance (SVR).
radial pressure gradient, the cardiac output declined (Table III), SVR increased (Table III), and hemodynamics became similar to that of Group B. Group B patients, on the other hand, exhibited the typical reported postoperative hemodynamic behavior'? "; that is, cardiac index immediately postoperatively is lower and SVR is higher than the preoperative values. Postoperative core (~34° C) and toe temperatures (~24° C) were equal in Groups A and B, which indicates similar constriction, and probably resistance, of the vascular bed in the extremities. A possible explanation for the great discrepancy between the cold constricted extremities and the normal calculated SVR 0,181.8 ± 218.4) in Group A may be the presence of a central shunt. In addition, hypovolemia, cold temperature, and central shunting can explain inadequate filling of the low peripheral arterial tree and the observed low radial pressures. Not all patients who have a femoral-radial gradient belong to the type of Group A patient of the present series. A femoral-radial gradient may also be present in patients with poor cardiac function and low central arterial pressure. Patients with low femoral arterial pressure required catecholamine support and were not included in the study. In conclusion, the radial artery pressure measurement is frequently misleading. In every situation in which, on being weaned from cardiopulmonary bypass, the heart is beating well, the palpable aortic pressure is good, but the radial pressure is unexpectedly low, a femoral line should be inserted. Correction of the femoral-radial gradient will be achieved by volume load in most instances.
REFERENCES 1. Kroeker EJ, Wood EH. Comparison of simultaneously recorded central and peripheral arterial pressure pulses during rest, exercise and tilted position in man. Circ Res 1955;3:623-32. 2. Pascarelli EF, Bertrand CA. Comparison of blood pressures in the arm and legs. N Engl J Med 1964;270:6938. 3. Stern DH, Gerson JI, Allen FB, Parker FB. Can we trust the direct radial artery pressure immediately following cardiopulmonary bypass? Anesthesiology 1985;62:55761. 4. Goor DA, Lavee J, Mohr R. Enhanced protection of myocardial function by systemic deep hypothermia during cardioplegic arrest in multiple coronary bypass grafting. J THORAC CARDIOVASC SURG 1982;84:237-42. 5. Gardner RM. Direct blood pressure 'measurement: dynamic response requirements. Anesthesiology 1981; 54:227-36. 6. O'Rourke MF. Pressure and flow waves in systemic arteries and the anatomical design of the arterial system. J Appl Physiol 1967;23:139-49. 7. Bruner JMR: Handbook of blood pressure monitoring. Littleton: PSG Publishing, 1978:63-7. 8. O'Rourke MF. Arterial function in health and disease. New York: Churchill Livingstone, 1982:134-8. 9. Hamilton WF, Dow P. An experimental study of the standing waves in the pulse propagated through the aorta. Am J Physiol 1939;125:48-59. 10. Roberts AJ, Spies SM, Sanders JH, et al. Serial assessment of left ventricular performance following coronary artery bypass grafting. J THORAC CARDIOVASC SURG 1981;81:69-84. 11. Czer L, Hamer A, Murphy F, et al. Transient hemodynamic dysfunction after myocardial revascularization. J THORAC CARDIOVASC SURG 1983;86:226-34.