Improvement in Blood Flow Through a Critical Arterial Stenosis by Defibrination With Ancrod William W.
Barrie, MB, ChB, FRCS, Worthington G. Schenk, Jr, MD
\s=b\ Using electromagnetic flow probes, cardiac output and hind limb blood flow were measured in dogs in which one hind limb had been rendered ischemic. Four dogs served as controls; seven were defibrinated by intravenous infusion of ancrod, 1 unit/kg, over a 30-minute period. In both groups, hematocrit readings remained constant, but cardiac output fell (this was attributed to barbiturate anesthesia), as did flow in the normal hind limb. In the controls after three hours, flow in the ischemic hind limb had decreased by 34%, but in the treated animals it had increased by 20%. The difference was statistically significant (P < .001). The selective increase in blood flow in the ischemic limb may be explained by the greater reduction in blood viscosity at low shear rates achieved by defibrination.
(Arch Surg 111:561-563, 1976) reduced blood flow is a factor common to a variety of clinical conditions including obstructive arterial disease, venous thrombosis, and the poor periph¬ eral perfusion associated with shock. Treatment is usually aimed at improving blood flow by lowering the vascular resistance, as in the reconstruction of an arterial stenosis, or improving the perfusion pressure by the infusion of plasma volume expanders. However, since the flow rate of a fluid is inversely related to its viscosity, it should also be possible to improve blood flow by lowering the viscosity of
Greatly
for publication Dec 4, 1975. From the Department of Surgery, State University of New York at Buffalo, and the Surgical Research Laboratories, E. J. Meyer Memorial Hospital, Buffalo, NY. Reprint requests to State University of New York, School of Medicine, 462 Grider St, Buffalo, NY 14215 (Dr Schenk).
Accepted
whole blood or plasma. Originally, it was thought that this was the mode of action of low molecular weight dextran, but subsequent studies have shown that the beneficial effect of this agent is more related to its action as a plasma volume expander.1 Since the contribution of fibrinogen to normal blood viscosity is second in importance only to the concentration of red blood cells (RBCs),'- the introduction of a drug, ancrod, which specifically reduces the plasma fibrinogen level, again raises the possibility of therapeutically altering the flow properties of blood. Ancrod, a purified enzymatic derivative of the venom of the Malayan pit viper, breaks down fibrinogen to degrada¬ tion products that are removed from the circulation by the reticuloendothelial system; the resultant effect is virtually complete defibrination.' Unlike the syndrome of dissemi¬ nated intravascular coagulation, other clotting factors are not affected, although the split products formed by the degradation of fibrinogen may exert a transient anticoag¬ ulant and antiplatelet action46. Early clinical reports suggest that therapeutic reduction of plasma fibrinogen may be of value in the treatment of peripheral arterial disease7 and in the prophylaxis of postoperative thromboembolism,8 but a direct improve¬ ment of in vivo blood flow has not previously been demon¬ strated. The purpose of this study was to measure blood flow through a critically ischemie limb during and after defibrination with ancrod. MATERIALS AND METHODS Eleven mongrel dogs, each weighing 18 to 22 kg, were anesthe¬ tized with pentobarbital sodium, 30 mg/kg, and ventilated with air
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
Treated Control Treated
Control
c
E
-
2
1
10-
D
O >
U -40 30
60
120
90
Minutes
,
150
180
Time after infusion
1—Mean cardiac output (± 2 SE) for control and treated animals. Fall in cardiac output is greater, but not significantly so, in treated group.
Fig
via a mechanical respirator. Thoracotomy was performed through the left fifth interspace and a noncannulating electromagnetic flow probe placed around the ascending aorta. A lower midline abdominal incision was made and the aortic trifurcation exposed. The incision was extended into the left groin to allow exposure of the left iliac and femoral arteries. A polyethylene catheter (internal diameter, 0.076 cm) was inserted into the abdominal aorta via the median sacral artery and the common internal iliac artery was ligated. The catheter was connected to an inductancetype pressure transducer to allow continuous monitoring of aortic pressure. Electromagnetic flow probes were placed around both iliac arteries just distal to the aortic trifurcation. On the left side, all the iliac and femoral arterial branches (including the profunda femoris artery) between the probe and the superficial femoral artery were ligated. The right iliofemoral arterial supply was not disturbed. After an interval to allow stabilization of the prepara¬ tion, the values of cardiac output, arterial pressure, and each iliac arterial flow were recorded. A "critical" arterial stenosis was produced by tying a ligature around the left superficial femoral artery and reducing the flow to the minimum that could be obtained without occluding the vessel completely. The preparation was then allowed to stabilize for approximately 45 minutes. Animals comprising the treated group, a total of seven, were infused intravenously with ancrod, 1 unit/kg in 50 ml of normal saline solution for a period of 30 minutes. Four control animals were given a similar volume of normal saline solution only. The hemodynamic variables of all the animals were recorded on a direct writing recorder for the remaining three hours of the experiment. Zero flow references were obtained periodically for the iliac probes by occlusion of the arteries just distal to the probes. On the left side, the absence of collateral supply was
Time minutes Fig 2.—Mean flow (± 2 SE) in ischemie limbs is shown as percentage change of pretreatment value. After 60 minutes, difference between groups is statistically significant (P < .001). -
confirmed by checking that zero flow was obtained by occlusion of the superficial femoral artery immediately proximal to the steno¬ sis. The diastolic quiet interval was taken as zero flow for the ascending aortic probe.
RESULTS The Table shows the mean changes, with standard deviations, in cardiac output and iliac arterial flow in the
four control and seven treated animals. In both groups, cardiac output fell during the experiment; the decrease was greater, but not significantly so, in the treated group (Fig
1).
Although there was considerable individual variation, in both treated and untreated animals the mean iliac arterial flow in the nonischemic limb decreased by a greater amount than cardiac output; there was no significant difference between the treated and untreated animals (Table). By contrast, there
was a distinct difference between treated and untreated animals with regard to flow though the ischemie limb (Fig 2 and Table). During the three hours after ancrod infusion, the mean iliac flow of the control group decreased by 34%, while the mean iliac flow of the treated group increased by 20%. Individually, blood flow decreased in all the ischemie limbs in the control group and increased in all the treated group. After 60 minutes, the difference between the two groups is statistically signifi¬ cant (P < .001). Since cardiac output decreased by a
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
Iliac Flow Iliac Flow Time Nonischemic Ischemie Cardiac Period HematoOutput, urnl), Limb, Affer ml/min* ml/min* ml/min* crit, %* Ancrod ,_«_v_«_„_*_y_«_, ConConConConInfusion, min trol Treated trol Treated trol Treated trol Treated re¬
treatment 30 60
218
204
74
49
3520
(108)
(79)
(13)
(22)
(920) (750)
(8.2) (5.3)
221
185
70
51
3430
2800
44.8
(94)
(66)
(14)
(18)
(710) (800)
(8.2)
(5.1)
202
170
62
53
3280
2780
44.8
44.3
2860
44.8
43.6 44.2
(82)
(61)
(12)
(22)
(670) (790)
(8.2)
(5.2)
90
184
156
60
55
3230
45.1
44.0
(75)
(48)
(13)
(25) (740) (620)
(8.0)
(5.8)
120
176
161
57
60
46.8
43.8
3050
2450 2560
(77)
(43)
(15)
(29)
(780) (770)
(7.9) (5.0)
173
153
53
60
3000
2129
46.0
(70)
(53)
(10)
(28)
(670) (660)
(6.0)
(5.0)
174
148
50
66
3110
2000
45.5
45.0
_(56)
(68)
(9)
(37) (880) (570)
(6.0)
(5.0)
150 180
"Numbers in
parentheses
in cardiac output to 50% the original value after three hours of pentobarbital anesthesia. Also, in our experiments, to prevent possible hemodilution and resul¬ tant decrease in blood viscosity, virtually no supportive intravenously administered fluids were given. In humans, therapeutic defibrination with ancrod has been used in the treatment of thromboembolism,13 priapism,14 retinal vascular occlusive disease,10 and following the insertion of cardiac valve prostheses.16 Toxic effects have been minimal with the doses required for defibrina¬ tion, and bleeding problems have not been greater than with conventional anticoagulants. Although the flow of blood has been extensively studied in the laboratory and to a lesser extent in normal circula¬ tion, little is known of the behavior of blood in the grossly ischemie circulation. These preliminary results suggest that therapeutic defibrination may be of value in the critically ischemie limb. Further investigation of the in vivo flow properties of blood is required, and studies are in progress to evaluate blood flow properties in the intact ischemie circulation.
steady decline
Comparative Effects of Ancrod on Blood Flow Through Ischemie vs Nonischemic Limb
44.8
indicate 1 SD.
greater amount in the treated group, this increase is even
striking. During the three hours of the experiment, the hemato¬ crit reading remained constant in each group (Table). more
COMMENT The dose of ancrod administered in this study, 1 unit/kg, is sufficient to reduce the plasma fibrinogen concentration in dogs to very low levels (unmeasurable by normal labora¬ tory methods) within two hours (unpublished data). Although viscosity measurements of blood differ with the type of viscometer used and the rate of shear at which they were measured, there is general agreement that defibrina¬ tion reduces blood viscosity. At the shear rates encountered in the normal limb (which are believed to vary from 100 sec-1 to 10 sec1),9 the reduction in whole blood viscosity achieved by defibrination is in the area of 20%.7 However, as shear rate is reduced, the viscosity of normal blood increases and there is good evidence that a yield stress exists.10 It seems likely that these effects are largely dependent on the presence of fibrinogen, since suspensions of RBCs in saline solution have been shown to behave in an almost Newtonian fashion.11 Thus, at extremely low shear rates, the concentration of fibrinogen may have a profound effect on the flow properties of blood. This may explain why, in our experiments, defibrination with ancrod increased blood flow through the ischemie limb, but not through the normal limb of the treated animals. The decrease in cardiac output that occurred in the ancrod-treated animals is thought not to be related to defibrination, since almost as large a decrease occurred in the control animals; but this may be attributable to the pentobarbital anesthesia. Indeed, Nash et al1- found a
This investigation was supported in part by grant-in-aid HL-03181 from the Heart and Lung Institute of the Public Health Service. The ancrod used in this investigation was supplied as Venacil through John Donahoe, MD, of Abbott Laboratories, North Chicago, 111.
References 1. Kilman JW, Waldhausen JA, Shumaker HB: Effects of low molecular weight dextran on peripheral blood flow with controlled cardiac output. Ann Surg 166:190-194, 1967. 2. Merrill EW: Rheology of blood. Physiol Rev 49:863-870, 1969. 3. Reid HA, Chan KE: The paradox in therapeutic defibrination. Lancet
1:485-486, 1968.
4. Kowalski E, Kopec M, Wegrzynowicz Z: Influence of fibrinogen degradation products (FDP) on platelets aggregation, adhesiveness and viscous metamorphosis. Thromb Diath Haemorrh 10(suppl):406-423, 1964. 5. Larrieu MJ, Marder VJ, Inceman S: Effects of fibrinogen degradation products on platelets and coagulation. Thromb Diath Haemorrh 20(suppl):215-223, 1966. 6. Prentice CRM, Hassanein AA, Turpie AGG, et al: Changes in platelet behaviour during Arvin therapy. Lancet 1:644-647, 1969. 7. Ehringer H, Dudczak R, Lechner K: A new approach in the treatment of peripheral arterial occlusions: Defibrination with Arvin. Angiology
25:279-281, 1974. 8. Barrie WW, Wood EH, Crumlish P, et al: Low dosage ancrod for prevention of thrombotic complications after surgery for fractured neck of femur. Br Med J 4:130-133, 1974. 9. Wells RE, Merrill EW: Influence of flow properties of blood upon viscosity hematocrit relations. J Clin Invest 41:1591-1598, 1962. 10. Cokelet GR, Merrill EW, Gilliland EP, et al: The rheology of human blood measurement near and at zero shear rate. Trans Soc Rheol 7:303-310, 1963. 11. Merrill EW, Cokelet GR, Britten A, et al: Non-Newtonian rheology of human blood: Effect on fibrinogen deduced by "substraction." Circ Res 13:48-55, 1963. 12. Nash CB, Davis F, Woodbury RA: Cardiovascular effects of anesthetic doses of pentobarbital sodium. Am J Physiol 185:107-112, 1956. 13. Bell WR: Treatment of thromboembolic disease by defibrinogenation with Arvin. Ann Intern Med 74:825-826, 1971. 14. Bell WR, Pitney WR: Management of priapism by therapeutic defibrination. N Engl J Med 280:649-650, 1969. 15. Bowell RE, Marmion VJ, McCarthy CF: Treatment of central retinal vein thrombosis with ancrod. Lancet 1:173-174, 1970. 16. Singh MP, Pitney WR, Melrose DG: Further experience in the use of ancrod (Arvin) to prevent thrombosis on prosthetic heart valves. Thorax 26:167-171, 1971.
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
Barrie, MB, ChB, FRCS, Worthington G. Schenk, Jr, MD
\s=b\ Using electromagnetic flow probes, cardiac output and hind limb blood flow were measured in dogs in which one hind limb had been rendered ischemic. Four dogs served as controls; seven were defibrinated by intravenous infusion of ancrod, 1 unit/kg, over a 30-minute period. In both groups, hematocrit readings remained constant, but cardiac output fell (this was attributed to barbiturate anesthesia), as did flow in the normal hind limb. In the controls after three hours, flow in the ischemic hind limb had decreased by 34%, but in the treated animals it had increased by 20%. The difference was statistically significant (P < .001). The selective increase in blood flow in the ischemic limb may be explained by the greater reduction in blood viscosity at low shear rates achieved by defibrination.
(Arch Surg 111:561-563, 1976) reduced blood flow is a factor common to a variety of clinical conditions including obstructive arterial disease, venous thrombosis, and the poor periph¬ eral perfusion associated with shock. Treatment is usually aimed at improving blood flow by lowering the vascular resistance, as in the reconstruction of an arterial stenosis, or improving the perfusion pressure by the infusion of plasma volume expanders. However, since the flow rate of a fluid is inversely related to its viscosity, it should also be possible to improve blood flow by lowering the viscosity of
Greatly
for publication Dec 4, 1975. From the Department of Surgery, State University of New York at Buffalo, and the Surgical Research Laboratories, E. J. Meyer Memorial Hospital, Buffalo, NY. Reprint requests to State University of New York, School of Medicine, 462 Grider St, Buffalo, NY 14215 (Dr Schenk).
Accepted
whole blood or plasma. Originally, it was thought that this was the mode of action of low molecular weight dextran, but subsequent studies have shown that the beneficial effect of this agent is more related to its action as a plasma volume expander.1 Since the contribution of fibrinogen to normal blood viscosity is second in importance only to the concentration of red blood cells (RBCs),'- the introduction of a drug, ancrod, which specifically reduces the plasma fibrinogen level, again raises the possibility of therapeutically altering the flow properties of blood. Ancrod, a purified enzymatic derivative of the venom of the Malayan pit viper, breaks down fibrinogen to degrada¬ tion products that are removed from the circulation by the reticuloendothelial system; the resultant effect is virtually complete defibrination.' Unlike the syndrome of dissemi¬ nated intravascular coagulation, other clotting factors are not affected, although the split products formed by the degradation of fibrinogen may exert a transient anticoag¬ ulant and antiplatelet action46. Early clinical reports suggest that therapeutic reduction of plasma fibrinogen may be of value in the treatment of peripheral arterial disease7 and in the prophylaxis of postoperative thromboembolism,8 but a direct improve¬ ment of in vivo blood flow has not previously been demon¬ strated. The purpose of this study was to measure blood flow through a critically ischemie limb during and after defibrination with ancrod. MATERIALS AND METHODS Eleven mongrel dogs, each weighing 18 to 22 kg, were anesthe¬ tized with pentobarbital sodium, 30 mg/kg, and ventilated with air
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
Treated Control Treated
Control
c
E
-
2
1
10-
D
O >
U -40 30
60
120
90
Minutes
,
150
180
Time after infusion
1—Mean cardiac output (± 2 SE) for control and treated animals. Fall in cardiac output is greater, but not significantly so, in treated group.
Fig
via a mechanical respirator. Thoracotomy was performed through the left fifth interspace and a noncannulating electromagnetic flow probe placed around the ascending aorta. A lower midline abdominal incision was made and the aortic trifurcation exposed. The incision was extended into the left groin to allow exposure of the left iliac and femoral arteries. A polyethylene catheter (internal diameter, 0.076 cm) was inserted into the abdominal aorta via the median sacral artery and the common internal iliac artery was ligated. The catheter was connected to an inductancetype pressure transducer to allow continuous monitoring of aortic pressure. Electromagnetic flow probes were placed around both iliac arteries just distal to the aortic trifurcation. On the left side, all the iliac and femoral arterial branches (including the profunda femoris artery) between the probe and the superficial femoral artery were ligated. The right iliofemoral arterial supply was not disturbed. After an interval to allow stabilization of the prepara¬ tion, the values of cardiac output, arterial pressure, and each iliac arterial flow were recorded. A "critical" arterial stenosis was produced by tying a ligature around the left superficial femoral artery and reducing the flow to the minimum that could be obtained without occluding the vessel completely. The preparation was then allowed to stabilize for approximately 45 minutes. Animals comprising the treated group, a total of seven, were infused intravenously with ancrod, 1 unit/kg in 50 ml of normal saline solution for a period of 30 minutes. Four control animals were given a similar volume of normal saline solution only. The hemodynamic variables of all the animals were recorded on a direct writing recorder for the remaining three hours of the experiment. Zero flow references were obtained periodically for the iliac probes by occlusion of the arteries just distal to the probes. On the left side, the absence of collateral supply was
Time minutes Fig 2.—Mean flow (± 2 SE) in ischemie limbs is shown as percentage change of pretreatment value. After 60 minutes, difference between groups is statistically significant (P < .001). -
confirmed by checking that zero flow was obtained by occlusion of the superficial femoral artery immediately proximal to the steno¬ sis. The diastolic quiet interval was taken as zero flow for the ascending aortic probe.
RESULTS The Table shows the mean changes, with standard deviations, in cardiac output and iliac arterial flow in the
four control and seven treated animals. In both groups, cardiac output fell during the experiment; the decrease was greater, but not significantly so, in the treated group (Fig
1).
Although there was considerable individual variation, in both treated and untreated animals the mean iliac arterial flow in the nonischemic limb decreased by a greater amount than cardiac output; there was no significant difference between the treated and untreated animals (Table). By contrast, there
was a distinct difference between treated and untreated animals with regard to flow though the ischemie limb (Fig 2 and Table). During the three hours after ancrod infusion, the mean iliac flow of the control group decreased by 34%, while the mean iliac flow of the treated group increased by 20%. Individually, blood flow decreased in all the ischemie limbs in the control group and increased in all the treated group. After 60 minutes, the difference between the two groups is statistically signifi¬ cant (P < .001). Since cardiac output decreased by a
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
Iliac Flow Iliac Flow Time Nonischemic Ischemie Cardiac Period HematoOutput, urnl), Limb, Affer ml/min* ml/min* ml/min* crit, %* Ancrod ,_«_v_«_„_*_y_«_, ConConConConInfusion, min trol Treated trol Treated trol Treated trol Treated re¬
treatment 30 60
218
204
74
49
3520
(108)
(79)
(13)
(22)
(920) (750)
(8.2) (5.3)
221
185
70
51
3430
2800
44.8
(94)
(66)
(14)
(18)
(710) (800)
(8.2)
(5.1)
202
170
62
53
3280
2780
44.8
44.3
2860
44.8
43.6 44.2
(82)
(61)
(12)
(22)
(670) (790)
(8.2)
(5.2)
90
184
156
60
55
3230
45.1
44.0
(75)
(48)
(13)
(25) (740) (620)
(8.0)
(5.8)
120
176
161
57
60
46.8
43.8
3050
2450 2560
(77)
(43)
(15)
(29)
(780) (770)
(7.9) (5.0)
173
153
53
60
3000
2129
46.0
(70)
(53)
(10)
(28)
(670) (660)
(6.0)
(5.0)
174
148
50
66
3110
2000
45.5
45.0
_(56)
(68)
(9)
(37) (880) (570)
(6.0)
(5.0)
150 180
"Numbers in
parentheses
in cardiac output to 50% the original value after three hours of pentobarbital anesthesia. Also, in our experiments, to prevent possible hemodilution and resul¬ tant decrease in blood viscosity, virtually no supportive intravenously administered fluids were given. In humans, therapeutic defibrination with ancrod has been used in the treatment of thromboembolism,13 priapism,14 retinal vascular occlusive disease,10 and following the insertion of cardiac valve prostheses.16 Toxic effects have been minimal with the doses required for defibrina¬ tion, and bleeding problems have not been greater than with conventional anticoagulants. Although the flow of blood has been extensively studied in the laboratory and to a lesser extent in normal circula¬ tion, little is known of the behavior of blood in the grossly ischemie circulation. These preliminary results suggest that therapeutic defibrination may be of value in the critically ischemie limb. Further investigation of the in vivo flow properties of blood is required, and studies are in progress to evaluate blood flow properties in the intact ischemie circulation.
steady decline
Comparative Effects of Ancrod on Blood Flow Through Ischemie vs Nonischemic Limb
44.8
indicate 1 SD.
greater amount in the treated group, this increase is even
striking. During the three hours of the experiment, the hemato¬ crit reading remained constant in each group (Table). more
COMMENT The dose of ancrod administered in this study, 1 unit/kg, is sufficient to reduce the plasma fibrinogen concentration in dogs to very low levels (unmeasurable by normal labora¬ tory methods) within two hours (unpublished data). Although viscosity measurements of blood differ with the type of viscometer used and the rate of shear at which they were measured, there is general agreement that defibrina¬ tion reduces blood viscosity. At the shear rates encountered in the normal limb (which are believed to vary from 100 sec-1 to 10 sec1),9 the reduction in whole blood viscosity achieved by defibrination is in the area of 20%.7 However, as shear rate is reduced, the viscosity of normal blood increases and there is good evidence that a yield stress exists.10 It seems likely that these effects are largely dependent on the presence of fibrinogen, since suspensions of RBCs in saline solution have been shown to behave in an almost Newtonian fashion.11 Thus, at extremely low shear rates, the concentration of fibrinogen may have a profound effect on the flow properties of blood. This may explain why, in our experiments, defibrination with ancrod increased blood flow through the ischemie limb, but not through the normal limb of the treated animals. The decrease in cardiac output that occurred in the ancrod-treated animals is thought not to be related to defibrination, since almost as large a decrease occurred in the control animals; but this may be attributable to the pentobarbital anesthesia. Indeed, Nash et al1- found a
This investigation was supported in part by grant-in-aid HL-03181 from the Heart and Lung Institute of the Public Health Service. The ancrod used in this investigation was supplied as Venacil through John Donahoe, MD, of Abbott Laboratories, North Chicago, 111.
References 1. Kilman JW, Waldhausen JA, Shumaker HB: Effects of low molecular weight dextran on peripheral blood flow with controlled cardiac output. Ann Surg 166:190-194, 1967. 2. Merrill EW: Rheology of blood. Physiol Rev 49:863-870, 1969. 3. Reid HA, Chan KE: The paradox in therapeutic defibrination. Lancet
1:485-486, 1968.
4. Kowalski E, Kopec M, Wegrzynowicz Z: Influence of fibrinogen degradation products (FDP) on platelets aggregation, adhesiveness and viscous metamorphosis. Thromb Diath Haemorrh 10(suppl):406-423, 1964. 5. Larrieu MJ, Marder VJ, Inceman S: Effects of fibrinogen degradation products on platelets and coagulation. Thromb Diath Haemorrh 20(suppl):215-223, 1966. 6. Prentice CRM, Hassanein AA, Turpie AGG, et al: Changes in platelet behaviour during Arvin therapy. Lancet 1:644-647, 1969. 7. Ehringer H, Dudczak R, Lechner K: A new approach in the treatment of peripheral arterial occlusions: Defibrination with Arvin. Angiology
25:279-281, 1974. 8. Barrie WW, Wood EH, Crumlish P, et al: Low dosage ancrod for prevention of thrombotic complications after surgery for fractured neck of femur. Br Med J 4:130-133, 1974. 9. Wells RE, Merrill EW: Influence of flow properties of blood upon viscosity hematocrit relations. J Clin Invest 41:1591-1598, 1962. 10. Cokelet GR, Merrill EW, Gilliland EP, et al: The rheology of human blood measurement near and at zero shear rate. Trans Soc Rheol 7:303-310, 1963. 11. Merrill EW, Cokelet GR, Britten A, et al: Non-Newtonian rheology of human blood: Effect on fibrinogen deduced by "substraction." Circ Res 13:48-55, 1963. 12. Nash CB, Davis F, Woodbury RA: Cardiovascular effects of anesthetic doses of pentobarbital sodium. Am J Physiol 185:107-112, 1956. 13. Bell WR: Treatment of thromboembolic disease by defibrinogenation with Arvin. Ann Intern Med 74:825-826, 1971. 14. Bell WR, Pitney WR: Management of priapism by therapeutic defibrination. N Engl J Med 280:649-650, 1969. 15. Bowell RE, Marmion VJ, McCarthy CF: Treatment of central retinal vein thrombosis with ancrod. Lancet 1:173-174, 1970. 16. Singh MP, Pitney WR, Melrose DG: Further experience in the use of ancrod (Arvin) to prevent thrombosis on prosthetic heart valves. Thorax 26:167-171, 1971.
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 05/23/2015
