- Email: [email protected]
Effects of dopamine, ethanol, and mannitol on cardiopulmonary function in patients with adult respiratory distress syndrome
J
THoRAc CARDIOVASC SURG
82:203-210, 1981
Effects of dopamine, ethanol, and mannitol on cardiopulmonary function in patients with adult respiratory distress syndrome Dopamine, ethanol, and mannitol were investigated to determine if they could increase pulmonary blood flow and oxygen delivery without significantly increasing intrapulmonary shunt. These drugs were studied in adult patients with respiratory distress following trauma, operation, or sepsis. Intravascular pressure, cardiac output, oxygen consumption and delivery, and limb blood flow and peripheral oxygen delivery were measured in all patients. Hypotensive patients received dopamine in incremental doses of 2 IJ-g/kg/min until either mean arterial pressure increased 15 mm Hg or heart rate increased by more than 15 beats/min. Ethanol was given as 10% ethanol in 5% dextrose at 2 ml/kg/hr. Mannitol was given as 25 gm of a 25% solution in a single bolus followed by infusion of 8 to 25 gm of 20% solution (mean 10 ± 2 gm) as a continuous intravenous drip over I hour. No drug produced a significant change in intrapulmonary shunt. Ethanol produced significant (p < 0.05) increases in cardiac index, heart rate, oxygen consumption, and oxygen delivery. Dopamine significantly decreased pulmonary vascular resistance while increasing systemic blood pressure. Visceral blood flow apparently increased while the peripheral vascular response to ischemia remained intact. Mannitol increased oxygen delivery and consumption in both the total body and limb. Thus in patients with adult respiratory distress syndrome (ARDS), increases in pulmonary blood flow can be achieved with several distinct pharmacologic agents without significant increases in intrapulmonary shunt. These increases in flow are generally accompanied by increases in oxygen delivery without increased pulmonary vascular resistance.
G. R. Rhodes, M.D., M. Taylor, M.D., J. C. Newell, Ph.D., D. M. Shah, M.D., W. A. Scovill, M.D., and S. R. Powers, M.D.,t Albany and Troy, N. Y.
Adult respiratory distress syndrome (ARDS) is commonly treated by the application of end-expiratory pressure either in the form of positive end-expiratory pressure (PEEP) 1 during mechanical ventilation or continuous positive airway pressure (CPAP)2 during spontaneous ventilation. End-expiratory pressure increases alveolar volume and usually results in increased oxygen delivery through a decrease in intrapulmonary
From the Departments of Physiology and Surgery, Albany Medical College, Albany, N. Y., and Centerfor Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, N. Y. Supported by the following grants: P-50-GM 15426, T-32GM07033, MO-I-RR-00749, National Institutes of Health. Received for publication May 21, 1979. Accepted for publication Feb. 3, 1981. Address for reprints: D. M. Shah, M.D., Department of Surgery, Albany Medical College, Albany, N. Y. 12208. Address for correspondence: J. C. Newell, Ph.D., Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, N. Y. 12181. tDeceased.
shunt. It is also possible to increase oxygen delivery by increasing pulmonary blood flow. This study examines the cardiopulmonary effects of three drugs, dopamine, an inotropic agent, ethanol, a vasodilator, and mannitol, an osmotic agent, in a series of patients in whom ARDS developed following trauma, operation, or sepsis. These drugs were investigated to determine if they could increase pulmonary blood flow and oxygen delivery without a significant increase in intrapulmonary shunt. The effects of mannitol, although previously published," are presented here to permit a comparison among all three drugs studied.
Methods-General Twenty patients were studied who had significant hypoxemia following trauma, surgical operation, or the development of sepsis (Table I). Informed consent was obtained either directly or by proxy, according to institutional guidelines. All patients had intrapulmonary shunts greater than 13%, together with radiologic findings of bilateral diffuse pulmonary infiltrates that
0022-5223/8I!080203+08$OO.80/0 © 1981 The C. V. Mosby Co.
203
The Journal of
204 Rhodes et at.
Thoracic and Cardiovascular Surgery
Table I Case No.
Age and sex
Injuries and operations
19, M
2
3
4 5
6
7
8
9
10
II
12 13 14 IS 16 17 18 19 20
Fracture of right and left fibula and tibia, comminuted fracture of left humerus, cerebral concussion, aspiration 21, M Fractured mandible, scalp and facial lacerations, traumatic abdominal aortic aneurysm necessitating resection and grafting, ruptured spleen necessitating splenectomy 58, M Fractures of left clavicle, right ribs 2-6, right tibia and fibula, right pulmonary contusion, scalp and facial lacerations; exploratory laparotomy with finding of mesenteric hematoma; cerebral concussion Blunt abdominal trauma with renal contusion and ruptured spleen 52, F necessitating splenectomy; multiple pelvic fractures 16, M Right hemopneumothorax, right radial and ulnar fractures, C.-C a fracture with paralysis and probable T 7 fracture; repair lacerated colon and lacerated duodenum 28, M Left parietal skull fracture and scalp laceration, right forearm lacerations, right femoral neck and midshaft fracture, right femoral supercondylar fracture, fractured left clavicle, and dislocated left ankle; exploratory laparotomy with cauterization of liver laceration 22, M Blunt abdominal injury with rupture of cisterna chyli, puncture of duodenum, retroperitoneal bleed; above-knee amputation of right leg, open debridement, and internal fixation of left leg 38, F Parietal scalp laceration, depressed skull fracture, facial fractures, fracture of both pubic rami, vaginal lacerations, partial transection of urethra, multiple rib fractures, fractured sternum, right pneumothorax Severe crush injury to lower pelvis and left leg with multiple fracIS, F tures of the left femur, destruction of the left acetabulum, crush injury to almost all left adductor and gluteal muscle groups, left sciatic nerve transection, devitalization of iliac crest, retroperitoneal hematoma 53, F Blunt abdominal injury necessitating removal of 50% of small bowel, mid-transverse colonic tear, right knee puncture necessitating open exploration, late ischemic injury to right forearm necessitating open exploration, late ischemic injury to right forearm necessitating above-elbow amputation of right arm 29, F Multiple rib fractures, facial fractures, rupture of stomach and duodenum with chemical peritonitis, crushed right kidney necessitating right nephrectomy, gastric resection, Biliroth II gastrojejeunostomy, and tube duodenostomy 64, M Bilateral fractured ribs, cerebral contusion, left lung contusion Perforated small bowel, intra-abdominal sepsis 62, F 70, M Ruptured abdominal aneurysm 56, M Common bile duct diversion; sepsis, dehiscence Exploratory laparotomy, lysis of adhesions, ileal resection; post48, F operative sepsis 6O,M Esophagectomy, colon interposition, partial liver resection Portocaval shunt, excessive intraoperative hemorrhage 51, F 71, M Exploratory laparotomy for subphrenic abscess, lysis of adhesions, continual sepsis Emergency portocaval shunt with hemorrhagic shock 45, F
Legend: ARDS, Adultrespiratory distress syndrome.
Principal cause of ARDS
Protocols
Mannitol
I
Ethanol
I
Sur-
Dopamine
vived
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
+
Trauma
+
+
+
Sepsis
+
+
+
Trauma Sepsis Operation Sepsis Sepsis
+ + + + +
Operation Operation Sepsis
+ +
Operation
+ + + + + + +
+
Volume82 Number2
August. 1981
could not be ascribed to acute inflammatory disease, congestive heart failure, pneumonia, or large pulmonary emboli. The findings in these patients constitute a syndrome consistent with ARDS. Drugs were studied on separate days in almost all instances. Mannitol usually was evaluated before ethanol, and dopamine was evaluated only in patients with specific indications (described subsequently). Some drugs were not studied in certain patients because of the patient's clinical improvement or unstable clinical condition. The initial studies in all patients were performed within 48 hours of the development of hypoxemia. No patient was studied while in shock. In addition, no patient died within 3 days of the studies. All patients were studied while recumbent with the head elevated 30 degrees. Changes were considered statistically significant at a level of p < 0.05 according to Student's paired t test. A femoral venous catheter was inserted in all patients receiving dopamine or mannitol and in six of 10 patients receiving ethanol. An arterial catheter and a triple-lumen flow-directed pulmonary arterial catheter were placed in all patients. No attempt was made to determine the altitude of the tip of the Swan-Ganz catheter. Pressure transducers were zeroed to left atrial level, estimated to be 5 em below the angle of Lewis. Mean arterial (MAP), central venous (CVP), pulmonary artery (PAP), and pulmonary wedge (PWP) pressures were measured by strain-gauge transducers and recorded on a polygraph. Pressures were recorded at end-expiration. Cardiac output (CO) was determined by indocyanine green (Cardio-green) dye-dilution technique, and the mean of duplicate values is reported. Pulmonary vascular resistance (PVR) and systemic vascular resistance (SVR) were calculated as follows: PVR = (80 [PAP - PWP])/CO; SVR = (80 [MAP CVP])/CO. Arterial, mixed venous, and femoral venous blood samples were analyzed for P0 2, Pc0 2, and pH with standard electrodes and were corrected to body temperature. Intrapulmonary shunt was calculated from the shunt equation at the therapeutic inspired oxygen level. Hemoglobin concentration (Hb) was measured by spectrophotometric technique. Limb blood flow was measured in the calf with venous occlusion plethysmography and expressed in milliliters per 100 gm of tissue per minute." Limb blood flow was measured both at rest and 20 seconds after a 2 minute period of ischemia as a measure of the reactive hyperemic response. Ischemia was produced by inflating the blood pressure cuff above arterial systolic pressure. Total body oxygen consumption ("\10 2) and oxygen
Adult respiratory distress syndrome
205
delivery (02DTB) and limb oxygen consumption (VLo.) and oxygen delivery (02DL) were calculated as follows: 02D = Q (0.031 Pao. + 13.9 [Hb] [Sa o.ll00]) where Q is either cardiac output for total body or blood flow for the limb. Venous oxygen return = Q (0.031 Pv.; + 13.9 [Hb] Sv o./100]) and V0 2 = arterial oxygen delivery - venous oxygen return. For total body calculations, PVa. = PVa., while for the limb, PVa.was that in femoral venous blood. Patients were maintained on a volume-cycled ventilator with tidal volumes of 12 to 15 ml/kg, respiratory rates of 12 to 15 breaths/min, PEEP of 5 to 15 cm H20, and fractions of inspired oxygen of 0.40 to 0.55. All ventilatory parameters were held constant for the duration of the study. The PEEP level was chosen in each case to maximize oxygen delivery. Procedure Dopamine. Seven patients were studied, all of whom had been requiring dopamine to maintain mean arterial pressure above 70 mm Hg. They were studied with the drug infusion at a time when all but one had a mean arterial pressure less than 70 mm Hg. This patient had an unstable blood pressure necessitating intermittent dopamine support. Following baseline measurements with no drug infusion, a dopamine infusion (100 JLg/ml) was begun and increased at increments of 2 JLg/kg/min until either mean arterial pressure increased 15 mm Hg or heart rate increased by more than 15 beats/min. After 15 minutes had elapsed to allow achievement of a steady state, all measurements were repeated. Ethanol. Ten patients were studied. Pulmonary capillary blood was examined histologically for the presence of fat emboli during the first 3 days following admission to the trauma center. * This technique consists of the slow withdrawal of pulmonary capillary blood from the Swan-Ganz catheter with the balloon inflated, followed by sectioning of the clotted blood with a cryostat and staining for neutral fat. Serum ethanol and lactate were measured by standard techniques. Measurements were performed before and 1, 3, and 6 hours after the beginning of an infusion of 10% ethanol in 5% dextrose solution infused intravenously at 2 ml/kg/hr for 6 hours. Because of its diuretic effect, these patients required from 500 to 1,000 ml of Ringer's lactate solution as additional fluid over the 6 hour period to maintain the pulmonary capillary wedge pressure at control levels. *Dutton RE: Diagnosis of fat embolism with a Swan-Ganz catheter. Personal communication.
The Journal of Thoracic and Cardiovascular Surgery
206 Rhodes et al.
Table II. Cardiopulmonary effects of ethanol
_______________I'-----L.-. MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (Lzrnin/m") PVR (dyne-sec/em") SVR (dyne-sec/em") Arterial P0 2 (mm Hg) Mixed venous P0 2 (mm Hg) Pulmonary shunt (%) Total oxygen consumption (ml/Os/rnin) Total body oxygen delivery (m1/0 2/min) Limb blood flow, rest (ml/loo gm/min) Limb blood flow, ischemia (ml/IOO gm/min) Limb oxygen consumption (ml O2/ 100 gm/min) Limb oxygen delivery (mI/02/100 gm/min) Ethanol (mg/loo ml) Lactate (mg/loo ml)
Control
One hour
85 ± 4 4±1 19 ± 3 4±1 100 ± 7 3.84 ± 0.36 192 ± 27 1,IOO±110 98 ± II 40 ± 2 24 ± 4 221 ± 24 980 ± 100 0.9±0.13 2.68 ± 0.48 0.027 ± 0.006 0.16 ± 0.06
82 ± 4 5 ± I 20 ± 3 4 ± I 117 ± 7 4.56 ± 0.51 181 ± 37 920 ± 130* 93 ± 9 40 ± 2 27 ± 5 276 ± 29* 1,210± 140* 1.3 ± 0.45 2.02 ± 0.38 0.048 ± 0.015 0.26 ± 0.07 8.5 ± 1.2 23 ± 4
o
20 ± 3
.l...-
Three hours 74 ± 5 5 ± 3 20 ± 3 4 ± I 118 ± 5 4.52 ± 0.4* 187 ± 34 830 ± 140* 98 ± 12 36 ± 3 26 ± 6 284 ± 27* 1, 170 ± 120* 1.47 ± 0.39* 2.52 ± 0.69 0.06 ± 0.02 0.27 ± 0.07 12.1 ± 2.6* 25 ± 4
L..-
Six hours
_
77 ± 5 4±1 19 ± 3 4±1 125 ± 6* 4.28 ± 0.43 190 ± 32 930 ± 120 98 ± II 40 ± 6 26 ± 5 277 ± 29* 1,130 ± 130 1.16 ± 0.26 2.32 ± 0.65 0.037 ± 0.009 0.21 ± 0.04 12.0 ± 2.7* 25 ± 4
Legend: CI, Cardiac index. CVP, Central venous pressure. HR, Heart rate. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. 'p < 0.05; statistical significance based on comparison with control values.
Mannitol. Fourteen studies were performed in II patients. Studies were repeated in two patients at least 24 hours apart after the patient had a septic episode associated with hypotension and necessitating additional fluid therapy. Control measurements were performed before drug administration. Then 25 gm of 25% mannitol was infused as a single bolus, followed by the infusion of 8 to 25 gm of 20% mannitol (mean 10 ± 1.9 gm) as a continuous intravenous drip over I hour. All measurements were repeated during the last 10 minutes of this period. Results Dopamine. During dopamine infusion at rates between 4 and 15 ILg/kg/min, cardiac index increased from a baseline of 4.3 ± 0.7 to 5.7 ± 1.0 L'rnin/rn" (p < 0.05) (Table II). Oxygen consumption rose from 261 ± 34 to 305 ± 38 mllmin (p < 0.005). There were no significant changes in arterial, mixed venous, or femoral venous P0 2 or oxyhemoglobin saturation. Intrapulmonary shunt remained unchanged. Central venous and pulmonary wedge pressures did not change significantly. Mean heart rate rose from 106 ± 2 to 114 ± 5 beats/ min, but this change was not statistically significant. Despite a significant rise in pulmonary artery pressure, 21 ± 2 to 24 ± 2 mm Hg (p < 0.01), the pulmonary vascular resistance fell from 200 ± 50 to 170 ± 50 dyne-sec/em" (p < 0.005).
Limb blood flow, limb oxygen consumption and delivery, and the postischemic hyperemic response were unchanged by dopamine infusion. Ethanol. Following I hour of ethanol infusion, total body oxygen delivery increased significantly from 980 ± 100 to 1,200 ± 100 ml/Oj/rnin associated with an increase in oxygen consumption from 220 ± 20 to 290 ± 30 mIl0 2/min (Table III). Systemic vascular resistance decreased as a result of the combined effects of increasing cardiac output and decreasing mean arterial pressure (Table III). All other parameters remained unchanged at I hour. By 3 hours, there were significant increases in cardiac index, accompanied by increases in total body oxygen delivery and oxygen consumption. The increase in cardiac index was due principally to the increase in heart rate, from a mean 110 ± 10 beats/min at control to 125 ± 10 beats/min at 6 hours. There were no significant changes in any pressure measured throughout the study, although mean arterial pressure tended to decrease from 84 ± 4 mm Hg at control to 76 ± 5 mm Hg at 6 hours. Neither pulmonary vascular resistance nor intrapulmonary shunt changed significantly: however, increases in intrapulmonary shunt and pulmonary vascular resistance occurred in four of five patients whose initial intrapulmonary shunt exceeded 20%, Stains of pulmonary capillary blood from two of the 10 patients were positive for fat. These two patients had multiple long bone fractures. Their initial intrapul-
Volume 82 Number 2 August, 1981
Adult respiratory distress syndrome
Table III. Cardiopulmonary effects of dopamine
MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (Lzrnin/m") PVR (dyne-see-em") SVR (dyne-see-em') Arterial P0 2 (mm Hg) Mixed venous P0 2 (mm Hg) Femoral venous P0 2 (mm Hg) Hemoglobin (gm/loo ml) Pulmonary shunt (%) Total body oxygen consumption (ml 02/min) Total body oxygen delivery (ml Oj/rnin) Limb blood flow, rest* (ml/ 100 gm/min) Limb blood flow, ischemia" (mllloo gm/min) Limb oxygen consumption, rest (ml 02/min/100 gm) Limb oxygen delivery, rest (ml 02/min/100 gm)
I
Control 63 9.0 21.4 5.9 106 4.3 200 690 98 43 44 11.9 26.0 261
± ± ± ± ± ± ± ± ± ± ± ± ± ±
4 2.0
I.7 1.3 2 0.7 50 104 9 4 4 0.5 5.0 34
1,170 ± 210
I
During dopamine 78 10.0 23.7 8.0 114 5.7 170 740 97 43 45 11.9 29.0 305
± ± ± ± ± ± ± ± ± ± ± ± ± ±
4t 2.0 2.2t 1.7 5
i.ot
50t 170 9 3 3 0.5 6.0 38t
1,600 ± 320t
1.02 ± 0.16
1.16 ± 0.24
1.84 ± 0.42
1.97 ± 0.50
0.036 ± 0.008
0.035 ± 0.007
0.16 ± 0.03
0.19 ± 0.05
Legend: CI, Cardiac index. CVP, Central venous pressure. HR, Heart rate. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. • p < 0.05 comparing control limb blood flow at rest versus control limb blood flow after ischemia; p < 0.05 comparing effect of dopamine on limb blood flow at rest versus limb blood flow after ischemia. t p < 0.05; statistical significance based on comparison with control values.
monary shunts were 20% and 23%, which were not markedly different from the mean shunt of 24%. Changes in cardiopulmonary parameters following ethanol infusion in these two patients were in the same direction as in the whole group. We found no fat in the pulmonary capillary blood of two other patients in this group with long bone fractures. Reactive hyperemia was present before drug infusion, since limb blood flow after ischemia significantly exceeded that at rest; however, at 3 and 6 hours, reactive hyperemia was abolished, as indicated by similar values of limb blood flow after ischemia and at rest (Table III). Serum lactate levels were unchanged by ethanol (Table III). Mannitol. Mannitol produced a significant increase in cardiac output (p < 0.005) without changes in either right (central venous) or left (pulmonary capillary wedge) side filling pressure (Table IV). There was no significant change in either mean arterial or pulmonary
207
Table IV. Cardiopulmonary effects of mannitol _ _ _ _ _ _ _ _ _ _1
MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (liters/rnin/m") PVR (dyne-sec/em') SVR (dyne-sec/em') Arterial P0 2 (mm Hg) Mixed venous P0 2 Femoral vein P0 2 (mm Hg) Hemoglobin (gm/ I00 ml) Pulmonary shunt (%) Total body oxygen consumption (ml 02/min) Total body oxygen delivery (rnl Og/min) Limb blood ftlow, rest* (ml/IO) gm/min) Limb blood flow ischemia" (ml/lOO gm/min) Limb oxygen consumption, (ml/loo gm/min) Limb oxygen delivery, (mil 100 gm/min) Limb oxygen eonsumptiont (mllIOO gm/min)
Control 81 8 20 7 107 4.21 163 888 104 40 40 11.6 22.5 261
± 4 ± I ± I ± I ± 6 ± 0.36 ± 23 ± 98 ± 7 ± 2 ± 3 ± 0.5 ± 2.1 ± 25
1,200 ± 105
1 During mannitol
81 8 21 8 113 4.94 133 745 101 41 40 11.4 24.2 298
± 4 ± I ± I ± I ± 6 ± 0.39* ± 18 ± 73* ± 7 ± I ± 3 ± 0.4 ± 2.2 ± 27*
1,350 ± 110*
1.06 ± 0.23
1.20 ± 0.22
2.02 ± 0.49
2.12 ± 0.46
0.040 ± 0.007
0.043 ± 0.005
0.17 ± 0.03
0.19 ± 0.04
0.029 ± 0.006
0.042 ± 0.004*
Legend: CI, Cardiac index. CVP, Central venous pressure. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. HR, Heart rate. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. LBF, Limb blood flow. * P < 0.01 comparing control limb blood flow at rest versus control limb blood flow after ischemia; p < 0.05 comparing the effect of mannitol on limb blood flow at rest versus limb blood flow after ischema. t These values represent changes in limb oxygen consumption within the nine patients in whom limb oxygen delivery increased. *p < 0.05; statistical significance based on comparison with control values.
artery pressure. Systemic vascular resistance fell significantly without a change in pulmonary vascular resistance. Total body oxygen delivery increased significantly, without an increase in pulmonary shunt. Arterial and mixed venous blood gases were unchanged by mannitol infusion. This increase in total body oxygen delivery was seen in all but two patients studied. Among the exceptions, there was a decrease in cardiac output in one and a decrease in hemoglobin concentration in the other. There was a significant increase in total body oxygen consumption. Mean limb blood flow and limb oxygen delivery and consumption increased in eight of the nine studies in which blood flow increased. Chi square analysis demonstrated that oxygen consumption and delivery changed in the same direction (X 2 = 7 [p < 0.05]) for the total body and (X 2 = 9.5 [p < 0.01]) for the limb.
The Journal of
208
Rhodes et al.
Thoracic and Cardiovascular Surgery
Discussion
The major emphasis of therapy for ARDS has been the use of increased inspired oxygen together with PEEP or CPAP to decrease intrapulmonary shunting, raise arterial Po 2, and thereby increase oxygen delivery to the tissues. These techniques tend to minimize the effect of ventilationlperfusion inequalities, a significant cause of pulmonary shunting. In contrast, the studies presented here suggest that a pharmacologic approach may be used also to increase oxygen delivery by increasing pulmonary perfusion without causing a significant increase in shunt. In most patients studied, any of these increases in oxygen delivery were associated with significant increases in oxygen consumption for total body and limb. Although pulmonary vascular resistance was initially high in most patients, increased pulmonary blood flow did not result in additional increases in pulmonary vascular resistance with any drug studied. A strict comparison of the three drug groups is complicated by several factors. Not all patients were studied with each drug because of clinical instability, rapid clinical improvement, or technical problems. There were varying causes for the development of ARDS both within and between groups. Chest roentgenograms indicative of an increase in interstitial and alveolar fluid were common in all patients. Initial central venous and pulmonary wedge pressures were lower in the ethanol group than in the other groups, for unknown reasons. Despite these differences among groups, the cardiorespiratory responses to the three drugs were similar. Dopamine. In our hypotensive patients with ARDS, dopamine increased cardiac output and reduced pulmonary vascular resistance, as it does in most patients with cardiac disease. Intrapulmonary shunt and limb blood flow remained unchanged. The latter suggests that the increased cardiac output predominantly perfused central organs. In some animal models'': 6 and groups of patients such as those with sepsis," higher doses of dopamine may not reduce pulmonary vascular resistance. However, four of seven of the patients in our study were in a septic condition. Nevertheless, the significant increase in limb blood flow following ischemia, both before and during dopamine infusion, suggests that the ability of muscle capillary beds to dilate in response to an ischemic stimulus is not adversely affected by dopamine at the moderate concentrations employed. In previous studies of the effects of dopamine on cardiopulmonary function in different patient populations,8-13 cardiac output has increased, but the changes in pulmonary vascular resistance have been variable. In
a study of critically ill patients who were in a septic condition, dopamine infusion at rates of 2 to 55 JLgl kg/min produced a marked increase in mean arterial pressure, cardiac output, and pulmonary artery pressure but no significant changes in systemic or pulmonary vascular resistance." Most patients in that study were receiving dopamine when baseline studies were performed and measurements were repeated at a higher dosage. In contrast, our patients received no dopamine during the control measurements. In patients with congestive heart failure secondary to cardiac disease, a significant rise in mean arterial pressure and stroke volume was found at infusion rates greater than 5 JLg/kg/min. However, no significant changes in pulmonary vascular resistance or pulmonary artery, pulmonary wedge, or central venous pressure were noted. None of their patients were described as having abnormal pulmonary function and all had higher mean arterial and pulmonary artery pressures than the patients in the present study. In contrast, dopamine decreased pulmonary vascular resistance in postoperative patients undergoing cardiac bypass, II in patients in cardiogenic shock? resulting from myocardial infarction, and in patients undergoing cardiac catheterization.l? Ethanol. Ethanol is a potent vasodilator used to treat ischemic vascular disease. 14. IS Its use has been proposed in the therapy of pulmonary fat embolism.v" " either because of its possible role as a pulmonary vasodilator or as a specific treatment of ARDS associated with fat embolism. IS However, there is little physiological evidence for the use of ethanol in treating ARDS with or without associated fat embolism. This study evaluates the ability of low doses of intravenous ethanol to increase oxygen delivery and consumption in patients with ARDS of multiple origins, including fat embolism and sepsis. The effect of ethanol on cardiopulmonary function has been studied in animals.w?' normal subjects ,22. 23 and patients with chronic obstructive lung disease and coronary artery disease. Several investigations have demonstrated its effect as a systemic vasodilator. 14. IS. 22-24 These investigations, which have been performed in spontaneously breathing subjects, have generally not addressed changes in pulmonary blood flow, oxygen consumption, and oxygen delivery. In studies of the effect of ethanol on cardiac function.':" 19-24 the doses of ethanol used were as high as ten times those used in this study, so that comparison is difficult. The route and concentration of ethanol used to evaluate cardiovascular function may be important. Our study evaluated the effects of a relatively low dose of ethanol, which has been reported." to produce sig-
Volume82 Number 2 August. 1981
nificant systemic vasodilation without marked impairment in cardiovascular function. Turnipseed, Vasko, and Evans!" used a dose of 10% ethanol infused at 2 ml/kg/hr to produce a decrease in blood volume of 17 ml/kg. We tried to prevent a blood volume decrease in the present study by infusing a balanced salt solution at rates sufficient to keep the pulmonary wedge pressure constant. The volume of excess intravenous solution required was about the same as the decrease in blood volume reported in the earlier study. 15 This added fluid, which kept wedge pressure at a steady value in our study, may explain why we found an increased cardiac output without a fall in mean blood pressure; by contrast, in previous studies, cardiac output and/or mean blood pressure decreased following ethanol administration.I9-21. 23 Pulmonary vascular resistance was above normal levels at the beginning of this study and remained unchanged with ethanol infusion despite the increase in pulmonary blood flow. It is possible that ethanol may have increased blood flow through pulmonary capillaries that were either fully or partially closed by vasoconstriction, microemboli, or other mechanisms. Although the mean change in pulmonary shunt was not statistically significant, increases in shunt were frequently seen in patients with shunts initially greater than 20%. It remains uncertain whether the increase in pulmonary blood flow occurred through previously patent pulmonary gas exchange units with adequate ventilation or whether a shunt increase may have occurred in some of these lungs. Mannitol. In this study, infusion of 25 gm of hypertonic mannitol significantly increased cardiac output without increasing pulmonary shunt or pulmonary vascular resistance. These results differ from the immediate effect of a 25 gm bolus injection of hypertonic mannitol, which decreased pulmonary vascular resistance as well as its vasodilator effects seen in the kidney, heart, brain, and lung. 25-29 The mechanism for this reduced vascular resistance may involve the osmotic removal of water from endothelial or adjacent cells. The increase in total body oxygen delivery associated with an increase in oxygen consumption earlier reported in post-trauma patients I is confirmed here in a more diverse group of critically ill patients. In addition, when limb blood flow increases following administration of hypertonic mannitol, there is the same pattern of increased oxygen consumption and delivery for the limb as is found in the total body. Thus in patients with ARDS, including those requiring cardiovascular support, increases in pulmonary blood flow were achieved with several distinct pharma-
Adult respiratory distress syndrome
209
cologie agents without significant increases in intrapulmonary shunt. These increases in flow were generally accompanied by increases in oxygen delivery without increased pulmonary vascular resistance. On the contrary, decreases in pulmonary vascular resistance were seen with dopamine and mannitol. In critically ill patients, increases in oxygen delivery may be associated with significant increases in oxygen consumption for all three drugs studied. REFERENCES Powers SR, Mannal R, Neclerio M, English M, Marr C, Leather R, Udea H, Williams G, Custead W, Dutton R: Physiologic consequences of positive end-expiratory pressure (PEEP) ventilation. Ann Surg 178:265-272, 1973 2 Shah DM, Newell JC, Dutton RE, Powers SR Jr: Con-
tinuous positive airway pressure versus positive end-expiratory pressure in respiratory distress syndrome. J THORAC CARDIOVASC SURG 74:557-562, 1977 3 Rhodes GR, Newell JC, Shah DM, Scovill W, Tauber J, Dutton RE, Powers SR Jr: Increased oxygen consumption accompanying increased oxygen delivery with hypertonic mannitol in adult respiratory distress syndrome. Surgery 84:490-497, 1978 4 English MC, Lozman J, Powers SR Jr: A refined technique for the determination of limb blood flow by venous occlusion impedance plethysmograph. Proceedings of the Second Annual New England Bioengineering Conference, Worcester, Mass., 1974 5 Kaumann AJ, Ochoa E: Similar receptors mediating norepinephrine and dopamine contraction in main pulmonary artery of dog. Acta Physiol Lat Am 20:451, 1970 6 Mentzer RM Jr, Alegre C, Nolan SP: The effects of dopamine and isoproterenol on the pulmonary circulation. J THoRAc CARDIOVASC SURG 71:807-814, 1976 7 Wilson RF, Sibbalb WJ, Jaanimagi JL: Hemodynamic effects of dopamine in critically ill septic patients. J Surg Res 20:163-172, 1976 8 Beregovich J, Bianchi C, Rubler S, Lomnitz E, Cagin N, Levitt B: Dose-related hemodynamic and renal effects of dopamine in congestive heart failure. Am Heart J 87: 550-557, 1974 9 Flammang D, Sebastian PH, Bouvain Y: Etude hemodynamique de la dopamine employee dans les cardiopathies chroniques et dans la choc cardiogenique compliquant I'infarctus aigu du myocarde. Ann Anesthesiol Fr 16:669-672, 1975 10 Harrison DC, Pirages S, Robison SC, Wintroub BU: The pulmonary and systemic circulatory response to dopamine infusion. Br J Pharmacol 37:618-626, 1969 II Holloway EL, Stinson EB, Derby GC, Harrison DC: Action of drugs in patients early after cardiac surgery. I. Comparison of isoproterenoland dopamine. Am J Cardiol 35:656-659, 1975
2 10
The Journal of Thoracicand Cardiovascular Surgery
Rhodes et al.
12 Loeb HS, Winslow EB, Rahimtoola SH, Shahbudin H, Rosen KM, Gunnar RM: Acute hemodynamic effects of dopamine in patients with shock. Circulation 44: 163-173, 1971 13 Talley RC, Goldberg LI, Johnson CE, McNay JL: A hemodynamic comparison of dopamine and isoproterenol in patients with shock. Circulation 39:361-378, 1969 14 Cook EN, Brown GE: Vasodilating effects of ethyl alcohol on peripheral arteries. Mayo Clin Proc 7:449-452, 1932 15 Turnipseed W, Vasko JS, Evans WE: Hemodynamic effects of intravenous alcohol on patients with ischemic limb disease. J Surg Res 20:477-480, 1976 16 Herrmann LG: Effects of dextrose-alcohol mixture upon pulmonary fat embolism. Proc Soc Exp Bioi Med 30:558-559, 1933 17 Yale CE, Herrmann LG: Pathologic physiology of pulmonary fat embolism. Mechanical and chemical alterations with an effective antidote. Surg Forum 12:451-453, 1961 18 Alder F, Lai SP, Peltier LF: Fat embolism. Prophylactic treatment with lipase inhibitors. Surg Forum 12:453-455, 1961 19 Liedtke AJ, DeMuth WE: Effects of alcohol on cardiovascular performance after experimental nonpenetrating chest trauma. Am J CardioI35:243-250, 1975 20 Gettler DT, Allbritten FF Jr: Effect of alcohol intoxication on the respiratory exchange and mortality rate associated with acute hemorrhage in anesthetized dogs. Ann Surg 158:151-158, 1963 21 Moss LK, Chenault OW Jr, Gaston EA Jr: The effects of
22 23
24
25
26
27
28
29
alcohol ingestion on experimental hemorrhagic shock. Surg Forum 10:390-392, 1960 Juchems R, Klobe R: Hemodynamic effects of ethyl alcohol in man. Am Heart J 78:133-135, 1969 Riff DP, Jain AC, Doyle JT: Acute hemodynamic effects of ethanol on normal human volunteers. Am Heart J 78:592-597, 1969 Horwitz 0, Montgomery H, Longaker ED, Sayen A: Effects of vasodilator drugs and other procedures on digital cutaneous blood flow, cardiac output, blood pressure, pulse rate, body temperature, and metabolic rate. Am J Med Sci 218:669-682, 1949 Moore FD, Lyons JH, Pierce EF Jr, Morgan AP, Drinker PA, MacArthur JD, Mannin GJ: Post traumatic respiratory insufficiency. Philadelphia, 1969, W. B. Saunders Company Powers SR Jr, Ryon D, Newell JC, Ralph C, Scovill W, Dutton RE: Hypertonic mannitol in the therapy of the acute respiratory distress syndrome. Ann Surg 185:619625, 1977 Ames A III, Wright RL, Kowada M, Thurston JM, Majno G: Cerebral ischemia. II. The no-reflew phenomenon. Am J PathoI53:437-453, 1968 Hostnik WJ, Powers SR Jr, Boba A, Stein AA: Observations on the effect of mannitol on renal hemodynamics and O 2 tension in the urine and renal vein. Surg Forum 10:872-875, 1960 Powell WJ Jr, DiBona DR, Flores J, Frega N, Leaf A: Effects of hyperosmotic mannitol on reducing ischemic cell swelling and minimizing myocardial necrosis. Circulation 53:Suppl 1:45-49, 1976
THoRAc CARDIOVASC SURG
82:203-210, 1981
Effects of dopamine, ethanol, and mannitol on cardiopulmonary function in patients with adult respiratory distress syndrome Dopamine, ethanol, and mannitol were investigated to determine if they could increase pulmonary blood flow and oxygen delivery without significantly increasing intrapulmonary shunt. These drugs were studied in adult patients with respiratory distress following trauma, operation, or sepsis. Intravascular pressure, cardiac output, oxygen consumption and delivery, and limb blood flow and peripheral oxygen delivery were measured in all patients. Hypotensive patients received dopamine in incremental doses of 2 IJ-g/kg/min until either mean arterial pressure increased 15 mm Hg or heart rate increased by more than 15 beats/min. Ethanol was given as 10% ethanol in 5% dextrose at 2 ml/kg/hr. Mannitol was given as 25 gm of a 25% solution in a single bolus followed by infusion of 8 to 25 gm of 20% solution (mean 10 ± 2 gm) as a continuous intravenous drip over I hour. No drug produced a significant change in intrapulmonary shunt. Ethanol produced significant (p < 0.05) increases in cardiac index, heart rate, oxygen consumption, and oxygen delivery. Dopamine significantly decreased pulmonary vascular resistance while increasing systemic blood pressure. Visceral blood flow apparently increased while the peripheral vascular response to ischemia remained intact. Mannitol increased oxygen delivery and consumption in both the total body and limb. Thus in patients with adult respiratory distress syndrome (ARDS), increases in pulmonary blood flow can be achieved with several distinct pharmacologic agents without significant increases in intrapulmonary shunt. These increases in flow are generally accompanied by increases in oxygen delivery without increased pulmonary vascular resistance.
G. R. Rhodes, M.D., M. Taylor, M.D., J. C. Newell, Ph.D., D. M. Shah, M.D., W. A. Scovill, M.D., and S. R. Powers, M.D.,t Albany and Troy, N. Y.
Adult respiratory distress syndrome (ARDS) is commonly treated by the application of end-expiratory pressure either in the form of positive end-expiratory pressure (PEEP) 1 during mechanical ventilation or continuous positive airway pressure (CPAP)2 during spontaneous ventilation. End-expiratory pressure increases alveolar volume and usually results in increased oxygen delivery through a decrease in intrapulmonary
From the Departments of Physiology and Surgery, Albany Medical College, Albany, N. Y., and Centerfor Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, N. Y. Supported by the following grants: P-50-GM 15426, T-32GM07033, MO-I-RR-00749, National Institutes of Health. Received for publication May 21, 1979. Accepted for publication Feb. 3, 1981. Address for reprints: D. M. Shah, M.D., Department of Surgery, Albany Medical College, Albany, N. Y. 12208. Address for correspondence: J. C. Newell, Ph.D., Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, N. Y. 12181. tDeceased.
shunt. It is also possible to increase oxygen delivery by increasing pulmonary blood flow. This study examines the cardiopulmonary effects of three drugs, dopamine, an inotropic agent, ethanol, a vasodilator, and mannitol, an osmotic agent, in a series of patients in whom ARDS developed following trauma, operation, or sepsis. These drugs were investigated to determine if they could increase pulmonary blood flow and oxygen delivery without a significant increase in intrapulmonary shunt. The effects of mannitol, although previously published," are presented here to permit a comparison among all three drugs studied.
Methods-General Twenty patients were studied who had significant hypoxemia following trauma, surgical operation, or the development of sepsis (Table I). Informed consent was obtained either directly or by proxy, according to institutional guidelines. All patients had intrapulmonary shunts greater than 13%, together with radiologic findings of bilateral diffuse pulmonary infiltrates that
0022-5223/8I!080203+08$OO.80/0 © 1981 The C. V. Mosby Co.
203
The Journal of
204 Rhodes et at.
Thoracic and Cardiovascular Surgery
Table I Case No.
Age and sex
Injuries and operations
19, M
2
3
4 5
6
7
8
9
10
II
12 13 14 IS 16 17 18 19 20
Fracture of right and left fibula and tibia, comminuted fracture of left humerus, cerebral concussion, aspiration 21, M Fractured mandible, scalp and facial lacerations, traumatic abdominal aortic aneurysm necessitating resection and grafting, ruptured spleen necessitating splenectomy 58, M Fractures of left clavicle, right ribs 2-6, right tibia and fibula, right pulmonary contusion, scalp and facial lacerations; exploratory laparotomy with finding of mesenteric hematoma; cerebral concussion Blunt abdominal trauma with renal contusion and ruptured spleen 52, F necessitating splenectomy; multiple pelvic fractures 16, M Right hemopneumothorax, right radial and ulnar fractures, C.-C a fracture with paralysis and probable T 7 fracture; repair lacerated colon and lacerated duodenum 28, M Left parietal skull fracture and scalp laceration, right forearm lacerations, right femoral neck and midshaft fracture, right femoral supercondylar fracture, fractured left clavicle, and dislocated left ankle; exploratory laparotomy with cauterization of liver laceration 22, M Blunt abdominal injury with rupture of cisterna chyli, puncture of duodenum, retroperitoneal bleed; above-knee amputation of right leg, open debridement, and internal fixation of left leg 38, F Parietal scalp laceration, depressed skull fracture, facial fractures, fracture of both pubic rami, vaginal lacerations, partial transection of urethra, multiple rib fractures, fractured sternum, right pneumothorax Severe crush injury to lower pelvis and left leg with multiple fracIS, F tures of the left femur, destruction of the left acetabulum, crush injury to almost all left adductor and gluteal muscle groups, left sciatic nerve transection, devitalization of iliac crest, retroperitoneal hematoma 53, F Blunt abdominal injury necessitating removal of 50% of small bowel, mid-transverse colonic tear, right knee puncture necessitating open exploration, late ischemic injury to right forearm necessitating open exploration, late ischemic injury to right forearm necessitating above-elbow amputation of right arm 29, F Multiple rib fractures, facial fractures, rupture of stomach and duodenum with chemical peritonitis, crushed right kidney necessitating right nephrectomy, gastric resection, Biliroth II gastrojejeunostomy, and tube duodenostomy 64, M Bilateral fractured ribs, cerebral contusion, left lung contusion Perforated small bowel, intra-abdominal sepsis 62, F 70, M Ruptured abdominal aneurysm 56, M Common bile duct diversion; sepsis, dehiscence Exploratory laparotomy, lysis of adhesions, ileal resection; post48, F operative sepsis 6O,M Esophagectomy, colon interposition, partial liver resection Portocaval shunt, excessive intraoperative hemorrhage 51, F 71, M Exploratory laparotomy for subphrenic abscess, lysis of adhesions, continual sepsis Emergency portocaval shunt with hemorrhagic shock 45, F
Legend: ARDS, Adultrespiratory distress syndrome.
Principal cause of ARDS
Protocols
Mannitol
I
Ethanol
I
Sur-
Dopamine
vived
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
Trauma
+
+
+
Trauma
+
+
+
Sepsis
+
+
+
Trauma Sepsis Operation Sepsis Sepsis
+ + + + +
Operation Operation Sepsis
+ +
Operation
+ + + + + + +
+
Volume82 Number2
August. 1981
could not be ascribed to acute inflammatory disease, congestive heart failure, pneumonia, or large pulmonary emboli. The findings in these patients constitute a syndrome consistent with ARDS. Drugs were studied on separate days in almost all instances. Mannitol usually was evaluated before ethanol, and dopamine was evaluated only in patients with specific indications (described subsequently). Some drugs were not studied in certain patients because of the patient's clinical improvement or unstable clinical condition. The initial studies in all patients were performed within 48 hours of the development of hypoxemia. No patient was studied while in shock. In addition, no patient died within 3 days of the studies. All patients were studied while recumbent with the head elevated 30 degrees. Changes were considered statistically significant at a level of p < 0.05 according to Student's paired t test. A femoral venous catheter was inserted in all patients receiving dopamine or mannitol and in six of 10 patients receiving ethanol. An arterial catheter and a triple-lumen flow-directed pulmonary arterial catheter were placed in all patients. No attempt was made to determine the altitude of the tip of the Swan-Ganz catheter. Pressure transducers were zeroed to left atrial level, estimated to be 5 em below the angle of Lewis. Mean arterial (MAP), central venous (CVP), pulmonary artery (PAP), and pulmonary wedge (PWP) pressures were measured by strain-gauge transducers and recorded on a polygraph. Pressures were recorded at end-expiration. Cardiac output (CO) was determined by indocyanine green (Cardio-green) dye-dilution technique, and the mean of duplicate values is reported. Pulmonary vascular resistance (PVR) and systemic vascular resistance (SVR) were calculated as follows: PVR = (80 [PAP - PWP])/CO; SVR = (80 [MAP CVP])/CO. Arterial, mixed venous, and femoral venous blood samples were analyzed for P0 2, Pc0 2, and pH with standard electrodes and were corrected to body temperature. Intrapulmonary shunt was calculated from the shunt equation at the therapeutic inspired oxygen level. Hemoglobin concentration (Hb) was measured by spectrophotometric technique. Limb blood flow was measured in the calf with venous occlusion plethysmography and expressed in milliliters per 100 gm of tissue per minute." Limb blood flow was measured both at rest and 20 seconds after a 2 minute period of ischemia as a measure of the reactive hyperemic response. Ischemia was produced by inflating the blood pressure cuff above arterial systolic pressure. Total body oxygen consumption ("\10 2) and oxygen
Adult respiratory distress syndrome
205
delivery (02DTB) and limb oxygen consumption (VLo.) and oxygen delivery (02DL) were calculated as follows: 02D = Q (0.031 Pao. + 13.9 [Hb] [Sa o.ll00]) where Q is either cardiac output for total body or blood flow for the limb. Venous oxygen return = Q (0.031 Pv.; + 13.9 [Hb] Sv o./100]) and V0 2 = arterial oxygen delivery - venous oxygen return. For total body calculations, PVa. = PVa., while for the limb, PVa.was that in femoral venous blood. Patients were maintained on a volume-cycled ventilator with tidal volumes of 12 to 15 ml/kg, respiratory rates of 12 to 15 breaths/min, PEEP of 5 to 15 cm H20, and fractions of inspired oxygen of 0.40 to 0.55. All ventilatory parameters were held constant for the duration of the study. The PEEP level was chosen in each case to maximize oxygen delivery. Procedure Dopamine. Seven patients were studied, all of whom had been requiring dopamine to maintain mean arterial pressure above 70 mm Hg. They were studied with the drug infusion at a time when all but one had a mean arterial pressure less than 70 mm Hg. This patient had an unstable blood pressure necessitating intermittent dopamine support. Following baseline measurements with no drug infusion, a dopamine infusion (100 JLg/ml) was begun and increased at increments of 2 JLg/kg/min until either mean arterial pressure increased 15 mm Hg or heart rate increased by more than 15 beats/min. After 15 minutes had elapsed to allow achievement of a steady state, all measurements were repeated. Ethanol. Ten patients were studied. Pulmonary capillary blood was examined histologically for the presence of fat emboli during the first 3 days following admission to the trauma center. * This technique consists of the slow withdrawal of pulmonary capillary blood from the Swan-Ganz catheter with the balloon inflated, followed by sectioning of the clotted blood with a cryostat and staining for neutral fat. Serum ethanol and lactate were measured by standard techniques. Measurements were performed before and 1, 3, and 6 hours after the beginning of an infusion of 10% ethanol in 5% dextrose solution infused intravenously at 2 ml/kg/hr for 6 hours. Because of its diuretic effect, these patients required from 500 to 1,000 ml of Ringer's lactate solution as additional fluid over the 6 hour period to maintain the pulmonary capillary wedge pressure at control levels. *Dutton RE: Diagnosis of fat embolism with a Swan-Ganz catheter. Personal communication.
The Journal of Thoracic and Cardiovascular Surgery
206 Rhodes et al.
Table II. Cardiopulmonary effects of ethanol
_______________I'-----L.-. MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (Lzrnin/m") PVR (dyne-sec/em") SVR (dyne-sec/em") Arterial P0 2 (mm Hg) Mixed venous P0 2 (mm Hg) Pulmonary shunt (%) Total oxygen consumption (ml/Os/rnin) Total body oxygen delivery (m1/0 2/min) Limb blood flow, rest (ml/loo gm/min) Limb blood flow, ischemia (ml/IOO gm/min) Limb oxygen consumption (ml O2/ 100 gm/min) Limb oxygen delivery (mI/02/100 gm/min) Ethanol (mg/loo ml) Lactate (mg/loo ml)
Control
One hour
85 ± 4 4±1 19 ± 3 4±1 100 ± 7 3.84 ± 0.36 192 ± 27 1,IOO±110 98 ± II 40 ± 2 24 ± 4 221 ± 24 980 ± 100 0.9±0.13 2.68 ± 0.48 0.027 ± 0.006 0.16 ± 0.06
82 ± 4 5 ± I 20 ± 3 4 ± I 117 ± 7 4.56 ± 0.51 181 ± 37 920 ± 130* 93 ± 9 40 ± 2 27 ± 5 276 ± 29* 1,210± 140* 1.3 ± 0.45 2.02 ± 0.38 0.048 ± 0.015 0.26 ± 0.07 8.5 ± 1.2 23 ± 4
o
20 ± 3
.l...-
Three hours 74 ± 5 5 ± 3 20 ± 3 4 ± I 118 ± 5 4.52 ± 0.4* 187 ± 34 830 ± 140* 98 ± 12 36 ± 3 26 ± 6 284 ± 27* 1, 170 ± 120* 1.47 ± 0.39* 2.52 ± 0.69 0.06 ± 0.02 0.27 ± 0.07 12.1 ± 2.6* 25 ± 4
L..-
Six hours
_
77 ± 5 4±1 19 ± 3 4±1 125 ± 6* 4.28 ± 0.43 190 ± 32 930 ± 120 98 ± II 40 ± 6 26 ± 5 277 ± 29* 1,130 ± 130 1.16 ± 0.26 2.32 ± 0.65 0.037 ± 0.009 0.21 ± 0.04 12.0 ± 2.7* 25 ± 4
Legend: CI, Cardiac index. CVP, Central venous pressure. HR, Heart rate. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. 'p < 0.05; statistical significance based on comparison with control values.
Mannitol. Fourteen studies were performed in II patients. Studies were repeated in two patients at least 24 hours apart after the patient had a septic episode associated with hypotension and necessitating additional fluid therapy. Control measurements were performed before drug administration. Then 25 gm of 25% mannitol was infused as a single bolus, followed by the infusion of 8 to 25 gm of 20% mannitol (mean 10 ± 1.9 gm) as a continuous intravenous drip over I hour. All measurements were repeated during the last 10 minutes of this period. Results Dopamine. During dopamine infusion at rates between 4 and 15 ILg/kg/min, cardiac index increased from a baseline of 4.3 ± 0.7 to 5.7 ± 1.0 L'rnin/rn" (p < 0.05) (Table II). Oxygen consumption rose from 261 ± 34 to 305 ± 38 mllmin (p < 0.005). There were no significant changes in arterial, mixed venous, or femoral venous P0 2 or oxyhemoglobin saturation. Intrapulmonary shunt remained unchanged. Central venous and pulmonary wedge pressures did not change significantly. Mean heart rate rose from 106 ± 2 to 114 ± 5 beats/ min, but this change was not statistically significant. Despite a significant rise in pulmonary artery pressure, 21 ± 2 to 24 ± 2 mm Hg (p < 0.01), the pulmonary vascular resistance fell from 200 ± 50 to 170 ± 50 dyne-sec/em" (p < 0.005).
Limb blood flow, limb oxygen consumption and delivery, and the postischemic hyperemic response were unchanged by dopamine infusion. Ethanol. Following I hour of ethanol infusion, total body oxygen delivery increased significantly from 980 ± 100 to 1,200 ± 100 ml/Oj/rnin associated with an increase in oxygen consumption from 220 ± 20 to 290 ± 30 mIl0 2/min (Table III). Systemic vascular resistance decreased as a result of the combined effects of increasing cardiac output and decreasing mean arterial pressure (Table III). All other parameters remained unchanged at I hour. By 3 hours, there were significant increases in cardiac index, accompanied by increases in total body oxygen delivery and oxygen consumption. The increase in cardiac index was due principally to the increase in heart rate, from a mean 110 ± 10 beats/min at control to 125 ± 10 beats/min at 6 hours. There were no significant changes in any pressure measured throughout the study, although mean arterial pressure tended to decrease from 84 ± 4 mm Hg at control to 76 ± 5 mm Hg at 6 hours. Neither pulmonary vascular resistance nor intrapulmonary shunt changed significantly: however, increases in intrapulmonary shunt and pulmonary vascular resistance occurred in four of five patients whose initial intrapulmonary shunt exceeded 20%, Stains of pulmonary capillary blood from two of the 10 patients were positive for fat. These two patients had multiple long bone fractures. Their initial intrapul-
Volume 82 Number 2 August, 1981
Adult respiratory distress syndrome
Table III. Cardiopulmonary effects of dopamine
MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (Lzrnin/m") PVR (dyne-see-em") SVR (dyne-see-em') Arterial P0 2 (mm Hg) Mixed venous P0 2 (mm Hg) Femoral venous P0 2 (mm Hg) Hemoglobin (gm/loo ml) Pulmonary shunt (%) Total body oxygen consumption (ml 02/min) Total body oxygen delivery (ml Oj/rnin) Limb blood flow, rest* (ml/ 100 gm/min) Limb blood flow, ischemia" (mllloo gm/min) Limb oxygen consumption, rest (ml 02/min/100 gm) Limb oxygen delivery, rest (ml 02/min/100 gm)
I
Control 63 9.0 21.4 5.9 106 4.3 200 690 98 43 44 11.9 26.0 261
± ± ± ± ± ± ± ± ± ± ± ± ± ±
4 2.0
I.7 1.3 2 0.7 50 104 9 4 4 0.5 5.0 34
1,170 ± 210
I
During dopamine 78 10.0 23.7 8.0 114 5.7 170 740 97 43 45 11.9 29.0 305
± ± ± ± ± ± ± ± ± ± ± ± ± ±
4t 2.0 2.2t 1.7 5
i.ot
50t 170 9 3 3 0.5 6.0 38t
1,600 ± 320t
1.02 ± 0.16
1.16 ± 0.24
1.84 ± 0.42
1.97 ± 0.50
0.036 ± 0.008
0.035 ± 0.007
0.16 ± 0.03
0.19 ± 0.05
Legend: CI, Cardiac index. CVP, Central venous pressure. HR, Heart rate. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. • p < 0.05 comparing control limb blood flow at rest versus control limb blood flow after ischemia; p < 0.05 comparing effect of dopamine on limb blood flow at rest versus limb blood flow after ischemia. t p < 0.05; statistical significance based on comparison with control values.
monary shunts were 20% and 23%, which were not markedly different from the mean shunt of 24%. Changes in cardiopulmonary parameters following ethanol infusion in these two patients were in the same direction as in the whole group. We found no fat in the pulmonary capillary blood of two other patients in this group with long bone fractures. Reactive hyperemia was present before drug infusion, since limb blood flow after ischemia significantly exceeded that at rest; however, at 3 and 6 hours, reactive hyperemia was abolished, as indicated by similar values of limb blood flow after ischemia and at rest (Table III). Serum lactate levels were unchanged by ethanol (Table III). Mannitol. Mannitol produced a significant increase in cardiac output (p < 0.005) without changes in either right (central venous) or left (pulmonary capillary wedge) side filling pressure (Table IV). There was no significant change in either mean arterial or pulmonary
207
Table IV. Cardiopulmonary effects of mannitol _ _ _ _ _ _ _ _ _ _1
MAP (mm Hg) CVP (mm Hg) PAP (mm Hg) PWP (mm Hg) HR (beats/min) CI (liters/rnin/m") PVR (dyne-sec/em') SVR (dyne-sec/em') Arterial P0 2 (mm Hg) Mixed venous P0 2 Femoral vein P0 2 (mm Hg) Hemoglobin (gm/ I00 ml) Pulmonary shunt (%) Total body oxygen consumption (ml 02/min) Total body oxygen delivery (rnl Og/min) Limb blood ftlow, rest* (ml/IO) gm/min) Limb blood flow ischemia" (ml/lOO gm/min) Limb oxygen consumption, (ml/loo gm/min) Limb oxygen delivery, (mil 100 gm/min) Limb oxygen eonsumptiont (mllIOO gm/min)
Control 81 8 20 7 107 4.21 163 888 104 40 40 11.6 22.5 261
± 4 ± I ± I ± I ± 6 ± 0.36 ± 23 ± 98 ± 7 ± 2 ± 3 ± 0.5 ± 2.1 ± 25
1,200 ± 105
1 During mannitol
81 8 21 8 113 4.94 133 745 101 41 40 11.4 24.2 298
± 4 ± I ± I ± I ± 6 ± 0.39* ± 18 ± 73* ± 7 ± I ± 3 ± 0.4 ± 2.2 ± 27*
1,350 ± 110*
1.06 ± 0.23
1.20 ± 0.22
2.02 ± 0.49
2.12 ± 0.46
0.040 ± 0.007
0.043 ± 0.005
0.17 ± 0.03
0.19 ± 0.04
0.029 ± 0.006
0.042 ± 0.004*
Legend: CI, Cardiac index. CVP, Central venous pressure. MAP, Mean arterial pressure. PAP, Pulmonary artery pressure. PWP, Pulmonary wedge pressure. HR, Heart rate. PVR, Pulmonary vascular resistance. SVR, Systemic vascular resistance. LBF, Limb blood flow. * P < 0.01 comparing control limb blood flow at rest versus control limb blood flow after ischemia; p < 0.05 comparing the effect of mannitol on limb blood flow at rest versus limb blood flow after ischema. t These values represent changes in limb oxygen consumption within the nine patients in whom limb oxygen delivery increased. *p < 0.05; statistical significance based on comparison with control values.
artery pressure. Systemic vascular resistance fell significantly without a change in pulmonary vascular resistance. Total body oxygen delivery increased significantly, without an increase in pulmonary shunt. Arterial and mixed venous blood gases were unchanged by mannitol infusion. This increase in total body oxygen delivery was seen in all but two patients studied. Among the exceptions, there was a decrease in cardiac output in one and a decrease in hemoglobin concentration in the other. There was a significant increase in total body oxygen consumption. Mean limb blood flow and limb oxygen delivery and consumption increased in eight of the nine studies in which blood flow increased. Chi square analysis demonstrated that oxygen consumption and delivery changed in the same direction (X 2 = 7 [p < 0.05]) for the total body and (X 2 = 9.5 [p < 0.01]) for the limb.
The Journal of
208
Rhodes et al.
Thoracic and Cardiovascular Surgery
Discussion
The major emphasis of therapy for ARDS has been the use of increased inspired oxygen together with PEEP or CPAP to decrease intrapulmonary shunting, raise arterial Po 2, and thereby increase oxygen delivery to the tissues. These techniques tend to minimize the effect of ventilationlperfusion inequalities, a significant cause of pulmonary shunting. In contrast, the studies presented here suggest that a pharmacologic approach may be used also to increase oxygen delivery by increasing pulmonary perfusion without causing a significant increase in shunt. In most patients studied, any of these increases in oxygen delivery were associated with significant increases in oxygen consumption for total body and limb. Although pulmonary vascular resistance was initially high in most patients, increased pulmonary blood flow did not result in additional increases in pulmonary vascular resistance with any drug studied. A strict comparison of the three drug groups is complicated by several factors. Not all patients were studied with each drug because of clinical instability, rapid clinical improvement, or technical problems. There were varying causes for the development of ARDS both within and between groups. Chest roentgenograms indicative of an increase in interstitial and alveolar fluid were common in all patients. Initial central venous and pulmonary wedge pressures were lower in the ethanol group than in the other groups, for unknown reasons. Despite these differences among groups, the cardiorespiratory responses to the three drugs were similar. Dopamine. In our hypotensive patients with ARDS, dopamine increased cardiac output and reduced pulmonary vascular resistance, as it does in most patients with cardiac disease. Intrapulmonary shunt and limb blood flow remained unchanged. The latter suggests that the increased cardiac output predominantly perfused central organs. In some animal models'': 6 and groups of patients such as those with sepsis," higher doses of dopamine may not reduce pulmonary vascular resistance. However, four of seven of the patients in our study were in a septic condition. Nevertheless, the significant increase in limb blood flow following ischemia, both before and during dopamine infusion, suggests that the ability of muscle capillary beds to dilate in response to an ischemic stimulus is not adversely affected by dopamine at the moderate concentrations employed. In previous studies of the effects of dopamine on cardiopulmonary function in different patient populations,8-13 cardiac output has increased, but the changes in pulmonary vascular resistance have been variable. In
a study of critically ill patients who were in a septic condition, dopamine infusion at rates of 2 to 55 JLgl kg/min produced a marked increase in mean arterial pressure, cardiac output, and pulmonary artery pressure but no significant changes in systemic or pulmonary vascular resistance." Most patients in that study were receiving dopamine when baseline studies were performed and measurements were repeated at a higher dosage. In contrast, our patients received no dopamine during the control measurements. In patients with congestive heart failure secondary to cardiac disease, a significant rise in mean arterial pressure and stroke volume was found at infusion rates greater than 5 JLg/kg/min. However, no significant changes in pulmonary vascular resistance or pulmonary artery, pulmonary wedge, or central venous pressure were noted. None of their patients were described as having abnormal pulmonary function and all had higher mean arterial and pulmonary artery pressures than the patients in the present study. In contrast, dopamine decreased pulmonary vascular resistance in postoperative patients undergoing cardiac bypass, II in patients in cardiogenic shock? resulting from myocardial infarction, and in patients undergoing cardiac catheterization.l? Ethanol. Ethanol is a potent vasodilator used to treat ischemic vascular disease. 14. IS Its use has been proposed in the therapy of pulmonary fat embolism.v" " either because of its possible role as a pulmonary vasodilator or as a specific treatment of ARDS associated with fat embolism. IS However, there is little physiological evidence for the use of ethanol in treating ARDS with or without associated fat embolism. This study evaluates the ability of low doses of intravenous ethanol to increase oxygen delivery and consumption in patients with ARDS of multiple origins, including fat embolism and sepsis. The effect of ethanol on cardiopulmonary function has been studied in animals.w?' normal subjects ,22. 23 and patients with chronic obstructive lung disease and coronary artery disease. Several investigations have demonstrated its effect as a systemic vasodilator. 14. IS. 22-24 These investigations, which have been performed in spontaneously breathing subjects, have generally not addressed changes in pulmonary blood flow, oxygen consumption, and oxygen delivery. In studies of the effect of ethanol on cardiac function.':" 19-24 the doses of ethanol used were as high as ten times those used in this study, so that comparison is difficult. The route and concentration of ethanol used to evaluate cardiovascular function may be important. Our study evaluated the effects of a relatively low dose of ethanol, which has been reported." to produce sig-
Volume82 Number 2 August. 1981
nificant systemic vasodilation without marked impairment in cardiovascular function. Turnipseed, Vasko, and Evans!" used a dose of 10% ethanol infused at 2 ml/kg/hr to produce a decrease in blood volume of 17 ml/kg. We tried to prevent a blood volume decrease in the present study by infusing a balanced salt solution at rates sufficient to keep the pulmonary wedge pressure constant. The volume of excess intravenous solution required was about the same as the decrease in blood volume reported in the earlier study. 15 This added fluid, which kept wedge pressure at a steady value in our study, may explain why we found an increased cardiac output without a fall in mean blood pressure; by contrast, in previous studies, cardiac output and/or mean blood pressure decreased following ethanol administration.I9-21. 23 Pulmonary vascular resistance was above normal levels at the beginning of this study and remained unchanged with ethanol infusion despite the increase in pulmonary blood flow. It is possible that ethanol may have increased blood flow through pulmonary capillaries that were either fully or partially closed by vasoconstriction, microemboli, or other mechanisms. Although the mean change in pulmonary shunt was not statistically significant, increases in shunt were frequently seen in patients with shunts initially greater than 20%. It remains uncertain whether the increase in pulmonary blood flow occurred through previously patent pulmonary gas exchange units with adequate ventilation or whether a shunt increase may have occurred in some of these lungs. Mannitol. In this study, infusion of 25 gm of hypertonic mannitol significantly increased cardiac output without increasing pulmonary shunt or pulmonary vascular resistance. These results differ from the immediate effect of a 25 gm bolus injection of hypertonic mannitol, which decreased pulmonary vascular resistance as well as its vasodilator effects seen in the kidney, heart, brain, and lung. 25-29 The mechanism for this reduced vascular resistance may involve the osmotic removal of water from endothelial or adjacent cells. The increase in total body oxygen delivery associated with an increase in oxygen consumption earlier reported in post-trauma patients I is confirmed here in a more diverse group of critically ill patients. In addition, when limb blood flow increases following administration of hypertonic mannitol, there is the same pattern of increased oxygen consumption and delivery for the limb as is found in the total body. Thus in patients with ARDS, including those requiring cardiovascular support, increases in pulmonary blood flow were achieved with several distinct pharma-
Adult respiratory distress syndrome
209
cologie agents without significant increases in intrapulmonary shunt. These increases in flow were generally accompanied by increases in oxygen delivery without increased pulmonary vascular resistance. On the contrary, decreases in pulmonary vascular resistance were seen with dopamine and mannitol. In critically ill patients, increases in oxygen delivery may be associated with significant increases in oxygen consumption for all three drugs studied. REFERENCES Powers SR, Mannal R, Neclerio M, English M, Marr C, Leather R, Udea H, Williams G, Custead W, Dutton R: Physiologic consequences of positive end-expiratory pressure (PEEP) ventilation. Ann Surg 178:265-272, 1973 2 Shah DM, Newell JC, Dutton RE, Powers SR Jr: Con-
tinuous positive airway pressure versus positive end-expiratory pressure in respiratory distress syndrome. J THORAC CARDIOVASC SURG 74:557-562, 1977 3 Rhodes GR, Newell JC, Shah DM, Scovill W, Tauber J, Dutton RE, Powers SR Jr: Increased oxygen consumption accompanying increased oxygen delivery with hypertonic mannitol in adult respiratory distress syndrome. Surgery 84:490-497, 1978 4 English MC, Lozman J, Powers SR Jr: A refined technique for the determination of limb blood flow by venous occlusion impedance plethysmograph. Proceedings of the Second Annual New England Bioengineering Conference, Worcester, Mass., 1974 5 Kaumann AJ, Ochoa E: Similar receptors mediating norepinephrine and dopamine contraction in main pulmonary artery of dog. Acta Physiol Lat Am 20:451, 1970 6 Mentzer RM Jr, Alegre C, Nolan SP: The effects of dopamine and isoproterenol on the pulmonary circulation. J THoRAc CARDIOVASC SURG 71:807-814, 1976 7 Wilson RF, Sibbalb WJ, Jaanimagi JL: Hemodynamic effects of dopamine in critically ill septic patients. J Surg Res 20:163-172, 1976 8 Beregovich J, Bianchi C, Rubler S, Lomnitz E, Cagin N, Levitt B: Dose-related hemodynamic and renal effects of dopamine in congestive heart failure. Am Heart J 87: 550-557, 1974 9 Flammang D, Sebastian PH, Bouvain Y: Etude hemodynamique de la dopamine employee dans les cardiopathies chroniques et dans la choc cardiogenique compliquant I'infarctus aigu du myocarde. Ann Anesthesiol Fr 16:669-672, 1975 10 Harrison DC, Pirages S, Robison SC, Wintroub BU: The pulmonary and systemic circulatory response to dopamine infusion. Br J Pharmacol 37:618-626, 1969 II Holloway EL, Stinson EB, Derby GC, Harrison DC: Action of drugs in patients early after cardiac surgery. I. Comparison of isoproterenoland dopamine. Am J Cardiol 35:656-659, 1975
2 10
The Journal of Thoracicand Cardiovascular Surgery
Rhodes et al.
12 Loeb HS, Winslow EB, Rahimtoola SH, Shahbudin H, Rosen KM, Gunnar RM: Acute hemodynamic effects of dopamine in patients with shock. Circulation 44: 163-173, 1971 13 Talley RC, Goldberg LI, Johnson CE, McNay JL: A hemodynamic comparison of dopamine and isoproterenol in patients with shock. Circulation 39:361-378, 1969 14 Cook EN, Brown GE: Vasodilating effects of ethyl alcohol on peripheral arteries. Mayo Clin Proc 7:449-452, 1932 15 Turnipseed W, Vasko JS, Evans WE: Hemodynamic effects of intravenous alcohol on patients with ischemic limb disease. J Surg Res 20:477-480, 1976 16 Herrmann LG: Effects of dextrose-alcohol mixture upon pulmonary fat embolism. Proc Soc Exp Bioi Med 30:558-559, 1933 17 Yale CE, Herrmann LG: Pathologic physiology of pulmonary fat embolism. Mechanical and chemical alterations with an effective antidote. Surg Forum 12:451-453, 1961 18 Alder F, Lai SP, Peltier LF: Fat embolism. Prophylactic treatment with lipase inhibitors. Surg Forum 12:453-455, 1961 19 Liedtke AJ, DeMuth WE: Effects of alcohol on cardiovascular performance after experimental nonpenetrating chest trauma. Am J CardioI35:243-250, 1975 20 Gettler DT, Allbritten FF Jr: Effect of alcohol intoxication on the respiratory exchange and mortality rate associated with acute hemorrhage in anesthetized dogs. Ann Surg 158:151-158, 1963 21 Moss LK, Chenault OW Jr, Gaston EA Jr: The effects of
22 23
24
25
26
27
28
29
alcohol ingestion on experimental hemorrhagic shock. Surg Forum 10:390-392, 1960 Juchems R, Klobe R: Hemodynamic effects of ethyl alcohol in man. Am Heart J 78:133-135, 1969 Riff DP, Jain AC, Doyle JT: Acute hemodynamic effects of ethanol on normal human volunteers. Am Heart J 78:592-597, 1969 Horwitz 0, Montgomery H, Longaker ED, Sayen A: Effects of vasodilator drugs and other procedures on digital cutaneous blood flow, cardiac output, blood pressure, pulse rate, body temperature, and metabolic rate. Am J Med Sci 218:669-682, 1949 Moore FD, Lyons JH, Pierce EF Jr, Morgan AP, Drinker PA, MacArthur JD, Mannin GJ: Post traumatic respiratory insufficiency. Philadelphia, 1969, W. B. Saunders Company Powers SR Jr, Ryon D, Newell JC, Ralph C, Scovill W, Dutton RE: Hypertonic mannitol in the therapy of the acute respiratory distress syndrome. Ann Surg 185:619625, 1977 Ames A III, Wright RL, Kowada M, Thurston JM, Majno G: Cerebral ischemia. II. The no-reflew phenomenon. Am J PathoI53:437-453, 1968 Hostnik WJ, Powers SR Jr, Boba A, Stein AA: Observations on the effect of mannitol on renal hemodynamics and O 2 tension in the urine and renal vein. Surg Forum 10:872-875, 1960 Powell WJ Jr, DiBona DR, Flores J, Frega N, Leaf A: Effects of hyperosmotic mannitol on reducing ischemic cell swelling and minimizing myocardial necrosis. Circulation 53:Suppl 1:45-49, 1976