JOURNAL

OF SURGICAL

RESEARCH

50,344-349

(19%)

The Effect of Low Dose Dopamine on Gut Hemodynamics during PEEP Ventilation for Acute Lung Injury DANIEL

J.

M.D., JAY A. JOHANNIGMAN, DAVIS, JR., M.D., AND

JOHNSON,

KENNETH University Presented

of Cincinnati, at the Annual

Department Meeting

of Surgery,

of the Association

Division

and Critical

Surgery,

Houston,

D. BRANSON, M.D. Care,

Cincinnati,

Texas,

R.R.T., Ohio 45267

November

14-17,

1990

32% with 10 cm H,O PEEP [l]. Clinical and animal studies have characterized this pressure-related drop in splanchnic perfusion, which may ultimately have an effect on organ function [l-4]. Gut ischemia has been implicated in the development and propagation of the multiple organ failure syndrome [5]. Prolonged ischemia may hasten translocation of bacterial products across the mucosal barrier. Mechanical ventilation with PEEP may, therefore, aggravate preexisting gut ischemia brought on by shock or operative trauma. Although volume loading can offset some, but not all, of the PEEP-induced changes in gut and liver perfusion, other methods of preventing or ameliorating further gut ischemia need to be investigated. This study was designed to examine the effects of pharmacologic manipulation of PEEP-induced gut hypoperfusion with low dose dopamine. Dopamine, given systemically in doses below 8 pg/kg/min, is a known splanchnic vasodilator [6]. The effect of dopamine on gut hemodynamics in the PEEP-created, flow-limited environment is unknown. MATERIALS

AND

METHODS

Animal Care and Preparation

INTRODUCTION

In the early postinjury period, acute respiratory failure often complicates the management of the injured patient. Mechanical ventilation with positive pressure ventilation and positive end-expiratory pressure (PEEP) are necessary to reduce intrapulmonary shunt and maintain adequate oxygenation. PEEP, however, reduces cardiac output under isovolemic circumstances by increasing mean airway pressure. Additionally, there is a reduction in portal and total liver blood flow corresponding to the reduction in cardiac output. Portal flow may be reduced as much as 344 Inc. reserved.

of Trauma

for Academic

Mechanical ventilation with positive end-expiratory pressure (PEEP) diminishes gut and hepatic blood flow and redistributes cardiac output away from the splanchnic circulation. This flow-limited environment can aggravate underlying hypoperfusion and &hernia in the postinjury setting. To examine the effects of low dose dopamine on a lung injury PEEP model of gut hypoperfusion, six anesthetized, splenectomized canines were instrumented with arterial, pulmonary artery, portal vein, and hepatic vein catheters. Electromagnetic flow probes were placed around the hepatic artery and portal vein for continuous flow measurements. Gut and hepatic blood flow, oxygen delivery, oxygen consumption, and extraction ratio were calculated at four time points: baseline, 1 hr after lung injury with oleic acid, 1 hr after ventilation with 10 cm H,O PEEP, and 1 hr after the continuous infusion of dopamine. Portal flow and gut oxygen delivery fell significantly with the infusion of PEEP. These values returned to near baseline levels with the addition of dopamine. Gut oxygen extraction increased from 16 f 2% to 35 + 3% with PEEP but returned to near baseline with dopamine (20 +- 4%, P < 0.01 compared to PEEP). We conclude that dopamine improves blood flow and oxygen delivery to the gut in this flow-limited model. This may preserve splanchnic physiology during PEEP ven0 1991 Academic Press, IN. tilation for acute lung injury.

0022-4804/91$1.50 Copyright 0 1991 by Academic Press, All rights of reproduction in any form

M.D., RICHARD M. HURST,

JAMES

Six mongrel dogs were used for this study. Animal subjects were cared for in accordance with the institutional animal care and use committee guidelines. These animals were male or nonpregnant females weighing between 20 and 30 kg. They were housed in individual cages until use. All animals were observed for 3 to 5 days and found to be free of infestation. Food was removed at 5 PM the night before the study, but the animals were allowed free access to water. Model Preparation On the morning of the study, animals were anesthetized with pentobarbital (30 mg/kg, intravenous) and intubated with a cuffed 8.0 mm o.d. endotracheal tube. They were then positioned supine on an operating table and placed on a volume-cycled ventilator with 40% in-

JOHNSON

ET

AL.:

DOPAMINE

AND

GUT

345

HEMODYNAMICS

Lung injury was induced by bolus infusion of 0.08 ml/ kg oleic acid into the central vein lumen of the thermodilution catheter. This produced a consistent degree of hypoxemia 60 to 90 min after administration of the oleic acid. At this point (T = 1) all measurements and blood samples were repeated. Inspired oxygen maintained at 40%. PEEP was added in 2.5-cm H,O increments until P,O,/F,O, > 200 (QJQ, < 20%). PEEP averaged 8-12 cm H,O. Tidal volume was adjusted to maintain normocarbia. All animals were ventilated at this level of PEEP for 1 hr at which time a third set of measurements was made (T = 2). After T = 2 an infusion of dopamine hydrochloride (5 pg/kg/min) was initiated and continued for 1 hr. The final set of measurements was made (T = 3).

spired oxygen (FiO,), a respiratory rate of 16, and a tidal volume of 15 cc/kg. No PEEP was used initially. Tidal volume was adjusted initially to maintain a normal pH and PCO,. Supplemental pentobarbital was given intravenously throughout the study to prevent spontaneous respirations. A 20-gauge polyethylene catheter was then placed percutaneously in the femoral artery for continuous blood pressure measurements and arterial sampling. A 7.5-F introducer was then placed percutaneously into the femoral vein through which a 7-F balloon-tipped flow-directed thermodilution catheter was advanced. The proximal and distal lumens in this catheter were connected to transducers. Pressure tracings were continuously monitored on a six-channel monitor-recorder. The balloon tip was flow-directed through the right heart and into the pulmonary artery (PA) until satisfactory PA, wedge (PCWP), and central venous pressure (CVP) tracings were obtained. Intravenous sodium chloride (0.9%) was administered via the sheath introducer at 8 cc/kg/hr during the entire study. A midline laparotomy was performed and the spleen was removed. A 16-gauge polyethylene catheter was placed into the portal vein via the splenic vein stump and the tip positioned at the portal vein bifurcation. A second catheter was placed in the left lateral hepatic vein through the liver parenchyma and secured to the liver with a silk suture. These lines were flushed with heparinized saline and used for blood sampling and pressure monitoring. The hepatic artery and portal vein were carefully isolated to prevent damage to the perivascular neural plexus. Electromagnetic flow probes were placed around these vessels and connected to flow meters (Carolina Medical Instruments, King, NC). Probes were factory calibrated and zero flow references were obtained in saline and in vivo at the beginning and end of each experiment. The gastroduodenal artery was ligated to isolate hepatic flow for accurate measurement. The abdomen was then loosely closed to allow access to the probes and venous catheters and to prevent increases in intraabdominal pressure.

Cardiac output, hepatic artery, and portal vein blood flows were indexed and expressed per kilogram of animal weight. Oxygen content on all samples was calculated according to the following equation: (Saturation (%) X Hgb X 1.39) + (PO, X 0.003). Systemic oxygen delivery (SO,D) was calculated as the product of arterial oxygen content and cardiac index (C,O, X CI). Systemic oxygen consumption (SO,V) was calculated as the product of arteriovenous oxygen content difference and cardiac index [ (C,O, - C,O,) X CI]. Gut oxygen delivery (GO,D) was determined as the product of portal blood flow index (PFI) and C,O,. Gut oxygen consumption (GO,V) was calculated as the product of portal flow index and arterial-portal venous oxygen content difference [(C,O, - C,O,) X PFI]. Hepatic oxygen delivery (HO,D) was determined as the sum of hepatic artery delivery (hepatic artery flow index HFI X C,O,) and portal vein delivery (PFI X C,,O,>. Hepatic oxygen consumption (HO,V) was calculated by subtracting hepatic venous oxygen effluent (total hepatic flow index x C,, 0,) from HO,D. Splanchnic oxygen consumption, therefore, was the sum of GO,V and HO,V.

Experimental

Statistics

Protocol

All animals were allowed to equilibriate for 30 min after surgical preparation. At this point (7’ = 0), the following set of measurements were taken: CVP, PCWP, PA pressure, hepatic vein pressure, portal vein pressure, and mean arterial pressure. Cardiac output was determined (x3) by thermodilution technique and hepatic artery end portal vein flows were recorded as the mean flow over two respiratory cycles. Mean airway pressure and peak inspiratory pressure were measured at the endotracheal tube. Blood was drawn from the femoral artery, pulmonary artery, portal vein, and hepatic vein for blood gasses (ABL-30, Radiometer America, Inc., Westlake, OH) and cooximeter panel (OSM3, Radiometer America, Inc.).

Calculations

The four time points for each variable were examined by repeated measure analysis of variance. Significant differences were then subjected to paired t tests for comparisons between time points. RESULTS

Hemodynamic Data Sixty to ninety minutes after bolus infusion of oleic acid, a significant fall was noted in cardiac index and mean arterial pressure (Table 1). CVP and PCWP remained unchanged. After 1 hr of ventilation with 10 cm H,O of PEEP (T = 2), cardiac index was reduced further

346

JOURNAL

OF

SURGICAL

RESEARCH:

VOL.

TABLE

Cardiac index (ml/min/kg) Mean arterial pressure (mm I-k) CVP (mm Hg) PCWP (mm Hg) Heart rate (bpm) Portal pressure (mm Hd Hepatic vein pressure (mm Hd * P < 0.05 compared y P < 0.05 compared

Data

1991

(n = 6)

T=l lung injury

After

4, APRIL

1

Hemodynamic T=O Baseline

50, NO.

After

T=2 PEEP

T=3 dopamine

After

166 -t 11

125 ? 8*

103 r 9*

118 + 7**#

156 + 7 721 9fl

114 + 8* 7fl 10 & 2

114 + 5* 11+ 1* 13 + 2*

107 + 9* 13 * 1* 15 + 2*

117 f 3

98 + 9

114 + 10

145 _t 13**”

7*1

7&l

10 + 1*

12 + le.”

4fl

321

7 + 1*

9 + 1-t

to T = 0 by paired to T = 2.

t test.

without a change in arterial pressure. Filling pressures increased to an appropriate level corresponding to the increase in airway pressures. The addition of low dose dopamine provided a slight (15%), but significant, increase in cardiac index, little change in arterial pressure, and a marked increase in heart rate. CVP and PCWP changed little with this intervention.

changed. Low dose dopamine (T = 3) increased P,O, compared to T = 2, increased mixed venous PO, slightly, and increased portal venous PO, by 33%. Hepatic venous PO, and saturation did not change significantly during this intervention.

Blood Gas Data

Hepatic artery blood flow changed little with acute lung injury, and the addition of PEEP produced a slight decrease which was not significant compared to baseline (Table 3). The administration of low dose dopamine (T = 3) had no effect on hepatic artery flow compared to PEEP alone (7’ = 2). On the other hand, portal flow decreased significantly at T = 2 compared to baseline, but the addition of dopamine returned portal blood flow

The administration of oleic acid produced a lung injury which resulted in a dramatic fall in P,O, (Table 2); P,CO, and pH remained constant. Mixed, portal, and hepatic venous PO, and saturation also fell. The addition of PEEP (T = 2) produced an increase in P,O, while mixed, portal, and hepatic venous PO, remained un-

Blood Flow Data

TABLE Blood T=O Baseline Arterial pH Arterial PCO, (mm Hg) Arterial PO, (mm Hg) Saturation (S,O*) Mixed venous PO, (mm Hd Saturation (S,O,) Portal venous PO, (mm I-k) Saturation (S,O,) Hepatic venous PO, (mm I-k) Saturation C&O,) * P < 0.05 compared #P