In-house versus on-call attending trauma surgeons at comparable level I trauma centers: a prospective study.

BACKGROUND: The purpose of this study was to prospectively compare patient outcomes based on the presence of in-house versus on-call attending trauma surgeons at comparable Level I trauma centers. METHODS: Two designated Level I trauma centers agreed to prospectively review trauma admissions over a 6-month period, one institution with 24-hour in-house trauma attending surgeons (IH), and the other with trauma-attending surgeons taking call from home (OC) available to the hospital within 15 minutes of notification. A 6-month prospective study was conducted reviewing all trauma patients admitted to both trauma centers with an Injury Severity Score > or =16. Comparisons were made between institutions utilizing admission demographics, clinical presentation, times to clinical care, and mortality rates. RESULTS: In comparison, OC and IH institutions were distinctly different in geographic environment, size, and number of patients admitted. As a group, IH patients were significantly older, with higher Injury Severity Scores and lower Glasgow Coma Scale scores than the OC group. In all comparisons, OC trauma attending surgeons responded to the trauma room with equal speed or more rapidly when compared with IH trauma attending surgeons. There were no other significant differences in either population in times to provision of clinical care or in clinical outcome. CONCLUSION: The ability of the OC institution to be similar to the IH institution in its provision of clinical care and mortality rate is accomplished in an environment where trauma attending surgeons live within a 15-minute response time to the trauma center. Using a voice-paged trauma alert activation with accurate information and sufficient warning, evaluation, provision of care, and clinical outcome of the acutely injured patient can be provided equally by in-house trauma attending surgeons and trauma attending surgeons on-call from home.

Leukocyte responses to hypoxic/ischemic conditions.

Ischemia is an interruption of oxygen and nutrient supply to a determined area of tissue for a period of time. Because of the heterogeneity of various tissues with regard to their microvascular flow reserve and oxidative capacity, as well as their markedly different metabolic needs, a single critical Po2 level below which ischemia occurs is unlikely. This is why there are variations of tolerance to hypoxia within and among organs. In general, when Pao2 reaches approximately 5 torr there is already evidence, in some organs, of altered cellular energetics. In addition, cessation of flow impairs the incoming transfer of nutrients such as glucose, and cells must depend on their own intracellular stores of carbon radicals, if available. Epidemiologic data suggest that there are deleterious effects of hypoxia on the immune system and that these effects result in increased susceptibility to infection. The histology of ischemic tissues demonstrates intravascular neutrophil (PMN) accumulation, vascular damage, and increased vascular permeability. Expression of PMN adhesion receptors is increased when oxygen is nearly completely removed from the medium. Expression of integrins on the cell surface is regulated by intracellular calcium; hypoxia causes a sustained and prolonged increase of intracellular calcium levels. Because both granule movement and functional expression of adhesion receptors on the cell surface are important in leukocyte motility, chemotaxis, and phagocytosis, these functions may be impaired by hypoxia. Exposure of a human macrophage cell line to nonlethal levels of hypoxia causes in vitro release of significant amounts of biologically active cytokines tumor necrosis factor (TNF) alpha, interleukin (IL)-1 and IL-8, as well as expression of intercellular adhesion molecule-1 and bound and soluble receptors for TNF alpha. Hypoxia markedly decreases T-lymphocyte IL-2 messenger RNA, a key cytokine responsible for B-cell proliferation and immunoglobulin secretion.

Should critical care medicine be a formal part of the undergraduate curriculum?

Medical students were surveyed shortly after completing the third year of medical school. TIle survey was designed to identify those areas of critical care medicine students had been exposed to and expressed interest in learning more about. In addition, the surveys sought to discern the level of confidence students felt with respect to different critical illnesses and intensive care unit (lCU) therapeutic modalities.Finally, the students were asked their opinion regarding the possibility or need for critical care medicine as pan of their medical school curriculum.The three most common topics of interest among medical students who had recently ftnished their third year in medical school were shock, hemodynamic monitoring, and mechanical ventilation. Less than 30% of the students surveyed felt "better-than-average" confidence on anyone of a number of critical care topics and treatment modalities. Of the 80% of students (n = 70) who completed the survey, 91% (n = 64) felt that critical care medicine should be made a pan of the medical school curriculum,6% (n = 4) felt it should not, and 3% (n = 2) were undecided.TIle survey results and the finding that most of the relevant literature acknowledges the need for critical care medicine in medical school has led us to conclude that a national core clerkship or a didactic lecture series in critical care medicine should be carefully designed and implemented into the undergraduate curriculum.

Decreasing unplanned extubations in the surgical intensive care unit.

BACKGROUND: Unplanned extubations are common, but can be life-threatening. METHODS: We conducted a prospective evaluation of all intubated patients in our surgical intensive care unit to examine the effects of three parameters on the likelihood of accidental extubation. The parameters were the method of endotracheal tube fixation, the use of sedation/paralysis, and the use of hand restraints. During the baseline period, tubes were secured with cloth or velcro ties, sedation was used conservatively, and hand restraints were used routinely. A change in one study parameter was made prior to each period. Thus, in period II, tubes were secured using waterproof tape; in period III, tubes were secured with waterproof tape and sedation/paralysis was used liberally; and in period IV, tubes were secured with waterproof tape and limited use was made of hand restraints. RESULTS: Accidental extubations were significantly less frequent when tubes were secured with waterproof tape (P < 0.0001). No difference was seen when sedation was instituted liberally. Restricted use of hand restraints was associated with significantly increased accidental extubations (P < 0.001). CONCLUSIONS: Our data support the use of water resistant tape to secure endotracheal tubes and the routine use of hand restraints.

Ethanol added to resuscitation improves survival in an experimental model of hemorrhagic shock.

Ethyl alcohol induces systemic vasodilation, decreases platelet aggregation, and inhibits neutrophil activation in vivo. Alcohol may thus be of potential benefit in resuscitation from shock by improving microcirculation. The purpose of this study was to test the effects of ethanol (ETOH) in resuscitation from hemorrhagic shock. Blood pressure, tissue pO2, white blood cell (WBC) and platelet adhesiveness, and survival were measured for 60 male Sprague-Dawley rats in a blinded and randomized study. Anesthetized animals were phlebotomized to 60 per cent of their blood volume, and maintained in shock for 45 minutes. Resuscitation was by continuous infusion of Lactated Ringers (LR) at 2 x shed blood volume over 1 hour. The experimental group received LR and ETOH (1.25 mL/kg). Control rats received LR and placebo. Mean arterial pressure was not significantly different, nor was WBC adhesiveness index different. However, postresuscitation platelet adhesiveness index was significantly higher in control rats than in ETOH rats. Postresuscitation total platelet arterial-venous difference was also greater in controls than in ETOH rats. Average tissue pO2 for ETOH rats (47 +/- 8.2 mm Hg) was significantly higher than controls (39.0 +/- 9.8 mm Hg) during resuscitation (P = 0.0001). Survival for ETOH rats (70%) was significantly higher than controls (20%) (P = 0.003). Our data suggests that ETOH added to resuscitation from shock improves survival by inhibiting platelet activation and increasing tissue perfusion.

Effects of trauma on leukocyte intercellular adhesion molecule-1, CD11b, and CD18 expressions.

BACKGROUND: During traumatic injury, a multitude of events, including ischemia, may cause leukocyte adhesion and margination. In this study, alterations of surface receptors involved in leukocyte adhesion were studied in traumatized patients. In an attempt to discern the role of hypoxia, additional experiments were conducted in which normal human leukocytes were subjected to hypoxic stress in vitro. METHODS: Venous blood was obtained from 10 trauma patients within 2 hours of blunt injury (mean Injury Severity Score of 17 +/- 8) and from 8 normal volunteers (controls). Leukocytes were isolated from patients and controls. To assess the effect of hypoxia, normal leukocytes were placed in hermetically sealed environments containing 100% nitrogen. All leukocytes were labeled with phycoerythrin- or fluorescein-bound monoclonal antibodies to intercellular adhesion molecule-1 (ICAM-1), or to integrins CD18 and CD11b. Receptor concentration was measured by flow cytometry. Results were expressed as percentage of receptor-positive cells (%) and mean fluorescence channel units, which directly correlate with monoclonal antibody cell surface density. Significance of differences was tested by analysis of variance/Kruskal-Wallis test. RESULTS: Compared with the normal controls, circulating leukocytes obtained from traumatized patients showed decreased expression of ICAM-1, CD11b, and CD18 2 hours after injury. In contrast, normal leukocytes exposed to hypoxic stress in vitro exhibited a marked increase in CD11b and CD18 expression and no change in ICAM-1 expression. CONCLUSIONS: Leukocytes obtained from traumatized patients showed a significant decrease in cell surface expression of adhesion receptors. This phenomenon is unlikely to be a direct consequence of hypoxia alone, because exposure to isolated hypoxia in vitro actually increased expression of CD11b and CD18.

Hypoxia-induced alterations of neutrophil membrane receptors.

Extravasation of leukocytes at sites of ischemia may mediate tissue injury. To determine how leukocyte accumulation may be induced by ischemia, effects of hypoxia on basal neutrophil expression of adhesion and activation receptors were examined. Effects of hypoxia upon preactivated cells were also studied. To determine whether regulation of expression is dependent on oxygen availability or on mitochondrial respiration, the effects of physical hypoxia (substitution of O2 by nitrogen) were compared with those of chemical hypoxia with sodium cyanide (NaCN). Leukocytes in whole blood (eight volunteers) were exposed either to hypoxia alone or to priming concentrations of lipopolysaccharide (LPS, 1 microgram/ml) followed by chemical hypoxia (NaCN, 1 mM) or physical hypoxia (PO2 of 1-10 torr) for various time intervals. Room air was controlled and hypoxic cells were labeled with fluorescent monoclonal antibodies to integrins CD18 and CD11b or to the 55-kDa TNF alpha cell surface receptor (TNFR). Receptor concentrations were measured by flow cytometry. Data were analyzed by ANOVA/Student's t test. Physical hypoxia increased expression of both CD11b and CD18 over time and augmented their LPS-induced up-regulation. Isolated chemical hypoxia did not change neutrophil expression of CD11b or CD18, but partially inhibited neutrophil CD11b and CD18 up-regulation by LPS. LPS-induced TNFR down-regulation was not affected by physical hypoxia, which failed to alter TNFR expression in this model.(ABSTRACT TRUNCATED AT 250 WORDS)

Soluble cytokine receptors and receptor antagonists are sequentially released after trauma.

Cytokine receptors and receptor antagonists (RAs) have been identified in trauma patients. We hypothesized that after traumatic injury, a sequential release of soluble cytokine receptors and RAs may exist that mirrors the release of the primary cytokines themselves. Twenty-two patients were included in the study: 14 males and 8 females. The mean age was 30.1 +/- 12.5 (range, 19 to 71), and the mean Injury Severity Score was 28.7 +/- 12.6 (range, 4 to 57). There were 15 survivors and 7 nonsurvivors. Samples were collected on arrival to the emergency department and at serial intervals for up to 7 days. Monoclonal antibody enzyme-linked immunosorbent assay kits to tumor necrosis factor (TNF), soluble TNF-receptor (sTNF-R) 55 kd and 75 kd, interleukin (IL)-1 and IL-1 RA, and IL-2 and IL-2r were used. Sera from 22 healthy individuals were used as normal controls. No TNF, IL-1, or IL-2 could be detected in any patient sera after injury. Control levels for the soluble cytokine receptors and RAs were as follows: sTNF-R 55 kd, 607 +/- 89 pg/mL; sTNF-R 75 kd, 2,141 +/- 169 pg/mL; IL-1 RA, 291 +/- 35 pg/mL; and IL-2r, 426 +/- 53 U/mL. In trauma patients, both 55 kd and 75 kd sTNF-R were significantly elevated on arrival to the emergency department, with values of 2,441 +/- 506 pg/mL (p < 0.001) and 4,736 +/- 537 pg/mL (p < 0.001), respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

A technique for repair of traumatic parasternal lung herniation: case report.

A case of traumatic lung herniation through an area of costalsternal separation in a 36-year-old male is presented. Persistent pain and the threat of strangulated lung tissue prompted repair that was accomplished with an expanded polytetrafluoroethylene Gortex tissue patch.

Respiratory distress in traumatized and burned children.

Incidence and significance of respiratory failure after trauma in children was the subject of this study. One thousand nine hundred eighty-nine pediatric trauma patients (aged 18 years or less) were treated at the authors' level I trauma center between 1985 and 1993. Of these, 364 (18%) were intubated. Their mechanisms of injury were: motor vehicle accidents in 93 (25%), pedestrians struck by vehicles in 93 (25%), motorcycle or bicycle accidents in 55 (15%), gunshot and stab wounds in 43 (12%), major burns (> 20% BSA) in 31 (9%), 14 of whom also had smoke inhalation, falls in 25 (7%), sport-related injuries in 9 (2%), and child abuse in 8 (2%). Average injury severity score of intubated patients was 27.0 +/- 21.4. Average trauma score was 11.7 +/- 4.1. Of the intubated patients, 248 (68%) had head injuries, 153 (42%) chest injuries, and 114 (31%) abdominal and pelvic injuries. Ninety-three (25%) of intubated patients died within 5 days of injury: 70 of head injury, 23 of multiple major organ injury. Intubation was required for more than 5 days in 77 patients (21%); 50 (14%) of these patients met criteria for respiratory distress syndrome (RDS): 12 (24% of RDS patients) died. Two of the deaths were multiply traumatized patients, and 10 were patients with burns and smoke inhalation. The authors conclude that RDS is uncommonly the cause of death in pediatric trauma patients. Burned patients with RDS are an exceptional group, with significant mortality.

A method to determine the adequacy of resuscitation using tissue oxygen monitoring.

The change in tissue PO2 in response to an increased inspired O2 challenge may be related to the state of cellular oxygenation, and hence the adequacy of resuscitation. To test this hypothesis, we measured tissue PO2 during inspired O2 challenges in 29 injured patients during acute resuscitation or intensive care unit monitoring. The O2 challenge test had 100% sensitivity and specificity in detecting flow-dependent O2 consumption in invasively monitored patients in the intensive care unit. During acute resuscitation, 60% of patients had negative initial O2 challenge test results, indicating that flow-dependent O2 consumption might have been present. Of nine such patients, five had subsequent positive O2 challenge test results after fluid resuscitation, indicating successful resuscitation. Four patients (27% of acute resuscitations), however, had repeatedly negative findings, possibly indicating persistent inadequate cellular oxygenation despite fluid resuscitation. Other commonly measured variables did not differentiate these patients. Monitoring of tissue PO2 during an inspired O2 challenge may be a useful test for determining the adequacy of resuscitation from hypovolemic shock.

Radiation exposure from cervical spine radiographs.

Radiation exposure to hospital personnel during 41 cervical spine radiographs of 30 multiply injured patients was prospectively evaluated. A digital dosimeter was attached to the exposed torso of personnel applying upper extremity traction or managing the airway. Radiation exposure was measured during each radiograph. Any exposure of one or more milliroentgen was detectable. No radiograph resulted in a measurable radiation exposure. Multiple radiographs (up to five exposures) also did not register even the minimum recordable exposure, demonstrating that exposure is less than 1.0 mR per radiograph (P < .05). We conclude that hospital personnel, even those applying traction or managing the airway, are not at risk of significant radiation exposure at the time of cervical spine radiographs.

Continuous arteriovenous hemodiafiltration in postoperative and traumatic renal failure.

Management of acute renal failure (ARF) in surgical patients has relied on supportive measures including hemodialysis and peritoneal dialysis. An alternative technique currently available is continuous arteriovenous hemodiafiltration (CAVH-D). Records of 44 surgical patients with ARF who were treated with CAVH-D in our surgical intensive care unit from 1989 to 1992 were reviewed. Thirty-five patients underwent emergency operations, and 4 patients underwent elective operations. Thirty-three patients were hemodynamically unstable immediately prior to the institution of CAVH-D, making hemodialysis a contraindication. A total of 565 CAVH-D days with an average of 13 days per patient were evaluated. Seventeen patients survived, with recovery of renal function in 13 patients. Vascular access was obtained via 227 percutaneous femoral catheters and 4 Scribner shunts. Seven vascular complications occurred, including arteriovenous fistula, pseudoaneurysm, limb ischemia, femoral artery hemorrhage, and femoral vein thrombosis. Based on these data, we conclude that CAVH-D is a safe and effective alternative in surgical patients with ARF.

Pentoxifylline alone versus pentoxifylline combined with superoxide dismutase prolongs survival in a rat hemorrhagic shock model.

Pentoxifylline (PTX) and superoxide dismutase (SOD) have each proven effective in improving survival when administered during resuscitation in animal models of hemorrhagic shock. This study was conducted to determine if PTX and SOD combined would have synergistic effectiveness in the treatment of hemorrhagic shock. Sprague-Dawley rats (n = 40) were phlebotomized at 25 ml/kg for 2 min, then subjected to a 45-min ischemic period, and resuscitated with lactated Ringer's solution (LR) (50 ml/kg) over 1 h. This model resulted in 70% mortality over 72 h when resuscitation was with LR alone. Animals were randomized into groups to receive one of the following agents during resuscitation: PTX in LR, SOD in LR, a combination of PTX and SOD in LR, or LR alone. PTX or SOD alone were effective in prolonging survival. However, the combination of PTX and SOD did not prolong survival above LR control.

Emergency thoracotomy with lung resection following trauma.

Over a 7-year period, 9443 trauma patients were evaluated with 2934 (31%) sustaining chest trauma. Of these, 347 (12%) patients required thoracotomy, with 12 patients undergoing emergency lung resection. Mean age was 23.1 years with mean Injury Severity Score of 32. Mechanism of injury was blunt in three (25%), gunshot wound in seven (58%), and stab wound in two (17%). Associated injuries included head injury in two (17%), intra-abdominal injury requiring laparotomy in four (33%), cardiac injury in three (25%), and great vessel injury in one (8%). Indications for operation included persistent hemorrhage in 11 and suspected tracheobronchial disruption in one. Non-anatomic lung resection was performed in five patients, lobectomy in three patients, and pneumonectomy in four patients. Overall mortality was 33 per cent: 20 per cent for non-anatomical lung resection, 33 per cent for lobectomy, and 50 per cent for pneumonectomy. All survivors fully recovered except for one patient with an associated head injury. Our experience supports the selective use of lung resection, including pneumonectomy, to immediately control hemorrhage and to impact survival in severe chest trauma.

Orotracheal intubation in trauma patients with cervical fractures.

OBJECTIVE: To evaluate orotracheal intubation with in-line stabilization of the cervical spine for emergency airway treatment of trauma patients with cervical spine injuries. DESIGN: Of 7518 trauma patients examined, 81 patients with cervical spine injuries received emergency orotracheal intubation. All intubations were performed by experienced anesthesiologists, with a separate individual maintaining in-line stabilization. Neurologic examination was documented before and after intubation. RESULTS: Peripheral neurologic deficit was present from the outset in 20 patients. There were unstable cervical fractures in 38 patients with no neurologic deficit. Twenty-three patients were neurologically intact with fractures that were later judged stable. In no instance was there a deterioration of neurologic status following intubation. Peripheral neurologic deficits improved after intubation in four patients. CONCLUSION: Orotracheal intubation, performed with manual in-line stabilization by trained and experienced personnel, is a safe emergency procedure in patients with cervical fractures.

Tumor necrosis factor and monocytes are released during hemorrhagic shock.

Tumor necrosis factor (TNF) is a key mediator involved in many physiologic processes including immunity, inflammation, and metabolism. A relationship between TNF and hemorrhagic shock has not been clearly demonstrated. To help understand the role of TNF in hemorrhagic shock we developed a hemorrhagic shock model to measure TNF and monocyte levels during hemorrhage and resuscitation. Male Sprague-Dawley rats were anesthetized and subjected to a 50% blood loss (30 ml/kg) over 2 min and left in shock for 58 min. The animals were then resuscitated with two times blood loss (60 ml/kg) using lactated Ringers over 1 h. This model results in 75% mortality within 3 days (LD 75). Blood samples (2 ml) were obtained at intervals during shock and resuscitation, and assayed for TNF concentrations and white blood cell counts. Despite a marked fall in total leukocytes (24,600 pre-hemorrhage to 11,300 post-hemorrhage, P < 0.005), monocytes increased in percentage and in total count. Blood levels of TNF were initially undetectable but rose within 10 min after hemorrhage, peaked at 30 min after hemorrhage, and then became undetectable during resuscitation. In this model, macrophages and TNF are released into the circulation after hemorrhagic shock. TNF may play a role as a mediator in the pathophysiology of hemorrhagic shock.

Trauma causes early release of soluble receptors for tumor necrosis factor.

The importance of tumor necrosis factor (TNF) in the pathophysiology of trauma and hemorrhagic shock is not known. In addition, TNF bioactivity may be modulated by soluble forms of the 55-kd and 75-kd membrane receptors (TNFR). This study was undertaken to determine circulating levels of TNF and TNFR after trauma. Nine severely injured male patients were studied. The mean age was 30 +/- 10 years (range, 15-45). The mean Injury Severity Score (ISS) was 31.3 +/- 17.6 (range, 10-59), and the mean Revised Trauma Score (RTS), 5.7 +/- 2.2 (range, 0.7-7.8). Serum was obtained immediately upon arrival at our trauma center, within 1 hour of injury. The TNF and TNFR levels in the serum were measured using ELISA techniques. After trauma, 55-kd and 75-kd TNFR levels were significantly elevated above those of controls (6.99 +/- 4.57 ng/mL and 5.42 +/- 1.88 ng/mL, respectively, p < 0.01); TNF levels were not increased. Patient serum containing TNFR inhibited in vitro TNF cytotoxicity and correlated with 55-kd TNFR levels (p < 0.05). We conclude that TNF is a strong releasing factor for TNFR; the presence of TNFR may be indirect evidence that TNF is present after trauma, despite low measured levels. Both TNF and TNFR may be more important in trauma and hemorrhagic shock than previously thought.

Hypoxia induces a human macrophage cell line to release tumor necrosis factor-alpha and its soluble receptors in vitro.

Tissue hypoxia following hemorrhage and trauma is a possible initiating factor of the generalized inflammatory response seen after shock. The role of hypoxia in the release from a human monocyte cell line (THP-1) of tumor necrosis factor-alpha (TNF alpha) and its soluble membrane receptors (TNF alpha R) in-vitro is investigated in this study. Flat-bottom plates with 500,000 THP-1 cells/ml were placed in air-tight sealed boxes and exposed to hypoxia (O2 = 1%) or controls (O2 = 9%) for up to 24 hr. Supernatants were tested for TNF alpha, as well as 55- and 75-kDa soluble receptors for TNF alpha, by ELISA. Cell viability was assessed by vital dye uptake and was found to be maintained throughout hypoxic exposure. Medium pH levels were within normal range. In eight experiments conducted in duplicate, minimal change over 24 hr occurred in control samples. Control mean and SD were: TNF alpha = 12.0 +/- 4.2, 55-kDa R = 34.6 +/- 2.03, and 75-kDa R = 38.88 +/- 9.68 pg/ml. During hypoxia, TNF alpha was released as early as the first 30 min of exposure (41.3 +/- 2.3 pg/ml) with a small peak at 1 hr (52 +/- 5.0 pg/ml) and a later more pronounced peak at 18 hr (526 +/- 48 pg/ml). Both 55- and 75-kDa R were released by the hypoxic monocytes; release was progressive and was maximal at 24 hr in this study. Maximal release value of 55-kDa R was 236 +/- 15 pg/ml, while for 75-kDa R it was 2450 +/- 63 pg/ml.(ABSTRACT TRUNCATED AT 250 WORDS)

Regional flow during experimental hemorrhage and crystalloid resuscitation: persistence of low flow to the splanchnic organs.

UNLABELLED: BACKGROUND AND METHODS. Rapid changes in cardiac output (CO) and organ perfusion occur with hemorrhagic shock and fluid resuscitation. To assess regional alterations of flow, 40 Sprague-Dawley male rats were subjected to hemorrhagic shock and crystalloid resuscitation under halothane anesthesia. Polyethylene microspheres were injected before and after hemorrhage and after resuscitation. At sacrifice, brain, lungs, heart, liver, intestine, spleen and kidneys were harvested, weighed and radioactivity counted. Changes in mean arterial pressure, oxygen consumption, organ flow and CO were also measured. RESULTS: Cardiac output decreased during hemorrhage (P less than 0.01), it increased with resuscitation but did not return to baseline even with infusion of fluid volumes of three times the blood loss. Flow decreased during hemorrhage in all organs, but the difference was not statistically significant in the liver (P greater than 0.05), since a larger percentage of CO was maintained as hepatic perfusion. During resuscitation, flow to brain and kidneys increased over the percentage values expected by increased CO (P less than 0.01), but flow to the liver did not increase significantly. Flow to small bowel remained depressed (P less than 0.005). CONCLUSIONS: Following hemorrhage there is hypoperfusion of all splanchnic organs; however, flow to the liver decreases least. Crystalloid resuscitation in our model failed to return CO to baseline. Blood supply to intestine remained depressed in disproportion to CO both after hemorrhage and resuscitation and hepatic blood flow remained decreased after resuscitation.