Adequate crystalloid resuscitation restores but fails to maintain the active hepatocellular function following hemorrhagic shock

Studies have shown that active hepatocellular function is depressed early after trauma-hemorrhage and persists despite resuscitation with two or three times (x) the volume of maximum bleedout (MB) with lactated Ringer's solution (LR). However, it is not known if a larger volume of fluid resuscitation corrects this dysfunction. To study this, rats were bled to and maintained at a mean arterial pressure of 40 mm Hg until 40% of the MB volume was returned in the form of LR, and then resuscitated with 4x or 5x the volume of MB with LR. Three doses of indocyanine green (ICG) were given intravenously and [ICG] measured in vivo using an in-vivo hemoreflectometer. The initial velocity of the clearance of ICG was calculated. Maximal velocity of the clearance (Vmax: the number of functional ICG receptors) and kinetic constant (Km: the efficiency of the active transport) were determined from the Lineweaver-Burk plot. Vmax decreased during hemorrhage, was restored to control levels at 0-4 hours after resuscitation, but decreased at 4-8 hours after resuscitation despite restoration of cardiac output following resuscitation with 5x LR. This could be the result of increased TNF release. The Km also decreased during hemorrhage, but increased at 0-1.5 hours and remained at control levels even 4-8 hours after resuscitation. Thus the failure of Vmax to remain at control levels following adequate fluid resuscitation may form the basis of cellular dysfunction and multiple organ failure after severe hemorrhagic shock.     

Pentoxifylline but not saralasin restores hepatic blood flow after resuscitation from hemorrhagic shock

After determining that hepatic blood flow remains impaired after resuscitation from hemorrhagic shock, we used the angiotensin II receptor antagonist saralasin and pentoxifylline to investigate their respective effects on hepatic blood flow responses after resuscitation from hemorrhagic shock. Rats were bled to 50% of baseline blood pressure for 60 min and resuscitated with shed blood and an equal volume of lactated Ringer's solution. Saralasin [10 micrograms/kg per min (n = 6)], pentoxifylline [25 mg/kg bolus and 12.5 mg/kg per hr (n = 7)], or saline (n = 11) were started with the onset of resuscitation. Total hepatic blood flow measured by ultrasonic transit time flow meter, effective nutrient hepatic blood flow measured by galactose clearance, mean arterial pressure, and cardiac output were recorded at 15-min intervals for 2 hr after resuscitation. Hemorrhage decreased cardiac output 57% below baseline and decreased total hepatic blood flow 64% below baseline. Resuscitation restored cardiac output to baseline levels in all three groups. Despite restoration of cardiac output, total hepatic and effective hepatic blood flow remained significantly below baseline in the saline control and saralasin groups but was restored to baseline levels in the pentoxifylline group. These data indicate that angiotensin II does not contribute significantly to the hepatic blood flow impairment after resuscitation from hemorrhagic shock. Improvement in flow with pentoxifylline implies that hemorrhage and resuscitation impair hepatic microvascular hemorrheology and that addition of pentoxifylline to standard resuscitation corrects the impairment.     

ATP-MgCl2 restores the depressed hepatocellular function and hepatic blood flow following hemorrhage and resuscitation

Although ATP-MgCl2 produces a myriad of beneficial effects following organ ischemia and simple hemorrhagic shock in animal models which involved heparinization and/or blood resuscitation, it is not known whether ATP-MgCl2 has any salutary effect on the depressed active hepatocellular function (AHF) and hepatic microvascular blood flow (HMBF) in a nonheparinized model of trauma and severe hemorrhage in the absence of blood resuscitation. To determine this, rats underwent a midline laparotomy (i.e., trauma induced) and were bled to and maintained at a mean arterial pressure of 40 mm Hg until 40% of the maximum shed blood volume was returned in the form of Ringer's lactate (RL). The animals were then resuscitated with four times the volume of shed blood with RL. ATP-MgCl2, 50 mumoles/kg body weight (BW) each or an equivalent volume of normal saline, was infused intravenously for 95 min during and following crystalloid resuscitation. At 1.5 and 4 hr after resuscitation, AHF (Vmax, maximal velocity of indocyanine green clearance; Km, efficiency of the active transport process) was determined without blood sampling by using an in vivo indocyanine green clearance technique. HMBF was measured with laser Doppler flowmetry. Results indicate that Vmax, Km, and HMBF decreased significantly at 1.5-4 hr after hemorrhage and resuscitation. ATP-MgCl2 infusion restored the depressed Vmax, Km, and HMBF and prevented the occurrence of hepatic edema. The restoration of AHF with ATP-MgCl2 treatment may be due to its direct salutary effect on the active indocyanine green transport process and/or due to improvement in hepatic microcirculation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Fluid resuscitation after an otherwise fatal hemorrhage: I. Crystalloid solutions

One half of deaths among trauma victims occur within 1 hour of injury and are due to rapid hemorrhage or CNS trauma. We developed a rapid hemorrhage model in unanesthetized swine to simulate human exsanguination. We compared the ability of four crystalloid solutions to prevent death after an otherwise fatal hemorrhage: normal saline (NS), Ringer's lactate (RL), Plasmalyte-A (PA), and Plasmalyte-R (PR). Five days before hemorrhage swine received an aortic sideport and a central venous treatment catheter. Aortic blood (54 ml/kg) was removed in 15 minutes from 116 swine. The percentages of shed blood replaced were 14% in 5 minutes with NS, 100% in 20 minutes with NS, and 300% in 30 minutes with NS, RL, PA, or PR. We found that all mortalities were determined within 2 hours after hemorrhage and that RL provided the best survival rate of 67% (NS 300% = 50%, PR = 40%, and PA = 30%.) After an analysis of arterial blood gas, lactate, acid-base, heart rate, and aortic pressure measurements, we conclude that RL is the superior crystalloid solution because of its decreased chloride load (compared to NS) and because of the absence of acetate or magnesium (compared to PA and PR).     

Pneumoperitoneum in healthy humans does not affect central blood volume or cardiac output

BACKGROUND: This study addresses the question of whether the elevation of the mean arterial pressure and central venous pressure in response to pneumoperitoneum for laparoscopic surgery is caused by increases in central blood volume and/or cardiac output. METHODS: Eleven patients in good cardiopulmonary health and scheduled for laparoscopic cholecystectomy, with a mean age of 42 years, were included. After induction of anaesthesia with fentanyl and propofol, radial arterial and central venous lines were introduced. The central blood volume and cardiac output were determined by the indicator-dilution technique, using inline densitometric measurements of indocyanine green (ICG). The measurements were made before and after the establishment of pneumoperitoneum by insufflation of carbon dioxide to an intra-abdominal pressure level of 11-13 mmHg. RESULTS: The mean arterial pressure (62+/-6 mmHg) increased after induction of pneumoperitoneum by 40+/-26% (P<0.05) and the central venous pressure increased from 6+/-4 mmHg to 8+/-6 mmHg (P<0.05). The cardiac output (4.3+/-0.9 L/min) and central blood volume (1.5+/-0.5 L) were not affected by the induction of pneumoperitoneum. CONCLUSIONS: In healthy anaesthetized subjects, the elevation of mean arterial pressure and central venous pressure in response to pneumoperitoneum was not caused by enhancement in cardiac output or central blood volume.     

Urine output following severe burns

A difference of opinion exists concerning the usefulness of hourly urine volume as an index of effective resuscitation following burns. An investigation is described in which the volume and osmolality of the hourly urine output of severely burned patients was measured. The determinants of urine output in normal and injured patients are described and it is emphasized that volume is only one parameter of urine output. A clear distinction is made between oliguria and antidiuresis, and the significance of free osmolal output as an index of renal performance is explained.It is pointed out that the nature of the infusion fluid, the methods of assessing effective resuscitation, and the nature of the urine output are closely interrelated in any one scheme of resuscitation. From this it follows that a urine volume judged to indicate adequate resuscitation in one scheme may have a different significance in another. The view is expressed that if effective resuscitation means adequate blood flow to all tissues, the kidney should not be made an exception. Adequate renal perfusion is proved only when the urine contains a free osmolal output sufficient to permit maintenance of normal body osmolal balance in the absence of gross changes in solute-free water intake. It is suggested that urine output (volume, osmolality and free osmolal output) is a useful index of the effectiveness of resuscitation since it provides a reliable indication of renal perfusion except when renal function itself is impaired.For patients resuscitated with plasma, the range in which an ‘adequate’ urine output can be expected to lie, is given, and the possible affects of other fluids on the urine output are discussed. It is stressed that when renal function is impaired, consideration of the urine output is a more useful prognostic aid than is a study of the blood urea levels.     

Intravenous iron following cardiac surgery does not increase the infection rate

BACKGROUND: Intravenous iron (FeIV) has been used increasingly, alone or in combination with recombinant erythropoietin, to promote red cell production as part of a blood conservation program. Given the important role that iron plays in the growth of bacteria, it has been hypothesized that this use of FeIV may promote surgical site infection. However, this hypothesis has not yet been tested appropriately. To assess this hypothesis, postoperative infection rates in patients undergoing cardiothoracic surgery were analyzed. METHODS: Data were collected on 863 patients undergoing cardiopulmonary bypass surgery in 2001. Patients were either enrolled voluntarily in a blood conservation program in which they received either postoperative FeIV and erythropoietin (n=302), as indicated, or blood transfusions and no FeIV (n=561), as indicated, to correct postoperative anemia. Infections were defined according to the U.S. Centers for Disease Control and Prevention guidelines. RESULTS: Thirty-nine infections developed. The overall infection rate was 4.52%, with an infection rate of 3.97% in the iron-treated group (n=12) and a rate of 4.81% in the untreated group (n=27). When the impact of gender, age, diabetes mellitus, operating time, type of surgery, and blood transfusions were controlled for, FeIV did not increase the risk of infection (odds ratio of 1.031 for each increment of 125 mg of FeIV; 95% confidence interval 0.908, 1.170; p=0.64). CONCLUSIONS: There was no impact of FeIV on the subsequent infection rate in a cardiac surgery patient cohort, indicating its safety for use in the postoperative setting.     

Core-peripheral temperature gradient does not predict cardiac output or systemic vascular resistance in children

We prospectively measured toe temperature, rectal temperature, systemic arterial pressure and cardiac output on two occasions one hour apart in 136 children who had had phenoxybenzamine after cardiac surgery while on cardiopulmonary bypass. Repeated measures analysis showed that there was no significant correlation between the change in temperature gradient over one hour in each patient and the change in cardiac index (r = 0.03, P greater than 0.1) or systemic vascular resistance (r = 0.007, P greater than 0.1). Although the peripheral temperature (toe temperature), and the core-peripheral temperature difference are simple, safe and inexpensive to measure in the population studied, they did not provide any guide to either cardiac output or systemic vascular resistance.     

Continuous cardiac output measurements do not agree with conventional bolus thermodilution cardiac output determination

PURPOSE: To evaluate the performance of two different continuous cardiac output monitoring systems based on the thermodilution principle in critically ill patients. METHODS: Nineteen cardiac surgical patients were randomly assigned to continuous cardiac output monitoring using one of the two systems under study (group I, IntelliCath(TM) catheter, n=9; group II, Opti-Q(TM) catheter, n=10). Each patient was studied over a period of three hours. Conventional bolus thermodilution cardiac output measurements were carried out every 15 min leading to 13 measurements in each patient. The continuous cardiac output values were compared with the bolus thermodilution measurements. Bias (mean difference between continuous and bolus thermodilution) and precision (SD of differences) were calculated as a measure of agreement between the respective continuous method and conventional bolus thermodilution. RESULTS: The range of measured cardiac outputs was 3.8-15.4 L*min(-1) (IntelliCath(TM)) and 3.5-8.3 L*min(-1) (OptiQ(TM)). Bias and precision was 0.06 +/- 0.76 L*min(-1) (IntelliCath(TM)) and -0.04 +/- 0.74 L*min(-1) (OptiQ(TM)), respectively. There was no difference in bias between the two systems (P=0.38). +/- 2 SD of the differences (i.e., 95% of the differences) did not fall within the predetermined limits of agreement of +/- 0.5 L*min(-1). CONCLUSIONS: There was no difference between the two systems regarding the agreement with conventional bolus thermodilution as the standard. A discrepancy between bolus and continuous thermodilution cardiac output measurement techniques above the clinically acceptable limits suggest that they are not interchangeable.     

Elevated activated partial thromboplastin time does not correlate with heparin rebound following cardiac surgery

Purpose

Exposure to cardiopulmonary bypass (CPB) is associated with postoperative coagulopathy and hemorrhage. Recent literature indicates that heparin rebound occurs almost universally following cardiac surgery. We conducted this pilot study to evaluate if the presence of residual circulating heparin following cardiac surgery can be diagnosed by elevation of activated partial thromboplastin time (APTT).

Method

After obtaining Research Ethics Board approval, blood samples from 30 patients receiving heparin for CPB were evaluated at the time of intensive care unit admission and 2, 4, and 6 hr thereafter. Activated clotting time, whole blood heparin concentration (Hepcon HMS Plus, Medtronic), anti-Xa levels, and APTT were measured at each time point. Samples with prolonged APTT were subjected to mechanistic studies with heparin adsorption and 1:1 mixing.

Results

Anti-Xa was elevated in 52 of the 120 blood samples (0.08 ± 0.08 U · mL^?1, mean ± SD). APTT was elevated in 49 (40.8%) samples with an average of 51.4 ± 31.9 sec. At all time points, the APTT correlated poorly with anti-Xa levels with correlation coefficients ranging from ?0.26 to ?0.05. Mean APTT was modestly, but not significantly, associated with total dose of protamine with r = 0.34 (CI: ?0.03, 0.62). After 1:1 mixing studies, APTT returned to normal in most (82%) samples tested.

Conclusion

Circulating residual heparin is commonly presented following cardiac surgery and does not correlate with APTT. Considering that mixing studies normalize APTT in most samples, elevated APTT following CPB may reflect deficiency of coagulation factors or presence of a coagulation inhibitor such as protamine. Further studies are required to confirm this observation.     

Large volume crystalloid resuscitation does not increase extravascular lung water

The purpose of this study was to determine whether Ringer's lactate solution increases extravascular lung water (EVLW) during resuscitation after hemorrhagic shock. Ten sheep anesthetized with thiamylal were bled to a mean arterial pressure (MAP) of 50 mm Hg; further bleeding maintained that pressure for 30 min. Resuscitation fluid consisted of Ringer's lactate solution in volumes necessary to restore and maintain for 1 hr MAP, pulmonary capillary wedge pressure (PCWP), and cardiac index at levels equal to those measured before bleeding. After volume replacement, the colloid oncotic pressure (COP) - PCWP gradient (COP - PCWP) decreased from 12 +/- 3 to 2 +/- 5 mm Hg (P less than 0.001). After volume restoration, COP decreased from 19 +/- 8 mm Hg to 12 +/- 2 mm Hg (P less than 0.001). Despite the large volume of fluid administered, EVLW did not increase. Crystalloid resuscitation does not necessarily increase EVLW despite significant decreases in COP and COP - PCWP gradient.