Uwe Kreimeier

Changes in blood volume and hematocrit during acute preoperative volume loading with 5% albumin or 6% hetastarch solutions in patients before radical hysterectomy.

BackgroundThe impact of acute preoperative volume loading with colloids on blood volume has not been investigated sufficiently. MethodsBefore surgery, in 20 patients undergoing major gynecologic procedures, volume loading was performed during anesthesia by infusing approximately 20 ml/kg of colloid at a rate of 90 ml/min (group I: 5% albumin solution; group II: 6% hetastarch solution; n = 10 each). Plasma volume (indocyanine green dilution technique), erythrocyte volume (labeling erythrocytes with fluorescein), hematocrit, total protein, and hetastarch plasma concentrations (group II) were measured before and 30 min after the end of infusion. ResultsMore than 1,350 ml of colloid (approximately 50% of the baseline plasma volume) were infused within 15 min. Thirty minutes after the infusion had been completed, blood volume was only 524 ± 328 ml (group I) and 603 ± 314 ml (group II) higher than before volume loading. The large vessel hematocrit (measured by centrifugation) dropped more than the whole body hematocrit, which was derived from double-label measurements of blood volume. ConclusionsThe double-label measurements of blood volume performed showed that 30 min after the infusion of approximately 20 ml/kg of 5% albumin or 6% hetastarch solution (within 15 min), only mean 38 ± 21% and 43 ± 26%, respectively, of the volume applied remained in the intravascular space. Different, i.e., earlier or later, measuring points, different infusion volumes, infusion rates, plasma substitutes, or possibly different tracers for plasma volume measurement might lead to different results concerning the kinetics of fluid or colloid extravasation.

Changes in intravascular volume during acute normovolemic hemodilution and intraoperative retransfusion in patients with radical hysterectomy.

Background: Changes in blood volume during acute normovolemic hemodilution (ANH) and their consequences for the perioperative period have not been investigated sufficiently. Methods: In 15 patients undergoing radical hysterectomy, preoperative ANH to a hematocrit of 24% was performed using 5% albumin solution. Intraoperatively, saline 0.9% solution was used for volume substitution, and intraoperative retransfusion was started at a hematocrit of 20%. Plasma volume (indocyanine green dilution technique), hematocrit, and plasma protein concentration were measured before and after ANH, before retransfusion, and postoperatively. Red cell volume (labeling erythrocytes with fluorescein) was determined before and after ANH and postoperatively. Results: Mean normal plasma volumes (1,514 ± 143 ml/m2) and reduced red cell volumes (707 ± 79 ml/m2) were measured preoperatively. Blood (1,150 ± 196 ml) was removed and replaced with 1,333 ± 204 ml of colloid. Blood volume before and after ANH was equal and amounted to 3,740 ml. Intraoperatively, plasma volume did not increase until retransfusion despite infusing 3,389 ± 1,021 ml of crystalloid (corrected for urine output) to compensate for an estimated surgical blood loss of 727 ± 726 ml. Postoperatively, after retransfusion of all autologous blood, blood volume was 255 ± 424 ml higher than preoperatively before ANH. Despite mean calculated blood loss of 1,256 ± 892 ml, only one patient received allogeneic blood. Conclusions: During ANH, normovolemia was exactly maintained. After surgical blood loss of 1,256 ± 892 ml, crystalloid and colloid supplies of 5,752 ± 1,462 ml and 1,667 ± 548 ml, respectively, and complete intraoperative retransfusions of autologous blood in every patient, mean blood volume was 250 ml higher than preoperatively before ANH.

Acid-base changes caused by 5% albumin versus 6% hydroxyethyl starch solution in patients undergoing acute normovolemic hemodilution: a randomized prospective study.

BackgroundPreoperative acute normovolemic hemodilution (ANH) is an excellent model for evaluating the effects of different colloid solutions that are free of bicarbonate but have large chloride concentrations on acid–base equilibrium. MethodsIn 20 patients undergoing gynecologic surgery, ANH to a hematocrit of 22% was performed. Two groups of 10 patients each were randomly assigned to receive either 5% albumin or 6% hydroxyethyl starch solutions containing chloride concentrations of 150 and 154 mm, respectively, during ANH. Blood volume (double label measurement of plasma and red cell volumes), pH, Paco2, and serum concentrations of sodium, potassium, chloride, lactate, ionized calcium, phosphate, albumin, and total protein were measured before and 20 min after completion of ANH. Strong ion difference was calculated as serum sodium plus serum potassium minus serum chloride minus serum lactate. The amount of weak plasma acid was calculated using a computer program. ResultsAfter ANH, blood volume was well maintained in both groups. ANH caused slight metabolic acidosis with hyperchloremia and a concomitant decrease in strong ion difference. Plasma albumin concentration decreased after hemodilution with 6% hydroxyethyl starch solution and increased after hemodilution with 5% albumin solution. Despite a three-times larger decrease in strong ion difference after ANH with 6% hydroxyethyl starch solution, the decrease in pH was nearly the same in both groups. ConclusionsANH with 5% albumin or 6% hydroxyethyl starch solutions led to metabolic acidosis. A dilution of extracellular bicarbonate or changes in strong ion difference and albumin concentration offer explanations for this type of acidosis.

Instantaneous restoration of regional organ blood flow after severe hemorrhage: effect of small-volume resuscitation with hypertonic-hyperoncotic solutions.

The acute effects of small-volume infusion of hypertonic-hyperoncotic solutions on central hemodynamics, regional organ blood flow (RBF; 15-microns-diameter radiolabeled microspheres), and respiratory function following severe hemorrhage (MAP = 40 mm Hg for 45 min, approx 50% blood loss) were analyzed in anesthetized beagles. Treatment regimens used were: 10% Dextran 60 in 7.2% NaCl (HHS); 10% Dextran 60 in 0.9% NaCl (HDS); or 7.2% NaCl (HSS) alone, administered over 2 min in a volume equivalent to 10% of the blood loss (4 ml/kg iv). Within 5 min, cardiac output reached (HSS, HDS) or even exceeded prehemorrhage values (HHS), and MAP increased to 56% (HDS)-74% (HHS) of baseline. At the same time, RBF in kidneys (all groups), pancreas (HHS, HSS) and gastric mucosa (HHS) was completely restored, while flow in myocardium, brain, skeletal muscle, adrenal glands (all groups), and small intestine and colon (HHS) rose even above baseline values (P less than 0.05). Fractional blood flow (percentage of cardiac output) favored heart and brain in all three groups. These effects tended to persist for at least 30 min. Respiratory function was not affected by either of the three solutions, and no adverse effects were noted. Small-volume resuscitation with 7.2% saline/10% Dextran 60 provides instantaneous restitution of regional organ blood flow; it appears to result in a more uniform circulatory response than 7.2% saline or 10% Dextran 60 alone and might ensure improved organ perfusion during evacuation of patients from the accident site.

Hypertonic saline dextran resuscitation during the initial phase of acute endotoxemia : effect on regional blood flow

Background and MethodsSmall-volume resuscitation by means of bolus application of hypertonic saline solutions has been demonstrated to restore central hemodynamics and regional blood flow in severe hemorrhagic and traumatic shock. The aim of this study was to elucidate the potential of this new concept for treatment of profound hypovolemia and microcirculatory deterioration associated with sepsis and endotoxic shock.In a porcine model of acute hyperdynamic endotoxemia (elicited by continuous i v infusion of Salmonella abortus equi endotoxin for 3.5 hrs), small-volume resuscitation applying hypertonic-hyperoncotic solutions was analyzed for its effect on central hemodynamics, oxygen delivery (Do2), and regional blood flow. Fluid therapy was initiated when the pulmonary artery occlusion pressure (PAOP) tended to decrease (at 43 to 52 mins of endotoxemia), and consisted of 4 mL/kg bolus infusion of either 7.2% sodium chloride, 10% dextran, or 10% dextran in 7.2% sodium chloride; thereafter, PAOP was maintained by controlled infusion of 6% dextran-60. In a control group, 6% dextran-60 was given without preinjection of hypertonic-hyperoncotic solutions. ResultsOn small-volume resuscitation, cardiac index significantly increased within 5 mins in all groups, while mean arterial pressure remained unchanged. Fluid requirements were significantly reduced after small-volume resuscitation and the hyperdynamic circulatory state was maintained until the end of the observation period; Do2 as well as blood flow to heart, kidneys, and splanchnic organs remained high. ConclusionSmall-volume resuscitation by means of hypertonic saline-dextran proved the most effective, and seems to be an attractive supportive therapy to prevent microcirculatory failure in sepsis and endotoxemia.

First Clinical Implications of Perioperative Red Cell Volume Measurement with a Nonradioactive Marker (Sodium Fluorescein)

PhD Institute of Anesthesiology, Ludwig-Maximilians-Universitat, Klinikum Grophadern, Miinchen, Germany We measured red cell volume (RCV) with the nonradio- active marker sodium fluorescein (SoF) in 30 patients undergoing gynecological operations. Sixteen patients underwent preoperative isovolemic hemodilution (PIHD). RCV measurements were performed before and after PIHD and at the end of the operation. All RCVs were related to corresponding hematocrit (hct) levels. We report a simplified method for its clinical ap- plication by reducing the number of blood samples re- quired. To validate our method, we compared RCV within the PIHD bags (bag RCV) with the difference of the patients’ RCV before and after PIHD. Bag RCV ob- tained during PIHD (mean 399 -’ 81 mL) was measured with a precision of 4.2% by using SoF. There was a sig- nificant difference (mean 286 ? 401 mL; P < 0.05) be- tween intraoperatively estimated and measured blood loss. The blood loss tended to be underestimated and, in some cases, was underestimated or overestimated sub- stantially. Preoperative and postoperative hct values only offered an imprecise estimation of the patients’ RCV. We conclude that RCV measurement using SoF is a precise method for monitoring changes in RCV dur- ing PIHD and surgical operation. Implications: We measured red cell volume changes of 30 patients with the nonradioactive marker sodium fluorescein before and after hemodilution and postoperatively with a high precision. We frequently found large differences be- tween intraoperatively estimated and measured blood loss. Preoperative hematocrit values offered an impre- cise estimation of the patients’ red cell volume. (Anesth Analg 1998;87:1234-8)

Effects of adenosine on cardiopulmonary functions and oxygen-derived variables during endotoxemia

Abstract Objectives: To determine the effects of a prophylactic intravenous infusion of adenosine on cardiopulmonary functions and oxygen‐derived variables in a porcine model of endotoxemia. Design: Prospective, randomized, placebo‐controlled, unblinded study. Setting: University research laboratory. Subjects: Thirty country bred pigs, aged 6 to 7 wks, weighing 24.9 +/‐ 0.65 (SEM) kg body weight. Interventions: Pigs were anesthetized by iv pentobarbital and fentanyl, intratracheally intubated, and mechanically normoventilated with a gas mixture of nitrous oxide/oxygen = 1:1. Intravascular catheters were inserted to allow for determination of arterial, central venous blood pressure, pulmonary artery occlusion pressure, cardiac output, and sampling of blood for gas analyses. Group 1 (n = 10) received a 330‐min intravenous infusion of Salmonella abortus equi endotoxin (5 [micro sign]g/kg body weight x hr). Group 2 (n = 10) received an additional intravenous infusion of adenosine (150 [micro sign]g/kg body weight x min), started 30 mins before the infusion of endotoxin. Control groups 3 and 4 (n = 5 for both groups) received adenosine or physiologic saline, respectively. Measurements and Main Results: Parameters of cardiopulmonary function and oxygen‐derived variables were calculated from pulmonary artery catheter measurements and blood gas analyses using standard formula. Plasma concentrations of purine compounds (adenosine, inosine, hypoxanthine) were determined by high‐performance liquid chromatography. Since tumor necrosis factor‐alpha plays a central role in the development of endotoxic shock, concentrations of this cytokine were determined in serum by enzyme‐linked immunosorbent assay. Infusion of adenosine before the beginning of the infusion of endotoxin increased plasma concentrations of the nucleoside from 193 +/‐ 72 to 553 +/‐ 65 nmol/L and decreased the systemic vascular resistance by 50%. Although acting as a potent vasodilator under control conditions, adenosine did not aggravate the arterial hypotension elicited by endotoxemia but significantly increased cardiac output by a comparably small decrease in systemic vascular resistance, prevention of pulmonary vasoconstriction, and improvement of left ventricular performance. Despite significant pulmonary vasodilation, gas exchange was not worsened but slightly improved by adenosine. With the increase in cardiac output and arterial oxygenation, systemic oxygen delivery almost doubled. This adenosine‐induced oxygen flux was not a surplus but was most likely utilized by tissues, as suggested by the much earlier beginning of the increase in the systemic oxygen consumption and the attenuation of the decrease in the gastric mucosal pHi. No effects of adenosine were observed on the endotoxin‐induced increase in serum concentrations of tumor necrosis factor‐alpha. Conclusions: Infusion of adenosine might be useful to improve flow‐dependent oxygen delivery and tissue oxygenation during endotoxic shock without the induction of adverse cardiopulmonary side effects. The beneficial hemodynamic effects of adenosine appear not to be mediated by the inhibition of the release of tumor necrosis factor‐alpha. (Crit Care Med 1998; 26:322‐337) Endotoxic or septic shock represents a peripheral vascular failure which is characterized by compromised arteriolar vasomotion, persistent vasodilation, vasoconstriction, and microembolization, as well as an increase in capillary permeability [1]. The loss in the surface area of transcapillary exchange, the overloading of the interstitium with fluid, the formation of cellular edema, and the direct inhibition of mitochondrial respiration by bacterial toxins may all combine to contribute to the development of functional shunts of oxygen at the cellular level. Besides studies in animals [2,3], which demonstrate the early onset of severe vasoconstriction and the decrease in blood flow to the mucosa of the intestinal tract within a few minutes in response to endotoxemia, the measurement of gastric mucosal pHi in septic patients also provides indirect evidence for impaired mucosal oxygenation [4,5]. In addition, using laser Doppler fiberoptic flow probes in skeletal muscle, blood flow, and reactive hyperemia were shown to be severely depressed in septic patients [6]. Based on the assumption that early vasoconstriction and obstruction of the capillary bed (caused by the hosts inflammatory humoral and cellular response) play a pathophysiologically relevant role in compromising the microcirculation, and hence, in the subsequent development of tissue hypoxia, any therapeutic pharmacologic interventions should be aimed at increasing nutritive blood flow to tissues, e.g., by the use of potent vasodilators. Accordingly, vasodilatory prostaglandins (PgI2, PgE1/2) were shown in numerous animal models of endotoxic shock and in critically ill patients to improve significantly cardiac output, systemic oxygen delivery, and‐although not consistently‐systemic oxygen consumption [7‐9]. In addition to their vasodilatory action, the prostaglandins mentioned above have been shown in vitro to exert a variety of anti‐inflammatory effects on thrombocytes and neutrophils, which is considered to be useful in protecting the microcirculation from inflammatory injury. Thus, the administration of vasodilatory prostaglandins appears to be promising. However, this therapeutic approach is limited by hemodynamic side effects in the pulmonary and systemic circulation [10,11]. In general, irrespective of the vasodilatory compound applied intravenously (e.g., prostaglandins, calcium antagonists, beta2‐agonists, nitric oxide donors, etc.), dilation of lung blood vessels is likely to hamper hypoxic pulmonary vasoconstriction, to increase pulmonary shunt perfusion, and hence, to impair pulmonary gas exchange. Moreover, indistinct use of vasodilators may cause profound systemic arterial hypotension. Another endogenously produced, potent vasodilator with a broad spectrum of anti‐inflammatory actions [12] is adenosine which, like prostaglandins, might increase blood flow to tissues. In states of shock, however, adenosine might also cause hemodynamic instability and worsen gas exchange by vasodilation in the systemic and pulmonary circulations. Because exogenous adenosine has not been tested during endotoxic shock so far, it was the aim of the present study to characterize the effects of a continuous prophylactic infusion of this nucleoside on cardiopulmonary function, oxygen‐derived variables, and on the pHi, as a measure of the mucosal acid‐base balance, in a standardized porcine model of endotoxemia. The results obtained suggest that the iv infusion of adenosine might be useful to improve flowdependent oxygen delivery and tissue oxygenation during endotoxic shock without the induction of adverse cardiopulmonary side effects.

Four cases of radical hysterectomy with acute normovolemic hemodilution despite low preoperative hematocrit values.

The study was approved by the institutional ethics committee at our institution with all patients giving their written, informed consent. Four patients from a continuing study with preoperative diagnosis of carcinoma of the cervix are presented. They had a preoperative Hct of ,33%, were ASA physical status I or II, were without cardiovascular, pulmonary, or hormonal dysfunctions, and were scheduled for radical hysterectomy. They fasted for 10 h, and neither preoperative bowel cleansing nor any preoperative infusions were administered before measurement. After the induction of general anesthesia and insertion of central and arterial catheters, baseline measurements of plasma volume (PV), Hct, and RCV were performed. (For measuring procedures, see below.) These baseline measurements showed that all four patients had low preoperative Hct values because of extraordinarily large PVs rather than small RCVs. Therefore, ANH to a target Hct of 20% was performed. A median of 1,300 mL of blood was removed and simultaneously replaced by 1,550 mL of colloid (5% albumin solution or 6% hydroxyethylstarch). After completion of ANH and a steady-state interval of 30 min without any further infusions, simultaneous measurements of PV, Hct, and RCV were performed. After these measurements, the surgical procedure began. The intraoperative transfusion trigger for beginning retransfusion was a Hct of 16%. After having reached this Hct value, the fraction of inspired oxygen was switched from 0.5 to 1.0. All autologous blood was retransfused in every patient late in surgery when major blood loss had ceased. Postoperative measurements of PV, Hct, and RCV were performed immediately after closure of the abdominal wall during a period of stable anesthesia without obvious blood loss. Preoperative PV and Hct measurements before ANH and postoperative measurements were performed in duplicate in a time interval of 30 min without any infusions.

Influence of hydroxyethyl starch (6% HES 130/0.4) administration on hematology and clinical chemistry parameters

Abstract Background: The chemical inertness of hydroxyethyl starch (HES) might cause interferences of the colloid with a variety of laboratory tests. We aimed to evaluate potential influences of HES 130/0.4, the newest HES type, on several common hematology and clinical chemistry parameters. Methods and results: A convenient sample of 25 patients scheduled for rheological therapy with 500 mL 6% HES 130/0.4 was evaluated. Blood samples were drawn before and after colloid application. Comparing pre- and post-infusion values of a battery of laboratory tests (i.e., hematology and hemostasis parameters, electrolytes, enzymes, kidney and metabolic parameters, lipids, etc.) in time course, a median difference greater than the reference change value for a specific parameter was considered clinically relevant. Among all parameters tested, only serum amylase activity displayed a clinically relevant difference between pre- and post-infusion values (median increase of 85% due to HES administration). By applying in vitro experiments, we demonstrated that serum amylase values obtained in the samples diluted in a 1:1 ratio with HES 130/0.4 and in samples diluted in a 1:1 ratio with 0.9% NaCl displayed a negligible median difference of 3%. Conclusions: The in vivo effect of HES 130/0.4 administration on serum amylase activity observed in our study was pharmacological (real) in nature. With the exception of the influence of HES 130/0.4 on amylase activity, the effects of HES 130/0.4 on other parameters tested in this study can be interpreted as having no clinical relevance. Clin Chem Lab Med 2008;46:558–62.

Buflomedil hydrochloride reduces systemic activation of polymorphonuclear leukocytes during hyperdynamic endotoxemia

The purpose of the study was to examine the effects of buflomedil hydrochloride (BFL) on the expression of adhesion molecules (β2-integrins) and oxygen radical production of circulating and emigrated intra-abdominal polymorphonuclear leukocytes (PMNL) in a standardized porcine model of hyperdynamic endotoxemia. A total of 20 anesthetized pigs were randomly assigned to one of the following groups: endotoxin group (endotoxin 5 μg/kg-h, intravenously (i.v.), n = 7), BFL group (BFL (3 mg/kg initial i.v. bolus followed by a continuous infusion (.1 mg/kg.h) and endotoxin 5 μg/kg.h, i.v., n = 7), and control group (NaCI.9%; n = 6). Experiments were terminated at 330 min. Infusion of endotoxin alone resulted in the activation of circulating and emigrated PMNL as evidenced by neutropenia, functional and numerical up-regulation of β32-integrins, and enhanced oxygen radical production. BFL was able to attenuate functional and numerical up-regulation of β2-integrins as well as oxygen radical production of circulating and emigrated PMNL. An unexpected decrease in the plasma concentration of BFL during the experiments was associated with an increase in the oxygen radical production of PMNL at the end of the experiments. In addition, BFL attenuated the fall in the intramucosal pH and the increase in plasma concentration of lactate observed in the late phase of endotoxemia. These findings suggest that BFL is able to decrease systemic activation of PMNL and to improve local tissue oxygenation during endotoxemia.