Splanchnic ischaemia and its role in multiple organ failure

Multiple organ failure remains the leading cause of death in the intensive care unit. Increasing numbers of investigators have focused their attention on the role of gastrointestinal tract in the pathogenesis of this syndrome. Their data indicate that inadequate gut perfusion leads to a measurable imbalance between oxygen delivery and the needs of the tissues, i.e., ischaemia. Gut ischaemia of sufficient duration impairs gastrointestinal tract barrier function, facilitating the passage of enteric bacterial endotoxin into the circulation. It has been hypothesized that production of tumor necrosis factor alpha, and other biologic mediators by endotoxin-stimulated macrophages, triggers a generalized and uncontrolled inflammatory response that ultimately leads to multiple organ failure. Preliminary evidence suggests that survival can be improved significantly if gut ischaemia is promptly identified and aggressively treated by administration of fluids and inotropic drugs, using gastric intramucosal pH as the therapeutic endpoint. Future studies are needed to determine whether additional treatment modalities can improve outcome once the inflammatory response has fully developed.

Safety and efficacy of intravenous Carbicarb in patients undergoing surgery: comparison with sodium bicarbonate in the treatment of mild metabolic acidosis. SPI Research Group. Study of Perioperative Ischemia

OBJECTIVES

To compare the safety and efficacy of intravenous Carbicarb with intravenous sodium bicarbonate in well-oxygenated patients who developed metabolic acidosis while undergoing major surgery. Carbicarb is an equimolar solution of sodium bicarbonate and sodium carbonate (Na2CO3). It does not undergo significant breakdown to CO2 and H2O, nor does it increase CO2 concentrations to the same extent as does pure sodium bicarbonate. Because of these characteristics, Carbicarb may be a more suitable agent than bicarbonate in the treatment of metabolic acidosis.

DESIGN

Prospective, double-blind, randomized, multicenter trial.

SETTING

Veterans Affairs Medical Center (a teaching hospital of the University of California, San Francisco), and the University of Massachusetts Medical Center, Worcester, MA.

PATIENTS

We prospectively studied 36 patients who underwent either cardiac surgery or major noncardiac surgery and developed intraoperative metabolic acidosis (pH < 7.35 and whose serum bicarbonate concentration decreased by > 3 mmol).

INTERVENTIONS

Patients were randomly assigned to receive either sodium bicarbonate (1 mEq sodium/mL, n = 18) or 1 mol Carbicarb (1 mEq sodium/mL, n = 18) administered by intravenous bolus over a 30-sec period.

MEASUREMENTS AND MAIN RESULTS

For Carbicarb-treated patients, the mean arterial pH increased from 7.31 +/- 0.008 (baseline) to 7.36 +/- 0.009 10 mins after treatment; for the sodium bicarbonate-treated patients, the mean pH increased from 7.31 +/- 0.006 to 7.37 +/- 0.01. The increases in pH were statistically significant for both groups (p = .0001). There was no statistically significant difference between treatment groups in the number of repetitions of initial dose that was required to correct acidosis. Hemodynamic variables remained unchanged in both treatment groups during the study period, with the exception of the mean cardiac output which increased from 6.1 +/- 0.4 (baseline) to 6.9 +/- 1.4 L/min (60 mins after treatment) in a subset of Carbicarb-treated patients and decreased from 6.7 +/- 1.3 to 6.0 +/- 1.2 L/min in a subset of sodium bicarbonate-treated patients, p = .048 (between groups); and the mean pulmonary artery occlusion pressure decreased from 19 +/- 2 mm Hg (baseline) to 8 +/- 3 mm Hg (45 mins after treatment) in the Carbicarb-treated patients, and decreased from 18 +/- 2 to 13 +/- 4 mm Hg in the sodium bicarbonate-treated patients, p = .012 (between groups). Systemic utilization of lactate increased from 0.3 +/- 1.0 mmol/min (baseline) to 5.6 +/- 4.3 mmol/min (45 mins after treatment) in Carbicarb-treated patients, and increased from 1.0 +/- 0.6 mmol/min (baseline) to 1.5 +/- 1.3 mmol/min in the sodium bicarbonate-treated patients, p = .033 (between groups). The administration of Carbicarb was safe. No patients were discontinued from the study because of adverse events.

CONCLUSIONS

Carbicarb corrects metabolic acidosis as well as sodium bicarbonate. However, the potential therapeutic advantage of Carbicarb remains to be determined, especially in patients with more severe metabolic acidosis.

Correction of intramyocardial hypercarbic acidosis with sodium bicarbonate

Although it has been hypothesized that exogenously administered bicarbonate can exacerbate intramyocardial acidosis and compromise contractile function, this phenomenon has not been demonstrated in an intact model in which intramyocardial pH (pH(int)), regional venous pCO2, and regional contractile function have been simultaneously monitored. In 20 anesthetized dogs, we studied the effects of intracoronary infusions of sodium bicarbonate NaHCO3 30 mEg over 15 min, on regional pH(int), (glass electrode) and regional stroke work (SW, sonomicrometry) before and after creating systemic hypercarbic acidosis by hypoventilation. During NaHCO3 administration, regional coronary venous pCO2 increased rapidly during the first minute (eucapnea; 34 +/- 7 to 55 +/- 18 mm Hg; hypercapnea: 70 +/- 15 to 98 +/- 23 mm Hg, P < 0.05 for both increases). Regional venous pH rose from 7.36 +/- .04 to 7.55 +/- .06 (P < 0.05) after the first minute of NaHCO3 infusion during eucapnea and from 7.09 +/- .09 to 7.22 +/- .09 (P < 0.05) during hypercapnea. During the first minute of NaHCO3 infusion, pH(int) declined minimally. However, during the remaining 14 min of each infusion, pH(int) increased significantly (eucapnea: 7.19 +/- 0.10 to 7.43 +/- 0.12; hypercapnea: 6.86 +/- 0.14 to 7.02 +/- 0.15, P < 0.05 for both changes). Regional SW decreased significantly during the first minute of infusion, both during eucapnea (23,400 +/- 7,400 to 18,000 +/- 6,300 ergs/cm2, P < 0.05) and hypercapnea (27,000 +/- 9,100 to 25,000 +/- 10,000 ergs/cm2, P < 0.05). The first minute of contractile dysfunction was followed by recovery and ultimately supranormal contractile function during the remainder of each bicarbonate infusion. To test the hypothesis that transient intracellular acidosis during bicarbonate infusions was underestimated by measurements of pH(int), measurements of intracellular pH using the pH-sensitive dye, BCECF, were performed in isolated guinea pig papillary muscles incubated in vitro. These measurements confirmed the presence of transient intracellular acidosis during bicarbonate infusion. In conclusion, (1) the intracoronary administration of sodium bicarbonate causes a transient depression in myocardial contractile function that is related to transient intracellular acidosis; and (2) despite exacerbating hypercarbia, sodium bicarbonate ultimately neutralizes intracellular acid and augments myocardial contractile function.

Thoracic impedance and pulmonary atrial natriuretic peptide during head-up tilt induced hypovolaemic shock in humans

Head up and down tilts were used for manipulating the central blood volume in eight volunteers. During head-up tilt thoracic electrical impedance (TI) increased from 36.7 (33.9-52.1) ohm (mean and range) to 41.9 (36.9-59.2) ohm, heart rate from 60 (49-72) to 80 (65-90) beats min-1 (P < 0.05) and decreased again to 57 (48-67) beats min-1 accompanying a fall in mean arterial pressure from 86 (76-97) to 54 (41-79) mmHg and in cardiac output from 9.2 (5.9-12.1) to 6.9 (3.4-8.8) 1 min-1 (n = 7, P < 0.07). Central venous pressure did not change significantly. Pulmonary arterial mean, 6 (3-12) mmHg, and wedge pressures, 4 (1-9) mmHg, decreased to 4 (1-11) and 1 (0-7) mmHg, respectively, and mixed, 78 (77-79%), and central venous oxygen saturations, 72 (71-73)%, fell to 62 (46-75) and 54 (44-58)%, respectively (P < 0.05). Atrial natriuretic peptide (ANP) was determined from blood of the superior vena cava and pulmonary and brachial arteries. Pulmonary artery ANP, 18.4 (7.5-30.7) pmol l-1, was higher than in vena cava, 13.3 (5.2-20.9) pmol l-1 (P < 0.05). At the time of presyncope, pulmonary artery ANP decreased from 20.8 (37.4-10.1) to 13.7 (19.7-5.7) pmol l-1, in vena cava from 13.8 (23.1-7.1) to 10.2 (17.9-6.7) pmol l-1 and in the brachial artery from 16.9 (34.1-5.2) to 11.3 (18.5-5.1) pmol l-1 (P < 0.05). Head-down tilt did not affect the recorded variables significantly. Thoracic electrical impedance, pulmonary artery pressure and venous oxygen saturations were sensitive indices of the central blood volume as reflected in the release of atrial natriuretic peptide from the right side of the heart.

Use of ultrahigh frequency ventilation in patients with ARDS. A preliminary report

STUDY OBJECTIVE

Our objective was to compare the efficacy of ultrahigh frequency ventilation (UHFV) (frequencies > 3 Hz) with respect to oxygenation, airway pressures, and hemodynamic parameters in patients with adult respiratory distress syndrome (ARDS) who were not responding to conventional ventilation.

DESIGN

We used a prospective, multicenter, nonrandomized study design in which each patient served as his own control.

SETTING

Three university-affiliated, tertiary-care medical centers participated.

PATIENTS

Persons aged 16 to 79 years old with ARDS and unresponsive to conventional ventilation, as defined by a Food and Drug Administration (FDA) approved protocol, were included.

INTERVENTIONS

Ninety patients who were not responding to conventional ventilation were changed to UHFV using a microcomputer-controlled device.

MEASUREMENTS AND RESULTS

The patient's blood gas, hemodynamic, and airway pressure variables were measured just before, and at 1 and 24 h after the switch to UHFV. We demonstrated clinically significant improvements in arterial oxygen tension (PaO2) and reductions in peak and mean inspiratory pressures.

CONCLUSIONS

In a multicenter study, UHFV improved respiratory gas exchange and reduced airway pressure variables at both 1 h and 24 h after the onset of UHFV when compared with conventional ventilation just prior to the change and without hemodynamic deterioration, in patients with severe ARDS.

Sodium bicarbonate versus Carbicarb in canine myocardial hypercarbic acidosis

The objective of this study was to compare the in vivo effects of sodium bicarbonate (NaHCO3) and Carbicarb infusion on regional contractile performance and acid-base status in the setting of hypercarbic acidosis. Animals (N = 9) were anesthetized and paralyzed using sodium pentothal, halothane, and pancuronium bromide, and mechanically ventilated with an air-O2 mixture so that arterial PO2 was > or = 300 mm Hg. Following beta-adrenergic blockade, alveolar ventilation was gradually reduced over a 50-minute period to increase arterial PCO2 to 60 to 80 mm Hg. Each of the following solutions was then infused in consecutive order directly into the left anterior descending artery coronary artery for 15 minutes: (1) 8.4% NaHCO3 at 2 mL/min; (2) 5% sodium chloride at 2 mL/min, equivalent to NaHCO3 in osmolality; (3) 6.3% Carbicarb at 0.5 mL/min, equivalent to NaHCO3 in buffer capacity; and (4) 6.3% Carbicarb at 2 mL/min, equivalent to NaHCO3 in volume. Regional stroke work analog (ultrasonic dimension transducers), interstitial myocardial pH (Khuri electrode), coronary blood flow (doppler flow probe), and hemodynamic/metabolic variables (heart rate, blood pressure, arterial and coronary venous blood gases) were measured at 1, 5, 10, and 15 minutes during each infusion and 10 minutes after the infusion was discontinued, ie, at 25 minutes. Animals were allowed to recover for 45 minutes between interventions. Values at each time point were compared with baseline for statistical significance. Small reductions in interstitial myocardial pH (P < .05) and stroke work (P > .05) were observed within 1 minute of NaHCO3 administration. Both parameters increased significantly from baseline levels thereafter, ie, interstitial myocardial pH at 5 minutes and stroke work at 15 minutes. Infusion of Carbicarb invariably was associated with an increase (P < .05) in interstitial myocardial pH. Stroke work increased (P < .05) during low-dose Carbicarb administration, but infusion of the higher dose was accompanied by a biphasic response, ie, an increase (P < .05) from 0 to 5 minutes, followed by a gradual decrease that achieved statistical significance 10 minutes after termination of the infusion. End-diastolic length was inversely proportional to changes in stroke work, and coronary blood flow varied directly with changes in coronary venous Pco2. Myocardial O2 consumption decreased (P < .05) during Carbicarb infusion, but changes during NaHCO3 did not reach statistical significance. Our findings lend support to the hypothesis that intramyocardial pH determines myocardial function independent of CO2 production by buffer therapy.(ABSTRACT TRUNCATED AT 400 WORDS)

Association between gastric intramucosal pH and splanchnic endotoxin, antibody to endotoxin, and tumor necrosis factor-alpha concentrations in patients undergoing cardiopulmonary bypass

OBJECTIVES

To determine the association between gastric intramucosal pH, a minimally invasive marker reflecting the adequacy of oxygen delivery to the gastrointestinal tract, and splanchnic endotoxin, antibody to endotoxin, and tumor necrosis factor (TNF)-alpha concentrations in patients undergoing cardiopulmonary bypass.

DESIGN

Single-arm, prospective study.

SETTING

University hospital.

PATIENTS

Adults (n = 10) free of hepatic, pulmonary, and renal disease undergoing nonemergent coronary artery bypass surgery.

INTERVENTIONS

After induction of general anesthesia and endotracheal intubation, a tonometer nasogastric tube was positioned in the stomach, and triple-lumen fiberoptic catheters were inserted into the hepatic vein and pulmonary artery. Hepatic venous and mixed venous blood samples were analyzed for endotoxin, antibody to endotoxin, and TNF-alpha at six times: 30 mins after induction of anesthesia (time 1); during vena caval cannulation (time 2); after 15 mins of hypothermic cardiopulmonary bypass (time 3); during spontaneous left ventricular ejection after release of the aortic cross-clamp, but before termination of cardiopulmonary bypass (time 4); 15 mins after termination of cardiopulmonary bypass (time 5); and 1 hr after termination of cardiopulmonary bypass (time 6). Gastric intramucosal pH, systemic oxygen delivery (DO2), mixed venous oxygen saturation, hepatic venous oxygen saturation, and hepatic venous lactate concentrations were recorded at these same times. Data for each variable were compared with baseline values (time 1) for statistical significance.

MEASUREMENTS AND MAIN RESULTS

Cardiopulmonary bypass was associated with an increase (p < .05) in systemic endotoxin concentrations from ventricular ejection until the end of the study. Virtually identical changes in the splanchnic circulation at this time approached, but did not reach, statistical significance, because hepatic venous endotoxin concentrations were higher than the mixed venous endotoxin concentrations at baseline (41.6 +/- 11.2 vs. 16.9 +/- 4.9 pg/mL). Gastric intramucosal pH was abnormal (< 7.35) at 15 mins (p > .05) and at 1 hr after termination of cardiopulmonary bypass (p > .05). The relationship between endotoxin and gastric intramucosal pH was not statistically significant (p = .15). The decrease in endotoxin antibody was small and statistically insignificant. TNF-alpha was not detected in any patient. Systemic DO2 decreased (p < .05) after 15 mins of hypothermic cardiopulmonary bypass, but returned to baseline values thereafter. There were no significant changes in mixed venous and hepatic venous oxygen saturation values. Splanchnic lactate concentrations increased at cannulation (p < .05), after 15 mins of hypothermic cardiopulmonary bypass (p < .05), and 15 mins after termination of cardiopulmonary bypass (p < .05).

CONCLUSIONS

These observations are consistent with the hypothesis that impaired gut-barrier function is responsible for endotoxemia occurring during cardiopulmonary bypass. It is unclear whether increased mucosal permeability and mucosal acidosis are causally related phenomena or simply independent markers of damage to gut epithelium.

Splanchnic lactate production in cardiac surgery patients

OBJECTIVE

To review the pathophysiology of lactic acidosis in patients undergoing open-heart surgery, with special reference to the splanchnic circulation.

DATA SOURCE

MEDLINE search of pertinent experimental and clinical research studies.

RESULTS

Lactate is an end-product of anaerobic metabolism and is in dynamic equilibrium with its precursor, pyruvate. The ratio of serum lactate to pyruvate concentrations in arterial blood is normally < or = 10:1. In patients with lactic acidemia, measurement of serum pyruvate concentrations may yield valuable clinical information. Lactate/pyruvate ratios > 10:1 suggest that oxygen delivery (DO2) is inadequate to meet metabolic demand, whereas increases in both lactate and pyruvate values with preservation of normal lactate/pyruvate ratios suggest a defect in oxidative utilization (e.g., a fractional increase in the inactive form of the pyruvate dehydrogenase enzyme complex) despite adequate DO2. Meaningful changes in regional oxygen kinetics occur during extracorporeal circulation. Increased production of endogenous vasoconstrictors in response to nonpulsatile flow, together with a decrease in arterial oxygen content resulting from the addition of 2 L of pump prime to the patient's circulation at the beginning of cardiopulmonary bypass, decrease DO2 to the gastrointestinal tract. The effect of this reduction is mitigated, in part, by lowering core temperature and reducing tissue oxygen demand.

CONCLUSIONS

The abdominal organs tolerate reductions in DO2 when they are cold (25 degrees C), since gastric intramucosal pH (a marker of inadequate DO2), and hepatic venous lactate/pyruvate ratios and oxygen saturation during the first half of cardiopulmonary bypass are normal. As surgery nears completion and core temperature is increased, tissue oxygen demands escalate. The presence of gastric mucosal acidosis, coupled with lactic acidemia and oxygen desaturation of hepatic venous blood, suggest that delivery of oxygen to the abdominal organs at the conclusion of cardiopulmonary bypass is insufficient to meet demand. A growing proportion of cardiac surgery patients are older and many have concomitant medical problems that can impair their recovery. Useful strategies are needed to reduce the occurrence of splanchnic ischemia in these and other high-risk populations if surgical outcome is to improve in the future.

Gastric tonometry and venous oximetry in cardiac surgery patients

OBJECTIVE

To determine the relationship between gastric intramucosal pH and several other indices of splanchnic perfusion in patients undergoing cardiopulmonary bypass.

DESIGN

Prospective, single-arm study.

SETTING

University Hospital.

METHODS

Elective cardiac surgery patients (n = 8), free of hepatic disease, were studied. Before anesthetic induction, a triple-lumen, heparin-bonded fiberoptic catheter was inserted into the hepatic vein under fluoroscopic guidance. An identical catheter was inserted into the pulmonary artery. After endotracheal intubation, a nasogastric tube modified to permit measurement of gastric intramucosal pH was inserted into the stomach. Systemic oxygen delivery (DO2), and arterial, mixed venous, hepatic venous, and femoral venous blood gases and lactate concentrations were recorded at the following times: immediately before induction of anesthesia (time 1); during atrial cannulation (time 2); after 30 mins of hypothermic cardiopulmonary bypass (time 3); 15 mins after termination of cardiopulmonary bypass (time 4); and 1 hr after arrival in the ICU (time 5). Hepatic venous hemoglobin saturation (SO2) and mixed venous hemoglobin saturation (SvO2) were monitored continuously from times 1 to 5. Gastric intramucosal pH was recorded at times 2, 3, 4, and 5. The hepatic catheter was removed as soon as the last samples were collected in the ICU.

RESULTS

The square of the weighted mean correlation coefficients (rw)2 for gastric intramucosal pH vs. hepatic venous lactate concentrations, gastric intramucosal pH vs. hepatic venous PO2, and gastric intramucosal pH vs. hepatic venous pH were (rw)2 = .50, (rw)2 = .58, and (rw)2 = .32, respectively. Systemic DO2, hepatic venous lactate concentrations, hepatic venous PO2, and hepatic venous pH were significant determinants in the multiple regression model for gastric intramucosal pH (r2 = .89). There were significant differences between SvO2 and hepatic venous SO2 at times 4 and 5.

CONCLUSION

Gastric intramucosal pH may provide a minimally invasive way to monitor the adequacy of splanchnic DO2 in patients undergoing cardiopulmonary bypass. Additional data are necessary to determine whether low gastric intramucosal pH is truly a marker of supply-dependent oxygen uptake across the hepatosplanchnic vascular bed under these conditions.

An improved lidocaine infusion protocol for cardiac surgical patients

A new protocol for lidocaine administration was tested to determine whether it would provide higher free and total serum lidocaine concentrations during and after weaning from cardiopulmonary bypass (CPB), without leading to accumulation toxicity, than those resulting from a conventional protocol (1.5 mg/kg loading dose bolus + 2 mg/min infusion rate). Ten elective adult cardiac surgical patients were studied. Ten seconds prior to aortic cross-clamp release (0 min), each patient received a lidocaine bolus (1.5 mg/kg) and simultaneous lidocaine infusion (5 mg/min for 1 hour, followed by 2 mg/min for 23 hours). Weaning occurred 20 to 30 minutes after cross-clamp release. Lidocaine levels were determined from arterial blood samples at 0, 10, 20, 30, and 60 minutes, and at 2, 4, and 24 hours postcross-clamp release. Compared with the conventional protocol, the new protocol showed a significant increase in mean total lidocaine concentration (P less than or equal to 0.02) at 30 and 60 minutes, and a significant increase in mean free lidocaine concentration (P less than or equal to 0.005) at 20, 30, and 60 minutes. No patient had toxic levels at 24 hours. Results of this study suggest that in patients undergoing cardiac surgery, a "5 + 2" mg/min infusion protocol is safe and superior to a conventional protocol in achieving therapeutic serum levels of free and total lidocaine.

Free and total lidocaine levels in cardiac surgical patients

A lidocaine bolus of 2 mg/kg has been shown to decrease the incidence of ventricular fibrillation in cardiac surgical patients during reperfusion following aortic cross-clamp release (CCR). In an effort to extend this effect into the immediate postoperative period, many centers routinely administer a prophylactic lidocaine infusion for the first 12 to 24 hours after surgery. Sequential total serum lidocaine levels were measured over 6 hours in 28 adult cardiac surgical patients without liver, kidney, or respiratory disease, and free serum lidocaine levels were measured in 15 of these patients. In addition, pharmacokinetic data were collected and analyzed in 8 of these patients. Each patient received a lidocaine bolus of 1.5 mg/kg and an infusion at 2 mg/min immediately before CCR. Serum levels were determined just before CCR, and at 10, 20, 30, 60, 120, 240, and 480 minutes after CCR. More than 50% of patients had subtherapeutic total serum lidocaine levels from 20 to 120 minutes after CCR, but free lidocaine levels remained within the therapeutic range. No patient showed malignant ventricular dysrhythmias during the study. The results suggest that (1) it may be misleading to estimate free lidocaine concentration based on total lidocaine levels; and (2) free drug is the moiety responsible for lidocaine's antiarrhythmic effect.

Efficacy of large-bore intravenous fluid administration sets designed for rapid volume resuscitation

We tested the hypothesis that large-bore iv tubing sets provide greater in vitro flow rates than conventional sets currently utilized for the administration of normal saline. The study was conducted in two parts. First, flow rates were measured using a conventional iv tubing set under gravity and 300 mm Hg of pressure, and four large-bore iv tubing sets under gravity. Additional measurements were made after attaching various catheters and sheaths ranging in size from 14 ga to 9 Fr. Five trials were conducted for each combination. Second, we measured the amount of time taken in a mock drill by 28 randomly chosen anesthesiology residents and staff to load a unit of packed RBC into a pressure infusor bag, inflate to 300 mm Hg, and deflate to ambient pressure. Data were analyzed for individual tubing sets as a whole and individual catheter-tubing set combinations, using the Mann-Whitney U test with Bonferroni adjustment. Statistically significant (p less than .03) differences in flow rates were noted between the conventional set under gravity compared with both the pressurized conventional set and the large-bore tubing sets under gravity. There was no statistical difference in flow rates between pressurized conventional iv sets and large-bore iv sets. Mean time necessary for the 28 anesthesiology staff and residents to pressurize and deflate a conventional tubing set was 65 sec/bag of packed RBC. We conclude that a considerable amount of time can be saved by utilizing large-bore iv tubing sets instead of conventional pressurized sets during massive volume resuscitation. Clinical trials are indicated to corroborate these results.

Current Trends in Neurosurgical Critical Care Medicine

We conducted a survey of neurosurgeons to discover how they manage specific problems commonly seen in critically ill neurosurgery patients and compared the responses with relevant investigations published in the neurosurgical literature over the past 10 years. The 101 Chiefs of Neurosurgery of every North American academic training program were sent copies of two consecutive surveys for distribution to their attending staff. Among the topics covered were indications for steroid administration, prophylaxis of gastrointestinal bleeding, indications for insertion of ICP monitoring devices and use of prophylactic antibiotics while they are in place, and fluid restriction in patients with cerebral edema. Replies were strictly anonymous. A Medline Computer search for the period 1976-1989 was carried out for each topic. We received 136 and 67 replies each to our two surveys. Over 90 clinical and laboratory studies published in major medical journals, many randomized and prospective, were reviewed. Two trends were evident: (a) a noticeable lack of uniformity among neurosurgeons with respect to management of the same problem; (b) little or no basis of support in the literature for many management practices. Based on these studies and the responses to our survey, we suggest an examination of current clinical practices.

Perioperative hemodilution

Perioperative hemodilution is one way of minimizing the morbidity and mortality associated with blood transfusion. A review of the literature indicates that, for many patients, substitution of crystalloid solutions for surgical blood loss is safe and physiologically advantageous as long as euvolemia is maintained. Increased stroke volume is the principal mechanism responsible for augmented tissue oxygen delivery. Based on this review the surgical population can be divided into three groups: most patients under 60 years of age (group 1); those with important cardiac disease (group 2); and the elderly (group 3). Patients in group 1 can safely undergo hemodilution to a hematocrit of 0.25; patients in the other groups should probably not be subjected to hemodilution. Implementation of this proposal could diminish the risk and expense of many surgical procedures.