Major improvements have been made over the last 30 years in the care of severely injured patients in the areas of preoperative management, [1-4] operative therapy, [5-9] and injury classification. [10-12] Although intraoperative techniques and trauma scoring systems have become more sophisticated,  current approaches to the evaluation of circulatory failure in trauma patients remain rather imprecise. Shock is commonly diagnosed by subjective clinical signs and symptoms, such as cold clammy skin, pallor, unstable vital signs, and mental confusion. [4,13] Furthermore, normal values of heart rate (HR), blood pressure (BP), and urine output may be inappropriate as resuscitation goals. [4,13] Moreover, clinical signs of hypovolemia are also nonspecific and insensitive;  heart rate and blood pressure measurements may remain normal despite significant volume loss. [14,15] In the surgical intensive care unit (SICU) setting, normal central venous pressure (CVP), pulmonary artery occlusion pressure (PAOP), and arterial blood gas (ABG) values are also commonly used as endpoints for volume therapy.  Because mean values of CVP, PAOP, and ABGs are similar in surviving and nonsurviving trauma patients,  their usefulness as therapeutic endpoints is questionable. Most importantly, using these parameters as resuscitation goals has not been shown to decrease the incidence of multiple organ failure, which accounts for a large number of trauma fatalities in the SICU. Clearly, more effective, outcome-related resuscitation goals are needed.
Much evidence exists which suggests that early postinjury increases in cardiac index (CI) and oxygen delivery are primary compensations that attempt to increase tissue oxygen consumption in order to replenish the oxygen debt incurred by the trauma. Eight years ago Schultz et al.  suggested that physiologic monitoring of patients with hip fractures might improve outcome. Scalea et al.  then observed that geriatric blunt trauma patients had a lower mortality when they underwent invasive monitoring in the early postinjury period. Moore et al.  reported a series of 34 severely injured patients in whom decreased oxygen consumption was associated with increased shock-related organ failures. Rady et al.  described 24 patients with severe blunt thoracic injury in which depressed CI and oxygen delivery index (Dosub 2 I) were associated with a poor outcome. Our group empirically described a series of 90 consecutively monitored severe trauma patients in which the 60 survivors had significantly higher postinjury values of CI, Dosub 2 I, and oxygen consumption index (Vosub 2 I) than did the 30 nonsurvivors; more importantly those patients who reached the mean survivor values CI, Dosub 2 I, and Vosub 2 I within 24 hours of admission had a significantly lower mortality and significantly fewer organ failures than those that did not.  Subsequently we reported, in a preliminary prospective trial including 67 patients,  a trend toward improved outcome when the empirically described survivor values of CI, Dosub 2 I, and Vosub 2 I were used as therapeutic endpoints of resuscitation; this trend did not reach statistical significance.
The present study describes a prospective trial of 115 patients in which the protocol group, using empirically described survivor values of CI, Dosub 2 I, and Vosub 2 I as resuscitation goals, had a significantly lower mortality and significantly fewer shock-related organ failures, intensive care unit days, and days on mechanical ventilation than did the control group who were managed with normal values as therapeutic goals.
MATERIALS AND METHODS
Criteria for Entrance into Study
The study protocol was approved by the hospital's Institutional Review Board. All trauma admissions and deaths were tracked over a one-year period. All patients greater than or equal to16 years of age were entered into the study if they met one or more of the following criteria: (1) an intraoperative blood loss (EBL) greater than or equal to2000 mL, and (2) a pelvic fracture or two or more major long bone fractures with transfusion of greater than or equal to four units packed red blood cells needed during the first six hours after admission to maintain the whole blood hemoglobin (Hgb) greater than or equal to10 gm/dL. Patients meeting the above criteria but with suspected intracranial injury were excluded if they had one or more of the following: (1) an acute traumatic lesion (subdural, epidural, or subarachnoid bleeding or cerebral contusion) visible on computed tomographic scan, (2) generalized cerebral edema and an abnormal Glasgow Coma Scale (GCS) score, or (3) a skull fracture and an abnormal GCS score. Patients qualifying for the study were prospectively randomized into either the protocol or the control arm depending on the day of admission.
For each study patient the following time intervals were recorded: hospital admission to operating room (OR) entry, OR entry to incision, length of operation, length of stay in postanesthetic recovery room (PAR), and hospital admission to SICU entry. The first measured BP and HR, the lowest systolic BP and highest HR, and the duration of hypotension (systolic BP < 100 mm Hg) were recorded in the ED, OR, and PAR. Blood loss in the peri-injury period was estimated from measured operative losses and/or the amount of blood needed in the first day after admission to maintain a Hgb greater than or equal to 10 gm/dL. Fluid intake (separated into crystalloids, colloids, and blood), output, and net balance were recorded in the ED, OR, and PAR and then daily in the SICU; cumulative fluid balance was also tracked. The Revised Trauma Score (RTS) and Injury Severity Score (ISS) were calculated for each patient.
In the SICU, all study patients underwent continuous BP monitoring using a percutaneously inserted radial or femoral artery catheter. Hemodynamic monitoring was performed using a balloon-tip flotation pulmonary artery catheter inserted via the subclavian, internal jugular, or femoral vein.
Final outcome, hospital days, SICU days, ventilator days, organ failures, and other complications were tracked for all patients. Organ failures were defined as follows: pulmonary, Paosub 2 /FIosub 2 less than or equal to 250 and/or pulmonary venous admixture (Qs/Qt) greater than or equal to20% with PAOP less than or equal to 18 mm Hg; [22,23] hepatic, serum bilirubin greater than or equal to2 gm/dL with serum transaminases greater than or equal to twice normal;  renal, serum creatinine greater than or equal to 2 mg/dL, or with pre-existing renal disease a serum creatinine twice that on admission;  hematologic, spontaneous bleeding (other than from gastrointestinal tract) in the face of normothermia with either serum fibrinogen less than or equal to100 mg/dL, serum fibrin split products or fibrin monomers present, and/or platelet count < 50,000/microL;  cardiac, CI less than or equal to 2.0 L/min/msup 2 with PAOP > 18 mm Hg;  central nervous system, GCS score less than or equal to 6 in the absence of intracranial pathology, pain medication, muscle relaxants or paralytic agents.  Sepsis was defined as follows:  presence of identifiable infection with two or more of the following conditions: (1) temperature > 38degreesC or <36degreesC, (2) heart rate > 90 beats/minute, (3) respiratory rate > 20 breaths/minute, (4) Pacosub 2 < 32 mm Hg, and (5) white blood cell count > 12,000/mmsup 3 , <4,0000/mmsup 3 , or with >10% immature neutrophils.
Management of Protocol Patients
All protocol patients were monitored with a pulmonary artery (PA) catheter. Catheters were inserted in either the PAR or SICU within 6 hours after the operation, or within 12 hours after admission in those patients not undergoing immediate surgery. The goal of resuscitation for the protocol patients was to achieve a Dosub 2 I greater than or equal to 670 mL/min/msup 2 , a Vosub 2 I greater than or equal to 166 mL/min/msup 2 , and a CI greater than or equal to 4.5 L/min/msup 2 within 24 hours of admission and to maintain them for at least 48 hours thereafter. These goals were pursued first by volume loading to a PAOP of 18 mm Hg, and then by administering dobutamine (starting at 5 mcg/kg/min and increasing until desired effect was achieved or until HR > 130 beats/min) to increase the CI. Further increases in Dosub 2 I were attained by increasing the Hgb to as high as 14 gm/dL. Measurements of PAOP, CVP, pulmonary artery pressure (PAP), arterial and mixed venous blood gases, and CI were performed 15 to 20 minutes after each therapeutic intervention during the acute resuscitation and then every 6 hours for the first 3 days after admission. Values of Dosub 2 I and Vosub 2 I were calculated using standard formulas. The time from admission to attainment of the resuscitation goals was recorded for each patient.
Management of Control Patients
The resuscitation goals of the control group were a systolic BP greater than or equal to 120 mm Hg, a HR less than or equal to 110 beats/minute, a Hgb greater than or equal to 10 mg/dL, a urine output of 30 to 50 mL per hour, and, when monitored, a CVP of 8 to 12 mm Hg and a PAOP of 8 to 12 mm Hg. In the control patients monitored with a PA catheter (placed at the discretion of the particular surgeon managing the control patient), Dosub 2 I, Vosub 2 I, and CI were periodically measured but were not used as goals of resuscitation. Except for the differences in resuscitation goals, management of the control and protocol patients was identical.
All recorded time intervals, physiologic measurements, and fluid data were analyzed for protocol vs. control groups and again for survivors vs. nonsurvivors. The study group was also stratified by EBL (less than or equal to3000 mL, >3000 mL but less than or equal to5000 mL, >5000 mL but less than or equal to10,000 mL, and >10,000 mL) and compared for survivors vs. nonsurvivors. Mean values for RTS, ISS, EBL, BP, and HR data, and the measured time intervals at the ED, OR, and PAR were compared using the Student's t test. Mean values for all continuous physiologic variables (including CI, Dosub 2 I, and Vosub 2 I) were calculated over sequential intervals and compared using one-way analysis of variance. Mortality and the incidence of organ failures in the protocol and control groups were compared using2 analysis. Association between continuous variables was evaluated using linear regression. A p value < 0.05 was considered significant for all tests. Tabulated data is expressed as mean +/- SEM.
Characteristics of Study Group Patients
One hundred fifteen patients (50 protocol and 65 control patients) met entry criteria for the study. These 115 patients constituted 36% of the 321 SICU admissions and 4% of the 2784 hospital admissions during the 12-month study period. The 105 males and 10 females in the study group had a mean age of 30 +/- 2 years (range 16 to 85), a mean RTS of 6.6 +/- 1.6 (range 1 to 8), a mean ISS of 25 +/- 2 (range 16 to 60), and an EBL of 6273 +/- 478 mL (range 2000 to 27,000 mL). Thirty-four of 115 patients (30%) had an EBL less than or equal to 3000 mL (mean 2501 +/- 68), 33 of 115 patients (29%) had an EBL > 3000 mL and less than or equal to 5000 mL (mean 4336 +/- 107), 35 of 115 patients (30%) had an EBL > 5000 and less than or equal to 10,000 mL (mean 7411 +/- 278), and 13 of 115 patients had an EBL > 10,000 mL (mean 17,993 +/- 1680). Ninety-two of 115 (81%) patients were admitted with penetrating trauma; 23 of 115 (20%) patients sustained blunt injuries (Table 1). All 92 (100%) patients with penetrating injuries underwent emergent surgery, while 20 of 23 (83%) patients with blunt injury underwent major procedures (Table 1). Twenty of 115 (18%) study group patients underwent two or more major operations (range of two to six) during their hospital course.
The 50 protocol patients and the 65 control patients had comparable admitting diagnoses and had similar numbers of organ injuries (Table 1) and operations performed. The two study groups were well-matched with respect to RTS, ISS, TRISS, and ASCOT values, EBL, number of packed red blood cells transfused, and duration of hypotension (Table 2). The protocol and control patients spent similar amounts of time in the ED, OR, and PAR and had comparable vital signs in each of these departments.
The 50 protocol patients had a significantly lower mortality than did the 65 control patients (Table 3). Nine of 50 (18%) protocol patients died, while 24 of 65 (37%) control patients died (p = 0.03,2 analysis).
The 50 protocol patients had a total of 37 organ failures (19 respiratory, 6 renal, 7 hepatic, 3 septic, and 2 hematologic), or 0.74 +/- 0.28 organ failures per patient (Table 3). The 65 control patients had a total of 105 organ failures (38 respiratory, 16 renal, 29 hepatic, 12 sepatic, and 10 hematologic), or 1.62 +/- 0.28 organ failures per patient. This difference was statistically significant (p < 0.05). Nine of 16 (56%) protocol patients who had one or more organ failures died, while 25 of 41 (61%) control patients with one or more organ failure died.
The protocol group had significantly less SICU days and ventilator days per patient than did the control group (6 +/- 1 days vs. 11 +/- 2 days and 4 +/- 1 days vs. 9 +/- 1 days, respectively; p < 0.05 for both).
Predicted Mortality by TRISS and ASCOT Scoring
The RTS and ISS of the 34 nonsurvivors were lower and higher, respectively (5.5 +/- 0.4 vs. 7.0 +/- 0.2 and 31 +/- 4 vs. 23 +/- 3) than those of the survivors (Table 4). The overall survival in the protocol group was similar to that predicted by TRISS (81% and 84%, respectively) and ASCOT (81% and 89%, respectively); however, survival in the control group was significantly lower than the predicted values for TRISS (63% vs. 89%, p < 0.01) and ASCOT (63% vs. 91%, p < 0.01).
Evaluation of Protocol Group Resuscitation
Thirty-five of 50 (70%) protocol patients reached the resuscitation goals within 24 hours of hospital admission. Only five of these 35 patients (14%) died. Another six protocol patients (12%) attained the goals but did so in more than 24 hours; one of these six patients (17%) died. Of the nine (18%) protocol patients who never reached the resuscitation goals, three (33%) died.
Evaluation of Control Group Resuscitation
All (100%) of the 65 control patients attained their designated goals within 24 hours of admission. Fifty-two (80%) of the 65 control patients were also hemodynamically monitored during their resuscitation. Fifteen of these 52 (29%) control patients reached survivor values in 24 hours or less; five (33%) of these patients died. Four of these five patients, however, reached the survivor the survivor values only transiently (one set of measurements) and maintained lower values of CI, Dosub 2 I, and Vosub 2 I at all other times. Nine of 52 (17%) control patients reached supranormal values in more than 24 hours, and five (56%) of them died; again, three of these five patients had only one set of measurements demonstrating survivor values. Twenty-seven of 52 (52%) control patients never reached supranormal values; 12 of 27 (44%) of these patients died. Fourteen of 65 (22%) control patients were treated without a pulmonary artery catheter; one of these 14 (7%) patients died. The fourteen unmonitored control patients had significantly less blood loss than both the monitored control patients (2871 +/- 142 mL vs. 6600 +/- 639 mL, p < 0.01) and the protocol patients (2871 +/- 142 mL vs. 6818 +/- 844 mL, p < 0.05).
Fluid Therapy in Protocol and Control Groups
The protocol study patients received significantly more colloid solutions (fresh frozen plasma, hydroxyethyl starch, and/or 5% albumin) from 12 to 72 hours after admission (Table 5). They were also given significantly more blood (packed red blood cells and/or whole blood) and had a higher total fluid intake during the second day after admission. The protocol patients had a significantly higher cumulative fluid balance than did the control patients upon leaving the OR (21975 +/- 4746 mL vs. 10876 +/- 1572 mL, p = 0.036); they also received significantly more blood (664 +/- 103 mL vs. 362 +/- 75, p = 0.017) and colloids (692 +/- 86 mL vs. 287 +/- 40 mL, p < 0.001) and had a higher net fluid intake (1768 +/- 263 mL vs. 925 +/- 163 mL, p = 0.005) while in the SICU.
Mortality with Respect to Estimated Blood Loss
Thirty-four of 115 (30%) study group patients (11 protocol, 23 control) had an EBL of <3000 mL. Only three of 34 (9%) of these patients (all control patients) died. Thirteen of 115 (11%) patients (eight protocol and five control) had an EBL > 10,000 mL; four of five (80%) control patients and five of eight (63%) protocol patients died. However, of the 33 patients (21 protocol, 12 control) with an EBL > 3000 mL and less than or equal to5000 mL, only one of 21 (5%) protocol patients in this group died, while five of 12 (42%) control patients died (p < 0.01). Also, 35 of 115 (30%) patients (10 protocol, 25 control) had an EBL > 5000 mL and less than or equal to10,000 mL; 3 of 10 (30%) protocol patients in this group died, while 12 of 25 (48%) control patients died (p < 0.05).
Patterns of Cl, Dosub 2 I, and Vosub 2 I in Study Group Patients
The mean Dosub 2 I of the 50 protocol patients was higher than that of the control patients over all measured intervals from 0 to 72 hours after admission, while the CI and Vosub 2 I of the protocol patients was significantly higher than those of the control patients from 12 to 72 hours after admission (Table 6) and Figure 1Figure 2Figure 3) The mean CI, Dosub 2 I, and Vosub 2 I, of the survivors was generally higher than those of the nonsurvivors during the first 48 hours after admission (Figure 4Figure 5Figure 6.
Nine protocol patients never reached the resuscitation goals; two of these nine exsanguinated from massive abdominal injuries and one died of respiratory failure shortly after pneumonectomy. The six survivors each lost <3000 mL blood intraoperatively and did not reach a Vosub 2 I of 166 mL/min/msup 2 despite a supranormal Dosub 2 I and CI.
The lower mortality of the protocol patients compared with the control group (18% vs. 37%) in our study supports the hypothesis that the increased CI, Dosub 2 I, and Vosub 2 I seen in survivors of severe trauma are primary physiologic compensations, and suggests that the use of empirically observed survivor values of CI, Dosub 2 I, and Vosub 2 I as endpoints of resuscitation decreased mortality compared with conventional therapy. Using these survivor values of CI, Dosub 2 I and Vosub 2 I may also result in less shock-related organ failures, less time spent in the intensive care unit, and less time on mechanical ventilation.
The time frame in which the survivor values of CI, Dosub 2 I, and Vosub 2 I are reached appears to be as important as the values themselves. In a previous series  of monitored trauma patients we described significant differences in survivor and nonsurvivor CI, Dosub 2 I, and Vosub 2 I occurring within 24 hours after admission. However, in another series of trauma patients Abramson et al.  found few differences in survivor and nonsurvivor patterns of CI, Dosub 2 I, and Vosub 2 I starting at 48 hours after admission. Therefore, resuscitation of trauma patients to survivor values would probably be most beneficial during the first 24 hours after admission. This hypothesis was supported by the 20% mortality in the 50 patients reaching survivor values within 24 hours and 40% mortality in the 15 patients reaching them after the first day. The survivor pattern of CI, Dosub 2 I, and Vosub 2 I in trauma patients is different from that found in other high-risk patient groups both in absolute value and in temporal relationships. [30-33] Both the values themselves and their time relationships are important to patient care.
The greatest difference in outcome between the protocol and control patients (13% vs. 46%) was seen in the group losing between 3000 and 10,000 mL of blood. Most of these patients had successful repair of their injuries but sustained significant volume loss and tissue hypoperfusion which were correctable by aggressive intravascular resuscitation. Patients losing less than 3000 mL of blood probably accumulated minimal oxygen debts and were able to compensate adequately with or without aggressive therapy. On the other hand, patients losing more than 10,000 mL of blood had extensive injuries and often developed coagulopathies. These patients either remained hypovolemic or failed to restore tissue perfusion as measured by Dosub 2 I and Vosub 2 I despite massive fluid replacement.
The protocol and control patients spent similar amounts of time in the ED, OR, and PAR, had similar mean times from hospital to SICU admission, and underwent comparable numbers of ED and OR procedures. This reflects similar management of the two study groups with respect to diagnosis and operative intervention; it further strengthens the hypothesis that the improved outcome of the protocol patients is due to the resuscitation itself rather than to other treatment differences.
The 115 study group patients were selected for their large amount of blood loss; patients with similar anatomic injury but less volume loss were excluded. Because blood loss is an additional risk factor not included in RTS or ISS scoring, the expected mortality of these patients should be higher than that predicted by TRISS and ASCOT. While the control group did have a higher than predicted mortality, the protocol patient survival was actually similar to that predicted by TRISS and ASCOT. This suggests that the aggressive volume resuscitation and restoration of tissue oxygenation in the protocol patients effectively counteracted the increased risk of mortality due to blood loss. Conventional trauma scoring evaluates the number of organs injured or bones fractured but not the degree of circulatory impairment. [11,12]
The mean values of CI, Dosub 2 I, and Vosub 2 I used as protocol resuscitation goals in this trial were empirically described from a series of trauma patients with different injuries and variable amounts of volume loss.  Since trauma patients accumulate tissue oxygen debts proportional to the degree and duration of hypovolemia and the extent of tissue hypoperfusion,  the actual values of CI, Dosub 2 I, and Vosub 2 I needed for each patient to survive will vary above and below the goals used in this study. The mean values of CI, Dosub 2 I, and Vosub 2 I used in this trial represent a reasonable ``first approximation'' of optimal goals.
1. Davis JH: History of trauma. In Moore EE, Mattox KL, Feliciano DV (eds): Trauma. East Norwalk, Conn., Appleton & Lange, 1991, pp 109-126
2. Peltier LF, Davis JH: A history of the American association for the surgery of trauma: The first 50 years. J Trauma 29:143, 1989
3. Collicott PE: Initial assessment of the trauma patient. In Moore EE, Mattox KL, Feliciano DV (eds): Trauma. East Norwalk, Conn., Appleton & Lange, 1991, pp 109-126
4. American College of Surgeons Committee on Trauma: Advanced Trauma Life Support Course. Chicago, American College of Surgeons, 1988
5. Section III: Management of Specific Injuries. In Moore EE, Mattox KL, Feliciano DV (eds): Trauma. East Norwalk, Conn., Appleton & Lange, 1991, pp 245-663
6. Baxter BT, Moore EE, Moore JB, et al: Emergency Department thoracotomy following injury: Critical determinants for patient salvage. World J Surg 12:671, 1988
7. Burch JM, Feliciano DV, Mattox KL: The atriocaval shunt. Ann Surg 207:555, 1988
8. Kram HB, Nathan RC, Klein SR, et al: Clinical use of nonautologous fibrin glue. Am Surg 54:570, 1988
9. Hess PJ, Howe HR, Robicsek F, et al: Traumatic tears of the thoracic aorta: Improved results using the biomedicus pump. Ann Thorac Surg 48:6, 1989
10. Champion HR, Sacco WJ, Copes WS, et al: A revision of the Trauma Score. J Trauma 29:623, 1989
11. Copes WS, Champion HR, Sacco WJ, et al: The Injury Severity Score revisited. J Trauma 28:69, 1988
12. Champion HR, Copes WS, Sacco WJ, et al: A new characterization of injury severity. J Trauma 30:539, 1990
13. Fantini GA, Shires GT III, Shires GT: Management of Shock. In Moore EE, Mattox KL, Feliciano DV (eds): Trauma. East Norwalk, Conn., Appleton & Lange, 1991, p 148
14. Shoemaker WC: Shock states: Pathophysiology, monitoring outcome prediction, and therapy. In Shoemaker WC, Ayres SA, Grenvik A (eds): Textbook of Critical Care. Philadelphia, W.B. Saunders Co., 1989, pp 977-992
15. Wo CJ, Shoemaker WC, Appel PL, et al: Unreliability of blood pressure and heart rate for evaluation of circulatory stability in emergency resuscitation and critical illness. Crit Care Med 21:95, 1992
16. Bishop MH, Wo CJ, Appel PL, et al: Relationship between supranormal circulatory values, time delays, and outcome in severely traumatized patients. Crit Care Med 21:56, 1993
17. Schultz R, Whitfield GI, LaMura JJ, et al: The role of physiologic monitoring in patients with fractures of the hip. J Trauma 25:309, 1985
18. Scalea TM, Simon HM, Duncan AO, et al: Geriatric blunt multiple trauma: Improved survival with early invasive monitoring. J Trauma 30:129, 1990
19. Moore FA, Haenel JB, Moore EE, et al: Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury organ failure. J Trauma 33:58, 1992
20. Rady MY, Edwards JD, Nightingale P: Early cardiorespiratory findings after severe blunt thoracic trauma and their relation to outcome. Br J Surg 79:65, 1992
21. Fleming AW, Bishop MH, Shoemaker WC, et al: Prospective trial of supranormal values as goals of resuscitation in severe trauma. Arch Surg 127:1175, 1992
22. Hyers TH: Prediction of survival and mortality in patients with adult respiratory distress syndrome. New Horizons 1:466, 1993
23. Rinaldo JE: Adult respiratory distress syndrome. In Shoemaker WC, Ayres SA, Grenvik A (eds): Textbook of Critical Care. Philadelphia, W.B. Saunders Co., 1989, pp 500-505
24. Fry DE: Multiple system organ failure. In Fry DE (ed): Multiple System Organ Failure. St. Louis, Mosby Year Book, 1992, pp 3-15
25. Johnson PC: Disseminated intravascular coagulation. In Fry DE (ed): Multiple System Organ Failure. St. Louis, Mosby Year Book, 1992, pp 291-299
26. Meuller, HS: Management of acute myocardial infarction. In Shoemaker WC, Ayres SA, Grenvik A (eds): Textbook of Critical Care. Philadelphia, W. B. Saunders Co., 1989, pp 341-353
27. Knaus WA, Wagner DP: Multiple systems organ failure: Epidemiology and prognosis. Crit Care Clin 5:221, 1989
28. Bone RC, Balk RA, Cerra FB, et al: Definitions for sepsis and organ failure and guidelines for the use for innovative therapies in sepsis. The ACCP/SCCM Concensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 101:1644, 1992
29. Abramson D, Scalea TM, Hitchcock R, et al: Lactate clearance and survival following injury. J Trauma 35:584, 1993
30. Shoemaker WC, Appel PL, Kram HB, et al: Prospective trial of supranormal values of survivors as therapeutic goals in high risk surgical patients. Chest 94:1176, 1988
31. Tuschmidt J, Fried J, Astiz M, et al: Elevation of cardiac output and oxygen delivery improves outcome in septic shock. Chest 1102:216, 1992
32. Berlauk JF, Abrams JH, Gilmour IF, et al: Preoperative optimization of cardiovascular hemodynamics improves outcome in peripheral vascular surgery. Ann Surg 214:189, 1991
33. Creamer JE, Edwards JD, Nightingale P: Hemodynamic and oxygen transport variables in cardiogenic shock secondary to acute myocardial infarction and response to treatment. Am J Cardiol 65:1297, 1990
© Williams & Wilkins 1995. All Rights Reserved.
34. Shoemaker WC, Appel PL, Kram HB: Role of oxygen debt in the development of organ failure, sepsis, and death in high-risk surgical patients. Chest 102:208, 1992