Massive Transfusion and Control of Hemorrhage in the Trauma Patient

N A L T R A I O U T M A A N R C E A T R

N E

Based on Special ITACCS Seminar Panels. The International Trauma Anesthesia and Critical Care Society (ITACCS) is accredited by the Accreditation Council for Continuing Medical Education (ACCME) for physicians. This CME activity was planned and produced in accordance with the ACCME Essentials. ITACCS designates this CME activity for 15 credit hours in Category 1 of the Physicians Recognition Award of the American Medical Association.

CME QUESTIONS INCLUDED JANUARY 2003 LEARNING OBJECTIVES OF THE MONOGRAPH Chapter 6 Atraumatic blood salvage and autotransfusion in trauma and surgery ...... Page 17 Sherwin V. Kevy, MD, and Robert Brustowicz, MD, Trans- After completion of this activity, the participant will be able to: fusion Service, Children’s Hospital Department of Anes- thesia, Harvard Medical School, Boston, Massachusetts 1. Evaluate the etiology and pathophysiology of traumatic shock. 2. Describe the management of massive transfusion in the trauma patient. Section III: Transfusion: Clinical Practice 3. Discuss the clinical indications and problems related to the use of blood, blood components, hemostatic agents, oxygen-carrying vol- Chapter 7 Current practices in blood and blood ume expanders, and venous thromboembolism prophylaxis. component therapy ...... Page 18 Charles E. Smith, MD, FRCPC, Department of Anesthesi- EDITORS ology, MetroHealth Medical Center, Case Western Reserve University School of Medicine, Cleveland, Ohio Charles E. Smith, MD, FRCPC, Professor of Anesthesiology, Chapter 8 Immunomodulatory effects of transfusion .. Page 22 MetroHealth Medical Center, Case Western Reserve University School David T. Porembka, Do, FCCM, FCCP, Associate Professor of Medicine, Cleveland, Ohio; Chair, ITACCS Special Equipment/Tech- of Anesthesia and Surgery, Associate Director of Surgical niques Committee Intensive Care, University of Cincinnati Medical Center, Cincinnati, Ohio Andrew D. Rosenberg, MD, Chairman, Department of Anesthesi- ology, Hospital for Joint Diseases Orthopaedic Institute, Associate Pro- Chapter 9 Blood transfusions ...... Page 27 fessor of Clinical Anesthesiology, New York University School of Medi- Andrew D. Rosenberg, MD, Department of Anesthesiol- cine, New York, New York ogy, Hospital for Joint Diseases/Orthopaedic Institute, New York, New York Christopher M. Grande, MD, MPH, Lecturer, Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women’s Chapter 10 Vascular access in trauma: options, risks, Hospital, Harvard Medical School, Boston, Massachusetts; Professor, benefits, complications ...... Page 28 Department of Anesthesiology, State University of New York, Buffalo, Maureen Nash Sweeney, MD, Attending Anesthesiologist, Buffalo, New York; Professor of Anesthesiology, West Virginia University Department of Anesthesiology, Department of Veterans School of Medicine, Morgantown, West Virginia; Executive Director, Affairs Medical Center, New York, New York International Trauma Anesthesia and Critical Care Society (ITACCS), World Headquarters Baltimore, Maryland Chapter 11 Principles of fluid warming ...... Page 30 Charles E. Smith, MD, Department of Anesthesiology, MetroHealth Medical Center, Case Western Reserve Uni- CONTENTS AND CONTRIBUTORS versity, Cleveland, Ohio Chapter 12 Management of massive hemorrhage and Section I: Etiology and Pathophysiology transfusion in trauma ...... Page 34 Chapter 1 Trauma, a disease of bleeding...... Page 3 Georges Desjardins, MD, FRCPC, Division of Trauma Anes- Thomas M. Scalea, MD, Physician-in-Chief, Professor of thesia and Critical Care, Ryder Trauma Center, University Surgery, R Adams Cowley Shock Trauma Center, Baltimore, of Miami/Jackson Memorial Medical Center, Miami, Florida Maryland Chapter 13 Rapid infusion and point-of-care chemistry testing monitoring in massive transfusion: Chapter 2 Pathophysiology of traumatic shock ...... Page 5 avoiding common pitfalls ...... Page 38 Richard P. Dutton, MD, Associate Director, Division of Jeffrey R. Jernigan, MD, and John G. D’Alessio, MD, De- Anesthesiology, R Adams Cowley Shock Trauma Center, partment of Anesthesiology, Elvis Presley Memorial Trauma Baltimore, Maryland Center, Memphis, Tennessee Section II: Therapeutic Strategies Section IV: New Horizons in Synthetic Blood Substitutes Chapter 3 Surgical perspectives to control Chapter 14 Hemoglobin-based oxygen-carrying bleeding in trauma ...... Page 7 solutions and hemorrhagic shock ...... Page 40 Brian R. Plaisier, MD, Department of Surgery, Bronson Colin F. Mackenzie, MB, ChB, FRCA, FCCM, Director, Methodist Hospital, Kalamazoo, Michigan National Study Center for Trauma and Emergency Medi- cal Systems, University of Maryland School of Medicine, Chapter 4 Hemostatic drugs in trauma and Baltimore, Maryland orthopaedic practice ...... Page 11 David Royston, MB, FRCA, Consultant Anaesthetist, Royal Chapter 15 Hemoglobin therapeutics, blood substitutes, Brompton and Harefield NHS Trust, Harefield, Middlesex, and high-volume blood loss ...... Page 44 United Kingdom Armin Schubert, MD, MBA, Chairman, Department of Gen- eral Anesthesia, Cleveland Clinic Foundation, Cleveland, Ohio Chapter 5 Antithrombotics in Trauma Care: Benefits and Pitfalls ...... Page 14 CME Questions ...... Page 48 John K. Stene, MD, PhD, Past President, ITACCS, Associate Professor of Anesthesia and Director of Trauma Anesthesia, The drug and dosage information presented in this publication is Milton S. Hershey Medical Center, Hershey, Pennsylvania believed to be accurate. However, the reader is urged to consult the full prescribing information on any product mentioned in this publication for recommended dosage, indications, contraindications, warnings, precautions, and adverse effects. This is particularly important for drugs that are new or prescribed infrequently.

2 Massive Transfusion and Control of Hemorrhage in the Trauma Patient Massive Transfusion and Control of Hemorrhage in the Trauma Patient

fluids, endpoints of fluid and blood resuscita- potensive versus normotensive resuscitation, Introduction tion, complications of transfusion therapy, and the benefits of point-of-care testing, and the Priorities in trauma patient management clinical strategies to reduce complications. use of guidelines (in conjunction with the are to ensure adequate ventilation and oxygen- The section on “Transfusion: Clinical Prac- blood bank) for managing trauma patients who ation, control hemorrhage, and restore tissue tice” begins with a discussion on the immuno- require “rapid infusion.” perfusion to vital organs. The most familiar logic consequences of transfusions and con- In the final section on “New Horizons in means to control hemorrhage are surgical liga- cludes that allogeneic transfusions have a dy- Synthetic Blood Substitutes,” Dr. Mackenzie tures and clips. Other means include namic immunomodulatory effect on the recipi- reviews the complex issues surrounding the transcatheter embolization, appropriate blood ent and that leukocytes are the chief mediator use of hemoglobin solutions and hemorrhagic component therapy, maintenance of normo- of these effects. Dr. Rosenberg reviews the sci- shock. He states that, although many of the thermia, and pharmacologic agents. Finally, entific literature and his own personal experi- problems associated with oxygen-carrying so- attention must also be directed toward treat- ence with the concept of “decreasing the lutions have been overcome, there is a paucity ment of the hypercoaguable state that follows amount of blood transfused to trauma pa- of published data concerning the use of oxy- major traumatic injury and can lead to deep tients” in light of transfusion-related immuno- gen-carrying solutions in humans with hem- venous thrombosis and pulmonary embolism. suppression and other risks. Dr. Sweeney orrhagic shock. Dr. Schubert concludes the The management of massive transfusion evaluates the options, risks, and potential com- monograph by examining the potential clini- and control of hemorrhage in the trauma pa- plications of obtaining vascular access in cal uses and effectiveness of hemoglobin-based tient were discussed during two special trauma, illustrating the different approaches oxygen carriers and perfluorocarbons. The ITACCS seminars. The 15 reports in this in pediatric and adult trauma patients. The long shelf life, long circulation half-life, and monograph summarize the state-of-the art principles of warming IV fluid and blood are good oxygen-carrying capacity and tissue oxy- knowledge and clinical practice issues regard- reviewed by Dr. Smith, with special emphasis gen delivery make these compounds particu- ing surgical and nonsurgical management of on the thermal stress of infusing cold or inad- larly attractive in patients with high blood loss, massive transfusion and control of hemor- equately warmed fluids, and the safety and ef- i.e., trauma patients. In his manuscript, Dr. rhage in the injured patient. ficacy of fluid warmers and rapid infusion de- Schubert evaluates the different hemoglobin In the section on “Etiology and Patho- vices. Dr. Desjardins focuses on the manage- solutions and the pitfalls associated with their physiology,” Dr. Scalea reviews the physiologic ment of exsanguinating hemorrhage (other- clinical use. importance of recognizing and restoring he- wise known as “massive, massive transfusion”) As editors and principal organizers of this mostasis following injury and discusses the and reports on the washing and centrifuging special ITACCS symposium, we have attempted American College of Surgeons classification of packed red blood cells prior to rapid infu- to provide a concise, up-to-date reference on scheme for hemorrhage, as well as operative sion in order to decrease adverse metabolic massive transfusion and management of hem- and nonoperative (e.g., embolization) tech- consequences such as hyperkalemia. Drs. orrhage in the trauma patient—a reference that niques for treatment of ongoing blood loss. Jernigan and D’Alessio discuss their experience integrates both basic science and clinical prac- Dr. Dutton discusses the four phases of trau- using rapid infusion devices to deliver massive tice. We sincerely hope that you, the reader, matic shock and reviews the macro- and mi- quantities of fluids, blood, and blood products will obtain essential knowledge from this cro-circulatory responses to traumatic to maintain circulating blood volume. These monograph that will improve your clinical shock—responses that ultimately determine authors point out the controversies over hy- practice when caring for trauma patients. patient outcome. The “Therapeutic Strategies” section begins with a report on surgical perspectives to control bleeding in trauma. In that article, Dr. SECTION I: Etiology and Pathophysiology Plaisier describes the benefits and risks of topi- 1 cal hemostatic agents such as oxidized cellu- Trauma, A Disease of Bleeding lose, collagen sponges, thrombin, denatured gelfoam, and fibrin glue. Dr. Royston reviews Thomas M. Scalea, MD tablished organ failure has not changed since the hemostatic and anti-inflammatory effects Physician-in-Chief it was first described almost 25 years ago.2 of a variety of drugs in trauma. There appears R Adams Cowley Shock Trauma Center Thus, it is imperative that hemorrhage is rec- to be a significant benefit of high-dose University of Maryland School of Medicine ognized and treated early. aprotinin therapy to reduce blood loss and the 22 South Greene Street The recognition of acute hemorrhage can need for blood and blood product transfusion. Baltimore, MD 21201 USA be difficult. The American College of Surgeons Major post-traumatic morbidity and mortality has developed the classification scheme for may result from venous thromboembolism, Acute blood loss is a very common prob- hemorrhage, stratifying blood loss from Stage and Dr. Stene discusses therapeutic strategies lem following injury. Rapid recognition and 1 (less than 15% of total circulating blood vol- to prevent and treat deep venous thrombosis restoration of homeostasis is the cornerstone ume) to Stage 4 (more than 40% of total circu- and pulmonary embolism in the injured pa- of the initial care of any badly injured patient. lating blood volume).3 Changes in various tient. In the article on atraumatic blood sal- Untreated, hemorrhage robs the cardiovascu- physiologic parameters as hemorrhage vage and autotransfusion, Drs. Kevy and lar system of the preload necessary to ensure volume increases are listed in Table 1. Unfor- Brustowicz critique the use of surgical suction adequate cardiac output and peripheral oxy- tunately, many of these signs and symptoms systems as a means of reducing (or supple- gen delivery. Inadequate perfusion, even if it are nonspecific. In addition, a number of other menting) allogeneic blood use. Dr. Smith ana- is not associated with overt hypotension, can parameters will affect the patient’s vital signs lyzes the use of fluid and blood component set off the neurohumoral cascade, ultimately and physical findings. For instance, the rapid- therapy in trauma and addresses various issues leading to sequential organ failure.1 This is ity of volume loss may be as important as the such as delayed fluid resuscitation, hypertonic especially important, as the mortality from es- total volume of hemorrhage.2 Underlying car-

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 3 hypotension produces a more injurious shock Table 1. American College of Surgeons Classification of Acute Hemorrhage insult than do multiple episodes of shock and resuscitation.18 Thus, the clinician must esti- Class I II II IV mate the degree of hemorrhage, the depth of Blood loss (ml) <750 750-1,500 1,500-2,000 ≥ 2,000 shock, and the time to definitive hemostasis when making a decision. % Blood volume lost <15% 15-30% 30-40% ≥ 40% Regardless of the resuscitation decision, Pulse rate <100 >100 >120 ≥ 140 patients who demonstrate ongoing bleeding Blood pressure Normal Normal Decreased Decreased require definitive hemostasis. Serial blood gas determinations and/or central venous oxygen Pulse pressure Normal or Decreased Decreased Decreased saturation determination may be very helpful (mmHg) increased in determining whether blood loss is continu- Capillary refill Normal Delayed Delayed Delayed ing.8,9 Unfortunately, the relationship between blood loss and physiologic parameters may be Respiratory rate 14-20 20-30 30-40 >35 different after resuscitation than they were Urine output >30 20-30 5-15 Negligible during hemorrhage. For instance, approxi- Mental status Slightly anxious Mildly anxious Anxious, Confused, mately 12 to 16 hours following resuscitation, confused lethargic the relationship changes between base deficit and anion gap versus serum lactate, and an- Recommended fluid 0.9% saline, 3:1 0.9% saline, 3:1 0.9% saline 0.9% saline ion gap and base deficit no longer correlate replacement + red cells +red cells with lactate.19 During this time, one must directly measure serum lactate, as it cannot be Amounts are based on the patient’s initial presentation. Assumes 70-kg male with a blood inferred from either of the other two measure- volume of ~70 ml/kg. ments. When resuscitation decisions are based Adapted from the American College of Surgeons Committee on Trauma: Advanced Trauma on these parameters, therapy will be inappro- Life Support Program for Physicians, Student and Instructor Manual, Chicago, American Col- priate almost 50% of the time. lege of Surgeons, 1993. Elderly patients with poor underlying cardiovascular reserve often require invasive diovascular reserve also plays a role. Young surements. That is, flow from all vascular beds monitoring to precisely measure the physi- people with very compliant blood vessels may contributes to this determination. However, ologic deficits and to guide therapy. In fact, in compensate extremely well for large-volume some vascular beds are more sensitive than high-risk elderly patients (Table 2), monitor- blood loss, even as much as 40% to 50% of others to the effects of hemorrhage. Thus, ing must be instituted extremely early, within total circulating blood volume.4 They then de- shock may be detected earlier if we are able to 2 to 3 hours of injury if possible. There is a velop sudden cardiovascular compromise recognize a local decrease in perfusion. Shock statistically significant decrease in survival when compensatory mechanisms fail. Elderly is defined as inadequacy of peripheral oxygen when monitoring is delayed as long as 6 hours.5 people, on the other hand, will develop car- delivery. Clinically, we use indirect measure- Even young people may have inadequate cardiovascular insufficiency and hypotension with ments to gauge hemorrhage. Target organ func- diovascular response to substantial injuries. A much smaller blood loss.5 Prescription medi- tion such as urine output or mental status are surprising percentage of young patients with cation and/or illicit drugs will also influence examples of this. Unfortunately, urine output either blunt or penetrating trauma benefit from the cardiovascular response to injury.6,7 The is extremely variable and nonspecific. Although invasive monitoring and require volume and amount of resuscitation, if any, the patient re- oliguria almost certainly indicates hypov- pharmacologic therapy to support cardiovas- ceives in the field will affect cardiovascular re- olemia, normal urine output or polyuria is in- cular performance and clear lactate.20,21 sponse as well.4 conclusive. Renal tubular function is affected Clearly, achieving hemostasis is the most Data from the past 10 years strongly sug- by as little as a 20% acute loss of blood vol- important part of resuscitating the trauma vic- gest that normally followed vital signs are a ume. The kidney develops a salt-wasting neph- tim. Resuscitation efforts will not be success- very poor indication of the depth of hemor- ropathy, and the patient makes more urine ful until blood loss is arrested. Substantial hem- rhage.8 In particular, blood pressure and pulse than is appropriate for this degree of physi- orrhage usually requires operative therapy. rate, the two vital signs often used in the emer- ologic insult.15 Blood flow to the gastrointesti- Recently, however, other techniques have gency department to gauge hemorrhage, are nal tract, however, is a relatively sensitive indi- emerged and should be considered, even in tremendously nonspecific. Central venous oxy- cator of the loss of circulating blood volume. patients with hypotension. The diagnosis of gen saturation and mixed venous oxygen satu- Intracellular pH, as measured in the stomach, ongoing blood loss with angiography and he- ration are far more sensitive and reliable mea- small bowel, or colon, is a very sensitive mea- mostasis with transcatheter embolization is a surements of acute volume loss.8,9 Degree of sure of hemorrhage.16 Current technology does real alternative to standard operative therapy.22 metabolic acidosis, as measured by the base not allow us to measure intracellular pH in real This has been a mainstay of therapy for many deficit from an arterial blood gas, is also ex- time. However, that technology may be forth- years in patients bleeding from a blunt pelvic tremely helpful in gauging the degree of coming in the not-too-distant future. injury. Retroperitoneal exploration in these shock.10 Base deficit has been shown to corre- Once the clinician has made the diagno- late with transfusion requirements, ICU stay, sis of acute blood loss, several issues become and ultimate outcome.11,12 During initial resus- important. Traditional dogma suggests that Table 2. High-Risk Geriatric Patients citation, base deficit should also correlate with restoration of forward flow by crystalloid re- serum lactate level. The ability to clear lactate suscitation followed by blood is optimal Initial systolic blood pressure <130 mmHg to normal is one of the most important pre- therapy. However, increases in blood pressure Closed head injury dictors of survival following hemorrhage and produced by fluid may, in fact, increase blood injury.13,14 loss by displacing the hemostatic clot that was Multiple long-bone fractures Measures such as mixed venous oxygen formed at the time of hypotension.17 This is- Metabolic acidosis content, venous oxygen saturation, blood pres- sue will be discussed in Chapter 3. However, Pedestrian–motor vehicle mechanism sure, and lactate concentration are global mea- there are now data to suggest that sustained

4 Massive Transfusion and Control of Hemorrhage in the Trauma Patient patients is fraught with danger, and emboliza- References mission base deficit predicts transfusion tion is far preferable in almost every case. These 1. Dantzker D. Oxygen delivery and utilization requirements and risk of complications. J techniques have been extended to other areas in sepsis. Crit Care Clin 1989; 5:81–98. Trauma 1996; 41:769–74. of the body. More recently, transcatheter em- 2. Scalea TM, Henry SM. Inotropes in the 13. Iberti TJ, Leibowitz AB, Papdakos PJ, et al. bolization has been used for nonoperative intensive care unit. In Advances in Low cardiac sensitivity of the anion gap as a management of solid visceral injuries within Trauma and Critical Care, vol. 7. St. screen to detect hyperlactatemia in critically the abdomen. Treatment algorithms using Louis, Mosby, 1992. ill patients. Crit Care Med 1990; 18:275–7. splenic artery embolization in patients man- 3. Committee on Trauma, American College 14. Abramson D, Scalea TM, Hitchcock D, et aged nonoperatively have resulted in a greater of Surgeons. The Advanced Trauma Life al. Lactate clearance and survival follow- than 90% rate of splenic salvage.23 This is far Support Program, Instructors Manual. ing injury. J Trauma 1993; 35:584–9. higher than any series utilizing observation Chicago, American College of Surgeons, 15. Sinert R, Baron B, Low R, et al. Is urine and/or operation alone. In addition, 1988, pp 59–62. output a reliable index of blood volume embolotherapy may be extremely helpful in 4. Lewis FR. Prehospital intravenous fluid in hemorrhagic shock? Acad Emerg Med patients with vascular injuries in relatively in- therapy: a physiologic computerized 1996; 3:448. accessible areas. Exposure of the carotid ar- model. J Trauma 1986; 26:804–11. 16. Baron BJ, Scalea TM. Acute blood loss. tery in Zone 3 of the neck is extremely diffi- 5. Scalea TM, Simon HM, Duncan AL, et al. Emerg Med Clin North Am 1996; 14:35–54. cult. Embolotherapy has a real role in manag- Geriatric blunt trauma: improved survival 17. Shaftan GW, Chui C, Dennis C, et al. Fun- ing these injuries. Temporary hemostasis can with early invasive monitoring. J Trauma damentals of physiologic control of arte- be achieved with percutaneous balloons used 1990; 30:129–36. rial hemorrhage. Surgery 1965; 58:851. at the time of diagnostic angiography. This tem- 6. Horton JW. Ethanol impairs 18. Sinha HA, Baron BJ, Buckley MC, et al. Fluid porary control of bleeding allows further im- cardiocirculatory function in treated canine restriction versus early resuscitation in hem- aging, ongoing resuscitative efforts, and time hemorrhagic shock. Surgery 1986; 100:520. orrhagic shock. J Trauma 1994; 37:1015. to plan definitive therapy. In addition to its 7. Sloan EP, Zalenski RJ, Smith RF, et al. Toxi- 19. Mikulaschek A, Henry SM, Donovan R, usefulness in Zone 3 of the neck, angiographic cology screening in urban trauma pa- Scalea TM. Serum lactate is not predicted hemostasis has great utility in injuries to the tients: drug prevalence and its relation- by anion gap or base excess after trauma thoracic outlet and deep within the pelvis. ship to trauma severity and management. resuscitation. J Trauma 1996; 40:218–24. Embolization techniques can be com- J Trauma 1989; 29:1647. 20. Abou-Khalil B, Scalea TM, Trooskin SZ. He- bined with surgery, allowing the patient to 8. Scalea TM, Holman M, Fuortes M, et al. modynamic responses to shock in young benefit from both techniques. Ideally, this Central venous blood oxygen saturation: trauma patients: the need for invasive moni- should be done in the operating room and, in an early accurate measurement of volume toring. Crit Care Med 1994; 22:633–9. some centers, biplanar angiography is avail- during hemorrhage. J Trauma 1988; 21. Scalea TM, Maltz S, Yelon J, et al. Resusci- able. Patients who may benefit from this tech- 28:725–32. tation of multiple trauma and head inju- nology are those with a combination of intra- 9. Scalea TM, Hartnett RW, Duncan AO, et al. ries: role of crystalloid fluid and inotropes. abdominal blood loss and pelvic blood loss. Central venous oxygen saturation: a use- Crit Care Med 1994; 22:1610–5. The pelvic blood loss can be embolized while ful clinical tool in trauma patients. J 22. Panetta T, Sclafani SJA, Goldstein AJ, et al. intra-abdominal blood loss is treated directly Trauma 1990; 30:1529–44. Percutaneous transcatheter embolization via surgery. Sometimes patients are too pro- 10. Rutherford EJ, Morris JA, Reed GW, et al. for massive bleeding from pelvic fractures. foundly ill to allow definitive surgery. Damage Base deficit stratifies mortality and deter- J Trauma 1985; 25:1021. control techniques should then be employed. mines therapy. J Trauma 1992; 33:417. 23. Sclafani SJA, Scalea TM, Herskowitz M, et In these settings, major vascular injuries are 11. Davis JW, Shackford SR, MacKersie RC, Hoyt al. Salvage of CT-diagnosed splenic inju- repaired and gastrointestinal contamination DB. Base deficit as a guide to volume re- ries: utilization of angiography for triage controlled. The patient is then packed with suscitation. J Trauma 1988; 28:1464–7. and embolization for hemostasis. J laparotomy pads and taken to the intensive 12. Davis JW, Parks SN, Kaups KL, et al. Ad- Trauma 1995; 39:818–27. care unit for ongoing resuscitation and warming techniques. Once patients are resuscitated, they can return to the operating room for un- packing, gastrointestinal reconstruction, and 2 any other procedures necessary. Angiographic Pathophysiology of Traumatic Shock embolotherapy has a role in these patients as well and can be utilized postoperatively to Richard P. Dutton, MD hemorrhagic: the patient bleeds, and perfusion supplement surgical hemostasis. Injuries deep Director, Trauma Anesthesia decreases. This may be followed by an anemic within the substance of the liver, in the R Adams Cowley Shock Trauma Center phase as the patient is resuscitated with crys- retroperitoneum, or in the pelvis may be more University of Maryland School of Medicine talloid solutions and simultaneously mobilizes easily controlled via embolization than surgery. Baltimore MD 21201 USA interstitial fluid into the vasculature. A cardio- Early recognition of hemorrhage is key to e-mail: [email protected] genic or neurogenic component may be the optimal care of trauma patients. Ongoing present initially due to specific injuries to the controversies exist as to the ideal resuscitation Shock—a condition of decreased total heart or central nervous system (CNS) or may scheme. In fact, there is probably no one ideal body oxygen delivery—can be brought on by be the secondary result of hypoperfusion and strategy. Care must be tailored to the patient’s a number of mechanisms. These include fail- the release of toxic factors. It is important to mechanism of injury and physiology. ure of the heart to pump blood through the recognize that the traumatic shock seen clini- Nonoperative homeostasis can supplement sur- body (cardiogenic), loss of circulating fluid cally in severely injured patients may be quite gical techniques and its use should be consid- volume (hemorrhagic), decreased oxygen car- different from the induced shock seen in labo- ered. Normally followed vital signs are very poor rying capacity (anemic), or loss of vascular tone ratory animals hemorrhaged under controlled indicators of the degree of hemorrhage and the (neurogenic).1 “Traumatic shock”—shock conditions. adequacy of resuscitation. Invasive monitoring brought on by an injury in an otherwise healthy is often necessary to precisely determine the patient—is best thought of as a combination Stages of Shock physiologic deficit and guide therapy. of these factors. The initial phase is usually Traumatic shock may be thought of as

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 5 occurring in four phases (Fig. 1). In compensated traumatic shock, an increase in heart Figure 1 rate and vasoconstriction of nonessential and ischemia-tolerant vascular beds will allow prolonged survival and easy recovery once coagulation occurs and adequate fluids and nutri- tion are provided. Decompensated traumatic shock, also known as progressive shock, is a transitory state in which the lack of perfusion to certain tissues is building up a debt of local cell damage that will produce a toxic effect on the organism when perfusion is reestablished. Shock is still reversible at this stage. In subacute irreversible shock, the patient can be resuscitated hemodynamically but succumbs at a later time to multiple organ system failure as a result of the toxic effects of ischemia and reperfusion. Finally, acute irreversible shock is the condition of ongoing hemorrhage, aci- Traumatic shock and its potential outcomes. A. In early shock there is only a small drop in dosis, and coagulopathy that spirals steadily oxygen delivery due to compensation by the cardiovascular system. B. Decompensated downward to the patient’s demise.1,2 shock is characterized by an accelerating defect in oxygen delivery. C. Recovery from decompensated shock includes a hyperdynamic period as the body’s oxygen debt is repaid. D. The patient whose hemorrhage has pro- In subacute irreversible shock, the macrocirculation is restored and bleeding stopped, but ceeded to the point of decompensated shock hypoperfusion has been severe enough that oxygen debt cannot be repaid. Lethal multiple represents a surgical and metabolic emergency. organ system failure develops. E. Acute irreversible shock occurs when hemodynamic con- If the loss of blood (and thus the loss of oxy- trol is never regained. The patient exsanguinates and dies in cardiovascular collapse. gen-delivering capacity) can be reversed before the inflammatory cascade begins, the patient will survive. Adequate volume resuscitation equate control of surgical bleeding and volu- Organ System Responses to Traumatic leads the patient into higher-than-normal oxy- minous blood product replacement.1 Shock gen consumption—a hypermetabolic state— Specific organ systems respond to trau- for hours to days after the acute injury, as the The Body’s Response to Shock matic shock in specific ways. The CNS is the body repays the metabolic debt built up dur- The stages of traumatic shock are directly prime trigger of the neuroendocrine response ing the period of ischemia.3 Figure 1 shows related to the body’s response to hemorrhage. to shock, which maintains perfusion to the this patient following curve C and eventually The initial response is on the macrocirculatory heart and brain at the expense of other tissues.8 achieving normal equilibrium. level and is mediated by the neuroendocrine Regional glucose uptake in the brain changes A few minutes too late, however, and sub- system. Decreased blood pressure leads to during shock.9 Reflex activity and cortical elec- acute irreversible shock will have occurred, as vasoconstriction and catecholamine release. trical activity are both depressed during hy- represented by curve D. Bleeding may be con- Heart and brain blood flow is preserved, while potension; these changes are reversible with trolled and vital signs may be normal or even other regional beds are constricted. Pain, hem- mild hypoperfusion, but become permanent hypernormal, but the damage has been done orrhage, and cortical perception of traumatic with prolonged ischemia. Failure to recover on the cellular level. Some tissues will continue injuries lead to the release of a number of hor- preinjury neurologic function is a marker for to be ischemic due to lack of reflow caused by mones, including renin–angiotensin, vaso- subacute irreversible shock, even if the cellular swelling and microcirculatory obstruc- pressin, antidiuretic hormone, growth hor- patient’s hemodynamic functions are normal.10 tion. When flow is successfully restored on the mone, glucagon, cortisol, epinephrine and The kidney and adrenal glands are prime cellular level, the process of reperfusion be- norepinephrine.5 This response sets the stage responders to the neuroendocrine changes of gins. This washout of toxins and inflammatory for the microcirculatory responses that will shock, producing renin, angiotensin, aldoster- factors is as dangerous to the patient as the ultimately determine the patient’s outcome. one, cortisol, erythropoietin, and catechola- hemorrhage itself. This is the patient who de- On the cellular level the body responds mines.11 The kidney itself maintains glomeru- velops adult respiratory distress syndrome then to hemorrhage by taking up interstitial fluid, lar filtration in the face of hypotension by se- progresses to acute renal failure, gut dysfunc- causing cells to swell.6 This may choke off ad- lective vasoconstriction and concentration of tion, immunosuppression, cardiac failure, and jacent capillaries, resulting in the “no-reflow” blood flow in the medulla and deep cortical eventual death due to multiple organ system phenomenon that prevents the reversal of is- area. Prolonged hypotension leads to de- failure. Even though the hemorrhage is chemia even in the presence of adequate creased cellular energy and an inability to con- stopped short of exsanguination, the body is macro flow.7 Ischemic cells produce lactate and centrate urine, followed by patchy cell death, unable to survive the ischemic insult.2,4 free radicals, which are not cleared by the cir- tubular epithelial necrosis, and renal failure.8,12 Curve E in Figure 1 represents acute irre- culation. These compounds cause direct dam- The heart is relatively preserved from is- versible traumatic shock. Prolonged hypoten- age to the cell, as well as comprising the bulk chemia during shock because of maintenance sion is followed by progressive vasodilatation, of the toxic load that will be washed back to or even increase of nutrient blood flow, and loss of response to fluids and catecholamines, the central circulation when perfusion is rees- cardiac function is generally well preserved capillary leak, diffuse coagulopathy, cardiac tablished. The ischemic cell will also produce until the late stages.8,11 Lactate, free radicals, dysfunction, and early death. These patients and release a variety of inflammatory factors: and other humoral factors released by ischemic are usually said to have exsanguinated, al- prostacyclin, thromboxane, prostaglandins, cells all act as negative inotropes, however, and though in the presence of modern rapid infu- leukotrienes, endothelin, complement, in the decompensated patient may produce sion techniques and aggressive transfusion, this interleukins, tumor necrosis factor, and oth- cardiac dysfunction as the terminal event in is not strictly true. Rather, the patient dies from ers.1 These are the ingredients of acute and the shock spiral.13 the acute metabolic consequences of failed subacute irreversible shock. The lung, which cannot itself become is- perfusion, frequently in the presence of ad- chemic, is nonetheless the downstream filter

6 Massive Transfusion and Control of Hemorrhage in the Trauma Patient for the inflammatory byproducts of the is- 2. Shoemaker WC, Peitzman AB, et al. Resus- monary response to major injury. Arch chemic body. The lung is often the sentinel citation from severe hemorrhage. Crit Surg 1974; 108:349–55. organ for the development of multiple organ Care Med 1996; 24:S12–23. 15. Thorne J, Blomquist S, Elmer O. Polymor- system failure.4,14 Immune complex and cellu- 3. Cerra FB. Metabolic response to injury. In phonuclear leukocyte sequestration in the lar factors accumulate in the capillaries of the Cerra FB, ed. Manual of Critical Care. St. lung and liver following soft tissue trauma: lung, leading to neutrophil and platelet aggre- Louis, CV Mosby, 1987, pp 117–45. an in vivo study. J Trauma 1989; 29:451– gation, increased capillary permeability, de- 4. Demling R, Lalonde C, Saldinger P, Knox 6. struction of lung architecture, and the acute J. Multiple organ dysfunction in the sur- 16. Martin BA, Dahlby R, Nicholls I, Hogg JC. respiratory distress syndrome.15,16 The pulmo- gical patient: pathophysiology, preven- Platelet sequestration in lungs with hem- nary response to traumatic shock is the lead- tion, and treatment. Curr Probl Surg 1993; orrhagic shock and reinfusion in dogs. J ing evidence that this disease is not just a dis- 30:345–424. Appl Physiol 1981; 50:1306–12. order of hemodynamics: pure hemorrhage, in 5. Peitzman AB. Hypovolemic shock. In Pinsky 17. Fulton RL, Raynor AVS, Jones C. Analysis the absence of hypoperfusion, does not pro- MR, Dhainaut JFA, eds. Pathophysiologic of factors leading to posttraumatic pulmo- duce pulmonary dysfunction.14,17 Foundations of Critical Care. Baltimore, nary insufficiency. Ann Thorac Surg 1978; The intestine is one of the earliest organs Williams & Wilkins, 1993, pp 161–9. 25:500–9. affected by hypoperfusion and may be one of 6. Shires GT, Cunningham N, Baker CRF, et 18. Reilly PM, Bulkley GB. Vasoactive media- the primary triggers of multiple organ system al. Alterations in cellular membrane func- tors and splanchnic perfusion. Crit Care failure. Intense vasoconstriction occurs early, tion during hemorrhagic shock in pri- Med 1993; 21:S55–68. and frequently leads to a “no-reflow” phenom- mates. Ann Surg 1972; 176:288–95. 19. Redan JA, Rush BF, McCullogh JN, et al. enon even when the macrocirculation is re- 7. Shires GT, Coln D, Carrico J, et al. Fluid Organ distribution of radiolabeled enteric stored.18 Intestinal cell death causes a break- therapy in hemorrhagic shock. Arch Surg Escherichia coli during and after hemor- down in the barrier function of the gut, which 1964; 88:688–93. rhagic shock. Ann Surg 1990; 211:663–8. results in increased translocation of bacteria 8. Runciman WB, Sjowronski GA. Patho- 20. Korinek AM, Laisne MJ, Nicholas NH, to the liver and lung.19 The impact of this on physiology of haemorrhagic shock. Raskine L, Deroin V, Sanson-Lepors MJ. Se- the development of multiple organ failure is Anaesth Intensive Care 1984; 12:193–205. lective decontamination of the digestive controversial at present.20 9. Bronshvag MM. Cerebral pathophysiology tract in neurosurgical intensive care pa- The liver has a complex microcirculation in haemorrhagic shock: nuclide scan data, tients: a double-blind, randomized, pla- and has been demonstrated to suffer fluorescence microscopy, and anatomic cebo-controlled study. Crit Care Med reperfusion injury during recovery from correlations. Stroke 1980; 11:50–9. 1993; 21:1466–73. shock.21 Hepatic cells are also metabolically 10. Peterson CG, Haugen FP. Hemorrhagic 21. Chun K, Zhang J, Biewer J, Ferguson D, active and contribute substantially to the inflam- shock and the nervous system. Am J Surg Clemens MG. Microcirculatory failure deter- matory response to decompensated shock. Ir- 1963; 106:233–9. mines lethal hepatocyte injury in ischemic- regularities in blood glucose levels following 11. Collins JA. The pathophysiology of hem- reperfused rat livers. Shock 1994; 1:3–9. shock are attributable to hepatic ischemia.22 orrhagic shock. Prog Clin Biol Res 1982; 22. Maitra SR, Geller ER, Pan W, Kennedy PR, Failure of the synthetic functions of the liver 108:5–29. Higgins LD. Altered cellular calcium regu- following shock are almost always lethal. 12. Troyer DA. Models of ischemic acute relation and hepatic glucose production Skeletal muscle is not metabolically active nal failure: do they reflect events in hu- during hemorrhagic shock. Circ Shock during shock, and tolerates ischemia better man renal failure? J Lab Clin Med 1987; 1992; 38:14–24. than other organs. The large mass of skeletal 110:379–80. 23. Peitzman AB, Corbett WA, Shires GT III, muscle, though, makes it important in the gen- 13. Lefer AM, Martin J. Origin of a myocardial Illner H, Shires GT, Inamder R. Cellular eration of lactate and free radicals from is- depressant factor in shock. Am J Physiol function in liver and muscle during hem- chemic cells. The classic cellular response to 1970; 218:1423–7. orrhagic shock in primates. Surg Gynecol shock of increasing intracellular sodium and 14. Horovitz, JH, Carrico CJ, Shires GT. Pul- Obstet 1985; 161:419–24. free water were first elucidated in skeletal muscle cells.23

Conclusion SECTION II: Therapeutic Strategies Traumatic shock is a disease not just of 3 hemorrhage but also of tissue ischemia. Bleed- Surgical Perspectives to ing can be controlled surgically and oxygen delivery restored through adequate transfu- Control Bleeding in Trauma sion, and the patient can still die as a result of the accumulated metabolic load of prolonged Brian R. Plaisier, MD nal – blood loss onto the “street” or the trauma hypoperfusion. Although control of bleeding Department of Surgery room floor, 2) left and right hemithoraces, 3) and restoration of the circulating blood vol- Bronson Methodist Hospital peritoneal cavity, 4) pelvis and retroperitoneum, ume must remain the cornerstones of care for 252 East Lovell, Box 67 and 5) long-bone fracture sites. Methods of de- the traumatized patient, we must build on this Kalamazoo MI 49007 USA finitive hemostatic control may be very simple, foundation techniques for the management of e-mail: [email protected] as in the application of direct pressure to a lac- reperfusion injury, the inflammatory cascade, eration, or very complex, such as in the patient and “no reflow” if we are truly going to im- After establishing a secure airway and en- with a pelvic fracture who requires emboliza- prove long-term survival. suring adequate oxygenation and ventilation, tion. This article addresses surgical, pharmaco- the highest priority in the trauma patient is to logic, and various other nonsurgical methods References control hemorrhage. Because patients may to control bleeding. 1. Peitzman AB, Billiar TR, Harbrecht BG, bleed from multiple sites simultaneously, it is Kelly E, Udekwu AO, Simmons RL. Hem- imperative that the surgeon establish a strategy Hemostasis orrhagic Shock. Curr Probl Surg to address all possible sources of bleeding and Hemostasis is the process that terminates 1995;929–1002. control them. These sources include 1) exter- blood loss from an injured blood vessel. Sur-

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 7 geons depend greatly on normal hemostasis, often taking it for granted, so that surgery may Figure 1. Complex interaction of vasoconstriction, be conducted safely. The process is very effi- platelet factors, and coagulation reactions. cient and utilizes circulating proteins, cellular (Reproduced with permission from Mosby-Year Book Inc.) elements, and the endothelial lining (Fig. 1).1 The first response to injury is vasoconstriction, which decreases blood flow distal to the laceration. The mechanism for vasoconstriction involves both direct injury and reflex responses. Platelets are exposed to subendothelial collagen and quickly adhere to each other and the blood vessel wall. Von Willebrand’s factor acts as a bridge between the subendothelium and the platelet membrane, where it binds to receptor sites made available as a result of platelet activation. Other platelets are then recruited from the blood, and a loose plug forms to seal the blood vessel. If this response reaches sufficient intensity, the platelet release reaction occurs whereby the contents of the platelet and its granules are lib- erated into the surrounding microenvironment. This is a complex reaction involving adenosine diphosphate, serotonin, platelet factor 4, platelet-derived growth factor, thrombin, calcium, and magnesium.1,2 The result is the formation of a stable platelet plug, which, unlike the initial loose plug, is no longer reversible. Figure 2. Coagulation reactions. Platelet reactions occur simultaneously (Reproduced with permission from Mosby-Year Book, Inc.) with the events of the coagulation cascade. Co- agulation serves to convert prothrombin into thrombin, which, in turn, converts fibrinogen to fibrin. This process utilizes circulating inac- tive proenzymes, which are converted into an active form and then, in turn, activate the next proenzyme in the sequence. There are two distinct divisions of the coagulation process: 1) the intrinsic pathway and 2) the extrinsic pathway (Fig. 2). The intrinsic pathway is initiated by the interaction of Factor XII and nonendothelial surfaces, which induces a conformational change in Factor XII. The complicated reactions that follow lead to clotting, kinin formation, complement activation, and fibrinolysis.1 The extrinsic pathway is the more important pathway in hemostasis. Thromboplastin, a lipoprotein, is released from cells in response to tissue trauma. When thromboplastin is present, Factor VII becomes active and the sequence ensues. The two pathways merge into a common pathway with the activation of Factor X, which, in turn, converts prothrombin to thrombin. Fi- brinogen is then acted upon by thrombin, resulting in the formation of fibrin monomers. Polymerization of the fibrin monomers occurs, resulting in a cross-linked, stable, fibrin clot. Fibrinolysis is the process that limits the hemostatic response to the local area of injury and maintains vascular patency throughout the organism. This system is initiated simultaneously the site of clotting. A complex inhibition system Abnormalities of Hemostasis Resulting with the clotting mechanism and is under the inactivates any plasmin that gains access to the from Injury influence of numerous circulating mediators. The general circulation. Other methods the body uses Injury triggers a vast array of responses release of plasminogen activator from injured to limit coagulation, which are beyond the scope that affect hemostasis. Patients may exhibit ei- endothelium and activation of Factor XII initiate of this discussion, include products of the cyclo- ther a hypercoagulable or hypocoagulable state fibrinolysis. These convert plasminogen to plas- oxygenase enzyme pathway, protein C, and anti- following trauma. Severely injured patients min, which can digest fibrin and fibrinogen at thrombin III. have elevated serum fibrin degradation prod-

8 Massive Transfusion and Control of Hemorrhage in the Trauma Patient ucts, lowered platelet counts, and activation of The most familiar means of achieving de- the kallikrein–kinin system.3 In survivors, these Table 1. finitive hemostasis is the placement of surgi- values normalize within the first few days after Surgical Priorities at cal ligatures and clips. These must be placed injury but continue to worsen in nonsurvivors. Laparotomy in the Trauma Patient very accurately so as not to endanger surround- Interestingly, a hypercoagulable state is seen in ing structures. Small vessels may be managed patients with less severe injuries. This is caused • Control of exsanguinating with simple ligatures; large arteries should be by a suppression of thrombolysis.3 hemorrhage controlled with a suture ligature to prevent Coagulopathy after injury may also result • Stop gastrointestinal contamination slippage of the tie. In very confined spaces from the patient’s abnormal physiology • Thorough exploration of entire where the placement of ties would be difficult, (hypoperfusion or hypothermia) or the inter- abdomen surgical clips may be applied. ventions used to treat the patient (massive trans- • Definitive repair of all injuries Occasionally an organ such as the liver or fusion). It has become clear that prophylactic spleen may be lacerated, but removal may not administration of fresh frozen plasma or plate- be necessary. Organ-wrapping methods utilize lets in the absence of clinical bleeding is not a mesh net to envelope the liver or spleen to warranted.4,5 Hypothermia has been shown to Table 2. gain tamponade. This method may be used adversely affect coagulation, and it is important Surgical Methods to when other methods to achieve hemostasis fail that surgeons and anesthesiologists strive to Control Bleeding or where splenectomy or extensive maintain normothermia during treatment. In- hepatorraphy would otherwise be required. juries to the brain and liver and those that cause • Proper exposure Thermal agents such as electrocautery either hypoperfusion or tissue devitalization • Digital pressure produce hemostasis by heating and denatur- have significant potential to induce • Sutures and clips ing proteins, resulting in coagulation. Both al- coagulopathy.4,5 The importance of the restora- • Thermal coagulation ternating and direct current may be employed tion of tissue perfusion and debridement of • Topical hemostatic agents for this purpose. This method allows rapid ces- devitalized tissue cannot be overemphasized. • Organ wrapping sation of bleeding but may result in large areas of tissue necrosis if applied carelessly. Priorities in the Operating Room Occasionally an organ such as the liver or At laparotomy it is absolutely necessary to trol of surgical bleeding and if coagulopathy, spleen may be lacerated, but removal may not control hemorrhage and gastrointestinal con- hypothermia, and acidosis are present, laparo- be necessary. Hemostasis may be accomplished tamination in the most rapid fashion possible. tomy sponges may be placed between the ab- by several methods, such as direct pressure or Dr. William Halsted6 considered this absolutely dominal wall and the bleeding organ to gain suture repair. The liver or spleen may also be essential for all types of surgery and eloquently tamponade. The laparotomy is terminated wrapped with a mesh netting to envelop the stated the rationale: quickly to allow transfer to the intensive care organ to gain tamponade. This method may be The confidence gradually acquired unit so that coagulopathy, acidosis, and hypo- used when other methods to achieve hemosta- from masterfulness in controlling thermia may be corrected. A second operation sis fail or where splenectomy of extensive hemorrhage gives to the surgeon the is required to remove the packs once the hepatorraphy would otherwise be required but calm which is so essential for clear patient’s condition is more stable. may compromise chances for survival. thinking and orderly procedure at the operating table. Table 3. It is only after hemorrhage is controlled that Comparison of Topical Hemostatic Agents a patient’s injuries may be addressed in an orderly fashion (Table 1). Control of gastrointestinal contamination is the next goal. Only after these goals are accomplished can a thorough exploration of the abdomen can be conducted and all injuries addressed definitively. The surgeon has a wide range of tools to employ in order to control bleeding (Table 2). The most obvious method is the application of digital pressure. Although not definitive control for large vessels, the surgeon’s finger is the most atraumatic instrument available and will control bleeding temporarily while the blood vessel is exposed. The offending blood vessel must be exposed properly prior to repair or ligation. Occasionally, one may need to gain control of the aorta at the diaphrag- matic hiatus to allow the anesthesiologist time to replace blood and fluids while exposure is being accomplished. The patient’s condition may not allow all injuries to be addressed fully at initial exploration. An abbreviated laparotomy to control hemorrhage, followed by continued resuscitation in the intensive care unit, is now an established concept in trauma surgery.7 If the patient’s condition is deteriorating after con- Reproduced with permission from Innovative Publishing Incorporated.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 9 Pharmacologic agents have gained an im- antifibrinolytic agent such as aprotinin may be portant place in the surgeon’s armamentarium. added to the second solution, depending on Table 4. The mechanisms of action are widely varied: specific requirements.9 When these two parts Nonsurgical Interventions some act by vasoconstriction, some by supply- are combined, clotting ensues. The glue may to Achieve Hemostasis ing a scaffold for attracting blood elements, and be applied by two methods: 1) In the “sand- still others by promoting coagulation per se wich technique,” the fibrinogen is spread onto • Pneumatic antishock garment (Table 3).8 The ideal topical hemostat would the surface to be sealed and the thrombin solu- • External pelvic fixator have several properties: 1) rapid hemostasis, 2) tion spread over it. 2) The premixed method • Angiography and embolization easily applied and manipulated, 3) holds su- uses two syringes joined by a Y-connector. • Temporary balloon occlusion tures, 4) little tissue reaction, 5) low infectious Ochsner et al used fibrin glue as the pri- risk, 6) absorbable, and 7) easily removed. Each mary hemostatic agent or as an adjunct to con- of these agents has particular advantages and ventional suture repair in 26 patients with he- disadvantages, which will be discussed in brief. patic and splenic trauma.10 Seventeen patients Other Invasive Interventions Topical epinephrine is used commonly in had liver injuries (6 blunt and 11 penetrating) Numerous other tools for hemorrhage applications such as burn surgery and exerts and 9 had splenic injuries (7 blunt and 2 pen- control may be used in the field, emergency its action by promoting vasoconstriction. The etrating). Liver injuries ranged from moderate department, or radiologic suite (Table 4). Al- drug can be used to cover wide surfaces, but it to severe and the splenic injuries were all though the surgeon does not necessarily per- must be applied with caution because systemic moderate. Fibrin glue achieved hemostasis in form all of these procedures, he or she should effects may result if excess drug is used. 21 patients with the first application and with be responsible for combining them into a logi- Oxidized cellulose (i.e., Surgicel®) acts by the second in the remaining five. No patients cal strategy for prompt control of bleeding forming a gelatinous mass on contact with were re-explored for bleeding. Eight patients when surgical methods cannot be used. blood. This compound conforms well to ir- had postoperative coagulopathy and thromb- The pneumatic antishock garment is used regular surfaces, is relatively inert, causes little ocytopenia, but the fibrin glue hemostasis re- to control bleeding temporarily in patients with tissue reaction, and is absorbed in 1 to 2 weeks. mained effective. pelvic and lower extremity fractures by acting In addition, cellulose holds sutures relatively A controlled in vitro review of topical he- as a splint to tamponade bleeding. It can be well and is bactericidal. mostatic agents was undertaken by Wagner et used for hypovolemic shock, but it is only a Collagen sponges (Actifoam®, Helistat®, al.11 The tested agents included three types of temporizing measure. Prolonged use may be Instat®) and microfibrillar collagen (i.e., collagen sponges (Actifoam®, Helistat®, associated with numerous complications, such Avitene®) have a rapid time to hemostasis and Instat®), microfibrillar collagen (Avitene®), a as compartment syndrome. are absorbed in approximately 8 to 12 weeks. gelatin sponge (Gelfoam®), and oxidized re- The external pelvic fixator may be defini- Sponges are easy to apply and they remove and generated cellulose (Surgicel®). Actifoam® and tive in stopping bleeding from veins lining frac- hold sutures well. Microfibrillar collagen packs Avitene® caused the greatest response (both tured pelvic bones. It is most effective in pa- easily into small spaces but is difficult to re- statistically similar) in an in vitro platelet ag- tients with fractures associated with a diasta- move and sticks to gloves and instruments. gregation test. Gelfoam® exhibited an interme- sis of the pubic symphysis (“open-book” pel- Thrombin is a protein that converts fi- diate response, whereas Helistat®, Surgicel®, vic fractures), since it draws the anterior ele- brinogen to fibrin, resulting in clot formation. and Instat® caused a lesser degree of platelet ments together. This decreases the potential Thrombin may be applied as a liquid or pow- aggregation. In a similar test using thrombin to space into which bleeding may occur. The ex- der or combined with another carrier such as presoak each agent, platelet aggregation oc- ternal fixator is not effective for fractures in- Gelfoam®. Hemostasis is rapid and wide sur- curred at a more rapid rate for all agents tested. volving only the posterior elements of the pel- faces may be treated. The agents were also tested in their abil- vic ring or in controlling bleeding from the Denatured gelatin (i.e,. Gelfoam®) pos- ity to induce gross blood coagulation (Lee– arteries coursing through the pelvis. sesses no clotting activity itself but provides a White clotting time). Actifoam®, Avitene®, and For patients with bleeding from pelvic scaffold on which clot can form. It also helps Helistat® responded in a manner similar to fractures in whom an external fixator is not plug small blood vessels by virtue of its bulk thrombin, but Instat®, Gelfoam®, and effective, bleeding from arteries in the pelvis when moistened. It may be used as a carrier Surgicel® demonstrated no significant impact must be suspected and angiography should be for other compounds such as thrombin. The on clotting time. performed. If an offending vessel is identified, sponge should be pre-moistened with either Wagner et al, using the above assays as well embolization may be carried out with either saline or thrombin and all air should be re- as tests of platelet deposition and platelet ad- Gelfoam® or metal microcoils (Fig. 3). While moved from the interstices by compressing the enosine triphosphate secretion, constructed an sponge. Gelatin conforms well to surfaces, but overall ranking of these hemostatic agents: Fig. 3. it does not hold sutures. Actifoam® ~ Avitene® > Helistat® >> Microcoils used in pelvic Fibrin sealants have numerous applica- Gelfoam® > Instat® > Surgicel®. It should arterial embolization. tions within the field of surgery, including be noted that, although this ranking notes dif- (Photograph courtesy of nerve anastomoses, intracranial operations, ferences between the agents for these in vitro James Newman, MD, PhD.) skin grafting, and cardiovascular procedures.9 assays, it is certainly limited when considering Fibrin glue has also been used as a hemostatic the numerous clinical situations encountered agent in trauma surgery for lacerations of the by surgeons in a wide variety of subspecialties. liver and spleen. In the presence of calcium Heat energy has a significant role in ions, fibrinogen and Factor XIII are activated treating the hypothermic trauma patient. by thrombin. Fibrinogen is converted to fibrin Hypothermia causes platelet dysfunction and monomers and these, in turn, are polymerized prolongs clotting times.12 Laboratory assays to form a stable clot. underestimate the extent to which hypoth- Fibrin glue has two components that must ermia affects bleeding, since the plasma and be mixed together for clotting to occur. The test reagents are heated to 37°C prior to run- primary parts of the first component are fibrino- ning the assay. Because of this, coagulation gen and Factor XIII. The second component test results and platelet counts may not cor- consists of thrombin and calcium chloride. An relate with nonsurgical bleeding.

10 Massive Transfusion and Control of Hemorrhage in the Trauma Patient not usually used for pelvic arteries, balloon tion on pharmacologic methods, specifically Transfusion therapy. In Maull KI, occlusion may be used by the angiographer as topical hemostatic agents. Each of these agents Rodriguez A, Wiles CE, eds. Complications a temporizing measure to achieve hemostasis has particular advantages and disadvantages in Trauma and Critical Care. Philadel- in arteries of the chest, neck, and extremities and must be applied to the appropriate situa- phia, WB Saunders, 1996. before the causative lesions are controlled in tion. There are other invasive techniques that 6. Halsted WS. The Johns Hopkins Hospital the operating room. may not be performed by the surgeon but that Reports 1920; 19:71. Cited in Schwartz SI, I would be remiss if I did not emphasize must be orchestrated by the surgeon into a Green RM. Biology of hemostasis. In the importance of the anesthesia service in the clear strategy for hemorrhage control. The an- Schwartz SI, ed. Techniques of Hemosta- management of these patients. Surgeons must esthesiologist has an important role in help- sis. West Berlin, New Jersey, Innovative focus on control of bleeding at the surgical site. ing to control hemorrhage by appropriate Publishing Incorporated, 1993. Anesthesiologists provide the necessary factors transfusion therapy but, more importantly, pre- 7. Rotondo MF, Schwab CW, McGonigal, et to assist in the correction of surgical bleeding venting bleeding at the surgical site by meth- al. “Damage control”: An approach for im- and the prevention of nonsurgical bleeding. ods such as maintaining normothermia. proved survival in exsanguinating pen- The proper transfusion of blood component etrating abdominal injury. J Trauma 1993; therapy has important implications for control References 35:375. of bleeding, since platelets and coagulation 1. Clagett GP. Hemostasis in surgical patients. 8. Schwartz SI, Moore EE. Local hemostasis. factors may be required by severely injured In Miller TA, ed. Physiologic Basis of Mod- In Schwartz SI, ed. Techniques of Hemo- patients. Anesthesiologists must also focus on ern Surgical Care. St. Louis, Mosby, 1988. stasis. West Berlin, New Jersey, Innovative the maintenance of normothermia to help pre- 2. Schwartz SI, Green RM. Biology of hemo- Publishing Incorporated, 1993. vent coagulopathy. Effective communication stasis. In Schwartz SI, ed. Techniques of 9. Lerner R, Binur NS. Current status of sur- between the surgeon and anesthesiologist is Hemostasis. West Berlin, New Jersey, In- gical adhesives. J Surg Res 1990; 48:165. essential. The surgeon must alert the anesthe- novative Publishing Incorporated, 1993. 10. Ochsner MG, Maniscalco-Theberge ME, siologist to bleeding at the surgical site, so that 3. Rutledge R, Sheldon GF. Bleeding and co- Champion HR. Fibrin glue as a hemostatic corrective methods may be undertaken. agulation problems. In Feliciano DV, agent in hepatic and splenic trauma. J Moore EE, Mattox KL, eds. Trauma, 3rd Trauma 1990; 30:884. Summary ed. Stamford, Connecticut, Appleton and 11. Wagner WR, Pachence JM, Ristich J, et al. Control of blood loss is one of the most Lange, 1996. Comparative in vitro analysis of topical important priorities in the trauma patient. We 4. Knudson MM. Coagulation disorders. In hemostatic agents. J Surg Res 1996; 66:100. have discussed several methods of obtaining Ivatury RR, Cayten CG, eds. The Textbook 12. Gentilello LM. Advances in the manage- hemostasis. These include standard surgical of Penetrating Trauma. Baltimore, Mary- ment of hypothermia. Surg Clin North Am techniques such as digital pressure and su- land, Williams & Wilkins, 1996. 1995; 75:243. tures. We have focussed much of our atten- 5. Phillips GR, Rotondo MF, Schwab CW.

4 Haemostatic Drugs in Trauma and Orthopaedic Practice

Dr. David Royston dose aprotinin therapy reduced drain losses Many forms of bone and joint surgery are Consultant Anaesthetist (range, 35%–81%), the total amount of trans- associated with a significant risk of bleeding Royal Brompton and Harefield NHS Trust fusions (range, 35%–97%), and the propor- and thus the use of blood and blood prod- Harefield, Middlesex UB9 6JH tion of patients requiring transfusions of ucts.5 A number of systems have been used to United Kingdom blood or blood products (range, 40%–88%). reduce this probability. Some of these are al- e-mail: [email protected] Since the first description of the haemostatic most unique to orthopaedic surgery, such as actions of high-dose aprotinin therapy in pa- creating a bloodless field by tourniquet ap- Aprotinin is a naturally occurring serine tients undergoing re-operation4 or high-risk plication in limb surgery. In addition, in many protease inhibitor. It is found in the mast cells cardiac procedures, this agent has been the countries, orthopaedic and trauma surgeons of all mammalian species as well as many lower standard of care in this situation and is the have become the principle users of orders of life. Unfortunately, at this time, we only product licensed for use for this indica- predonated blood and blood product sys- do not understand the true physiologic role tion in North America. tems. However, there is still significant scope of aprotinin in nature. The aim of this article is to discuss the for the use of other techniques and methods, What is known is that high doses of the potential for this anti-inflammatory and such as pharmacologic intervention, to inhibit drug inhibit a number of the inflammatory haemostatic action to benefit patients having bleeding and minimize the need for blood and processes involved with open heart surgery elective orthopaedic and trauma surgery and blood product transfusions. and also modify the haemostatic system to al- also following trauma itself. The article is di- low reductions in bleeding and thus the need vided into three major sections dealing with Nonemergency Orthopaedic Surgery for blood and blood products. The use of The three most commonly used pharma- high-dose aprotinin therapy followed reports • The use of drugs to prevent bleeding dur- cologic interventions in nonemergency ortho- of the potential benefit of this approach in ing elective surgery paedic surgery are tranexamic acid, traumatically injured patients.1 Large-dose desmopressin (DDAVP), and aprotinin. Each aprotinin therapy has been shown to be ex- • The potential for aprotinin therapy in pa- of these agents has a relatively unique mode tremely effective, and safe, in preventing tients who have sustained trauma of action, although there is overlap between blood loss and the need for blood and blood some of the physiologic events produced by products in patients undergoing open heart • The potential use of serine protease in- these agents. surgery. The current literature contains more hibitors to prevent certain sequelae of Desmopressin is a synthetic analogue of than 40 reports of randomised placebo-con- trauma and surgery of bones, joints, and the natural hormone argenine vasopressin and trolled studies2,3 that have shown that high- tendons has been shown to increase plasma levels of

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 11 factor VIII activity in patients with the early literature in which hemophilia and Von Willebrand’s Figure 1 aprotinin was administered to disease type I. Desmopressin had Total numbers of transfused red cells in patients undergoing patients who had sustained considerable support for use in orthopaedic surgery for removal of infected bioprosthesis or trauma, particularly road traffic complex heart surgery, but more resection of tumour with and without aprotinin therapy. crashes. The majority of these pa- recent data suggest that, overall, Individual patient data are shown and demonstrate an average pers are found in the German lit- this drug provided little, if any, three-fold reduction in need for blood products in patients given erature. In one multiple centre benefit to the patient. Conversely, aprotinin therapy. Data are drawn from Capdevila et al.17 study published in 1976,24 4,686 more recent data suggest that the patients were entered into a mul- use of desmopressin in patients tiple centre study to investigate who are currently taking aspirin the effects of aprotinin therapy therapy has significant benefit in the treatment of traumatic during and after heart surgery.6,7 shock. The dosage used was rela- The results with desmopressin in tively low—approximately 3 mil- orthopaedic surgery have been lion KIU over a 2-day period— universally poor.8 However, at but produced an impressive ben- present, there is little information efit to the patient outcome when about the use of this agent in pa- administered within a few hours tients taking aspirin or non-steroi- of the trauma. The most signifi- dal anti-inflammatory drugs cant benefits of the use of (NSAIDs) prior to surgery. This is aprotinin therapy were found in obviously an area that needs fur- patients with injuries to the up- ther investigation. per extremity and soft tissues, With regard to tranexamic but there were significant ben- acid, there are again few data to efits following trauma to the support its use for surgery of cen- lower limb and spine as well. The tral bones and joints, such as spine and hip Soft Tissue Injury and Disseminated study found no benefit of the use of aprotinin surgery. Some reports indicate a reduction in Intravascular Coagulopathy (DIC) therapy in patients with either chest or head the requirement for blood in patients who are The use of blood-sparing agents in trauma injury (Fig. 2). having knee replacement surgery under tour- surgery has potential in soft tissue injury and Most recently, a number of studies have niquet.9,10 in patients with intraabdominal (especially focused on aprotinin therapy following blunt Aprotinin therapy has been used with ef- hepatic) trauma. Severe soft-tissue injury pre- liver trauma. These investigations appear to be fect in a wide variety of surgeries, including sents a variety of challenges with problems a natural progression from studies that inves- orthopaedic surgery. There is, however, still associated with the initial event, the subse- tigated this therapy in patients having liver only one report from a randomised placebo- quent potential for ischaemia reperfusion in- transplantation. In both an animal25 and a hu- controlled study in patients undergoing pri- jury, and the development of a coagulopathy man26 study, significant benefits were achieved mary hip surgery.11 This shows a significant during resuscitation. in terms of reduction in bleeding and the need benefit of high-dose aprotinin therapy to re- Soft-tissue trauma is associated with the for donor blood in liver trauma and resection. duce drain losses and the need for donor blood release of a number of procoagulant media- These data suggest that aprotinin therapy and blood products. The dose of aprotinin tors, which can lead to a form of disseminated may be beneficial in certain patients with soft- used in this study from Belgium was intended intravascular coagulation and haemorrhage. tissue injury and intraabdominal trauma. to be equivalent to the high-dose regimen used The use of factors to promote haemostasis and during cardiac surgery. prevent bleeding in these circumstances is still Antiinflammatory Actions A number of other studies have shown an controversial. The use of apure In addition to the potential haemostatic effect of lower doses of aprotinin on variables antifibrinolytics, such as a lysine analogue, is benefits of the use of aprotinin in patients such as platelet function but without showing potentially lethal in these circumstances. These undergoing elective surgery and in those who consistent benefit to reduce the requirement drugs are therefore contraindicated in the pres- have been injured, there are also a number of for transfusions.12–16 It appears that a higher ence of intravascular thrombin generation. other actions of the drug that may benefit the dose of aprotinin is needed to ensure reduced Indeed, in animal models, the use of lysine patient. These are related to its anti-inflam- blood transfusions than the dose that will have analogue antifibrinolytics such as tranexamic matory and anticoagulant actions.27 significant effects on haemostatic processes. acid with excess thrombin generation leads to All serine protease inhibitors, including Similarly, there is evidence that the greater the death of the animal.19–21 aprotinin, will inhibit platelet function. This the surgical risk, the more benefit the high- In contrast to the effects of these lysine is achieved by a number of mechanisms re- dose regimen appears to demonstrate. For ex- analogue agents are the effects of serine pro- lated to the ability to inhibit surface recep- ample, a recent article17 showed that aprotinin tease inhibitors. A number of odious models tors and by intracellular metabolism pro- therapy produced a three-fold reduction in the of tissue injury in animals have shown signifi- cesses. Indeed, the first use of aprotinin need for blood and blood products in patients cantly high early mortality. These models in- therapy in patients having hip surgery was as undergoing hip replacement because the joint clude rotating drum experiments with rats and an adjunct to the use of heparin to prevent had become infected or invaded by tumor (Fig. fracture/sepsis models in sheep. In both these deep venous thrombosis after surgery.12,14 Pre- 1). This massive reduction in the requirements experimental models, the use of aprotinin liminary data from these studies (involving for blood and blood products is similar to the therapy prevented mortality and improved small numbers of patients) suggest a small but observations in heart surgery, where the higher outcome.22 A number of animal models to- statistically significant effect to reduce the the risk of bleeding, the more obvious is the gether with anecdotes about humans suggest incidence of venous thrombosis. This effect benefit of aprotinin therapy. In addition, that aprotinin therapy in addition to heparin needs to be investigated in larger groups of aprotinin therapy has been used with benefit inhibits the DIC associated with trauma and patients using various forms of antithrombotic in patients undergoing spinal surgery.18 sepsis.19,21,23 prophylaxis in addition to aprotinin therapy There are also a number of studies from to determine if there is a significant benefit

12 Massive Transfusion and Control of Hemorrhage in the Trauma Patient in this respect together paedic/trauma surgery suggest that the benefits with a reduction in the Figure 2 of this drug can be obtained without increas- requirement for blood Percentage mortality in groups of patients with trauma ing the risk of a thrombotic episode. Whether and blood products. and tissue injury with (hatched bars) and without (clear bars) the incidence of thrombosis can be reduced Other reports suggest aprotinin therapy. Data are drawn from more than further by co-administration of a protease in- that the incidence and 4,000 patients.24 Mortality rates are significantly lower in hibitor with other antithrombotic prophylaxis severity of pulmonary fat aprotinin-treated patients for lower-limb and soft-tissue remains to be investigated. embolism and the fat em- injury (p <0.01) and for lower-limb trauma and spinal bolus syndrome following injury (p <0.05). There was no significant difference References trauma are reduced sig- between treatment and no treatment in patients 1. Clasen C, Jochum M, Mueller Esterl W. nificantly with aprotinin with predominantly head injury. Feasibility study of very high aprotinin dos- therapy.28,29 age in polytrauma patients. Prog Clin Biol A further conse- Res 1987; 236a:175–83. quence of the use of 2. Davis R, Whittington R. Aprotinin: a re- aprotinin and its effects view of its pharmacology and therapeu- on intracellular metabo- tic efficacy in reducing blood loss associ- lism is inhibition of cer- ated with cardiac surgery. Drugs 1995; tain aspects of ischaemia 49:954–83. and reperfusion injury. In 3. Royston D. High-dose aprotinin therapy: particular there is consid- a review of the first five years’ experi- erable evidence to show ence. J Cardiothorac Vasc Anesth 1992; that aprotinin therapy is 6:76–100. associated with improved 4. Royston D, Bidstrup BP, Taylor KM, microvascular blood Sapsford RN. Effect of aprotinin on need flow.27 This improved for blood transfusion after repeat open- flow together with modi- heart surgery. Lancet 1987; 2:1289–91. fications to the metabolic 5. Clarke AM, Dorman T, Bell MJ. Blood loss process may explain why and transfusion requirements in total joint there is a significant re- arthroplasty. Ann R Coll Surg Engl 1992; duction in the amount of lactic acid produced homology with aprotinin.34 74:360–3. after ischaemia reperfusion in patients under- Preliminary data from human studies sug- 6. Dilthey G, Dietrich W, Spannagl M, Rich- going hip surgery12,30 and in those undergoing gest that the chronic injection of aprotinin into ter J. Influence of desmopressin acetate vascular surgery with aorto-bifemoral replace- the joint space is associated with a significant on homologous blood requirements in ment.31 inhibition of progression of disease.35 A simi- cardiac surgical patients treated with as- One potential area for the use of aprotinin lar mechanism may also play a part in the use pirin. J Cardiothorac Vasc Anesth 1993; as a treatment after bone surgery is to inhibit of aprotinin therapy to prevent adhesion for- 7:425–30. the oedema that occurs after trauma to bone mation and fibrosis following tendon repair.36 7. Laupacis A, Fergusson D. Drugs to mini- and periosteum. Oedema formation can be as- mize perioperative blood loss in cardiac sociated with considerable discomfort. A num- Summary and Conclusion surgery: meta-analyses using perioperative ber of studies suggest that the local infiltra- The use of aprotinin therapy in sufficiently blood transfusion as the outcome. The In- tion of aprotinin significantly reduces both the high doses is associated with an improvement ternational Study of Peri-operative Trans- oedema formation and the pain associated with in haemostatic function and a reduction in fusion (ISPOT) Investigators. Anesth Analg bone surgery. This is especially true for pa- drain losses and blood utilisation in patients 1997; 85:1258–67. tients requiring maxillofacial surgery and den- undergoing major trauma surgery and ortho- 8. Mannucci P. Hemostatic drugs. N Engl J tal extraction.32 paedic surgery. The anti-inflammatory actions Med 1998; 339:245–53. Finally, there is the potential for the use of aprotinin may also have significant benefit 9. Benoni G, Fredin H. Fibrinolytic inhibi- of aprotinin and other protease inhibitors to in reducing mortality after soft tissue trauma tion with tranexamic acid reduces blood be used prophylactically and in treatment of alone and especially in those traumas associ- loss and blood transfusion after knee ar- patients with progressive joint destruction or ated with increased risk of embolic phenom- throplasty: a prospective, randomised, following joint and tendon repair. It is becom- ena or intravascular coagulation. Although double-blind study of 86 patients. J Bone ing increasingly recognised that many of the drugs such as tranexamic acid have value in Joint Surg Br 1996; 78:434–40. cells in cartilage to produce proteolytic en- patients requiring certain joint replacement 10. Hiippala S, Strid L, Wennerstrand M, et zymes, which may be responsible for chronic surgeries, their safety in the presence of a al. Tranexamic acid (Cyklokapron) re- joint destruction.33 More modern methods of prothrombotic state is not proven. Therefore, duces perioperative blood loss associated molecular biology suggest that one of the ma- at this stage, it seems inappropriate to recom- with total knee arthroplasty. Br J Anaesth jor participants in this process is the genera- mend these drugs for patients with soft tissue 1995; 74(5):534–7. tion of plasmin from a urokinase plasminogen trauma. The use of drugs such as desmopressin 11. Janssens M, Joris J, David JL, Lemaire R, type activator. This activity is inhibited by in otherwise routine surgery has, as yet, no Lamy M. High-dose aprotinin reduces aprotinin therapy in tissue culture.33 The ra- proven benefit, although there may be some blood loss in patients undergoing total hip tionale for using intra-articular aprotinin benefit to patients who are taking anti-inflam- replacement surgery. Anesthesiology therapy is suggested by the observation that matory medicines. 1994; 80:23–9. chondrocytes produce a number of protease In addition to the benefit of reducing 12. Haas S, Ketterl R, Stemberger A, et al. The inhibitors of the proteolytic enzymes such as bleeding, protease inhibitors can improve pa- effect of aprotinin on platelet function, the plasminogen activators. One of the major tient outcome by reducing ischaemic injury blood coagulation and blood lactate level inhibitors thus far categorised from human and the oedema formation that may cause pain. in total hip replacement - a double blind chondrocytes is a 6-kD molecule that has re- At present, safety data on the use of aprotinin clinical trial. Adv Exp Med Biol 1984; markable, if not identical, amino acid sequence therapy in both open heart surgery and ortho- 167:287–97.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 13 13. Freick H, Reuter HD, Piontek R. [Supple- tion of thrombosis during intravascular proteinase inhibitors.] Langenbecks Arch mentary preoperative prevention of coagulation 1: comparative effects of Chir 1969; 325:369–72. thromboembolism through the use of norepinephrine in thrombin and adenos- 29. Weisz GM, Barzilai A. Fat embolism: phys- aprotinin in alloplastic hip joint prosthe- ine diphosphate (ADP) treated rabbits. iopathology, diagnosis with management. sis?] Med Welt 1983; 34:614–9. Pathology 1984; 16:411–7. Arch Orthop Unfallchir 1975; 82:217–23. 14. Ketterl R, Haas S, Lechner F, Kienzle H, 21. Moriau M, Rodhain J, Noel H, et al. Com- 30. Wendt P, Ketterl R, Haas S, et al. [Postop- Blumel G. [Effect of aprotinin on throm- parative effects of proteinase inhibitors, erative increase in lactate in total hip bocytic function during total plasminogen antiactivators, heparin and endoprosthesis operations: effect of endoprosthesis surgery of the hip.] Med acetylsalicylic acid on the experimental aprotinin. Results of a clinical double- Welt 1980; 31:1239–43. disseminated intravascular coagulation blind study.] Med Welt 1982; 33:475–9. 15. Hayes A, Murphy DB, McCarroll M. The induced by thrombin. Thromb Diath 31. Horl M, Sperling M, Herzog I, et al. Effect efficacy of single-dose aprotinin 2 million Haemorrh 1974; 32:171–88. of aprotinin on metabolic changes in KIU in reducing blood loss and its im- 22. Dwenger A, Remmers D, Grotz M, et al. blood following aortofemoral bypass op- pact on the incidence of deep venous Aprotinin prevents the development of eration. Eur Surg Res 1985; 17:186–96. thrombosis in patients undergoing total the trauma-induced multiple organ fail- 32. Brennan PA, Gardiner GT, McHugh J. A hip replacement surgery. J Clin Anesth ure in a chronic sheep model. Eur J Clin double blind clinical trial to assess the value 1996; 8:357–60. Chem Clin Biochem 1996; 34:207–14. of aprotinin in third molar surgery. Br J Oral 16. Utada K, Matayoshi Y, Sumi C, et al. 23. Kolbow H, Barthels M, Oestern HJ, et Maxillofac Surg 1991; 29:176–9. [Aprotinin 2 million KIU reduces al. [Early changes of the coagulation sys- 33. Ronday HK, Smits HH, Quax PH, et al. perioperative blood loss in patients un- tem in multiple injuries and their modi- Bone matrix degradation by the plasmi- dergoing primary total hip replacement.] fication with heparin and Trasylol.] Chir nogen activation system. Possible mecha- Masui 1997; 46:77–82. Forum Exp Klin Forsch, April 1977, pp nism of bone destruction in arthritis. Br 17. Capdevila X, Calvet Y, Biboulet P, et al. 119–23. J Rheumatol 1997; 36:9–15. Aprotinin decreases blood loss and ho- 24. Schneider B. [Results of a field study on 34. Rodgers KJ, Melrose J, Ghosh P. Purifica- mologous transfusions in patients under- the therapeutic value of aprotinin in trau- tion and characterisation of 6 and 58 kDa going major orthopedic surgery. Anesthe- matic shock (author’s transl).] forms of the endogenous serine protein- siology 1998; 88:50–7. Arzneimittelforschung 1976; 26:1606–10. ase inhibitory proteins of ovine articular 18. Llau JV, Hoyas L, Higueras J, et al. 25. Thomae KR, Mason DL, Rock WA Jr. Ran- cartilage. Biol Chem 1996; 377:837–45. [Aprotinin reduces intraoperative bleed- domized blinded study of aprotinin in- 35. Capasso G, Testa V. [Infiltrations in ing during spinal arthrodesis interven- fusion for liver crush injuries in the pig gonarthrosis, a therapeutic turning point: tions (letter).] Rev Esp Anestesiol Reanim model. Am Surg 1997; 63:113–20. the use of a proteinase inhibitor.] Arch 1996; 43:118. 26. Lentschener C, Benhamou D, Mercier FJ, Putti Chir Organi Mov 1990; 38:277–84. 19. Arnljots B, Wieslander JB, Dougan P, et al. Aprotinin reduces blood loss in pa- 36. Komurcu M, Akkus O, Basbozkurt M, et Salemark L. Importance of fibrinolysis in tients undergoing elective liver resection. al. Reduction of restrictive adhesions by limiting thrombus formation following Anesth Analg 1997; 84:875–81. local aprotinin application and primary severe microarterial trauma: an experi- 27. Royston D. Preventing the inflammatory sheath repair in surgically traumatized mental study in the rabbit. Microsurgery response to open-heart surgery: the role flexor tendons of the rabbit. J Hand Surg 1991; 12:332–9. of aprotinin and other protease inhibitors. Am 1997; 22:826–32. 20. Latour JG, Leger Gauthier C, Daoust Int J Cardiol 1996; 53(suppl):S11–37. Fiorilli J. Vasoactive agents and produc- 28. Morl FK, Heller W. [Fat embolism and

5 Antithrombotics in Trauma Care: Benefits and Pitfalls John K. Stene, MD, PhD morbidity, anesthesiologists are faced with Other high-risk conditions for DVT include Department of Anesthesia patients who receive LMWH prophylaxis for bed rest for longer than 72 hours; lower-extrem- The Milton S. Hershey Medical Center DVT, which may preclude the use of epidural ity fractures, especially pelvis and acetabular Hershey PA 17033 USA catheters.3,5 In this article, the risks and natu- fractures; penetrating venous injuries; head in- e-mail: [email protected] ral history of DVT and recommendations for juries inducing a low Glasgow Coma Scale use of continuous epidural anesthesia in con- score; family history of thrombi; and a history Although deep venous thrombosis (DVT) junction with LMWH will be reviewed. of DVT or PE. Also associated with appreciable and pulmonary embolism (PE) have always Venous thromboembolic disease—which DVT risk are comorbid conditions such as age been major complications of trauma, they have includes both DVT and PE—is a major post- greater than 40; obesity; malignancy; pregnancy, only recently become a major concern of traumatic morbidity and mortality issue.1,2 Di- up to 1 month postpartum; use of oral contra- trauma anesthesiologists because modern ef- rect trauma to blood vessels and thrombophilia ceptives; and lung operations.7 fective prevention of DVT affects anesthesia associated with the general inflammatory re- Virchow noted in the 19th century that DVT practice.1 Recently developed low-molecular- sponse to traumatic injury lead to an increased was initiated by one or more of the following weight heparins (LMWH) provide very effec- incidence of DVT and subsequent pulmonary conditions: stasis of blood flow in the deep veins tive prevention of posttraumatic DVT with far embolism (PE).7 The overall incidence of DVT of the leg, trauma to the endothelial lining of the fewer bleeding complications than intravenous in the North American and European popula- veins, and increased coagulability of the blood. unfractionated heparin; however, LMWH use tions is 1 in 1,000. It occurs more frequently Because trauma patients are at risk for the entire is also associated with epidural hematomas in older people, obese people, and patients Virchow’s triad, they are at increased risk of DVT from epidural catheters.2–8 Thus, at a time when with traumatic injury. Some injury patterns and thus PE. Aside from the obvious vascular the use of continuous regional anesthesia with such as spinal cord injury are associated with trauma and venous stasis caused by bed rest, the epidural catheters is shown to reduce trauma a very high incidence of DVT. inflammatory state of trauma (e.g., cytokine re-

14 Massive Transfusion and Control of Hemorrhage in the Trauma Patient lease) causes a generalized thrombophilia. This topic V/Q scans.11 Spiral CT of the chest en- thromobophilia may cause multiple microvascu- Table 1. Inborn Errors of Coagulation hanced with contrast medium may reveal a fill- lar thrombin formation and DIC or lead to large- ing defect of a major branch of the pulmonary scale thrombus in the deep veins of the leg. Factor Population artery in patients with PE.10 A pulmonary ar- Incidence Patients with congenital forms of throm- tery angiogram remains the gold standard for Activated protein bophilia are at especially high risk for DVT and diagnosing PE by revealing filling defects in the C resistance 3%-4% PE. These inborn errors of metabolism (Table pulmonary artery or its branches. (Factor V Leiden) 1) vary in incidence from 3% to 4% in the popu- Hyperhomocysteinemia 5% Treatment of known DVT is anticoagula- lation for activated protein C resistance (Fac- Protein C deficiency 0.2%-0.4% tion to prevent further propagation of the tor V Leiden) to 0.02% for antithrombin defi- Protein S deficiency 0.1% thrombus. Therapeutic heparinization is usu- 7 ciency. Although protein C and protein S de- Antithrombin deficiency 0.02% ally performed with intravenous heparin ti- ficiency were involved in the earliest descrip- trated to a PTT in the range of 60 to 80 sections of thrombophilia, they are not nearly as onds. Because patients with one episode of common as factor V Leiden deficiency. Dimer is released into the circulation when DVT are at risk for another episode, therapeu- Hyperhomocysteinemia, which is also a risk intravascular clots are broken down by throm- tic heparinization is usually followed by 3 factor for arterial thrombosis such as coronary bolysis.12 Venous duplex Doppler reveals months to a lifetime of warfarin or long-term arterial occlusion, may be related to folic acid, noncompressibility of flow in the proximal subcutaneous LMWH.14 vitamin B12, and vitamin B6 deficiency and its deep veins of the leg. Venography is the gold Treatment of PE is mostly supportive of incidence is not well defined.9 standard of diagnosis for DVT, allowing filling pulmonary function along with administration The main reason to worry about DVT in defects to be seen after injection of contrast of heparin to prevent further clot buildup. the trauma patient is fatal PE. PE occurs symp- dye in a peripheral limb vein. Approximately Heparin dose is adjusted for PTT of 60 to 80 tomatically in 30% of patients with a DVT; as- 2% of DVTs occur in the upper extremities, with seconds. Embolectomy of the pulmonary ar- ymptomatic PE increases the overall incidence a risk of 12% for PE from an upper-extremity tery has been used for “saddle emboli” ob- of PE to 50% to 60% in patients with DVT. The DVT.13 Upper-extremity DVT is diagnosed by structing both branches of the pulmonary ar- death rate caused by PE is 50,000 per year in detection of obstruction to flow in the deep tery, but embolectomy must be initiated almost the United States. Because there is no adequate veins of the shoulder or upper arm. immediately to be effective. treatment for a diagnosed PE and diagnosis is PE is diagnosed by physical examination, Prevention of DVT relies on a combination difficult, prevention of DVT is the most effec- radioisotopic ventilation perfusion (V/Q) scan, of mechanical methods such as stockings, se- tive measure to reduce the incidence of PE.7,8 spiral computed tomography (CT) of the chest, quential compression devices, or foot pumps Therefore, many trauma patients will receive or pulmonary angiography (Table 3). Physical and pharmacologic techniques (Table 4). Infe- DVT prophylaxis, and the use of LMWH is the signs and symptoms of PE include pleuritic rior vena caval filters are designed to prevent PE most effective prophylaxis. chest pain, dyspnea, hemoptysis, abnormal but have no effect on the development of DVT. DVT is diagnosed with physical examina- breath sounds, atrial dysrhythmias, hypoxia, Pharmacologic prevention of DVT has tion, chemical markers of coagulation such as and an increase in arterial to end tidal carbon been attempted with aspirin, dextran, heparin, D-dimer, venous duplex Doppler, and venog- dioxide gradient. The presence of a known warfarin, and LMWH, and thrombolytics have raphy (Table 2).10–12 Symptoms and signs of DVT increases the probability that these signs been used to lyse established DVT. Of these DVT include pain, a venous cord along the leg, and symptoms represent a PE. A perfusion pharmacologic agents, LMWH, low-dose sub- edema distal to the occlusion, and pain in- defect not matched with a simultaneous venti- cutaneous heparin, warfarin, and intravenous duced by forceful dorsiflexion of the foot. D- lation defect demonstrates a PE on radioiso- high-dose heparin have proven effective in preventing DVT.2,4,6,7,15 LMWH is more efficacious than low-dose heparin, especially in pre- Table 2. Diagnosis of Deep Venous Thrombosis venting PE, and has fewer complications.4,2,15 Because of its ease of use, efficacy, and low Physical examination Simple and inexpensive; needs laboratory confirmation incidence of side effects, LMWH is the drug of D-Dimer Indicates intravascular coagulation choice for DVT prophylaxis in trauma patients. Venous duplex Doppler Useful screening tool Table 5 lists the recommended doses and dos- Venography Gold standard ing intervals for available LMWH, as well as the current indications for these drugs. The complications of DVT prophylaxis in- Table 3. Diagnosis of Pulmonary Embolism clude bleeding, epidural hematoma, heparin-induced osteopenia, heparin-induced thrombocy- Investigation Comments topenia (HIT), and warfarin-induced skin necrosis. LMWH is much less likely to cause osteopenia History and physical examination Simple; easy to use than unfractionated heparin, but both are likely Dyspnea to cause HIT.7 For patients who develop HIT, DVT Chest pain propylaxis and treatment of PE can be controlled Hemoptysis with danaparoid or hirudin. Atrial dysrhythmias The use of regional anesthesia in trauma Friction rub patients receiving anticoagulation therapy for Hypoxia DVT prophylaxis requires the compulsive fol-

Decrease in PETCO2 lowing of guidelines to prevent epidural he- 3,5 V/Q scan Useful only to confirm physical findings matomas. Table 6 lists recommendations for Spiral CT of chest Good sensitivity and specificity epidural catheter use in patients receiving Pulmonary angiography Gold standard LMWH. With the use of sequential compression devices during periods when LMWH can-

PETCO2, end-tidal carbon dioxide; V/Q, ventilation-perfusion; CT, computed tomography not be administered safely, prophylaxis against venous thromboembolic disease as well as ex-

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 15 Table 4. Techniques to Prevent Posttraumatic DVT Table 6. Use of Neuraxial Block in Technique Effectiveness Anticoagulated Patients Sequential compression device Effective; easy to use with minimal complications 1. Do not mix ASA and other Foot pumps Effective; easy to use with minimal complications anticoagulants. Compression stockings Not effective; cheap; easy to use 2. Bloody tap requires a 24-hour Subcutaneous heparin Inexpensive; easy to use; high incidence of bleeding delay of anticoagulation. Low-molecular-weight heparins Easy to use; highly effective; low incidence 3. Delay needle placement for 12 of bleeding hours after LMWH administration. Vena caval filters Effective; highly invasive; may lead to chronic 4. Delay catheter withdrawal for 12 venous obstruction hours after LMWH administration. 5. Delay LMWH dosing until at least 2 hours after needle placement Table 5. Low-Molecular-Weight Heparin Preparations and/or catheter removal. 6. Be extremely vigilant in patients with epidural catheter who are Drug Indications Subcutaneous Injection Dose receiving LMWH.

Ardeparin DVT prophylaxis: Knee replacement urgery 50 antiXa U/kg BID LMWH, low-molecular-weight heparin

Dalteparin DVT prophylaxis: Hip replacement 2,500-5,000 IU QD 5. Vandermeulen EP, Van Aken H, Vermylen surgery, abdominal surgery J. Anticoagulants and spinal-epidural an- (at-risk patients*) esthesia. Anesth Analg 1994; 79:1165–77. 6. Angelli G, Piovella F, Buoncristiani P, et al. Danaparoid† DVT prophylaxis: Elective hip surgery 750 antiXa units BID Enoxaparon plus compression stockings compared with compression stockings Enoxaparin DVT prophylaxis: Hip and knee Prophylaxis hip and knee: replacement surgery, abdominal surgery 30 mg BID or 40 mg QD alone in the prevention of venous throm- (at-risk patients*) Prophylaxis abdominal: boembolism after elective neurosurgery. 40 mg QD N Engl J Med 1998; 338:80–5. 7. Hyers TM. Venous thromboembolism. Am Inpatient treatment of acute DVT with Inpatient treatment: J Respir Crit Care Med 1999; 159:1–14. or without PE, in conjunction with 1 mg/kg BID or 1.5 mg/kg QD 8. Goldhaber SZ. Pulmonary embolism. N warfarin Engl J Med 1998; 339:93–104. 9. Den Heijer M, Koster T, Blom HJ, et al. Outpatient treatment of acute DVT Outpatient Treatment: Hyperhomocysteinemia as a risk factor for without PE, in conjunction with warfarin 1 mg/kg BID deep-vein thrombosis. N Engl J Med 1996; Prevention of ischemic complications of Ischemia: 1mg/kg BID 334:759–62. unstable angina and non-Q wave MI 10. Remy-Jardin M, Remy J, Deschildre F, et (when used concurrently with aspirin) al. Diagnosis of pulmonary embolism with spiral CT: comparison with pulmonary DVT, deep vein thrombosis; MI, myocardial infarction; PE, pulmonary embolism. angiography and scintigraphy. Radiology *At-risk: age > 40, obesity, general anesthesia >30 minutes, history of malignancy or DVT or 1996; 200:699–706. pulmonary embolism. 11. PIOPED Investigators. Value of the venti- †Danaparoid is an antithrombotic agent with an average molecular weight of ~5,500 daltons. lation/perfusion scan in acute pulmonary embolism. JAMA 1990; 263:2753–9. 12. Ginsberg JS, Wells PS, Kearon C, et al. Sen- cellent analgesia from continuous epidural References sitivity and specificity of a rapid whole- analgesia can be provided to trauma patients. 1. Geerts WH, Code KI, Jay AM, et al. A pro- blood assay for D-dimer in the diagnosis spective study of venous thromboembo- of pulmonary embolism. Ann Intern Med Summary lism after major trauma. N Engl J Med 1998; 129:1006–11. The use of LMWH for reducing the risk of 1994; 331:1601–6. 13. Nemmers DW, Thorpe PE, Knibbe MA, DVT and PE has gained increasing popularity 2. Geerts WH, Jay RM, Code KI, et al. A com- Beard DW. Upper extremity venous throm- in trauma patients with pelvic fractures requir- parison of low-dose heparin with low-mo- bosis. case report and literature review. ing operative fixation or prolonged (>5 days) lecular-weight heparin as prophylaxis Orthop Rev 1990; 19:164–72. bed rest, in patients with complex lower extrem- against venous thromboembolism after 14. Kearon C, Gent M, Hirsh J, et al. A com- ity fractures requiring operative fixation or pro- major trauma. N Engl J Med 1996; parison of three months of anticoagulation longed bed rest, and in spinal-cord-injured pa- 335:701–7. with extended anticoagulation for a first tients with complete or incomplete motor pa- 3. Horlocker TT, Heit JA. Low molecular episode of idiopathic venous thromboem- ralysis. However, the use of LMWH in trauma weight heparin: biochemistry, pharmacol- bolism. N Engl J Med 1999; 340:901–7. can be a challenge, necessitating a fine balance ogy, perioperative prophylaxis regimens, 15. Imperiale TF, Speroff T. A meta-analysis of between bleeding risk and DVT/PE risk. There and guidelines for regional anesthetic man- methods to prevent venous thromboem- are many unresolved issues concerning the use agement. Anesth Analg 1997; 85:874–85. bolism following total hip replacement. of antithrombotics in trauma patients, which 4. Clagett GP, Anderson FA Jr, Heit J, et al. JAMA 1994; 271:1780–5. require further investigation, especially in pa- Prevention of venous thromboembolism. tients receiving continuous neuraxial analgesia. Chest 1995; 108(4 suppl):312S–334S.

16 Massive Transfusion and Control of Hemorrhage in the Trauma Patient 6 Atraumatic Blood Salvage and Autotransfusion in Trauma and Surgery

Sherwin V. Kevy, MD* Robert Brustowicz, MD** Table 1. Comparison of the BloodStream System with *Transfusion Service Wall Suction Using Yankauer (Y) and Frazier (F) Suction Wands **Department of Anesthesia Children’s Hospital Method/Pressure Flow rate (L/min)* Harvard Medical School SD ± ml/min Boston, Massachusetts BloodStream/100 mmHg 3.74 ± 25 The experience with many trauma victims has emphasized the need for a blood source Wall suction/100 mmHg ± other than banked blood. Cardiopulmonary Y50515 1.36 20 ± bypass and vascular surgery have established F3310 0.64 14 unwashed filter autotransfusion as a safe and Wall suction/150 mmHg† practical means to supplement homologous Y50515 1.75 ± 15 blood usage. During major orthopedic surgi- F3310 0.84 ± 19 cal procedures, autotransfusion has been demonstrated to reduce blood requirements. Wall suction/200 mmHg† The properties of an ideal autotransfusion Y50515 2.10 ± 18 system include 1) ease of operation, 2) rela- F3310 0.98 ± 20 tively low cost, 3) in-line filtration system, 4) simplified anticoagulation, 5) high fluid aspi- Wall suction/250 mmHg† ration rate and minimal hemolysis whether Y50515 2.19 ± 17 evacuating a pool of blood or surface skim- F3310 1.12 ± 22 ming from the operative field, and 6) the abil- Wall suction/450 mmHg† ity to concentrate the aspirated blood. Y50515 3.03 ± 18 The BloodStream Recovery System (BRS) F3310 1.66 ± 19 (Harvest Technologies LLC, Norwell, Massa- chusetts) (Fig. 1) is a surgical suction system that automatically senses the pressures re- *Mean flow rates observed during evacuation of a pool of blood (volume, 3,000 ml; hema- quired and adjusts from 20 to 40 mmHg dur- tocrit, 24%) ing surface skimming and a maximum of 100 †These pressure levels are not recommended for collection for autotransfusion. mmHg when evacuating a pool of blood. Dur- ing trauma and cardiovascular surgery, the BRS can be utilized as a stand-alone autotransfusion lection system to cell-washing systems by con- Results system by transferring from the collection res- necting the BRS reservoir to the intake line of Results obtained during evacuation of a ervoir to a reinfusion bag that contains an in- the cell-washing machine. pool of blood are shown in Tables 1 and 2. tegral 40-micron filter. During orthopedic sur- Flow rates obtained with the BRS are more than gery, the BRS can be used as the front-end col- Methods twice those obtained with a Yankauer or Frazier The BloodStream was compared with wall suction want at vacuum pressures of 100 and suction (SS) at vacuum pressures of 100 to 450 150 mmHg (Table 1). The latter level is greater mmHg during blood pool evacuation and sur- than that recommended for autotranfusion or Fig. 1. BloodStream Recovery System face skimming. A variety of suction wands were intraoperative blood salvage. When the (Harvest Technologies LLC, used with both suction systems. BloodStream serves as the vacuum source, flow Norwell, Massachusetts). Multiple red cell pools were required for rates obtained with Yankauer and Frazier suc- the studies. The pools are identified by duration wands are comparable to those obtained tion of storage, hematocrit, and pertinent con- with a wall suction system at 200 mmHg (Tables trol values. 1 and 2).

Table 2. Comparison of the flow rates obtained with the BloodStream, Yankauer (Y), and Fazier (F) Suction Wands when the BloodStream was the Vacuum Source

BloodStream as the Flow Rate (L/min)* Vacuum source (100 mmHg) SD ± ml/min

BloodStream wand 3.60 ± 223

Yankauer 50515 2.00 ± 19

Frazier 3310 0.92 ± 21

*Mean flow rates observed during evacuation of a pool of blood (volume, 3,000 ml; hematocrit, 25%).

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 17 Table 3 illustrates an example of the results obtained during surface skimming. As one in- Table 3. Comparison of the BloodStream System with Wall Suction creases the vacuum pressure, there is a tendency at Vacuum Pressures Between 100 and 450 mmHg Using Yankauer (Y50515) for the Frazier wand to grab onto tissue, reduc- and Frazier (F3310) Suction Wands During Surface Skimming ing the flow rate and increasing red cell damage. The BRS has significantly greater flow and Method/Pressure Flow rate (L/min)* Plasma Hemoglobin (mg/dl) results in significantly less damage to red cells, as determined by plasma hemoglobin levels. A Blood Stream/20-40 mmHg 270 ± 17 24.3 ± 2 wall suction vacuum pressure of 200 mmHg is Wall suction/100 mmHg required to achieve the flow rate obtained with Y50515 242 ± 21 32.7 ± 5 the BRS at 40 mmHg. However, this results in a F3310 165 ± 19 60.3 ± 3 two-fold increase in plasma hemoglobin. The BRS system can be used for Wall suction/150 mmHg* autotransusion of unwashed shed blood dur- Y50515 250 ± 19 40.0 ± 3 ing cardiac and vascular surgery. F3310 140 ± 18 53.0 ± 4

Conclusions Wall suction/200 mmHg* The BloodStream system can rapidly Y50515 268 ± 22 59.3 ± 5 evacuate a pool of blood at more than twice F3310 153 ± 19 97.6 ± 7 the flow rate achieved with a wall suction sys- Wall suction/300 mmHg* tem at recommended pressures. The Y50515 279 ± 24 63.9 ± 7 BloodStream results in significantly less dam- F3310 143 ± 16 117 ± 11 age to red cells as determined by plasma hemoglobin levels compared with wall suction. Wall suction/450 mmHg* The data demonstrate that the BloodStream Y50515 277 ± 27 109.0 ± 8 system can be used as a stand-alone mobile F3310 138 ± 13 500 ± 27 surgical suction system in operating rooms, emergency departments, and trauma centers. *These pressure levels are not recommended for collection for autotransfusion.

SECTION III: Transfusion: Clinical Practice 7 Current Practices in Fluid and Blood Component Therapy in Trauma

Charles E. Smith, MD, FRCPC Chapters 10 and 12). Monitoring of tempera- then be used for maintenance of general anes- Chair, ITACCS Special Techniques/ ture, urine output, arterial blood gases, hemo- thesia until the intravascular volume deficit has Equipment Committee globin, hematocrit, electrolytes, and param- been corrected and bleeding is under control. Department of Anesthesia eters of coagulation is routine in severely in- Neuromuscular relaxants, benzodiazepines, MetroHealth Medical Center jured patients. An arterial catheter is usually and other agents are given as clinically indi- Professor of Anesthesia warranted after basic management priorities cated.2 Case Western Reserve University are fulfilled. Consideration is given to place- Cleveland OH 44109 USA ment of invasive monitors (e.g., central venous Fluid Options e-mail: [email protected] catheter, pulmonary artery catheter), There is controversy about which IV solu- transesophageal echocardiography, and provi- tions should be used for resuscitation. During The acutely volume-depleted trauma pa- sion of anesthesia as needed. hemorrhage, the interstitial space, in addition tient will have cool, moist, pallid, or cyanotic For induction of anesthesia in hemody- to the intravascular compartment, is dimin- skin, especially at the extremities. Initial evalu- namically unstable patients, etomidate or ished, with a compensatory increase in reab- ation of the patient will include an estimate of ketamine is useful.2 Titrated opioids and am- sorption of fluid into the capillaries. To replete blood volume deficit (Table 1), rate of addi- nestic concentrations of volatile agents can the intravascular and interstitial compartment, tional blood loss, primary and secondary sur- crystalloid solutions such as iso- vey according to ATLS™ principles, and an tonic 0.9% saline are given ini- 3 evaluation of cardiopulmonary reserve and co- Table 1. tially. Glucose-containing solu- 1 existing hepatic or renal dysfunction. The Estimation of Blood Volume Deficit in Trauma tions are avoided because hyper- major goal in resuscitation is to stop the bleed- glycemia aggravates central ner- 4,5 ing and replete intravascular volume to maxi- Unilateral hemothorax 3,000 ml vous system injury. Lactated mize tissue oxygen delivery. Cardiac output, Ringer’s solution has an osmo- Hemoperitoneum with blood pressure, and oxygenated blood flow to lality of 273 mOsm/L, which is abdominal distension 2,000–5,000 ml vital organs are important determinants of out- hypotonic with respect to come. Full-thickness soft-tissue plasma. Moreover, lactated Management priorities in the trauma pa- defect 5 cm3 500 ml Ringer’s cannot be used to dilute tient who is bleeding acutely include ventila- Pelvic fracture 1,500–2,000 ml packed red blood cells. Thus, tion and oxygenation; measurement of blood Femur fracture 800–1,200 ml 0.9% saline is preferred. Colloid pressure; placement of ECG, pulse oximeter, solutions such as hetastarch have and capnograph; and establishment or verifi- Tibia fracture 350–650 ml been shown to be as effective as cation of adequate intravenous (IV) access for Smaller fracture sites 100–500 ml 5% albumin for volume expan- infusion of normothermic fluids (See also sion. Hetastarch is used after the

18 Massive Transfusion and Control of Hemorrhage in the Trauma Patient initial phase of resuscitation, which occurs af- Whole blood is not available at the author’s Cryoprecipitate may then be indicated to cor- ter cessation of bleeding, and is characterized institution. The initial choice will depend on rect specific factor deficiencies. It should be by interstitial fluid sequestration and maximal the degree of hemodynamic instability. One noted that 1 unit of fresh whole blood or 1 weight gain. Large amounts of hetastarch unit of packed red blood cells will usually in- unit of single-donor apheresis platelets also has (>15–20 ml/kg) are avoided because of the risk crease the hematocrit by ~3% or the hemo- similar factor levels as 1 unit of fresh frozen of coagulopathy.6 globin by 1 g/dl in a 70-kg non-bleeding adult. plasma. Similarly, 4 to 5 multiple donor plate- Type O-negative red cells have no major let units have similar factor levels as 1 unit of Delayed Fluid Resuscitation antigens and can be given reasonably safely to fresh frozen plasma because the platelets are The use of large quantities of fluids for patients with any blood type. Unfortunately, suspended in plasma. immediate resuscitation of victims of penetrat- only 8% of the population has O-negative Dilutional thrombocytopenia in the ing trauma before hemorrhage is controlled blood, and blood bank reserves of O-negative, trauma patient also occurs. Leslie and Toy16 by surgical means has been questioned.7 Dis- low-antibody-titer blood are usually very low. showed that platelet count was reduced to advantages of immediate fluid resuscitation are For this reason, O-positive red cells are fre- <50,000/µl after administration of 20 units of that inserting venous cannulae and giving fluid quently used. This is a reasonable approach in red cells. Platelet transfusions are usually in- boluses in the prehospital setting may delay males but may be a problem in childbearing- dicated in the presence of clinical bleeding and transfer and surgical intervention, may contrib- aged females who are Rh negative. a platelet count <75,000 to 100,000/µl. Plate- ute to secondary hemorrhage by disrupting or If 50% to 75% of the patient’s blood vol- let concentrates are stored at room tempera- decreasing resistance to flow around a partially ume has been replaced with type O blood (e.g., ture and contain about 70% of the platelets in formed thrombus or by increasing blood pres- ~10 units of red cells in an adult patient), one a unit of blood. One unit of platelets, equiva- sure, may dilute clotting factors, and can con- should continue to administer type O red cells. lent to 50 ml, increases the platelet count in tribute to hypothermia. In a randomized, pro- Otherwise, risk of a major cross-match reac- an adult by 5,000 to 10,000/µl, and is given spective trial of immediate versus delayed fluid tion increases since the patient may have re- through a 170-µ filter. resuscitation in patients with penetrating ceived enough anti-A or anti-B antibodies to trauma, there was increased mortality, length precipitate hemolysis if A, B, or AB units are Hypertonic Fluids of stay, and postoperative complication rate in subsequently given.2 Lesser amounts of hypertonic fluids, as the immediate versus the delayed group.7 How- Obtaining type-specific red cells requires opposed to isotonic fluids, can also provide ever, the study was limited to isolated torso 5 to 10 minutes in most institutions, and tem- rapid volume expansion and improved hemo- injuries, and the receiving trauma center had porizing measures can sometimes be em- dynamics and have the added advantage of a very rapid response time such that most pa- ployed to gain the necessary time. At our insti- decreasing tissue edema, intracranial pressure, tients were in the operating room within 1 tution, we use a tube system to deliver blood and brain water. These hypertonic solutions hour of injury. Therefore, results of this study samples and products to and from the operat- result in an osmotic translocation of extracel- may not be applicable to other types of inju- ing room or trauma resuscitation suite. This lular and intracellular water. Because the in- ries such as blunt trauma, head injury, and system significantly reduces delays and “lost” travascular half-life of hypertonic saline is simi- multiple sites of penetrating trauma or to pa- samples. The use of type-specific red cells is lar to that of isotonic saline, these fluids can tients in remote locations requiring long trans- preferred over O-negative and the risk of a be combined with colloid solutions such as port times. hemolytic transfusion reaction is very low.12 If hetastarch or dextran to prolong their plasma one can wait 15 minutes, typed and screened volume expansion effects. Hypertonic saline Red Cell Transfusion blood should be available. When blood is typed has been associated with bleeding, hemody- The lower limit of anemia is not estab- and screened, the patient’s blood group is namic deterioration, and increased mortality lished in humans. Oxygen delivery is gener- identified and the serum is screened for major in animal studies of uncontrolled hemorrhagic ally adequate with a hemoglobin of 7 g/dl, blood group antibodies. A full cross match shock.17 Further, it does not improve cerebral which corresponds to an oxygen delivery of generally requires about 45 minutes and in- oxygen delivery after head injury and mild ~500 ml/min in a 70-kg patient, assuming nor- volves mixing donor cells with recipient serum hemorrhage in animals.18 Nonetheless, hyper- mal cardiac output and hemoglobin/oxygen to rule out any antigen/antibody reactions.13 tonic saline combined with 6% hydroxyethyl saturation. In otherwise healthy, normovolemic starch has been shown to improve neurologic individuals, Messmer and colleagues8 demon- Coagulation Factors and Platelets function and cerebral perfusion pressure in strated that tissue oxygenation is maintained The primary cause of bleeding after patients with traumatic brain injury.18a This with hematocrit between 18% and 25%. The trauma is surgical, while the second leading hypertonic fluid solution is currently used in heart and brain are often considered to be most cause is hypothermia. In the past, coagulopathy Austria for resuscitation of all head-injured and vulnerable to the effects of anemia. The heart after massive transfusion with whole blood was major trauma patients in the field (Mauritz W, begins to produce lactic acid at hematocrits usually caused by dilutional thrombocytope- personal communication). between 15% and 20%,9 and heart failure gen- nia. However, this is not necessarily the case erally occurs at hematocrit of 10%.10 Generally, with red cells reconstituted in 0.9% saline. Endpoints of Resuscitation hematocrits between 25% and 30% result in Murray et al have shown that microvascular Blood and fluid resuscitation is continued optimal oxygen delivery, but therapy must be bleeding and clinical evidence of coagulopathy until perfusion has been improved and organ individualized.11 occurred in the setting of massive transfusion function has been restored. Manifestations of Factors affecting the transfusion trigger for and was associated with decreased coagulation improved cardiac output include improved red cells include the rate and magnitude of factor levels, decreased fibrinogen, elevated mental status; increased pulse pressure; de- blood loss; degree of cardiopulmonary reserve; prothrombin times and platelet counts creased heart rate; increased urine output; presence of atherosclerotic disease of the >100,000/µl.14,15 Moreover, administration of resolution of lactic acidosis and base deficit; brain, heart, and kidneys; and oxygen con- fresh frozen plasma corrected the microvascu- brisk capillary refill; and improvement in oxy- sumption.11 If the patient has lost large lar bleeding. Two units of fresh frozen plasma gen delivery, oxygen consumption, and cen- amounts of blood and is in class III or IV shock (10–15 ml/kg) will achieve 30% factor activity tral venous or pulmonary artery oxygen satu- (see table on page 4), blood administration is in most adults. Coagulation factor deficiencies ration (Table 2).19 required.2 Available options are type O-nega- may be present due to other causes such as tive, type-specific, typed and screened, or typed preexisting defects or disseminated intravas- Blood and Fluid Warmers and cross-matched packed red blood cells. cular coagulopathy resulting from tissue injury. Fluid and blood resuscitation of the

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 19 result in a decrease of 1.5oC in core tempera- Table 2. Resuscitation Endpoints Within the First 24 Hours After Trauma ture. Similarly, infusion of 2.3 liters of red cells could result in a core temperature decrease of Parameter Value between 1 and 1.5oC.32,33 Since the thermal Oxygen delivery index >600 ml/min/m2 stress of infusing fluids at normothermia is Oxygen consumption index >150 ml/min/m2 essentially zero, it follows that use of fluid- warming devices effective at delivering normo- Mixed venous oxygen tension >35 mmHg thermic fluids to the patient at clinically rel- Mixed venous oxygen saturation >65% evant flow rates permits more efficient rewarm- (central venous or pulmonary artery) ing of hypothermic trauma patients using other Base deficit >-3 mmol/L methods such as the patient’s own metaboli- Lactate <2.5 mmol/L cally generated heat, or externally provided heat such as convective warming.22 Adapted from Ivatury RR, Simon RJ. Assessment of tissue oxygenation (evaluation of the adequacy of resuscitation). In Ivatury RR, Cayten CGC, eds. The Textbook of Penetrating Citrate Intoxication, Hyperkalemia, and Trauma. Baltimore, Williams & Wilkins, 1996, pp 927–938. Acid–Base Abnormalities Blood is stored in citrate phosphate dextrose with adenine or adsol at 4oC. Citrate in- trauma patient is best accomplished with large- trauma victim.16 In the author’s opinion, the toxication is caused by acutely decreased se- gauge intravenous catheters and effective fluid platelet count should be monitored and main- rum levels of ionized calcium, which occurs warmers with high thermal clearances.20 Be- tained at or greater than 75,000 to 100,000/µl because citrate binds calcium.34 Administration cause alterations in red cell integrity are not to achieve adequate surgical hemostasis. It is of calcium is warranted during massive trans- apparent until 46oC,21 fluid warmers with set advisable that prothrombin time, activated fusion if the patient is hypotensive and mea- points of 42oC are now commonly used. Coun- partial thromboplastin time, fibrinogen, and sured serum ionized serum calcium is low or tercurrent water and other fluid warmers us- fibrin degradation products be monitored be- large amounts of blood are infused rapidly (50 ing 42oC set points will not damage red cells, cause deficiencies may be present due to dilu- to 100 ml/min). Ionized serum calcium levels will result in consistently warmer fluid deliv- tion, preexisting defects, or disseminated in- will usually return to normal when hemody- ery, and will allow the clinician to maintain travascular coagulopathy.24 Point-of-care test- namic status is improved. The potassium level thermal neutrality with respect to fluid man- ing and rapid reporting of coagulation test re- in stored blood rises with length of storage and agement over a wide range of flow rates.22 sults should be used to guide decisions regard- can be as high as 78 mmol/L after 35 days. The ing administration of fresh frozen plasma, potential for clinically important hyperkalemia Complications of Transfusion Therapy platelets, or cryoprecipitate. still exists in patients receiving blood administered at rates >120 ml/min35 and in patients Impaired Oxygen Release from Hemoglobin Hypothermia with severe acidosis. Monitoring the ECG for The ability of the red blood cell to store The adverse effects of hypothermia in the signs of hyperkalemia is always warranted, and and release oxygen is impaired after storage. trauma patient include major coagulation de- treatment of hyperkalemia with calcium chlo- The erythrocytic levels of 2,3-diphosphoglyc- rangements, peripheral vasoconstriction, meta- ride, bicarbonate, glucose, and insulin may be eric acid decrease both with CPD and CPDA-1 bolic acidosis, compensatory increased oxygen life saving. stored blood. Low levels of 2,3-diphosphoglyc- requirements during rewarming, and impaired The pH of bank blood decreases to about eric acid will shift the blood’s oxygen dissocia- immune response.25–27 Standard coagulation 6.9 after 21 days of storage because of accu- o tion curve to the left, and the red cell will have tests are temperature corrected to 37 C and mulation of CO2, lactic acid, and pyruvic acid a higher affinity for oxygen at physiologic PO2 may not reflect hypothermia-induced by red blood cell metabolism. Thus, the aci- and will release less oxygen at a given tissue coagulopathy.28-30 Hypothermia impairs coagu- dosis seen in stored blood is partly respiratory 23 PO2. Impaired oxygen release from hemoglo- lation because of slowing of enzymatic rates and partly metabolic. The respiratory compo- bin can be minimized by warming all blood and reduced platelet function. Hypothermia nent is of little consequence with adequate and by avoiding factors that shift the O2 disso- can cause cardiac dysrhythmias and even car- patient ventilation. The metabolic component ciation curve to the left, e.g., hypothermia. diac arrest due to electromechanical dissocia- is not usually clinically significant. It is unwise tion, standstill, or fibrillation, especially with to administer sodium bicarbonate on an em- Dilutional Coagulopathy core temperatures below 30oC. Hypothermia piric basis, because there is already a pool of Most coagulation factors are stable in also impairs citrate, lactate, and drug metabo- bicarbonate generated from the metabolism of stored whole blood, except factors V and VIII.13 lism; increases blood viscosity; impairs red citrate, which is present in large quantities in These factors gradually decrease to 15% and blood cell deformability; increases intracellu- stored blood. 50% of normal, respectively, after 21 days of lar potassium release; and causes a leftward storage. However, most centers today use shift of the oxyhemoglobin dissociation curve. Hemolytic Transfusion Reactions packed red blood cells and not whole blood A mortality of 100% has been reported in Immediate reactions occur from errors during massive transfusion. Microvascular trauma patients whose body temperature fell involving ABO incompatibility. More than half bleeding and clinical evidence of coagulopathy below 32oC, regardless of severity of injury, of these errors happen after the blood has been can occur in the setting of massive transfusion degree of hypotension, or fluid replacement.31 issued by the blood bank, which highlights the with 1 blood volume and are associated with The importance of fluid warming cannot importance of verifying and identifying each decreased levels of Factor V, VIII, and fibrino- be underestimated in the trauma patient. It and every donor unit for recipient compatibil- gen and increased prothrombin times.14–16 Mi- requires 16 kCal of energy to raise the tem- ity. Intravascular hemolysis occurs when recipi- crovascular bleeding in this case can be treated perature of 1 liter of crystalloid infused at 21oC ent antibody coats and immediately destroys appropriately with fresh frozen plasma. to body temperature and 30 kCal to raise the the transfused red cells. Classic signs of Dilutional thrombocytopenia is a cause of hem- temperature of cold 4oC blood to 37oC. Infu- hemolytic transfusion reaction are masked by orrhagic diathesis after 1.5 to 2.0 blood vol- sion of 4.3 liters of crystalloid at room tem- general anesthesia. The only evidence may be umes have been transfused. This corresponds perature to an anesthetized adult trauma pa- hemoglobinuria, hypotension, and a bleeding to ~15 to 20 units of red cells in an adult tient who cannot increase heat production can diathesis. Treatment is supportive and involves

20 Massive Transfusion and Control of Hemorrhage in the Trauma Patient stopping the transfusion and maintaining sys- transfusion, as well as following trends in vital shock. Crit Care Med 1983; 11:839. temic and renal perfusion. signs, urinary output, central venous pressures, 4. Lam AM, Winn HR, Cullen BF, et al. Hy- and arterial and central venous blood gas analy- perglycemia and neurological outcome in Microaggregates sis, are of vital importance to managing pa- patients with head injury. J Neurosurg Microaggregates begin forming after ap- tients with hemorrhagic shock. 1991; 75:545. proximately 2 days of blood storage. During 5. Michaud LJ, Rivara FP, Longstreth WT, et the first 7 days, microaggregates are mostly References al. Elevated initial blood glucose levels and platelets or platelet debris. After the first week, 1. Stene J, Smith CE, Grande CM. Evaluation poor outcome following severe brain inju- the larger fibrin–white blood cell–platelet ag- of the trauma patient. In Longnecker DE, ries in children. J Trauma 1991; 31:1356. gregates begin to accumulate.36 Whether these Tinker JH, Morgan GE, eds. Principles and 6. Wilson RF. Blood replacement. In Wilson microaggregates contribute to lung dysfunc- Practice of Anesthesiology, 2nd ed. Phila- RF, Walt AJ, eds. Management of Trauma. tion during blood transfusion and whether delphia, Mosby-Yearbook, 1997. Pitfalls and Practice, 2nd ed. Baltimore, they need to be removed by micropore filters 2. Grande CM, Smith CE, Stene J. Anesthe- Williams & Wilkins, 1996. is controversial. sia for trauma. In Longnecker DE, Tinker 7. Bickell WH, Wall MJ, Pepe PE, et al. Imme- JH, Morgan GE, eds. Principles and Prac- diate versus delayed fluid resuscitation for Infection tice of Anesthesiology, 2nd ed. Philadel- hypotensive patients with penetrating torso Hepatitis C accounts for more than 90% phia, Mosby-Yearbook, 1997. injuries. N Engl J Med 1994; 331:1105. of posttransfusion hepatitis. Every year, at least 3. Rackow EC, Falk JL, Fein IA et al. Fluid 8. Messmer K, Sunder-Plassmann L, Jesch F, 2,600 patients develop cirrhosis as a result of resuscitation in circulatory shock: a com- et al. Oxygen supply to the tissues during hepatitis after blood transfusions.37 Each unit parison of the cardiorespiratory effects of limited normovolemic hemodilution. Res of fresh frozen plasma or platelets has the same albumin, hetastarch, and saline solutions Exp Med 1973; 159:152. risk of infection as a unit of packed red cells. in patients with hypovolemic and septic 9. Jan KM, Heldman J, Chien S. Coronary Recent estimates of infectious rates per unit transfused include hepatitis C, 1:103,000; hepatitis B, 1:63,000; HIV, 1:493,000; and HTLV Table 3. I or II, 1:641,000.38 New screening tests using Clinical Strategies to Reduce Complications of Transfusion Therapy nucleic acid/genomic amplification techniques will shorten the window period and reduce Complication Clinical Strategies to Reduce Complication the risk for these viruses even further. The risk per unit for Yersinia, malaria, babesiosis, and Impaired oxygen release from Warm all blood. Avoid alkalosis. Maintain Chagas is estimated at <1:1,000,000. Other hemoglobin normothermia (core temperature 36-37°C) types of infectious diseases such as toxoplas- mosis and cytomegalovirus, Epstein-Barr virus, Dilutional coagulopathy Fresh frozen plasma for PT>1.5 x normal and and bacterial infections may also be transmit- clinically excessive bleeding. Platelets for ted via transfused blood and blood products. thrombocytopenia <75,000/µl and clinically The risk of bacterial contamination per unit of excessive bleeding. random donor platelets is 1:2,500. Hypothermia Warm all IV fluids and blood. Warm room Transfusion-Induced Immunosuppression >28°C. Convective warming. Humidify all (See also Chapters 8 and 9) inspired gases. Blood transfusion therapy is also associated with allosensitization, immunosuppres- Decreased ionized calcium Treat with calcium chloride, 20 mg/kg, in sion, and an increased incidence of postop- setting of massive transfusion and hypotension erative infections.39 These effects may be mediated by reduced lymphocyte function, down- Hyperkalemia Monitor ECG and treat with calcium chloride, regulation of macrophage function, and altered 20 mg/kg, if hemodynamically significant. cytokin production and activity. Strategies to Otherwise, monitor and treat with glucose and reduce the risk of immunomodulation include insulin and/or bicarbonate. the use of third-generation leukocyte filters, lower transfusion trigger, red cell salvage, and Hemolytic transfusion reaction Check and recheck every donor unit. Once 11,40–42 blood substitutes (Table 3). It is antici- occurred, stop transfusion and maintain pated that new devices for autotransfusion, systemic perfusion and renal blood flow. together with the introduction of hemoglobin- Alkalinize urine. Watch for DIC. Send suspected based red cell substitutes, will dramatically al- unit to blood bank for crossmatch. ter the current approach to fluid and blood component therapy in trauma. Infection Lower transfusion trigger. Red cell salvage. Avoid indiscriminate platelet transfusions. Summary Oxygen-carrying red blood cell substitutes. The bleeding trauma patient requires rapid evaluation and treatment to ensure ad- Transfusion-induced Lower transfusion trigger. Red cell salvage, equate tissue perfusion and successful out- immunosuppression oxygen-carrying red blood cell substitutes. come. Resources such as thermally efficient Third-generation leukocyte filters. fluid warmers, effective transfusion services, and rapid availability of coagulation tests are practical aspects of trauma resuscitation that DIC, disseminated intravascular coagulation deserve priority. Preventing hypothermia and recognizing other complications of massive

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 21 hemodynamics and oxygen utilization af- 26. Smith CE, Patel N. Hypothermia in adult sive blood replacement: correlation of ion- ter hematocrit variations in hemorrhage. trauma patients: anesthetic consider- ized calcium, citrate, and hydrogen ion con- Am J Physiol 1980; 239:H326. ations. Part I: Etiology and pathophysiol- centration. Anesth Analg 1979; 58:274. 10. Varat MA, Adolph RJ, Fowler NO. Cardio- ogy. Am J Anesthesiol 1996; 23:283. 35. Insalaco SJ. Massive transfusion. Lab Med vascular effects of anemia. Am Heart J 27. Smith CE, Patel N. Hypothermia in adult 1984; 15:325. 1972; 83:415. trauma patients: anesthetic consider- 36. Arrington P, McNamara JJ. Mechanism of 11. ASA Task Force. Practice guidelines for ations. Part II: Prevention and treatment. microaggregate formation in stored blood. blood component therapy. Anesthesiology Amer J Anesthesiol 1997; 24:29. Ann Surg 1974; 179:146. 1996; 84:732. 28. Reed RL, Johnston TD, Hudson JD, Fischer 37. Zuck TF, Sherwood WC, and Bove JR. A 12. Gervin AS, Fischer RP. Resuscitation of RP. The disparity between hypothermic review of recent events related to surro- trauma patient with type-specific coagulopathy and clotting studies. J gate testing of blood to prevent non-A, uncrossmatched blood. J Trauma 1984; Trauma 1992; 33:465. non-B posttransfusion hepatitis. Transfu- 24:327. 29. Reed RL, Bracey AW, Hudson JD, et al. Hy- sion 1987; 27:203. 13. Miller RD. Transfusion therapy. In Miller pothermia and blood coagulation: disso- 38. Schreiber GB, Busch MP, Kleinman SH, RD, ed. Anesthesia, 4th ed. New York, ciation between enzyme activity and clot- Korelitz JJ. The risk of transfusion-trans- Churchill Livingstone, 1994. ting factor levels. Circ Shock 1990; 32:141. mitted viral infections. The Retrovirus Epi- 14. Murray, DJ, Pennell BJ, Weinstein SL, Olson 30. Valeri CR, MacGregor H, Cassidy G, et al. demiology Donor Study. N Engl J Med JD. Packed red cells in acute blood loss: Effects of temperature on bleeding time and 1996; 334:1685–90. dilutional coagulopathy as a cause of sur- clotting time in normal male and female 39. Landers DF, Hill GE, Wong KC, Fox IJ. gical bleeding. Anesth Analg 1995; 80:336. volunteers. Crit Care Med 1995; 23:698. Blood transfusion-induced immunomo- 15. Murray DJ, Olson J, Strauss R, Tinker JH. 31. Jurkovich GH, Greiser WR, Luterman A et dulation. Anesth Analg 1996; 82:187. Coagulation changes during packed red al. Hypothermia in trauma victims: an 40. Lane TA Leukocyte reduction of cellular cell replacement of major blood loss. An- ominous predictor of survival. J Trauma blood components. Arch Pathol Lab Med esthesiology 1988; 69:839. 1987; 27:1019. 1994; 118:392. 16. Leslie SD, Toy PT. Laboratory hemostatic 32. Gentilello LM, Moujaes S. Treatment of 41. Kevy SV et al. Evaluation of a new abnormalities in massively transfused pa- hypothermia in trauma victims: thermo- atraumatic surgical suction system (ab- tients given red blood cells and crystal- dynamic considerations. J Intensive Care stract). Proceedings of the 10th Annual loid. Am J Clin Pathol 1991; 96:770. Med 1995; 10:5. Trauma Anesthesia and Critical Care Sym- 17. Gross D, Landau EH, Klin B, et al. Quanti- 33. Mendlowitz M. The specific heat of human posium, Baltimore, 1997. tative measurement of bleeding following blood. Science 1948; 107:97. 42. Cohn SM. Is blood obsolete? J Trauma hypertonic saline therapy in “uncon- 34. Kahn RC, Jasco HD, Carlon GC et al. Mas- 1997; 42:730. trolled” hemorrhagic shock. J Trauma 1989; 29:79. 18. Dewitt DS, Prough DS, Deal DD, et al. Hypertonic saline does not improve cerebral oxygen delivery after head injury and 8 Immunomodulatory Effects of Transfusion mild hemorrhage in cats. Crit Care Med 1996; 24:109. David T. Porembka, Do, FCCM, FCCP 18A.Hartl R, Ghajar J, Hochleuthner H, Mauritz Associate Professor of Anesthesia and Surgery Table 1. Risks of Transfusions W. Treatment of refractory intracranial Associate Director of Surgical Intensive Care hypertension in severe traumatic brain University of Cincinnati Medical Center Reactions Frequency: unit injury with repetitive hypertonic/ Cincinnati, Ohio, USA hyperoncotic infusions. Zentrabl Chir Febrile (FNHTR) 1–4:100 1997; 122:181–5. The administration of blood and its com- Allergic 1–4:100 19. Scalea TM, Hartnett RW, Duncan AO, et al. ponents can be life-saving, particularly dur- Delayed hemolytic 1:1,500 Central venous oxygen saturation: a use- ing resuscitation in trauma patients when Acute hemolytic 1:12,000 ful clinical tool in trauma patients. J blood loss can be severe enough to result in Fatal hemolytic 1:100,000 Trauma 1990; 30:1539. cellular hypoxia.1 Also, during other critical Anaphylactic 1:150,000 20. Patel N, Smith CE, Pinchak AC. Clinical illness such as systemic inflammatory re- comparison of blood warmer perfor- sponse syndrome, especially if the patient is Infections mance during simulated clinical condi- septic with significant acute lung injury, blood tions. Can J Anaesth 1995; 42:636. is administered to augment oxygen delivery Hepatitis C 1:103,000 21. Uhl L, Pacini DG, Kruskall MS. The effect to avoid cellular hypoxia and lactate produc- Hepatitis B 1:200,000 of heat on in vitro parameters of red cell tion.2 Even though there are risks following HIV-1 1:490,000 integrity. Transfusion 1993; 33:60S. blood transfusions, the benefits appear to be HIV-2 Unknown 22. Patel N, Knapke D, Smith CE, et al. Simu- insurmountable3 (Table 1). In spite of this, the HTLV-I (II) 1:641,000 lated clinical evaluation of conventional risks of infection, especially from HIV, have Malaria 1:4,000,000 and newer fluid warming devices. Anesth taken center stage even in the lay press. Thus, Analg 1996; 82:517–524. the immunologic effects of transfusion have Miscellaneous 23. Valeri CR, Collins FB. Physiologic effects of not gained the attention deserved. Nonethe- 2,3-DPG-depleted red cells with high affin- less, in certain disciplines—hematology, criti- RBC allosensitization 1:100 ity for oxygen. J Appl Physiol 1971; 31:823. cal care medicine, oncology, surgery, and par- HLA allosensitization 1:10 24. Sohmer PR, Scott RL. Massive transfusion. ticularly transplantation—have appreciated Graft vs. host disease Rare Clin Lab Med 1982; 2:21. the immunologic potential from its use. This 25. Sessler DI. Perianesthetic thermoregula- presentation will discuss the basics of immu- From Dzieczkowski and Anderson.3 tion and heat balance in humans. FASEB J nology, concentrating on the immunologic 1993; 7:638–644. consequences of transfusions, the clinical and

22 Massive Transfusion and Control of Hemorrhage in the Trauma Patient animal studies affecting tumor recurrence, and infection.

T-Cell Recognition and Activation T-cell recognition of an antigen with T-cell activation is key in the initiation of rejection Figure 1 and/or tolerance of foreign tissue. Typically, T Representation of the cells require two signals for activation. The first antigen-presenting cell occurs when an antigen is processed into pep- (APC) interacting with the tides by an antigen-presenting cell (APC) and major histocompatibility loaded into the groove of a major histocom- complex (MHC) molecule. patibility complex (MHC) molecule. The anti- In addition this interaction gen is then presented to the T cell, which is is presented to the T cell, recognized in the context of self-MHC (Fig. 1). which acts with the service The second signal occurs when the T cell re- molecules and APC. ceives stimulation by a cytokine or by the interaction of the T cell with surface molecules of an APC. Numerous cytokines (interleukins, alpha-tumor necrosis factor, and interferon) are involved in this process as well as cell surface receptors, adhesion molecules, and lymphocyte functioning antigen.4–6 Other signifi- However, successful transplantation of kidneys renal transplantation and at the same time, cant cell surface molecules are the CD3 com- from HLA-mismatched donors was not possible Opelz et al, following the success in animal plex and CD45. The former is associated (1963) until the advent of immunosuppressive models, provided evidence (by reviewing trans- noncovalently with the T-cell receptor on ma- agents, prednisone and azathioprine.15,16 The plant data from multiple centers) in humans ture T cells and is a target for OKT3, whereas immunosuppressive agents had to be contin- that blood transfusion prior to renal transplan- the latter does not have a known ligand and ued to ensure “acceptance” of the foreign tis- tation improved renal allograft survival. Com- allows continued activation of the T cell.7 sue or organ. In addition, early in transplanta- pared with patients who did not receive blood Generally, T cells recognize antigens pre- tion, efforts were directed to minimize expo- transfusions, the transfused patients (>5 trans- sented as short peptides that are bound in the sure to or sensitization from transfusions. fusions) had a higher survival rate of the renal MHC groove. Allo-MHC molecules stimulate a However, in 1972 two animal studies chal- allografts, approaching 20%. Interestingly greater response (in vitro mixed lymphocytes lenged that premise. Jenkins et al revealed that though in this study, this effect appeared to response and cytotoxic T-lymphocyte assay) transfusions administered prior to cardiac al- have a dose-response relationship.18,20,21 Even than antigens that are not foreign.8–10 The path- lografting improved survival of transplanted though this seminal publication was retrospec- ways for these alloreactivities are both indirect hearts in rats.17 Separately, Fabre and associ- tive, recently there appears to be more direct and direct.11–13 In the direct pathway of alloan- ates showed that rejection of the transplanted evidence for this response.22 In 1979, Cochrum tigen presentation, the T cells recognize intact kidney in rats can be diminished by and colleagues used pretransplantation-di- donor MHC molecules on the surface of the pretransplant transfusions.18 rected donor-specific whole blood in patients donor APC. This pathway may be responsible Possibly realizing these attributes, Newton with renal failure. In strong mixed lymphocyte for early acute rejection of grafts.12 Early in the and Anderson, in 1973, attempted to manipu- culture-responsive, one haplotype-mis- care of these patients, radiation and other im- late the immune response to renal allografts matched, and living-related donor transplants, mune modulation strategies were used to af- of four patients with donor-specific peripheral directed transfusions improved survival up to fect this pathway directly by removing or de- lymphocyte buffy-coat transfusion from their 90%. This rate is not that different from that in stroying these graft leukocytes.13 The exact potential living related donor over an extended HLA-identical siblings.23,24 Following this suc- mechanism is not known and is, without a time (22 to 66 days). Allosensitization did not cess, there was equivalent survival in patients doubt, multifactorial. In the indirect pathway, ensue. Critics believed that the addition of aza- with two haplotype-mismatched, related and T cells recognize processed donor allo-MHC thioprine contributed to the allograft’s success unrelated donor–recipient combinations.25,27 bound to and presented in the context of self- rather than the administration of blood.19 Sub- From these studies and others, the presence MHC molecules on the surface of self-APC. This sequent to this new era of cadaveric donor of leukocytes and one shared HLA-DR antigen pathway is normally associated with a nomi- within the transfusions are sufficient enough nal antigen. for optimal immunosuppression.20,28 Overall, Table 2. there is sufficient evidence documenting that History of Donor-Specific Transfusion Possible Mechanisms of transfusions prior to solid organ transplanta- As early as 1953, Billingham and associ- Transfusion-Associated tion improves survival and reduces the inci- ates demonstrated white blood cells as im- Immunomodulation dence of rejection. mune modulators when neonatal mice of one Although the precise mechanisms in- strain injected with blood from another sub- Anergy volved in tolerance and sensitization are not sequently accepted skin grafts from the immu- Tolerance completely understood, the laboratory findings nizing strain. This effect was long term only in Cytokines released during blood storage have been consistent29 (Table 2). Generally, the neonatal mice, not in the adults.14 The first Iron-mediated immune suppression blood transfusions induce predictable immune solid organ transplantation (kidney) was per- Suppressor cell network inhibition responses stimulating alloantibody production formed in 1954 between monozygotic twins. Anti-idiotypic and anti-clonotypic anti- when exposed to red cell, white cell, and plate- The success was probably related to matching bodies let alloantigens.30–33 Investigations have shown of the ABO blood type with compatibility of Clonal deletion the development of Fc receptor-blocking fac- the (MHC) antigens, not from immune sup- tors, lymphocyte activation, lymphocyte sub- pression. (The complexity of the immune sys- From Brennan et al.29 population changes, and down regulation of tem was not well understood during this era.) APC after transfusion34 (Table 3). These results

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 23 should be reviewed skeptically. To counteract survival rate was 34% with blood and 53% with- Table 3. or attempt to explain the association with tu- out blood. An apparent criticism (not minimiz- Immunologic Laboratory mor recurrence and blood, a meta-analysis was ing a retrospective analysis) was that 61% of Tests in Transfused Patients performed, reviewing retrospective studies of the transfused patients had femoral tumors colorectal cancer patients. In this statistical while the nontransfused group included only Decreased lymphocyte response to review, this association was not confirmed.40 50%.44 mitogens/alloantigens This study was in contradistinction to another In animal studies of tumor augmentation, Down-regulation of natural killer, meta-analysis review, in which there was a the data are provocative but still suggestive of cytotoxic T lymphocytes higher recurrence rate in patients with transfusions as a factor. One important issue Decreased IL-2 production colorectal and head and neck cancer.30 addressed in these animal models is the re- Decreased CD4 cells In a prospective, nonrandomized study, moval of leukocytes and its timing. In athymic Increased CD7 cells 315 consecutive patients with prostatic cancer mice transfused with either allogeneic or syn- Decreased natural killer cell number who underwent radical retropubic prostatec- geneic blood or saline prior to tumor cell in- Increased B cells tomy were analyzed. Group 1 received at least fusion, the subsequent tumor size was of equal Macrophage function one unit of allogeneic blood with or without dimensions. However, in immunocompetent Decreased migration to autologous blood; group 2 received autolo- mice, there were larger and heavier tumors chemotactic stimuli gous blood only or no blood. These patients after transfusion with allogeneic blood.45 Cor- Decreased deposition in received no adjuvant hormonal therapy or ra- respondingly, rats transfused with allogeneic inflammatory foci diotherapy. The incidence of reoccurrence was or syngeneic blood stored for 1 day had higher Increased macrophage prostaglandin similar: 25% vs. 23%, respectively. In addition, rates of tumor growth and shorter survival E2 production mortality was not affected by the administra- times than controls with saline infusion.46 Con- Decreased antigen-presenting cell activity tion of blood.41 trary to these studies, Shirwadkar et al gave Decreased delayed-type hypersensitivity In patients with high-grade soft-tissue sar- mice various doses of tumor cells with the comas of the extremities and osteosarcomas transfusions and concluded that the From Blumberg and Heal.34 of long bones, there is a suggestion that trans- immunomodulatory effect of transfusion is fusion can alter outcome.42–44 In Rosenberg’s solely dependent on the dose of the inoculated study of patients with soft-tissue tumors who cells.47 are not reproduced in the neonate as com- underwent various prospective, randomized In addressing the issue of leukocyte deple- pared with the adult immune system. Neonates treatment protocols, the patients without any tion, Blajchman and colleagues preempted 10 who received washed and irradiated blood transfusions had a 70% actuarial 5-year disease- days before the infusion of tumors cells either failed to exhibit similar effects seen in adult free survival rate while patients who received leukocyte-reduced or nonleukocyte-reduced recipients.35 1 to 3 units of blood had a 48% rate. The over- blood. The pulmonary metastatic nodules were all 5-year survival rates were 85% and 63%, assessed 3 weeks later. The recipients of allo- Tumor Recurrence respectively. As expected, tumor size correlated geneic transfusion had two- to five-fold in- There appears to be a beneficial effect of inversely with outcome, but after this was taken creases in these nodules compared with the blood transfusions on the immune systems in into consideration, the effect of transfusion still animals receiving either leukocyte-reduced al- solid-organ transplantation, but there is a was a negative prognostic indicator.42 In a re- logeneic or syngeneic blood.48 In an acute ex- down side—the reemergence of cancer cells lated study that focused primarily on the periment (tumor cells injected within 60 to 90 in patients with neoplastic disease. However, cardiotoxicity of doxorubicin in patients with minutes of transfusion), pulmonary metastatic the results are divergent, ranging from stimu- high-grade soft-tissue sarcoma, factors that nodules were greater (four- to seven-fold) in lation of tumor growth, suppression of growth, were associated with distal metastases included the group with allogeneic blood. In this inves- and varied to no response to the tumor cells. blood transfusion within 24 hours, tumors >5 tigation, the authors believed that the removal In 1982, Burrows and colleagues retrospec- cm, tumors extending into the deep fascia, and of allogeneic leukocytes ameliorated the tumor tively reviewed 122 patients following other histologic subtypes.43 Similar correlation growth potential.48 Consequently, these same colorectal surgery. They found a shorter dis- was seen between survival and transfusions in investigators found that removal of leukocytes ease-free interval (6 to 12 months) in patents patients with nonmetastatic osteosarcoma of following storage did not have similar extent who received a transfusion.36 A similar investi- long bones. In this retrospective study, the of amelioration.49 gation detected contrasting results and revealed that 43% of transfused patients developed recurrent disease or died, compared with 9% who did not receive a transfusion.37 How- Table 4. ever, a large multicenter, randomized, con- Multivariate Analysis of Factors Related to Disease-Free Survival in Patients trolled study of colorectal patients (n=475) Undergoing Colorectal Curative Surgery with cancer found no direct relationship between allogeneic transfusion and prognosis Factors Relative 95% Confidence p (cancer-free survival rates after 4 years were Recurrence Rate Interval no different between the two groups), but the data did suggest an increased in recurrence Transfusion group no matter if the blood was allogeneic or au- Allogeneic 1 — — tologous38 (Table 4). These studies did not fil- Autologous 1.1 0.7–1.6 0.74 ter the white cells from the blood components. Dukes’ classification In an investigation that did filter, the results A1—— were confusing because a number of patients B 4.0 1.7–9.5 0.002 received both types of blood products. This C 10.8 4.7–25.1 <0.001 study did not reach any conclusions.39 Unfor- tunately, retrospective studies have inherent From Busch et al.38 flaws and conflicting conclusions. These results

24 Massive Transfusion and Control of Hemorrhage in the Trauma Patient Tumor Recurrence and Infection mice to burn injury and infused Escherichia immunomodulatory effect on the recipient. Since there appears to be an coli. They found that enhanced gut permeabil- The leukocytes appear to be the culprit. The immunomodulatory effect of transfusions, the ity and bacterial challenge responded syner- reason why removing white cells prior to stor- question arises, particularly in regard to pa- gistically in secondary infection.54 A follow-up age to minimizes complications (infections, tients with cancer, is there a higher rate of in- investigation showed an association between recurrence of tumor) is not understood. One fection? In reviewing the data in Heiss’s series, allogeneic leukocytes and an adverse effect on thing certain is that the extent of transfusions the postoperative infection rate was higher in host bacterial mechanisms.55 correlates with these secondary problems, but the allogeneic group (27%) compared with the Clinically, the results are more confusing. in patients who receive a greater number of autologous group (12%). Multivariant regres- A retrospective analysis of orthopedic patients blood products, what is the predominant in- sion analysis revealed that infection was related revealed that allogeneic transfusions are asso- determinate factor: the underlying disease to transfusion, with an odds ratio of 2.84. Seg- ciated with increased frequency of postopera- process, the patient’s co-morbidity factors, or menting the groups revealed that the infection tive infections, including pneumonia and uri- the aggressiveness of surgical eradication? rate also increased with a greater certainty with nary tract infection.56 Other reports (one ret- No large clinical trials of transfusion in allogeneic blood compared with the autolo- rospective and one prospective) showed an trauma patients (who tend to be young and gous group.50 In a large prospective study of association.57,58 In a study of patients under- not have complicating medical diseases) have colorectal patients (n=871), patients were ran- going total hip replacement, Murphy and as- been undertaken to determine the incidence domly assigned either leukocyte-filtered blood sociates found a proven or suspected infection of infection when leukocytes are removed prior (<0.2 x 109 leukocytes) or blood without a in 32% of their select group, e.g., patients with- to storage. In the author’s opinion, one group buffy coat (0.8 x 109 leukocytes per unit). At 3- out prior infections, malignancy, and infusion will benefit, and that group includes patients year follow-up, there was no statistical differ- of <3 units of blood with allogeneic infusions. transfused with <10 units of blood. However, ence in the infection rate. It is interesting to This is in comparison to 3% of patients given this investigation must be initiated upon ar- note that in this study a certain number (>3) autologous blood. The hospital stay was con- rival to the definitive area and it must be of transfusions was a marker or independent siderably longer in patients with suspected in- blinded and prospective. Should the standard risk factor for survival as well, similar to tumor fection.57 In a nonrandomized, prospective trial, of blood banking include prestorage filtering location or size. This correlated with the inci- Triulzi and colleagues detected infection in of all blood? Economically, it would be feasible dence of infection in the curative surgery pa- 20.8% of the allogeneic blood recipients com- to filter the blood when the potential risk of tients.51 Even though statistical analysis se- pared with 4% of the nontransfused individu- infection and cancer is there. The precedent lected certain factors, such as >3 units of blood als. Apparently, the amount of units given cor- for accepting increased cost without clearly (which had greater postoperative morbidity related with the infection rate.58 In another ret- demonstrated benefit has already been set by and mortality), was this associated with the rospective review in patients undergoing either the much greater costs involved in the preven- more complex patient with extensive disease hip, spinal, or knee surgery, the data between tion of transmission of the AIDS virus by p24 and technically difficult surgery? the groups were not conclusive.59 Howard and antigen testing in blood banking64 (Table 5). Vamvakas corroborated these inconclusive find- Infection ings.60,61 In a different population (cardiac sur- References Similar controversy surrounds the associa- gery patients), the results are more revealing. A 1. Gutierrez G. Cellular metabolism during tion between blood transfused and the inci- multivariate analysis demonstrated contributing hypoxia. Crit Care Med 1991; 19:619. dence of postoperative infection. Animal mod- factors for postoperative infection as 2. Shoemaker WC, Appel PL, Bishop MH. els suggest that allogeneic transfusion in- reoperation, blood transfusions, early chest Temporal patterns of blood volume, he- creases the appearance of infection.52 In trau- reexploration, and sternal rewiring. The diffi- modynamics, and oxygen transport in matic burn or induced peritonitis experimen- culty with this statistical review is that one would pathogenesis and therapy of postopera- tal models, animals had shorter survival with expect the infection rate to be higher in emer- tive adult respiratory distress syndrome. allogeneic transfusions than the groups receiv- gency reoperations. Controlling clinical factors New Horizons 1993; 1:522. ing either crystalloid or syngeneic blood.30,31,33 would be almost impossible.62,63 3. Dzieczkowski JS, Anderson KC. Transfusion Interestingly, contrasting this traumatic model, and therapy. In Fauci AS et al, eds. Brunson did not induce injury and only in- Conclusions Harrison’s Principles of Internal Medicine, fused blood. They found that the addition of The weight of scientific evidence from 14th ed, New York, McGraw-Hill, 1998. trauma, not the dose of blood, altered the im- both basic science and clinical studies sug- 4. Bretscher P, Cohn M. A theory of self- mune system, suggesting trauma alone ablated gests that allogeneic transfusions have a sig- nonself discrimination. Science the immune response toward infection.53 To nificant but varied effect on the immune sys- 1979;169:1042–9. challenge this premise, Gianotti et al subjected tem. There is no doubt there is dynamic 5. Lafferty KJH, Cunningham A. A new analy-

Table 5. Allogeneic Transfusion Immunomodulation-Related Postoperative Infection and Cancer Recurrence: Theoretic Estimates of U.S. Mortality Rates

Infection Cancer Recurrence Total Death Rate Death Rate Death Rate Estimated % Deaths per Million Deaths per Million Deaths per Million Causal Contribution per Year Transfusions per Year Transfusions per Year Transfusions

100 1,500 250 20,000 33,000 21,500 33,250 50 750 125 10,000 16,667 10,750 16,892 10 150 25 2,000 3,300 2,150 3,325 1 15 2.5 200 330 215 332.5

From Blumberg.64

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 25 sis of allogenic interactions. Aust J Exp Biol 22. Quigley RL, Wood KJ, Morris PJ. Investi- Papendrecht MA, et al. Blood transfusions Med Sci 1975;53:27–42. gation of the mechanism of active en- and prognosis in colorectal cancer. N Engl 6. Nickerson PW, Steurer W, Steiger J, Strom hancement of renal allograft survival by J Med 1993; 328:1372–6. TB. In pursuit of the “Holy Grail”: allograft blood transfusion. Immunology 1988; 39. Heiss MM, Mempel W, Delanoff C, et al. tolerance. Kidney Int 1994; 45(suppl 63:373–81. Blood transfusion-modulated tumor re- 44):S40-9. 23. Salvatierra O, Vincenti F, Amend W. Delib- currence: First results of a randomized 7. Donovan JA, Koretzky GA.CD45 and the erate donor specific blood transfusions study of autologous versus allogeneic immune response. J Am Soc Nephrol 1993; prior to living renal transplantation: a new blood transfusion in colorectal cancer 4:976-85. approach. Ann Surg 1980; 192:543–52. surger. J Clin Oncol 1994; 12:1859–67. 8. Bach FH, Graupner EE, Klostermann H. Cell 24. Potter D, Garovoy M, Hopper S, Terasaki 40. Vamrakis E, Moore SB. Perioperative blood kinetic studies in mixed leukocyte cultures: P, Salvatierra O Jr. Effect of donor-specific transfusion and colorectal cancer recur- an in vitro model of homograf reactivity. Proc transfusions on renal transplantation in rence: A qualitative statistical overview and Natl Acad Sci USA 1969; 63:377–84. children. Pediatrics 1985; 76:402–5. meta-analysis. Transfusion 1993; 33:754–65. 9. Fischer-Lindahl K, Wilson DB. Histocom- 25. Anderson CB, Jendrisak MD, Flye MW, 41. Ness PM, Walsh PC, Zahurak M, et al. Pros- patibility antigens-activated cytotoxic T Hanto DW, Anderman CK, Rodey GE, tate cancer recurrence in radical surgery lymphocytes. II. Estimates of the fre- Sicard GE. Renal allograft recipient patients receiving autologous or homolo- quency and specificity of precursors. J Exp immunomodulation by concomitant im- gous blood. Transfusion 1992; 32:31–6. Med 1977; 145:508–22. munosuppression and donor-specific 42. Rosenberg SA, Seipp CA, White DE, et al. 10. Widmer MB, Donald HRM. Cytolytic T lym- transfusions. Transplant Proc 1988; Perioperative blood transfusions are asso- phocyte precursors reactive against mu- 20:1074–8. ciated with increased rates of recurrence tant Kb alloantigens are as frequent as 26. Sollinger HW, Kalayoglu M, Belzer FO. Use and decreased survival in patients with those reactive against a whole foreign hap- of the donor specific transfusion protocol high-grade soft-tissue sarcomas of the ex- lotype. J Immunol 1980; 127:48–51. in living-unrelated donor recipient combi- tremities. J Clin Oncol 1985; 3:698–709. 11. Lechler RI, Batchelor JR. Restoration of nations. Ann Surg 1986; 204:315–21. 43. Casper ES, Gaynor JJ, Hajdu SI, et al. Pro- immunogenicity to passenger cell-de- 27. Belzer FO, Kalayoglu M, Sollinger HW. spective randomized trial of adjuvant che- pleted kidney allografts by the addition of Donor-specific transfusion in living-unre- motherapy with bolus versus continuous donor strain dendritic cell. J Exp Med lated renal donor-recipient combinations. exclusion doxorubicin in patients with 1982; 155:31–41. Transplant Proc 1987; 19:1514–5. high-grade extremity soft tissue sarcoma 12. Shoskes A, Wood KJ. Indirect presentation 28. Van Twuyver E, Mooijaart RJD, ten Berge and an analysis of prognostic factors. Can- of MHC antigens in transplanation. IJM, et al. Pretransplantation blood trans- cer 1991; 68:1221–9. Immunol Today 1994; 15:32. fusion revisited. N Engl J Med 1991; 44. Chesi R, Cazzola A, Bacci G, et al. Effect of 13. Sayegh M, Watschinger B, Carpenter CB. 325:1210–3. perioperative transfusions on survival in Mechanisms of T cell recognition of al- 29. Brennan DC, Mohanakumar T, Flye MW. osteosarcoma treated by mutlimodal loantigen: the role of peptides. In-depth review donor-specific transfusion therapy. Cancer 1989; 64:1727–37. Transplanation 1994; 57:1295–302. and donor bone marrow infusion in re- 45. Francis DMA, Burren CP, Clunie GJA. Ac- 4. Billingham RE, Brent L, Medawar PB. Ac- nal transplantation tolerance: a review of celeration of B16 melanoma growth in tively acquired tolerance of foreign cells. efficacy and mechanisms. Am J Kidney Dis mice after blood transfusion. Surgery Nature 1953; 172:603–6. 1995; 26:701–15. 1987; 102:485–92. 15. Calne RY. The rejection of renal 30. Blumberg N, Heal JM. Effects of transfu- 46. Waymack JP, Chance WT. Effect of blood homografts inhibition in dogs with BM6- sion in immune function. Arch Pathol Lab transfusions on immune function: IV. Ef- mercaptopurine. Lancet 1960; 1:417–8. Med 1994; 118:371–9. fect on tumor growth. J Surg Oncol 1988; 6. Zukoski C, Lee HM, Hume DM. The pro- 3. Vamvakas EC, Moore SB. Blood transfu- 39:159–64. longation of functional survival of canine sion and postoperative septic complica- 47. Shirwadkar S, Blajchman MA, Frame B, et renal homografts with BM6 mercaptopu- tions. Transfusion 1994; 34:714–27. al. Effect of blood transfusions on experi- rine. Surg Forum 1960; 11:470–2. 32. Klein HG. Wolf in wolf’s clothing: is it time mental pulmonary metastases in mice. 17. Jenkins A. McL., Woodruff MFA. The effect to raise the bounty on the passenger leu- Transfusion 1990; 30:188–90. of prior administration of donor strain kocyte? Blood 1992; 80:1865–8. 48. Blajchman MA, Bardossy L, Carmen R, et blood or blood constituents on the sur- 33. Bordin JO, Heddle NM, Blajchman MA. al. Allogeneic blood transfusion-induced vival of cardia allografts in rats. Transplan- Review: biologic effects of leukocytes enhancement of tumor growth: two ani- tation 1972; 12:57–60. present in transfused cellular blood prod- mal models showing amelioration by 8. Fabre JW, Morris PJ. The effect of donor uct. Blood 1994; 84:1703–21. leukodepletion and passive transfer using strain blood pretreatment on renal al- 34. Blumberg N, Heal JM. Transfusion and spleen cells. Blood 1993; 81:1880–2. lograft rejection in rats. Transplantation recipient immune function. Arch Pathol 49. Bordin JO, Bardossy L, Blajchman MA. 1972; 14:608–17. Lab Med 1989; 113:246–53. Growth enhancement of established tu- 19. Newton WT, Anderson CB. Planned 35. Zinkernagel RM, Doherty PC. Restriction mors by allogeneic blood transfusion in preimmunization of renal allograft recipi- of in vitro T cells mediated cytotoxicity in experimental animals and its amelioration ents. Surgery 1973; 74:430–6. lymphocytic choriomeningitis within a by leukodepletion: the importance of the 20. Lagaaij EL, Hennemann PH, Ruigrok M, syngeneic or semiallogeneic system. Na- timing of the leukodepletion. Blood 1994; et al. Effect of one-HLA-DR-antigen- ture 1974; 248:701–2. 84:344–8. matched and completely HLA-DR-mis- 36. Burrows L, Tartter P. Effects of blood trans- 50. Heiss MM, Mempel W. Jauch KW, et al. matched blood transfusions on survival of fusions on colonic malignancy recurrence Beneficial effect of autologous blood heart and kidney allografts. N Engl J Med rate (letter). Lancet 1982 ;2:662. transfusion on infectious complications 1989; 321:701–5. 37. Blumberg N, Agarwal MM, Chuang C. Re- after colorectal cancer surgery. Lancet 21. Opelz G, Sengar DP, Mickey MR, et al? Ef- lation between recurrence of cancer of the 1993; 342:1328–33. fect of blood transfusions on subsequent colon and blood transfusion. Br Med J 51. Houbiers JGA, Brand A, van de Watering kidney transplants. Transplant Proc 1973; 1985; 290:1037–9. LMG, et al. Randomised controlled trial 5:253–9. 38. Busch OR, Hop WC. Hoynck van comparing transfusion of leucocyte-de-

26 Massive Transfusion and Control of Hemorrhage in the Trauma Patient pleted or buffy-coat-depleted blood in 33:458–65. Blood transfusion and septic complica- surgery for colorectal cancer. Lancet 1994; 56. Blumberg N, Heal JM. Transfusion and host tions after hip replacement surgery. Trans- 344:573–8. defenses against cancer recurrence and infusion 1995; 35:15–6. 52. Waymack JP, Warden GD, Alexander JW, et fection. Transfusion 1988; 29:236–45. 61. Howard HL, Rushambuza FG, Martlew VJ, al. Effect of blood transfusion and anes- 57. Murphy P, Heal JM, Blumberg N. Infection et al. Clinical benefits of autologous blood thesia on resistance to bacterial peritoni- or suspected infection after hip replace- transfusion: an objective assessment. Clin tis. J Surg Res 1987; 42:528–35. ment surgery with autologous or homolo- Lab Haematol 1993;15:165–71. 53. Brunson ME, Ing R, Tchervenkov JL, et al. gous blood transfusions. Transfusion 62. Vamvakas EC, Carven JH, Hibberd PL. Variable infection risk following allogeneic 1991; 31:212–7. Blood transfusion and infection after blood transfusions. J Surg Res 1990; 58. Triulzi DJ, Vanek K, Ryan DH, et al. A clini- colorectal cancer surgery. Transfusion 48:308–12. cal and immunologic study of blood trans- 1996; 36:1000–8. 54. Gianotti L, Pyles T, Alexander JW, et al. Im- fusion and postoperative bacterial infec- 63. Ottino G, De Paulis R, Pansini S, et al. Major pact of blood transfusion and burn injury tion in spinal surgery. Transfusion 1992; sternal wound infection after open-heart on microbial translocation and bacterial 32:517–24. surgery: A multivariate analysis of risk fac- survival. Transfusion 1992; 32:312–7. 59. Fernandez MC, Gottlieb M, Menitove JE. tors in 2,579 consecutive operative proce- 55. Gianotti L, Pyles T, Alexander JW, et al. Blood transfusion and postoperative in- dures. Ann Thorac Surg 1987; 44:173–9. Identification of the blood component re- fection in orthopedic patients. Transfu- 64. Blumberg N. Allogeneic transfusion and sponsible for increased susceptibility to sion 1992; 32:318–22. infection: economic and clinical implica- gut-derived infection. Transfusion 1993; 60. Vamvakas EC, Moore SB, Cabanela M. tions. Sem Hematol 1997; 34:34.

9 from 182 “seronegative” donors. This resulted with colorectal cancer and head and neck can- Blood Transfusions from the long period necessary to develop de- cer.3 Osteosarcoma patients who receive tectable levels of antibodies to HIV. Current perioperative blood transfusions have an in- Andrew D. Rosenberg, MD antibody testing has diminished the window to creased incidence of metastases and shorter Department of Anesthesiology 22 days. In March 1996, the U.S. Food and Drug survival time.2 Hospital for Joint Diseases/Orthopaedic Institute Administration mandated P24 antigen testing, An altered immunologic state results from New York, New York, USA which decreased the window period to 16 days. receiving a blood transfusion. Allogeneic blood In addition to HIV transmission, hepatitis transfusions have been associated with de- The trauma patient frequently requires B, hepatitis C, and CMV can be transmitted creases in cell-mediated immunity, macroph- multiple blood transfusions during resuscita- easily if blood is not tested adequately. CMV is age migration, and natural killer cell activity. tion to achieve a stable hemodynamic state. frequently present in transfused blood, its Additionally, allogeneic transfusion affects the Adequate oxygen-carrying capacity necessitates prevalence determined by geographic location. cells that incite B-lymphocytes to differentiate transfusion based on the patient’s pathophysi- Special care must be taken in the immunosup- and produce antibodies. These immunosup- ology after being injured, the patient’s baseline pressed patient to ensure that CMV is not pressive effects are thought to be the result of medical condition, and actual and anticipated present in transfused blood. Bacterial and para- either antigen excess, a graph-versus-host phe- blood loss. sitic infections can also be transmitted. Other nomenon, reactivation of immunosuppressive Transfusion is often necessary, but it is not complications known to occur with transfu- viruses, or the white blood cells that are always benign. To even consider the concept sions include allergic reactions, hemolytic transfuseed along with red blood cells.2 of decreasing the amount of blood transfused transfusion reactions, and volume overload. Allogeneic blood transfusions have also to trauma patients, we must determine Transfusions may also result in immuno- been implicated in postoperative infections. whether we can accomplish this goal without suppression or immunomodulation of the re- Independently, Murphy and Triulzi, in sepa- affecting outcome. Obviously, many patients cipient. Studies have demonstrated that renal rate studies on orthopaedic patients, demon- would die without transfusion. Although blood transplant patients had improved allograft sur- strated the effect of allogeneic blood transfu- transfusions increase oxygen-carrying capacity, vival times and lower allograft rejection rates if sions in producing postoperative infection.4,5 massive transfusion is associated with physi- they received transfusions of bank blood (allo- A significant increase in postoperative infec- ologic alterations, immunomodulation, and geneic blood) prior to receiving their allograft.1 tion rates occurred in patients who received postoperative infection. Two questions have During the 1970s, some protocols required allogeneic blood transfusions during either become important in transfusion medicine: 1) patients receiving cadaveric renal transplants to total hip or spine surgery compared with pa- What is in the blood? and 2) What are the sys- receive transfusions prior to the transplant pro- tients who did not receive allogeneic blood. temic effects of transfusion other than increas- cedure. Transfusion of whole blood was a stron- Of patients who received allogeneic transfu- ing the hematocrit? ger enhancer of allograft survival than packed sions, there was an infection rate of 20.8% in a Despite safeguards and tests to ensure that red blood cells. The prevalent thought was that study of 102 patients undergoing 109 spinal blood is not contaminated, blood is being re- transfusion induced an immunosuppressive fusions. The infection rate was only 3.5% in leased that is in fact contaminated. Transfusion effect in the patient and thus, after the patient those who did not receive allogeneic blood. of tainted blood can transmit the human im- received the transplant, rejection did not oc- Natural killer cell activity, an indicator of im- munodeficiency virus (HIV), hepatitis, cytome- cur. Fortunately, the need for preoperative al- munologic function, decreased in the patients galovirus (CMV) and syphilis. Testing for HIV logeneic transfusion has been mitigated by the who received allogeneic transfusion. A specific has become increasingly accurate, so the win- introduction of cyclosporin.2 dose-response curve demonstrated that pa- dow period for possible infection has been In addition to evidence that allogeneic tients who received two transfusions had a shortened because of earlier dectection. The transfusions result in immunosuppression, higher infection rate than patients who re- window period for HIV is that time in which a there is evidence that cancer patients who received either one or no transfusion at all.4,5 person is infected with the virus but has not ceive these transfusions at the time of surgery Fernandez demonstrated that patients who yet demonstrated infectivity by available test- have lower survival rates and an increased in- received homologous whole blood had a ing methods. In a study conducted a number cidence of recurrence.2 Meta-analysis has dem- higher incidence (20%) of infection compared of years ago, 39 patients became seropositive onstrated this finding to be true in patients with the overall (6.1%) infection rate for all

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 27 patients in the study.6 Some orthopaedic stud- fusion? In order to tolerate low hemoglobin, 299:798–803. ies do not demonstrate an association between patients must be able to compensate for the 2. Landers DF, Hill GE, Wong KC, Fox IJ. Blood allogeneic transfusion and infection. In a meta- decreased oxygen-carrying capacity associated transfusion-induced immunomodulation. analysis, Vamvakkas et al were unable to dem- with decreased concentrations of red blood Anesth Analg 1996; 82:187–204. onstrate a clear relationship between transfu- cells. Healthy patients can frequently compen- 3. Vamvakas EC, Moore SB, Cabanela M. sion and infection. Their study criteria, how- sate, but this ability becomes compromised Blood transfusion and septic complica- ever, defined a significant relationship occur- with age and cardiac and respiratory disease. tions after hip replacement surgery. Trans- ring between transfusion and infection as one Increases in cardiac output must be sufficient fusion 1995; 35:150–6. that would result in an infection rate more than to overcome existing deficits. Since oxygen 4. Triulzi DJ, Vanek K, Ryan DH, Blumberg double the baseline occurrence rate.3 delivery depends on cardiac output and arte- N. A clinical and immunologic study of There is significant evidence that transfu- rial oxygen concentration, in addition to sup- blood transfusions and postoperative bac- sions are associated with immunomodulation plying enhanced oxygen concentration, the terial infection in spine surgery. Transfu- and increased infection in trauma patients.7–11 patient must be able to increase stroke volume sion 1992; 32:517–24. Rosemurgy demonstrated an increased inci- and heart rate. The trauma patient is faced with 5. Murphy P, Heal JM, Blumberg N: Infection dence in postoperative infection in a popula- acute decreases in hemoglobin levels and not or suspected infection after hip replace- tion of 390 uncrossmatched trauma patients afforded the ability to compensate, as do pa- ment surgery with autologous or homolo- who received type O blood. In the 61% of patients with chronic anemia. Once volume sta- gous blood transfusions. Transfusion tients who survived at least 7 days, the infec- tus is repleted, hemoglobin (Hb) levels must 1991; 31:212–7. tion rate was higher in those who received be evaluated to determine the need for trans- 6. Fernandez MC, Gottlieb M, Menitove JE. seven or more units of packed red blood cells.7 fusion. Most patients require transfusion when Blood transfusion and postoperative in- Dellinger noted that, while wound infections the Hb is less than 6 gm/dl and few require it fection in orthopedic patients. Transfu- after open fractures of the arm or leg were af- when the Hb is more than 10 gm/dl. Transfu- sion 1992; 32:318. fected by local factors, nosocomial infections sion in the intermediate area requires consid- 7. Rosemurgy AS, Hart ME, Murphy CG, et al. were related to Injury Severity Score (ISS), the eration of physiologic status and the Infection after injury associated with blood incidence of blood transfusion, patient age, individual’s ability to ensure adequate oxygen- transfusion. Am Surg 1992; 2:104–7. and the mode of injury. Edna and Bjerkeset, ation to vital organs. 8. Phillips TF, Soulier G, Wilson. Outcome in a Norwegian study of 484 trauma patients of massive transfusion exceeding two who survived longer than 2 days, demonstrated Conclusion blood volumes in trauma and emergency a 9.5% infection rate, with a univariate rela- Many trauma patients require blood trans- surgery. J Trauma 1987; 27:903–10. tionship between infection and transfusion.10 fusions to replenish massive blood loss from 9. Wudel JH, Morris JA, Yates K, et al. Mas- This relationship was independent of ISS, age, wounds. The advantages of predonation and sive transfusion: outcome in blunt trauma and surgical procedure. The risk of infection cell salvage techniques are not present under patients. J Trauma 1991; 31:1–7. after colon injury is associated with blood emergency conditions or are inappropriate 10. Edna TH, Bjerkeset T. Association between transfusion, age, and the number of associated based on the type of injury. Currently, this leaves blood transfusion and infection in injured injuries and splenic injury. Agarwal, in a study banked blood as the source of blood for trans- patients. J Trauma 1992; 33:659–61. of 5,366 consecutive trauma patients, fusion. The advantages afforded by administer- 11. Agarwal N, Murphy J, Cayten L, et al. Blood documentrf that blood transfusion was a pre- ing allogeneic blood to enhance oxygen-carry- transfusion increases the risk of infection dictor of infection after controlling for patient’s ing capacity must be weighed against its adverse after trauma. Ann Surg 1993; 128:171–7. age, sex, mechanism, or severity of injury.11 side effects, which include immunomodulation, 12. Koval KJ, Rosenberg AD, Zuckerman JD, In a study of 619 geriatric patients with transmission of infectious diseases, and the et al. Does blood transfusion increase the hip fracture, a study at the author’s institution possibility of a transfusion reaction. risk of infection after hip fracture? J documented a significantly higher incidence Trauma 1997; 11:260–6. of urinary tract infections in patients who re- References 13. Riska EB, Bostman O, von Bonsdorff H, ceived allogeneic transfusion compared with 1. Opelz G, Terasaki PI. Improvement of kid- et al. Outcome of closed injuries exceed- those who did not require any transfusion.12 ney-graft survival with increased numbers ing 20-unit blood transfusion need. Injury Riska demonstrated a linear relationship of blood transfusions. N Engl J Med 1976; 1988; 19:273–6. between the number of units transfused and mortality, with 21 to 39 units being associated with a 25% mortality and more than 40 units associated with a 52% mortality.13 Wudel 10 documented 5 survivors of more than 50 units Vascular Access in Trauma: Options, of blood after massive transfusion.9 Blunt and penetrating trauma patients receiving mul- Risks, Benefits, Complications tiple transfusions had similar survival rates (59%). Shock, closed head injury, and age Maureen Nash Sweeney, MD In the trauma patient presenting with predicted increased mortality but did not pre- Attending Anesthesiologist multiple serious injuries and hemorrhagic clude survival. Anesthesiology Department shock, vascular access is necessary to restore Massive transfusion may be associated Department of Veterans Affairs Medical Center circulatory volume rapidly. The urgency of the with high citrate and acid load, possible he- New York, New York, USA placement and the size and number of intra- mostatic failure, disseminated intravascular venous (IV) lines is dictated by the degree of coagulation, large amounts of infused blood Vascular access in the trauma patient is shock, the apparent rate of bleeding, and the debris, inadequate 2,3-DPG levels, and throm- essential for three reasons: type of injury. Advanced Trauma Life Support bocytopenia. Thus, although multiple transfu- • administration of intravenous fluids (ATLS™) protocol recommends proceeding sion is indicated under many conditions, we • administration of drugs with attempts at percutaneous peripheral ac- need to consider what are appropriate trans- • measurement and monitoring of cardiac cess, followed by a surgical venous cutdown fusion triggers. What factors are considered parameters before resorting to central venous access. The important in determining the need for trans- rationale is that, in a hypovolemic patient, the

28 Massive Transfusion and Control of Hemorrhage in the Trauma Patient likelihood of success with a venous cutdown placement of a large-bore IV should be veri- endocardium of the atrium or ventricle. Proper is greater than with a central line. Addition- fied by checking for backflow. An IV site should positioning of the catheter in the superior vena ally, the rate of complications (e.g., pneu- infuse easily without added pressure. Intrave- cava (SVC) usually abolishes this problem. mothorax and arterial puncture) is higher with nous fluids can extravasate into soft tissues Myocardial perforation and tamponade rarely central IV access.1 However, the most impor- when pumped under pressure through an in- occur. tant factor in considering the procedure and filtrated IV line, and a compartment syndrome Thrombosis or thrombophlebitis occurs route for vascular access is the individual can result. It is always best to have intravenous with malpositioned or misdirected catheters. physician’s level of skill and expertise. sites out where they can be examined. The subclavian catheter is often malpositioned Location of the injury must be considered into the internal jugular vein. When the cath- when choosing a site for venous access. Venous Central Venous Access eter is placed properly in the SVC, thrombosis access must never be initiated in an injured limb. Rapid peripheral percutaneous IV access usually does not occur because of the high In patients with injuries below the diaphragm, may be difficult to achieve in patients with hy- flow and large caliber of the vessel. A kinked at least one IV line should be placed in a tribu- povolemia and venous collapse, edema, obe- or knotted catheter in the SVC may lead to tary of the superior vena cava, as there may be sity, scar tissue, history of IV drug abuse, or thrombosis. vascular disruption of the inferior vena cava. burns. Under such circumstances, central ac- Injury to the brachial plexus or phrenic Patients with upper thoracic and neck injuries cess with wide-bore catheters can be advanta- nerve may result from attempts to place a sub- should have large-bore access in the lower ex- geous. An additional benefit is the ability to clavian line. The nerves are posterior to the tremity, as there may be superior vena cava dis- monitor central venous pressure. However, vein, and injury occurs when the needle has ruption. In patients with severe multitrauma in subclavian and internal jugular catheterization penetrated both walls. Left-sided central line whom occult thoracoabdominal damage is sus- should not be used routinely in trauma pa- attempts can result in thoracic duct injury. pected, it is recommended to have one IV ac- tients, as the complications can be dangerous. Infectious complications associated with cess site above the diaphragm and one below line placement can be prevented by using the diaphragm, thus accessing both the supe- Subclavian Catheterization proper sterile technique. Any lines placed dur- rior vena cava and inferior vena cava, respec- Subclavian catheterization provides rapid ing resuscitation of a trauma patient without tively.2,3 and safe venous access in experienced hands. strict aseptic technique should be removed. For rapid administration of large amounts The most frequent complication of subclavian of intravenous fluids, short large-bore catheters venipuncture is pneumothorax. Pneumotho- Internal Jugular Vein Catheterization should be used. Based on Poiseulle’s law, the rax is more likely to occur on the left side be- Percutaneous placement of internal jugu- rate of fluid flow is inversely proportional to cause the left pleural dome is anatomically lar (IJ) catheters is also an excellent means of the length of the catheter and directly propor- higher. Subclavian and internal jugular cath- attaining rapid large-bore catheter access. Cer- tional to its internal diameter: eters should be inserted on the side of injury vical trauma is a contraindication for internal in patients with chest wounds, reducing the jugular placement. Trendelenburg position ∏r4(∆P) chances of collapse of the uninjured lung. A and Valsalva maneuver help to distend the in- Q = simple pneumothorax may result in respira- ternal jugular vein and improve the rate of 8nL tory compromise in individuals with pulmo- success for venipuncture. nary contusions or a pneumothorax in the con- Carotid artery puncture is a common com- where Q=flow, r=radius of the catheter, tralateral hemithorax.2 A suspected injury to the plication of IJ catheter placement. Local direct P=driving pressure through the catheter (grav- subclavian vein is an exception to this principle pressure can prevent hematoma formation. ity or externally applied), n=viscosity of the because the infused fluid may extravasate into Carotid puncture is a contraindication to at- solution, and L=length of IV tubing. Doubling the mediastinum or thoracic cavity. tempting IJ catheter placement on the oppo- the internal diameter of the venous cannula A hemothorax may result from laceration site side, because bilateral hemorrhages could increases the flow through the catheter 16-fold. of the subclavian vein or subclavian artery. If compress the airway. A 14-gauge, 5-cm catheter in a peripheral vein the subclavian catheter is placed inadvertently Other complications from IJ venipuncture will pass fluid twice as fast as a 16-gauge, 20- in the thoracic cavity, subsequent infusions of are similar to those associated with subclavian cm catheter passed centrally. Although resis- blood or crystalloids will produce a hemotho- venipuncture. The incidence of pneumotho- tance to flow is added by multiple stopcocks rax or hydrothorax. Catheter placement should rax is less with IJ catheter placement than with and connections, stopcocks are recommended be ensured prior to IV infusions, whether by placement of a subclavian line. The incidence for universal precautions. When using 8.5 aspiration or by lowering the IV infusion bag of hemothorax, mediastinal migration of the French pulmonary catheter introducers, the below the patient and verifying backflow. These catheter, and intrapleural catheter placement side port should be removed, as this increases tests are suggestive of IV placement but none tends to be greater with left IJ placement than the resistance roughly four-fold. For subcla- is diagnostic.4 When inserting introducers over right because the left IJ is more circuitous, and vian, internal jugular, femoral, and antecubital guide wires, it is important not to force the advancement of a catheter can rupture the lines, 8.5 French introducers can be used.4 introducer if resistance is encountered. Forc- vessel. Stellate ganglion injury is a possible ing the introducer could result in perforation complication. Percutaneous Intravenous Insertion of large veins or arteries and bleeding. ATLS™ guidelines recommend rapid Venous air embolism is another complica- Femoral and Basilic-Cephalic placement of two large-bore (16-gauge or tion of central line insertion. Occlusion should Central Lines larger) IV catheters in the patient with serious be maintained over the catheter lumen with a Femoral vein cannulation is another alter- injuries and hemorrhagic shock. The first gloved finger or by increasing the pressure in native for line placement and is associated with choice for IV insertion should be a peripheral the subclavian vein by Trendelenburg position fewer acute complications. Bowel perforation extremity vein. The most suitable veins are at or Valsalva maneuver. Even with prompt can occur, especially in patients with femoral the wrist, the dorsum of the hand, the antecu- therapy, the fatality rate with significant air em- hernia. Penetration of the hip could result in bital fossa in the arm, and the saphenous in bolism is high.5 Embolization of catheter frag- septic arthritis. Thrombophlebitis occurs more the leg. These sites can be followed by the ex- ments can occur when withdrawing a catheter often with femoral than with IJ or subclavian ternal jugular and femoral vein. with a through-the-needle technique. catheters; however, this is most likely with pro- The complication rate of properly placed Arrhythmia may occur during line place- longed use. intravenous catheters is low. Intravascular ment when the catheter or wire contacts the Basilic-cephalic catheterization may be

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 29 used for central access and central venous pres- 6 years of age.1 Scalp veins should not be used The incidence of osteomyelitis is low when sure monitoring with a “long-arm” catheter. when rapid fluid administration may be catheters are removed early. Standard periph- Introducers can also be inserted safely. They needed. Internal jugular and subclavian cath- eral or central venous placement should be are easily placed and associated with a low eterization can be done in children but should attempted when the patient is stable. Bones complication rate. be performed only by experienced personnel. with fractures and sites with open wounds In awake children, there is a higher incidence should be avoided.5 Venous Cutdowns of pneumothorax and arterial puncture. Venous cutdowns can be performed when Intraosseous catheters can be used in all References rapid, secure, large-bore venous cannulation age groups but are most successful in those 1. Alexander RH, Proctor HJ. Advanced is desirable, such as in hemodynamic shock younger than 2 because the cortical bone is Trauma Life Support Program for Physi- and in situations where percutaneous periph- softer. Fluids and drugs can be given through cians, Instructor Manual. Chicago, Ameri- eral or central access is either contraindicated the catheter. Specially designed intraosseous can College of Surgeons, 1993. or impossible to achieve. needles are available but 18- to 20-gauge 2. Lucas CE, Ledgerwood AM. Initial evalua- Most favored sites for cutdowns are the needles, bone marrow aspiration needles, and tion and management of severely injured cephalic, basilic, and median antecuital veins 18-gauge spinal needles can be used. Eighteen- patients. In Wilson RF, Walt AJ, eds.. Man- in the upper extremity and the greater saphe- gauge spinal needles are readily available, but agement of Trauma: Pitfalls and Practice. nous in the lower. These veins can accept large they often bend and make placement difficult. Baltimore, Williams & Wilkins, 1996. catheters, allowing rapid infusion. Strict asep- In children younger than 6 years of age, the 3. Abrams KJ. Preanesthetic evaluation. In tic technique should be used. Surgical masks locations of choice are the proximal tibia and Grande, CM, ed. Textbook of Trauma An- and caps should be worn.6 the distal femur. When using the proximal tibial esthesia and Critical Care. St. Louis, Venous cutdown has a low potential for plateau, the needle should be placed 2 to 3 Mosby, 1993. anatomic damage. Cutaneous nerve injury is the cm distal to the level of the tibial tuberosity on 4. Kollef MH. Fallibility of persistent blood most common problem. The infection rate is the anterior medial surface of the proximal return for confirmation of intravascular relatively low when used acutely but increases tibia. In adults, a site 2 cm proximal to the tip catheter placement in patients with hem- precipitously over time. Therefore, it is recom- of the medial malleoli is selected, with the orrhagic thoracic effusions. Chest 1994; mended that venous cutdown catheters be re- needle directed slightly cephalad. The distal 106:1906–8. moved as soon as it is possible to achieve IV tibia, distal femur, sternum, clavicle, and 5. Bickell W, Pepe PE, Mattox KL. Complica- access through standard percutaneous IV cath- humerous can also be used. Pressure and a tions of resuscitation. In Mattox KL, ed. eters or a central venous catheter.5 rotary motion should be used until there is a Complications of Trauma. New York, decrease in resistance, indicating that the med- Churcill Livingstone, 1994. Vascular Access in Pediatric Patients ullary cavity has been entered. It is not always 6. Mackersie RC. Venous and arterial cut- Ideally, venous access in severely injured possible to aspirate marrow, but IV fluid should down. In Benumof JL, ed. Clinical Proce- children should be established via a percuta- run easily without a pump.7 dures in Anesthesia and Intensive Care. neous route. Unfortunately, this often proves Complications of intraosseous infusions Philadelphia, J.B. Lippincott, 1992. to be a difficult task. ATLS™ recommends that include extavasation of fluids into surround- 7. Benumof JL. Intraosseous infusion. In after two unsuccessful percutaneous attempts, ing tissues, cellulitis, and osteomyelitis. Mul- Benumof JL, ed. Clinical Procedures in consideration should be given to intraosseous tiple attempts at insertion should be avoided Anesthesia and Intensive Care. Philadel- infusion in children younger than 6 years of since the other holes in the bone could allow phia, J.B. Lippincott, 1992. age or direct venous cutdown in children over leakage of fluid into the adjacent soft tissue.

11 Principles of Fluid Warming in Trauma

Charles E. Smith, MD, FRCPC principles of fluid warming as they apply to may contribute to a hypothermia-induced or MetroHealth Medical Center the trauma patient. dilutional coagulopathy, although experimen- Case Western Reserve University tal evidence suggests that hydraulic factors may Cleveland, OH 44109 USA Importance of Warming IV Fluids play a more important role (e.g., disruption of e-mail: [email protected] Conclusive evidence demonstrating the soft clot, decreased resistance to flow around harmful effects of cold fluid infusion was pro- a partially formed thrombus).18 [Editors’ note: Dr. Smith has received research vided by Boyan and Howland.14 In their study, support from SIMS Level I, Augustine Medical infusion of 0.5 L of cold blood reduced core Thermal Stress of Infusing Cold or Inad- Mallinckrodt, and Belmont Instruments.] temperature of anesthetized cancer patients by equately Warmed Fluids and Blood 0.5 to 1.0oC. When 3.0 L or more of cold blood The theoretical impact of infusing fluids on Hypothermia occurs frequently in trauma was administered, esophageal temperature de- body temperature can be calculated as follows: patients because of exposure, infusion of cold creased markedly and was associated with a high Change in body temperature = fluids and blood, opening of body cavities, incidence of cardiac arrests (12 of 25 patients).14 Thermal stress of infused fluids / decreased heat production, and impaired ther- When blood was warmed, the incidence of car- (Weight x Sp heat) moregulatory control.1–7 Infusion of unwarmed diac arrests in a matched group of patients with where: or inadequately warmed IV fluids and cold similar surgeries, blood loss, anesthesiologist, Thermal stress = Temperature difference blood is a well known cause of hypothermia and surgeon was only 3 of 105 patients.15,16 between core and infused fluids x and may contribute to the multiple adverse The use of large quantities of unwarmed specific heat of infused fluid x volume consequences of hypothermia such as periph- fluids for immediate resuscitation of patients of fluid infused eral vasoconstriction, metabolic acidosis, with penetrating trauma prior to emergency Weight = weight of patient in kg coagulopathy, wound infection, and cardiac surgical intervention has been discouraged.17 Sp heat = specific heat of the patient morbidity.1–3,8–13 This manuscript reviews the It is possible that the use of unwarmed fluids (0.83 kcal/L/oC)19,20

30 Massive Transfusion and Control of Hemorrhage in the Trauma Patient According to the specific heat of water, 1 kCal of heat is required to raise the tempera- Table 1. Commercially Available Warming Devices ture of 1 kg of water by 1oC. Assuming that 1 L of crystalloid weighs 1 kg and that its specific Instrument Technology Comments heat is the same as water, one needs 16 kCal of energy to raise the temperature of 1 liter of Flotem IIe Dry heat IV tubing sandwiched between heating plates crystalloid infused at 21oC to body temperature (37oC).19-22 Similarly, infusion of 4.3 L of DW-100 Dry heat Plastic bag wrapped around heating cylinder crystalloid at room temperature to an adult Fenwall Dry heat Plastic bag with channels sandwiched between trauma patient would require 71 kCal, the heating plates equivalent of 1 hour of heat production in an awake adult, or 1.5 hour of heat production Dupaco Water bath Coiled IV tubing immersed in a bath in an anesthetized adult male (heat production decreased by 33%). Level 1 H250 Countercurrent Tube in tube heat exchange The negative thermal balance of 4.3 L of water bath room temperature fluids is thus equivalent to a decrease of 1oC body temperature in an Level 1 H500 Countercurrent Tube in tube heat exchange, larger heater than H250 awake individual and a 1.5oC temperature de- water bath crease in an anesthetized patient. Conversely, 30 kCal are required to raise the temperature Hotline Countercurrent Entire 254-cm patient IV line is warmed to ensure of cold 4oC blood to 37oC, such that infusion water bath delivery of warm fluids at flow rates between 5 and of 2 L could result in a body temperature de- 90 ml/min (300-5000 ml/hr) crease of between 1.0 and 1.5oC.19-22 Level 1 H1000 Countercurrent Tube in tube heat exchange combined with a Temperature Setpoints of Warmers water bath 254-cm patient IV line with Hotline characteristics In the United States, blood can be warmed to prevent heat loss at moderate flow rates safely so as not to cause hemolysis using a tem- (<100 ml/min) perature setpoint of 42oC in conjunction with BairHugger Convective air Spiral IV tubing suspended in same convective an FDA-cleared blood warming device. This 2.4.1 warming hose that delivers forced air to a warming setpoint is based on observations by Uhl and blanket colleagues23 and is supported by a large body of experience with cardiac perfusion. In the FMS 2000 Magnetic High-speed volumetric pump with automatic air 23 study by Uhl et al, red cells were incubated induction detectors, line pressure sensor, and flow rate o at 37, 40, 42, 44, 46, 48, and 50 C for up to 2 control up to 500 mL/min; 122-cm patient line hours in a constant-temperature waterbath. Even subtle alterations in red cell integrity such R.I.S. Countercurrent High-efficiency pump with 3-L reservoir, three air/ as increased plasma hemoglobin and osmotic (Haemonetics) water bath bubble detectors, line pressure sensor, and fragility were not apparent until 46oC.23 automatic flow rate control up to 1500 ml/min There has been renewed interest in delivering very hot fluids in an attempt to transfer Arrow In-line Direct microwave energy transferred in a heating heat to hypothermic patients. For example, Thermostat microwave chamber to coils of IV tubing wound on a infusion of crystalloid at 54oC will transfer ~21 900 disposable cartridge kCal/L to a hypothermic patient whose core temperature is 33oC. This technique has been Baxter Dry heat Disposable canister sets that fit over the heating shown to be relatively safe in a series of pa- Thermacyl unit; bubble trap to remove microbubbles tients undergoing operative burn debridement Mallinckrodt Countercurrent Metal foil cassette inserted between two heated and immediate skin grafting.24 Fluids were in- Warmflo metal plates; IV tubing inserts directly into metal fluid fused at a rate of 110 ml/hr. In the study, there FW538 channels within the cassette was no evidence of intravascular hemolysis or other overt complications such as excessive Alton Dean Countercurrent Metal foil cassette inserted between two heated 24 bleeding or hyperkalemia. There is currently metal plates; IV tubing inserts directly into metal fluid not enough safety information to recommend channels within the cassette this technique and there is danger that very hot fluids may result in local vascular damage Ranger Countercurrent Cartridge-style plastic disposable set inserted and other complications such as hemolysis. metal between conductive warming plates

Fluid Warming Devices (Table 1) Intravenous administration of large vol- tubing through a conductive surface interfaced such as plastic, limited surface area of the heat umes of inadequately warmed fluid can lead with a counter-current heated water bath, mag- exchange mechanism, inadequate heat trans- to significant hypothermia. Several methods to netic induction, and inline microwaving.25–29 fer of the exchange mechanism at high flow warm IV fluids are currently available. These The ideal fluid warmer should be capable rates, erythrocyte damage, and heat loss after methods include immersing coiled IV tubing of safely delivering fluids and blood products the IV tubing exits the warmer.25,30–32 For ex- in a warm water bath, microwaving the bag of at normothermia at both high and low flow ample, adding warmed saline to blood could fluid to be infused, adding heated saline to rates. The ability of blood warmers to safely have catastrophic results unless the saline is blood to be infused, passing the IV tubing deliver normothermic fluids over a wide range not heated above a certain temperature—the through a heating block or through a plastic of flows is limited by several factors, including maximum safe temperature would be highly tube warmed with forced air, passing the IV limited heat-transfer capability of materials dependent on the relative volume of saline and

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 31 Figure 1. Table 2. Implications of Using Warming Devices for Crystalloid Schematic of the Level 1 250 and Fluid Resuscitation (5 and 10 L) in Anesthetized Adult Trauma Patients 500 warmer. The device consists of a heater that warms water and circulates Device Flow Rate Outlet Decrease in Decrease in it through a pump and a heat-exchange (ml/min) Temperature MBT* (5 L MBT* (10 L segment with a central tube for water flow infusion, °C) infusion, °C) (countercurrent heat exchange technology). Fluid flows through the outer sheath, Flotem IIe which surrounds the water core. Pressure 260 24 -1.12 -2.24 Note the filter and air eliminator distal Gravity 90 27 -1.03 -2.06 to the heat exchanger. Astotherm Pressure 260 25 -1.03 -2.06 Gravity 90 30 -0.60 -1.20

BairHugger 2.4.1 Pressure 360 24.2 -1.10 -2.20 Gravity 80 29.6 -0.63 -1.27

Hotline Pressure 220 29.8 -0.62 -1.24 Gravity 80 34.8 -0.19 -0.38

Level 1 250 Pressure 600 33 -0.34 -0.69 Gravity 290 36 -0.09 -0.17

Level I H1000 Pressure 470 39.5 +0.22 +0.43 Gravity 150 39.4 +0.21 +0.41

FW537 Pressure 580 38.9 +0.16 +0.32 Gravity 200 39.9 +0.25 +0.49 blood. The dangers of using unproven meth- Cardioplegia ods and nonapproved approaches to blood Heat Exchanger warming cannot be overemphasized. Pressure 700 35 -0.17 -0.34 The heat-transfer capabilities of warming Gravity 150 35 -0.17 -0.34 devices using dry heat exchange technology is limited by use of poorly conducting materials For all devices, fluids were infused during two conditions—pressure-driven infusion and such as plastic and by limited heat transfer sur- gravity-driven infusion with the roller clamp wide open. Data from references 32 and 33. face area. Warming devices that utilize counter- *Change in mean body temperature (MBT) was calculated as follows: current heat exchange (Level 1 H250 [Fig. 1], H1000 [Fig. 2], and FW537) are capable of (Tfluid - Tpatient) Sfluid / Weight x Spatient warming fluids even at very rapid flow rates due where to better conduction materials interposed be- Tfluid = Outlet temperature of fluid delivered to the patient tween the heating elements and the infused o Tpatient = Temperature of the patient, assumed to be 37 C fluid.32,33 Therefore, both these devices are o Sfluid = Specific heat of infused fluid, 1 kcal/L/ C appropriate for situations where rapid (>100 o 19,20 Spatient = Specific heat of the patient, 0.83 kcal/l/ C ml/min) volume resuscitation is necessary. Weight of patient was assumed to be 70 kg At moderate flows (<100 ml/min), there is significant heat loss after the IV tubing exits the warmer. The continual countercurrent warming of fluids in the tubing (Hotline [Fig. 3] and the second scenario, the fluid and blood vol- core temperature, the greater the drop in mean H1000) essentially eliminates the loss of heat ume deficit is not as severe, although ongoing body temperature. As well, the greater the fluid along the tubing distal to the warmer.33 blood loss may necessitate moderately fast in- requirement relative to body weight, the greater Table 2 summarizes the implications of us- fusions with the roller clamp wide open to main- the potential drop in body temperature. ing various fluid warmers during commonly en- tain normovolemia and hemodynamic stability. Because of the marked inefficiencies of countered clinical situations: pressure-driven in- It can be seen from the calculations in Table 1 conventional warming devices such as the fusion, and gravity-driven infusion with the that the thermal stress of infusing cold fluids Flotem IIe, Astotherm (Fig. 4), and others roller clamp wide open.32,33 In the first situa- may result in considerable changes in mean (Fig. 5), these devices are no longer in use at tion, the patient presents with severe circula- body temperature, especially if the patient is the author’s institution and have been re- tory shock due to massive blood loss. Fluid re- unable to increase heat production or prevent placed with the Level 1 H250 and H1000 for suscitation via large-bore IV cannulas is required further heat loss. The larger the gradient be- rapid infusion (>100 ml/min or 6 L/hr) and to prevent acidosis and irreversible shock. In tween the temperature of the infused fluid and the Hotline device for all other situations.

32 Massive Transfusion and Control of Hemorrhage in the Trauma Patient Figure 2a and b. Figure 4. Level 1 H1000 warmer. The device consists of a cylindrical aluminum heat exchanger Schematic of the Astotherm warmer. mounted on the warming unit and heated by a countercurrent water bath, similar to the This device consists of IV tubing coiled Level 1 250. After the fluid exits this first heat exchanger, it enters a 254-cm patient line around a circular heating element in which heat loss is prevented by surrounding the central lumen with warmed water (dry heat technology). circulating in a countercurrent direction, similar to the Hotline device.

Figure 3a and b. Figure 5. Hotline warmer. This device consists of a water bath and an L-70 disposable. Schematic of the modified cardioplegia The L-70 disposable heats fluid within the 254-cm patient line, which consists of a heat exchange warmer. The device consists central lumen for the IV fluid surrounded by an outer layer through which warm water of a water bath that circulates water circulates down one side and then back up to the warm water reservoir in a through a stainless steel cardioplegia heat countercurrent fashion (countercurrent heat exchange technology). exchanger in a countercurrent fashion. This device is no longer used at the author’s institution because of delays in setup and de-airing and high disposable costs.

Safety of Rapid Infusion Devices with author’s experience, it is wise to have one in- sion devices not be used unless the patient is Constant Pressure dividual solely responsible for pressurized in- in profound hemorrhagic shock, and all air has Because of the high flow rates generated fusion of fluids. This individual must utilize been removed from the fluid to be infused rap- by newer warmers when used with constant- extreme vigilance and caution because of the idly. The automatic air eliminator incorporated pressure devices, the limiting factor in fluid danger of infusing air at these high flow rates. into the design of the Haemonetic RIS and resuscitation is the time required to identify This author is aware of four cases of massive Level 1 devices make these units somewhat red cell donor and recipient information, to air embolus at other institutions following use safer, but does not eliminate the risks of mas- spike and hang the fluid, and to ensure ab- of pressurized infusions. Therefore, it is the sive air embolus. sence of air from the fluid system. In the author’s belief that constant pressurized infu-

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 33 Summary Perioperative maintenance of normother- McNiece WL. Evaluation of a new fluid Adverse consequences of perioperative mia reduces the incidence of morbid car- warmer effective at low to moderate flow hypothermia include myocardial ischemia, cardiac events: a randomized clinical trial. rates. Anesthesiology 1993; 78:974–80. diac arrhythmias, coagulopathy, shivering, in- JAMA 1997; 227:1127–34. 26. Patel N, Smith CE, Pinchak AC, Hagen JF: creased oxygen consumption, alteration in 14. Boyan CP, Howland WS. Blood tempera- Prospective, randomized comparison of drug metabolism and increased wound infec- ture: a critical factor in massive transfu- the Flotem IIe and Hotline fluid warmers tion. Administration of cold or inadequately sion. Anesthesiology 1961; 22:559–63. in anesthetized adults. J Clin Anesth 1996; warmed intravenous fluids contributes to hy- 15. Boyan CP, Howland WS. Cardiac arrest and 8:307–16. pothermia, whereas administration of normo- temperaturre of bank blood. JAMA 1963; 27. Smith CE, Holbrook C, Radesic B, thermic fluids may reduce both the incidence 183:58–60. Raghupathy A, Sweda S, Botero CA, Patel and complications of hypothermia. Therefore, 16. Boyan CP. Cold or warmed blood for massive N, Punjabi A, Thompson L, Hagen JF, infusion of adequately warmed fluids is impor- transfusions. Ann Surg 1964; 160:2882–6. Pinchak AC. Comparison of perioperative tant in order to minimize thermal stress and 17. Bickell WH, Wall MJ, Pepe PE, et al. Imme- heating modalities in anesthetized adult maintain thermal homeostasis. diate versus delayed fluid resuscitation for patients: a prospective randomized study. hypotensive patients with penetrating torso Am J Anesthesiol 1998; 25:62–8. Acknowledgement injuries. N Engl J Med 1994; 331:1105–9. 28. Smith CE, Desai R, Glorioso V, Cooper A, The secretarial assistance of Fran Hall is 18. Martin RR, Bickell WH, Pepe PE, et al. Pro- Pinchak AC, Hagen JF. Preventing hypoth- very much appreciated. spective evaluation of preoperative fluid ermia: convective and intravenous fluid resuscitation in hypotensive patients with warming versus convective warming References penetrating truncal injury. J Trauma 1992; alone. J Clin Anesth 1998; 10:380–5. 1. Danzl DF, Pozos RS. Accidental hypother- 33:354–62. 29. Smith CE, Patel N: Hypothermia in adult mia. N Engl J Med 1994; 331:1756–60. 19. Mendlowitz M. The specific heat of human trauma patients: anesthetic consider- 2. Luna Gk, Maier RV, Pavlin EG, Anardi D, blood. Science 1948; 107:97. ations. Part II, Prevention and treatment. Copass MK, Oreskovich MR. Incidence and 20. Gentilello LM, Cortes V, Moujaes S, Am J Anesthesiol 1997; 24:29–36. effect of hypothermia in seriously injured Viamonte M, Malinin TL, Ho CH, Gomez 30. Fildes J, Fisher S, Sheaff CM, Barrett JA. patients. J Trauma 1987; 27:1014–8. GA. Continuous arteriovenous rewarm- Effects of short heat exposure on human 3. Jurkovich GJ, Greiser WB, Luterman A, ing: experimental results and thermody- red and white blood cells. J Trauma 1998; Curreri PW. Hypothermia in trauma vic- namic model simulation of treatment for 45:479–84. tims: an ominous predictor of survival. J hypothermia. J Trauma 1990; 30:1436–49. 31. Herron DM, Grabowy R, Connolly R, Trauma 1987; 27:1019–24. 21. Dubois EF. Basal Metabolism in Health Schwaitzberg SP. The limits of 4. Gregory JS, Flancbaum L, Townsend MC, and Disease. Philadelphia, Lee and bloodwarming: maximally heating blood Cloutier CT, Jonasson O. Incidence and Febiger, 1924, p 324. with an inline microwave bloodwarmer. J timing of hypothermia in trauma patients 22. Uhl L, Pacini DG, Kruskall MS. The effect Trauma 1997; 43:219–26. undergoing operations. J Trauma 1991; of heat on in vitro parameters of red cell 32. Patel N, Smith CE, Pinchak AC: Compari- 31:795–800. integrity. Transfusion 1993; 33:60S. son of fluid warmer performance during 5. Pavlin EG. Hypothermia in traumatized pa- 23. Gore DC, Beaston J. Infusion of hot crys- simulated clinical conditions. Can J tients. In Grande CM, ed. Textbook of talloid during operative burn wound de- Anaesth 1995; 42:636–42. Trauma Anesthesia and Critical Care. St. bridement. J Trauma 1997; 42:1112–5. 33. Patel N, Knapke DM, Smith CE, Napora Louis, Mosby-Year Book, 1993, chapter 94, 24. Uhl L, Pacini D, Kruskall MS. A compara- TE, Pinchak AC, Hagen JF: Simulated clini- pp 1131–9. tive study of blood warmer performance. cal evaluation of conventional and newer 6. Little RA, Stoner HB. Body temperature Anesthesiology 1992; 77:1022–8. fluid warming devices. Anesth Analg 1996; after accidental injury. Br J Surg 1981; 25. Presson RG, Bezruczko AP, Hillier SC, 82:517–24. 68:221–4. 7. Sessler DI. Mild perioperative hypothermia. N Engl J Med 1997; 336:1730–7. 8. Smith CE, Patel N: Hypothermia in adult 12 trauma patients: Anesthetic consider- Management of Massive Hemorrhage ations. Part 1, Etiology and Pathophysiol- ogy. Am J Anesthesiol 1996; 23:283–90. and Transfusion in Trauma 9. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the Georges Desjardins, MD, FRCPC improvements in emergency prehospital care, incidence of surgical- wound infection and Attending Anesthesiologist initial resuscitation of trauma victims in emer- shorten hospitalization. N Engl J Med Boca Raton, Florida, USA gency departments, and rapid transport to op- 1996; 334:1209–15. erating rooms. It is not uncommon to care for 10. Sessler DI. Consequences and treatment Trauma is the most common cause of death patients with blunt injuries to the great vessels, of perioperative hypothermia. Anesth Clin in Americans under the age of 45.1 In the United penetrating injuries to the heart, severe blunt North Am 1994; 12:425–56. States, deaths from unintentional injuries are injuries to the liver, severe open-book pelvic 11. Watts DD, Trask A, Soeken K, et al. Hypo- most often the result of motor vehicle crashes, fractures, and penetrating injuries to the trunk thermic coagulopathy in trauma: effect of falls, poisoning, fires, or drowning. Although and to then see these patients leave the hospi- varying levels of hypothermia on enzyme the number of deaths from motor vehicle tal to lead constructive, functional lives. speed, platelet function, and fibrinolytic crashes has decreased over the past few years, Hypotension and hypovolemia are gener- activity. J Trauma 1998; 44:846–54. there has been an alarming increased in fire- ally regarded as detrimental to the brain and 12. Frank SM, Higgins MS, Breslow MJ, et al. arm-related deaths. If this trend continues, other organs and are associated with worse The catecholamine, cortisol, and hemody- deaths from firearms are likely to exceed those outcome, particularly in association with se- namic responses to mild perioperative hy- from motor vehicle crashes by the year 2003.1 vere head injury. In recent reports, there is pothermia. Anesthesiology 1995; 82:83–9. Trauma anesthesiologists are faced today speculation that hypovolemia and associated 13. Frank SM, Fleisher LA, Breslow MJ, et al. with sicker patients than in the past because of hypotension are beneficial in some circum-

34 Massive Transfusion and Control of Hemorrhage in the Trauma Patient stances when hemorrhage is uncontrolled.2,3 and induction techniques to set new priori- being applied (cervical collar, backboard, triple The most frequently stated example is a lacer- ties and techniques for the resuscitation. This fixation of the cervical spine), and unstable vital ated major artery, when the administration of modification is the crash emergency anesthe- signs are present, several problems can be an- fluid and associated increase in blood pressure sia technique. It is in fact a combination of the ticipated. Moving the head and neck or open- might dislodge a clot from the area of injury, ATLS™ initial evaluation7 and the regular an- ing the cervical collar would be necessary to increase the hemorrhage, and turn stable hy- esthesia induction set-up. The first priorities perform easy and timely cannulation of either potension into lethal recurrent hemorrhage. will be evaluating and managing the airway, the external or the internal jugular vein. Re- The evidence for the occurrence of these theo- oxygenation, ventilation, followed by measur- moving some of the spinal precautions before retic effects from fluid resuscitation is stron- ing the blood pressure; sorting out the intra- clinical or radiologic clearance would not be ger for penetrating trauma than for blunt venous lines already in place; finding access ideal and waiting for the radiologic evaluation trauma, which is more common in patients for drug injection; attaching an ECG; infusing would be impractical. In these circumstances, with head injuries. Currently there is general fluids through blood warmers; getting blood the femoral vein could be an excellent second support for fluid administration as a mainstay in the room; checking the patient identity and choice for venous access because of its large of initial resuscitation after blunt trauma. The history of allergy; placing an arterial catheter; size and easy access. However, a major con- initial hemodynamic stabilization is still intra- drawing blood for blood gases, hematocrit, and cern with the use of the femoral vein as the venous access, correction of hypovolemia, and other lab tests; titrating an anesthetic, if pos- “main IV” in the acute phase of resuscitation hemorrhage identification and control. sible; checking temperature and urine output; is the possibility of vascular injuries from the This review focuses on the management inserting a central venous or pulmonary artery original trauma in the pelvic and/or the abdomi- of exsanguinating hemorrhage and massive catheter or the TEE probe for monitoring nal region, especially in patients with penetrat- transfusion from blunt or penetrating trauma needs; and finally inserting a gastric tube. ing trauma to the abdomen and in patients who after the patient’s arrival in the operating room. The route for fluid administration in have sustained major pelvic fractures, in whom trauma is a source of controversy. There is gen- associated vascular injuries are frequent. Rely- Definition eral consensus that the first choice for cannu- ing mainly on femoral access in this situation Resuscitation of the severely injured patient lation is a vein that is visible, which most often might lead to loss of resuscitation fluid into the with fluids and blood products for hemorrhagic means a peripheral vein on the upper extremi- extravascular space. Use of a venous cutdown shock is often associated with complex meta- ties. Two large-bore peripheral intravenous in the lower extremities has the same limita- bolic alterations. Several definitions for massive catheters (16 gauge or larger) should be placed tion. Although a venous cutdown in the upper blood transfusions have been proposed.4,5 as quickly as possible for the administration extremities would avoid this problem, it is tech- These range from the replacement of the of fluids and blood. Using 14- or 16-gauge 2- nically more difficult and therefore often more patient’s whole blood volume in 24 hours to inch peripheral catheters should allow a flow time consuming. When there is inadequate in- replacement of 50% of the volume in 3 hours. rate of 300 ml/min of crystalloid or 150 ml/ travenous access in the severely injured patient When reviewing the physiologic conse- min of blood when used in combination with with suspected intra-abdominal injuries, it is our quences of massive transfusion, knowledge of a pressure bag.8 In areas with well-developed practice to use the subclavian vein as our sec- these different definitions would seem essential, emergency medical systems, most trauma vic- ond choice for fluid administration6 (Table 1). as they are quite different. Another definition that tims arrive at the hospital with these intrave- In situations of advanced hypovolemic shock we would like to submit is the concept of mas- nous catheters already in place.9 or exsanguination, where percutaneous tech- sive massive transfusion, which we define as the If peripheral intravenous access was un- niques of IV insertion via peripheral central replacement of a patient’s estimated blood vol- successful in the field or in the resuscitation veins are unsuccessful, venous cutdown at the ume in less than 30 or 60 minutes. Certainly, the room or if hypotension persists, additional saphenofemoral junction may be used.10 metabolic abnormalities associated with blood sites should be considered to ensure immedi- The use of an 8.5 or 9.0 French introducer replacement and resuscitation in this type of pa- ate intravenous access. Some authors suggest, allows a flow rate higher than 500 ml/min with tient should be anticipated to be worse than for as a second choice, the cannulation of the ex- the use of a pressure bag and large-caliber IV patients who get 20 units of packed red blood ternal jugular vein; as third choice, the use of tubing.8,11 Strict aseptic technique should be used cells (PRBCs) in 24 hours. the femoral vein; and as last choices, venous even in emergency situations. As a general rule, cutdown and catheterization through the in- all intravenous catheters placed in the prehospital Initial Evaluation ternal jugular or subclavian veins.9 phase and in the resuscitation room should be After initial evaluation and resuscitation, the If the patient does not have adequate in- changed in the first 24 hours after insertion, be- trauma victim requiring surgery should be moni- travenous access, spinal precautions are still cause they may have been inserted under less- tored during transport and be accompanied by members of the trauma team. The basic monitoring devices for transport include an electro- Table 1. Intravenous Access in the Patient with Multiple Injuries cardiogram, automated blood pressure device, Option 1 — Peripheral IV x 2 in visible vein of the upper extremities and pulse oximeter. End-tidal CO2 monitoring should be considered if the patient is intubated. Ideally, the trauma team leader should transfer Option 2 — If unsuccessful, suggested second choice: care directly to the anesthesiologist involved in the case. Information accompanying the trans- • If cervical spine injury is unlikely: fer should include mechanism of injury, injuries External or internal jugular vein access with large-bore IV catheter that have been identified, results and omissions of investigations, medical history, and allergies. • If abdominal or pelvic injuries are unlikely: Giving this advance notice to the team in the Femoral vein access with large-bore IV catheter operating room before the actual arrival of the or patient will avoid delays and keep the focus on Venous cutdown in the lower extremities continuity and quality of care.6 After the arrival of an exsanguinating pa- • If abdominal or pelvic injuries are suspected: tient in the operating room, the anesthesiolo- Subclavian vein with large-bore IV catheter gist will have to modify his or her evaluation

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 35 than-ideal aseptic conditions.8 Our practice is to and increased hemolysis is linked to the de- hemoglobin for oxygen and impairs clotting provide the history of all IV lines to the ICU or creased levels of intracellular ATP. This in turn function. Low temperature also increases the ward teams, who will then change all central is linked to the increased levels of potassium, potential for hypocalcemia because of de- catheters over a guidewire, culture the intracu- ammonia, and hemoglobin in the supernatant creased hepatic metabolism of citrate. Preven- taneous segments and tip of the catheter with a plasma or preservative solution. The change tion of hypothermia is essential and can be semiquantitiative techniques, and remove the in oxygen affinity of hemoglobin for oxygen achieved by warming intravenous fluids and peripheral lines placed during the prehospital is, in large part, a consequence of decreased blood during administration, warming the and resuscitation phases of care.6 These changes levels of intracellular 2,3-DPG. The increase in operating room to 30°C, and using convective should be done only after relative hemodynamic vasoactive substances is a result of their release warming blankets in all cases of severe trauma. stability has been established and/or additional from leukocytes and platelets contained in the As the amount of blood replacement in- “clean” intravenous access has been secured. blood or red cell concentrate. Finally, the decreases, the trauma patient’s own blood be- All intravenous fluids and blood products velopment of microaggregates is due to the gins to take on characteristics of bank blood, should be warmed. The H1000 infusion system formation of small amounts of fibrin strands with low levels of 2,3-DPG and low activities (Level One Technologies, Inc., Rockland, Mas- during storage and the adherence of senescent of Factor V and VIII, as well as dilutional throm- sachusetts) is capable of infusing and heating platelets and leukocytes to them. bocytopenia. When blood is stored at 4°C for 800 ml/min of crystalloid or 500 ml/min of Massive transfusion of blood components 24 to 48 hours, the platelets have only 5% to blood. The Rapid Infusion System (RIS, containing sodium citrate can lead to transiently 10% of normal activity. Following transfusion, Haemonetics Corporation, Braintree, Massachu- decreased levels of ionized calcium. Hypocal- these platelets are essentially nonfunctional. setts) can infuse blood products (red cells, fresh cemia can cause hypotension, narrowed pulse The massive transfusion of packed RBCs will frozen plasma), crystalloids, or colloids at rates pressure, and biventricular dysfunction. Elec- rapidly dilute the patient’s existing platelet up to 1,500 ml/min. This system is extremely trocardiographic abnormalities such as pro- pool. The decrease is often less than expected useful in the management of exsanguinating longed QT interval can occur. Adults who have on the basis of simple dilution because of some hemorrhage. The use of blood warming/high- normal hepatic function, are normothermic, release of platelets from the spleen and bone volume infusion systems in addition to warm- and are not in shock can tolerate the infusion marrow. Prompt platelet administration should ing the resuscitation room or operating room of one unit of PRBCs every 5 minutes (20 units/ be considered once abnormal bleeding is to temperatures at high as 30°C is essential if hr) without developing hypocalcemia.17 noted. In the patient who has microvascular hypothermia is to be prevented effectively dur- Since stored blood commonly has el- bleeding without hypothermia, a platelet count ing resuscitation of the trauma patient. evated potassium concentration, up to 30 to below 50,000/µl or a falling count below 40 mEq/L by 3 weeks of storage, hyperkalemia 100,000/µl indicates the need for platelet trans- Metabolic and Hemostatic Effects of is possible with massive transfusion. Hyper- fusion. Indications for fresh frozen plasma Massive Blood Transfusions kalemia may cause elevated peaked T waves (FFP) and cryoprecipitate are not clear. In Since banked blood undergoes a number on the electrocardiogram. It can significantly trauma patients who receive between one and of metabolic and structural changes over time, alter cardiac function, especially if associated two blood volume replacement, dilutional multiple severe derangements of physiology with hypocalcemia. The incidence of intraop- thrombocytopenia and fibrinogen levels below are theoretically possible when large volumes erative hyperkalemia increases with infusion 75 mg/dl often occur.20 Low levels of coagula- of banked blood are given to critically ill or rate of PRBCs above 150 ml/min. Hyperkale- tion Factors V and VIII are usually a clinical injured patients. Although the volume of blood mia can be treated early with intravenous cal- problem after two blood volume replacement. transfused may lead to a variety of problems cium, insulin, and bicarbonate and with PRBC Fibrinogen can be replaced with FFP or cryo- (Table 2), both the depth and duration of shock washing before administration.18 precipitate. In trauma patients, low coagula- appear to be more significant determinants of Although stored PRBCs have an acid pH tion factors are usually replaced with FFP. physiologic derangements than the transfusion (about 6.3), alkalosis is the usual result of mas- In addition to the metabolic changes ob- of blood itself.12 If the patient receiving massive transfusion without shock. Sodium citrate served with massive transfusion, infectious and sive transfusion receives adequate fluid resus- contained in the anticoagulant is converted to immunologic effects can complicate the care of citation and maintains oxygen delivery and sodium bicarbonate in the liver. The alkalosis trauma patients. Viral hepatitis remains the organ perfusion, the sequelae of massive trans- initially increases the oxygen affinity of hemo- major infectious risk of transfusion. With bet- fusion may be minimized. The volume of blood globin, resulting in less oxygen off-loading to ter donor blood screening in the United States, products that the patient receives should not the tissues. The clinical significance of this al- the estimated risks (per unit of blood trans- be the primary determinant of therapeutic kalosis is unknown. fused) of transmission of viral infection are as decisions or prognosis.13–15 Hypothermia may occur with rapid trans- follows21: HIV, 1:493,000; hepatitis B, 1:63,000; The ability to provide massive transfusion fusion of large volumes of cold blood compo- hepatitis C, 1:103,000; and HTLV, 1:641,000. is a relatively recent medical accomplishment nents. It remains the most under-recognized (See Chapters 8 and 9.) Transfusion has the resulting from a series of advances (large blood and under-treated cause of coagulopathy in potential to modify the recipients’ immune re- banks, rapid infusers of warm fluids, and bet- trauma patients.19 It increases the affinity of sponse. This is a potentially serious problem in ter understanding of the physiology of trans- many survivors of massive transfusion, who gen- fusion). The varied definitions of massive trans- erally develop immune compromise and are at fusion, the numerous associated clinical con- Table 2. high risk for sepsis and multiple organ failure. ditions, and the relative lack of detailed rigor- Metabolic and Hemostatic Effects ous studies have crated controversy in the lit- of Massive Blood Transfusions Management of Massive Transfusion erature regarding the metabolic effects of mas- One thing is clear: the goal of hemorrhagic sive transfusion. The confusion is com- Decreased oxygen dissociation shock resuscitation is prompt restoration of pounded by the use of blood of varied storage Hypocalcemia adequate perfusion and oxygen transport. The life, nonuniform resuscitation protocols, and Hyperkalemia objective of resuscitation is to reestablish oxi- comparison of patients suffering from shock Derangement of acid–base balance dative metabolism by providing adequate oxy- of differing severity and duration. Hypothermia (<35°C) gen flow to cells, preventing reperfusion dam- The storage and refrigeration of pRBCs Dilutional coagulopathy (platelets, age, and avoiding blood loss. results in progressive changes that are termed coagulation factors) Patients in hemorrhagic shock develop storage lesions.16 The change in deformability low pH from the buildup of intracellular hy-

36 Massive Transfusion and Control of Hemorrhage in the Trauma Patient drogen ions, which occurs during the anaero- products at normothermia. At our institution, Critical Care of the Neurological Injury. bic conversion of glucose to lactate. Some of the system is primed with a crystalloid solu- Futura Publishing, 1997, pp 95–120. the intracellular lactate and associated hydro- tion and blood products are added to the 3- 7. American College of Surgeons. Advanced gen ions eventually leave the cell and produce liter reservoir as indicated during the resusci- Trauma Life Support Student Manual. Chi- the characteristic metabolic acidosis of hem- tation. Platelets are not infused with the RIS cago, American College of Surgeons, 1993. orrhagic shock. The pH, lactate level, and base device. They are infused through a separate 8. Palter MD et al. Secondary triage of the deficit are highly correlated with mortality and intravenous access. The usual ratio of blood trauma patient. In Civetta JM, Taylor RW, are thought to be an underlying cause of de- products used with the RIS follows the Uni- Kirby RR, eds. Critical Care. Philadelphia, creased cardiac contractility and eventual mor- versity of Pittsburgh protocol, with 2 units of Lippincott, 1992, pp 611–25. 9. Calcagni De et al. Resuscitation: blood, tality. However, the clinical hemodynamic con- packed red cells (600 ml), 2 units of FFP (400 blood component and fluid therapy. In sequences of low serum pH are unclear. Many ml), and 500 ml of a colloid or crystalloid so- Grande CM, ed. Textbook of Trauma An- clinicians give bicarbonate to increase cardiac lution. The hematocrit of this solution is 28%. esthesia and Critical Care. St. Louis, contractility. There is some evidence that con- All blood is filtered through a 150-micron fil- Mosby, 1993, pp 381–416. tractility does not decrease substantially until ter as it is introduced into the reservoir. It then 10. Rogers FB. Technical note: a quick and the pH is 6.9 or 6.8, unless adequate oxygen is passes through a 40-micron filter. The heat simple method of obtaining venous access 22 not available. The most significant determi- exchanger system also uses countercurrent in traumatic exsanguination. J Trauma nants of depressed cardiac contractility in technology. The fluid is infused with the aid of 1993; 34:142–3. shock appear to be hypercarbia and hy- a roller pump from a minimal rate of 10 ml/hr 11. Milikan JS et al. Rapid volume replace- poxia.23,24 Clinically, if perfusion has been re- to a maximum of 1,500 ml/min. To our knowl- ment for hypovolemic shock: a compari- stored, oxygen delivery is adequate, and the edge, at present, no other infusing system can son of techniques and equipment. J patient is well ventilated, pH correction with deliver normothermic units at this rate. Trauma 1984; 24:428. exogenous bicarbonate is unnecessary. The use of the RIS has introduced new 12. Collins JA. Recent developments in the Conventional fluid warmers, such as those problems during resuscitation of trauma vic- area of massive transfusion. World J Surg in which fluid (crystalloid, colloid, or blood) tims. Although coagulopathies, hyperkalemia, 1987; 11:75–81. is passed within plastic tubing through heat- and hypocalcemia have been well described in 13. Canizaro PC, Pessa ME. Management of ing blocks or those in which the tubing is sub- the literature as rare phenomena, we have no- massive hemorrhage associated with ab- merged in warm water, are inefficient in deliv- ticed a high incidence of them after massive dominal trauma. Surg Clin North Am ering normothermic fluids at fast flow rates transfusions. As discussed previously, 1990; 70:621–34. (≥ 250 ml/min). With aggressive fluid resusci- coagulopathies and hypocalcemia are well 14. Patterson A. Massive transfusion. Int tation and blood transfusions, clinicians are known problems associated with rapid and Anesthesiol Clin 1987; 25:61–74. 15. Practice Guidelines for Blood Component confronted with five distinct problems: hypo- massive transfusions. Hyperkalemia is a rela- Therapy. A report by the American Soci- volemia, hypothermia, coagulopathy, hyper- tively new phenomenon. Its incidence is high ety of Anesthesiologists Task Force on kalemia, and hypocalcemia. Fluid warmers are when using flow rates of 500 to 1,000 ml/min. 18 Blood Component Therapy. Anesthesiol- designed to prevent and treat some of these As described by Jameson et al, for prevention ogy 1996; 84:732–47. problems. The H1000 infusion system (Sims of transfusion-associated hyperkalemia, our 16. Lovric V. Alterations in Blood components Level One Technologies, Inc., Rockland, Mas- practice is to use the Haemonetics Cellsaver during storage and their clinical significance. sachusetts) is a very effective fluid-warming blood salvage system in combination with the Anaesth Intensive Care 1984; 12:246–51. device. It consists of a cylindric aluminum heat RIS. The Cellsaver system is used not only to 17. Denlinger JK et al. Hypocalcemia during exchanger mounted on the warming unit and recycle blood from the surgical field but also, rapid blood transfusion in anaesthetized heated by a countercurrent water bath with a and more importantly, to wash the blood bank man. Br J Anaesth 1976; 48:995. set point of 42°C. To decrease heat loss even PRBCs before transfusion to the trauma victim. 18. Jameson LD et al. Hyperkalemic death more, a second device can be added— the Washing the PRBCs decreases the H+ and K+ during use of high-capacity fluid warmer Hotline warmer (Sims Level One Technolo- concentrations of the blood transfused and, in for massive transfusion. Anesthesiology gies)—on the 254-cm line between the H1000 our experience, decreases the incidence of se- 1990; 73:1050–2. and the patient. The central lumen of the in- vere transfusion-associated hyperkalemia. 19. Wilson RF et al. Electrolytes and acid-base travenous line is warmed by water circulating changes with massive blood transfusion. in a countercurrent direction. The countercur- References Am Surg 1992; 58:535–45. rent circulation water is warmed by a heated 1. Capan LM, Miller SM. Trauma and burns. 20. Murray DJ et al. Coagulation changes dur- reservoir, with a set point of 42°C. In Barash PG, Cullen BF, Stoelting RK, eds. ing packed red cells replacement of major Countercurrent water fluid warmers us- Clinical Anesthesia, 3rd edition. Philadel- blood loss. Anesthesiology 1988; 69:839. ing 42°C set points do not damage red cells, phia, Lippincott-Raven, 1997, pp 1173–204. 21. Schreiber GB, Busch MP, Kleinman SH, Korelitz JJ. The risk of transfusion-trans- deliver warm intravenous fluids, and allow the 2. Bickell WH, Wall MJ, Pepe PE, et al. Imme- mitted viral infections. The Retrovirus Epi- clinician to maintain thermal neutrality with diate versus delayed fluid resuscitation for demiology Donor Study. N Engl J Med respect to fluid management up to 400 ml/min. hypotensive patients with penetrating torso injuries. N Engl J Med 1994; 331: 1105–9. 1996; 334:1685–90. With flow rates above that, the infusion fluid 22. Downing SE et al. Influences of hypox- temperature will decrease slightly in propor- 3. Martin RR, Bickell WH, Pepe PE, et al. Pro- spective evaluation of preoperative fluid emia and acidemia on left ventricular function to the increase in flow rate. resuscitation in hypotensive patients with tion. Am J Physiol 1966; 210:1327–34. The H1000 infusion system is very useful penetrating truncal injury: a preliminary 23. Siegel HW, Downing SE. Contributions of for resuscitation of trauma victims, as it deliv- report. J Trauma 1992; 33:354–62. coronary perfusion pressure, metabolic ers warm fluid at rapid rates. It takes care of 4. Rutlege R et al. Massive transfusion. Crit acidosis and adrenergic factors to the re- hypovolemia and prevention of hypothermia Care Clin 1986; 2:791–805. duction of myocardial contractility during very well. Unfortunately, it is difficult to de- 5. Wudel JH et al. Massive transfusion: out- hemorrhagic shock in cats. Circ Res 1970: liver more than 800 ml/min with this infusion come in blunt trauma patients. J Trauma 27:875–89. system. To achieve infusion rates above this 1991; 31:1–7. 24. Prezlosi MP et al. Metabolic acidemia with level, our practice is to use the Rapid Infusion 6. Desjardins G, Varon AJ. Immediate hypoxia attenuates the hemodynamic re- System (RIS, Haemonetics). This device is ca- intrahospital management. In Abrams KJ, sponses to epinephrine during resuscitation pable of delivering 1,500 ml/min of blood Grande CM, eds. Trauma Anesthesia and in lambs. Crit Care Med 1993; 21:1901–7.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 37 13 Rapid Infusion and Point-of-Care Chemistry Testing in Massive Transfusion: Avoiding Common Pitfalls Jeffery R. Jernigan, MD specifically developed for volume infusion >500 dramatically improved ability to maintain cir- John G. D’Alessio, MD cc/min is the RIS. This device utilizes roller culating blood volume. The ease and efficiency Elvis Presley Memorial Trauma Center pumps that propel fluids from a 3-liter reser- with which these volumes are administered al- Memphis, Tennessee voir through two limbs of high-capacity tubing lows the anesthesia team to devote more men- at rates of up to 1,500 cc/min. The system also tal and physical energy toward other critical For multiple trauma patients with massive delivers 100-cc or 500-cc boluses over 1 minute. aspects of the case in progress. Further, with hemorrhage presenting for surgery, preopera- Additionally, there are three air detectors, which the RIS there is much greater flexibility in the tive efforts have been directed at stabilizing (or automatically stop the infusion in the event of rates of infusion. If one accepts the notion that at least temporizing) hemodynamic status. Ob- bubbles in the infusion path. While some plan- hypotensive resuscitation is desirable, this taining adequate intravenous access and infus- ning and a brief set-up period of 3 to 5 minutes would appear to be all the more reason to use ing crystalloid and packed red blood cells are are required, it was readily apparent that sig- the RIS in such a scenario. In addition, the abil- important measures in supporting circulating nificantly greater volumes of fluid could be in- ity of the RIS to arrest and reverse hypothermia blood volume. However, anesthesiologists are fused in a short time. However, this device is to the point of warming a cold patient to nor- still faced with precarious situations in which relative large and expensive, and it requires mothermia is significant and cannot be ignored. either all the above has taken place in the face maintenance of an adequate supply of dispos- Thus, it was evident that the RIS possesses of ongoing hemorrhage, or some stabilization able tubing/reservoir set-ups. several undeniably desirable characteristics. has occurred but surgical management will, of We currently employ both of these sys- Indeed, when one considers the five major necessity, entail increased blood loss. In either tems, the Level I System 1000 being the more problems encountered during massive trans- case, the clinical sequelae of hemorrhage and widespread of the two, with units in each op- fusion (hypovolemia, hypothermia, shock (such as acidosis, hypothermia, and erating room (OR), shock trauma admitting, coagulopathy, hyperkalmia, and hypocalce- coagulopathy) will begin to present at this point, and the intensive care unit. The combination mia), our experience has been that the RIS problems that become all the more difficult if of the two systems has proven very useful, the addresses hypovolemia and hypothermia effec- not managed early and effectively. Level I being used perioperatively, with the tively which, in turn, has beneficial effects in In this section, we will discuss our expe- option of large-volume infusions with the RIS dealing with acidosis and coagulopathy.2 How- rience in the clinical management of patients’ if need for massive transfusion arises in the ever, we concluded the increased risks of sig- problems regarding massive transfusion. This operating room. nificant hyperkalemia and hypocalcemia discussion is not intended to represent a de- Questions arose when we began using the needed to be addressed separately. finitive management protocol, since much RIS routinely in the OR. We found we had al- Since there is greater risk of physiologic debate continues about such topics as appro- tered the dynamics of blood administration in derangement in this setting, we felt a need for priate resuscitation strategies, desired clinical our trauma OR, in that we were no longer the closer monitoring of physiologic parameters end-points, and proper use of blood products. “rate-limiting step.” The blood bank raised by laboratory tests. To obtain turnaround times Rather, it is a “walk through” of the questions, concerns regarding appropriate use of blood faster than the hospital laboratory could pro- trials, and decision-making processes that have products and maintenance of an adequate sup- vide, we considered point-of-care testing de- led us to our current use of the Rapid Infusion ply of these valuable resources. Additionally, vices. Point-of-care testing has gained favor in System (RIS) (Haemonetics Corporation, some of our surgical colleagues expressed con- recent years, one example being glucometers Braintree, Massachusetts) in conjunction with cern about striving for normotension with ag- developed for home use, which enable diabet- point-of-care chemistry-testing devices in man- gressive fluid administration and the effects this ics to monitor their glucose levels. Newer tech- aging these difficult problems. may have on hemostatis. These issues were a nologies have expanded this concept into When our trauma center opened in 1983, direct result of our dramatically increased abil- other areas involving a variety of laboratory pa- we were using standard pressure bags con- ity to infuse large volumes. rameters relevant to intensive care and surgi- nected to a pneumatic pump with six outlets Other questions arose regarding some of cal settings. and infusing fluids through a separate blood the problems well known to be associated with After discussion with our laboratory direc- warmer. Although this approach was adequate massive transfusion,1-5 which are discussed tor, we chose the i-STAT Portable Clinical Ana- for several years, we were constantly struggling elsewhere in this monograph. We noted clini- lyzer (i-STAT Corp, Princeton, New Jersey). This with problems of acidosis, hypothermia, and cally significant hyperkalemia on at least one device is hand-held and battery powered and coagulopathy in the face of ongoing and, occa- occasion. Such related complications previ- comes with a portable printer. It is easy to use sionally, exsanguinating hemorrhage. Therefore, ously thought to be infrequent were now more and relatively inexpensive and provides reliable we began looking for ways to improve our abil- likely to be encountered as infusion capability accurate results in 2 minutes. The system em- ity to keep up with massive hemorrhage. increased.1 ploys a “thin film” biosensor housed in a small We initially considered the fluid-warming As we worked through these issues, Hambly cartridge. Two to three drops of blood are placed pressure infusers manufactured by Level I and Dutton concluded that using the RIS was into the cartridge, which is inserted into the ana- (Level I Technologies, Rockland, Massachu- associated with increased mortality. They also lyzer. The lab values obtained depend on the setts). This system consisted of two pressure asked the question (raised by others7–11) whether particular cartridge used. There are several types infusers connected to a blood warmer we had hypotensive resuscitation may be advantageous available. The cartridge we use measures sodium, already been using. The advantages of this sys- in this setting. This followed the article by Dun- potassium, ionized calcium, arterial blood gases, tem were ease of use and portability. However, ham and associates,12 which showed a positive hematocrit, and hemoglobin. There were early only two pressurized bags could be connected outcome associated with fluid administration concerns about the biosensor technology regard- to this system at any one time. Infusion rates through the RIS. These considerations led to a ing manufacturing and failure rate. We have had were comparable to or slightly faster than the reassessment of our use of the RIS. no problems in these areas. However, the i-STAT pneumatic pumps used previously (approxi- Despite the problems we encountered, we does not provide point-of-care testing for coagu- mately 500 cc/min). felt there were distinct advantages in using the lation studies, so we continue to send these to The only commercially available system RIS. The primary, overriding advantage is the our trauma laboratory.

38 Massive Transfusion and Control of Hemorrhage in the Trauma Patient In addition to federally mandated quality * We aim for a hematocrit in the low to trol of blood loss was obtained. This would not assurance guidelines, there are a number of mid-20s. appear to be applicable to intraoperative use of point-of-care testing guidelines, which vary the RIS as studied by Hambly and Dutton, since from state to state. Federal guidelines were set * FFP are infused through the RIS in addi- their patients were, presumably, resuscitated in forth in the Clinical Laboratory Improvement tion to the NS. the usual fashion prior to and after arrival at Act of 1967 and amended in 1988. The cur- the Shock Trauma Center. Before conclusions rent rules and regulations are referred to as * Platelets are infused separately. can be drawn regarding the appropriateness the CLIA ’88 (Clinical Laboratory Improvement and timing of use of the RIS, more uniformity Amendments of 1988). They divide laboratory * With each five units of packed cells given between Bickell’s and Dutton’s patients would tests into three categories: 1) waived (no spe- in 15 minutes or less, 1 gram of CaCl2 is have to be demonstrated. cial qualifications to run tests); 2) moderately given. Unanswered questions remain, along with complex (requires high school diploma); and the need for further controlled, well-focused 3) highly complex (requires an associate de- * Labs are repeated after each 10 units studies. Whatever strategy is employed during gree in laboratory science). The federal gov- PRBCs. fluid resuscitation of the trauma patient and ernment may inspect, fine, and even close fa- massive transfusion, it is important to remem- cilities found not to be in compliance.13 An * Hyperkalemia (>6.0) is treated with 10 ber to treat each patient individually, globally, institution that performs laboratory tests is units regular insulin with D5W. and according to clinical judgment rather than responsible for compliance regardless of where by strict protocol. Use of the RIS together with within the facility that testing is done. In addi- * Acidosis is treated with volume infusion point-of-care testing and improved communi- tion, four states (California, Florida, New York, and sodium bicarbonate as deemed ap- cation with blood bank personnel, laboratory and Tennessee) require that anyone not a cer- propriate. personnel, and surgeons improves our ability tified medical technologist (including MDs and to manage trauma patients requiring massive CRNAs) must be granted a waiver in order to * Cryoprecitipate is given based on fibrino- transfusion. run lab tests. Thus, in order to avoid these gen levels. types of problems, we recommend consulting References the lab director of your institution if one of * We continue to strive to maintain a rela- 1. Jameson LC et al. Hyperkalemic death these devices is being considered. tively normotensive state in this setting. during use of a high-capacity warmer for Another option available in avoiding com- Communication with the surgical team, massive transfusion. Anesthesiology 1990; plications of massive transfusion is washing red monitoring of urine output, and consid- 73:1050–2. blood cells (RBCs) prior to infusion. Storage eration of cerebral perfusion help guide 2. Ferrara A et al. Hypothermia and acidosis of packed red blood cells (PRBCs) results in decisions regarding target pressures. worsen coagulopathy in the patient requir- accumulation of potassium over time.14 Wash- ing massive transfusion. Am J Surg 1990; ing RBCs prior to administration removes As to the controversies concerning hypoten- 160:515–8. much of this potassium as well as a significant sive resuscitation, use of the RIS in this scenario, 3. Phillips GR et al. Massive blood loss in proportion of existing citrate, which, in some and possible increased mortality associated with trauma patients: the benefits and dangers cases, can result in hypocalcemia and cardio- its use, close examination of the pertinent litera- of transfusion therapy. Post-Graduate vascular depression.15 The removal of these ture led us to the following analysis. Medicine: Transfusion Therapy 1994; agents can preempt some of the problems as- The conclusions reached in the study by 95(4):61–70. sociated with massive transfusion. This option Hambly and Dutton are clouded by two prob- 4. Hamilton SM. The use of blood in resus- has been employed successfully in a variety of lems. First, selection bias may have played a citation of the trauma patient. Can J Surg clinical settings.1,16,17 We do not perform this significant role, as noted by the authors. Sec- 1993; 36(1):21–7. routinely, except when treating patients with ond, their findings are predicated on a com- 5. Wilson RF et al. Electrolyte and acid-base a history of renal insufficiency. parison of expected versus observed mortality changes with massive blood transfusions. In using the RIS in conjunction with the i- between the study groups. They defined ex- Am Surg 1992; 58(9):535–45. STAT, we employ the following strategy when pected mortality in this population based on a 6. Hambly PR, Dutton RP. Excess mortality massively transfusing a patient: logistic regression equation published by Dun- associated with the use of a rapid infusion ham et al from their institution in 1986.18 This system at a level I trauma center. Resusci- * When the decision is made to use the RIS, equation was written as a statistical descriptor tation 1996; 31:127–33. we notify the blood bank than an RIS case of observed events at that institution, not as a 7. Bickell WH et al. Immediate versus de- is starting. The blood bank then sets up predictor of mortality. They state, “To ensure layed resuscitation of hypotensive patients what are termed “RIS units” consisting of validity of the equation used to determine the with penetrating torso injuries. N Engl J 10 units PRBCs, 4 units FFP, and 4 units of probability of death, a prospective assessment Med 1994; 331:1105–9. platelets (not to be infused through the needs to be performed on another popula- 8. Bickell WH. Are victims of injury some- RIS). The blood bank continues to hold tion.” A search of the literature and conversa- times victimized by attempts at resuscita- one of these units until informed by us tions with the author have not revealed such a tion? Ann Emerg Med 1993; 22:225–6. that we are no longer in a massive trans- study. Therefore, the applicability of this equa- 9. Bickell WH et al. Intravenous fluid admin- fusion mode. tion in predicting mortality in this population istration and uncontrolled haemorrhage. must be questioned. Such an equation or simi- J Trauma 1989; 38:227–33. * Baseline labs are drawn, consisting of ar- lar predictive tool remains elusive. 10. Stem A et al. Effect of blood pressure of terial blood gasses, complete blood count, Regarding the question of hypotensive haemorrhagic volume in a near-fatal PT/PTT, fibrinogen, potassium, and ion- resuscitation, it should be noted that the study haemorrhage model incorporating a vas- ized calcium. by Bickell et al deals with penetrating trauma, cular injury. Ann Emerg Med 1993; whereas Hambly and Dutton raise this issue in 22:155–163. * In filling the pump reservoir, PRBCs are their study on blunt trauma patients. Further, 11. Capone A et al. Treatment of uncontrolled diluted with 500 cc normal saline per unit. Bickell found increased survival with minimal haemorrhagic shock: improved outcome resuscitation prior to, not in, the operating with fluid restriction. J Trauma 1993; room, and full resuscitation once surgical con- 35:984.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 39 12. Dunham CM et al. The Rapid Infusion blood transfusion during pediatric liver 17. Ramsay AE, Swygert TH. Anesthesia for System: a superior method for the resus- resection. Surgery 1986; 99(5):664–9. hepatic trauma, hepatic resection and liver citation of hypovolaemic trauma patients. 15. Westphal RG. Special topics. In Westphal transplantation. Balliere’s Clinical Anes- Resuscitation 1991; 21:207–27. RG, ed. Handbook of Transfusion Medi- thesiology 1992; 6:863-94. 13. Passey RB. CLIA ’88 penalities and how to cine, 3rd ed. The American Red Cross, 18. Dunham, CM, Cowley R, Gens DR, et al. avoid them. In Coping with CLIA: An 11- 1996; p 106. Methodologic approach for a large func- Part series. Medical Laboratory Observer. 16. Kang YG. Hemodynamic instability during tional trauma registry. Md Med J 1989; Medical Economics Publishing, June 1993. liver transplantation. TransProc 1989; 38:227–33. 14. Estrin JA et al. A new approach to massive 21(3):3489–92.

SECTION IV: New Horizons in Synthetic Blood Substitutes 14 Hemoglobin-Based Oxygen-Carrying Solutions & Hemorrhagic Shock

Colin F. Mackenzie, MB, ChB, FRCA, FCCM hemoglobin and stimulation of the reticuloen- Director, National Study Center for Trauma dothelial system and macrophages have not Table 2. and Emergency Medical Systems been resolved. Leland Clark demonstrated that Currently Available Products University of Maryland School of Medicine a mouse could survive while breathing liquid That Can Be Used As Oxygen-Carrying Baltimore, MD 21201 USA perflurocarbons saturated with oxygen.5 In Solutions in Humans e-mail: [email protected] 1972, Benesch discovered reagents that could bind the 2,3-DPG binding site so that they Whole blood [Editors’ note: Dr. Mackenzie receives grant could reduce hemoglobin affinity for oxygen. Liquid red cells support from Biopure Corporation and The most widely used agent is pyridoxal, 5 PO 4 Frozen red cells Anjinomoto Corporation.] (so-called pyridoxalation), which is used to reduce oxygen affinity.6 The normal P5O (the Lyophilized red cells There has been only one reported use, in partial pressure of oxygen when Hb is 50% Free hemoglobin 1949, of a hemoglobin solution for resuscita- saturated) of blood is 26.7 mmHg. P5O is in- Encapsulated hemoglobin tion of a human in hemorrhagic shock.1 A creased by pyridoxalation. Human stroma-free Perfluorocarbons woman suffering from postpartum hemorrhage hemoglobin has a P5O of 12 to 15 mmHg and was given 2.3 liters of 9% hemoglobin solution therefore has a high oxygen affinity and tends in saline after all available compatible blood had to hold onto the oxygen rather than give oxy- (Table 3) include vasoactivity, with binding of been given. Consciousness returned, her blood gen up at the tissue level. nitric oxide by free hemoglobin being the main pressure rose, and her heart rate fell. However, suspect causing vasoconstriction.7 Nephrotox- the patient died 9 days later from renal failure. General Properties icity from stromal remnants is probably of only Attempts to develop blood substitutes go Red cells can be stored in liquid form with historical interest, because better purification back many hundreds of years2 (Table 1). In citrate phosphate dextrose adenine (CPDA) techniques have resulted in lack of renal tox- 1916, hemoglobin solutions were given in anticoagulant for 35 days and in AS-1 for 42 icity with newer hemoglobin-based oxygen small quantities to 33 subjects to determine days. They can also be frozen after addition of carriers.8 In human volunteers given recombi- the renal threshold for hemoglobin without glycerol to prevent lysis or they can be instantly adverse effects. Many studies, however, using freeze-dried or lyophilized. Oxygen-carrying larger quantities of hemoglobin solutions, had solutions (Table 2) may consist of free hemo- Table 3. adverse effects, including hypertension, brady- globin from which the stroma or cell wall has Toxicities and Interferences of cardia, oliguria, and anaphylaxis.3 In 1957, been removed, or liposome-encapsulated he- Hemoglobin-Based Oxygen-Carrying Chang encapsulated hemoglobin,2 and since moglobin containing hemoglobin with a syn- Solutions then development of liposome-encapsulated thetic membrane. Perfluorocarbons are organic hemoglobin has continued. The problems as- solutions with high oxygen solidity. Vaso-activity sociated with disposal of the encapsulated Toxicities of free hemoglobin solutions Interference with mononuclear phagocyte system Antigenicity Table 1. History of Transfusion and Oxygen-Carrying Solution Use Oxidation to methemoglobin Activation of complement, kinin, 1667 First human blood transfusion (Denis), causing death and moratorium and coagulation 1863 Gum-saline transfusion (Ludwig) Thrombocytopenia Red cell and platelet aggregation 1916 Hemoglobin infusion in humans (Sellards and Minot) Histamine release 1941–45 Albumin and hemoglobin infusion Fever, chills, gastrointestinal upset, 1957 Encapsulated hemoglobin described (Chang) headache, backache 1966 Perfluorocarbon “bloodless mouse” (Clark and Gollman) Iron deposition 1969 Amberson’s report of hemoglobin infusion in human hemorrhagic shock Binding nitric oxide 1972 Pyridoxalation to reduce hemoglobin affinity (Benesch et al) Colorimetric interference, pulse and fiberoptic oximetry 1978 Human safety trial, unmodified hemoglobin (Santsky et al) Interference with liver function, 1980–97 Human trials with human and bovine hemoglobin-based solutions blood compatibility, and Human trials with second-generation perfluorocarbons chemical testing

40 Massive Transfusion and Control of Hemorrhage in the Trauma Patient nant hemoglobin, 0.23 g/kg, there was no evi- prolong intravascular retention time but also in hemorrhagic shock? The objectives of suc- dence of nephrotoxicity. Immunologic effects make the hemoglobin more rigid. The polymer- cessful resuscitation from hemorrhagic shock of hemoglobin-based oxygen-carrying solu- ization reaction is very difficult to control, so include 1) restoration of intravascular pres- tions remain somewhat of an unknown. In fact, there is some batch-to-batch variability. An al- sures, 2) increase in cardiac output, and 3) the immunologic effects of blood transfusion ternative to polymerization is conjugation to a reversal of the increased oxygen extraction that have been extensively explored only recently. larger molecule, and this also prolongs reten- occurs in hemorrhagic shock. When studies Interferences occur with free hemoglobin so- tion time. Some solutions can be polymerized using red cell substitutes to achieve the first lutions (Table 3). Use of hemoglobin-based and conjugated. Intravascular retention time two of these objectives are examined, difficul- oxygen-carrying solutions interferes with can be prolonged from 7 hours in the unmodi- ties in interpretation occur. The protocol and fiberoptic oximetry because of the red color.9 fied form to about 36 hours after modification. animal model can influence the judgment of Mixed venous oxygen saturation is overesti- The hemoglobin can be incorporated into efficacy. In one study in which a hemoglobin mated at low levels of 60% to 70%—a danger- an artificial cell, and liposome encapsulation solution was tested, the protocol specified that ous situation that may cause patients to be is currently under study. However, the lipo- fluid resuscitation should be given to restore underresuscitated. The interference is nonlin- somes cause substantial drops in platelet cardiac filling pressures to baseline values.16 If ear, as it overestimates oxygen saturation at counts, and during excretion, they block the a vasoconstrictor response occurred with in- high venous oxygen tension. Pulse oximetry reticuloendothelial system.4 A hemoglobin so- fusion of the hemoglobin-based oxygen-carry- interference occurs because of methemoglo- lution that has been studied in hemorrhagic ing solution, it would appear very efficacious bin.10 Hemoglobin-based oxygen-carrying so- shock is a pyridoxalated hemoglobin at restoring vascular pressures. In addition, the lutions make it impossible to carry out some polyoxyethylene conjugate made from stroma- evidence for a vasoconstrictor response would liver function tests such as alkaline phosphate free hemoglobin by conjugation with be minimized. Furthermore, if an awake dehy- measurement11 and coagulation tests such as polyoxethylene to increase its half-life from 7 drated pig model had been used instead of a partial thromboplastin time.12 They can also to 36 hours and by pyridoxalation to increase dog, as other studies have shown, the animal interfere with cross-matching, but this can be P5O from 15 to 20 mmHg. Maltose is added to may have died as a result of the hemoglobin- overcome with dilution. prevent oxidation to methemoglobin.15 based oxygen-carrying solution causing pro- The stimulus for all recent activity in de- found vasoconstriction and reduced cardiac Methods to Prevent Complications of velopment of hemoglobin-based oxygen-car- output.17 Oxygen-Carrying Solutions rying solutions is reduction of disease trans- If cardiac output and arterial blood pres- For many reasons, including avoidance of mission, particularly of human immunodefi- sure changes during resuscitation with oxygen- human disease transmission, sources other ciency virus (HIV) and hepatitis virus. From the carrying solutions are examined, confounding than outdated human blood have been used perspective of the manufacturers of oxygen- data are also obtained. In two studies, cardiac to produce hemoglobin solutions. Transgenic carrying solutions, there is much interest be- output or blood pressure was less with hemo- pigs and mice have been bred to produce hu- cause it is estimated to be a potential $12 bil- globin solution infusion than with autologous man hemoglobin, and recombinant hemoglo- lion a year industry. Their use in hemorrhagic blood transfusion.18,19 In these four studies, bins can be produced from bacteria and yeast shock is important because huge quantities of cardiac output and arterial pressure changes by modifications that incorporate globin genes. blood are currently used for this. In 1993, at were no different with hemoglobin solution For example, it is possible to express both the Shock Trauma Center at the University of and blood resuscitation.20–23 Only one study human a and b globin chains in Escherichia Maryland, 1,300 patients were given 8,500 showed that the hemoglobin solution sus- coli; however, the yields are still very low. units of blood, an average of 6.5 units per pa- tained oxygen transport at higher levels than About 750 liters of cell culture would be tient, or about 50% to 60% of blood volume did non-oxygen-carrying solution volume ex- needed to produce 1 unit of blood. Endotoxin replacement. The potential for replacing some panders such as albumin or lactated Ringer’s contamination may also occur.2 of this blood use with an alternative is very solution.16 Transient cardiac output and blood Sources of hemoglobin other than human enticing for the manufactures of oxygen-carry- pressure increases were greater half an hour include bovine hemoglobin. In addition, any of ing solutions and is also of interest to the Red after resuscitation began with hemoglobin so- these hemoglobins can be modified to optimize Cross, which goes to great efforts to maintain lution resuscitation compared with autologous their characteristics such as retention time; oxy- this vital supply. blood reinfusion in another study.15 So there gen affinity, reduction of dimer conversion into are no clear-cut data showing what effects he- tetramers, and prevention of oxidation to meth- Vascular and Other Physiologic Effects moglobin solutions in general have on arte- emoglobin. Bovine hemoglobin has a high P5O of Hemoglobin-Based rial pressure and cardiac output, nor is there without modification and is therefore of inter- OxygenSubstitutes much information showing they are conclu- est since it is also in plentiful supply.13 How do we judge whether hemoglobin- sively more beneficial than non-oxygen-carry- Because of osmotic effects, most hemoglo- based oxygen-carrying solutions are efficacious ing volume expanders. In some studies, oxy- bin-based oxygen-carrying solutions are in con- gen transport was significantly impaired com- centrations no greater than 7 to 8 g/dl. pared with whole blood because of a fall in Perfluorocarbons have a linear oxygen dissocia- Table 4. Characteristics of Some hematocrit,15 whereas in other studies oxygen tion curve, and a relatively high oxygen content Clinically Used Hemoglobin-Based transport is no different than with autologous of 50% or more is required for them to carry Oxygen-Carrying Solutions blood resuscitation.20–23 equivalent amounts of oxygen to hemoglobin. How can oxygen-carrying solutions have The second-generation perfluorocarbons 1. Stroma-free hemoglobin (SFH): added value over blood as a means of deliver- (Perflubron) have more efficient oxygen carriage, simple removal of the cell wall ing oxygen to tissues? There are several po- even breathing 50% oxygen, whereas the first- Human and bovine SFH tential ways, some of which have been con- generation (Fluosol) required 100% oxygen 2. Modifications include firmed by experiments in animals. Because breathing to achieve even one-fourth the oxy- a. Cross-links oxygen-carrying solutions are acellular, they gen carriage of blood.2 (alpha-alpha and beta-beta) are less viscous than blood and flow more eas- Hemoglobin may be modified by polymer- b. Polymerization ily through narrow vessels and the microcir- ization14 (Table 4). Polymerized hemoglobin is c. Configuration culation. It is therefore possible that oxygen- produced by addition of reactive groups to the d. Encapsulation carrying solutions may be useful in hemor- surface of hemoglobin. These reactive groups rhagic shock. There is experimental evidence

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 41 that hemoglobin-based oxygen-carrying solu- were examined with resuscitation from severe plasma. None was identified by hemoglobin tions can enhance oxgyen diffusion from the hemorrhagic shock in dogs.24 The solutions electrophoresis (which can clearly distinguish vascular to the intracellular space.20 In addi- used were 8% pyridoxalated hemoglobin the two types of hemoglobin). In addition, tion, when compared with whole autologous polyoxyethylene conjugate and 8% maltose, measurements of plasma hemoglobin gave blood, a hemoglobin-based oxygen carrier pre- known as PHP88, a 4% solution of the same values consistent with the quantities of hemo- served exercise capacity in humans. Diffusion solution made by diluting PHP88 with equal globin-based oxygen-carrying solutions in- of carbon monoxide across the alveolar-capil- volume of Plasmalyte A (PHP44), and stroma- fused. Red cell counts show the same picture lary membrane (DLCO) and blood lactate lev- free hemoglobin (SFH), a simple non-conju- as Hct. These data strongly suggest that cells els were measured during exercise in hu- gated hemoglobin solution. Both hemoglobin were being removed from the circulation. It mans.25 There was a greater oxygen uptake and solutions were highly purified and endotoxin was postulated that they may be sequestered for CO2 production and normal lactate levels free. Use of these three hemoglobin solutions in circulatory beds as a result of endothelial or were maintained in those given hemoglobin- was compared with re-infusion of autologous other interactions. based oxygen carriers in comparison with sub- blood. The volume of blood removed to pro- Why thrombocytopenia occurred and jects given autologous blood. Infusion of 1 duce 2 hours of shock was 63% of the estimated why there was a reduction in all other cellu- gram of hemoglobin-based oxygen carriers in- blood volume. Resuscitation began with flu- lar components remains the cause of much creased DLCO as much as 3 grams of autolo- ids infused at 20 ml/min by infusion pump; in speculation and investigation. Many factors gous blood. four dogs, no resuscitation was given. Samples are known to give rise to platelet adhesive- In addition to the short-term benefits of for coagulation and hematology profiles and a ness and rouleaux formation, including re- enhanced diffusion, hemoglobin-based oxy- blood smear were taken one-half hour after lease of thromboxane, and endothelial reac- gen carriers may have longer lasting effects, resuscitation began, when all the hemoglobin- tions, including binding of nitric oxide by free as it has been shown that serum iron, fer- based oxygen-carrying solutions were infused hemoglobin; the presence of free heme also ritin, and erythropoietin increase in parallel or, in the case of non-resuscitated dogs, no induces platelet aggregation. Hemodilution with plasma levels of hemoglobin. The iron additional fluids were given. Measurements is another important factor causing thromb- infusion adds the equivalent of one unit of were repeated at 2, 4, and 6 hours after resus- ocytopenia, as this was a severe hemorrhagic blood transfusion within 1 week of hemo- citation, and then daily for 7 days after awak- shock model in which 63% of the circulating globin infusion.25 ening from anesthesia. blood volume, and therefore cellular compo- Newly discovered allosteric and electric All dogs not resuscitated died within 2 nents of the blood, were removed from the properties of hemoglobin appear to control hours. All autologous blood and PHP44 dogs circulation. In addition, the high colloid blood pressure and may facilitate tissue oxy- survived 8 days, while mortality among PHP88 oncotic pressure of these hemoglobin solu- genation. S-Nitrous-hemoglobin (SNO-Hb) is dogs was 63% and among SFH dogs, 14%. tions may have further accentuated the circu- free of vasoactivity and may be a route for effi- Clinical coagulopathy occurred in all dogs lating volume increase and dilution of cellu- cient delivery of nitric oxide to the mitochon- given PHP88 and in four of the six dogs given lar components.15 It was speculated that, as a dria. Nitric oxide controls mitochondrial res- SFH, and there was evidence of hematoma for- result of platelet aggregation and rouleaux piration. Cell-free SNO-Hb may be a good he- mation around cannulation sites in all six dogs formation, platelets and red cells are trapped moglobin-based oxygen-carrying solution.7. given PHP44 and five of the six dogs given au- in the microcirculation and this sequestration Other properties besides oxygen transport tologous blood when autopsy was performed. prevents their subsequent employment in affect assessment of efficacy of red cell substi- Clinical coagulopathy with spontaneous devel- coagulation and oxygen transport. A factor tutes. Profound increases in pulmonary pres- opment of oozing from percutaneously placed that may additionally or singularly be the sures can cause fatalities in some animals and cannulae, spontaneous development of he- cause of the problem is the polyoxyethylene prevented any benefit from being realized due matomas in the femoral areas where catheters moiety attached to hemoglobin in PHP. It is to hemoglobin solution infusion.17 When were placed, and in some dogs receiving both used to increase molecular size and prolong changes in pulmonary artery pressure are com- PHP and SFH petechiae were visible subcuta- vascular retention time. However, it may also pared after resuscitation from hemorrhagic neously all over the body, and submucosally cause electrostatic charges that increase the shock with oxygen-carrying solutions, in the mouth. In the dogs that died, exsan- likelihood of platelet aggregation and red cell interspecies differences as well as protocols guination was the major cause of mortality rouleaux formations. Mediators released dur- and models become confounding variables. secondary to thrombocytopenia. Death oc- ing hemorrhagic shock are probably an im- The rise in pulmonary artery pressure after curred between 7 and 254 hours after infusion portant determinant of the cell aggregation resuscitation in the swine model greatly ex- of the hemoglobin solution. seen with infusion of PHP. No coagulopathy ceeds that seen in the dog.17 In some studies, There was a fall in hematocrit (Hct) in all or mortality was found in dogs undergoing fluid resuscitation is given with the objective animals resuscitated with these cellular fluids. exchange transfusions with PHP of 80% of of returning filling pressures to baseline val- However, the fall in animals given PHP88 was blood volume or in nonvolemic dogs given ues, so that if a vasoconstrictor response oc- significantly greater than in those receiving the 20 ml/kg of PHP. Changes that occur during curred with infusion, the protocol used would other solutions, with an average Hct of 3% af- hemorrhagic shock exacerbate the effects of prevent this difference from becoming appar- ter resuscitation.15 Since 63% of the estimated large doses of PHP. ent.20 Several investigators have also noted blood volume was removed and because he- The whole issue is very complex. An ob- thrombocytopenia after infusion of red cell matocrit was, on average, about 40% before vious possible mechanism of platelet aggre- substitutes, and clearly it is critical that red cell resuscitation, it was expected on the basis of gation, namely, binding of nitric oxide by free substitutes for use in the management of hem- hemodilution alone, that Hct would be about hemoglobin, has not been excluded. Free orrhagic shock should not interfere with resi- 25%. The finding that Hct was between 9% heme can cause platelet aggregation, as can dent blood cells or the coagulation system, as and 11% with PHP44 and SFH suggests that thromboxane release secondary to hemor- these toxicities would preclude their use in the these hemoglobin solutions also had some rhagic shock. The coagulopathy that occurred management of patients with trauma or those adverse effect on red cells. The possibility of with PHP may be a combination of some or undergoing surgery. hemolysis occurring was explored by hemo- all of these mechanisms. globin electrophoresis of the plasma samples— Hemostatic Effects the PHP and SFH were both derived from hu- Conclusion The effects of free hemoglobin solutions man hemoglobin. If hemolysis had occurred, The studies discussed illustrate some very on coagulation and blood cellular components canine hemoglobin would be found in the important facts about the data that are avail-

42 Massive Transfusion and Control of Hemorrhage in the Trauma Patient able on oxygen-carrying solutions. First, there real organ perfusion, and cardioplegia. Poten- 14. Winslow RM. Hemoglobin-Based Red Cell is virtually no published data on use of any of tial future uses also include transfusion alter- Substitutes. Baltimore, Johns Hopkins these products in humans for resuscitation native in patients with red cell incompatibili- University Press, 1992, pp 72–95. from hemorrhage shock. Second, much of ties. For Jehovah’s Witnesses, perfluorocarbon, 15. Sprung J, Mackenzie CF, Barnas GM, et the evidence for the oxygen-carrying solu- but not hemoglobin-based oxygen-carrying al. Oxygen transport and cardiovascular tions currently under study in humans un- solutions, are an acceptable alternative to effects of resuscitation from severe hem- dergoing Phase 2 and Phase 3 trials to ob- blood transfusion. orrhagic shock using hemoglobin solu- tain FDA approval is proprietary. As a result, tion. Crit Care Med 1995; 23:1540–53. the data in the literature may not be scien- References 16. Harringer W, Hodakowski GT, Svizzero T, tifically valid, as adverse effects may be mini- 1. Amberson WR, Jennings JJ, Rhode CM. et al. Acute effects of massive transfusion mized. Third, there are interspecies varia- Clinical experience with hemoglobin-sa- of a bovine hemoglobin blood substitute tions, so that toxicities or benefit seen with line solutions. J Appl Physiol 1949; in a canine model of hemorrhagic shock. a product in animal studies may not trans- 1:469–89. Eur J Cardiothorac Surg 1992; 6:649–53. late into reality in human studies. Fourth, 2. Winslow RM. Hemoglobin-Based Red Cell 17. Hess JR, MacDonald VW, Brinkley WW. endothelial interactions are still not com- Substitutes. Baltimore, Johns Hopkins Systemic and pulmonary hypertension pletely understood and could result in ad- University Press, 1992, pp 1–16. after resuscitation with cell free hemo- verse reactions to oxygen-carrying solutions 3. Sellards AW, Minot GR. Injection of he- globin. J Appl Physiol 1993; 74:1769–78. that preclude their use in shock. Fifth, me- moglobin in man and its relation to blood 18. Nho K, Glower D, Bredehoeft S, et al. diators released during reperfusion or con- distribution, with especial reference to PEG-bovine hemoglobin: safety in a ca- ditions that develop in hemorrhagic shock the anemias. J Med Res 1916; 34:469–94. nine dehydrated hypovolemic-hemor- may accentuate the toxicities of hemoglobin- 4. Rabinovici R, Rudolph AS, Vernick J, et rhagic shock model. Biomat Art Cells based oxygen-carrying solutions, since ad- al. A new salutary resuscitative fluid: li- Immobilization Biotechnol 1992; ministration of similar quantities of one he- posome encapsulated hemoglobin/hy- 20:511–24. moglobin-based oxygen-carrying solution pertonic saline solution. J Trauma 1993; 19. Ning J, Anderson PJ, Biro GP. Resuscita- (PHP) to animals not in shock does not re- 35:121–7. tion of bled dogs with pyridoxalated-po- sult in coagulopathy or mortality. Sixth, mi- 5. Clark LC, Gollman F. Survival of mammals lymerized hemoglobin solution. Biomat nor changes among several potential modi- breathing organic liquids equilibrated Art Cells Immobilization Biotechnol fications can significantly alter the toxicities with oxygen at atmospheric pressure. Sci- 1992; 20:525–30. of oxygen-carrying solutions. ence 1996; 152:1755–6. 20. Teicher RA et al. Oxygenation of tumors What then is the future of hemoglobin and 6. Benesch RE, Benesch R, Renthal RD, by a hemoglobin solution. J Cancer Res perfluorocarbon-based oxygen-carrying solu- Maeda N. Affinity labeling of the Clin Oncol 1993; 120:85–90. tion? From news through the proprietary grape- polyphosphate binding site of hemoglo- 21. Marks DH, Lynett JE, Letscher RM, et al. vine, it appears that an equivalent of a two-unit bin. Biochemistry 1972; 11:3576–82. Pyriodoxalated polymerized stroma-free transfusion of oxygen-carrying solution is well 7. Jai L et al. S. Nifroso Hb. A dynamic activ- hemoglobin solution (SFHS-PP) as an tolerated by the majority of individuals when ity of blood involved in vascular control. oxygen-carrying fluid replacement for given in elective surgical circumstances. The side Nature 1996, 380:221–6. hemorrhagic shock in dogs. Milit Med effects are relatively minor, including gas- 8. Viele MK, Weiskopf RB, Fisher D. Recom- 1987;152:265–71. trointestinal upset, musculoskeletal aches, and binant human hemoglobin does not af- 22. Nees JE, Hauser CJ, Shippy C, et al. Com- headache. One worrisome side effect that is fect renal function in humans. Anesthe- parison of cardiorespiratory effects of rumored has been the development of pancre- siology 1997; 86:848–58. crystalline hemoglobin, whole blood, al- atitis or signs of pancreatic changes seen in a 9. Kang LS, Ryder IG, Kahn R, et al. In vitro bumin, and Ringer’s lactate in the resus- very few individuals receiving some hemoglo- oxyhemoglobin saturation measure- citation of hemorrhage shock in dogs. bin-based oxygen-carrying solutions. Another ments in hemoglobin solutions using Surgery 1978; 83:639–47. worrisome issue was the indefinite postpone- fiberoptic pulmonary artery catheters. Br 23. Greenberg AG, Schooley M, Ginsburg KA, ment of a Phase 3 trial of a hemoglobin-based J Anaesth 1995; 74:201–8. Peskin GW. Pyridoxalated stroma-free he- oxygen solution in patients with trauma and 10. Barker SJ, Tremper KK, Hyatt J. Effects of moglobin in resuscitation of hemorrhagic hemorrhagic shock, presumably because of methemoglobinemia on pulse oximetry shock. Surg Forum 1978; 29:44–6. increased mortality in the study group (Wall and mixed venous oximetry. Anesthesi- 24. Mackenzie CR, Parr M, Christenson R, et Street Journal, February 6, 1998). The data ology 1989; 70:112–7. al. The effect of free hemoglobin solu- that led to the action have not been made 11. Bucci E, Fronticelli C, Razynska A, tions on coagulation and hematology af- public to date. Urbaitis B. Overview of chemically ob- ter resuscitation from severe hemor- Other potential future uses include man- tained oxygen carriers from hemoglobin: rhagic shock in dogs. Anesthesiology agement of ischemic disease and angioplasty. pseudo cross-linked tetramers. Biomater 1993; 79(suppl):A269. The acellular oxygen-carrying fluids have very Artif Cells Artif Organs 1989; 17:637–9. 25. Hughes GS Jr, Yancey EP, Albrecht R, et favorable rheologic properties and enhance 12. Eldridge J, Russell R, Christianson R, et al. Hemoglobin-based oxygen carrier pre- mitochondrial oxygenation. In some tumors, al. Liver function and monorphology fol- serves submaximal exercise capacity in radiosensitivity is increased by means of in- lowing resuscitation from severe hemor- humans. Clin Pharmacol Ther 1995; creased oxygen levels. A further potential use rhagic shock with hemoglobin solutions 58:434–43. of oxygen-carrying solutions is as an adjunct or autologous blood. Crit Care Med 1996; 26. Hughes GS Jr, Francome SF, Antal EJ, et to radiation therapy for certain tumors. In 24:663–71. al. Hematologic effects of a novel hemo- sickle cell crisis, perfusion and oxygenation 13. Alonsozana GL, Elfarth MD, Mackenzie globin-based oxygen carrier in normal may be improved with oxygen-carrying solu- CF, et al. In vitro interference of the red male and female subjects. J Lab Clin Med tions and hematopoietic stimulation may be cell substitute pyridoxalated hemoglobin- 1995; 126:444–51. a result of infusion of a hemoglobin-based polyoxethylene with blood compatibility, oxygen-carrying solution. Because of the abil- coagulation, and clinical chemistry test- ity to carry oxygen, these solutions may also ing. J Cardiothoracic Vasc Anesth 1997; be useful for organ preservation, extracorpo- 11:845–50.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 43 15 Hemoglobin Therapeutics, Blood Substitutes, and High Volume Blood Loss

Armin Schubert, MD, MBA States. Oxygen-carrying volume expanders may undertaken a variety of strategies to modify the Chairman, Department of General Anesthesia be particularly helpful in situations where native hemoglobin molecule in order to stabi- Cleveland Clinic Foundation blood is not available (remote areas; difficult lize it, extend intravascular residence time, and Cleveland, Ohio, USA cross match; rare blood type, etc.). Further- return its oxygen-unloading properties into the more, a national blood shortage is predicted range of erythrocyte-based hemoglobin. One [Editors’ note: Dr. Schubert is a consultant to with the aging of America, since the over-65 such method is intramolecular cross-linking Biopure (Hemosol).] age group has a high demand for blood. This between alpha and beta chains. Other meth- age group represents 12.5% of the population ods involve polymerization, pyridoxylation, or Definitions but receives 50% of all blood transfusions. The conjugation to larger molecules, including Technically, a blood substitute is a sub- risk of infection from blood has decreased dra- polyethylene glycol (PEG). Encapsulation of stance that can effectively replace most func- matically, but potentially could be eliminated hemoglobin into a liposome or polymer struc- tions of human blood. However, oxygen-carry- with blood substitutes (although it is recog- ture has also been pursued. There is a dilemma ing modified hemoglobin solutions and nized that prions and other agents appear to in the trade-off between desirable properties: perfluorocarbons have been referred to as resist sterilization). Allogeneic blood also is Larger hemoglobins and liposomes may have “blood substitutes.” Since these recently devel- associated with a higher surgical infection rate, longer half-lives and are less active in scaveng- oped solutions can only carry out selected func- presumably related to the immunosuppressive ing nitric oxide (NO) from the endothelium tions of blood, they are more accurately referred effects of white blood cells contained in non- (which limits their hypertensive properties). to as “oxygen-carrying volume expanders.” leuko-depleted blood. Unfortunately, they also undergo accelerated Hemoglobin-based oxygen carriers Desirable “blood substitutes” have a long auto-oxidation, hemoglobin peroxidation, and (HBOCs) are modified hemoglobin solutions shelf life, a long circulation half-life, good oxy- heme loss.1 On the other hand, smaller spe- or hemoglobin packaged into liposomes, gen carrying capacity and tissue oxygen deliv- cies are less antigenic but can be filtered by which are able to deliver oxygen to tissues. A ery, few side effects, and reasonable cost. Fur- the kidneys, are more oncotically active, and hemoglobin therapeutic is a hemoglobin so- thermore, their use should not interfere with have shorter vascular residence times. lution optimized through chemical modifica- diagnostic tests or the clinical diagnosis of se- Such “designer” modifications stabilize the tion to bring about certain pharmacologic and rious disease processes. molecule’s tetrameric structure and affect mo- therapeutic effects. Hemoglobin therapeutics lecular size, renal filtration, P50 (defined as the may possess a combination of therapeutically Hemoglobin-Based Oxygen Carriers oxygen tension at which hemoglobin oxygen active properties such as oxygen-carrying ca- Structure and Design saturation is 50%), affinity to NO binding, cir- pacity, favorable rheologic properties, and Free, unmodified human tetrameric he- culation half life, and more. The raw material pressor action. moglobin rapidly dissociates into dimers and for hemoglobin solutions can be human red monomers when removed from its normal blood cells, bovine red blood cells, or recombi- Need for Blood Substitutes environment inside the erythrocyte. Dissocia- nant Escherichia coli bacteria. To date, no he- Although blood transfusions represent a tion into hemoglobin fragments leads to renal moglobin solution has been approved for hu- life-saving measure for many medical and sur- toxicity and greatly increased oxygen affinity, man use, although several are being investigated gical patients, there are still problems with ho- precluding effective tissue oxygen delivery. for safety and efficacy (Table 1). mologous blood transfusions in the United Manufacturers of HBOCs therefore have

Table 1. Hemoglobin Solutions Undergoing Clinical Testing

HBOC Raw Material Structure for Size (kD) T1/2 (hr) Oncotic Viscosity P50* Stabilization Pressure vs. blood (mmHg) (mmHg)

Hemassist (DCLHb™; Human RBC Alpha-alpha cross-linked 64 4-16‡ 42 50% 32 Baxter)†

Optro (Baxter-Somatogen) Recombinant Alpha-alpha cross-linked; 64 12-24 <20 50% 33 E coli Hb Presbyterian mutation

Hemopure (HBOC-201; Bovine RBC Glutaraldehyde- >150 8-17§ 17 30% 34 Biopure) polymerized

Hemosol (Fresenius) Human RBC o-Raffinose cross-linked >150 10-11 24 25% 34 polymerized

Polyheme (Northfield) Human RBC Polymerized Hb; >150 24 20 30%-40% 28-30 pyridoxylated 2,3-DPG site

*Normal human P50 = 28 mmHg †Baxter has discontinued the DCLHb program in favor of developing a second-generation hemoglobin. ‡Varies directly with dose (0.1–1.0 g/kg) §Dose = 0.2–0.6 g/kg

44 Massive Transfusion and Control of Hemorrhage in the Trauma Patient Properties lirubin, and amylase.15–17 In a study of patients without antioxidants) may lead to a potentially Although there are product-specific varia- undergoing high-blood-loss (approximately higher risk of reperfusion injury.30 tions, the P-50s of HBOC solutions are gener- half of an adult’s blood volume) surgical pro- Although many early trials indicate that ally similar to those of fresh blood but higher cedures, 1 g/kg DCLHb™ was associated with some HBOCs have not been associated with than those of stored blood. Circulation half- transient elevations in serum LDH, AST, total severe toxicity, more study in a wide variety of lives are measured in hours (4–24 hours, of- bilirubin, CK, BUN, and amylase; a high inci- clinical situations is required before their side ten dose dependent) rather than days, as dence of yellow skin discoloration; and asymp- effects are fully known. Investigation into the would be the case for red blood cells. tomatic hemoglobinuria.9 effects of HBOCs on the gastrointestinal sys- All currently investigated hemoglobins Gastrointestinal side effects include flatu- tem, pulmonary vasculature, and organ func- elevate systemic and pulmonary vascular re- lence, nausea, vomiting,6,7 and possibly pan- tion during hemorrhagic and other stress is sistance, resulting in a mild reduction in car- creatitis. However, pancreatitis occurs fre- particularly needed. Furthermore, the charac- diac index. For example, diaspirin cross-linked quently after major abdominal surgery18-20 even teristics of HBOC-assisted oxygen delivery and hemoglobin (DCLHb™), an alpha-alpha cross- in the absence of HBOC administration. The tissue oxygen availability during supply-depen- linked tetramer, produces a predictable, rapid, gastrointestinal side effects of DCLHb™ may dent conditions need additional investigation. and sustained rise in mean arterial pressure be related to its ability to interfere with NO (MAP) and in systemic and pulmonary vascu- production and signaling,21 thus possibly af- Perfluorocarbons (PFCs) lar resistance.2 At the microcirculatory level, fecting gastrointestinal and biliary motility. Perfluorocarbons are inert aromatic or functional capillary density is reduced.3 The Judging from preclinical studies22,23 of intesti- aliphatic chemicals that can dissolve oxygen pressor response is dose dependent and phar- nal and portal system blood flow after admin- and carry it in solution throughout the body. macologically reversible and exhibits a “ceil- istration of DCLHb™, gastrointestinal side ef- They typically carry 4 to 50 vol% at a PaO2 of ing effect.”4,5 In human volunteers, 100 mg/kg fects are unlikely the result of tissue ischemia. 160 mmHg; their ability to carry oxygen is di- DCLHb™ raised median systolic BP maximally rectly proportional to their concentration in by no more than 10 mmHg and diastolic BP by Toxicity blood and, importantly, to the partial pressure no more than about 15 mmHg.6 Biopure’s Toxicity of hemoglobin solutions has his- of oxygen. The first fluorocarbon to be ap- HBOC-201 raised MAP by about 10 mmHg torically been related to impurities such as RBC proved for clinical use (during percutaneous when a dose of 0.6 g/kg was administered to membrane residues, endotoxin, free dimers, transluminal coronary angioplasty) was fluosol healthy volunteers,7 but it had no significant and monomers. With vastly improved purifi- DA-20, which contains 20% emulsified fluoro- effect on blood pressure when given to surgi- cation procedures, concern over toxicity from carbon. When used as an oxygen-carrying vol- cal patients.8 In the author’s clinical investiga- impurities is waning. In particular, the issue ume expander, fluosol DA was associated with tive experience with 1g/kg DCLHb™ adminis- of renal toxicity appears to have been over- a number of limitations, including low oxygen- tered to patients undergoing major orthope- come. In rats, 0.4 g/kg DCLHb™ did not affect carrying capacity, short shelf life, temperature dic and urologic surgery, MAP was elevated by renal blood flow.23 Creatinine clearance was instability, and serious side effects. Second- an average of about 20 mmHg, the hyperten- neither decreased by 0.1 g/kg DCLHb™4 nor generation perfluorocarbons, such as sive effect persisting for 24 to 30 hours after by 0.32 g recombinant hemoglobin24 in human perfluoro-octylbromide (PFOB; Alliance Phar- administration.9 Although HBOC-associated volunteers. It was similarly unaffected by up maceuticals), are being investigated and show hypertension has not been associated with to 0.7 g/kg in critically ill patients with sepsis promise because of a much higher oxygen-car- adverse cardiac events, selected patients are syndrome25 by 750 ml DCLHb™ in cardiac pa- rying capacity, a 2- to 4-year refrigerated shelf likely to require treatment of hemoglobin-in- tients,26 and by 1.0 g/kg DCLHb™ in patients life, low viscosity, and less interference with duced systemic and pulmonary hypertension. undergoing high-blood-loss surgery, despite normal pulmonary surfactant mechanisms.31 The mechanisms thought to account for the occurrence of hemoglobinuria at the higher Since PFCs are not metabolized, but ex- this pressor effect are the scavenging of NO doses.9 Neither was renal toxicity observed creted unchanged via the lungs, their poten- from vascular endothelium, facilitation of with polymerized hemoglobin.27 tial for cytotoxicity is thought to be limited. endothelin production and, possibly, a sym- Free hemoglobin, then directly applied to There is no antigenicity. However, since PFCs pathomimetic effect. The smaller the hemoglo- central nervous system tissue, is neurotoxic. are taken up avidly by the reticuloendothelial bin molecule the more effectively it interacts It stimulates leukocyte migration and vascular system, they increase liver enzymes and result with the endothelium, penetrating it and scav- adherence. Hemoglobin also activates plate- in hepatosplenomegaly. Because of the exten- enging endothelial NO to form met-hemoglo- lets, promoting aggregation.28 Circulating fer- sive uptake in the reticuloendothelial system bin and NO-hemoglobin.10 rous hemoglobin, even when highly purified, and impairment of neutrophil function, they In the operative setting, several factors undergoes a number of reactions that may may interfere with host defense mechanisms. may blunt HBOC-associated hypertensive ten- contribute to toxicity.29 Ferrous hemoglobin Monocyte and macrophage activation may lead dencies. The hypotensive action of surgical binds NO about 3,000 times more tightly than to release of prostaglandins, endoperoxides, hemorrhage,11 as well as volume depletion,12,13 carbon monoxide and therefore effectively re- and cytokines, which probably accounts for the may diminish hypertension. Furthermore, ha- moves any NO in its vicinity, accounting for symptoms of flushing, backache, fever, chills, lothane and propofol, but not isoflurane, have the vasoactive properties. Free hemoglobin is headaches, and nausea observed in clinical tri- been shown to decrease the hypertensive ac- converted to methemoglobin at a rate as fast als. Platelet count decreases by as much as 40% tion of DCLHb™ on pulmonary vein rings.14 as 4% per hour; this reaction can lead to the due to increased platelet clearance from PFC- Hemoglobin solutions have colloidal generation of free radicals. Hemoglobin also induced modification of platelet surfaces.32 properties (Table 1), are highly purified, gen- has a number of “pseudo-enzymatic” proper- PFCs also may prolong the effects of certain erally do not affect coagulation, and are only ties, which could lead to oxygenation, lipid drugs, including barbiturates. weakly antigenic. Modified molecular hemo- peroxidation, and cytotoxicity. Further possi- globin undergoes oxidation to methemoglo- bilities for toxicity arise from the degradation Potential Clinical Uses and Effective- bin and leaves the circulation primarily products of hemoglobin’s heme moiety such ness of HBOCs Major Surgical Bleeding through the reticuloendothelial system. Pre- as hemin. Red blood cells contain antioxidant and Hemorrhagic Shock clinical and clinical studies indicate that modi- enzymes such as catalase and superoxide Fluid therapy for the acutely bleeding pa- fied hemoglobins can mildly increase the con- dismutase, which may help limit ischemia tient can be accomplished initially with either centrations of plasma CPK (but not MB frac- reperfusion injury. It has been speculated that crystalloid or colloid solutions. Blood transfu- tion), hepatic enzymes, reticulocyte count, bi- the administration of pure hemoglobin (i.e., sion is begun when, despite volume resuscita-

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 45 tion with non-oxygen carrying solutions, there them over” a limited period of intraoperative is evidence of tissue ischemia and resultant Table 2. or postoperative bleeding, after which autolo- organ dysfunction. Accumulation of base defi- Use of Hemoglobin-Based Oxygen gous or allogeneic blood would be administered cit and serum lactate and low central venous Carriers in Patients with High Blood Loss if still needed. Furthermore, volume replace- oxygen concentration are all indices of tissue ment with HBOCs (compared with crystalloid ischemia, which should be taken into consid- Emergency administration or colloid) during ANH for autologous collec- eration in the transfusion decision.33 Alterna- • Trauma, especially penetrating tion would likely result in a greater yield of tively, blood is transfused when organ ischemia • Unexpected surgical bleeding pheresed blood components. can be anticipated, given the extent and rapid- • Unexpected bleeding from ity of ongoing bleeding. disease (e.g., gastrointestinal tract) Caveats Regarding the Use of HBOCs Hemoglobin solutions are as effective as • Difficult cross-match for Major Blood Loss (Table 3) whole blood in restoring MAP in animals34-36 and The safety of large-scale and rapid transfu- humans.12 In contrast to typical catecholamine Elective administration sion of HBOCs in human traumatic injury re- effects, the pressor response of DCLHb™ is as- • Acute normovolemic hemodilution* mains to be demonstrated. While the author’s sociated with an increase in perfusion (as indi- • Acute hypervolemic hemodilution* small series of patients undergoing high-blood- cated by organ flow measurements) in both top- • Replacing blood transfusion during loss elective surgery tolerated up to 1g/kg load and hemorrhagic, hypovolemic, animal expected active surgical bleeding DCLHb™ relatively well,9 a phase III trial of models.36–40 DCLHb™, compared with non-oxy- • Replacing blood transfusion DCLHb™ for resuscitation of traumatically in- gen-containing crystalloid or colloid solutions, postoperatively jured patients was halted among concerns about resulted in substantially better survival from increased mortality in the study group. experimental hemorrhagic shock.37,38 This salu- *Especially in patients presenting with Because HBOCs are associated with sys- tary effect may be related to DCLHb™’s effect low initial hematocrit temic hypertension, concern has been raised on tissue perfusion and peripheral oxygenation. over a potential for increased blood loss in For example, tissue oxygenation, measured di- hemorrhage. This concern could not be cor- rectly by a fluorescence-quenching optode, was surgery was approximately 10 hours.45 Admin- roborated in preclinical40 or clinical studies9 restored more effectively in a rat hemorrhagic istration of DCLHb™ after bypass spared nearly conducted in a setting of hemorrhage, but the shock model treated with DCLHb™ compared 20% of cardiac patients from allogeneic trans- issue has yet to be clarified in the setting of with lactated Ringer’s solution and albumin.41 fusion.46 Nevertheless, there is also concern penetrating trauma. Despite increased total peripheral vascular re- that the administration of modified hemoglo- The clinical use of HBOCs with relatively sistance, rat coronary blood flow23 was aug- bins merely delays blood transfusion rather short half-lives must take into account their mented after DCLHb™ and human cerebral than truly substituting for it. tendency for transvascular migration and their blood flow was unchanged after infusion of Preoperative acute normovolemic hemodi- rapid clearance through the reticuloendothe- polymerized hemoglobin.42 lution (ANH) is likely to become more attrac- lial system. Although the initial effect of trans- Despite their vasoconstrictive properties, tive with the use of modified hemoglobins as a fusion may be an immediate increase in vascu- HBOCs may counteract tissue hypoperfusion diluent. The short half-life of HBOCs does not lar volume (enhanced by some HBOCs’ col- with added blood oxygen-carrying capacity and present a significant liability for this clinical ap- loidal properties) and blood pressure (medi- better rheologic properties. In spontaneously plication. Patients with low preoperative hema- ated by the HBOCs’ NO-scavenging effect), hypertensive rats subjected to middle cerebral tocrit might receive an infusion of HBOC to “tide rapid dissipation of the HBOC requires care- artery occlusion, hemodilution with DCLHb™ (to hematocrits of 30, 16, or 9%) resulted in a significant dose-dependent reduction in the Table 3. Caveats and Potential Remedies in the Clinical Use of HBOCs extent of brain injury and cerebral edema.43 The most effective reductions in ischemic in- Caveats Remedies jury occurred in those animals in which the inherent hypertensive response to DCLHb™ Hypertensive tendency; Co-administration of nitroglycerin, was not inhibited. cardiac afterload stress other vasodilator The effect of HBOCs on cardiac index is more controversial, with some studies report- Pulmonary hypertension, Co-administration of pulmonary ing a slight decrease,44 others no change.25 Cal- right ventricular dysfunction vasodilator culated oxygen delivery generally follows cardiac output, thus accounting for the slight de- Short intravascular residence time; More frequent assessment and creases reported. However, increased tissue- recurrence of hypovolemia adjustment of intravascular volume dffusing capacity has been shown.7 Further- more, the equivalent or enhanced oxygen-un- Interference with diagnostic Avoidance of photospectrometric methods; loading capacity of HBOCs compared with blood tests removal of free hemoglobin from blood should allow favorable tissue oxygen specimens; other correction algorithms delivery, or at least counteract vasoconstrictive effects of free hemoglobins. Therefore, their use Hemoglobinurla interfering with Special pre-arranged testing protocol in trauma patients and in those with substan- diagnosis of transfusion reaction tial surgical bleeding would seem reasonable. Table 2 suggests potential uses of HBOCs for Immune depression as larger Hb species Unknown therapy of patients suffering large blood losses. overwhelm the reticuloendothelial system However, because of their short half-lives, current hemoglobin solutions are likely to be Possible NO-related gastrointestinal or Co-supply NO donor or precursor; used essentially as a “bridge to transfusion.” other organ injury redesign molecule For example, the half-life of DCLHb™ administered to patients undergoing high-blood-loss NO, nitric oxide; Hb, hemoglobin

46 Massive Transfusion and Control of Hemorrhage in the Trauma Patient ful and frequent monitoring of circulatory ad- dose-escalation study to evaluate the ki- circulatory and systemic hemodynamic equacy, since hypovolemia may re-manifest netics and hemodynamic effects of hemo- effects of diaspirin crosslinked hemoglo- rather quickly.28 globin-based oxygen carrier-201. Anesthe- bin in the rat. Artif Cells Blood Subs At least within the first 24 to 36 hours of siology 1997; 87:A214. Immobil Biotech 1994; 22:593–602. administration, free hemoglobins can interfere 9. Schubert A, Bedocs N, O’Hara JF Jr, Tetzlaff 23. Sharma AC, Gulati A. Effect of diaspirin with the photospectrometric methods used in JE, Marks KE, Novick AC. Effect of crosslinked hemoglobin and norepineph- a variety of clinical laboratory tests. Interfer- perioperative administration of diaspirin rine on systematic hemodynamics and ence with laboratory testing constitutes an cross-linked hemoglobin on indices of or- regional circulation in rats. J Lab Clin Med important limitation for the potential clinical gan function. Anesthesiology 1997; 1994; 123:299–308. use of artificial hemoglobin species. 87:A220. 24. Viele MK, Weiskopf RB, Fisher D. Recom- 10. Scott MG, Kucik DF, Goodnough LT, Monk binant human hemoglobin does not affect Other Uses: Hemoglobin Therapeutics? TG. Blood substitutes: evolution and fu- renal function in humans: analysis of Since NO plays a part in the pathogenesis ture applications. Clin Chem 1997; safety and pharmacokinetics. Anesthesiol- of septic shock, modified hemoglobins may 43:1724–31. ogy 1997; 86:848–58. become useful in the treatment or prevention 11. Malcolm D, Kissinger D, Garrioch M. 25. Rhea G, Bodenham A, Mallick A, Przybelski of severe septic shock. Artificial oxygen carri- Diaspirin crosslinked hemoglobin solu- R, Daily E. Initial evaluation of diaspirin ers also may be used for oxygen delivery to tion as a resuscitative fluid following se- crosslinked hemoglobin (DCLHb) as a ischemic tissues (as in stroke or intestinal is- vere hemorrhage in the rat. Biomat, Artif vasopressor in critically ill patients. Crit chemia) and tumor cells to improve their sus- Cells, Immmob Biotech 1992; 20:495–7. Care Med 1997; 25:1480–8. ceptibility to radiation and chemotherapy. Be- 12. Swan SK, Halstenson CE, Collins AJ, 26. Baron JF, Berridge J, Brichant JF, cause iron is one of the breakdown products Colburn WA, Blue J, Przybelski RJ. Phar- Demeyere R, Lamy M, et al. The use of of hemoglobin metabolism, hemoglobin macologic profile of diaspirin crosslinked diaspirin crosslinked hemoglobin therapy may stimulate erythropoiesis under hemoglobin in hemodialysis patients. Am (DCLHb) as an alternative to blood trans- certain circumstances. These and other poten- J Kidney Dis 1995; 26:918–23. fusion in cardiac surgery patients follow- tially salutary effects of HBOCs, which tran- 13. Malcolm D, Kissinger D, Garrioch M. ing cardiopulmonary bypass: a pivotal ef- scend basic oxygen-carrying properties, have Diaspirin crosslinked hemoglobin solu- ficacy trial. Anesthesiology 1997; 87:A217. led to an emerging interest in the area of “he- tion as a resuscitative fluid following se- 27. Gonzalez P, Hackney AC, Jones S. moglobin therapeutics.” vere hemorrhage in the rat. Biomat, Artif Strayhorn D, Hoffman EB, Hughes G, Cells, Immmob Biotech 1992; 20:495–7. Jacobs EE, Oringer EP. A phase I/II study References 14. Jing M, Ledvina MA, Bina S, Hart JL, of polymerized bovine hemoglobin in 1. Sanders KE, Ackers G, Sligar S. Engineer- Muldoon SM. Effects of halogenated and adult patients with sickle cell disease not ing and design of blood substitutes. Curr non-halogenated anesthetics on diaspirin in crisis at the time of study. J Inv Med Op Struc Biol 1996; 6:534–40. crosslinked hemoglobin induced contrac- 1997; 45:258–64. 2. Hess JR, Macdonald VW, Brinkley WW. Sys- tions of porcine pulmonary veins. Artif 28. Hess JR, Reiss RF. Resuscitation and the temic and pulmonary hypertension after Cells Blood Subs Immobil Biotech 1995; limited utility of the present generation resuscitation with cell-free hemoglobin. J 23:487–94. of blood substitutes. Transfusion Med Rev Appl Physiol 1993; 74:1769–78. 15. Leone BJ, Chuey C, Gleason D, Steele SM, 1996; 10:276–85. 3. Tsai AG, Kerger H, Intaglietta M. Microcir- et al. Can recombinant human hemoglo- 29. Everse J, Hsia N. The toxicities of native culatory consequences of blood substitu- bin make ANH more effective? An initial and modified hemoglobins. Free Radical tion with alpha-alpha-hemoglobin. In feasibility and safety study. Artif Cells Biology & Medicine 1997; 22:1075–99. Winslow RM, Vandegraff KD, Intaglietta M, Blood Subs Immobil Biotech 1996; 30. Chang TM. Recent and future develop- eds. Current Issues in Blood Substitute Re- 24:A379. ments in modified hemoglobin and mi- search—1995. Boston, Birkhauser, 1995, 16. Lessen R, Williams M, Seltzer J, Lessin J, croencapsulated hemoglobin as red blood pp 155–74. et al. A safety study of recombinant hu- cell substitutes. Artif Cells Blood Subs 4. Malcolm DS, Hamilton IN, Schultz SC, et man hemoglobin for intraoperative trans- Immobil Biotech 1997; 25:1–24. al. Characterization of hemodynamic re- fusion study. Artif Cells Blood Subs 31. Spiess BD. Perfluorocarbon emulsions: sponse to intravenous diaspirin Immobil Biotech 1996; 24:A380. one approach to intravenous artificial res- crosslinked hemoglobin solution in rats. 17. Gould SA, Moore EE, Moore FA, Haenel piratory gas transport. Inter Anesth Clin Artif Cells Immobil Biotech 1994; 22:91– JB, et al. The clinical utility of human po- 1995; 33:103–13. 107. lymerized hemoglobin as a blood substi- 32. Smith DJ, Lane TA. Effect of a high con- 5. Hamilton I, Schultz SC, Cole F, Burhop K, tute after acute trauma and urgent surgery. centration perfluorocarbon emulsion on Malcolm D. Characterization of diaspirin J Trauma 1995; 39:157. platelet function. Biomater Artif Cells cross-linked hemoglobin’s pressor re- 18. Jurkovich GJ, Carrico CJ. Pancreatic trauma. Immobil Biotech 1991; 19:383–5. sponse. Crit Care Med 1992; 20:S106. Surg Clin North Am 1990; 70:575–93. 33. Baron BJ, Scalea TM. Acute blood loss. 6. Przybelski RJ, Daily EK, Kisicki JC, Mattia- 19. Miyagawa S, Makuuchi M, Kawasaki S, Emerg Med Clin North Am 1996; 14:35–55. Goldberg C, Bounds MJ, Colburn WA. Kakazu T. Changes serum amylase level 34. Barve A, Sen AP, Saxena PR, Gulati A. Dose Phase I study of the safety and pharmaco- following hepatic resection in chronic liver response effect of diaspirin crosslinked logic effects of diaspirin crosslinked he- disease. Surgery 1994; 129:634–8. hemoglobin (DCLHb) on systemic hemo- moglobin solution. Crit Care Med 1996; 20. Akagi Y, Yamashita Y, Kurohiji T, et al. Inves- dynamics and regional blood circulation 24:1993–2000. tigation of hyperamylasemia after hepatec- in rats. Artif Cells Blood Substit Immobil 7. Hughes GS, Antal EJ, Locker PK, Francom tomy. J Jpn Soc Clin Surg 1991; 52:314–8. Biotechnol 1997; 25(1-2):75–84. SF, Adams WJ, Jacobs EE Jr. Physiology and 21. Sharma AC, Sinh G, Gulati A. Role of NO 35. Chang TMS, Varma R. Effect of a single pharmacokinetics of a novel hemoglobin- mechanism in cardiovascular effects of replacement of one of Ringer lactate, hy- based oxygen carrier in humans. Crit Care diaspirin crosslinked hemoglobin in anes- pertonic saline/dextran, 7g% albumin, Med 1996; 24:756–64. thetized rats. Am J Physiol 1995; 269(4 Pt stroma-free hemoglobin, o-raffinose 8. Monk TG, Goodnough LT, Peruzzi WT, 2):H1379–88. polyhemoglobin or whole blood on the Byers PM, Jacobs Jr EE, Silverman MH. A 22. Sharma AC, Rebello S, Gulati A. Regional long term survival of unanesthetized rats

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 47 with lethal shock after 67% acute blood Biotech 1994; 22:A260. 43. Cole DJ, Schell RM, Drummond JC, loss. Biomater Artif Cells Immobil Biotech 39. McKenzie JE, Scandling DM, Rohrer MJ, Przybelski RJ, Marcantonio S. Focal cere- 1992; 20:503–10. Jacot JL. Diaspirin crosslinked hemoglobin bral ischemia in rats: effect of hemodilu- 36. Przybelski RJ, Malcolm DS, Burris DG, et (DCLHb™) during hypovolemic shock in tion with (-( crosslinked hemoglobin on al. Cross-linked hemoglobin solution as a swine. ISBS 1993 Program and Abstracts. brain injury and edema. Can J Neurol Sci resuscitative fluid after hemorrhage in the 40. Schultz SC, Powell CC, Burris DG, et al. The 1993; 20:30–6. rat. J Lab Clin Med 1991; 117:143–7. efficacy of diaspirin crosslinked hemoglobin 44. Lamy M. Personal communication, 1997. 37. McKenzie JE, Scandling DM, Ahle NW. solution resuscitation in a model of uncon- 45. O’Hara JF, Tetzlaff JE, Udayashankar SV, Diaspirin crosslinked hemoglobin trolled hemorrhage. J Trauma 1994; 37:408. Connors DF, Bedocs NM, Schubert A. The (DCLHb™): Improvement in regional 41. Powell C, Schultz SC, Burris DG, Drucker effect of diaspirin cross-linked hemoglobin blood flow following administration in hy- WR, Malcolm DS. Subcutaneous oxygen on coagulation in surgical patients. Anes- povolemic shock in the swine (abstract). tension: a useful adjunct in assessment of thesiology 1997; 87:A230. 5th International Symposium Blood Sub- perfusion status. Crit Care Med, in press. 46. Baron JF, Berridge J, Brichant JF, et al. The stitutes, San Diego, 1993. 42. Brauer P, Standl T, Wilhelm S, Burmeister use of diaspirin crosslinked hemoglobin 38. Schultz SC, Powell CC, Bernard E, Malcolm MA, Schulte am Esch J. Transcranial doppler (DCLHb) as an alternate to blood transfu- D. Diaspirin crosslinked hemoglobin sonography mean flow velocity during in- sion in cardiac surgery following cardiop- (DCLHb™) attenuates bacterial transloca- fusion of ultrapurified bovine hemoglobin. ulmonary bypass. Anesthesiology 1997; tion in rats. Artif Cells Blood Subs Immobil J Neurosurg Anesth 1998; 10:146–52. 87:A217.

CME Questions This monograph can be used to earn 15 AMA category 1 credit hours. The International Trauma Anesthesia and Critical Care Society (ITACCS) is accredited by the Accreditation Council for Continuing Medical Education (ACCME) for physicians. This CME activity was planned and produced in accordance with the ACCME Essentials. ITACCS designates this CME activity for 15 credit hours in Category 1 of the Physicians Recognition Award of the American Medical Association.

Educational Objectives thesia and Critical Care Society (ITACCS). pleted disclosure forms are on file in the man- This activity is designed to provide trauma ITACCS is accredited by the ACCME to sponsor aging editor’s office. care professionals interested in the treatment continuing medical education (CME) for phy- of critically ill trauma patients with a regular sicians and takes responsibility for the content, INSTRUCTIONS overview and critical analysis of the most cur- quality, and scientific integrity of this CME ac- • Print the answer form and the evaluation rent, clinically useful information available, cov- tivity. form on page 51. ering strategies and advances in the diagnosis • Complete both forms. On the answer form, of traumatic injuries and the treatment of Credit Designation Statement circle only one response next to each num- trauma patients. Controversies, advantages, and ITACCS designates this educational activ- ber. disadvantages of diagnosis and treatment plans ity for a maximum of 15 hours in category 1 • Sign and date the certification statement are emphasized. There are no prerequisites for credit toward the AMA Physicians Recognition on page 51. participation in this activity. Award. • Write a check for $150 (or $75 accompa- After reading this document, participants nied by verification of current ITACCS should have a working familiarity with the most Faculty Disclosure Statement membership), payable to the International significant information and perspectives pre- It is the policy of ITACCS that faculty mem- Trauma Anesthesia and Critical Care Soci- sented and be able to apply what they have bers disclose real or apparent conflict of inter- ety. learned promptly in clinical practice. est relating to the optics of this educational • Mail the forms and your check (and mem- activity and also disclose discussions of unla- bership verification, if applicable) to Accreditation Statement beled/unapproved uses of drugs or devices in ITACCS, Department of CME Credit, PO This activity is planned and produced in their presentations. Sincere effort was made to Box 4826, Baltimore, MD 21211. accordance with the Essential Areas and Poli- contact the contributors to this publication. Any • The completed text will be accepted for cies of the Accreditation Council for Continu- responses that could possibly suggest conflict grading if received by January 31, 2004. ing Medical Education (ACCME) through the of interest are published on the opening pages • Please allow 4 to 6 weeks for processing. sponsorship of the International Trauma Anes- of the individual articles. The authors’ com-

1. The most sensitive measure of acute 3. The relationship between serum lactate 5. Interventional radiologic techniques can blood loss is and base deficit remains constant for how be useful in which body area? a. blood pressure. long following resuscitation? a. Zone 3 of the neck b. urine output. a. 12 hours b. Zone 2, the thoracic outlet c. heart rate. b. 24 hours c. deep in the pelvis d. mixed venous oxygen saturation. c. 36 hours d. all of the above d. 48 hours 2. High-risk elderly patients require invasive 6. Stages of traumatic shock include all of the hemodynamic monitoring 4. Which of the following is the most accu- following except a. only if they have a history of coronary rate predictor of survival following injury? a. subacute irreversible shock. artery disease. a. serial blood pressure determination b. compensated shock. b. as early as possible, following emer- b. adequacy of urine output c. decompensated shock. gency department admission. c. ability to clear lactate to normal d. cardiogenic shock. c. once they are evaluated for injuries. d. resolution of tachycardia e. acute irreversible shock. d. if they have evidence of hypotension and tachycardia.

48 Massive Transfusion and Control of Hemorrhage in the Trauma Patient 7. Of the following organ systems, the one possesses clotting activity similar catheter most directly affected by decreased blood to collagen preparations. d. Placement of the ECG flow in traumatic shock is c. Thrombin is effective only if e. Obtaining large-bore venous access a. cardiac. combined with a carrier b. intestinal. such as Gelfoam®. 22. Appropriate intraoperative fluid manage- c. pulmonary. d. Fibrin glue has been shown to be ment for a 70-kg multiple blunt trauma d. central nervous. effective as either the primary patient in class 4 hemorrhagic shock in- e. skeletal muscle. hemostatic agent or as an adjunct to cludes which of the following: conventional suture repair in patients a. Hetastarch, 2.5-L bolus 8. Which of the following statements is correct? with hepatic or splenic trauma. b. Lactated Ringer’s, 5 L a. In compensated shock, the body is not e. In vitro testing reveals that oxidized c. 7.5% saline, 1.5 L developing an oxygen debt. regenerated cellulose (Surgicel®) is d. Two units type-specific b. In subacute irreversible shock, normal more effective than collagen uncrossmatched red blood cells and hemodynamics are never achieved. preparations for inducing platelet 3 L normal saline c. In neurogenic shock, ischemia is caused aggregation and clotting. e. Four units of fresh frozen plasma by decreased oxygen-carrying capacity. d. Traumatic shock is the same as hem- 15. Severely injured patients 23. Resuscitation endpoints after major trauma orrhagic shock. a. have been shown to have elevated se- include which of the following: e. Decompensated shock is a stable clini- rum fibrin degradation products (FDP). a. Resolution of lactic acidosis and cal state that can persist for many days. b. may exhibit thrombocytopenia. base deficit c. may progress to death if FDP and b. Mixed venous oxygen saturation 45% 9. Which of the following is not an inflamma- platelet assays trend in an abnormal c. Normalization of ventilation tory mediator produced by ischemic cells? manner. –perfusion mismatch a. Prostacyclin d. benefit from prophylactic transfusion of d. All of the above b. Tumor necrosis factor fresh frozen plasma and platelets even e. a and c c. Complement in the absence of pathologic bleeding. d. Thromboxane e. a, b, c 24. The differential diagnosis of hypotension e. Angiotensin II in the setting of massive transfusion after 16. Risk factors for DVT in trauma patients in- major blunt trauma includes all of the fol- 10. Acute irreversible shock includes all of the clude lowing except: following clinical signs except a. spinal cord injury. a. Hypocalcemia a. hyperthermia. b. prolonged bed rest. b. Transfusion reaction b. coagulopathy. c. hypercoagulability. c. Hypovolemia c. hypotension not responsive to fluids. d. lower extremity fractures. d. Tension pneumothorax d. hypotension not responsive to e. all of the above. e. All of the above inotropes. e. diffuse edema. 17. The most common inborn metabolic er- 25. Which of the following products carries the ror that causes thrombophilia is highest risk of infection? 11. The first response of the body to obtain a. activated protein C resistance. a. 5 units packed red blood cells hemostasis is b. protein C deficiency. b. 2 units fresh frozen plasma a. initiation of the coagulation cascade. c. protein S deficiency. c. 6 units platelets b. platelet aggregation. d. hypohomocysteinemia. d. 3 units whole blood c. initiation of fibrinolysis. e. serum porciline deficiency. e. 2 L 0.9% saline d. platelet release reaction. e. vasoconstriction. 18. Physical examination is the most accurate 26. Is there an exact transfusion trigger HCT method of diagnosing DVT. at which all patients should be transfused: 12. Concerning platelets and hemostasis: a. True b. False a. Yes b. No a. Exposure of platelets to subendothelial collagen leads to 19. The primary reason to provide prophylac- 27. Can hepatitis C be transmitted via blood adherence between platelets and tic treatment to prevent DVT in trauma transfusion? the blood vessel wall. patients is to a. Yes b. No b. Platelet release reaction refers to a. prevent leg swelling. the liberation of platelets b. enhance fracture healing. 28. Does transfusion result in immunosup- sequestered in the spleen. c. prevent fatal pulmonary embolism. pression of the recipient? c. The intrinsic coagulation pathway d. increase billable services. a. Yes b. No is the predominant pathway in the e. decrease length of hospitalization. coagulation cascade. 29. In general, will patients with histories of d. The intrinsic and extrinsic pathways 20. Epidural analgesia in the patient impaired cardiac function or cardiac is- merge with the activation of Factor IX. receiving LMWH chemia require transfusion at higher or e. Fibrinolysis occurs only after the a. is absolutely contraindicated. lower HCT levels? clotting mechanism is completed. b. is associated with epidural abscesses. a. Higher b. Lower c. is no problem. 13. The ideal topical hemostatic agent pos- d. may be performed at least 12 hours 30. Is there a relationship between the num- sesses which of the following properties: after the last dose. ber of units of blood transfused and infec- a. Rapid time to hemostasis e. is not associated with problems of tion in trauma patients? b. Easily applied and manipulated catheter removal. a. Yes b. No c. Holds sutures d. Low infectious risk and minimal 21. Management priorities in the acutely 31. Complications of subclavian and internal tissue reaction bleeding trauma patient include all of the jugular catheterization include e. All of the above following except: a. air embolism a. Measurement of BP b. hemothorax 14. Concerning topical hemostatic agents: b. Securing the airway and verifying c. pneumothorax a. Collagen sponges are unique in that adequacy of ventilation and d. sepsis they are bactericidal. oxygenation e. all of the above b. Denatured gelatin (Gelfoam®) c. Insertion of a pulmonary artery

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 49 32. In a patient with multiple stab wounds to rent heat exchanger combined with c. Infusion rates of up to 1,500 cc/min the abdomen, which of the following heated patient line can be achieved. would provide adequate venous access? c. Metal foil countercurrent heat ex- d. 100-cc and/or 500-cc boluses over 1 a. a large-bore femoral catheter change minute can be infused periodically. b. two upper extremity 14-gauge d. IV tubing sandwiched between alu- e. All forms of blood components may IV catheters minum heating plates in a serpentine be infused through it. c. a saphenous cutdown fashion—dry heat technology d. a right internal jugular triple lumen e. Countercurrent heated patient line 47. True statements regarding the i-STAT Por- e. none of the above to insure delivery of 37°C fluid at table Clinical Analyzer (i-STAT Corp., flow rates of 5–80 ml/min (300– Princeton, NJ) include all of the follow- 33. Choose the incorrect statement regarding 5,000 ml/hr) ing except: venous access in the trauma patient: a. It is a hand-held unit. a. Venous cutdowns provide rapid, se- 41. When a critically injured patient enters the b. It utilizes a “thin film” biosensor re- cure, large-bore venous access. operating room for emergency surgical quiring 2 to 3 drops of blood in or- b. Two large-bore percutaneous cath- therapy, which of the following should be der to give results over a variety of eters should be placed immediately. the anesthesiologist’s #1 priority? laboratory parameters. c. A central line should be inserted in a. TEE probe insertion c. Coagulation studies available include all trauma patients. b. Pulmonary artery catheter insertion PT, PTT, and fibrinogen levels. d. Main complications of venous cut- c. ECG monitoring d. Blood chemistry results are obtained down are nerve injury and infection. d. Blood pressure measurement within 2 minutes. e. The major complications of internal e. Evaluation and management of e. Various laboratory results can be ob- jugular cannulation are pneumotho- the airway, oxygenation, and tained, depending on the particular rax and carotid puncture. administration cartridge inserted into the unit.

34. Choose the incorrect statement: 42. Which of the following is not ideal as a 48. Current guidelines regarding quality con- a. Long, large-bore IV catheters should route for fluid administration in trauma? trol in laboratory testing are mandated be used for rapid IV fluid infusion. a. Use of two peripheral intravenous through b. Thrombosis of femoral catheters oc- catheters in the upper extremities a. The National Committee for Clinical curs more often than with subclavian b. Use of the internal jugular vein with Laboratory Standards catheters. a short, large-bore IV catheter b. The Health Care Financing Adminis- c. Subclavian catheterization should be c. Use of the femoral vein with a large- tration attempted in the side of injury in a bore IV catheter in a patient with a c. The clinical director of an individual patient with a chest wound. gunshot wound in the neck laboratory facility d. Strict aseptic technique should always d. Use of the femoral vein with a large- d. The 1988 Amendment to the Clinical be used in central line placement. bore IV catheter in a patient with sus- Laboratory Improvement Law of 1967 e. Venous air embolism is often a fatal pected cervical spine, abdominal, and e. The Department of Health and Hu- complication. pelvic injuries mane Services

35. Intraosseous catheters 43. Which of the following is a known stor- 49. In the massive transfusion scenario, true a. are recommended only in children. age lesion for PRBCs? statements regarding banked red blood b. should be considered after two un- a. decreased pH cells include all of the following except successful percutaneous IV attempts b. hemolysis a. Pre-washing RBCs removes a signifi- in the pediatric trauma patient. c. increased concentration of potassium cant proportion of citrate that may be c. do not provide adequate venous ac- d. decreased 2,3-DPG present in the infused blood. cess for fluid administration. e. all of the above b. May be indicated in patients with a d. should be used only as a last resort history of renal insufficiency. in a trauma patient. 44. The following are true regarding sodium c. Pre-washing RBCs decreases K+ con- e. none of the above. citrate except centration of blood administered to a. Calcium chloride should always be the patient. 36. Regarding the impact of rapidly infusing given when more than 2 units of d. K+ concentration is unrelated to unwarmed IV fluids in a 70-kg anesthe- blood are transfused to an adult the length of time a unit of blood has tized patient: trauma patient. been stored. a. 4.5 L of 21°C crystalloid will result in b. Sodium citrate transiently decreases e. The risk of untoward effects of mas- ~1.0-1.5°C decrease in mean body ionized calcium. sive transfusion of banked red blood temperature. c. Hypocalcemia can cause hypotension. cells increases with rapidity of trans- b. 4 units of 4°C red cells diluted in 0.9% d. Hypocalcemia can cause a prolonged fusion. saline will result in ~ 1.0-1.5°C de- QT interval. crease in mean body temperature. e. Hypocalcemia can cause biventricular 50. Key points of the rapid infusion strategy c. Red cells may be warmed safely to a cardiac dysfunction. employed by anesthesia personnel at the maximum temperature of 42°C. Elvis Presley Memorial Trauma Center in- d. Gas embolism may occur, especially 45. Which of the following infection is the clude all of the following except with the use of constant pressurized most frequently associated with blood a. Transfusion of blood products through infusion devices. transfusion in the United States? the Rapid Infusion System in units of e. All of the above a. HIV 10 units PRBCS, 4 units fresh frozen b. Hepatitis B plasma, and 7 units pooled platelets. For questions 37-40, match the fluid/blood c. Hepatitis C b. Dilution of each unit of red blood warmer with the one best answer. d. HTLV 1 cells with 500 cc of normal saline. Answers may be used only once. e. HTLV 2 c. Maintenance of relative normotension. 37. Hotline d. Communication with surgeons, the 38. Flotem IIe 46. True statements regarding the Rapid In- blood bank, and lab personnel re- 39. Level 1- H1000 fusion System (Haemonetics Corporation) garding use of the RIS. 40. Alton Dean/Mallinkrodt FW537 or FW538 include all of the following except e. Infusion of fluids through the RIS at a. Coiled IV tubing immersed in a wa- a. It features a roller pump mechanism. 1,500 cc/min until hemostasis is ter bath b. Fluids are pumped from a 3-liter hard achieved. b. Aluminum tube in tube countercur- shell reservoir.

50 Massive Transfusion and Control of Hemorrhage in the Trauma Patient CME ANSWER FORM

Answer Form: Please circle the one best answer for each question. Massive Transfusion Monograph — 1999/2002

1. a b c d 18. a b 35. a b c d e 2. a b c d 19. a b c d e 36. a b c d e 3. a b c d 20. a b c d e 37. a b c d e 4. a b c d 21. a b c d e 38. a b c d e 5. a b c d 22. a b c d e 39. a b c d e 6. a b c d e 23. a b c d e 40. a b c d e 7. a b c d e 24. a b c d e 41. a b c d e 8. a b c d e 25. a b c d e 42. a b c d 9. a b c d e 26. a b 43. a b c d e 10. a b c d e 27. a b 44. a b c d e 11. a b c d e 28. a b 45. a b c d e 12. a b c d e 29. a b 46. a b c d e 13. a b c d e 30. a b 47. a b c d e 14. a b c d e 31. a b c d e 48. a b c d e 15. a b c d e 32. a b c d e 49. a b c d e 16 a b c d e 33. a b c d e 50. a b c d e 17. a b c d e 34. a b c d e

Evaluation Form: Please rate this self-study activity by marking one response for each statement.

Did the articles meet the monograph’s stated objectives? ___Yes ___No How do you rank the quality of this educational activity? ___5 (high) ____4 ___3 ___2 ___1 (low) Comments: ______

Did you perceive any evidence of bias for or against any commercial products? ___Yes ___No If yes, please explain. Comments: ______How do you rank the effectiveness of this activity as it pertains to your practice? ___5 (high) ____4 ___3 ___2 ___1 (low) Did this material stimulate your intellectual curiosity? ___5 (high) ____4 ___3 ___2 ___1 (low)

Additional comments about this activity: ______

Name: ______

Address: ______

______

City: ______

State: ______

Phone: ______

I certify that I have completed the “Massive Transfusion Monograph” activity as designed and claim 15 credit hours in Category 1 of the Physicians Recognition Award of the American Medical Association. ______Signature Date

Mail answer form and check ($75, members; $150, nonmembers) to ITACCS Department of CME Credit, P.O. Box 4826, Baltimore, MD 21211. Allow 4 to 6 weeks for processing. Credit for this activity is offered until January 31, 2004.

Massive Transfusion and Control of Hemorrhage in the Trauma Patient 51 THIS EDUCATIONAL PROGRAM SUPPORTED, IN PART, THROUGH UNRESTRICTED EDUCATIONAL GRANTS FROM

Orginally Published October 1999 Reviewed and “reprinted” on the web January 2003