HEMORRHAGIC SHOCK
SUMMARY: The Authors examined hemorrhagic shock
The first record of successful treatment of hemorrhagic shock is from the early 1800s when James Blundell transfused a woman with life-threatening postpartum hemorrhage. During World War I, W. B. Cannon recognized the importance of controlling blood loss, was the first to associate acidosis and hypothermia with hemorrhage, and advocated treating hemorrhage with alkalotic intravenous fluid. He also was the first to recommend delaying fluid administration until after operative intervention. During World War II, a greater understanding was developed of the fluid shifts that accompany hemorrhage and of the need to resuscitate with both crystalloid and blood. During the Korean and Vietnam conflicts there was greater emphasis on aggressive volume replacement and rapid access to definitive care, principles that have become standards in the treatment of acute hemorrhage. The military antishock trouser garment (MAST) was also introduced as a mechanism to maintain blood pressure prior to operative intervention.
In the early 1970s, principles of hemorrhage resuscitation learned in Vietnam were widely adopted by trauma surgeons and emergency physicians in the treatment of civilian trauma victims, with a marked improvement in their overall level of care. Recently questions have been raised about some of the standard assumptions upon which treatment is based, most notably the principle of rapid early restoration of blood pressure. Recent clinical studies have also questioned the value of interventions such as MAST and aggressive fluid therapy.
Acute hemorrhage is defined as a rapid blood loss that may accompany a wide variety of medical and surgical conditions. The most common causes of significant hemorrhage include trauma, disorders of the gastrointestinal and reproductive tracts, and vascular disease. Hemorrhagic shock occurs when blood loss is of sufficient magnitude to overcome normal physiologic compensatory responses and compromise tissue perfusion and oxygenation.
Acute hemorrhage triggers a series of physiologic responses involving the cardiovascular, respiratory, renal, hematologic, and neuroendocrine systems. The net effect of these responses is an increase in cardiac rate and contractility, a redistribution of blood flow to preserve vital organ function, conservation of water and sodium, and control of blood loss at the site of injury.
One of the very first responses observed in animal studies of acute hemorrhage is a fall in blood pressure that cannot be accounted for simply by the initial reduction in intravascular volume. It is likely that the fall in blood pressure is caused by a sudden reduction in systemic vascular resistance, although the mechanism has not been explained. The fall in blood pressure is sensed by high-pressure baroreceptors in the carotid artery sinus and the aortic arch and low-pressure baroreceptors in the left atrium and pulmonary veins. Stimulation of the baroreceptors causes disinhibition of the medullary vasomotor center, a subsequent decrease in vagal tone, and an increase in the secretion of norepinephrine (NE). Decreased vagal tone increases heart rate and cardiac output. Norepinephrine increases heart rate and myocardial contractility, stimulates renin secretion, and causes intense vasoconstriction, especially of splanchnic and musculoskeletal blood vessels. Between 20 and 30 percent of the circulating blood volume is in the splanchnic bed providing a functional reservoir that can be used to compensate for acute blood loss.
Cardiac output typically falls during hemorrhage despite increases in myocardial rate and contractility, because of the decrease in atrial filling or preload. It is commonly taught that afterload rises during acute hemorrhage in order to maintain blood pressure. There is little evidence to support this claim. It is more likely that early in the course of hemorrhage, total systemic vascular resistance falls or remains at near-normal levels in order to facilitate flow to vital organs. There are, however, increases in regional resistance that cause redistribution of blood flow away from skin, muscle, and gut to favor the brain, heart, and kidneys.
Conservation of sodium and water during hemorrhage is mediated by an increase in the levels of aldosterone and antidiuretic hormone. Stretch receptors in the afferent arterial walls of the juxtaglomerular apparatus (JGA) respond to a drop in blood pressure by stimulating an increase in renin secretion. Renin converts angiotensinogen to angiotensin I, which is then converted in the lung and liver to angiotensin II. The effects of angiotensin II include intense vasoconstriction of arteriolar smooth muscle and stimulation of aldosterone secretion by the adrenal cortex. Aldosterone is also secreted in response to elevation of potassium and adrenal corticotropin hormone (ACTH) levels, both of which occur during acute hemorrhage. Aldosterone increases the reabsorption of sodium and the excretion of potassium in the distal convoluted tubule. Water passively follows sodium and is therefore reabsorbed and conserved. Aldosterone also stimulates the secretion of the hydrogen ion, thus decreasing acidosis.
Osmo- and baroreceptors also regulate the release of arginine vasopressin or antidiuretic hormone (ADH), which is synthesized in the hypothalamus and stored in the posterior pituitary. Release of ADH occurs in response to both a fall in blood pressure and a decrease in sodium concentration. ADH increases the permeability of the renal distal tubule collecting ducts and loop of Henle to NaCl and water, with the net result being fluid and NaCl retention. In higher concentration, ADH is also a vasoconstrictor.
Acute hemorrhage causes local activation of the coagulation system. In response to injury, affected blood vessels contract, and activated platelets rapidly adhere to the edges of the damaged vessels. Platelets release thromboxane A2, which is a potent local vasoconstrictor and further platelet activator. Platelets form an unstable jelly-like plug during the first 20 min after injury. Control of hemorrhage during this period depends upon a regional reduction in flow caused by systemic hypotension and local vasoconstriction. Vessel injury exposes collagen and releases tissue thromboplastin, causing fibrin deposition in the platelet plug and gradual formation of a stable clot. The entire process for complete fibrinous transformation takes approximately 24 h.
The compensatory mechanisms described above are quite effective at maintaining critical organ perfusion even in the face of severe hemorrhage. Animal studies demonstrate complete recovery without intervention in animals bled as much as 40 percent of their estimated blood volumes. However, if the hemorrhage is not controlled, a vicious cycle of increased myocardial work and decreased perfusion eventually develops. Progressive increases in heart rate shorten diastole, with a resultant decrease in myocardial perfusion and oxygenation, as well as cardiac filling and output. The low perfusion state increases acidemia, which in turn decreases myocardial contractility.
Eventually, cardiac output becomes inadequate to maintain cellular oxygen delivery, and characteristic changes occur. The first cellular response to hypoperfusion is an attenuation of the cell membrane and an increase in sodium influx. Adenosine triphosphate (ATP) is utilized to maintain function of the sodium-potassium pump, but during periods of low flow it cannot be regenerated in sufficient quantities through the normal oxygen-dependent pathways. As the supply of oxygen and high-energy substrates diminishes, the cells revert to anaerobic metabolism to generate ATP, resulting in accumulation of lactic acid. As ATP availability decreases, sodium continues to enter the cells, causing progressive swelling, first of the cytoplasm, then the endoplasmic reticulum, and finally the mitochondria. Eventually the cells undergo clumping of mitochondria, loss of membrane integrity, and death.
There appears to be a point of no return for individual cells as well as for the overall organism in shock. Although this point is well defined for the cell, the clinician caring for patients in shock is less able to identify this landmark. It has been suggested that a sudden and substantial decrease in oxygen consumption may be a reliable marker of irreversible shock.
Factors that affect the clinical presentation of acute hemorrhage include the etiology, duration, and severity of hemorrhage and the patient's age and underlying medical condition. Acute hemorrhage most often accompanies blunt or penetrating trauma and is generally the presumed cause of shock in the trauma patient. Hemorrhage must be differentiated from other causes of shock associated with trauma, including cardiac tamponade (distinguished by elevated central venous pressure), tension pneumothorax (distinguished by unilaterally diminished breath sounds), and spinal cord injury (distinguished by the presence of neurologic deficits, warm skin, and a lower-than-expected pulse rate).
Hemorrhage not associated with trauma may present with a myriad of complaints depending upon the primary organ system involved. Nausea and vomiting; dizziness; syncope; pain in the chest, abdomen, or back; and rectal or vaginal bleeding are some common chief complaints.
The classic clinical features of acute hemorrhage include tachycardia, tachypnea, a narrow pulse pressure, decreased urine output, cool clammy skin, poor capillary refill, low central venous pressure, and, in the later stages, hypotension and altered mentation. Elderly patients and those with preexisting cardiac disease may show more severe signs and symptoms with less blood loss. Medications such as b-blockers can mask some of the early signs and symptoms of hemorrhage. On the other hand, young athletic patients can lose considerable amounts of blood before they appear ill. In general, however, there is a somewhat predictable and orderly progression of pathophysiologic events through which the patient passes as organ and cellular perfusion deteriorates and shock develops. Blood loss less than 20 percent of circulating blood volume causes cool clammy skin, delayed capillary refill, and decreased pulse pressure. Tachycardia may be present, and generally the blood pressure is normal, although the pulse pressure may narrow. As hemorrhage progresses, with blood loss of 20 to 40 percent, patients are tachycardic and tachypneic, have postural changes in blood pressure, and may be confused or agitated. If the patient is not resuscitated and the bleeding continues, hypotension and oliguria develop, respirations quicken and become deeper, tachycardia worsens, and the skin becomes mottled. With hemorrhage exceeding 40 percent of the circulating blood volume, patients commonly demonstrate tachycardia, profound hypotension, either tachypnea or irregular respirations, markedly decreased urine output, decreased or absent peripheral pulses, pallor, and lethargy or obtundation. Death from severe hemorrhage is generally marked by respiratory arrest prior to circulatory arrest, due to fatigue of the respiratory musculature and, sometimes, bradyasystolic rhythms.
There are two goals in the treatment of hemorrhagic shock: control of hemorrhage and maintenance of oxygen delivery. The definitive therapy of hemorrhage is control of the source of bleeding, and generally this requires operative intervention. Thus for most patients with hemodynamic instability secondary to hemorrhage, prompt surgical consultation and intervention are mandatory. Maintenance of tissue oxygen delivery requires, first and foremost, an assessment of the adequacy of oxygenation and ventilation. Items requiring immediate evaluation include airway patency, skin color, depth and rate of respiration, presence of any mechanical obstruction to ventilation, and presence of other factors compromising ventilation including pneumothorax, hemothorax, or flail chest. Supplemental oxygen should be administered to all patients in shock and to most patients who are acutely hemorrhaging. Many patients will require endotracheal intubation and ventilatory support. Respiratory arrest caused by fatigue of the intercostal muscles and diaphragm commonly precedes cardiac arrest as a terminal event in hemorrhagic shock, and therefore it is essential that liberal guidelines for ventilatory support are applied. Tissue oxygenation also requires restoration of circulating blood volume, and it is routine to place at least two large-bore intravenous (IV) lines for infusion of crystalloid and perhaps blood. As the IV lines are placed, initial blood samples should be drawn for type and cross matching, prothrombin time (PT), partial thromboplastin time (PTT), and a baseline complete blood count (CBC) with platelet count. All women of child-bearing age need a pregnancy test. If the shock is caused by massive gastrointestinal bleeding, liver function tests may also be helpful.
Parameters that should be routinely monitored during resuscitation of acute hemorrhage include vital signs, mentation, skin temperature, capillary refill, pulse oximetry, and urine output. There is a tendency to follow blood pressure quite closely and to gauge the adequacy of therapy by the extent to which blood pressure returns to normal levels. It is important to recognize, however, that blood pressure is an extremely crude index of the state of cellular metabolism. Also, restoration of normotension in the presence of a vascular injury may merely increase the severity of hemorrhage. Central venous pressure monitoring may be of some value in confirming the diagnosis of hypovolemia and judging the response to therapy. Swan-Ganz catheterization is usually not necessary in the acute setting except in elderly patients or those with respiratory or cardiac disease.
There has been much debate in recent years over the extent to which patients should be resuscitated prior to operative intervention, both in the prehospital setting and the emergency department. The concept of field stabilization of trauma victims has become popular despite the fact that blood loss generally cannot be controlled or corrected in the field. Standard prehospital interventions directed at restoring blood pressure, such as application of MAST and infusion of intravenous fluids, are now being subjected to careful scientific investigation.
The MAST became standard prehospital therapy in the late 1970s based on anecdotal reports of efficacy. While there is no doubt that application of the MAST often raises blood pressure, most likely through a rise in systemic vascular resistance, there is no evidence that use of MAST improves outcome. In the presence of shock and chest trauma, MAST use may increase hemorrhage severity and mortality. Thus, enthusiasm for the MAST has begun to wane except for patients with unstable pelvic fractures, for whom it may stabilize the fractures and tamponade retroperitoneal hemorrhage.
There has also been debate over the efficacy of prehospital IV line placement and fluid resuscitation. Proponents of field resuscitation state that skilled paramedics are able to place IV lines with little or no delay in transport. Opponents state that since blood loss cannot be controlled in the field, any delay in definitive treatment is excessive. Clinical studies have shown that the amount of fluid infused en route is usually minimal when compared to the total fluid requirement, and one randomized study of penetrating trauma victims has failed to show any benefit associated with preoperative fluid therapy. It is likely that prehospital fluid therapy does not affect outcome in the vast majority of cases, but it may be valuable given a specific combination of hemorrhage severity and distance from the hospital. Until conclusive data for a particular position can be obtained, it is reasonable to place IV lines once en route to the hospital whenever possible. This practice avoids potentially lethal delays in the field and grants the patients the potential benefits of prehospital fluid therapy.
It should be noted, however, that the benefits of early and aggressive fluid replacement in victims of acute hemorrhage remain unproved whether given in the prehospital setting or the emergency department. Many animal studies have shown that raising the blood pressure with either vasopressors or fluid also worsens mortality, sometimes dramatically. Investigations continue in an attempt to determine the ideal rate and volume of fluid administration as well as the appropriate therapeutic endpoint of resuscitation. At the present time the amount and type of volume expander used depends primarily on the clinical status of the patient and to a lesser extent on individual institutional preference. In most hospitals, isotonic crystalloids (0.9 % NaCl or Ringer's lactate) are the agents of choice for the initial management of acute hemorrhage. Standard therapy of the hemodynamically stable patient is rapid infusion of 20 to 40 mL/kg. If the adult patient continues to show signs of impaired perfusion after a total of 30 mL/kg (roughly 2 L), it is likely that blood loss exceeds 15 percent of the total blood volume. At this point, it is appropriate to begin red blood cell transfusions, particularly if blood loss has not been controlled. If the patient appears to be stable, it is usually possible to wait for fully cross-matched blood, but that decision must be individualized, based on the assessment of ongoing blood loss and the efficiency of the local blood bank. When in doubt, it is advisable to use type-specific blood. Several studies have shown this to be a very safe practice, and delays in providing needed oxygen-carrying capacity are potentially more harmful to the patient. Early blood therapy is particularly important in the elderly and in those with significant respiratory or cardiac disease.
More aggressive therapy is mandated in the hemorrhaging patient exhibiting any degree of hemodynamic instability or signs of end organ hypoperfusion. These patients almost always require blood transfusions, and it is appropriate to begin type-specific blood early unless there is a prompt and persistent improvement in perfusion with saline solution alone. Type-specific blood is indicated in patients who are profoundly hypotensive on initial presentation, those who remain in shock after crystalloid infusion, and those who demonstrate rapid ongoing hemorrhage. Continued administration of crystalloid without blood may result in profound dilution of the remaining red blood cell mass, platelets, and coagulation factors. It may also disrupt clot formation in the injured vessels. Volume restored at the expense of oxygen-carrying capacity and hemostasis is of questionable therapeutic value.
The moribund patient requires even more prompt restoration of circulating red blood cell mass. In this case, type O blood should be used immediately if it is available. Type O Rh-negative blood should be given to females of child-bearing age. In most other situations type O Rh-positive blood is preferred because of its greater availability. A sample for type and cross match should always be drawn and sent before administration of type O blood.
Clotting abnormalities are commonly noted after massive transfusion (equivalent to one blood volume, or 70 to 80 mL/kg) requiring transfusion of fresh-frozen plasma (FFP). One unit of FFP and six platelet packs are often recommended for every 5 to 10 units of blood transfused. These are general guidelines, however, and transfusion of these agents is based ideally on clinical evidence of impaired hemostasis and frequent monitoring of coagulation parameters. Platelets are indicated in the actively bleeding trauma patient with a platelet count of 50,000 or less. FFP is indicated if the PT is prolonged more than 1.5 times normal (usually >18). When an underlying coagulation disorder is suspected, such as in patients taking coumadin or with evidence of severe liver disease, it may be appropriate to administer FFP without waiting for laboratory confirmation.
Autologous whole blood may be given if the hemorrhage is intrathoracic and the capabilities for autotransfusion exist Autotransfusion decreases the risk of transmission of diseases such as AIDS and hepatitis and it also decreases the demand on the blood bank. There has been some discussion concerning autotransfusion in patients with intraabdominal injuries. It can be difficult to determine, especially in the emergency department, if there is fecal contamination of intraabdominal blood. Transfusion of contaminated blood has not been shown to be safe and it may be more prudent to autotransfuse blood from intraabdominal injuries only in the operating suite, after the source of blood has been discovered and the risk of transfusing contaminated blood is known.
Although isotonic saline solution is most commonly used in the initial management of hemorrhagic shock, debate continues over the fluid of choice (crystalloid vs. colloid). Albumin has fallen into disfavor, but purified protein fraction (PPF) and FFP continue to be recommended and used. Central to the issue are the effects of fluid resuscitation on the pulmonary interstitium. Proponents of protein replacement argue that saline resuscitation of hemorrhage results in a fall in intravascular oncotic pressure and a reversal of the normal gradient favoring intravascular fluid retention. Theoretically, this may lead to pulmonary edema and impaired tissue oxygenation. Colloid administration is advocated because it raises oncotic pressure in the pulmonary capillary bed. This argument ignores the fact that the pulmonary capillary endothelium permits considerable flow of fluids, including plasma proteins, between the capillaries and the interstitium. A fall in intravascular oncotic pressure is compensated for by a fall in pulmonary interstitial oncotic pressure, thereby minimizing changes in the pressure gradient. It appears likely that pulmonary capillary hydrostatic pressure (measured as pulmonary artery wedge pressure) is far more important than pulmonary capillary oncotic pressure in determining the amount of fluid flowing to the interstitium. Maintenance of the wedge pressure below 15 mmHg is probably the most important factor in preventing pulmonary edema. To date, there are few data to show that colloids are harmful, but the inability to convincingly demonstrate beneficial effects in scores of animal and clinical studies suggests that benefits are minimal. Clinicians inclined to use albumin, PPF, or FFP in the resuscitation of hemorrhagic shock should question whether the undocumented benefits of this therapy are worth the substantial increase in cost or, in the case of FFP, the risk of disease transmission.
Alternatives to the use of naturally occurring colloid preparations include synthetic colloid solutions such as hydroxyethyl starch (HES) and dextran 70. The volume-expanding properties of HES are equivalent to those of 5% albumin. These agents differ significantly from albumin, however, in that they remain predominantly in the intravascular space because of their high molecular weight and branched structure. Their plasma-expanding effects are more prolonged than those of albumin, and interstitial edema is not a significant concern.
The combination of 7.5% hypertonic saline and 6% dextran 70 (HSD) is an agent that has shown considerable promise in animal studies of acute hemorrhage. Given in small amounts HSD causes a prompt and long-lasting shift of fluid from the interstitial to the intravascular compartment, and therefore it was initially thought to be ideal for the prehospital resuscitation of hemorrhage. Thus far, clinical studies have not demonstrated an improvement in outcome with HSD, but there may be particular injury patterns (such as combined hypovolemia and head injury) in which a benefit exists.
- Bellamy RF, Maningas PA, Wenger BA: Current shock models and clinical correlations. Ann Emerg Med 15:12, 1392, 1986.
- Cayten CG, Murphy JG, Stahl WM: Basic life support versus advanced life support for injured patients with an injury severity score of 10 or more. J Trauma 35:460, 1993.
- Development Task Force of the College of American Pathologists: Practice parameters for the use of fresh-frozen plasma, cryoprecipitate, and platelets. JAMA 271:(10), 777, 1994.
- Maddox KL, Bickell WH, Pepe PE, et al: Prospective randomized evaluation of antishock MAST in postraumatic hypotension. J Trauma 26:779, 1986.
- Martin RR, Bickell WH, Pepe PE, et al: Prospective evaluation of preoperative fluid resuscitation in hypotensive patients with penetrating truncal injury: A preliminary report. J Trauma 33:3, 354, 1992.
- Peter RM, Hargens AR: Protein vs electrolytes and all of the Starling forces. Arch Surg 116:1293, 1981.
- Stern SA, Dronen SC, Birrer P, Wang X: Effect of blood pressure on hemorrhage volume and survival in a near-fatal hemorrhage model incorporating a vascular injury. Ann Emerg Med 22:155, 1993.