fluid replacement for the treatment of heamorrhagic shock: a 21st century perspective
Matthew G. Ratzlaff UCD School of Medicine and Medical Sciences, University College Dublin, Belfield, Dublin 4, Ireland
Jennifer Thompson MD FRCSI UCD School of Medicine and Medical Sciences, University College Dublin, Belfield, Dublin 4, Ireland
Haemorrhagic shock is a common and life-threatening consequence of trauma, gastrointestinal bleeds, and other pathologies. The main goals of treating haemorrhagic shock are to achieve haemostatic control and restore intravascular volume. The most common fluids currently used in fluid resuscitation are crystalloids, colloids, hypertonic solutions, allogeneic blood, and artificial oxygen carriers. While blood is the best form of fluid replacement, especially in severe haemorrhagic shock, its limited availability necessitates the use of other fluids. Crystalloids are the most commonly used fluid replacement class in the pre-hospital setting but must be delivered in large quantities and can therefore cause pulmonary oedema and other complications. Artificial oxygen carriers are a relatively recent development and show promise of providing optimal oxygen delivery to tissues until blood transfusions are available. Since World War II, the accepted treatment approach for haemorrhagic shock has been to restore physiological parameters as soon as possible with aggressive fluid replacement. While this strategy may be beneficial for patients with controlled haemorrhagic shock, recent research shows that patients with uncontrolled haemorrhagic shock may require hypotensive resuscitation to achieve the optimal balance between haemostasis and perfusion of vital organs. If intravenous administration of fluids is impossible, alternate routes include intraosseous needles, intraperitoneal delivery, nasogastric tubes, and rectal delivery. Finally, under-resuscitation can be avoided by monitoring more accurate end-point markers of fluid replacement, such as oxygen debt, lactate, and inflammatory markers.
Shock is a state of circulatory failure that leads to life-threatening hypoperfusion of vital organs . Initially, in 1934, Alfred Blalock identified 4 categories of shock: hypovolemic, vasogenic, cardiogenic and neurogenic . The commonest of these, hypovolemic shock, results in a loss of circulating blood volume due to either haemorrhage (internal or external) or increased vascular permeability and dilatation (1). Haemorrhage is regularly encountered by physicians in emergency departments, operating rooms, and intensive care units and requires rapid medical intervention . The estimated blood volume for a 70 kg person is approximately 5 L and represents 7% of total body weight . Haemorrhage is divided into 4 classes (see Table 1), with Class I (blood loss <750 mL) being a non-shock state, while Class IV (blood loss >2 L) is characterised by signs of severe shock (eg, tachycardia, hypotension, tachypnea, anuria, lethargy) and requires immediate therapy .
The most common cause of haemorrhagic shock is trauma, which in turn is the leading cause of death worldwide in people between the ages of 5 and 44 years . Particularly important mechanisms of trauma include lacerations, penetrating wounds to the abdomen and chest, and ruptured major vessels . Other common causes of haemorrhagic shock include gastrointestinal bleeding (eg, from oesophageal varices), massive internal bleeding from long bone fractures and solid organ injuries, antithrombotic therapy, coagulopathies, obstetric/ gynaecologic causes (eg, ruptured ectopic pregnancy), pulmonary causes (eg, lung cancer), ruptured aneurysms and retroperitoneal bleeding . The pathophysiological consequences of prolonged hypoperfusion include metabolic acidosis, loss of cell membrane integrity , and an aggressive inflammatory response that can result in irreversible multi-organ failure . Therefore, basic therapeutic goals for haemorrhagic shock are to provide fluid resuscitation as well as control bleeding, coagulation support, and maintenance of normothermia . In recent years, there have been many important advances in fluid resuscitation with regards to classes of fluid replacements available, strategies for fluid administration, and therapeutic end-points for which to monitor .
Classes of fluid replacement for hemorrhagic Shock
The most common types of fluids used in the treatment of haemorrhagic shock include crystalloids, colloids, hypertonic solutions, allogeneic blood, and artificial oxygen carriers .
For the past 40 years, the gold standard for treating trauma victims in haemorrhagic shock has been to infuse large volumes of crystalloids early and rapidly, especially when blood products are not available . The most commonly used crystalloids are lactated Ringer’s solution and normal isotonic saline solution . Lactated Ringer’s solution, a mixture of salts that is isotonic with blood, is relatively safe and inexpensive, and equilibrates rapidly through the extracellular compartment, restoring the extracellular fluid deficit that accompanies haemorrhage . Another advantage of lactated Ringer’s solution is the generation of bicarbonate from the metabolised lactate, which buffers against metabolic acidosis associated with haemorrhagic shock . Unfortunately, lactated Ringer’s solution has negative effects on the immune response to haemorrhagic shock, including increased neutrophil superoxide burst activity, increased neutrophil adherence, and increased cytokine activation (eg, IL-1, IL-6, and TNF) . The large volumes of crystalloid required for adequate resuscitation can lead to decreased intravascular oncotic pressure  and subsequent pulmonary oedema . Furthermore, a study using an experimental haemorrhagic shock model in pigs showed that fluid resuscitation with lactated Ringer’s solution was inferior to blood or gelatine (a colloid) at improving mucosal tissue oxygenation of the small intestine .
Because crystalloids were known to cause pulmonary oedema (“shock lung”), in the 1970s, focus was placed on the development of hyperosmotic solutions called colloids (eg, gelatins, dextrans, and hydroxyethyl starches) with a primary goal of improving pulmonary function during fluid resuscitation for haemorrhagic shock . Their use has since been advocated because unlike crystalloids, colloids remain in the intravascular compartment and so a lower volume is required to attain hemodynamic stability . On the other hand, colloids are expensive, lower calcium and immunoglobulin levels in the blood, and may deplete the extracellular fluid volume . Furthermore, a recent Cochrane review assessing fluid resuscitation in critically ill patients with trauma, burns, or following surgery found no reduction in mortality with use of colloids compared to crystalloids . There is also concern that colloids can cause hyperoncotic acute renal failure, but a recent study using an experimental model of severe haemorrhagic shock in rabbits showed that 6% hydroxyethyl starch either alone or in combination with Ringer’s lactate improved renal glomerular function and did not have a harmful effect on the kidney .
The use of hypertonic solutions (eg, hypertonic saline) has been investigated since the 1980s, and while their use in humans has not yet been approved by the US Food and Drug Administration, they show promise as a fluid replacement therapy . Experimental models of haemorrhagic shock in animals have shown that small volume hypertonic saline is as effective as large volume crystalloids in expanding plasma volume, improving cardiac output and microcirculation, restoring renal function, and reducing acute lung injury and red blood cell injury . Hypertonic saline is shown to be of particular benefit to trauma patients with combined head injury and haemorrhagic shock, because it increases cerebral perfusion while decreasing intracranial pressure and cerebral oedema . Furthermore, a recent study comparing the ability of hypertonic saline plus dextran (HSD) and crystalloids to prevent harmful immunologic effects in haemorrhagic trauma patients found that HSD significantly inhibited neutrophil activation, inhibited proliferation and cytokine production of pro-inflammatory monocytes, and stimulated proliferation and cytokine production of anti-inflammatory monocytes .
Allogeneic blood transfusion is invaluable for the pre-hospital and in-hospital treatment of severe haemorrhagic shock because blood is currently the only substance that can fully re- store oxygen-carrying capacity . A recent study using an experimental haemorrhagic shock model in dogs showed that while blood, Oxyglobin (an artificial oxygen carrier), saline (a crystalloid), and 6% hetastarch (a colloid) were able to restore microvascular and systemic function, only blood was able to restore oxygenation changes to pre-haemorrhagic levels . Unfortunately, blood is not readily available in the pre-hospital setting due to the necessity of refrigeration and blood typing .
In patients with no known risk factors, blood transfusions should be delivered when blood loss from haemorrhage exceeds 30% of blood volume (Class III haemorrhage) or when the haemoglobin level drops below the threshold of 7-8 g/dL . However, blood transfusions should be used to maintain haemoglobin at 10 g/dL in patients who are actively bleeding, the elderly, and those at risk for myocardial infarction . If type and crossmatched blood is unavailable, O-negative blood should be given . Despite the advantages of allogeneic blood, it is in limited supply and carries multiple potential adverse effects . These include infectious complications (eg, hepatitis, HIV, and bacterial contamination), immune reactions, metabolic complications (eg, hyperkalemia, hypocalcemia, and citrate toxicity) and mistransfusion .
Artificial oxygen carriers
Artificial oxygen carriers are a relatively new development and show promise of providing a better oxygen carrying capacity than crystalloids, colloids, and hypertonic solutions while avoiding the problems of storage, compatibility, and disease transmission associated with blood transfusions . The three types of artificial oxygen carriers currently under investigation are perfluorocarbons, haemoglobin-based oxygen carriers, and haemoglobin vesicles .
Perfluorocarbons are synthesized by halogenating cyclic or straight-chain hydrocarbons (5). They have a long shelf life, minimal infectious or immunogenic effects (2), and are an effective initial replacement for blood transfusions in patients with haemorrhagic shock . However, perfluorocarbons undergo rapid plasma clearance  and are known to be associated with neurological complications and postoperative ileus .
Haemoglobin-based oxygen carriers (eg, PolyHeme) are derived from human or bovine sources and are thought to act by scavenging nitric oxide, increasing release of endothelin, and stimulating endothelin receptors and adrenoreceptors . While they have a high oxygen carrying capacity and prolonged shelf life, some disadvantages include hypertensive effects, immunogenic effects, and potential renal toxicity . Moreover, a recent study in patients with haemorrhagic shock revealed a significantly higher risk of myocardial infarction associated with fluid resuscitation using PolyHeme plus blood (3%) compared to crystalloid plus blood (1%) .
Lastly, haemoglobin vesicles consist of purified human haemoglobin encapsulated by phospholipid vesicles . While their use has been limited to experimental studies of haemorrhagic shock, haemoglobin vesicles have been shown to maintain systemic oxygenation without producing hypertensive or immunogenic effects .
Treatment strategies for fluid RESUSCITATION
Aggressive versus hypotensive resuscitation
From the time of World War II until recently, the accepted therapeutic dogma for the treatment of haemorrhagic shock has been to replenish blood volume rapidly and to attain normal physiological parameters . This view was reinforced during the Vietnam War due to the observation that aggressive fluid resuscitation with red blood cells, plasma, and crystalloid solutions appeared to improve survival in trauma patients . However, this dogma is now being challenged by clinical trials and experimental animal models that have differentiated between controlled and uncontrolled haemorrhagic shock and identified key differences in their responses to aggressive fluid replacement .
In controlled haemorrhagic shock, fluid resuscitation is aimed at normalising haemodynamic parameters  because the source of bleeding has been occluded (eg, by the formation of a clot) . In contrast, uncontrolled haemorrhagic shock is characterised by ongoing bleeding, and therefore a failure to achieve haemostasis . As such, attempts to normalise vital signs in someone with uncontrolled haemorrhagic shock may lead to volume overload that prevents clot formation at the site of injury and leads to renewed bleeding .
It is now becoming clear that using smaller doses of fluid replacement—hypotensive resuscitation—achieves optimal survival in uncontrolled haemorrhagic shock . A recent study using an experimental uncontrolled haemorrhagic shock model in guinea pigs compared the efficiency of aggressive fluid resuscitation, low-volume fluid resuscitation and permissive hypotensive resuscitation therapy approaches using crystalloids and colloids and found that survival time was significantly higher in the permissive hypotensive resuscitation groups . In the pre-hospital setting, it is recommended that trauma victims with uncontrolled haemorrhagic shock receive repeated aliquots of 250 mL of lactated Ringer’s solution during evacuation, with a goal of maintaining a systolic blood pressure of 80 mmHg as well as controlling bleeding .
Immediate versus delayed resuscitation
Related to the dogma of aggressive resuscitation for haemorrhagic shock during World War II and the Vietnam War was the belief that early resuscitation was key to survival . For many years, the thinking of trauma surgeons was dominated by the concept of the “golden hour”, the time period in which shock could be reversed and organ damage prevented . However, it was shown that in patients with haemorrhagic shock from penetrating truncal injuries, a higher discharge rate and fewer complications were seen in those who received delayed resuscitation with lactated Ringer’s solution compared to those who received immediate resuscitation . Findings such as this indicate that patients with moderate hypotension from modest bleeding may benefit from postponing heavy fluid replacement until arrival at a definitive care facility . However, early provision of intravenous crystalloids, colloids, or blood products can be life saving in patients who are in severe haemorrhagic shock .
Routes of delivery of fluids
While large-bore peripheral intravenous lines are the most widely used method of administration of fluid replacement, alternate routes must be sought when haemorrhagic shock and compensatory vasoconstriction make intravenous access exceedingly difficult . A common route of fluid replacement in the paediatric setting is via an intraosseous needle, which is rapid but requires special equipment and sterile conditions and carries a risk of osteomyelitis and compartment syndromes . Intraperitoneal fluid replacement is used to improve organ perfusion and like intraosseous needles, requires special skills and equipment, sterile conditions, and increases the risk of peritonitis . Nasogastric tubes allow easy administration of fluids and do not require sterile conditions, but the amount of fluid necessary to reverse haemorrhagic shock increases the risk of vomiting and aspiration . A relatively new route for administration of fluid resuscitation and anti-shock drugs is across the rectal mucosa, which is relatively easy, painless, efficient, and does not require special skills or sterile conditions . A study using an experimental model of hypovolemic shock in rabbits showed that compared to no treatment, administration of 0.9% sodium chloride solution via the rectum resulted in a significantly raised mean arterial pressure .
Monitoring end points FOR FLUID RESUSCITATION
The goals of fluid resuscitation are to normalise blood pressure, heart rate, and urine output while keeping central venous pressure in an adequate range (8-15 mmHg) (5). If blood pressure and urine output alone are used as end-points for fluid replacement in haemorrhagic shock, up to 85% of patients may be under-resuscitated . This occurs because measured normalisation of gross physiologic parameters does not necessarily correspond to reperfusion of cells via the microvasculature . Normalisation of oxygen transport variables, oxygen delivery, cardiac index, oxygen consumption, lactate, base deficit, and mucosal gastric pH represent more accurate therapeutic end-points . Some of these variables are currently monitored indirectly, but experimental and clinical studies using new technologies are uncovering more direct and accurate methods (eg, transcutaneous partial oxygen tension) of monitoring patients receiving fluid resuscitation .
Other important end-points are pro-inflammatory cytokine levels (eg, TNF-alpha, IL-1, and IL-6), as excessive inflammatory reactions in ischemic tissues have been shown to adversely influence prognosis in patients being treated for haemorrhagic shock . New biotechnological methods are being utilised to identify immune cell surface markers (eg, HLA-DR), expression of pro- and anti-inflammatory mediators, and genes for cytokine expression .
Haemorrhagic shock is a common and life-threatening consequence of trauma, gastrointestinal bleeds, and other pathologies. The main goals of treating haemorrhagic shock are to stop the bleeding and restore intravascular volume. While blood is the best form of fluid replacement, especially in severe haemorrhagic shock, its limited availability necessitates the use of other fluids. Crystalloids are the most commonly used fluid replacement class in the pre-hospital setting, but the risk of pulmonary oedema and other complications has motivated the development of new classes such as hypertonic solutions and artificial oxygen carriers. The old dogma of early and aggressive fluid resuscitation for all haemorrhagic shock victims is now being challenged by an evolving understanding of controlled and uncontrolled haemorrhagic shock, with the latter requiring hypotensive resuscitation to achieve the optimal balance between haemostasis and perfusion of vital organs. Ongoing research into the optimal routes of delivery and end-point markers of fluid resuscitation show promise of improving outcomes for distinct populations of patients suffering from haemorrhagic shock.
Thank you to Dr. Jennifer Thompson for her editing suggestions.
1. Holmes CL, Walley KR. The evaluation and management of shock. Clin Chest Med 2003;24(4):775-789.
2. Krausz MM. Initial resuscitation of hemorrhagic shock. World Journal of Emergency Medicine 2006;1(1):14-18.
3. Gutierrez G, Reines HD, Wulf-Gutierrez ME. Clinical review: hemorrhagic shock. Critical Care 2004;8(5):373-381.
4. Committee on Trauma: Advanced Trauma Life Support Manual. Chicago: American College of Surgeons; 1997:103-112.
5. Angele MK, Schneider CP, Chaudry IH. Bench-to-bedside review: latest results in hemorrhagic shock. Critical Care 2008;12(4):218-230.
6. Knotzer H, Pajk W, Maier S, Dunser MW, Ulmer H, Shwarz B, et al. Comparison of lactated Ringer’s, gelatine and blood resuscitation on intestinal oxygen supply and mucosal tissue oxygen tension in haemorrhagic shock. British Journal of Anaesthesia 2006;97(4):509-516.
7. Perel P, Roberts I. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database of Systematic Reviews 2011, Issue 3.
8. Adanir T, Aksun M, Cirit M, Alkan Tasli F, Sahin O, Kestelli M, et al. The renal effect of replacement fluids in controlled severe hemorrhagic shock: an experimental study. Turkish Journal of Trauma & Emergency Surgery 2009;15(5):423-432.
9. Rizoli SB, Rhind SG, Shek PN, Inaba K, Filips D, Tien H, et al. The immunomodulatory effects of hypertonic saline resuscitation in patients sustaining traumatic hemorrhagic shock: a randomized, controlled, double-blind trial. Ann Surg 2006;243(1):47-57.
10. Cheung ATW, To PL, Chan DM, Ramanujam S, Barbosa MA, Chen PCY, et al. Comparison of treatment modalities for hemorrhagic shock. Artificial Cells, Blood Substitutes, and Biotechnol- ogy 2007;35(2):173-190.
11. Moore EE, Moore FA, Fabian TC, Bernard AC, Fulda GJ, Hoyt DB, et al. Human polymerized hemoglobin for the treatment of hemorrhagic shock when blood is unavailable: the USA multicenter trial. J Am Coll Surg 2009;208(1):1-13.
12. Durusu M, Eryilmaz M, Ozturk G, Mentes O, Ozer T, Deniz T. Comparison of permissive hypotensive resuscitation, low-volume fluid resuscitation, and aggressive fluid resuscitation therapy approaches in an experimental uncontrolled hemorrhagic shock model. Turkish Journal of Trauma & Emergency Surgery 2010;16(3):191-197.
13. Orgiler UPE, Navarro AJM, De Haro MS. The intraosseal route. When the veins have disappeared. Enferm Intensiva 2001;12:31-40. Cited by: Girisgin AS, Acar F, Cander B, Gul M, Kocak S, Bodur S. Fluid replacement via the rectum for treatment of hypovolemic shock in an animal model. Emerg Med 2006;23(11):862-864.
14. Banerjee S, Singhi SC, Singh S, Singh M. The intraosseous route is a suitable alternative to intravenous route for fluid resuscitation in severely dehydrated children. Indian Pediatr 1994;31:1511-1520. Cited by: Girisgin AS, Acar F, Cander B, Gul M, Kocak S, Bodur S. Fluid replacement via the rectum for treatment of hypovolemic shock in an animal model. Emerg Med 2006;23(11):862-864.
15. Garrison RN, Conn AA, Harris PD, Zakaria, ER. Direct peritoneal resuscitation as adjunct to conventional resuscitation from hemorrhagic shock: a better outcome. Surgery 2004;136:900-908. Cited by: Girisgin AS, Acar F, Cander B, Gul M, Kocak S, Bodur S. Fluid replacement via the rectum for treatment of hypovolemic shock in an animal model. Emerg Med 2006;23(11):862-864.
16. Nager AL, Wang VJ. Comparison of nasogastric and intravenous methods of rehydration in pediatric patients with acute dehydration. Pediatrics 2002; 109:566-572. Cited by: Girisgin AS, Acar F, Cander B, Gul M, Kocak S, Bodur S. Fluid replacement via the rectum for treatment of hypovolemic shock in an animal model. Emerg Med 2006;23(11):862-864.
17. Girisgin AS, Acar F, Cander B, Gul M, Kocak S, Bodur S. Fluid replacement via the rectum for treatment of hypovolemic shock in an animal model. Emerg Med 2006;23(11):862-864.