Kristan Staudenmayer, MD Stanford University, Stanford, CA

Fluid resuscitation Variety of fluids How to administer What you do DOES matter

WWII 1942 North Africa high mortality from hemorrhaghic shock By 1943, blood transfusions for shock with improved survival Korean War Renal failure in 1:200 casualties who survived

Moyer 1 & Shires 2 Hypothesized that irreversible shock was due to an imbalance in salt content in extracellular fluids. Showed could improve survival from shock with blood + crystalloid 1 Moyer CA, Burns, shock, and plasma volume regulation, 1967 2 Shires et al., Fluid therapy in hemorrhagic shock. Arch Surg, 1964

Vietnam War Use of crystalloids was the standard of care Improved survival and less renal failure Da Nang Lung Moved into civilian practice ARDS Abdominal compartment syndrome Coagulopathy Decreased gut motility Hyperglycemia

Start by talking about which fluids to use Crystalloids Colloids Hypertonic solutions Blood

Crystalloids, collioids, blood?

Crystalloids NS and Lactated ringers 1/3 remains intravascular Normal Saline Large volumes of NaCl nongap acidosis

Hypertonic Saline Solutions Resuscitation: 7.5% +/ Dextran TBI: 3%

Hopeful laboratory findings Intravascular expansion superior to LR and NS Decreased inflammation

Hypertonic Saline Safety concerns: Central pontine myelinolysis Hypertension No evidence to date have documented any adverse effects of hypernatremia (Not all trials have been designed to detect this complication among study subjects)

The largest hypertonic saline studies to date: Trauma Head injuries Pre hospital randomized controlled clinical trials Compared: 8oz NaCl vs.8oz HTS vs. 8oz HTS w/ dextran These trials was halted after enrollment of 1,000 patients because of an inability to detect a 28 day survival benefit on interval

Reports on interval outcomes still pending Hypertonic saline unproven, but still candidate for future research Military interested in continuing HTS research

Function Vary by MW in length of time they remain in the circulatory system Different molecular structures = different complications Types Albumin 6% Hydroxyethylstarch (HES) solutions Hextend: HES in lactated electrolyte solution Hespan: HES in NS Dextran Gelatins Combined Hypertonic Dextran solutions

Ex: Albumin (MW 70,000 daltons) Volume Expansion 100 ml of 25% albumin 3 5 X 500 ml of 5% albumin 500cc However, rapidly diffuses out: Estimated ~30 minutes in normal subjects Likely higher in subjects with increased intravascular permeability (eg, critically ill, sepsis, trauma, burn)

Ex: Hespan (6% hetastarch in 0.9% sodium chloride; 600,000 daltons) Volume expansion Same effect as 5% albumin Diffusion Urinary clearance 33% to 40% within 24 hours; ~62% within 72 hours Overall duration 6 36 hours

Associated complications Coagulopathy Bleeding Anaphylaxis Complications associated with specific formulas Large volumes of Hespan, Hextend may interfere with platelet function, prolong PT/PTT. Clinically Hextend has not been associated with coagulation abnormalities in doses >20 ml/kg up to a total of 5000 ml

Kaplan-Meier Estimates of the Probability of Survival Randomized Controlled Trial 16 ICUs in Australia and New Zealand 7,000 patients 4% albumin vs. 0.9% NaCl Mortality at 28 days The SAFE Study Investigators,. N Engl J Med 2004;350:2247-2256

Kaplan-Meier Estimates of the Probability of Survival The SAFE Study Investigators,. N Engl J Med 2004;350:2247-2256

The SAFE Study Investigators. N Engl J Med 2007;357:874-884

Crystalloid vs Number trials Pooled RR Mortality 95% CI Albumin 23 1.01 0.92 1.10 HES 17 1.18 0.96 1.44 Gelatin 11 0.91 0.49 1.72 Dextran 9 1.24 0.94 1.65 HTS + Dextran 8 0.88 0.74 1.05 Perel et al., Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database of Systematic Reviews 2007

Summary LR and NS are still considered the top choices Make sure brand of LR does not contain D lactate Hypertonic and saline and colloids: No evidence to support their use over crystalloids Likely that we will see further studies with more well defined populations that do benefit from these solutions. HTS and newer synthetic colloids likely will also be shown to be safe. Harm with albumin in head injured patients.

The bleeding patient who needs blood

Whole blood transfusions component therapy 1980s Allowed longer storage Improved resource utilization Products administered based on lab values But this does not work as well for actively bleeding patients

Acquired or Congenital Coagulopathy Injury Resuscitation Coagulopathy Bleeding Diatheses, medications, etc Coagulopathy of Trauma Lethal Triad

Brohi K, et al. J Trauma. Jun 2003;54(6):1127

3 4x increased mortality 8x higher mortality within 24 hours Higher incidence MOF Increased transfusion requirements Brohi K, et al., Ann Surg. May 2007;245(5):812 818. Maegele M, et al. Injury. Mar 2007;38(3):298 304

Bleeding Tissue Hypoxia Fluid Replacement RBC Transfusion Acidosis Dilution Hypothermia Coagulopathy

Borgman MA, et al. J Trauma. 2007;63:805 813.

Pulmonary morbidity may be higher with high Plasma resuscitation Transfusion related acute lung injury (TRALI) TRALI is the commonest cause of transfusion related death Anti neutrophil antibodies are commonly in the plasma from multiparous female donor TRALI frequency is higher in recipients from female donors Male only plasma policy has been adopted in many countries with marked reductions in TRALI

Gonzalez et al., J Trauma 2007 Advocated correcting INR to <1.3 with FFP Reported a high incidence of early and lethal acute respiratory distress syndrome [10]. Th e aggressive FFP transfusion was aimed Sperry, J Trauma 2008 Better survival with 1.5:1 resuscitation ARDS more common with patients with more FFP

When a patient is actively bleeding, blood products are preferred over crystalloids to better manage coagulopathy Know that bleeding trauma patients likely have a coagulopathy that needs to be addressed early. Many centers now administer plasma early during the resuscitation

The ideal use of plasma is still a matter of debate and does not come without risk. If you don t need blood or plasma, don t use blood products for volume expansion!

Maybe more fluid is not better

Pop the clot Timing and volume Early vs. delayed Large vs. small

Number of trauma victims with hypotension: 6 8% 1/3 are likely non survivable 1/3 are hypotensive due to other causes (tension pneumothorax, tamponade, drugs/etoh) Hypotensive resuscitation as a strategy is only relevant in: 1/3 of hypotensive patients 2 3% all trauma patients

Mapstone et al., J Trauma 2003 Meta analysis hypotensive strategies Normotensive versus hypotensive resuscitation Pooled risk ratio death was 0.37 (95% CI, 0.27 0.52)

Mapstone et al., J Trauma 2003 But also compared 44 animal trials of fluid versus no fluid There was marked heterogeneity in the effect of fluid resuscitation on the risk of death, much of which was explained by the hemorrhage model used.

Bickell W et al. N Engl J Med 1994;331:1105-1109

Revelation of methodological errors: Did not consider patients who died in the field Did not point out overlapping confidence intervals 95% CI Immediate resus 57 68% 95% CI Delayed resus 65 75% Re evaluation found: Initial statistics do not confirm difference When consider those who died in the field, lose all statistical differences in mortality.

Population OR death 95% CI Significant? Bickell 1994 Hypotensive pen. trauma 1.26 (1.0 1.6) No Blair 1986 Hypotensive GI bleeding 5.4 (0.3 10.7) No Turner 2000 Hypotensive trauma patients 1.06 (0.8 1.47) No Dunham 1991 Hypotensive trauma patients 0.8 (0.3 2.3) No Dutton 2002 Hypotensive trauma patients 1.0 (0.3 4.0) No Kwan I, et al., Timing and volume of fluid administration for patients with bleeding. Cochrane Database of Systematic Reviews 2003, Issue 3. Art. No.: CD002245. DOI: 10.1002/14651858.CD002245

Marshall et al. asked the question: Is it blood pressure or is it HEMODILUTION?

Used uncontrolled hemorrhage model in rats Found hemodilution was more associated with mortality and markers of shock vs. blood pressure

Hemodilution vs. Pressure Effects of Hemodilution on Long Term Survival in an Uncontrolled Hemorrhagic Shock Model in Rats. Marshall, Harry; Capone, Antonio; Courcoulas, Anita; MD, MPH; Harbrecht, Brian; Billiar, Timothy; Udekwu, Anthony; Peitzman, Andrew Journal of Trauma Injury Infection & Critical Care. 43(4):673 679, October 1997. Group I, MAP = 80 mm Hg with LR resuscitation only Group II, MAP = 80 mm Hg with Whole Blood + LR; Goup III, MAP = 40 mm Hg with LR only ; Group IV, MAP = 40 mm Hg with Whole Blood + LR. Williams & Wilkins 1997. All Rights Reserved. Published by Lippincott Williams & Wilkins, Inc. 2

Effects of Hemodilution on Long Term Survival in an Uncontrolled Hemorrhagic Shock Model in Rats. Marshall, Harry; Capone, Antonio; Courcoulas, Anita; MD, MPH; Harbrecht, Brian; Billiar, Timothy; Udekwu, Anthony; Peitzman, Andrew Journal of Trauma Injury Infection & Critical Care. 43(4):673 679, October 1997. Survival Blood loss (ml/100g) Group 1 None 6.6 151 Group 2 100% 10.4 69 Group 3 75% 4.6 29 Group 4 87% 5.3 20 LR administered (ml) Group I, MAP = 80 mm Hg with LR only Group II, MAP = 80 mm Hg with Whole Blood + LR; Goup III, MAP = 40 mm Hg with LR only ; Group IV, MAP = 40 mm Hg with Whole Blood + LR. Williams & Wilkins 1997. All Rights Reserved. Published by Lippincott Williams & Wilkins, Inc. 2

Most people moving towards judicious use of fluids until have definitive control Normal BP might not be best target Before surgical control of bleeding, perhaps should target palpable pulse and adequate cerebral perfusion if no head injury is present Further work required to support this strategy as a global solution

Crystalloids vs. colloids Which crystalloid or which colloid? When to use blood and what products to administer? Hypotensive resuscitation?

Has a lot to do with the clinical scenario

The massively bleeding trauma patient Needs blood products Minimize crystalloid Use LR, but make sure does not have D lactate Control coagulopathy early Don t wait for labs Consider early use of Plasma If prehospital and far from definitive therapy, may err on the side of hypotensive resucitation mentating and with palpable pulse.

If hypotensive but not in hemorrhagic shock Don t give blood products! Crystalloid is the first line of therapy Again, probably LR over NS to minimize acidosis Colloids can be used if not responding But understand that risks will differ depending on solution (i.e. albumin in head injured patients) And understand that they might not stick around long No evidence to support one colloid over the other.

Riskin DJ, et al. JACS. 2009;209:198.

Dysregulation Immunosuppression Inappropriate inflammation Association between infection and transfusion in trauma