Page 32 - Journal of Special Operations Medicine - Fall 2014

crystalloid solution. Other HES variants may have The detrimental effect of crystalloids on TBI has been
different mean molecular weights or varying ratios of observed in animal models. In a swine model of TBI and
hydroxyethyl group substitutions. The HES molecules hemorrhagic shock (40% blood volume controlled hem-
may also be dissolved in different solutions. orrhage), the animals were resuscitated with NS, Hex-
tend, or FFP. The volumes of Hextend and FFP matched
A meta-analysis of 19 reports (1567 patients) studying the shed blood volume; NS was administered at 3 times
the use of 6% HES solution in surgical patients found the shed blood volume. The outcome measure was brain
no increase in the incidence of postoperative death or lesion size. Plasma reduced the size of the brain lesion.
acute kidney injury in patients who received HES. The Hextend did not reduce the size of the brain lesion but
130
HES solutions in this study had a variety of molecular reduced the amount of edema associated with the lesion
weights and molar substitutions. The other fluids used in comparison to that produced by NS resuscitation.
134
were an assortment of different colloids and crystal- Cerebral edema is a major concern in casualties who
loids. A Cochrane Review concluded that neither HES sustain moderate to severe TBI in addition to hemor-
nor dextran has been shown to improve survival in hy- rhagic shock.
povolemic patients compared with crystalloids. 131
Another retrospective study examined HES use in
A recent article by Zarychanski et al noted an associa- 2225 trauma patients; 497 patients (22%) received
tion between HES administration, acute kidney injury, 6% HES (450/0.7) within 24 hours of admission to
and increased mortality. This meta-analysis of 38 tri- the hospital. (Note that Hextend has a different mo-
als did not focus specifically on hemorrhagic shock; it lecular weight and molar substitution [670/0.75] than
also included patients with diagnostic descriptors such 6% HES [450/0.7].) Acute kidney injury was defined
as sepsis, burns, “ICU patients,” and “post-cardiac ar- as a rise in creatinine greater than 2 times baseline. ISS
rest” as well as trauma. Patients in some studies were was greater in the HES group (29.7) compared with
described as “trauma” or “hypovolemia.” Neither the no-HES group (27.5). Patients who died within 24
131
of these two terms is synonymous with “hemorrhagic hours of admission were excluded. This is a significant
shock.” Outcomes after resuscitation with HES in a het- limitation of the study because individuals who die
erogeneous patient population may not reflect the ef- from hemorrhagic shock often do so within the first
fects of HES in patients with hemorrhagic shock. 24 hours and exclusion of these patients introduces the
potential for a survival bias. The mortality was 21%
The Zarychanski et al. report included HESs of various in the HES group and 11% in the no-HES group. The
concentrations, various molecular weights, and various incidence of acute kidney injury was 13% in the HES
molar substitution ratios. As they note, different types group and 8 % in the no-HES group. The mean in-
of starch solutions may have different physiologic ef- fused volume of HES was 725mL. Other options for
fects. Results after treatment with an assortment of fluid resuscitation included RBCs and plasma; the re-
101
HES options do not necessarily reflect the effects of any port notes that “there were no resuscitation protocols
single solution. The total volume of HESs infused in all in place during the study period.” The conclusion from
of the trials reviewed by Zarychanski et al. was not well this study was: “Because of the detrimental association
captured, but some of the volumes noted were well in ex- with renal function and mortality, hetastarch should be
cess of that recommended for the prehospital treatment avoided in the resuscitation of trauma patients.” The
of hemorrhagic shock in battlefield trauma care. 13,14 The study also notes that: “It has been argued that damage
study done at Ryder Trauma center in Miami, FL, used control resuscitation of a massively bleeding patient
Hextend at the volume recommended by the US military with plasma and blood may be beneficial. In this re-
(a 500mL bolus followed by a second 500mL bolus if gard, abandoning synthetic colloids in favor of plasma
required) and found no increased incidence of acute kid- may be appropriate.” 135 Since this fluid resuscitation
ney injury due to Hextend. was carried out in the hospital where blood products
132
were available, both TCCC and the Joint Trauma Sys-
Of note also is that the Zarychanski et al. report did tem Clinical Practice Guidelines would recommend
not address other potential complications of crystalloid that damage control resuscitation be accomplished
or colloid fluid resuscitation such as abdominal com- with 1:1:1 plasma, PRBCs, and platelets. Crystalloids
partment syndrome, ARDS, and worsening of cerebral and colloids are clearly not the preferred fluid for re-
edema in TBI. The increased extravascular distribution suscitation from hemorrhagic shock when blood com-
131
of crystalloids must be considered in selecting a prehos- ponents are available. 13,49
pital resuscitation fluid; crystalloids have been shown to
produce an increase in these complications, 28,53,81,133 as The FDA issued a safety communication on HESs (Hes-
well as an increase in mortality. pan, Hextend, and Voluven) in November 2013. The
25

24 Journal of Special Operations Medicine Volume 14, Edition 3/Fall 2014