Page 89 - Journal of Special Operations Medicine - Spring 2014

and adjusted the pneumatic pressure bag, ensuring a Δ scores. A MANOVA with use of the Wilks’ λ test
constant 300mmHg of pressure was applied. The control were used to determine if there were differences in the Δ
group was exsanguinated the same as the other groups scores (2, 4, 6, and 8 minutes) relative to SBP, DBP, HR,
but given no Hextend. The same data were collected ev- MAP, CO, and SV. All data in the tables are presented in
ery 2 minutes for 8 minutes for all groups. A control means and SDs (Tables 1–6 and accompanying Figures).
group was used to make sure that any changes in the out-
come variables (vital signs and hemodynamics) were be-
cause of the Hextend administration and not because of Discussion
the effects of endogenous catecholamines or autotransfu- The purposes of this study were to compare the time
sion. Swine have a substantial amount of blood volume and hemodynamics of IV and IO routes with the admin-
(20%–25% of circulating red blood cells) sequestered in istration of 500mL of Hextend in a Class II hemorrhage
the spleen that is released during hypovolemic shock. 24,25 swine model.

Although the administration of Hextend via the IO route
Results
was faster than via the IV route, the difference was not
An α of .05 was used for all analyses. Means and stan- statistically significant. The reason that the time for the
dard deviations (SDs) were calculated for each group. A IV group was slower than the IO group may because the
multivariate analysis of variance (MANOVA) was used IV needle was smaller (18 gauge) compared with the IO
to determine if there were significant differences be- needle (15 gauge). The reason the same needle size for
tween the groups relative to weight, age, blood volume, both the IV and IO groups was not used because each is
the amount of exsanguinated blood, NPO time, NPO the standard of care according to the TCCC.
fluid replacement, oxygen saturation, end-tidal carbon
dioxide, SBP, DBP, HR, MAP, CO, SV, and tempera- When a patient is in hypovolemic shock, their veins
ture. There were no significant differences between the have collapsed, making IV access difficult and time
groups on these data, indicating that the groups were consuming. Our experience in this study was that the IO
equivalent on these parameters (p > .05). could be successfully inserted in about 10 seconds. We
consistently found no statistical significant difference in
The time to administer Hextend for the IO group was hemodynamics between the IO and IV routes of admin-
8 minutes 57 seconds (SD ±2 minutes 30 seconds), and istration but significant differences between IO versus
the time for the IV group was 9 minutes 53 seconds control and IV versus control groups. Therefore, based
(SD ±2 minutes 50 seconds). An independent t test indi- on these findings, we recommend that the IO might be
cated that there were no significant differences between considered the first choice for administration of Hex-
the IO and IV groups relative to the time to administer tend on the battlefield or in a mass casualty scenario.
Hextend (p = .78). The hemodynamic data were col- Even with a skilled clinician under the best of circum-
lected again immediately after 30% of the swine’s blood stances, the time to gain IV access would certainly be
volume was exsanguinated. A MANOVA and the use longer than 10 seconds.
of the Wilks’ λ indicated there were differences in the
groups relative to hemodynamics (p < .05); therefore, a Future studies should be implemented using a larger
post-hoc Least Significance Difference test (LSD) was sample size. Additionally, a longer period of time be-
performed. Because there were no significant differences tween the hemorrhage and the administration of Hex-
in the groups and no identified differences in treatment tend should be considered. Such an approach would
before hemorrhage (BH), we anticipated there would simulate a more realistic approach to the time it takes
not be any significant differences in the hemodynam- to render care to a patient on the battlefield. Other stud-
ics immediately after the hemorrhage (AH). However, ies using Class III and IV hemorrhages should be imple-
the SBP and MAP of the IV group were significantly mented using the same model as this study. Additionally,
lower than those of the control group AH (p < .05). sternal and tibial IO sites should be used to determine
Theoretically, these data should have been the same in which site is most effective.
each group because there were no differences by group
relative to weight, anesthesia, hemodynamics, amount Study Limitations
of blood volume, the amount of hemorrhage, etc. The This study was performed with swine, and the auricular
differences cannot be explained except by individual vein was used for the IV site. The site may not be the same
variations. Because there were significant differences diameter as the human antecubital veins in all individu-
between the groups in the hemodynamics AH, we cal- als. However, Coles and colleagues examined 24 arms and
culated Δ scores. Specifically, we calculated changes in found the median cubital vein to have a diameter of 2mm or
data collection at the 2, 4, 6, and 8 minutes subtracting larger. Larger adult pigs weighing between 67 and 80 kg
26
the AH data. Means and SDs were calculated on the have larger veins, including the auricular veins, compared

Hextend and Time of Administration and Hemodynamics in Swine Model 81
®