Page 82 - JSOM Winter 2021

FIGURE 4 Hemodynamic data for each of the
three hemorrhage volume groups including heart
rate (A–C), mean arterial pressure (D–F), and
cardiac output (G–I).

Metabolic Data injury is perhaps more important with respect to thresholds
We compared metabolic data trauma and hemorrhage as well of mortality. 12
as during the observation period across the three groups (Fig-
ure 5). During trauma and hemorrhage the 20%, 30%, and Our model is further reflective of this. The pulmonary con-
40% hemorrhage groups had a mean K of 6.7 ± 3.3, 7.7 ± 1.7, tusion dose finding component, despite being limited to one
and 7.2 ± 2.4 mmol/L (p = .737). The mean lactates were 3.9 ± system, produced variable results suggesting that injury sever-
3.1, 3.5 ± 0.1, and 3.8 ± 1.7 mmol/L (p = .888). ity correlated poorly with the measured reactive force. This is
likely due to the viscoelastic properties of the porcine thoracic
At death, the 20%, 30%, and 40% hemorrhage groups had cavity and the necessary viscous criteria to produce the desired
a mean K of 7.3 ± 2.5, 10.4 ± 0.7, and 7.2 ± 2.4 (p = .340). results. It has previously been suggested that this plastic de-
13
The mean lactates were 4.3 ± 3.8, 5.9 ± 1.5, and 3.8 ± 1.7 formation of the rib cage alone can account for the linear rela-
(p = .700). tionship of the Abbreviated Injury Scale (AIS) that comprises
the ISS but does not describe the underlying anatomic tension
that can lead to larger inflammatory and physiologic derange-
Discussion 10,14
ments. An injury described as an AIS of 3 as compared to
We developed a model of hemorrhage and polytrauma in the an AIS 4 is reflective only of a 10% change in compression
absence of fluid resuscitation that incorporates three compo- depth in a linear relationship, but does not necessarily reflect
nents: tibial fracture, pulmonary contusion, and controlled what this viscoelastic parameter does to the underlying paren-
hemorrhage. The group that was subjected to the 30% hemor- chyma as a whole. 15,16 This is such that lower velocity injuries
rhage demonstrated optimal survival time characteristics given repeated over time with sufficient strain patterns would be
our aims. However, importantly, two components of this in- capable of producing equivocal tissue destruction and similar
jury model are titratable (pulmonary contusion and controlled outcomes, despite different biomechanical patterns as in our
hemorrhage). This is reflected by the pulmonary contusion model.
dose-finding results with quantified contusion volumes, as
well as the stepwise progression of the mortality, hemody- Further, with the addition of an extremity injury and subse-
namic, and metabolic data. Thus, the model can be adjusted quent hemorrhagic shock, the effects on mortality become
as needed to suit an investigator’s needs. even more clearly independent of ISS and that predicted by
TRISS. As has been previously described, hemorrhagic shock,
Our summation model was designed to provide an Injury un-resuscitated as in our model, leads to a significant systemic
Severity Score (ISS) > 16 across all groups to be consistent inflammatory cascade that compounds multisystem organ
with polytrauma as previously defined. 10,11 This included ex- dysfunction. 10,17 This is through some combination of, as of
tremity fracture, which contributed 9 points, and pulmonary yet, incompletely understood mediators, endothelial dysfunc-
contusion with laceration, which contributed 16 points, for tion, and the acidosis and coagulopathy that portend a worse
11
an overall ISS of 25. The ISS and the Trauma Injury Severity outcome in hemorrhagic shock. Concomitantly, we found
Scale (TRISS), however, fail to predict the overall course of subsequently worse hemodynamic and metabolic disorders
morbidity in these injury patterns as a result of their inabil- as hemorrhage was allowed to continue unabated, leading to
ity to capture physiologic, metabolic, and kinematic loading subsequent cardiac collapse. This was also manifest with re-
parameters. Previous murine models have demonstrated spect to loss of cardiac output markers of left-ventricular func-
11
that the pattern of injury as opposed to the severity of the tion. The same molecular inflammatory mediators of loss of

80 | JSOM Volume 21, Edition 4 / Winter 2021