Page 39 - Journal of Special Operations Medicine - Summer 2016

Table 1 Mean Pressure Results as a Connecting Letters Report Two minor findings dealt with differences in biomechan-
Mean ics by model of tourniquet. A minor finding of mechani-
Model (mmHg) cal advantage was due to the difference in designs of the
CAT A 202 models studied; in other words, with the methods used,
pneumatic designs were able to achieve mechanical ad-
TPT3 A B 190 vantage more easily. The greater ease of achieving me-
TPT2 B 183 chanical advantage, namely through compression of the
EMT C 160 skin by well-distributed pressure, with pneumatic designs
compared with other designs was one reason pneumatic
designs were originally developed; another reason was
Figure 6 Results of blood loss by model of tourniquet.
that pneumatic designs were safer. Similarly, the second
26
minor finding related to mechanical advantage was that
pressure applied by tourniquets was inversely associated
with model width: wider tourniquets such as the EMT
(88mm wide), the widest model studied, applied the low-
est pressure; the least-wide model (C-A-T, 39mm) exerted
the highest pressure; and the two middle models by width
(TPT2, 50mm; TPT3, 75mm) applied the middle pres-
sures. This second minor finding reinforces a large and
growing body of evidence that wider tourniquets achieve
mechanical advantage more easily than narrow ones. 14,26
In fact, this finding is a phenomenon of collapsible tube
science, wherein the length of the tube compressed is a
major factor in whether the transmural pressure gradient
Results of blood loss by model of tourniquet were that the TPT3 had 27
the highest mean blood loss. Pairwise comparison revealed signifi- collapses the lumen of the tube (e.g., the artery).
cant differences among all models except the TPT2 and EMT pairing.
Analysis of variance placed the TPT3 alone in the high tier and all Another minor finding dealt with differences in user per-
others in the low tier. C-A-T, Combat Application Tourniquet; EMT,
Emergency and Military Tourniquet; TPT2, 2-in. Tactical Pneumatic formance. Despite the fact that only 10 tests were made
Tourniquet; TPT3, 3-in. Tactical Pneumatic Tourniquet. of each model of tourniquet, a difference between users
was found for a secondary outcome, pressure. The few
Figure 7 Results of pressure applied by user. tests limit the statistical power to detect user effects such
as differences in outcomes because of differences in user
that could be due to differences in user experience, skill
level, strategy of use, and techniques of use. Although
the clinician-scientist has been gathering such findings
from multiple studies over the years that indicate user
effects are common, when the studies have few tests,
like the 10 per model in the present study, then user ef-
fects are difficult to detect, likely due to lack of statisti-
cal power. For example, in another similar study led by
the clinician-scientist with 10 tests of improvised tour-
niquet use, no user effects were detected. However, re-
23
cently, more and more studies have reported findings on
user effects, although the main focus is still on materiel
The more-experienced user had the higher mean pressure by 13mmHg. difference, not on user differences. To mitigate user ef-
No statistically significant difference between users was found among fects such as learning, study designs can be changed and
mean values for other tested parameters. randomization of the order of models used is included.
More fundamentally, the design of the studies now can
Discussion
be widened, based on the scope of interest of the study,
The main finding was that all models of tourniquet per- to mitigate or to include user effects such as learning
formed equally well for the both the critical outcome curves. Because, apart from the materiel development,
(effectiveness) and the important outcome (pulse stop- the user’s skill development also plays an integral role
page distal to the tourniquet), whereas results for sec- in improving clinical care, study designs to understand
ondary outcomes (time to stop bleeding, pressure, and such learning patterns are also important. Specifically,
blood loss) differed by model. Only the EMT had best for hemorrhage care and tourniquet application, up
performance for every type of measurement studied. to 2001, the weakest link in the chain of care was the

Comparison of Pneumatic Tourniquet Models 25