Page 54 - Journal of Special Operations Medicine - Spring 2015
P. 54

required with higher­pressure applications and when   Figure 2  Correlation coefficients for Friction Pressure
          more bunching was present.                         versus Combat Application Tourniquet (CAT) windlass turns
                                                             or Tactical Ratcheting Medical Tourniquet (RMT) ladder
          CAT Base Versus Strap Pressures                    distance used were consistent and strong.
          Base (n = 84) and Strap (n = 84) Friction Pressures under
          the CAT were very highly correlated (r = 0.9362). All 84
          Base Friction Pressures were higher than Strap Friction
          Pressures. The strength of the correlation between Base
          and Strap pressures was considerably less at Occlusion
          Pressure (r = 0.6669). Only 75 of the 84 Base Occlusion
          Pressures were higher than Strap Occlusion Pressures.
          The number of higher pressures under the firmer Base
          location indicates that the under­tourniquet­pressure is
          not uniformly distributed. The change in strength of the
          relationship between pressure under the Base and pres­
          sure under the Strap indicates that the pressure under
          the Base cannot be determined by simply adding a con­
          stant amount to the pressure under the Strap.

          Tactical RMT Ladder Versus Strap Pressures
          Ladder (n = 71) and Strap (n = 71) Friction Pressures
          under the Tactical RMT were very highly correlated (r =
          0.9108). All 71 Ladder Friction Pressures were higher than   Table 2  CAT Windlass Turns and Tactical RMT Ladder
          Strap Friction pressures. The strength of the correlation   Distance Used at Completion With Different Target
                                                             Friction Pressures
          between Ladder and Strap pressures was considerably less
          at Occlusion Pressure (r = 0.6061). Only 67 of the 70 Lad­                       Tactical RMT Ladder
                                                                                              Distance Used
          der Occlusion Pressures were higher than Strap Occlusion   Target   CAT Windlass Turns   at Completion,
                                                                          at Completion,
          Pressures. The number of higher pressures under the firmer   Friction   Median, Minimum–  Median, Minimum–
          Ladder location indicates that the under­tourniquet pres­  Pressure  Maximum No.    Maximum cm
          sure is not uniformly distributed. The change in strength   25mmHg  Not applicable  7.7, 5.6–12.5
          of the relationship between pressure under the Ladder and
          pressure under the Strap indicates that the pressure under   50mmHg  3.5, 3–4       5.4, 4.5–7.5
          the Ladder cannot be determined by simply adding a con­  75mmHg     3, 2–4          4.4, 2.7–5.1
          stant amount to the pressure under the Strap.       100mmHg         2, 1–3          3.7, 2.5–7.5
                                                              125mmHg         2, 1–3          2.8, 1.4–4.0
          Friction Pressures Versus Windlass Turns
          or Ladder Distance                                  150mmHg         1, 1–2          2.4, 0.8–3.4
          Base, Ladder, and Strap Friction Pressures for the CATs   175mmHg   1, 1–2         Not applicable
          and the Tactical RMTs were both consistently and    200mmHg         1, 1–2         Not applicable
          highly correlated with the number of windlass turns or   Notes: CAT, Combat Application Tourniquet; RMT, Ratcheting Med­
          the ladder distance used to reach Completion (Figure   ical Tourniquet.
          2). Lower Friction Pressures resulted in more windlass
          turns or more ladder distance traveled. For one Tacti­  the two were highly correlated (r = −0.8521 to −0.8655).
          cal RMT 25mmHg target Friction Pressure application   The distributions of the individual correlation coeffi­
          (59.4cm­circumference thigh), the available ladder dis­  cients around the medians, however, were not as tight
          tance was insufficient, and Occlusion was not achieved.   as those for Friction Pressures versus windlass turns or
          For one CAT 50mmHg target Friction Pressure applica­  ladder distance traveled (Figure 2). As with windlass
          tion (62cm­circumference thigh), the windlass could not   turns or ladder distance traveled, Bunching Indices were
          be turned sufficiently beyond Occlusion to be physically   higher (more bunching) with lower target Friction Pres­
          secured (no Completion). The number of windlass turns   sures. At higher Friction Pressures, bunching became a
          or ladder distance traveled to reach Completion at each   nonmeasureable phenomenon with the CATs: The hook­
          Friction Pressure is shown in Table 2.             and­loop covered strap simply gained a slightly non­
                                                             uniform appearance moving away from the bar on the
          Friction Pressures and Bunching                    base under which the hook­and­loop covered strap lay.
          As shown in Figure 3, the median correlation coefficients   Strap lengths involved in bunching and Bunching Indices
          for Friction Pressures versus Bunching Indices indicated   across all Friction Pressures are shown in Table 3.



          44                                      Journal of Special Operations Medicine  Volume 15, Edition 1/Spring 2015
   49   50   51   52   53   54   55   56   57   58   59