Page 16 - JSOM Spring 2024

PW supervised every set of TPT2 inflations and pulse gone/ FIGURE 3 Subject data.
return pairs and was responsible for stopping any applications All subjects Subjects with thigh Subjects without thigh
for safety or futility reasons. Subjects had the option to stop 200 n=68 occlusion n=55 occlusion n=13
at any time.
180
Statistical Analysis 160
®
Data were organized in Microsoft Office Excel (Micro soft p=.038
Corp., www.microsoft.com). Graphing and statistical analy- 140
ses were performed with GraphPad Prism, version 7.04 for 120
Windows (GraphPad Software Inc., www.graphpad.com). Re- p=.019
peated measures analysis of variance (ANOVA) with Tukey’s 100
multiple comparisons test was used for pressure comparisons 80
(main column effect with columns containing the five repli- p<.0001
cates for each set of pulse gones and pulse returns). 60
p<.0001
40
We knew from preliminary work that second sets of pulse
gones/returns tended to have more consistent pressure traces 20
and, on the thigh, slightly higher pressures. We decided to as- p<.0001
sign the second sets of pressure gone values as the pressures 0
to use for threshold setting of potential emergency-use limb Ages (yr) Heights (cm) Weights (kg) Thighs (cm) Ages (yr) Heights (cm) Weights (kg) Thighs (cm) Ages (yr) Heights (cm) Weights (kg) Thighs (cm)
tourniquet certification devices. Systolics (mmHg) Forearms (cm) Systolics (mmHg) Forearms (cm) Systolics (mmHg) Forearms (cm)

Results
Data for each subject is shown with a symbol. The whiskers extend
Two trials were discontinued for vasovagal responses with au- from minimum to maximum. The boxes extend from 25th percentile
dible heart rate slowing and subjects reporting pre- syncopal to 75th percentile. The medians are inside the boxes and generally
symptoms. One was a female with a 19.5-cm forearm circum- obscured by the symbols. The data of all 68 subjects are shown in the
ference and a 55.0-cm thigh circumference with symptoms first third of the graph. The data of the 55 subjects whose thighs were
occluded are shown in the middle third of the graph. The data of the
during the first set of thigh pulse gones/returns. The other 13 subjects whose thighs were not occluded are shown in the last third
was a male with a 16.3-cm forearm circumference and a of the graph. The p values shown in the last third of the graph are
43.3-cm thigh circumference with symptoms during the first for comparisons between subjects whose thighs were occluded versus
set of forearm pulse gones/returns. No data were used from those whose thighs were not occluded.
either subject. The female reported having similar symptoms
when receiving vaccinations; the male reported no instances and returns p=.86). The medians of the second sets of forearm
of similar symptoms. We have not had any subjects in prior pulse gone pressures (n=68) ranged from 128–282mmHg with
tourniquet studies 3–12 develop pre-syncopal symptoms, so we a median of 176mmHg, a 5th percentile of 135mmHg, a 95th
suspect the cycling of pulse gone/pulse return played an im- percentile of 216mmHg, a mean of 176mmHg, a standard de-
portant role in triggering the vasovagal response. viation of 27mmHg, and a coefficient of variation of 15%.

Figure 3 shows subject data for the 68 subjects with com- Thigh Applications
pleted sets of forearm and thigh pulse gones/returns. Thirty Thigh arterial occlusion was not reached on any of the nine
were female, and 38 were male. As can be seen in Figure 3, thighs over 62.4-cm in circumference (Figure 4). Thigh arterial
most subjects were 18–22 years of age, but ranged up to 67 occlusion was reached on the 33 thighs smaller than 55.0-cm
years of age. For subjects with and without TPT2 thigh occlu- in circumference, on 1 of 2 thighs of 55.0-cm in circumfer-
sion, median systolic pressures were the same (126mmHg), but ence, on all 18 thighs with circumferences of 55.1-cm through
the minimum and maximum values differed (102–142mmHg 59.6-cm, and on 3 of 6 thighs with circumferences of 60.0-cm
with occlusion and 120–173mmHg without occlusion). Thigh through 62.4-cm.
circumferences were smaller in subjects with TPT2 thigh oc-
clusion (minimum 41.6-cm, 5th percentile 51.6-cm, median The source of 98.5% of the pressure variation was the sub-
54.3-cm, 95th percentile 56.5-cm, maximum 62.4-cm) than ject. Thigh pulse gone pressures were higher than pulse return
in subjects without TPT2 thigh occlusion (minimum 55.0-cm, pressures (p<.0001). Pressures from the first sets of five thigh
5th percentile 61.6-cm, median 63.3-cm, 95th percentile gones and returns were lower than those from the second sets
69.0-cm, maximum 72.9-cm). Of the 55 subjects with thigh (p<.0001). For occlusive applications (n=55), the medians of
occlusion, 28 were male (51%). Of the 13 subjects without the second sets of thigh pulse gone pressures ranged from
thigh occlusion, 10 were male (77%). 274–1158mmHg with a median of 574mmHg, a 5th percen-
tile of 373mmHg, a 95th percentile of 1118mmHg, a mean
Forearm Applications of 619mmHg, a standard deviation of 200mmHg, and a co-
All forearm applications reached arterial occlusion (Figure 4). efficient of variation of 32%. The peak pressures for the 13
The source of 88.9% of the pressure variation was the subject thigh applications that failed to reach occlusion ranged from
(i.e., there was more variation between subjects than for gones/ 958–1377mmHg with a median of 1220mmHg (p<.0001
returns within a subject). Forearm pulse gone pressures were versus sets of gones and returns). All of the non-occlusive ap-
higher than pulse return pressures (p<.0001). Pressures from plications were stopped by PW because of concerning appli-
the first sets of five forearm gones and returns were not sig- cation pressures combined with no audible indication that the
nificantly different from those of the second sets (gones p=.77 application was nearing arterial occlusion along with some

14 | JSOM Volume 23, Edition 1 / Spring 2024