Page 34 - JSOM Winter 2017

Table 2 Time to Determination of Bleeding Control by Glove Of the 18 trial status failures (i.e., unsatisfactory result), 17
Group, Using Dunnett Method a (94%) were due to pressure problems.
Absolute Difference Minus
Least Significant Difference Pressure amplitude results for users 1, 2, and 3 were in the
Glove Group (seconds) p Value b high level; those of users 2 and 4 were in the low level. Eight
Cold gloves and mittens 12.13 <.0001 percent of the variance of results in tourniquet pressure
Mittens 4.177 .0043 could be attributed to the users. Users 1 through 4 had two,
Cold gloves 2.827 .0102 four, four, and five tests with loose tourniquets, respectively,
Leather gloves −0.02 .0506 whereas only user 1 had tests (n = 2) with tight tourniquets
Glove liners and (high pressure). Of these 17 tests with pressures either too high
leather gloves −1.37 .0976 or too low, eight occurred in the user’s first half of uses and
Flight gloves −4.82 .3783 nine occurred in the second half.
Glove liners −6.27 .5750
Examination gloves −7.67 .7759 For gloves, Dunnett method showed that no mean was signifi-
Bare hands −13.7 1.000 cantly different than that of bare hands (p > .1586, all nine).
a Dunnett method was used to calculate the absolute difference between Among 17 tests with loose tourniquets, the count was two for
the mean of each experimental group and the mean of the control bare hands, three for flight gloves, three for mittens, three for
group. If positive, the absolute difference minus the least significant glove liners, two for leather gloves, one for glove liners layered
difference (LSD) determined that the experimental group mean was under leather gloves, and three for cold gloves layered under
more apart than the LSD from the control group mean and, therefore, mittens. The two tests with tight tourniquets were with bare
was significantly different. hands and flight gloves.
b Values are significant at p < .05.
For devices, U4-1 and U4-2 were slowest, and U1-1 and U2-2 For devices, the two levels of pressure amplitude had five de-
were fastest. Devices U2-1, U2-2, and U3-1 were second fast- vices each.
est, and U3-1 and U4-2 were second slowest. Again, user
effects dominated those of devices: intrauser results were con- Blood Loss
sistently in the same level, although some devices also were in
adjacent levels. The mean blood loss was 241 ± 18.0mL (median, 207.5mL;
minimum, 40mL; maximum, 1017mL; range, 977mL; Figure
1). Blood loss volumes ranged 25-fold, the largest variance of
Overall Time: Unwrapping Time Plus Trial Time
the study.
The mean overall time (i.e., time to unwrap the tourniquet
through time to trial completion) was 84 ± 35.6 seconds (me- Figure 1 Blood Loss by User.
dian, 77 seconds; minimum, 32 seconds; maximum, 234 sec-
onds; range, 202 seconds). Thus, there was a sevenfold range
in overall times.

For users, the results followed same pattern as time to deter-
mination of bleeding control. The analysis showed that 61%
of the variance of results in overall time could be attributed to
the users.

For gloves, again, bare hands and examination gloves were
in the fastest level with glove liners and flight gloves, whereas
cold gloves layered under mittens were alone in the slowest
level. Overall times with thicker gloves were slower. By the
Dunnett method, five glove groups (cold gloves layered under
mittens; cold gloves; mittens; leather gloves; and leather gloves
and glove liners) had means that were significantly slower than
that of bare hands (p < .0458, all five). The vertical box plots depict the 25th percentile as the box bottom,
75th percentile as the box top, 5th percentile as the down bar, 95th
percentile as the up bar, the dashed line as the mean, and the solid line
For devices, U4-1 and U4-2 were slow, and U1-1 and U2-2 as the median.
were fast. Devices U2-1, U2-2, and U3-1 constituted the mid-
level. Thus for devices, intrauser results were consistently in Users 1 and 2 were in the low level, user 3 was in the midlevel,
the same level, except those of user 2, who had devices in ad- and user 4 was in the high level. The analysis showed that
jacent levels. 68% of the variance of results in blood loss could be attrib-
uted to the users. In regression by use number, users showed
no or mild learning (R < .3503, all four).
2
Tourniquet Pressure
The mean tourniquet pressure was 203mmHg ± 18.0mmHg For gloves, blood loss was lowest when bare hands were used
(median, 205mmHg; minimum, 140mmHg; maximum, and highest when cold gloves layered under mittens were
274mmHg; range, 134mmHg; twofold range in pressure). used, but each level had three or five groups. By the Dunnett
Tourniquets were loose (categorically low pressure) in 15 tests; method, means of three glove groups—cold gloves layered
tight use (categorically high pressure) occurred in two tests. under mittens, mittens, and cold gloves—were significantly

32 | JSOM Volume 17, Edition 4/Winter 2017