Page 57 - JSOM Spring 2024

days 3 and 4. Students complete a postcourse survey at the end the present investigation. These procedures were repeated the
of training day 4. following day with a different role-player, so that each student
had an opportunity to apply SS and RL. On the second day,
Study Design after data collection, the students completed the exit (post-
This study incorporated a counterbalanced (crossover) within- course) survey.
subjects design so that each student applied the SS and RL
devices on consecutive days. Students were randomly assigned Measurement of Key Outcome Variables
to one of two sequences (SS then RL or RL then SS). A venipuncture attempt was defined as the needle (for SS)
or catheter (for RL) piercing the skin. Venipuncture success
Experimental Protocol (scored as 1 = success, 0 = otherwise) was operationally de-
Volunteer role-players signed a consent form after being briefed fined as consistent visible blood flow toward the collection
9
on the potential risks of autologous transfusion. Participants bag. The number of venipuncture attempts was documented.
then completed a precourse survey, which included training
level, deployments, and prior experience with trauma care and Three time periods were assessed: time to venipuncture, col-
blood collection. No sex or age data were collected. lection bag fill time, and total time. Time to venipuncture
comprised the total time from the start of the stopwatch to
Students and role-players were then paired and assigned to an venipuncture success, regardless of the number of attempts,
outdoor data collection station that included one chair and including preparation time (assembling materials from the
one calibrated digital weigh scale. SS and RL devices were in- kit, locating the venipuncture site, and skin cleaning via a 30-
cluded in the backpack of the student. Each station was super- second chlorhexidine scrub followed by 30-second dry time).
vised by a Valkyrie instructor, who carried a stopwatch and a
clipboard with standardized data collection sheets. Volume is the clinically relevant metric in whole blood col-
lection, but volume was not directly measured in this setting.
The role-player was seated on a chair, with the weigh scale Volume was extrapolated from mass (weight). Whole blood
placed on level ground on the same side as the arm used for ve- has a specific gravity of 1.05g/mL, so the target volume of
nipuncture (Figure 4). The scale was zeroed-out. The instructor 450mL of blood weighs 473g (450*1.05=472.5, rounded to
indicated the randomly assigned collection device that would 473g). The blood collection bag weighs 112g without blood,
be used, and the student retrieved that device from their back- so a full bag weighs 585g (473+112=585g). The target weight
pack. The instructor set the empty collection bag on the scale. in this study followed the Association for the Advancement
of Blood and Biotherapies (AABB) and Armed Services Blood
Program (ASBP) guidelines for collecting whole blood to a
combined total bag mass of 585g±10% (between 526.5g and
643.5g). 10,11 Collection bag fill times were recorded at every
100-g fill until the weight reached 585g. The initial weight of
the bag on the scale was about 85g, reflecting the total bag
FIGURE 4 Blood weight of 112g minus the weight of the tubing and needle,
collection conducted from which were suspended above the scale and attached to the do-
a seated position. The nor (Figure 4). Therefore, the actual amount of the collected
time to fill the collection blood was 476mL, not 450mL, but well within the ±10%
bag was recorded at
100-g increments and AABB guideline. Total time was calculated as time to veni-
stopped upon reaching puncture plus time to fill the bag to the target volume.
585g.
The postcourse survey included user perceptions of the rela-
tive ease of use, reliability, and speed of blood collection (1–5
scale). For each device, the students used a modified 3-point
Likert-type scale to indicate whether they would use the device
reluctantly, willingly, or eagerly in a combat environment. Stu-
dents indicated their preference (SS or RL) and detailed what
The instructor started the stopwatch immediately after reading they liked and disliked about SS and RL.
a casualty report and asking if the student was ready. The stu-
dent began the preparation steps and initiated the blood col- Analysis Plan
lection process. The instructor recorded venipuncture success, ANOVA was used for scaled outcomes (e.g., time to venipunc-
time to venipuncture success, and time to fill the collection ture, fill times, and user ratings). Chi-square, McNemar’s Test,
bag in 100-g increments. Timing stopped when the collection and binomial statistics were used for categorical outcomes
bag reached 585g. Fill times exceeding 25 minutes were coded (e.g., first-attempt success and preferences). Hypotheses were
as an “incomplete fill.” This cut-off was established by group evaluated at the p<.05 threshold for statistical significance. Re-
investigators’ consensus to avoid excessive skewing of fill time sults are expressed as percentages or as mean values (standard
data, either from lost or inadequate venous access or from ob- error of the mean [SE]). Analyses were conducted with SPSS
structed flow due to clotting in the needle or tubing. statistical software (version 23, IBM Inc., Chicago, Illinois).

Immediately after blood collection, the donor moved to a su- Results
pine position to assume the role of a casualty and received
autologous transfusion. No data on autologous transfusion All 86 enrollees completed one of three iterations of the Valky-
were recorded because this training task was not relevant to rie Emergency Whole Blood Transfusion Training course, with

Comparison of Whole Blood Collection Techniques | 55