Page 78 - JSOM Summer 2020

Step Duration Effects on Blood Loss in
Simulated Designs of Tourniquet Use Procedure

1
2
John F. Kragh Jr, MD *; James K. Aden 3rd, PhD ; Michael A. Dubick, PhD 3

ABSTRACT

Background: We sought new knowledge by further developing remain somewhat unclear, partly because most data on emer-
a model of using calculations in the simulation of a first-aid gency use are limited and of low certainty. 9
6–8
task. The purpose of this study was to develop the model to
investigate the performance of tourniquet use in its component Tourniquet use comprises component steps ordered routinely
steps. Methods: We aimed to design an experiment on a desk- by design. Step performance in simulation and clinical care
top computer by mathematically manipulating simulated data can be assessed using metrics. A common metric is the dura-
in tourniquet use. A time factor of tourniquet use was ranged tion between injury and bleeding control. A long time risks
widely through time challenges in five degrees from ideal to shock and death. An understanding of when a step is per-
worst performances. Redesigning the task was assessed by formed unsatisfactorily can inform where future developments
time costs and blood losses. Results: The step of tourniquet might improve care.
application took 17% of the trial time and securing the tour-
niquet after bleeding control took the longest amount of the In 2018, we made calculations to inform our understanding
trial time, 31%. A minority of the time (48% [17% + 31%] of tourniquet use by including the step of donning medical
10
to apply tourniquet plus secure it) was spent after the tour- gloves. Extra blood volume lost in simulation while donning
niquet touched the patient, whereas most of the time (52%) gloves was, on average, 239mL. Before steady bleeding was
was spent before the tourniquet touched the patient. The step stopped, a relationship among the steps became clear: step
of tourniquet application lost 14% of the total blood lost, times and blood loss were causally linked. However, after
whereas no blood was lost during securing the tourniquet, bleeding control, times and losses were unassociated. Donning
because that was the moment of bleeding control despite se- gloves controlled the risk of infection, but uncontrolled bleed-
curing the tourniquet taking much time (31%). Most (86%) ing was prolonged by the donning time. In the task’s design,
of blood lost occurred before the tourniquet touched the pa- the risk of infection was prevented but inadvertently at a cost
tient. But blood losses differed 10-fold, with a maximum of of blood loss. The design produced a profile of risks that re-
2,434mL, which, when added to a pretask indication blood vealed itself: donning gloves harmed the patient by delaying
loss of 177mL, summed to 2,611mL. Before redesigning the control of bleeding, yet infection control benefited both the
task, costs of donning gloves and calling 9-1-1 included un- caregiver and the patient. A tradeoff of risks and benefits was
controlled bleeding, but gloving mitigated risk of spreading seen between the steps and the persons.
pathogens among people. By step and person, redesigns of the
task altered the risk-benefit profile. Conclusions: The model Given that unexpected tradeoff, we sought new knowledge
was useful because it simulated where most of the bleeding of other possible tradeoffs by broadening our way of using
occurred before the tourniquet touched the patient. Model- calculations in simulation. The purpose of this study was to
ing simulated redesigns of the task, which showed changes in develop the previous mathematical model to investigate the
the task’s risk-benefit profile by step and among persons. The performance of tourniquet use in its component steps. The fo-
model generated hypotheses for future research, including the cus was on duration effects on bleeding.
capability to screen candidate ideas among task designs.
Methods
Keywords: tourniquet; first aid, bleeding control and preven-
tion; emergency; task deconstruction, simulation, modeling This study was conducted in December 2019 within the lim-
its of protocol guidelines at the US Army Institute of Surgical
Research. The design was an experiment of simulated data on
a desktop computer of tourniquet use by mathematical ma-
Introduction
nipulations. The study was designed to model a time factor
1,2
Anyone may have to give first aid in care for an injury. First- of tourniquet use through a wide spectrum of simulated per-
aid experts recommend tourniquets to help control bleeding formances. The spectrum was made large to simulate many
3–5
from limb wounds. However, tourniquet indications and use emergency caregiving episodes with different degrees of time
*Correspondence to 3698 Chambers Pass, Joint Base San Antonio Fort Sam Houston, TX; or john.f.kragh.civ@mail.mil
1 Dr Kragh is a researcher of bleeding control at the Institute of Surgical Research (ISR), Fort Sam Houston, TX, and an associate professor in
2
the Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD. Dr Aden is a statistician at the Brooke Army
Medical Center, Fort Sam Houston, TX. Dr Dubick is a researcher and chairperson of the Department of Hemorrhage Control and Resuscitation
3
at the ISR.
76