The Changing Theory of the Tourniquet

In April of 2013, a joint committee of law enforcement, military, trauma medicine, and first responders met in Hartford, CT to create a national policy to enhance survivability from active shooter scenarios. The document produced from this meeting was known as the Hartford Consensus. The net result was an acronym THREAT: threat suppression, hemorrhage control, rapid extrication to safety, assessment by medical providers, and transport to definitive care. Of those five items, the recommendation for hemorrhage control resulted in a change to the protocol for tourniquet application. This change in protocol represents theory choice in the danger level of tourniquets. In this particular circumstance, the factors in choosing one theory among a selection of theories is affected by changing concerns of trauma medicine.

The idea of theory choice is simple: given multiple scientific theories, we must decide on a theory, the best of the possible choices. Let’s say we look up at the stars. Two possible theories present themselves: the number of stars are odd or the number of stars are even. Based on some information about either choice, we believe in one theory over the other.

Two possible initial scenarios for theory choices are possible. In the absence of a consensus, theories are weighed against each other. In the presence of a consensus, there is the currently accepted theory and a newly introduced theory. The case of theory choice regarding the tourniquet is an instance of the latter as the Hartford Consensus shifted from an existing protocol to a new protocol. For clarity, the two protocols will be designed as Protocol A and Protocol B.

Protocol A views the tourniquet as a medical intervention with significant risk. The tourniquet is acknowledged to be an effective intervention for hemorrhage control of the extremities, but it should be used as a last resort due to the risk of limb loss. If hemorrhage is not controlled, the patient will die. The patient’s loss of a limb is considered preferable to her death; the tourniquet is used as a last resort after all other interventions have failed. Protocol A calls for elevation of the limb affected by hemorrhage. This is followed by direct pressure with a hand, ideally gloved. The next step calls for a pressure dressing to be placed over the site of injury. If this pressure dressing fails, then a tourniquet is applied several inches above the extremity. Protocol A was the popularly accepted theory prior to the Hartford Consensus.

Protocol B views the tourniquet as a medical intervention without significant risk. There is still a risk of limb loss. There is no conceivable way in which cutting off circulation to a limb could remove any risk of limb loss; it is reduced to a risk which is not significant. The tourniquet is not viewed as a limb or life decision. It is a safe intervention and should be used much earlier in the protocol. Protocol B skips several steps found in Protocol A. In the case of hemorrhage of the extremities, direct pressure with a hand is applied first. If that should fail, a tourniquet should be applied. The tourniquet remains the last intervention, but not the last resort. It is so because it is effective, and other interventions are available but unnecessary. Unlike Protocol A, the intermediate steps to avoid the tourniquet are skipped.

When I was training to be an EMT several years ago, Protocol A was still taught. By the time I was done with the training in 2014, Prince William County, VA had switched to Protocol B.

Protocol B is not entirely new. It was discussed by Lt. Colonel Douglas Lindsey with the Army Medical Service Board in Washington, DC in April 1957 (The American Journal of Nursing). Lindsey found that 5 to 10 percent of all combat casualties could have been saved by a tourniquet if someone had gotten to the patient in time. Elevation and pressure dressing are effective for venous bleeding. Arterial bleeding is another matter entirely. Lindsey cites the tourniquet, direct pressure, clamping, or ligation as the only effective interventions for arterial bleeding. Protocol B failed to become the dominant theory for 56 years.

If science makes progress, then the theories chosen later can be considered to be better to the earlier accepted theories. In this case, Protocol A is superseded by Protocol B; Protocol B is better than Protocol A. However, Protocol B existed for almost six decades along with theory A. That Protocol B could be set aside for Protocol A appears irrational.

However, that Protocol A persisted while a better theory existed is not irrational when properly understood. Science is a large scale endeavor. Scientists make theory choices based on their access to the total of scientific belief. They cannot be expected to base their choices on information foreign to them. The case of the tourniquet is an example of this large scale enterprise. The scientists in this case are the doctors, surgeons, and researchers. The individuals involved in the application of the tourniquets are not scientists. They are technicians more so than scientists. In the civilian world, an EMT receives about six months of training before testing to be certified as a medical provider. In the military world, battlefield trauma care may be handled by a battle buddy with rudimentary trauma medicine training. The role of these technicians are to stabilize the patient and transport to definitive care. They would be the ones applying the tourniquet using either protocols and delving their patient to the scientists at a hospital facility. The scientists’ view is partial. They receive the patients without participating in the protocol which their theory choice drives. Neither the technicians nor the scientists have a complete view of the problem domain.

Behavioral shifts in theory choice occur based on social factors. There is the accepted theory and a newly proposed theory. In the initial phase, the accepted theory is has broad support, and there is resistance to the newly proposed theory. Once the mean choice level shifts to the new theory, the social factors involved increases consensus around the new theory and away from the old theory. We should see Protocol A have consensus for a period of time. There is a point when Protocol B becomes more popular and social factors move the scientific community to increase consensus on the newly adopted Protocol B.

This pattern of behavior is observed in the case of the tourniquet. Lindsey’s discussion of Protocol B occurred in 1957. It is only accepted as consensus in 2013. The specific document of the Hartford Consensus is to disseminate the findings of the committee of eight people to establish a policy for the United States. The pattern of resistance, acceptance, and support can be found in the change in theory choice for the tourniquet. Lindsey’s paper and the Hartford Consensus mark two boundaries in time where the factors for the theory choice for the tourniquet shifted. The increase for consensus on the new theory occurs after the mean choice level shifts to the new theory. Then the Hartford Consensus is not the moment when consensus shifted; it is the moment when consensus is increased for the theory; it had occurred prior. By working backwards, that moment can be found and then looking at that moment, the factors for the theory choice for the tourniquet can be identified.

Science is ultimately concerned with evidence. There has to be evidence to support theory B. Beyond the difficulties presented by the large scale enterprise of science, there is the matter of data to be gathered as evidence for the theory. The battlefield presents an opportunity for trauma research that is unparalleled in the peace time. The difficulty in studying interventions for hemorrhage of the extremities is that the problem is time bound; people only have so much blood to bleed. In the civilian world, first responders who would provide immediate trauma care can be minutes away. The primary obstacle is traffic. Survivability for the patient is greatly improved if someone nearby is able to start care before first responders arrive.

The modern soldier is cross-trained in trauma care to provide for his buddies. This puts medical assistance much closer to the patient on the battlefield that in the civilian world. In contrast to the civilian world, injuries are not rare. The pace of data gathering occurs at a rate unmatched in quantity and degree of injury.

In 2010, the model of trauma assessment is cABC: catastrophic bleeding, airway, breathing, and circulation. This was a modification of the old assessment which was only ABC. Catastrophic bleeding can include external bleeding and internal bleeding. It is vital to address catastrophic bleeding first because airway and breathing lead to oxygen flow in the body with blood as the oxygen carrier. If blood is flowing out of the body, the oxygen is also flowing out of the body. By keeping blood inside the blood vessels oxygen is kept flowing inside the body.

It is clear that by 2010 it was clearly understood in military medicine that control of bleeding was important. It was in 2005 that the US Army conducted a study in commercial off-the-shelf tourniquets. The study acknowledges that 7 percent of combat deaths could be prevented with a properly applied tourniquet. This mirrors the statistic offered by Douglas Lindsey in 1957 revealing that the fatality rate over 60 years remains similar. The study acknowledged that the need for a “rapidly deployable tourniquet has been identified for at least half a century.” One important criteria listed was that the selected products must be effective and possibly be applied one-handed. Rather than develop a device for its need, the Army looked at commercially available products. Based on informal internet research and reports from military personnel from Iraq and Afghanistan, the Army tested products already in production.

This tells us several things about theory choice. The consensus had not change to Protocol B yet. The commercial market already had products to support the new protocols. It may be the case that commercial market had already chosen theory B. The three selected products were variations of the same kind of product. The Emergency Military Tourniquet (EMT), the Combat Application Tourniquet (CAT), and Special Operations Force Tactical Tourniquet (SOFTT) all featured a webbing system with a buckle creating a loop and a windlass system to tighten the tourniquet to control circulation. Lindsey identified these as ideal by Douglas Lindsey in his paper in 1957.

The mention of one-handed application is unique. Lindsey cautioned that following Protocol A risked the patient losing additional units of blood. He had assumed like others that trauma care would be by a “buddy or aidman.” The requirement by the Army to identify a tourniquet which could be applied one-handed hinted at a very different use case. In the civilian world operating with Protocol A pre-Hartford, an ambulance unit does not operate with an actual tourniquet device. It is improvised with a triangular or cravat bandage. This is tightened with a tongue depressor or pen. If a person is injured and alone in such a condition that she is unable to control the bleed with direct pressure, improvising a tourniquet from a bandage risks losing additional units of blood, as Lindsey suggested would happen if Protocol A was used. A prefabricated tourniquet would shorten the time to apply the tourniquet, skipping a time consuming step. Similarly, if a prefabricated tourniquet was available and the injury person were to follow Protocol A, additional units of blood would be loss.

It is necessary that the context of the Hartford Consensus is understood. The focus of the committee was to enhance survivability from active shooter scenarios. In such a scenario, there are the wounded and the first responders. The first responders fall into two categories: law enforcement and emergency services. The involvement of the military at Hartford was to provide the background data from battlefield research. Law enforcement is tasked with rendering the scene safe; they are ill equipped to provide medical care due to lack of equipment and training. A more complicated tourniquet protocol such as Protocol A is ill-suited for someone who does not render that kind of aid on a regular basis; Protocol B is simpler. A simplification of the hemorrhage control to Protocol B involving two steps makes it ideal for law enforcement to stabilize patients until they can be transported.

In a scenario where assistance was available additional blood loss may not be catastrophic. The immediate danger of severe blood loss is loss of consciousness. In a scenario where assistance is available, transportation is possible while the patient is unconscious. In a scenario where the injured person is alone, loss of consciousness with catastrophic bleeding will result in death. Protocol A and Protocol B shows significant efficacy when the issue of self-aid is considered. When there is aid, there is little difference in the efficacy of either protocols. An unconscious patient can be transported by others; she cannot transport herself while unconscious. This is the significant innovation of Protocol B, but it is not the sole cause of a change in protocols.

The tourniquet is often improvised from readily available component in aid bags such as bandages, torn strips of clothing, or rubber blood pressure cuffs. The development of a readily available device to be used as a tourniquet is a recent development, particularly a device which is conducive to self-application. This readily available device reduces the time between the decision to use a tourniquet and when the tourniquet is successfully applied. Combined with the simpler Protocol B, less units of blood is loss due to the protocol and readily available technology to address the circumstance.

The scientific change in the tourniquet protocol changed due to the concerns of trauma medicine. Trauma medicine in the field has expanded from buddy care to self-care due to the emerging circumstances of the active shooter and terrorism scenarios such as the Boston Marathon bombing. Those scenarios create environments where medical first responders are unable to get access to the wounded and complex protocols create difficulties for law enforcement agents who are best positioned to render aid in the civilian world. In light of these new problems, Protocol B emerges as a solution that better solves the puzzle of how to effectively maintain blood inside the body’s circulatory system.