LIMB COMPLICATIONS FOLLOWING PRE-HOSPITAL TOURNIQUET USE

Abstract

It has stated that the application of a pre-hospital tourniquet could prevent 7% of combat deaths, but their widespread use has been questioned due to the potential risk from prolonged ischaemia, or local pressure. The debate centres on their ability to improve survival after major haemorrhage, versus the potential risk of limb loss. A recent US military prospective study on their use demonstrated improved survival when a tourniquet was applied, and reported that no limb was lost solely from tourniquet use. However, this study focused on early limb loss, with a median follow-up of only 7 days, and so could not consider later morbidity. The aim of this study was to investigate if the pre-hospital application of a tourniquet resulted in an increase in morbidity following significant ballistic limb injury.

We reviewed members of the UK armed forces who sustained severe limb-threatening injuries in Iraq and Afghanistan, and based on the presence or absence of a pre-hospital tourniquet a cohort study was then performed. Of the 23 lower limbs that definitely had a pre-hospital tourniquet applied it was possible to match 22 limbs with 22 that did not have a pre-hospital tourniquet. The injuries were matched for anatomical location, severity of the bony injury, initial surgical management, Injury Severity Score and Mangled Extremity Severity Score as much as possible.

Of the 22 limbs with a pre-hospital tourniquet applied, 19 limbs had a least 1 complication. Of the 22 with no tourniquet applied, 15 had at least 1 complication (p=0.13). There were 10 limbs with at least 1 major complication in the pre-hospital tourniquet group but only 4 in the group with no tourniquet (p=0.045). There was no difference in the amputation rate.

The significant difference in the incidence of major complications is a concern, particularly as the difference was mainly due to a deep infection rate of 32% vs. 4.5%. Although there are a number of variables which could have influenced these small groups, such as choice of implant, method and timing of wound closure, the use of a cohort and a p < 0.05, does suggest the use of a pre-hospital tourniquet was a factor.

Although the use of pre-hospital tourniquets cannot be decried as a result of this study, there does remain the need to continually review their use, prospectively, to determine their risk/benefit ratio.