TACTICAL MEDICINE ACADEMIC FILES
Source library for Talking TACMED and MED-TAC published content
Every clinical claim we put on camera or in print is supposed to be traceable to something you can read yourself. This page is where those sources live.
Each entry lists what the study actually was, how many people were in it, what it measured, and what it could not tell us. That last part matters most. A study is not a fact, it is an argument with a method attached, and the method decides how far the argument reaches. We list the limitation because you deserve to weigh it yourself rather than take our word for the conclusion.
How to read the evidence tiers. They are not interchangeable, and we do not blur them.
A randomised human trial outranks a human observational study. A human study outranks a cadaver or animal model. An animal model outranks a bench or manikin test. A case report outranks nothing except an anecdote, and a manufacturer claim is not evidence at all — it is marketing with a citation format.
Where the evidence genuinely disagrees, we list both sides.
You will find pairs on this page that reach opposite conclusions. That is deliberate. Burn cooling duration is the clearest example: one systematic review found no benefit to twenty minutes or more of cooling, and another found a significant reduction in skin grafting. Both are here. When a question is unsettled, saying so is the accurate answer, and picking the side that supports a tidier teaching point is not.
This library grows as we publish. If you find an error, or a newer study that contradicts something we have shipped, tell us and we will correct the asset rather than leave it standing.
Tourniquets — safety, effectiveness and limb outcome
Kragh JF Jr, Walters TJ, Baer DG, Fox CJ, Wade CE, Salinas J, Holcomb JB. Practical use of emergency tourniquets to stop bleeding in major limb trauma. J Trauma. 2008;64(2 Suppl):S38-S49. 232 patients, 428 tourniquets, 309 limbs; no limbs lost to tourniquet use.
What it is. 232 casualties, 428 tourniquets, 309 limbs, Baghdad combat support hospital, 2006. Measured morbidity directly rather than inferring it.
Limitation. No limbs were lost because of tourniquet use, and duration was not associated with increased morbidity. Prospective but uncontrolled, and a combat population is not a civilian one.
https://pubmed.ncbi.nlm.nih.gov/18376170/
Referenced in episode(s): 007, 023, 025
Kragh JF Jr, Walters TJ, Baer DG, et al. Survival with emergency tourniquet use to stop bleeding in major limb trauma. Ann Surg. 2009;249(1):1-7.
What it is. The survival companion to the paper above. Asked whether emergency tourniquet use saved lives.
Limitation. Uncontrolled prospective survey. Survival benefit was strongest when the tourniquet went on before shock developed.
https://pubmed.ncbi.nlm.nih.gov/19106667/
Referenced in episode(s): 023, 061
King DR, Larentzakis A, Ramly EP; Boston Trauma Collaborative. Tourniquet use at the Boston Marathon bombing: Lost in translation. J Trauma Acute Care Surg. 2015;78(3):594-599. Level V epidemiologic study, 27 improvised tourniquets, 0% mortality, no control group.
What it is. Boston Marathon bombing, 2013. Every Level I trauma centre in the city. 27 tourniquets applied, all improvised, none commercial. Zero mortality.
Limitation. Level V is the weakest design there is: descriptive, no comparison group, no randomisation. Mean transport was 24 minutes with marathon medical tents already on scene and multiple Level I centres nearby. Roughly one in five improvised tourniquets went onto a limb with no vascular injury.
https://pubmed.ncbi.nlm.nih.gov/25710432/
Referenced in episode(s): 005, 027
Mikdad S, Mokhtari AK, Luckhurst CM, et al. Implications of the national Stop the Bleed campaign: the swinging pendulum of prehospital tourniquet application in civilian limb trauma. J Trauma Acute Care Surg. 2021;91(2):352-360. Retrospective observational cohort, two Boston-area Level I trauma centres, January 2015 to December 2019. 147 patients. Tourniquet was clinically indicated in 51%; 27% were inappropriately applied, including placements distal to the injury and venous-only tourniquets.
What it is. Two Boston-area Level I trauma centres, January 2015 to December 2019, 147 patients arriving with a prehospital tourniquet. Medically trained abstractors judged whether the tourniquet was clinically indicated. It was indicated in 51%, and 27% were inappropriately applied — including tourniquets placed distal to the injury and tourniquets tightened only enough to occlude venous return.
Limitation. Chart review can only see what somebody wrote down, so a tourniquet placed correctly but never documented scores the same as one placed without reason. Retrospective indication judgements also carry hindsight.
https://journals.lww.com/jtrauma/abstract/2021/08000/implications_of_the_national_stop_the_bleed.15.aspx
Referenced in episode(s): 028
Hedger DJ, Smith M, Weaver N, Bendall J, Balogh ZJ. Increasing prehospital tourniquet use attributed to non-indicated use: an 11-year retrospective study. Eur J Trauma Emerg Surg. 2025;51(1):71. Level 1 trauma centre, John Hunter Hospital, Newcastle, Australia, 2011-2021. 88 applications in 86 patients; 77% non-indicated (95% CI 67-86%), and the non-indicated proportion rose across the study period (p=0.03). Complication in 33/86 (38%); with prolonged tourniquet time, 11/13 (85%).
What it is. Level 1 trauma centre in Newcastle, Australia, 2011 to 2021. 88 tourniquet applications in 86 patients. 77% were judged non-indicated, and that proportion rose across the study period. At least one attributable complication occurred in 38%, and in the 13 patients with prolonged tourniquet time, 11 had a complication.
Limitation. Single centre, small numbers, and indication was judged retrospectively by reviewers who knew the outcome. Hindsight is generous to the reviewer.
https://pubmed.ncbi.nlm.nih.gov/39856364/
Referenced in episode(s): 047
Butler FK, Holcomb JB, Dorlac W, et al. Who needs a tourniquet? And who does not? Lessons learned from a review of tourniquet use in the Russo-Ukrainian war. J Trauma Acute Care Surg. 2024;97(2S Suppl 1). 2,496 casualties with limb injuries and tourniquets applied, 2014-2022; 84.4% lower extremity. Application duration 50 to 380 minutes, mean 205.9 minutes. 92 amputations (3.7%) mainly for extensive necrosis.
What it is. Tourniquet use in the Russo-Ukrainian war, 2014 to 2022. 2,496 casualties with limb injuries and tourniquets applied, 84.4% lower extremity. Application duration ranged 50 to 380 minutes with a mean of 205.9 minutes. 92 amputations, mainly for extensive necrosis.
Limitation. There is no combat trauma registry in the Ukrainian military, and the data cannot separate amputations caused by the wound itself from those caused by tourniquet ischaemia. The finding travels with an evacuation environment measured in many hours, not the sub-hour evacuation of Iraq and Afghanistan.
https://pubmed.ncbi.nlm.nih.gov/38996420/
Referenced in episode(s): 043
Appropriate tourniquet types in the pediatric population: a systematic review. Cureus. 2021. Two single-arm observational studies, 73 children aged 2-16, CAT Gen 7 on UNINJURED limbs: 71/71 (100%) upper extremity and 69/73 (94.5%) lower extremity distal-pulse cessation. Very low certainty, high risk of bias, no controls.
What it is. 73 children aged 2 to 16. CAT Generation 7 applied to uninjured limbs. Distal pulse cessation by Doppler in 71 of 71 upper extremities and 69 of 73 lower extremities.
Limitation. Very low certainty, high risk of bias, no control group. Uninjured limbs in a controlled setting, and a stopped pulse is a surrogate for bleeding control, not bleeding control itself.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8118807/
Referenced in episode(s): 006, 030
Feeney EV, et al. Pediatric tourniquet use: safe and effective. J Pediatr Surg. 2025;60(10):162494. Level 1 pediatric trauma center, ages 0-17, 2015-2022, screened for amputation, compartment syndrome, rhabdomyolysis, kidney and nerve injury.
What it is. Level 1 paediatric trauma centre, ages 0 to 17, 2015 to 2022. Screened specifically for amputation, compartment syndrome, rhabdomyolysis, kidney injury and nerve injury.
Limitation. Single centre, retrospective, and small numbers relative to adult series. It can only see what was documented.
https://pmc.ncbi.nlm.nih.gov/articles/PMC12412217/
Referenced in episode(s): 006, 030
Hypothermia and the trauma triad
Jose AM, Rafieezadeh A, Zeeshan M, Kirsch J, Froula G, Prabhakaran K, Zangbar B. Hypothermia on admission predicts poor outcomes in adult trauma patients. Injury. 2025;56(5):112076. ACS-TQIP 2017-2021, n=3,043,030; mortality 23.4% vs 2.3%; Level III retrospective, association not causation.
What it is. ACS-TQIP, 2017 to 2021, 3,043,030 patients. Hypothermia defined as under 35 degrees Celsius on arrival. Mortality 23.4% in hypothermic patients against 2.3% in normothermic patients.
Limitation. Registry data shows association, not cause. Cold patients are often cold because they are badly hurt. Rewarming was documented in only 3% of hypothermic patients.
https://pubmed.ncbi.nlm.nih.gov/39658434/
Referenced in episode(s): 017, 040
Hsieh TM, et al. Effect of hypothermia in the emergency department on the outcome of trauma patients: a cross-sectional analysis. Int J Environ Res Public Health. 2018. 401 hypothermic vs 13,368 normothermic; after propensity matching of 399 pairs the mortality difference lost significance (13.0% vs 9.3%, p=0.115).
What it is. 401 hypothermic against 13,368 normothermic patients. The raw mortality gap was large.
Limitation. This is the honest counterweight. Across 399 well-balanced matched pairs the mortality difference lost statistical significance (13.0% vs 9.3%, p=0.115). Adjust for how badly injured people are and much of the signal fades.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6121888/
Referenced in episode(s): 017, 040
Chest and airway
Kheirabadi BS, Terrazas IB, Koller A, et al. Vented versus unvented chest seals for treatment of pneumothorax and prevention of tension pneumothorax in a swine model. J Trauma Acute Care Surg. 2013;75(1):150-156. US Army Institute of Surgical Research. n=8 anaesthetised spontaneously breathing pigs, both seal types tested on each animal in series; air injected in 0.2 L increments. Vented seals prevented tension pneumothorax; unvented seals did not. CoTCCC graded the resulting recommendation Level of Evidence C.
What it is. 8 anaesthetised, spontaneously breathing pigs with a surgically created open chest wound. Both a vented and an unvented seal were tested on each animal, with air injected into the pleural cavity in 0.2 litre increments. Vented seals prevented tension pneumothorax; unvented seals did not.
Limitation. This is animal-tier evidence, and the guideline sits on top of it. There is no human trial. CoTCCC itself graded the resulting recommendation Level of Evidence C. Notably, the unvented seal still improved work of breathing and restored most respiratory and circulatory measures before tension developed.
https://pubmed.ncbi.nlm.nih.gov/23940861/
Referenced in episode(s): 011
Ahmad et al. Meta-analysis of the optimal needle length and decompression site for tension pneumothorax and consensus recommendations on current ATLS and ETC guidelines. World J Emerg Surg. 2025;20:39. Radiological data from 24 studies, n=8,046: 32.84% failure to penetrate the pleural cavity. Heterogeneity very high (I2 99.72%).
What it is. 51 studies reviewed; radiological data pooled from 24 of them, 8,046 patients. Needle failed to penetrate the pleural cavity 32.84% of the time across sites and needle lengths. Each additional centimetre of needle length cut failure by 7.76%.
Limitation. Heterogeneity was extreme (I-squared 99.72%), which means the pooled number hides very different populations, imaging methods and outcome definitions.
https://pubmed.ncbi.nlm.nih.gov/40383767/
Referenced in episode(s): 036
Roberts DJ, et al. Clinical presentation of patients with tension pneumothorax: a systematic review. Ann Surg. 2015;261(6):1068-1078. Contralateral tracheal deviation reported in 17.9% of unassisted-breathing cases vs 2.9% on assisted ventilation. Assembled from case reports and case series.
What it is. Tension pneumothorax presentations. Contralateral tracheal deviation was reported in 17.9% of patients breathing on their own, and 2.9% of those on assisted ventilation.
Limitation. Built from case reports and case series, not a prospective cohort. Case reports over-report dramatic findings, so the real-world number is plausibly lower still.
https://journals.lww.com/annalsofsurgery/fulltext/2015/06000/clinical_presentation_of_patients_with_tension.9.aspx
Referenced in episode(s): 045
Resuscitation and positioning
CRASH-2 trial collaborators; Shakur H, Roberts I, et al. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet. 2010;376(9734):23-32. n=20,211 across 274 hospitals in 40 countries.
What it is. 20,211 bleeding trauma patients, 274 hospitals, 40 countries. Tranexamic acid reduced all-cause mortality and death due to bleeding.
Limitation. A pragmatic trial: enrolment required the treating doctor to be genuinely uncertain, which introduces selection effects. Most enrolling sites were outside high-resource trauma systems.
https://pubmed.ncbi.nlm.nih.gov/20554319/
Referenced in episode(s): 044
CRASH-2 collaborators; Roberts I, Shakur H, et al. The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet. 2011;377(9771):1096-1101. Exploratory, not the trial's primary question.
What it is. The source of the three-hour window. Reported that late administration was less effective and possibly harmful.
Limitation. Exploratory, not the trial's primary question. Subgroup timing analyses are the single easiest place for a trial to generate a finding that does not reproduce.
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60278-X/abstract
Referenced in episode(s): 044
Dutton RP, Mackenzie CF, Scalea TM. Hypotensive resuscitation during active hemorrhage: impact on in-hospital mortality. J Trauma. 2002;52(6):1141-1146. n=110 randomised to target systolic 70 vs above 100; no survival difference.
What it is. 110 bleeding trauma patients randomised to a conventional systolic target above 100 against a low target of 70, held until surgical control.
Limitation. Small, single centre, and it found no survival difference between targets. The specific number is genuinely contested.
https://pubmed.ncbi.nlm.nih.gov/12045644/
Referenced in episode(s): 051
Tran A, Yates J, Lau A, Lampron J, Matar M. Permissive hypotension versus conventional resuscitation strategies in adult trauma patients with hemorrhagic shock: a systematic review and meta-analysis of randomized controlled trials. J Trauma Acute Care Surg. 2018;84(5):802-808. Five trials, 1,158 patients, pooled OR 0.70; studies of poor to moderate quality.
What it is. Five randomised trials, 1,158 patients, comparing permissive hypotension against conventional resuscitation.
Limitation. The authors rated the constituent trials poor to moderate quality. The direction favours restraint; the precise target does not follow from the data.
https://pubmed.ncbi.nlm.nih.gov/29370058/
Referenced in episode(s): 051
Hemodynamic impact of the Trendelenburg position: a systematic review and meta-analysis. J Cardiothorac Vasc Anesth. 2024. 16 prospective interventional studies, 333 adult patients. Statistically significant 11% rise in stroke volume versus supine (MD 8.27, 95% CI 1.79-14.76, p=0.012), with increases in cardiac output and mean arterial pressure.
What it is. 16 studies, 333 adult patients. Head-down tilt produced a statistically significant 11% rise in stroke volume against supine, with increases in cardiac output and mean arterial pressure.
Limitation. The patients had normal cardiac function and most were anaesthetised or undergoing surgery. This tells us the position moves numbers in stable people. It does not tell us it helps somebody who is bleeding, and it does not measure how long the effect lasts.
https://www.jcvaonline.com/article/S1053-0770(24)00788-2/abstract
Referenced in episode(s): 050
Geerts BF, van den Bergh L, Stijnen T, Aarts LP, Jansen JR. Comprehensive review: is it better to use the Trendelenburg position or passive leg raising for the initial treatment of hypovolemia? J Clin Anesth. 2012;24(8):668-674. 13 Trendelenburg studies, n=246; cardiac output rose 9% at one minute and decayed to 4% by 2-10 minutes.
What it is. 13 Trendelenburg studies, 246 subjects. Cardiac output rose roughly 9% at one minute and had decayed to about 4% by two to ten minutes.
Limitation. Almost none of these subjects were actively haemorrhaging, and this review is the duration counterweight to the one above: the effect is real but brief. A transient rise in a healthy or anaesthetised subject is not a resuscitation intervention in a bleeding patient.
https://pubmed.ncbi.nlm.nih.gov/23228872/
Referenced in episode(s): 050
Bridges N, Jarquin-Valdivia AA. Use of the Trendelenburg position as the resuscitation position: to T or not to T? Am J Crit Care. 2005;14(5):364-368.
What it is. Reviewed the use of Trendelenburg as a resuscitation position and the evidence behind it.
Limitation. A narrative review, not a pooled analysis. Useful for tracing where the practice came from rather than for effect size.
https://pubmed.ncbi.nlm.nih.gov/16120887/
Referenced in episode(s): 050
Burns
Djarv T, et al. Duration of cooling with water for thermal burns as a first aid intervention: a systematic review. Burns. 2022. Four observational studies; no benefit found for 20 minutes or more of cooling versus less than 20 minutes on burn size, burn depth, re-epithelialization or skin grafting. Very low certainty.
What it is. Four observational studies. Found no benefit for cooling 20 minutes or more against less than 20 minutes for burn size, burn depth, re-epithelialisation or skin grafting.
Limitation. Very low certainty, downgraded for study design, risk of bias and indirectness. Absence of demonstrated benefit is not demonstrated absence of benefit.
https://www.sciencedirect.com/science/article/pii/S0305417921002941
Referenced in episode(s): 057
Chai H, et al. The effect of 20 minutes of cool running water first aid within three hours of thermal burn injury on patient outcomes: a systematic review and meta-analysis. Burns Open. 2022. Seven studies; 20 minutes of cool running water within three hours significantly reduced the odds of skin grafting and surgical intervention. Reaches the opposite conclusion to the review above.
What it is. Seven studies. Found that 20 minutes of cool running water within three hours of injury significantly reduced the odds of requiring skin grafting or other surgical intervention.
Limitation. Reaches the opposite conclusion to the review above, on an overlapping observational evidence base. Both are listed here deliberately. This question is not settled.
https://pubmed.ncbi.nlm.nih.gov/35688782/
Referenced in episode(s): 057
Pelvis
Bonner TJ, Eardley WGP, Newell N, et al. Accurate placement of a pelvic binder improves reduction of unstable fractures of the pelvic ring. J Bone Joint Surg Br. 2011;93(11):1524-1528. 172 radiographs with a visible binder buckle; only 50% sat at the greater trochanters, 39% were placed too high.
What it is. 172 radiographs with a clearly visible binder buckle, military hospital, 30 months. Only 83 (50%) sat at the level of the greater trochanters; 65 (39%) were placed too high.
Limitation. Retrospective, and only films where the buckle was visible could be scored, which may not represent every application.
https://boneandjoint.org.uk/Article/10.1302/0301-620X.93B11.27023
Referenced in episode(s): 053
Identifying risk factors for suboptimal pelvic binder placement in major trauma. Injury. 2020. Royal Brisbane and Women's Hospital, 496 radiographs: 43.5% suboptimally placed, 39.7% above the greater trochanter line.
What it is. 496 radiographs, Royal Brisbane and Women's Hospital, 2012 to 2016. 43.5% were suboptimally placed; 39.7% sat above the greater trochanter line.
Limitation. Retrospective and single centre. 284 patients could not be assessed, which is a substantial exclusion.
https://www.sciencedirect.com/science/article/abs/pii/S0020138320301741
Referenced in episode(s): 053
Shock recognition and ventilation
Deakin CD, Low JL. Accuracy of the advanced trauma life support guidelines for predicting systolic blood pressure using carotid, femoral, and radial pulses: observational study. BMJ. 2000;321(7262):673-674. n=20 hypovolaemic patients with arterial lines; pulses palpated by an observer blinded to the pressure readings. The guideline overestimated systolic pressure by a mean of 34 mmHg, with the largest errors in the lowest-pressure patients.
What it is. 20 hypovolaemic patients with arterial lines. Pulses were palpated by an observer blinded to the pressure readings, to test the taught rule that a radial pulse means a systolic of at least 80.
Limitation. The rule overestimated systolic pressure by a mean of 34 mmHg, and the error was largest in the patients with the lowest pressures — the exact patients the rule is used on. Only 20 patients, single centre. Enough to stop trusting the shortcut, not enough to build a replacement.
https://pubmed.ncbi.nlm.nih.gov/10987771/
Referenced in episode(s): 019
Bossers SM, Schwarte LA, Loer SA, et al. Association between prehospital end-tidal carbon dioxide levels and mortality in patients with suspected severe traumatic brain injury. Intensive Care Med. 2023;49(5):491-504. BRAIN-PROTECT, observational multicentre, Dutch Helicopter Emergency Medical Services, February 2012 to December 2017. 1,776 patients analysed. L-shaped association between end-tidal CO2 and 30-day mortality (p=0.01), with a sharp rise in mortality below 35 mmHg and no significant association above 45 mmHg.
What it is. BRAIN-PROTECT. Dutch Helicopter Emergency Medical Services, 2012 to 2017, 1,776 patients with suspected severe traumatic brain injury. An L-shaped association between end-tidal carbon dioxide and 30-day mortality, with a sharp rise in mortality below 35 mmHg and no significant association above 45 mmHg.
Limitation. Observational, so association rather than cause. Two things are worth knowing on top of that. End-tidal CO2 correlates imperfectly with arterial CO2 in shocked trauma patients, so the number you see is not exactly the number the brain sees. And the low-CO2 mortality association held across subgroups including patients with signs of cerebral herniation, which sits in tension with the guideline exception permitting brief hyperventilation in active herniation.
https://link.springer.com/article/10.1007/s00134-023-07012-z
Referenced in episode(s): 052
Governing guidelines
Committee on Tactical Combat Casualty Care (CoTCCC) Guidelines, Joint Trauma System — current published set.
What it is. The current published Committee on Tactical Combat Casualty Care guideline set, hosted by the Joint Trauma System.
Limitation. Guideline text reflects committee consensus weighing evidence, feasibility and operational risk. Where we cite a guideline rather than a study, the content says so out loud.
https://jts.health.mil/index.cfm/committees/cotccc/guidelines
Referenced in episode(s): Multiple
This library is educational reference material. It is not medical advice and it does not create a clinician–patient relationship. Training beats equipment every time — get both.
MED-TAC International Corp. · South Florida, US of A · tactical-medicine.com