Author, date and country | Patient group | Study type (level of evidence) | Outcomes | Key results | Study Weaknesses |
---|---|---|---|---|---|
Johnson AR, et al. 2011 United States of America | 11 patients between the age of 19-64 at a single burn unit. Each patient was imaged with triple-phase bone scintiscan to identify digits at risk. They each received a 0.15 mg/kg bolus and a 0.15 mg/kg/hour infusion for 6 hours of IV tPA. | Retrospective Case Series | Amputation Rate | 43/73 digits at risk were amputated (59%) | No comparison group. No baseline characteristics collected on patients that could contribute likelihood to require amputation. Small sample size. |
Nygaard RM et al. 2017 United States of America | 73 patients ages 11-83 years old from a single centre burn centre that were admitted with frostbite and evidence of perfusion defects on triple phase bone scan. 45 received tPA and 28 did not. They each received a 0.15 mg/kg bolus and a 0.15 mg/kg/hour infusion for 6 hours of IV tPA. | Retrospective Cohort | Amputation Rate | There was no statistical (p=0.092) difference between those that received tPA (16/45, 36%) and those that did not (14/28, 50%) | Relatively small sample size. Unclear why the patients in the no tPA group did not receive tPA. Limited information on the nature (temperature and duration) of the cold exposure. |
Frostbite severity | Frostbite severity scores did not differ between those that received tPA and those that did not | ||||
Cauchy E, et al 2011 France | 47 adults (44 men, 3 women) following mountain rescue with at least one digit demonstrating grade 3 or 4 frostbite. They all initially received 250mg of ASA and 400mg of buflomedil. 15 were randomized to receive that same treatment for 8 days. 16 were randomized to receive 250mg of ASA and 0.5-2 ng/kg of Iloprost for 8 days. 16 received 250mg of ASA and 2 ng/kg of Iloprost for 8 days with the addition of 100mg of rt-PA on day one. | Open Label Randomized Controlled Trial | Amputation Rate Patient Level | Both Iloprost alone (0%, 0/16) and Iloprost with rt-PA (19%, 3/16) had statistically significant lower amputations than buflomedil alone (60%, 9/15). There was no statistically significant difference between Iloprost and Iloprost plus rt-PA. | No baseline characteristics presented, large potential for confounding. A very specific population of people requiring mountain rescue, not generalizable. No blinding or concealment of allocation. Very small sample size. |
Carmichael H, et al. 2022 United States of America | 199 patients between the ages of 29-52 that were admitted to a single burn centre with grade 3 or 4 frostbite. 40 received IV tPA prior to transfer to the hospital, 32 received IV tPA on arrival, and 127 did not receive IV tPA. | Case Control | Amputation | Pre-transfer tPA had an odds ratio of 0.14 (0.05-0.40) compared to no tPA. However, tPA on arrival compared to no tPA had similar odds of amputation OR 0.74 (0.28-1.93) | The group that did not receive tPA had statistically significant less severe frostbite and were older. Relatively small sample size. |
Rogers C, et al. 2022 United States of America | A convenience sample of 131 patients admitted to a single burn centre with a frozen limb or digit. 104 received intravenous thrombolytic therapy. | Case Control | Limb Salvage | The use of thrombolytics had an adjusted odds ratio of 11.9 (1.57,89.4) | Small sample size - a post hoc power calculation was completed indicating a required sample size of 364 to detect differences with a power of 80%. It is unclear how the two cohorts were selected - potential for selection bias. |