Secondary Logo

Journal Logo

Brief Report

Frostbite in Southwest China: A Single-Center Retrospective Analysis

Xiao, Yue1,2; Wen, Ding-Ke3; Hao, Dan1,2; Li, Yan-Mei1,2; Jiang, Xian1,2,∗

Author Information
International Journal of Dermatology and Venereology: June 2022 - Volume 5 - Issue 2 - p 82-86
doi: 10.1097/JD9.0000000000000195



Frostbite is caused by the skin and underlying tissue freezing, with ice crystals forming in and between cells.1 Severe frostbite is considered one of the most devastating injuries among all types of cold-induced abnormalities.2

The morbidity and mortality rates of severe frostbite are increasing in the northern regions of China.3 The present retrospective observational study aimed to describe the demographic features of patients treated for frostbite at a single center in Southwest China as the data of frostbite was scarce in this region. However, there is little documentation of frostbite in the southwestern regions of China. Southwest China differs drastically from the northern regions regarding its mountainous terrain and diverse ethnicity. In this retrospective observational study, we evaluated the demographic and clinical differences between two patient populations with frostbite: Tibetan and Han, and discussed the factors affecting the outcome of severe frostbite in Southwest China.

Patients and methods

Patient selection

Patients with frostbite admitted to West China Hospital from January 1st, 2009, to January 1st, 2019, were screened for inclusion. Patients’ medical records were retrieved from the Research System by searching with term “frostbite.” Patients without a complete description of the physical examination findings or laboratory test results were excluded (Supplementary Fig. 1, This study was approved by the department of clinical trial management, West China Hospital of Sichuan University, to obtain clinical data retrospectively.

Clinical data

Patient information was collected from the hospital information systems. Demographic and clinical data were summarized, including age, sex, ethnicity, time from frostbite onset to admission, frostbite location, frostbite severity, operative intervention, and length of stay. Admission blood test results were also documented, including the red blood cell (RBC) count, hemoglobin (Hb), hematocrit (Hct), mean corpuscular volume, mean corpuscular Hb, mean corpuscular Hb concentration, white blood cell count, neutrophil granulocyte percentage, platelet count, alanine aminotransferase level, aspartate aminotransferase level, albumin level, prothrombin time, activated partial thromboplastin time, and fibrinogen level.

The severity of frostbite was classified into four grades, which can be simply classified as superficial (1st and 2nd degree) and deep (3rd and 4th degree) frostbite.1 Deep frostbite often required surgery to remove the devitalized tissue. Two researchers independently and blindly evaluated the frostbite grade based on a description of the physical examination findings (Supplementary Table 1, We modified Glasgow Outcome Scale and built a functional independence scoring algorithm of amputees to retrospectively define patients with a poor outcome as those who were partially able to perform self-care but lacked the competence to return to work or study (Supplementary Table 2,

Statistical analyses

Categorical variables were shown as frequency (percentage) and analyzed using the Chi-squared test or Fisher exact test. Continuous data were presented as mean ± standard deviation and analyzed using the Student t-test. All analyses were performed using SPSS 23.0 (SPSS inc., Chicago, IL, USA), and the significance level was set as P < 0.05.


Demographic and clinical characteristics of patients with frostbite

In this study, data of 27 patients were collected, with age ranged from 14 to 81 years, and most patients were younger than 20 years (41%, 11/27). Most patients were male, while only 15% were female. The ratio of Tibetans versus Hans was 13:14. The most common frostbite site was the lower extremities (74%, 20/27) followed by the upper extremities (63%, 17/27), whereas the ears and perineum were least commonly affected. The mean prehospital delay was 10 days (range 1–48 days), suggesting that none of these patients met the criteria for thrombolytic therapy; this was mainly due to the delay in admission, frequent freeze-thaw cycles, and the variable length of stay ranging from 2 to 104 days (Table 1).

Table 1 - Demographic characteristics of frostbite in a single center.
Parameter Number, n (%)
Age (years), median 22.4 (14–81)
Age group (years)
 ≤19 11 (40.8)
 20-39 6 (22.2)
 40-59 8 (29.6)
 ≤60 2 (7.4)
 Male 23 (85.2)
 Female 4 (14.8)
 Tibetan 13 (48.15)
 Han 14 (51.85)
 Upper extremities 17 (63.0)
 Lower extremities 20 (74.1)
 Ears 1 (3.7)
 Perineum 1 (3.7)
 Prehospital delay time (days) 13 (0–48)
 Length of stay (days) 10 (2–104)
Degree of frostbite
 1st and 2nd 4 (14.8)
 3rdand 4th 23 (85.2)
 Dressing and topical care 2 (7.4)
 Skin graft, escharotomy 3 (11.1)
 Amputation 10 (37.0)
 Unspecified 12 (44.4)
 Amputation 10 (37.0)
 Poor prognosis 7 (25.9)
Total > 100% as some patients have multiple involved locations.

We classified the patients based on the classification scheme for severity of frostbite. Most patients were categorized as having severe frostbite (85%, 23/27), while only 15% were considered to have minor frostbite. As tissue loss is generally regarded as a major indicator of an unfavorable outcome for frostbite, we investigated the amputation rate of the patients with severe frostbite. Amputation surgery was performed for nearly half of the patients with severe frostbite (43%, 10/23) versus none of the patients with mild frostbite (Table 1). To clarify the potential factors contributing to amputation among patients with severe frostbite, we compared the blood test results of patients with and without amputation. However, the blood test results did not significantly differ between these two groups (Table 2).

Table 2 - Differences of clinical characteristics among patients with severe frostbite (mean ± standard).
Amputation Prognosis

Parameter With (n = 10) Without (n = 13) P value Good (n = 16) Poor (n = 7) P value
Age, years 29.48 ± 18.82 32.31 ± 19.27 0.727 30.58 ± 16.94 32.62 ± 14.37 0.784
Male, n (%) 9 (90.00) 11 (84.60) 0.602 13 (81.25) 7 (100.00) 0.316
Tibetan, n (%) 7 (70.00) 6 (46.20) 0.237 10 (62.50) 3 (42.90) 0.337
Hematological indexes
 RBC, 1012/L 4.78 ± 0.80 4.57 ± 1.00 0.592 4.42 ± 0.86 5.25 ± 0.80 0.043
 HCT, % 0.43 ± 0.08 0.41 ± 0.08 0.696 0.40 ± 0.08 0.47 ± 0.06 0.033
 MCV, fL 89.20 ± 5.49 91.20 ± 5.50 0.397 90.06 ± 5.99 90.71 ± 4.46 0.798
 HGB, g/L 142.70 ± 26.60 139.92 ± 29.97 0.819 134.00 ± 26.48 159.14 ± 25.39 0.046
 MCH, pg 29.75 ± 1.93 30.84 ± 2.89 0.317 30.43 ± 2.88 30.34 ± 1.65 0.945
 MCHC, g/L 333.70 ± 9.89 337.46 ± 17.88 0.557 337.38 ± 15.67 334.43 ± 15.02 0.679
 WBC, 109/L 7.58 ± 1.91 8.62 ± 2.53 0.292 8.60 ± 2.34 8.00 ± 1.71 0.551
 NEUT%, % 70.35 ± 9.14 72.15 ± 12.56 0.706 73.40 ± 10.17 68.80 ± 10.77 0.338
 PLT, 109/L 246.36 ± 126.57 275.85 ± 98.63 0.536 278.29 ± 119.37 238.43 ± 82.89 0.434
 ALT, u/L 45.00 ± 62.60 68.54 ± 101.59 0.527 77.31 ± 99.44 22.86 ± 5.84 0.045
 AST, u/L 38.70 ± 30.71 99.85 ± 133.46 0.133 92.00 ± 121.83 36.00 ± 21.48 0.094
 Albumin, g/L 35.77 ± 6.94 36.49 ± 6.83 0.805 34.94 ± 6.27 38.46 ± 6.83 0.247
 PT, seconds 12.07 ± 1.38 11.85 ± 0.86 0.660 12.01 ± 1.15 11.87 ± 0.97 0.789
 APTT, seconds 29.92 ± 8.86 31.45 ± 9.04 0.690 31.83 ± 10.26 28.64 ± 3.27 0.276
 FIB, g/L 5.23 ± 2.52 3.58 ± 1.33 0.085 4.53 ± 2.30 3.84 ± 1.33 0.472
P value < 0.1.ALT: alanine aminotransferase; APTT: activated partial thromboplastin time; AST: Glutamate aminotransferase; FIB: fibrinogen; HCT: red blood cell specific volume; HGB: hemoglobin; MCH: mean corpuscular hemoglobin; MCHC: mean corpuscular hemoglobin concentration; MCV: mean corpuscular volume; NEUT: neutrophils; PLT: platelet count; PT: prothrombin time; RBC: red blood count; WBC: white blood cell count.

Considering that amputation and tissue loss may not equal the eventual functional outcome, we used a functional independence scoring algorithm to further study the differences between patients with a poor functional outcome and those with a good functional outcome (Supplementary Table 2, Based on this scale, seven patients were categorized as the poor outcome group, while 20 were categorized as the good outcome group. A comparison of the blood parameters in patients with good versus poor functional outcomes revealed a significant intergroup difference in the RBC count, Hct, and Hb level (P < 0.05) (Table 2).

Time and reason distribution of frostbite

All frostbite injuries were incurred from September to April, with the incident rate peaking in December, January, and February (Fig. 1A). For Tibetans, cold injuries were mainly caused by winter outdoor activities like intentional long exposure to extreme weather during pilgrimage (70%, 9/13). For Hans, the most common reason for frostbite was accidental long-time cold exposure during work, like liquid nitrogen leakage repair, followed by psychological disorders such as dementia, schizophrenia, and drunkenness (Fig. 1B).

Figure 1:
The time distribution and reason analysis of frostbite in southwest China. (A) Time distribution of frostbite. (B) Reasons for frostbite in Tibetans and Hans.


This research revealed the distinctive demographic and clinical profile of patients with frostbite in a single medical center in the southwestern Chinese region inhabited by Tibetan and Han people.5-9 Despite similarities in residence and living habits, the two populations developed frostbite owing to specific reasons. In the current study, pilgrimage accounted for 70% of all frostbite cases in Tibetan patients, whereas the Han patients with frostbite shared more epidemiologic similarities with patients in other regions especially cities.8-10

Most Tibetans live in the high-altitude mountainous regions in the Sichuan province and Qinghai-Tibet region and follow Tibetan Buddhism. For Tibetan Buddhists, pilgrimage is an essential and common sacred ritual comprising a nonstop pedestrian march from their homes to the Potala Palace. This nonstop travel exposes the Tibetan Buddhists to extreme weather patterns such as blizzards, often without proper shelter for an extended period. Moreover, hospital admission may be delayed due to the intentional endurance of cold-induced symptoms, which contributes to patients missing a valuable time window for thrombolytic therapy. As our center is often not the primary care-providing institute for these patients, caution should be taken in these primary triage centers to rapidly identify these patients and perform timely intervention, including faster assessment, rewarming, prehospital thrombolysis, or telemedicine to improve the functional outcome.1,11

Under the standardized protocol, patients with severe frostbite who are rewarmed rapidly and transferred to a hospital within 6–24 hours without repeated freeze-thaw damage can receive tissue plasminogen activator, which dissolves microvascular thrombi and restores blood flow before the occurrence of irreversible tissue necrosis.1,12 However, patients with frostbite often do not meet the strict criteria for thrombolytic therapy.13 Most patients admitted to our center were in the final stage, with bad body conditions. For example, a 19-year-old Tibetan man who accidentally soaked his feet in a creek while on pilgrimage was transferred to our center 4 days after the onset of symptoms. In the current study, none of the patients received thrombolytic therapy due to inadequate basic medical protocols for frostbite in the local hospitals, exposure to freeze-thaw cycles, and delayed transfer to our center. This lack of thrombolytic therapy markedly affects the chances of salvaging the affected limb/digit and the outcome, especially for severe cases. The current findings strongly indicate that the frostbite management protocols used in regional hospitals in West China are inadequate, as these hospitals generally do not see many patients with frostbite.

We summarized the precautions and therapeutical measures used to treat severe frostbite.14-16.

Precautions to prevent frostbite: (1) Wear appropriate clothing, including warm clothing that is not tight, gloves, a mask, and a hat. (2) Keep dry. (3) Intake sufficient energy and oxygen. (4) Monitor for early symptoms of frostbite such as numbness, a stabbing sensation, and white skin appearance.

Prehospital therapeutic measures for frostbite: (1) Get out of the cold environment and into a shelter as soon as possible. (2) Remove wet clothes. (3) Soak limbs in warm water at 37–39°C. (4) Avoid freeze-thaw cycles.

Frostbite treatment: (1) Debridement. (2) Elevate extremities to decrease edema. (3) Administration of analgesia and antibiotics. (4) Tetanus toxoid prophylaxis. (5) Administer tissue plasminogen activator within 24 hours of rewarming. (6) Perform amputation surgery if necessary (watch and wait for tissue demarcation before amputation).

Although it is well-accepted that the hallmark of a poor outcome in patients with frostbite is amputation and tissue loss, the occurrence of systemic inflammatory response syndrome might lead to multiple organ failure. Most previous studies have used different parameters to assess the frostbite severity because this assessment largely relies on caregivers’ subjective assessments in each institute.6,17 Considering that amputation is also an appropriate management with which to prevent systemic infection when the affected extremities present with gangrene or irreversible tissue necrosis, the poor outcome does not necessarily indicate the incidence of amputation. The outcome of severe frostbite may be affected by multiple factors, including the duration of prehospital delay, length of stay, and level of amputation. Therefore, when trying to clarify the factors associated with poor outcome, rather than using the conventional amputation categorization, we adopted a functional independence assessment algorithm performed by two blinded investigators. We predominantly evaluated the self-care ability and capacity to return to society; thus, patients with partially amputated toes and fingers were considered to have a better functional outcome than those with amputation of the lower extremities. This scoring system revealed that the RBC count, Hct, and Hb level were significantly associated with the outcome of patients with frostbite. However, the limited sample size might have biased the present results. Therefore, we suggest that it is insufficient to use amputation as the sole endpoint to determine the outcome of patients with frostbite. A more universal and detailed functional evaluation should be performed to optimize the post-treatment outcome.

The present study had some limitations. The small number of patients and retrospective single-center design may have biased the results. The results showed that the incidence of frostbite is affected by factors such as temperature, occupation, and activities. However, we could only analyze the data and observe trends rather than make statistically significant conclusions. In addition, potential confounding factors like the lack of access to transportation in remote areas and factors affecting hospital attendance should also be considered. However, this study was an initial attempt to study frostbite demographics in a non-typical cold region in China. A large-scale, clinical study in these populations might be helpful for providing a better clinical reference for frostbite treatment in such regions.

In conclusion, to avoid severe frostbite damage, routine mandatory activities like pilgrimage should be conducted under proper protection in extreme weather. Hematologi-cal indexes such as the RBC count, Hct, and Hb level should be monitored closely by first-line medical personnel during hospitalization, as these indexes might indicate the outcome of frostbite. Further research is advocated for the better management of frostbite in similar regions.


The authors would like to express their gratitude to the patients and their relatives for understanding and participating in our study.

Source of funding

The study was supported by the 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University (No. ZYJC21036), and National Natural Science Foundation of China (No. 82003373).


[1]. Handford C, Thomas O, Imray CHE. Frostbite. Emerg Med Clin North Am 2017;35 (2):281–299. doi:10.1016/j.emc.2016.12.006.
[2]. Boles R, Gawaziuk JP, Cristall N, et al. Pediatric frostbite: a 10-year single-center retrospective study. Burns 2018;44 (7):1844–1850. doi:10.1016/j.burns.2018.04.002.
[3]. Su H, Li Z, Li Y, et al. Treatment of 568 patients with frostbite in northeastern China with an analysis of rate of amputation. Zhonghua Shao Shang Za Zhi 2015;31 (6):410–415. doi:10.1134/1.1479982.
[4]. McMillan T, Wilson L, Ponsford J, et al. The Glasgow Outcome Scale -40 years of application and refinement. Nat Rev Neurol 2016;12 (8):477–485. doi:10.1038/nrneurol.2016.89.
[5]. Li Q, Wang LF, Chen Q, et al. Amputations in the burn unit: a retrospective analysis of 82 patients across 12 years. Burns 2017;43 (7):1449–1454. doi:10.1016/j.burns.2017.04.005.
[6]. Petrone P, Asensio JA, Marini CP. Management of accidental hypothermia and cold injury. Curr Probl Surg 2014;51 (10):417–431. doi:10.1067/j.cpsurg.2014.07.004.
[7]. Nemethy M, Pressman AB, Freer L, et al. Mt Everest base camp medical clinic “Everest ER”: epidemiology of medical events during the first 10 years of operation. Wilderness Environ Med 2015;26 (1):4–10. doi:10.1016/j.wem.2014.07.011.
[8]. Curran-Sills GM, Karahalios A. Epidemiological trends in search and rescue incidents documented by the Alpine Club of Canada from 1970 to 2005. Wilderness Environ Med 2015;26 (4):536–543. doi:10.1016/j. wem.2015.07.001.
[9]. Brandstrom H, Johansson G, Giesbrecht GG, et al. Accidental cold-related injury leading to hospitalization in northern Sweden: an eight-year retrospective analysis. Scand J Trauma Resusc Emerg Med 2014;22:6. doi:10.1186/1757-7241-22-6.
[10]. Strohle M, Rauch S, Lastei P, et al. Frostbite injuries in the Austrian Alps: a retrospective 11-Year National Registry Study. High Alt Med Biol 2018;19 (4):316–320. doi:10.1089/ham.2018.0060.
[11]. Durrer B. Management of frostbite in and outside of the doctor's surgery. Ther Umsch 2015;72 (1):55–57. doi:10.1024/0040-5930/a000640.
[12]. Drinane J, Kotamarti VS, O’Connor C, et al. Thrombolytic salvage of threatened frostbitten extremities and digits: a systematic review. J Burn Care Res 2019;40 (5):541–549. doi:10.1093/jbcr/irz097.
[13]. Shenaq DS, Beederman M, O’Connor A, et al. Urban frostbite: strategies for limb salvage. J Burn Care Res 2019;40 (5):613–619. doi:10.1093/jbcr/irz062.
[14]. Emsen IM. The approach to frostbite in Turkey: a retrospective study. Can J Plast Surg 2006;14 (1):21–23. doi:10.1177/229255030601400111.
[15]. Zafren K. Frostbite: prevention and initial management. High Alt Med Biol 2013;14 (1):9–12. doi:10.1089/ham.2012.1114.
[16]. Torpy JM, Lynm C, Golub RM. JAMA patient page. Frostbite JAMA 2011;306 (23):2633. doi:10.1001/jama.2011.1799.
[17]. Manganaro MS, Millet JD, Brown RK, et al. The utility of bone scintigraphy with SPECT/CT in the evaluation and management of frostbite injuries. Br J Radiol 2019;92 (1094):20180545. doi:10.1259/bjr.20180545.

frostbite; amputation; prognosis; Southwest China

Supplemental Digital Content

Copyright © 2022 Hospital for Skin Diseases (Institute of Dermatology), Chinese Academy of Medical Sciences, and Chinese Medical Association, published by Wolters Kluwer, Inc.