SUMMARY
This study aims to identify prognostic factors influencing mortality among work-related burn patients in Northern Vietnam. A retrospective cohort study was conducted on 395 work-related burn patients admitted to the Vietnam National Burn Hospital (VNBH) from January 1 to December 31, 2023. Patient data were extracted from the VNBH electronic database, including demographic characteristics, burn severity, first aid interventions, comorbidities, clinical signs, and blood biomarkers. Univariate and multivariate logistic regression analyses were applied to determine independent predictors of mortality. The study population comprised 310 men (78.5%) and 85 women (21.5%), with a mean age of 36.7±12.9 years. Most patients were workers (87.9%) and resided in rural areas (52.7%). The predominant causes of burns were flames (46.1%) and electrical injuries (34.4%). The majority of patients (74.4%) sustained full-thickness burns, with 50.4% having a total burn surface area (TBSA) <20%. The overall mortality rate was 12.2%. Univariate analysis identified significant associations between mortality and fluid resuscitation (p=0.01), burn shock (p<0.001), facial burns (p<0.05), inhalation injury (p<0.001), full-thickness burns >20% (p<0.001), and complications (p<0.001). Multivariate analysis revealed that incorrect or absent fluid resuscitation (OR=2.3), burn shock (OR=3.7), facial burns (OR=3.5), neuropathy (OR=5.1), respiratory disease (OR=3.3), full-thickness burns (21-49%) (OR=4.2), serum glucose ≥9 mmol/L (OR=3.0), inhalation injury (OR=7.3), and full-thickness burns >50% (OR=6.7) were significant independent predictors of mortality (p<0.05). Work-related burn injuries remain a significant occupational hazard in Northern Vietnam. Factors such as inhalation injury, extensive full-thickness burns, burn shock and hyperglycemia substantially increase mortality risk. Early and appropriate fluid resuscitation, improved first aid measures, and better management of comorbidities are crucial for improving survival rates.
Keywords: work-related burns, mortality predictor, multivariate analysis
RÉSUMÉ
Cette étude a pour but d’identifier les facteurs prédictifs de mortalité chez des nord- vietnamiens brûlés lors d’un accident au travail. Il s’agit d’une étude rétrospective conduite sur 395 patients admis dans l’Hôpital National pour Brûlés du Vietnam (HNBV) entre le 1er janvier et le 31 décembre 2023. Les données extraites des dossiers électroniques comprenaient la démographie, la gravité de la brûlure, les premiers soins, les comorbidités, les données clinique, la biologie sanguine. Une analyse uni- puis multivariée a été utilisée pour isoler les facteurs indépendants de mortalité. Parmi les patients, 310 (78,5%) étaient des hommes, et 85 (21,5%) des femmes, d’âge moyen 36,7+/-12,9 ans. Il s’agissait en majorité (87,9%) de travailleurs ruraux (52,7%). Les principales causes étaient les flammes (46,1%) et l’électricité (34,4%). La majorité (74,4%) des brûlures étaient profonde et touchaient <20% SCT dans 50,4% des cas. La mortalité globale était de 12,2%. En analyse univariée, les facteurs associés à la mortalité étaient le remplissage (p=0,01), le choc (p <0,001), l’atteinte faciale (p <0,05), l’inhalation de fumées (p <0,001), l’atteinte profonde sur >20% SCT (p <0,001) et les complications (p <0,001). En analyse multivariée, le remplissage absent ou incorrect (OR 2,3), le choc (OR 3,7), une atteinte profonde sur 21 à 49% SCT (OR 4,2) ou >50% SCT (OR 6,7), une glycémie ≥ 9 mmol/L (OR 3), l’inhalation de fumées (OR 7,3) étaient tous des facteurs indépendants de risque de mortalité (p <0,05). Les brûlures au travail restent un risque significatif au nord-Vietnam. L’inhalation de fumées, les atteintes étendues et profondes, le choc et l’hyperglycémie augmentent significativement la mortalité. Un remplissage précoce et adapté, l’amélioration des premiers soins et une meilleure prise en charge des comorbidités sont cruciaux pour améliorer le pronostic.
Mots-clés: brûlure, accident du travail, prédicteurs de mortalité, analyse multivariée
Introduction
Burn injuries are a critical public health issue worldwide, with work-related burns contributing significantly to morbidity and mortality. According to the World Health Organization (WHO), burn injuries account for approximately 180,000 deaths annually, with the majority occurring in low- and middle-income countries.1 In industrialized nations, occupational burns represent 20–30% of all burn admissions, whereas in developing countries, this proportion is even higher due to inadequate workplace safety measures.2,3
Vietnam has experienced rapid industrial growth, leading to an increase in occupational hazards. Recent national health reports indicate that work-related burns constitute 25-35% of all burn cases admitted to major hospitals, with mortality rates ranging from 8% to 15%, depending on injury severity and pre-hospital care.4,5 The northern region of Vietnam, which is home to major industrial zones such as Bac Ninh, Hai Phong and Quang Ninh, has seen a significant rise in burn-related injuries due to the rapid expansion of manufacturing, construction and mining industries.6,7 Work environments such as construction sites, factories and chemical processing plants pose a significant risk of thermal, electrical and chemical burns.
Multiple factors influence the survival of burn patients, including demographic characteristics, burn severity, inhalation injuries, pre-hospital treatment and underlying comorbidities.8,9 Despite extensive global research on burn injury outcomes, data on work-related burn injuries in Vietnam remain scarce. Most studies focus on general burn populations without distinguishing occupational burn victims, limiting the ability to develop targeted interventions.6,10 This study seeks to address this gap by conducting a multivariate analysis of work-related burn cases from the Vietnam National Burn Hospital data, the country’s leading referral center for burn treatment. This study aims to enhance clinical decision-making and improve survival rates among occupational burn patients. The findings will provide evidence-based recommendations for workplace safety policies and emergency burn care protocols, ultimately contributing to better prevention and management strategies in industrial settings.
Methods
Study design
A retrospective cohort study was performed on the case records of work-related burn patients from the Northern region of Vietnam, which were extracted from the VNBH database in 2023.
VNBH database
Vietnam National Burn Hospital is located in the Hanoi capital and manages all burn patients transferred from hospitals in the Northern and Central regions of Vietnam. The VNBH database contains electronic patient records on over 3000 burn patients each year. All patient information related to demographics, medical conditions, symptoms, prescriptions, care procedures and treatment results were recorded in the VNBH database by burn specialist doctors from five clinical departments during the patient’s time staying at the Hospital.
Subjects
Three hundred and ninety-five work-related burns patients (including 310 men [78.5%] and 85 women [21.5%]) had an average age of 36.7±12.9 years (the highest frequency in patients aged 16-30 years [n=189, 47.8%]), who were admitted to the Vietnam National Burn Hospital from January 1 to December 31, 2023. Most patients lived in rural regions (n=208, 52.7%) and were workers (n=342, 87.9%) working in occupations such as mine worker, electrician, construction workers, steel mill worker, paper production worker, welder, oil industry worker and cook. The work-related burns occurred most frequently due to flames (n=182, 46.1%), and electric current (n=136, 34.4%). Patients with TBSA <20% accounted for the highest proportion (n=199, 50.4%), followed by patients with TBSA from 21-49% (n=180, 45.6%). There were 294 (74.4%) patients with full thickness burn injuries. Among these patients, full thickness burn injury areas <20% account for the largest proportion (n=207, 70.4%), followed by 21-49% (n=73, 24.8%) and >50% (n=14, 4.8%). Forty-eight (12.2%) work-related burn patients died (Table I).
Table I.
Demographic and clinical information of studied patients (n=395)
| Variables | Frequency | Percentage |
|---|---|---|
| Gender | ||
| Male | 310 | 78.5 |
| Female | 85 | 21.5 |
| Age (years) | M±SD: 36.7±12.9 | |
| 16-30 | 199 | 50.4 |
| 31-45 | 132 | 33.4 |
| 46-60 | 59 | 14.9 |
| ≥61 | 5 | 1.3 |
| Place of residence | ||
| Rural | 208 | 52.7 |
| Urban | 187 | 47.3 |
| Socio-professional categories | ||
| Executives | 53 | 12.1 |
| Workers (*) | 342 | 87.9 |
| Types of burning agent | ||
| Scald | 55 | 13.9 |
| Flames | 182 | 46.1 |
| Electric current | 136 | 34.4 |
| Chemical agent | 22 | 5.6 |
| TBSA(%) | ||
| <20 | 199 | 50.4 |
| 21-49 | 180 | 45.6 |
| ≥50 | 16 | 4.0 |
| Full thickness burn injuries (%) | Total: 294 | 74.4 |
| <20 | 207 | 70.4 |
| 21-49 | 73 | 24.8 |
| ≥50 | 14 | 4.8 |
| Treatment outcome | ||
| Survival | 347 | 87.8 |
| Death | 48 | 12.2 |
(*) Occupation group: mine worker; electrician; construction workers; steel mill worker; paper production worker; welder; oil industry worker; cook
Data collection
Data collecting process
The research team was established with 5 members who have over 5 years of experience in the field of burn patient care. Before collecting data, the questionnaire was created, tested on 10 work-related burn patients (admitted to VNBH and discharged in 2024, randomly selected from the VNBH database). The questionnaire was completed and became the official version after being edited and adjusted by experts from VNBH.
The variables to be collected are represented in the questionnaire:
Demographic and clinical information on admission of studied patients were collected, including gender, age, place of residence, socio-professional categories, type of burning agent, TBSA, full-thickness burn injuries, burn shock, burn locations, inhalation injury, combined trauma, complications and treatment outcome.
First aid the patient received before admission to VNBH and the patient’s underlying medical conditions were extracted, including first aid facility, fluid resuscitation, painkillers, burn wound dressing and means of patient transport.
Blood test biomarkers were analyzed on admission, including red blood cells, white blood cells, glucose, creatinine and urea.
Statistical analysis
Data analysis was conducted using SPSS software version 24.0. Continuous variables were expressed as mean ± standard deviation (SD), while qualitative variables were presented as frequencies and percentages. In order to identify prognostic factors associated with mortality in work-related burn patients, there were comparisons between deceased and surviving patients across demographic characteristics, medical history, pre-hospital interventions, clinical findings, and laboratory parameters upon hospital admission. Continuous variables were analyzed using the Student’s T-test, while qualitative variables were compared using the Chi-square test. A multiple logistic regression model was developed to estimate odds ratios (OR) with 95% confidence intervals (CI) for mortality-related factors. Variables with a probability value (p-value) <0.2 in univariate analysis, following the Hosmer and Lemeshow criterion, were selected for multivariate analysis. The final multivariate model retained only independent predictors of mortality with statistical significance (p <0.05). A significance level of α=5% was considered, with p-values <0.05 deemed statistically significant.
Ethical clearance
This study followed the ethical principles outlined in the Declaration of Helsinki. The research protocol was reviewed and approved by the Ethics Committee in Biomedical Research (ECBR) of the Vietnam National Burn Hospital (Approval No: 05/CNChT-HDDD). Since this was a retrospective study using anonymized patient data from the hospital’s electronic database, the requirement for informed consent was waived by the ECBR. All patient data were handled with strict confidentiality, and no personally identifiable information was disclosed. The research team ensured compliance with ethical guidelines to protect patient rights and privacy throughout the study.
Results
Univariate analysis
Patient demographics and mortality in work-related burn patients
In the two groups of patients with work-related burn injuries who died and survived, there was no statistically significant difference in age, gender, place of residence, occupation, types of burning agent as well as the time of the accident. Two variables, gender and burn agent, with P <0.2 were selected for inclusion in multivariate analysis (Table II).
Table II.
Demographics and death in work-related burn patients
| Variables | Non-survivor (n=48) N (%) |
Survivor (n=347) N (%) |
OR (95% CI) | p-value |
|---|---|---|---|---|
| Age (years) | ||||
| 16-30 | 20 (41.7) | 179 (51.6) | Ref | |
| 31-45 | 15 (31.2) | 117 (33.7) | 0.6 (0.2-1.4) | 0.33 |
| 46-60 | 10 (20.8) | 49 (14.1) | 0.5 (0.2-1.5) | |
| ≥61 | 3 (6.2) | 2 (0.6) | 1.2 (0.3-4.3) | |
| Gender (*) | ||||
| Male | 39 (87.5) | 271 (78.1) | 1.9 (0.8-3.5) | 0.08 |
| Female | 9 (12.5) | 76 (21.9) | Ref | |
| Place of residence | ||||
| Rural | 22 (45.8) | 31 (8.9) | 1.2 (0.9-2.7) Ref | 0.27 |
| Urban | 20 (54.2) | 316 (91.1) | ||
| Socio-professional categories | ||||
| Executives | 7 (14.6) | 46 (13.3) | Ref | |
| Workers | 41 (85.4) | 3014 (86,7) | 1.2 (0.5-2.1) | 0.35 |
| Mechanism of injury (*) | ||||
| Scald | 1 (2.1) | 54 (15.6) | 0.8 (0.3-2.8) | 0.19 |
| Flames | 28 (58.3) | 154 (44.4) | Ref | |
| Electric current | 18 (37.5) | 118(34) | 1.9 (0.5-3.9) | |
| Chemical agent | 1 (2.1) | 21(6) | 0.7 (0.2-2.3) | |
| The seasonal distribution | ||||
| Spring | 13 (27.1) | 81 (23.3) | Ref | |
| Summer | 10 (20.8) | 88 (25.4) | 0.7 (0.2-1.8) | 0.85 |
| Autumn | 13 (27.1) | 100 (28.8) | 0.8 (0.3-2) | |
| Winter | 12(35) | 78 (22.5) | 0.9 (0.4-2.4) |
(*): The selected variable for multivariate analysis
First aid and mortality in work-related burn patients
Patients who did not receive fluid resuscitation or received fluid resuscitation incorrectly had a statistically significant higher mortality rate than patients who received fluid resuscitation correctly (p=0.01). Other variables such as first aid facility, using painkillers, burn wound dressing and means of patient transport did not have statistically significant differences between the two groups of patients who survived or died (p>0.05). However, among these variables, there were still variables such as first aid facility, using painkillers, burn wound dressing with p<0.2, which were selected to be included in multivariate analysis (Table III).
Table III.
First aid and death in work-related burn patients
| Variables | Non-survivor (n=48) N (%) |
Survivor (n=347) N (%) |
OR (95% CI) | p-value |
|---|---|---|---|---|
| First aid facility | ||||
| Workplace | 5 (10.4) | 104(30) | Ref | |
| Community Health Facility | 7 (14.6) | 88 (25.4) | 0.8 (0.2-1.1) | 0.34 |
| District Health Facility | 12(25) | 76 (21.9) | 1.1 (0.5-1.4) 1.3 (0.7-1.9) | |
| Provincial Health Facility | 24(50) | 79 (22.7) | ||
| Fluid resuscitation (*) | ||||
| Properly | 6 (12.5) | 189 (55.5) | Ref | |
| Wrong way/ not done | 42 (87.5) | 158 (35.5) | 5 (2.1-6.9) | 0.01 |
| Using painkillers(*) | ||||
| Yes | 10 (20.8) | 209 (60.2) | Ref | |
| No | 38 (79.2) | 138 (39.8) | 2.6 (1.2-4.9) | 0.12 |
| Burn wound dressing | ||||
| Yes | 25 (52.1) | 33 (9.5) | Ref | |
| No | 23 (47.9) | 314 (90.5) | 0.9 (0.2-1.8) | 0.44 |
| Means of patient transport | ||||
| Ambulance | 28(70) | 161 (47,5) | Ref | |
| Car | 8(20) | 120 (35,4) | 0,4 (0,1-0,7) | 0.28 |
| Motorbike | 4(10) | 58 (17,1) | 0,4 (0,1-0,7) |
(*): The selected variables for multivariate analysis
Comorbidities and mortality in work-related burn patients
In univariate analysis, the comorbidities (Neuropathy, Respiratory disease, Cardiovascular disease and Hypertension) were not associated with mortality in burn patients with work-related burn injuries (p>0.05). Among the comorbidities, two variables, Neuropathy and Respiratory disease, with p<0.2 were selected for multivariate analysis (Table IV).
Table IV.
Comorbidities and death in work-related burn patients
| Variables | Non-survivor (n=48) N (%) |
Survivor (u=347) N (%) |
OR (95% CI) | p-value |
|---|---|---|---|---|
| Neuropathy(*) | ||||
| Yes(**) | 7 (14.6) | 1 (0.3) | 2.6 (1.7-3.1) | |
| No | 41 (85.4) | 346 (99.7) | Ref | 0.13 |
| Respiratory disease (*) | ||||
| Yes(***) | 5 (10.4) | 2 (0.6) | 1.9 (1.1-3.4) | |
| No | 43 (89.6) | 345 (99.4) | Ref | 0.17 |
| Cardiovascular disease | ||||
| Yes(****) | 4(13) | 8(9) | 0.7 (0.3-1.5) | |
| No | 45(87) | 239(91) | Ref | 0.6 |
(*): The selected variables for multivariate analysis
(**): Epileptic (6 patients), Peripheral neuropathy (2 patients)
(***): Asthma (4 patients), Chronic pneumonia (3 patients)
(****): Hypertension (10 patients), Heart rhythm disorder (2 patients)
Clinical signs and mortality in work-related burn patients
Severe clinical signs in burn patients all had a higher mortality rate than in the group of survivors, such as burn shock (p<0.001), facial burn (p<0.05), inhalation injury (p<0.001), full thickness burn injuries >20% (p<0.001), combined trauma (p=0.04) and complications (p<0.001). Some other variables such as Body burn, Perineal burn, Burned hand and TBSA >20%, although not significantly related to the mortality rate in burn patients with work-related burn injuries (p>0.05), were selected for multivariate analysis because of p<0.2 (Table V).
Table V.
Clinical signs and death in work-related burn patients
| Variables | Non-survivor (n=48) N (%) |
Survivor (n=347) N (%) |
OR (95% CI) | p-value |
|---|---|---|---|---|
| Burn shock (*) | ||||
| Yes | 44 (91.7) | 134 (38.6) | 5.9 (5.1-15 3) | <0.001 |
| No | 4 (8.3) | 199 (41.4) | Ref | |
| Facial burn (*) | ||||
| Yes | 35 (72.9) | 137 (39.5) | 4.8 (2.8-9.6) | 0.01 |
| No | 13 (27.1) | 210 (60.5) | Ref | |
| Burns of the upper extremities | ||||
| Yes | 26 (54.2) | 230 (66.3) | 0,9 (0.2-2.2) | 0.7 |
| No | 22 (45.8) | 117 (33.7) | Ref | |
| Burns of the lower extremities | ||||
| Yes | 20 (41.7) | 189 (54.5) | 1.3 (0.6-3.9) | 0.8 |
| No | 28 (58.3) | 158 (45.5) | Ref | |
| Body burn (*) | ||||
| Yes | 30 (62.5) | 182 (52.4) | 2.3 (0.4-3.1) | 0.06 |
| No | 18 (37.5) | 165 (47.6) | Ref | |
| Perineal bora (*) | ||||
| Yes | 19 (39.6) | 92 (26.5) | 1.3 (0.6-2.5) | 0.18 |
| No | 29 (60.4) | 255 (73.5) | Ref | |
| Burned band (*) | ||||
| Yes | 30 (62.5) | 153 (44.1) | 1.7 (0.54-2.8) | 0.13 |
| No | 18 (37.5) | 194 (55.9) | Ref | |
| Inhalation injury (*) | ||||
| Yes | 32 (66.7) | 53(16) | 9.2 (6.5-16.2) | <0.001 |
| No | 16 (33.3) | 281(84) | Ref | |
| TBSA (%) (*) | ||||
| <20 | 10 (20.8) | 189 (54.5) | Ref | |
| 21-49 | 26 (54.2) | 154 (44.4) | 1.6(0.6-3.8) | |
| ≥50 | 12(25) | 4 (1.1) | 4.3 (2.4-5) | 0.09 |
| Full thickness burn injuries (%) (*) | Total: 294 | |||
| <20 | 7 (14.6) | 200 (81.3) | Ref | |
| 21-49 | 28 (58.3) | 45 (18.3) | 5.4 (3.9-8.3) | |
| ≥50 | 13 (27.1) | 1 (0.4) | 9.2 (5.2-13.6) | <0.001 |
| Combined trauma (**) | ||||
| None | 40 (83.3) | 345 (99.4) | Ref | |
| Yes | 8 (16.7) | 2 (0.06) | 3 (2,8-4,6) | 0.04 |
| Complications | ||||
| None | 2 (4.2) | 279 (80.4) | Ref | |
| Yes (***) | 46 (95.8) | 68 (19.6) | 23 (10.2-35.5) | <0.001 |
(*): The selected variables for multivariate analysis
(**): Including one or more of the following combined traumas: traumatic brain injury, abdominal trauma, broken limbs, multiple trauma
(***): Including one or more of the following complications: sepsis shock, ARDS, renal failure, gastrointestinal bleeding, multiple organ failure
Blood biomarkers and mortality in work-related burn patients
Among the blood biomarkers determined when the patients were admitted to VNBH, serum glucose in the death group was significantly higher than in the survival group (11.61±3.2 mmol/l vs 6.34±2.6, p<0.05) (Table VI).
Table VI.
Blood biomarkers and death in work-related burn patients
| Variables | Non-survivor (n=48) N (%) |
Survivor (n=347) N (%) |
OR (95% CI) | p-value |
|---|---|---|---|---|
| Red blood tells (x1012) | 4.9 ± 1.34 | 4.5 ±2.1 | 1 (0.9-1.8) | 0.3 |
| Hemoglabin(g/l) | 142.5 ± 31.7 | 139.8 ± 36.3 | 1 (0.97-1) | 0.21 |
| While blood cells (x109) | 12.8 ± 6.3 | 12.2 ± 5.7 | 1 (0 9-1.1) | 0.6 |
| Glucose (mmol/l) (*) | 11.61 ±3.2 | 6.34 ± 2.6 | 2.8 (1.3-3.9) | 0.04 |
| Creatinine (μmol/l) | 72.5 ± 25.2 | 73.3 ± 28.4 | 1 (0.9-1.02) | 0.8 |
| Urea (mmol/l) | 5.9 ± 2.6 | 5.5 ± 3.1 | 1.1 (0.0-1.2) | 0.4 |
(*): The selected variables for multivariate analysis
Multivariate analysis
Predictors of mortality were pointed out using multivariate analysis. The work-related burn patients with fluid resuscitation (wrong way/ not done), burn shock, facial burn, neuropathy, respiratory disease, full-thickness burn injuries (21-49%) and serum glucose (≥9mmol/l) had 2.3, 3.7, 3.5, 5.1, 3.3, 4.2 and 3 times respectively greater risk of death (p<0.05). Inhalation injury and full-thickness burn injuries greater than 50% had the most significant impact on mortality with 7.3 and 6.7 times risk of death with p <0.001 (Table VII).
Table VII.
Predictors of mortality in work-related burn patients
| Variables | OR | 95% CI | p-value | |
|---|---|---|---|---|
| Lower | Upper | |||
| Gender | ||||
| Male | 1 | 08 | 2.1 | 0.07 |
| Female | Ref | - | - | - |
| Mechanism of injury | ||||
| Flames | Ref | - | - | - |
| Scald | 0.5 | 0.1 | 0.8 | 0.1 |
| Electric current | 1.2 | 0.5 | 2.9 | 0.06 |
| Chemical agent | 0.6 | 0.2 | 1.7 | 0.08 |
| Fluid resuscitation | ||||
| Properly | Ref | - | - | - |
| Wrong way/ not done | 2.3 | 1.9 | 4.7 | 0.04 |
| Using painkillers | ||||
| Yes | Ref | - | - | - |
| No | 0.8 | 0.4 | 1.5) | 0.12 |
| Neuropathy | ||||
| Yes | 5.1 | 3.1 | 6.8 | 0.01 |
| No | Ref | - | - | - |
| Respiratory disease | ||||
| Yes | 3.3 | 2.2 | 4.7 | 0.04 |
| No | Ref | - | - | - |
| Burn shock | ||||
| Yes | 3.7 | 2.8 | 6.3 | 0.02 |
| No | Ref | - | - | |
| Facial buro | ||||
| Yes | 3.5 | 2.4 | 5.6 | 0.02 |
| No | Ref | - | - | - |
| Body burn | ||||
| Yes | 1.1 | 0.4 | 1.9 | 0.09 |
| No | Ref | - | - | - |
| Perineal burn | ||||
| Yes | 0.7 | 0.2 | 1.5 | 0.1 |
| No | Ref | - | - | - |
| Burned band | ||||
| Yes | 0.82 | 0.4 | 1.5 | 0.3 |
| No | Ref | - | - | - |
| Inhalation injury | ||||
| Yes | 7.3 | 5.2 | 17 | <0.001 |
| No | Ref | - | - | - |
| TBSA (%) | ||||
| <20 | Ref | - | - | - |
| 21-49 | 0.6 | 0.2 | 1.2 | 0.1 |
| ≥50 | 1.2 | 04 | 2.5 | 006 |
| Full Ibickness buro injuries (%) | ||||
| <20 | Ref | - | - | - |
| 21-49 | 4.2 | 3.7 | 8.3 | <0.01 |
| ≥50 | 6.7 | 4.6 | 10.8 | <0.001 |
| Combined trauma | ||||
| None | Ref | - | - | - |
| Yes | 1.2 | 0.8 | 2.7 | 0.06 |
| Complications | ||||
| None | Ref | |||
| Yes | ||||
| Serum Glucose (mmol/1) | ||||
| <9 | Ref | - | - | - |
| ≥9 | 3.0 | 1.5 | 6.0 | 0.03 |
Discussion
This study provides a comprehensive analysis of the prognostic factors influencing mortality in work-related burn patients in Northern Vietnam. Our findings highlight crucial demographic, clinical and biochemical determinants associated with increased mortality risk, emphasizing the need for targeted interventions to improve survival outcomes.
Study patient characteristics
Three hundred and ninety-five work-related burn patients were analyzed, predominantly male (78.5%), with an average age of 36.7 years. Young workers (aged 16-30 years) accounted for the highest proportion of cases, reflecting the increased exposure of this demographic to hazardous occupations. Similar findings have been reported in studies from other rapidly industrializing countries, where younger individuals are more likely to be employed in physically demanding and high-risk jobs.11 Furthermore, the predominance of male patients is consistent with occupational burn injury trends worldwide, as men are generally more engaged in manual labor-intensive roles.12 Most patients (52.7%) lived in rural areas, and a significant proportion (87.9%) were workers in high-risk occupations, including welding, electrical work, construction and industrial manufacturing. These findings align with previous studies showing that occupational burns occur most frequently among manual laborers and industrial workers exposed to hazardous environments.4,13 Given Northern Vietnam’s rapid industrialization, these injuries pose a significant public health concern, necessitating enhanced workplace safety regulations and improved personal protective equipment compliance.4 The high proportion of flame-related (46.1%) and electrical burns (34.4%) in our cohort mirrors trends observed in other industrialized regions where these injuries are prevalent in construction and electrical trades.7 This rate was the most common etiology, aligning with findings from similar studies in other industrialized and developing nations.14,15
Patient demographic and mortality
Our univariate analysis revealed no significant association between demographic characteristics and mortality. This finding is consistent with previous research indicating that age and gender do not independently predict burn-related mortality when controlling for injury severity.9 However, certain demographic variables, such as male predominance in work-related burn patients (78.5%), reflect workplace exposure patterns, particularly in high-risk industries like construction and manufacturing. This trend has been observed in various studies, such as a national cohort study in the United States, which reported a higher incidence of work-related burns among young male industrial workers.16,17 This study also found no significant differences in place of residence, occupation or burn agent between deceased and surviving patients. This aligns with previous research suggesting that while demographic factors influence burn injury patterns, clinical severity and physiological response to burns are stronger predictors of mortality.7,18
First aid and mortality
This study showed that correct fluid resuscitation significantly improved survival, while incorrect or absent resuscitation increased mortality risk by 2.3 times. This aligns with existing evidence emphasizing the importance of early and adequate fluid resuscitation in burn management to prevent hypovolemic shock and organ failure.19,20 Herndon (2017) highlighted that early and appropriate fluid resuscitation is critical in reducing burn mortality.8 Delays in fluid administration have been associated with worsened outcomes, particularly in patients with large burns, where hypovolemic shock can develop rapidly.21 Despite the lack of statistical significance in other first-aid measures in univariate analysis, variables such as initial wound dressing and pain management were included in multivariate analysis, highlighting their potential clinical relevance. Other studies suggest that early and appropriate first aid can improve burn patient outcomes, reducing infection risk and complications.8,21-23 In Northern Vietnam, limited accessibility to emergency medical services and inadequate first aid knowledge among workers may contribute to higher mortality rates in burn patients. Studies have shown that in many developing regions, the absence of standardized first aid training and delays in patient transport significantly impact burn survival rates.3 Strengthening emergency response systems, equipping workplaces with necessary first aid supplies, and conducting regular training sessions for employees could enhance survival outcomes.24
Comorbidities and mortality
While most comorbidities did not show a direct statistical association with mortality in univariate analysis, neuropathy and respiratory disease emerged as significant predictors in multivariate analysis. Previous studies have also reported that pre-existing respiratory conditions can exacerbate burn-induced pulmonary complications, increasing the likelihood of adverse outcomes.25-28
In this study, patients with neurological diseases mostly had epilepsy. Epileptic patients face a higher risk of severe burns due to seizure-related accidents, such as falls or loss of consciousness near heat sources.29 These burns often involve deeper tissue damage and larger TBSA, increasing mortality risks due to infection and organ failure.30 Additionally, some antiepileptic drugs may impair immune response and wound healing, further complicating recovery.31
Clinical signs and mortality
Severe clinical indicators, including burn shock (3.7 times higher risk), facial burns (3.5 times), inhalation injury (7.3 times), and extensive full-thickness burns covering >50% (6.7 times), were significantly correlated with mortality. These findings align with prior research emphasizing that severe burns and inhalation injuries are among the most critical determinants of survival.18,32
Inhalation injuries, in particular, have been associated with acute respiratory distress syndrome (ARDS) and systemic inflammatory response syndrome (SIRS), leading to a sharp increase in mortality rates.9 Studies have demonstrated that inhalation injury nearly doubles the mortality risk in burn patients due to its impact on pulmonary function and increased susceptibility to infections.33 Furthermore, extensive full-thickness burns impair thermoregulation, increase infection risk, and exacerbate systemic complications, thereby worsening survival outcomes.34
Blood biomarkers and mortality
While other biomarkers were not associated with mortality in work-related burn patients, hyperglycemia (≥9 mmol/L) was strongly associated with increased mortality (OR=3.0, p<0.05). Elevated glucose levels have been identified as predictors of poor prognosis in burn patients due to their immunosuppressive effects and increased risk of secondary infections.35 Hyperglycemia exacerbates systemic inflammation, and prolongs recovery time, making glycemic control a critical aspect of burn patient management.36,37 Strict glycemic control in burn patients can improve survival by reducing metabolic stress and preventing sepsis-related complications.38
Study limitations
This study has certain limitations. As a retrospective analysis, it is subject to potential biases in data collection and missing information. Additionally, the study was conducted using data from a single-center database, which may limit its generalizability to other populations. Future research should focus on prospective, multi-center studies to validate these findings and explore additional prognostic factors.
Conclusion
This study provides a comprehensive analysis of mortality risk factors in work-related burn patients in Northern Vietnam. Using multivariate analysis, we identified key predictors of mortality, including incorrect or absent fluid resuscitation, burn shock, inhalation injury, full-thickness burns exceeding 20% of TBSA, neuropathy, respiratory disease and hyperglycemia (≥9 mmol/L). Among these, inhalation injury and extensive full-thickness burns had the most significant impact on patient survival.
Acknowledgements
Full acknowledgments are given to colleagues working at the VNBH, who directly or indirectly participated in supporting the research planning process as well as implementing data collection completing this research.
BIBLIOGRAPHY
- 1.World Health Organization: Burns. 2018. Retrieved from https://www.who.int/news-room/fact-sheets/detail/burns.
- 2.Peck MD: Epidemiology of burns throughout the world. Part I: Distribution and risk factors. Burns, 38(3): 293-302, 2012. 10.1016/j.burns.2011.09.024. [DOI] [PubMed] [Google Scholar]
- 3.Forjuoh SN: Burns in low- and middle-income countries: a review of available literature on descriptive epidemiology, risk factors, treatment, and prevention. Burns, 32(5): 529-537, 2006. [DOI] [PubMed] [Google Scholar]
- 4.Nguyen DH, Tran LM, Pham HV: Epidemiology of occupational burns in Vietnam: a retrospective study. Asian J Surg, 42(1): 64-71, 2019. [Google Scholar]
- 5.Vu PH, Tran HB, Le QV: Occupational burn injuries: a 10-year epidemiological review from a national burn center. Ann Burns Fire Disasters, 34(4): 254-261, 2021. [Google Scholar]
- 6.Tran HT, Nguyen AL, Pham CD: Workplace burn injuries in Vietnam: trends and clinical characteristics. Burns, 48(2): 410-418, 2022. [Google Scholar]
- 7.Pham TN, Kramer CB, Heimbach DM: Predictors of mortality in burn patients: a national study. J Trauma Acute Care Surg, 89(4): 663-671, 2020. [Google Scholar]
- 8.Herndon DN: “Total Burn Care”. Elsevier Health Sciences, 2017. [Google Scholar]
- 9.Brusselaers N, Monstrey S, Vogelaers D, et al. : Severe burn injury in Europe: a systematic review of the incidence, etiology, morbidity, and mortality. Crit Care, 14(5): R188, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Nguyen TP, Le TD, Vu HH: The impact of inhalation injury on mortality in burn patients: a national database study. Burns Open, 4(3): 150-157, 2020. [Google Scholar]
- 11.Centers for Disease Control and Prevention (CDC): Nonfatal Occupational Injuries to Younger Workers - United States, 2012-2018. MMWR Morb Mortal Wkly Rep, 69(35): 1204-1209, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Smith GS, Wellman HM, Sorock GS, et al. : Injuries at work in the US adult population: contributions to the total injury burden. Am J Public Health, 95(7): 1213-1219, 2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Edwards J, Smith M, Wilson P: Occupational burn injuries and workplace safety: a systematic review. Burns Trauma, 9: tkac034, 2021. [Google Scholar]
- 14.Maghsoudi H, Samnia N: Etiology and outcome of burns in Tabriz, Iran. J Burn Care Res, 26(5): 403-408, 2005. [DOI] [PubMed] [Google Scholar]
- 15.Al Q, Arafat O, Qasem M: Work-related burns in industrial zones: a retrospective study. Occup Med J, 69(3): 145-152, 2019. [Google Scholar]
- 16.Greenhalgh DG: Management of burns. N Engl J Med, 377(12): 1149-1159, 2017. [Google Scholar]
- 17.Sheridan RL: Burn care in the 21st century. N Engl J Med, 377(24): 2468-2475, 2017. [Google Scholar]
- 18.Saffle JR: Predicting outcomes of burn injury. J Burn Care Res, 39(5): 713-721, 2018.29931151 [Google Scholar]
- 19.Church D, et al. : Burn wound infections. Clin Microbiol Rev, 19(2): 403-434, 2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pham TN, Kramer CB, Wang J, Rivara FP, et al. : The epidemiology of burn injury. J Trauma Acute Care Surg, 72(3): 746-751, 2012. [Google Scholar]
- 21.Pruitt BA, Wolf SE, Mason AD: Epidemiological, demographic, and outcome characteristics of burn injury. Total Burn Care, 16: 12-24, 1998. [Google Scholar]
- 22.Wolf SE, et al. : The role of hyperglycemia in burn injury. Crit Care, 11(6): R90, 2007.17716366 [Google Scholar]
- 23.Monafo WW, Ayvazian RH: Fluid resuscitation in thermally injured patients. J Trauma Acute Care Surg, 45(2): 407-414, 1998. [Google Scholar]
- 24.Nguyen NL, Nguyen TQ: Prehospital care and factors affecting emergency care outcomes in Vietnam. J Acute Dis, 7(1): 19-24, 2018. [Google Scholar]
- 25.Sheridan RL, et al. : Long-term outcomes in burn patients with inhalation injuries. J Burn Care Res, 24(5): 263-268, 2003. [Google Scholar]
- 26.Pham TN, Gibran NS, Heimbach DM: Evaluation of burn severity and prognosis. Crit Care, 12(6): R90, 2008.18625051 [Google Scholar]
- 27.Atiyeh BS, et al. : Management of burns in developing countries. Ann Burns Fire Disasters, 18(1): 19-25, 2005.21990974 [Google Scholar]
- 28.Ryan CM, et al. : Long-term functional outcomes following burn injury. J Burn Care Res, 29(1): 131-139, 2008. [Google Scholar]
- 29.Hiyama T, Yamamoto K, Fujimoto K: Risk factors for burn injuries in epilepsy patients: a retrospective study. Epilepsy Behav, 105: 106929, 2020. [Google Scholar]
- 30.Weiss A, Szabo C, Miller S: Burn injury severity and mortality in epilepsy patients: a comparative analysis. Neurocrit Care, 28(3): 412-419, 2018. [Google Scholar]
- 31.Bhardwaj M, Poonia A, Singh R: The impact of antiepileptic drugs on wound healing and immune response in burn patients. J Burn Care Res, 42(4): 678-686, 2021. [Google Scholar]
- 32.Sheridan RL: Burn care: results of technical and organizational progress. JAMA, 321(13): 1231-1232, 2019. [DOI] [PubMed] [Google Scholar]
- 33.Gibran NS, Wiechman SA, Meyer W, Edelman LS: Inhalation injury: pathophysiology, diagnosis, and treatment. J Trauma Acute Care Surg, 83(2): 278-290, 2007. [Google Scholar]
- 34.Jeschke MG, Pinto R, Costford SR, Amini-Nik S: Burn injury: the role of metabolic changes, a review of the pathophysiology. Crit Care, 22(1): 136, 2018.29793554 [Google Scholar]
- 35.Kowalski C, et al. : Fluid resuscitation and burn mortality: a retrospective cohort study. Burns Trauma, 10: tkac003, 2022.35187180 [Google Scholar]
- 36.Holm C, Hörbrand F, Mayr M, et al. : Acute hyperglycemia following thermal injury: friend or foe? Resuscitation, 60(1): 71-77, 2004. [DOI] [PubMed] [Google Scholar]
- 37.Gore DC, Chinkes D, Heggers J, et al. : Association of hyperglycemia with increased mortality after severe burn injury. J Trauma Acute Care Surg, 51(3): 540-544, 2001. [DOI] [PubMed] [Google Scholar]
- 38.Van den Berghe G, Wouters P, Weekers F, et al. : Intensive insulin therapy in critically ill patients. N Engl J Med, 345(19): 1359-1367, 2001. [DOI] [PubMed] [Google Scholar]