Burn Injuries in Jordan: A 5‐Year Retrospective Analysis of Presentation, Management and Hospital Mortality

ABSTRACT

Burn injuries are a significant cause of morbidity and mortality globally; however, limited data are available from low‐ and middle‐income countries such as Jordan. This study aimed to describe burn patient presentation, initial management and factors associated with in‐hospital mortality. A retrospective descriptive study was conducted using records of 493 patients admitted to a national referral centre in Jordan between 2018 and 2022. The sample was predominantly male (61.5%) with a mean age of 19.6 years (SD = 21); children under 18 years comprised 58.4%. The mean total body surface area (TBSA) burned was 18%. Flame (50.1%) and scald (44.6%) injuries were most common. Inhalation injury occurred in 25.8% and 21.3% required mechanical ventilation. The hospital mortality rate was 15.6%, significantly associated with TBSA, age, inhalation injury and low serum total protein. Baux and revised Baux scores showed high predictive accuracy (AUC = 0.902 and 0.918). Logistic regression identified TBSA, age, inhalation injury and total protein level as independent predictors of mortality. Burn injuries in Jordan disproportionately affect children and are associated with substantial mortality. Early identification of high‐risk patients using validated scores and prompt nutritional and respiratory interventions are essential. Multicentre studies and a national burn registry are recommended to guide future policy and care improvements.

Summary

• Burn wounds in Jordan predominantly affect children and are associated with substantial mortality (15.6%), highlighting a major pediatric wound‐care burden

• Wound severity, inhalation injury, and patient age were strong independent predictors of death

• Low total serum protein was independently associated with mortality, suggesting nutritional and inflammatory status are critical determinants of burn wound outcomes.

• Baux and revised Baux scores demonstrated excellent predictive accuracy (AUC > 0.90), supporting their use for early risk stratification in acute burn care.

1. Background

Burn injuries represent one of the most severe and life‐threatening forms of trauma, with significant implications for both physical and psychological health. Globally, it is estimated that between 7 and 12 million people suffer burn injuries each year that require medical care [1]. Despite medical advancements, burn injuries continue to be a leading cause of injury‐related mortality, with approximately 180 000 deaths reported annually by the World Health Organization [2].

Burn injuries initiate complex local and systemic responses in the human body. Locally, the burn wound is traditionally described by Jackson's three zones: hyperaemia, stasis and coagulation, each indicating different levels of tissue viability and damage [3, 4]. Systemically, burns provoke a hypermetabolic and inflammatory state that can persist beyond the initial injury, leading to profound pathophysiological changes such as hypovolemia, cardiac dysfunction, renal failure and respiratory complications [5, 6]. The extent and depth of burn, along with other clinical features like inhalational injury and electrolyte imbalance, are crucial in determining patient outcomes.

Inhalational injury, caused by the inhalation of smoke and toxic substances, significantly worsens burn prognosis. It directly affects the pulmonary epithelium and may lead to hypoxia, increased oxygen demand and systemic inflammatory response, contributing to respiratory failure and increased mortality [7, 8]. The severity of burns is often quantified by total body surface area (TBSA) involvement, with TBSA and age being the most consistently validated predictors of mortality. These factors form the foundation for the Baux score and its revised version, which incorporate inhalational injury as an additional mortality determinant [5].

Electrolyte disturbances and laboratory abnormalities are also implicated in burn‐related mortality. Studies have shown that variables such as hypocalcaemia, hypoalbuminemia, elevated creatinine and hyperglycaemia are significantly associated with worse outcomes [9, 10]. For instance, hypocalcaemia has emerged as an independent predictor of trauma mortality and is linked to coagulopathy and myocardial depression [11]. Similarly, hypoalbuminemia reflects capillary leakage and protein loss due to systemic inflammation, correlating with increased risk of mortality [12].

Effective burn management begins in the first 24 h postinjury, where the focus is on fluid resuscitation, wound care and stabilisation. The Parkland formula, one of the most widely accepted resuscitation protocols, plays a key role in addressing burn shock through calculated fluid replacement [5]. Topical agents such as silver sulfadiazine and antiseptics like povidone‐iodine are commonly used to prevent infection and promote healing. Despite the adoption of standard care protocols, variability in clinical outcomes remains, often due to differences in injury severity, initial physiological responses and local healthcare infrastructure.

While extensive literature exists regarding burn management and mortality predictors in high‐income countries, there remains a gap in the evidence from low‐ and middle‐income countries, particularly in the Middle East. In Jordan, existing studies have largely focused on postburn psychological consequences, quality of life or limited demographic descriptions [13, 14]. Previous investigations reported mortality rates between 8.5% and 14.6% but did not extensively examine the relationships between early presentation, clinical characteristics, management protocols and in‐hospital mortality [15]. Given the paucity of comprehensive, data‐driven studies on burn patient outcomes in Jordan, there is an urgent need to investigate the acute presentation, management and mortality predictors in this setting. A context‐specific understanding of these factors is essential for guiding clinical practice, improving patient outcomes and informing public health strategies. Therefore, the objectives of this study were to describe burn patients' presentation, injury characteristics and initial management and to examine their association with in‐hospital mortality over a 5‐year period at a national referral centre in Jordan.

2. Methods

2.1. Design

This study employed a retrospective descriptive design to analyse hospital records over a 5‐year period.

2.2. Setting

The study was conducted at a specialised burn care unit in Jordan, which functions as a national referral centre. The unit adheres to the American Burn Association (ABA) criteria, is equipped for critical burn care, and receives patients from various healthcare sectors across the country.

2.3. Sample and Sampling

A census sampling strategy is used in this study. The study sample included all patients admitted to the selected burn unit between January 2018 and December 2022. The inclusion criteria were all patients admitted with burn‐related injuries during the specified period, regardless of age, sex or cause. Nonburn cases such as Stevens‐Johnson syndrome and toxic epidermal necrolysis were excluded.

2.4. Measurements

A structured data collection form was developed and validated by a research team based on the relevant literature and clinical guidelines. This tool served as the primary instrument for extracting variables from patient records. It comprises four domains: [1] demographic information, [2] clinical and burn‐specific variables, [3] laboratory indicators on admission and [4] patient outcomes. The demographic variables included age, sex and residence. Clinical variables included TBSA, type and mechanism of burns, presence of inhalational injury and comorbidities. Inhalation injury was identified based on the clinical diagnosis documented in the medical records. These included closed‐space exposure, presence of facial burns, singed nasal or facial hair, hoarseness of voice and signs of nasal or oral mucosal involvement. Laboratory variables included WBC, haemoglobin, haematocrit, platelet count, albumin, total protein, calcium, urea, creatinine, sodium, potassium and random blood glucose. Outcome variables included the length of hospital stay, ICU admission and in‐hospital mortality.

TBSA was measured using the Lund‐Browder chart, which provides an age‐adjusted estimation of burned body surface area. Inhalational injury was determined based on the clinical diagnosis documented in the patient's record, typically informed by signs such as facial burns, singed nasal hairs, carbonaceous sputum and findings on bronchoscopy or imaging when available. The data collection tool was piloted on a sample of 20 records to ensure content validity and identify and resolve ambiguities or inconsistencies in data extraction prior to full implementation.

2.5. Data Collection

Data were retrieved from electronic medical records archived in the hospital. The structured data collection form described above was used to guide consistent extraction of information across four domains: demographics, clinical presentation, initial clinical and laboratory findings and patient outcomes. To gain access to patient records, the researcher obtained formal administrative approvals from the hospital's research and medical records departments following ethical clearance. Access credentials and permissions were granted through secure institutional protocols in compliance with hospital privacy regulations. The researchers completed the training on data confidentiality and documentation protocols required by the institution. Data collection was conducted over a 3‐month period. The research team accessed the patient files through secure systems, and any missing or ambiguous entries were cross‐referenced with additional clinical documentation.

2.6. Ethical Considerations

Ethical approval was obtained from the relevant institutional review board. Patient confidentiality was maintained strictly. Data were anonymised and stored securely, and access was restricted to the research team.

2.7. Data Analysis

The data were analysed using IBM SPSS version 26. Descriptive statistics were used to summarise the patient characteristics. Inferential analyses included chi‐square tests and point‐biserial correlations to explore associations with mortality. A forward stepwise binary logistic regression approach was used to select variables for the final model. Composite mortality scores (Baux and revised Baux) were not included in the multivariable regression to avoid collinearity with their component variables (age and TBSA). Receiver operating characteristic (ROC) curves were used to evaluate the predictive accuracy of the Baux and revised Baux scores. Missing laboratory data were handled using case‐wise deletion for inferential analyses. The number of valid observations (n) is reported for each laboratory variable, and no imputation was performed.

3. Results

3.1. Participant Demographics

A total of 493 patients admitted with burn injuries between January 2018 and December 2022 were included in the analysis. The majority were male (61.5%, n = 303), and the mean age was 19.6 years (SD = 21). Children under 18 years comprised 58.4% (n = 288) of the sample, and 23.3% (n = 115) were younger than 2 years. Full demographic and clinical descriptors of the sample are presented in Table 1.

TABLE 1.

Sample characteristics.

Variable	Frequency (%)	Mean (SD)
Age (years)		19.6 (21)
Gender
Male	303 (61.5)
Female	190 (38.5)
Total body surface area burned		18 (19)
Length of stay (days)		17.9 (24.1)
Albumin		3.4 (0.9)
Total protein		5.7 (1.2)
Mechanism of injury
Direct flame burn	247 (50.2)
Scald	220 (44.7)
Electrical	13 (2.6)
Others	13 (2.6)
Inhalational injury
Yes	127 (25.8)
No	366 (74.2)
Need for mechanical ventilation
Yes	105 (21.3)
No	388 (78.7)
Hospital discharge outcome
Dead (mortality)	77 (15.6)
Survived	416 (84.4)

3.2. Burn Injury Presentation

The mean TBSA affected was 18% (SD = 18.98), ranging from 0.5% to 100%. The leading mechanisms of injury were direct flame burns (50.1%, n = 247) and scalds (44.6%, n = 220). Inhalation injury was identified in 25.8% (n = 127), and facial burns were reported in 36.7% of patients. A total of 21.3% (n = 105) required mechanical ventilation. These findings are detailed in Table 1.

3.3. Initial Management Interventions

Among the patients, 78% (n = 382) received fluid resuscitation using the Parkland formula during the first 24 h of care. Lactated Ringer's solution was administered in 40.4% of cases, while Dextrose 0.45% saline was used in 49%. Silver sulfadiazine was applied as a topical antimicrobial in 99.6% of cases (n = 491), and povidone‐iodine was the disinfectant used in all documented cases. A summary of initial management interventions is presented in Table 2.

TABLE 2.

Initial burn management interventions.

Intervention	N		Frequency (%)
Intravenous fluid	490
		Lactated ringer	198 (40.4%)
		Dextrose 0.9 saline	43 (8.8%)
		Dextrose 0.45 saline	240 (49%)
		Dextrose 0.225 Normal saline	9 (1.8%)
Topical antimicrobial	493
		Silver sulfadiazine	491 (99.6%)
		Neomycin‐ointment	2 (0.4%)
Antiseptic solution	380	Povidone iodine	380 (100%)

3.4. Hospital Mortality

The overall hospital mortality rate was 15.6% (n = 77). Mortality was significantly associated with the mechanism of injury (p ≤ 0.001), presence of inhalation injury (p ≤ 0.001) and need for mechanical ventilation (p ≤ 0.001). There was no significant association between sex and mortality (p = 0.397). Associations between categorical variables and mortality are reported in Table 3.

TABLE 3.

Chi‐square analysis of risk factors associated with mortality.

Variable		Survivors	Mortality	Person chi‐	DF	p
		N (%)	N (%)
Gender	Male	259 (62.2%)	44 (57.1%)	0.71	1	0.397
	Female	157 (37.8%)	33 (42.8%)
Mechanism of injury	DFB	180 (43.3%)	67 (87%)	50.09	1	≤ 0.001
	Scald	210 (50.4%)	10 (13%)
Facial burns	Yes	128 (30.6%)	53 (68.8%)	40.512	1	≤ 0.001
	No	288 (69.2%)	24 (31.2%)
Inhalational injury	Yes	65 (15.6%)	62 (80.5%)	143.07	1	≤ 0.001
	No	351 (84.4%)	15 (19.5%)
Mechanical ventilation	Yes	31 (7.5%)	74 (96.1%)	304.64	1	≤ 0.001
	No	385 (92.5%)	3 (3.8%)

Point biserial correlation analysis demonstrated that several continuous variables were significantly associated with mortality, including age, TBSA, albumin, total protein (p ≤ 0.05). Length of stay was not significantly associated with mortality. These results are presented in Table 4.

TABLE 4.

Point biserial correlation test for mortality associated risk factors.

Variable	n	Mean	SD	rpb	p
Age	493	19.61	21	0.31	< 0.001
TBSA	493	18	18.9	0.69	< 0.001
Burned area	493	17.8	24.1	0.003	0.971
Albumin	475	3.4	0.8	−0.54	< 0.001
Total protein	468	5.7	1.21	−0.5	< 0.001

3.5. Predictive Validity of Baux and Revised Baux Scores

The mean Baux score was significantly higher among nonsurvivors (83.02 ± 32.53; t = 13.85, p ≤ 0.001). Similarly, the revised Baux (r‐Baux) scores were also significantly higher among nonsurvivors, with a mean of 96.7 ± 35.94 (t = 15.08, p ≤ 0.001). ROC curve analysis demonstrated excellent predictive accuracy for mortality, with the area under the curve (AUC) for the Baux score was 0.902 (95% CI: 0.859–0.937), and for the revised Baux score was 0.918 (95% CI: 0.878–0.949). At a cut‐off point of 50.25, both scores achieved 85% sensitivity, with the Baux score yielding a slightly lower false‐positive rate. The ROC curve illustrating the performance of both scores is shown in Figure 1.

FIGURE 1.

ROC curve for the Baux and r‐Baux score.

3.6. Predictors of Mortality

A binary logistic regression model was used to identify independent predictors of hospital mortality. The final model included TBSA, age, presence of inhalation injury and serum total protein. The model was statistically significant (χ ² = 225.0, df = 5, p < 0.001) and demonstrated good calibration (Hosmer–Lemeshow χ ² = 10.55, df = 8, p = 0.225). The pseudo‐R‐squared values were Cox & Snell R ² = 0.403 and Nagelkerke R ² = 0.689, indicating that the model explained 40.3%–68.9% of the variance in mortality. The regression coefficients, odds ratios and 95% confidence intervals are detailed in Table 5.

TABLE 5.

Results of binary logistic regression analysis for independent predictors of hospital mortality.

Variable	Coefficient	Odds ratio	95% CI	p
TBSA	0.07	1.072	1.04–1.105	≤ 0.001
Age	0.45	1.046	1.025–1.067	≤ 0.001
Inhalational injury	1.291	3.637	1.49–8.875	0.005
Total protein	−0.882	0.414	0.275–0.623	≤ 0.001
Constant	−1.851

In this final model, TBSA significantly increased the odds of mortality (OR = 1.072; 95% CI: 1.04–1.11; p < 0.001), as did age (OR = 1.046; 95% CI: 1.025–1.067; p < 0.001). Inhalation injury was associated with a 3.6‐fold increase in the odds of mortality (OR = 3.637; 95% CI: 1.49–8.88; p = 0.005), while total protein was inversely associated with mortality (OR = 0.414; 95% CI: 0.275–0.623; p < 0.001).

4. Discussion

4.1. Presentations and Characteristics of Burn Patients

Burn injuries incidence and severity vary across countries, care settings and population subgroups [16]. Comprehensive knowledge of burn characteristics—including patient demographics, injury mechanisms and management practices—is essential to improve outcomes, reduce complications and lower mortality rates [17]. In Jordan, detailed data on burn patient profiles remain scarce, highlighting the importance of this study.

The predominance of male patients is consistent with global reports. The Global Burn Registry [18] showed 38% of cases occurred in females. Similar patterns were observed in Vietnam (69.3% male) [19], the United States [20] and China (57.3%–76.3% male) [21]. In contrast, a study in India reported a higher female incidence (58.9%) [22], possibly due to domestic exposures and socioeconomic factors. Our findings affirm a global male predominance.

Paediatric patients represented half of our sample, aligning with global trends. The Global Burn Registry reported 42% of burn injuries occurred in children under 18, with 62% in those under five [23]. In the United States, approximately 30% of burns affected children under five [24]. In China, 40.1% of burn cases were paediatric [21]. Children's developmental vulnerability, limited risk awareness and unsafe environments contribute to this high incidence.

Regarding burn mechanisms, flame injuries were most common, followed by scalds. This pattern mirrors reports from India, Iran, Italy and Australia [25, 26, 27, 28]. In contrast, studies in the United States, China and Saudi Arabia noted a higher prevalence of scalds, especially among children [21, 29, 30]. The Global Burn Registry also highlighted scalds as the predominant mechanism in paediatric burns [23].

The average TBSA burned in this study was comparable to Italy (18.4%) [27], but lower than Iran (35.9%–50.7%) [26] and India, where 33% of cases had TBSA > 75% [25]. China reported a lower TBSA average of 13.64% [21]. These differences reflect regional variation in injury severity and hospital referral patterns.

Inhalation injuries were diagnosed in 25.8% of patients, and 80.5% of those affected died. This incidence is higher than the 15.7% pooled rate from Galeiras et al. but aligns with ICU‐specific data. Dyamenahalli et al. reported inhalation injuries increased mortality risk 10‐fold [7, 31]. In Saudi Arabia, 72% of patients with inhalation injuries died [30]. Diagnostic criteria inconsistencies likely explain variability between studies.

4.2. Initial Management Used in Burn Management

Effective early management in burn care is crucial. Fluid resuscitation follows protocols like the Parkland formula, which accounts for TBSA, weight and injury severity [4]. In this study, patients were managed with Parkland, but only 40.4% received Ringer's lactate—the recommended fluid. In children under 12, glucose‐containing fluids were used, in line with paediatric guidelines [32, 33].

Topical treatments included silver‐based dressings and antiseptics. Povidone‐iodine was the most used antiseptic, consistent with global practices [34]. Its broad antimicrobial effect and moisture retention enhance healing. Compared to chlorhexidine—which may inhibit cytokine signalling and delay re‐epithelialisation—iodine has superior outcomes [35, 36]. The combination of iodine disinfection followed by silver‐based dressing, although effective, may require refinement to reduce preparation time and risk of iodine accumulation. Research into alternative antiseptics (e.g., vinegar‐based solutions) could offer safer, equally effective options.

4.3. Factors Associated With Burn‐Related Hospital Mortality

The mortality rate in this study was consistent with Saudi Arabia (17.6%) [30] and Iran (18.1%) [26] but higher than Ethiopia (6.6%) [37], China (0.81%) [21] and the United States (2.7%) [38]. Variation may reflect differences in study period, socioeconomic status, care infrastructure and burn centre capacity [16]. Standardised reporting using prediction models can enhance cross‐study comparisons [39]. Sex was not significantly associated with mortality, contradicting studies showing worse outcomes for females due to higher TBSA or treatment disparities [22, 40]. Notably, the Belgian model includes female sex as a mortality predictor [41].

Age and TBSA were major predictors of death. Larger TBSA reflects greater physiological insult, fluid needs and infection risk [5]. Our average TBSA was 18%, and the mean age was 19.6 years. These findings support the established relationship between these variables and mortality [42, 43]. Mechanism of injury, facial burns and inhalation injuries were also linked to mortality, supporting previous studies [7, 44]. Inhalation injury increased mortality odds 3.6‐fold, consistent with Galeiras et al., who reported a pooled relative risk of 2 [31].

Among all factors examined, total protein emerged as a strong protective predictor. Survivors had higher TP levels (5.97 vs. 4.29 g/dL). Gupta et al. similarly found low TP to be a poor prognostic sign. Since total protein reflects the collective status of albumin, globulin and prealbumin, it may serve as a faster indicator of nutritional status and systemic inflammation than albumin alone [45, 46, 47, 48]. Beyond its association with nutritional status, total protein may reflect the magnitude of systemic inflammatory response and capillary permeability following burn injury, both of which contribute to protein redistribution and loss. Its prognostic value may therefore lie in capturing the combined effects of inflammation, metabolic stress and early nutritional depletion, making it a pragmatic risk indicator in acute burn care, particularly in resource‐limited settings.

The Baux and r‐Baux scores—based on age, TBSA and inhalation injury—were validated in this study. The findings align with global findings reporting AUCs above 0.9 [49, 50]. A cut‐off of 50.25 yielded 85% sensitivity for both models, though other settings have reported thresholds between 70 and 85 [51, 52, 53]. General ICU scoring systems like qSOFA and APACHE II showed lower predictive power (AUCs 0.62–0.73) [52, 54], underscoring the need for burn‐specific models. Coagulopathy markers (INR, PTT) also showed promise in enhancing prediction, as reported by Kaita et al. [55].

4.4. Implications for Clinical Practice

This study identifies key clinical factors influencing hospital mortality in burn patients and provides actionable guidance for frontline healthcare providers. Implementing routine predictive tools like the Baux and revised Baux scores can optimise triage for high‐risk patients, guiding intensive monitoring and aggressive management. Notably, lower total protein levels correlate with increased mortality, emphasising the need for nutritional assessment and early intervention in acute burn care. While the widespread use of the Parkland formula and silver sulfadiazine dressings at the study site aligns with international protocols, the findings underscore the necessity for localised adaptation and individualised care—particularly for paediatric and geriatric populations, who represent a substantial proportion of cases.

4.5. Limitations

Although this study provides valuable insights, several limitations should be acknowledged. First, data on burn depth, time to definitive surgical intervention and long‐term outcomes were not consistently available and could not be included in the analysis, which may have limited the assessment of their impact on patient prognosis. Second, the diagnosis of inhalational injury relied on clinical judgement rather than bronchoscopy or standardised criteria, which may have introduced interobserver variability and potential misclassification. Third, although the study considered a wide range of variables, it did not explore long‐term outcomes, such as disability, psychological distress or readmission rates, which are crucial for a holistic understanding of burn recovery. The absence of data on the depth of burn and time to definitive surgical intervention are further limitations that may influence the interpretation of the mortality risk. Finally, as this study was conducted at a single national referral centre, the findings may be influenced by referral bias, with a potential overrepresentation of patients with more severe burn injuries.

5. Conclusion

This study underscores the complexity of burn injuries and identifies the critical factors that influence hospital mortality in patients with burns. This affirms the relevance and reliability of Baux and r‐Baux scores as mortality predictors and stresses the prognostic importance of both clinical presentation and early laboratory parameters. With a reported mortality rate of 15.6%, these findings emphasize the need for early identification of high‐risk patients, structured burn management protocols and resource investment in acute care and prevention.

Future studies should aim for multicentre designs, including long‐term patient outcomes and evaluate the impact of interventions targeting modifiable risk factors such as nutritional status and early respiratory support. The development of a national burn registry and continuous professional training in burn assessment and management is strongly recommended to improve patient outcomes and reduce burn‐related mortality in Jordan.

Funding

The authors have nothing to report.

Ethics Statement

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committees of the University of Jordan and the Royal Medical Services (consent Human Research Ethics Committee, Royal Medical Services, Amman, Jordan, Ref. n 2353, January 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

Alshdowh H., Alshraideh J. A., Al Qadire M., and Abdelrahman H., “Burn Injuries in Jordan: A 5‐Year Retrospective Analysis of Presentation, Management and Hospital Mortality,” International Wound Journal 23, no. 2 (2026): e70839, 10.1111/iwj.70839.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.