Title
Author
DOI
Article Type
Special Issue
Volume
Issue
Crush syndrome outcomes after earthquake
1Department of Emergency Medicine, Faculty of Medicine, Dicle University, 21280 Diyarbakır, Türkiye
2Department of Emergency Medicine, 25 Aralık Government Hospital, 27090 Gaziantep, Türkiye
*Corresponding Author(s):mustafa.icer@dicle.edu.tr (Mustafa İçer)
| History | Submitted: 18 February 2025 | Accepted: 27 April 2025 | Published: 08 October 2025 |
| Copyright: | ©2025 The Author(s). Published by MRE Press. |
Background: Earthquakes are catastrophic natural disasters that result in complex and severe medical conditions such as the crush syndrome. This study aimed to investigate the association between trauma scores, laboratory findings, mortality, and the development of crush syndrome in patients affected by the Kahramanmaraş Pazarcık earthquake. Methods: A retrospective analysis was conducted on trauma cases from the Kahramanmaraş Pazarcık earthquake, which occurred on 06 February 2023, and were treated at the Dicle University Hospital emergency department. Receiver operating characteristic (ROC) curve analysis and logistic regression were used to assess predictors of mortality and crush syndrome. Results: Of the 118 patients included, 110 survived and 8 died. The injury severity score (ISS) was the most accurate predictor of mortality (area under the curve (AUC): 0.935; sensitivity: 87.5%; specificity: 90%; positive predictive value (PPV): 38.89%; negative predictive value (NPV): 99%). Potassium level was the most predictive laboratory parameter (AUC: 0.894; sensitivity: 87.5%; specificity: 83.64%; PPV: 28%; NPV: 98.92%). However, multivariate analysis did not identify these variables as independent risk factors for mortality. Crush syndrome developed in 70 patients (59.3%), ISS was identified as the best trauma score for predicting crush syndrome (AUC: 0.822; sensitivity: 87.14%; specificity: 64.58%; PPV: 78.21%; NPV: 77.50%), and among laboratory markers, aspartate aminotransferase (AST) demonstrated the highest predictive ability (AUC: 0.850; sensitivity: 88.57%; specificity: 68.75%; PPV: 80.52%; NPV: 80.49%). Multivariate analysis revealed that C-reactive protein (CRP) (odds ratio (OR): 1.018; 95% confidence interval (CI): 1.005–1.030; p = 0.006), lactate dehydrogenase (LDH) (OR: 0.997; 95% CI: 0.994–1.000; p = 0.030), and ISS (OR: 1.105; 95% CI: 1.019–1.197; p = 0.015) were independent predictors of crush syndrome. Conclusions: In conclusion, CRP, LDH, and ISS were identified as independent risk factors for crush syndrome among earthquake victims. In contrast, neither trauma scores nor laboratory parameters independently predicted mortality.
Cite this article
Mustafa İçer, Tuğçe Bayrak. Crush syndrome outcomes after earthquake. Signa Vitae. 2025; 21(10): 40-52. doi: 10.22514/sv.2025.137
1. Introduction
Earthquakes are among the deadliest natural disasters worldwide, frequently resulting in substantial loss of life. To date, fatal earthquakes have been documented in 117 countries, with 52 of these experiencing at least one event leading to more than 1000 deaths [1]. Türkiye, located in a seismically active region, has endured more than 100 fatal earthquakes between 1927 and 2022, resulting in approximately 100,000 deaths [1].
On 06 February 2023, two devastating earthquakes struck the southeastern Anatolia region of Türkiye. The first, with a magnitude of Mw 7.7 and a focal depth of 8.6 km, occurred at 04:17 local time with its epicenter in Pazarcık (Kahramanmaraş). The second, registering Mw 7.6 at a focal depth of 7 km, followed later that day at 13:24 with its epicenter in Elbistan (Kahramanmaraş). Subsequently, a triggered earthquake of Mw 6.4 occurred on 20 February 2023, centered in Yayladağı (Hatay). Collectively, these earthquakes caused extensive destruction across 11 provinces, resulting in 50,783 deaths and 115,353 injuries [2], and injured individuals rescued from the rubble received immediate first aid from emergency teams before being transferred to hospitals in less affected provinces, such as Diyarbakır, for further management and treatment.
Crush syndrome is the second leading cause of earthquake-related mortality. The underlying pathophysiology involves ischemia-reperfusion injury, characterized by cellular influx of sodium and water, and efflux of metabolites and toxins such as myoglobin and potassium [3]. Earthquake-related trauma often presents with unique injury patterns and may lead to complications, including hemorrhagic shock, crush syndrome, compartment syndrome, and other serious conditions [3, 4]. Systemic complications such as acute kidney injury (AKI), rhabdomyolysis, sepsis, multiple organ failure (MOF), disseminated intravascular coagulation (DIC), venous thromboembolic events (VTE), acute respiratory distress syndrome (ARDS), and electrolyte imbalances are commonly observed in crush syndrome [5].
Although direct trauma remains the most common cause of death following earthquakes [3], musculoskeletal injuries are widespread, with approximately 70% involving the lower extremities [6]. To evaluate trauma severity, several scoring systems are utilized. The Revised Trauma Score (RTS) is a physiological index incorporating respiratory rate, pulse, and Glasgow Coma Scale (GCS) scores. The Injury Severity Score (ISS) is an anatomical scoring system that assesses injuries based on their distribution across body regions. Furthermore, the Trauma and Injury Severity Score (TRISS) integrates age, ISS, and RTS components to provide a comprehensive assessment of injury severity [7]. While the distribution of trauma scores according to age, treatment modalities, and injury sites in earthquake-related injuries has been previously examined, limited research has investigated their association with mortality [8, 9, 10, 11].
We designed this study to assess complications arising in earthquake-injured patients and determine the prognostic value of trauma scores and laboratory findings in predicting mortality and the development of crush syndrome.
2. Materials and methods
2.1 Study design and setting
Dicle University Hospital is located in the southeastern Anatolia region of Türkiye, in Diyarbakır province. The hospital has a total bed capacity of 1226, including 364 intensive care unit (ICU) beds. Diyarbakır was among the 11 provinces affected by the earthquake. During the disaster, the hospital provided medical services to patients from both within its province and from neighboring earthquake-affected regions. The study was conducted following the principles of the Declaration of Helsinki. Ethical approval was obtained from the local ethics committee before the start of the study.
2.2 Diagnostic criteria and definitions
Diagnostic criteria for clinical conditions were defined as follows:
AKI was diagnosed based on the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. AKI was defined as an increase in serum creatinine of more than 0.3 mg/dL within 48 hours, a rise to more than 1.5 times the baseline value within seven days, or a urine output of less than 0.5 mL/kg/h for six hours [12].
Crush syndrome was defined as a serum creatinine phosphokinase (CPK) level of ≥1000 IU/L accompanied by systemic complications such as AKI, ARDS, electrolyte disturbances, DIC, hypovolemic shock, arrhythmias, venous thromboembolism (VTE), hemorrhage, or sepsis, resulting from crush injury [5, 13].
Compartment syndrome was diagnosed when intramuscular pressure exceeded 30 mmHg and compatible clinical findings were present [14].
MOF was defined as progressive dysfunction in two or more organ systems that was potentially reversible with treatment [15].
Sepsis was defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, identified clinically by a Sequential Organ Failure Assessment (SOFA) score of ≥2 [16].
2.3 Patient selection and inclusion criteria
This retrospective study included trauma patients affected by the Kahramanmaraş Pazarcık earthquake on 06 February 2023, who presented to the emergency department of Dicle University Hospital. Patients were included consecutively. A total of 192 earthquake-related trauma cases were initially identified. Exclusion criteria included incomplete data (38 cases), death upon admission (7 cases), leaving the clinic without permission (13 cases), referral to an external center (1 case), and injuries from falls at ground level (15 cases). After applying these criteria, 118 patients were included in the analysis. The causes of trauma in these patients included entrapment under debris, traffic accidents, and falls from a height of 3 meters or more (Fig. 1).
Fig. 1.Flow chart of the study. ED: Emergency department.
2.4 Emergency department management and initial treatment protocol
Most patients arrived at the emergency department via ambulance, while a smaller proportion presented using private vehicles. Upon arrival, all patients were assessed and treated according to Advanced Trauma Life Support (ATLS) protocols [17]. Electrocardiography, arterial blood gas analysis, and serum electrolyte evaluations were performed immediately. Intravenous (IV) access was established, and urinary catheterization was carried out. Fluid resuscitation was initiated with isotonic saline at a rate of 1 L/h.
Tetanus prophylaxis was administered to all patients, and IV antibiotic prophylaxis was provided for those with open or contaminated wounds. Specialist consultations were obtained as required. Hemodialysis catheters were inserted in patients with uremia, severe hyperkalemia, or severe metabolic acidosis, and hemodialysis was initiated accordingly. Blood transfusions were provided to patients in hemorrhagic shock, while those requiring amputation, surgical management of compartment syndrome, or other urgent surgical interventions were transferred to the emergency operating room. The remaining patients were managed in the ICU with continued fluid replacement to maintain a urine output of at least 1 mL/kg/h.
2.5 Data collection and recorded variables
The following variables were recorded for each trauma case: date and location of presentation, sex, age, vital signs (temperature, heart rate, respiratory rate, and blood pressure), and affected body regions. Trauma severity was assessed using the GCS, RTS, ISS, and TRISS. Surgical interventions, including fasciotomy and amputation, were documented, as were complications such as hemorrhagic shock, crush syndrome, compartment syndrome, AKI, rhabdomyolysis, sepsis, ICU admission, and MOF. Additionally, the length of hospital stay, and survival outcomes (survival or death) were recorded. Laboratory parameters analyzed included whole blood and biochemical markers: urea, creatinine, alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), creatine kinase (CK), albumin, sodium, potassium, total calcium, C-reactive protein (CRP), pH, lactate, base deficit, hemoglobin, white blood cell (WBC) count, and platelet count.
2.6 Study outcomes
The primary objective of this study was to evaluate the predictive ability of trauma scores and biochemical markers for mortality in trauma patients. The secondary objective was to determine the predictive value of these variables for the development of crush syndrome.
2.7 Statistical analysis
Continuous variables were analyzed for normality. If normally distributed, they were presented as mean ± standard deviation (SD) and compared using Student’s t-test. Non-normally distributed continuous variables were expressed as median and interquartile range (IQR, Q1–Q3) and compared using the Mann-Whitney U-test. Categorical variables were reported as frequencies and percentages, and comparisons were performed using the chi-square (χ2) test. The diagnostic performance of trauma scores and laboratory markers in predicting mortality and crush syndrome was assessed using receiver operating characteristic (ROC) curve analysis and logistic regression analysis. The discriminative ability of each parameter was expressed as the area under the curve (AUC). The optimal cut-off value, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and odds ratio (OR) were calculated. A p-value of < 0.05 was considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (Armonk, NY, USA).
3. Results
3.1 Patient demographics and baseline characteristics
The mean age of the patients was 35.73 ± 18.28 years, and 66 (55.9%) were female. The basic demographic features of the study population are summarized in Table 1.
| Variables | N = 118 (%) | |
| Age, yr, mean ± SD | 35.73 ± 18.28 | |
| Sex, female | 66 (55.90) | |
| Distribution of patients by city | ||
| City in the hospital | 39 (33.10) | |
| Surrounding cities | 79 (66.90) | |
| Time to hospital admission after an earthquake | ||
| First day | 28 (23.70) | |
| Second day | 22 (18.60) | |
| Third day | 15 (12.70) | |
| Fourth day | 14 (11.90) | |
| Fifth day | 11 (9.30) | |
| Sixth day | 10 (8.50) | |
| Other days | 18 (15.30) | |
| Trauma causes | ||
| Fall from a height | 5 (4.20) | |
| Traffic accident | 3 (2.50) | |
| Stay under the debris | 113 (95.80) | |
| Injured body region | ||
| Head | 91 (77.10) | |
| Chest | 95 (80.50) | |
| Abdominal | 92 (78.00) | |
| Pelvic | 91 (77.10) | |
| Limb | 114 (96.60) | |
| Complications | ||
| Hemorrhagic shock | 2 (1.70) | |
| Crush syndrome | 70 (59.30) | |
| Compartment syndrome | 15 (12.70) | |
| Acute kidney injury | 26 (22.00) | |
| Rhabdomyolysis | 12 (10.20) | |
| Sepsis | 10 (8.50) | |
| Multiple organ failure | 8 (6.80) | |
| Admission to intensive care | 26 (22.00) | |
| Surgical interventions | ||
| Fasciotomy | 14 (11.90) | |
| Amputation | 12 (10.20) | |
SD: standard deviation. |
3.2 Association of clinical and laboratory parameters with mortality
Of the 118 patients analyzed, 110 survived and 8 died. Significant differences were observed between survivors and non-survivors with respect to trauma scores and several clinical and laboratory parameters. Specifically, AKI, rhabdomyolysis, sepsis, multiple organ failure (MOF), ICU admission, as well as levels of AST, LDH, potassium, and pH, were significantly associated with mortality (p < 0.001). Detailed comparisons of the clinics are given in Table 2.
| Variables | Survival (n = 110) | Mortality (n = 8) | p-value | |
| Age, (yr) | 35.83 ± 18.68 | 34.50 ± 12.30 | 0.844 | |
| Age ≥65, (yr) | 8.00 (7.30) | 0.00 (0.00) | ||
| Sex, female | 63.00 (57.30) | 3.00 (37.50) | 0.299 | |
| Time to hospital admission after an earthquake, (d) | 3.00 (2.00–4.75) | 3.00 (2.00–5.00) | 0.630 | |
| Vital signs | ||||
| Body temperature (°C) | 36.45 (36.20–36.70) | 36.50 (35.50–36.70) | 0.560 | |
| Heart rate, (beats/min) | 89.50 (81–104.25) | 110.00 (100.00–120.00) | 0.013 | |
| Respiration rate, (per min) | 14.00 (13.00–16.00) | 15.00 (14.00–17.00) | 0.225 | |
| SBP, (mmHg) | 123.75 ± 15.77 | 110.63 ± 31.81 | 0.039 | |
| DBP, (mmHg) | 75.47 ± 11.30 | 65.13 ± 16.36 | 0.017 | |
| Injured body region | ||||
| Head | 83.00 (75.50) | 8.00 (100.00) | 0.195 | |
| Chest | 87.00 (79.10) | 8.00 (100.00) | 0.352 | |
| Abdominal | 84.00 (76.40) | 8.00 (100.00) | 0.197 | |
| Pelvic | 83.00 (75.50) | 8.00 (100.00) | 0.195 | |
| Limb | 106.00 (96.40) | 8.00 (100.00) | 1.000 | |
| Complications | ||||
| Hemorrhagic shock | 1.00 (0.90) | 1.00 (12.50) | 0.132 | |
| Crush syndrome | 62.00 (56.40) | 8.00 (100.00) | 0.020 | |
| Compartment syndrome | 12.00 (10.90) | 3.00 (37.50) | 0.063 | |
| Acute kidney injury | 19.00 (17.30) | 7.00 (87.50) | <0.001 | |
| Rhabdomyolysis | 6.00 (5.50) | 6.00 (75.00) | <0.001 | |
| Sepsis | 2.00 (1.80) | 8.00 (100.00) | <0.001 | |
| Multiple organ failure | 1.00 (0.90) | 7.00 (87.50) | <0.001 | |
| Admission to ICU | 19.00 (17.30) | 7.00 (87.50) | <0.001 | |
| Surgical interventions | ||||
| Fasciotomy | 11.00 (10.00) | 3.00 (37.50) | 0.053 | |
| Amputation | 10.00 (9.10) | 2.00 (25.00) | 0.188 | |
| Trauma scores | ||||
| GCS | 15.00 (15.00–15.00) | 13.00 (6.00–15.00) | <0.001 | |
| RTS | 7.84 (7.84–7.84) | 7.84 (5.43–7.84) | <0.001 | |
| ISS | 26.00 (17.25–34.00) | 54.00 (48.00–66.00) | <0.001 | |
| TRISS | 97.40 (88.60–98.70) | 59.40 (36.30–84.70) | <0.001 | |
| Laboratory values | ||||
| Urea, (mg/dL) | 31.50 (22.65–59.47) | 89.40 (59.90–141.70) | 0.002 | |
| Creatinine, (mg/dL) | 0.68 (0.51–1.17) | 3.46 (0.82–5.08) | 0.003 | |
| ALT, (U/L) | 53.20 (26.77–164.07) | 334.10 (76.90–417.10) | 0.026 | |
| AST, (U/L) | 80.05 (33.92–397.10) | 546.30 (381.20–1333.60) | 0.001 | |
| LDH, (U/L) | 549.95 (289.42–1010.92) | 1727.90 (806.50–6801.80) | 0.001 | |
| Creatine Kinase, (U/L) | 2033.85 (410.20–15,325.75) | 40,515.20 (1201.10–181,948.40) | 0.158 | |
| Albumin, (g/L) | 36.00 ± 6.32 | 31.11 ± 7.36 | 0.039 | |
| Sodium, (mmol/L) | 135.45 (132.52–137.20) | 129.70 (125.70–141.10) | 0.419 | |
| Potassium, (mmol/L) | 4.26 (3.81–4.87) | 5.50 (5.07–6.35) | <0.001 | |
| Calcium total, (mg/dL) | 8.78 (7.88–9.26) | 7.51 (6.88–8.34) | 0.003 | |
| CRP, (mg/dL) | 59.10 (9.48–111.71) | 137.61 (119.11–163.26) | 0.024 | |
| pH | 7.38 (7.33–7.41) | 7.14 (7.10–7.25) | <0.001 | |
| Lactate, (mmol/L) | 1.90 (1.32–4.37) | 2.00 (1.10–11.30) | 0.179 | |
| Base deficit, (mEq/L) | −2.00 ((−3.00)–0.00) | −2.00 ((−5.00)–(−1.00)) | 0.558 | |
| Haemoglobin, (g/dL) | 12.80 ± 2.70 | 13.31 ± 4.88 | 0.628 | |
| White blood cell, (103/µL) | 13.68 (9.28–18.38) | 17.73 (6.57–34.08) | 0.140 | |
| Platelets, (103/µL) | 269.50 (195.75–335.50) | 153.00 (58.00–276.00) | 0.054 | |
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; CRP: c-reactive protein; DBP: diastolic blood pressure; GCS: glasgow coma scale; ICU: intensive care unit; ISS: injury severity score; LDH: lactate dehydrogenase; RTS: revised trauma score; SBP: systolic blood pressure; TRISS: trauma score and injury severity score. |
3.3 Association of clinical and laboratory parameters with complications
Complications developed in 38 patients (32.2%), and a statistically significant difference was observed in GCS scores between patients with and without complications; however, this is thought to be due to small differences in minimum and maximum values. However, ISS and TRISS, along with laboratory parameters including ALT, AST, LDH, CK, CRP, albumin, and total calcium, differed significantly between patients with and without complications (p < 0.001) (Table 3).
| Variables | Without complications (n = 38) | With complications (n = 80) | p-value | |
| Age, (yr) | 38.26 ± 18.81 | 34.53 ± 18.03 | 0.303 | |
| Age ≥65, (yr) | 4.00 (10.50) | 4.00 (5.00) | 0.269 | |
| Sex, female | 19.00 (50.00) | 47.00 (58.80) | 0.486 | |
| Time to hospital admission after an earthquake, (d) | 2.00 (1.00–8.00) | 3.00 (2.00–4.00) | 0.861 | |
| Vital signs | ||||
| Body temperature (°C) | 36.40 (36.20–36.50) | 36.50 (36.20–36.70) | 0.373 | |
| Heart rate, (beats/min) | 93.00 (89.00–105.00) | 90.00 (81.00–109.75) | 0.301 | |
| Respiration Rate, (per min) | 14.00 (14.00–16.00) | 14.00 (13.00–16.00) | 0.433 | |
| SBP, (mmHg) | 122.92 ± 14.74 | 122.84 ± 18.64 | 0.981 | |
| DBP, (mmHg) | 73.58 ± 11.21 | 75.34 ± 12.25 | 0.456 | |
| Injured body region | ||||
| Head | 23.00 (60.50) | 68.00 (85.00) | 0.006 | |
| Chest | 23.00 (60.50) | 72.00 (90.00) | <0.001 | |
| Abdominal | 20.00 (52.60) | 72.00 (90.00) | <0.001 | |
| Pelvic | 20.00 (52.60) | 71.00 (88.80) | <0.001 | |
| Limb | 34.00 (89.50) | 80.00 (100.00) | 0.010 | |
| Trauma scores | ||||
| GCS | 15.00 (15.00–15.00) | 15.00 (15.00–15.00) | 0.017 | |
| RTS | 7.84 (7.84–7.84) | 7.84 (7.84–7.84) | 0.085 | |
| ISS | 14.00 (9.00–19.00) | 32.50 (26.00–49.50) | <0.001 | |
| TRISS | 98.90 (97.40–99.40) | 95.50 (84.70–97.60) | <0.001 | |
| Laboratory values | ||||
| Urea, (mg/dL) | 28.50 (22.10–31.00) | 45.95 (25.25–126.60) | 0.003 | |
| Creatinine, (mg/dL) | 0.64 (0.55–0.78) | 0.81 (0.51–3.31) | 0.239 | |
| ALT, (U/L) | 27.00 (21.90–36.90) | 121.00 (37.00–271.67) | <0.001 | |
| AST, (U/L) | 23.90 (20.80–66.10) | 282.45 (62.32–564.37) | <0.001 | |
| LDH, (U/L) | 283.80 (193.20–350.00) | 832.70 (417.00–1701.47) | <0.001 | |
| Creatine Kinase, (U/L) | 186.80 (83.70–2104.00) | 5448.00 (1268.55–40,339.30) | <0.001 | |
| Albumin, (g/L) | 38.96 ± 4.66 | 34.10 ± 6.65 | <0.001 | |
| Sodium, (mmol/L) | 136.10 (134.40–136.90) | 134.10 (130.52–138.25) | 0.423 | |
| Potassium, (mmol/L) | 4.12 (3.49–4.61) | 4.33 (3.90–5.19) | 0.095 | |
| Calcium total, (mg/dL) | 9.22 (8.79–9.60) | 8.37 (7.59–8.91) | <0.001 | |
| CRP, (mg/dL) | 10.67 (2.53–33.38) | 88.22 (25.89–135.84) | <0.001 | |
| pH | 7.38 (7.33–7.39) | 7.37 (7.28–7.42) | 0.364 | |
| Lactate, (mmol/L) | 1.80 (1.30–3.30) | 1.95 (1.40–4.77) | 0.435 | |
| Base deficit, (mEq/L) | 0.00 (−2.00–(1.00)) | −2.00 (−5.00–(−1.00)) | 0.001 | |
| Haemoglobin, (g/dL) | 13.11 ± 2.27 | 12.70 ± 3.11 | 0.470 | |
| White blood cell, (103/µL) | 11.96 (9.04–17.00) | 14.19 (9.28–19.24) | 0.038 | |
| Platelets, (103/µL) | 309.00 (243.00–380.00) | 261.00 (177.25–296.25) | 0.098 | |
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; CRP: c-reactive protein; DBP: diastolic blood pressure; GCS: glasgow coma scale; ICU: intensive care unit; ISS: injury severity score; LDH: lactate dehydrogenase; RTS: revised trauma score; SBP: systolic blood pressure; TRISS: trauma score and injury severity score. |
3.4 Clinical and laboratory predictors of crush syndrome
Crush syndrome was diagnosed in 70 patients (59.3%), and a statistically significant difference was observed in GCS scores between patients with and without crush syndrome; however, this is thought to be due to small differences in minimum and maximum values. In contrast, ISS, TRISS, and laboratory markers, including ALT, AST, LDH, CK and CRP, were significantly elevated in patients with crush syndrome (p < 0.001). A comparative analysis of median (Q1–Q3) values in patients with and without crush syndrome revealed the following: GCS, 15 (15–15) vs. 15 (15–15); ISS, 11 (5.25–21.75) vs. 26 (19–43.50); TRISS, 99.05 (97.40–99.57) vs. 97.20 (84.70–98.70); ALT, 27.45 (20.37–44.70) vs. 80.00 (37.22–209.90); AST, 36.45 (20.82–57.40) vs. 197.35 (59.52–443.00); LDH, 283.40 (216.75–387.52) vs. 702.80 (415.15–1208.92); CK, 270.70 (121.05–1380.22) vs. 6372.00 (1249.00–19,019.12); and CRP, 10.82 (2.64–45.36) vs. 84.45 (38.05–136.29) (Table 4).
| Variables | Without crush syndrome (n = 48) | With crush syndrome (n = 70) | p-value | |
| Age, (yr) | 37.64 ± 18.59 | 34.42 ± 18.10 | 0.350 | |
| Age ≥65, (yr) | 5.00 (10.40) | 3.00 (4.30) | 0.268 | |
| Sex, female | 24.00 (50.00) | 42.00 (60.00) | 0.376 | |
| Time to hospital admission after an earthquake, (d) | 2.00 (1.00–7.75) | 3.00 (2.00–5.00) | 0.348 | |
| Vital signs | ||||
| Body temperature (°C) | 36.50 (36.20–36.70) | 36.50 (36.20–36.70) | 0.349 | |
| Heart rate, (beats/min) | 89.00 (78.00–98.75) | 89.50 (81.00–104.25) | 0.514 | |
| Respiration Rate, (per min) | 15.00 (13.25–16.00) | 14.00 (13.00–15.00) | 0.201 | |
| SBP, (mmHg) | 124.94 ± 13.90 | 121.44 ± 19.43 | 0.286 | |
| DBP, (mmHg) | 75.10 ± 11.69 | 74.54 ± 12.13 | 0.803 | |
| İnjured body region | ||||
| Head | 29.00 (60.40) | 62.00 (88.60) | 0.001 | |
| Chest | 30.00 (62.50) | 65.00 (92.90) | <0.001 | |
| Abdominal | 28.00 (58.30) | 64.00 (91.40) | <0.001 | |
| Pelvic | 27.00 (56.30) | 64.00 (91.40) | <0.001 | |
| Limb | 44.00 (91.70) | 70.00 (100.00) | 0.025 | |
| Complications | ||||
| Hemorrhagic shock | 0.00 (0.00) | 2.00 (2.90) | ||
| Compartment syndrome | 1.00 (2.10) | 14.00 (20.00) | 0.010 | |
| Acute kidney injury | 2.00 (4.20) | 24.00 (34.30) | <0.001 | |
| Rhabdomyolysis | 0.00 (0.00) | 12.00 (17.10) | ||
| Sepsis | 0.00 (0.00) | 10.00 (14.30) | ||
| Multiple organ failure | 0.00 (0.00) | 8.00 (11.40) | ||
| Admission to ICU | 8.00 (16.70) | 18.00 (25.70) | 0.348 | |
| Surgical interventions | ||||
| Fasciotomy | 1.00 (2.10) | 13.00 (18.60) | 0.015 | |
| Amputation | 2.00 (4.20) | 10.00 (14.30) | 0.119 | |
| Trauma scores | ||||
| GCS | 15.00 (15.00–15.00) | 15.00 (15.00–15.00) | 0.027 | |
| RTS | 7.84 (7.84–7.84) | 7.84 (7.84–7.84) | 0.211 | |
| ISS | 11.00 (5.25–21.75) | 26.00 (19.00–43.50) | <0.001 | |
| TRISS | 99.05 (97.40–99.57) | 97.20 (84.70–98.70) | <0.001 | |
| Laboratory values | ||||
| Urea, (mg/dL) | 28.30 (18.97–35.37) | 35.35 (23.25–83.70) | 0.002 | |
| Creatinine, (mg/dL) | 0.64 (0.53–0.78) | 0.78 (0.49–1.43) | 0.054 | |
| ALT, (U/L) | 27.45 (20.37–44.70) | 80.00 (37.22–209.90) | <0.001 | |
| AST, (U/L) | 36.45 (20.82–57.40) | 197.35 (59.52–443.00) | <0.001 | |
| LDH, (U/L) | 283.40 (216.75–387.52) | 702.80 (415.15–1208.92) | <0.001 | |
| Creatine Kinase, (U/L) | 270.70 (121.05–1380.22) | 6372.00 (1249.00–19,019.12) | <0.001 | |
| Albumin, (g/L) | 37.43 ± 6.30 | 34.46 ± 6.36 | 0.014 | |
| Sodium, (mmol/L) | 135.50 (133.22–136.55) | 134.30 (131.17–137.87) | 0.387 | |
| Potassium, (mmol/L) | 4.06 (3.64–4.52) | 4.24 (3.82–4.95) | 0.050 | |
| Calcium total, (mg/dL) | 9.02 (8.59–9.38) | 8.71 (7.68–9.22) | 0.011 | |
| CRP, (mg/dL) | 10.82 (2.64–45.36) | 84.45 (38.05–136.29) | <0.001 | |
| pH | 7.38 (7.34–7.41) | 7.40 (7.34–7.43) | 0.104 | |
| Lactate, (mmol/L) | 1.75 (1.30–3.17) | 1.70 (1.37–3.32) | 0.941 | |
| Base deficit, (mEq/L) | −1.00 (−2.00–0.00) | −2.00 (−5.00–(−1.00)) | 0.016 | |
| Haemoglobin, (g/dL) | 12.80 ± 2.61 | 12.85 ± 3.05 | 0.924 | |
| White blood cell, (103/µL) | 10.93 (8.14–15.30) | 13.96 (9.31–18.51) | 0.029 | |
| Platelets, (103/µL) | 256.50 (220.50–336.00) | 261.00 (190.50–305.00) | 0.475 | |
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; CRP: c-reactive protein; DBP: diastolic blood pressure; GCS: glasgow coma scale; ICU: intensive care unit; ISS: injury severity score; LDH: lactate dehydrogenase; RTS: revised trauma score; SBP: systolic blood pressure; TRISS: trauma score and injury severity score. |
3.5 Predictive accuracy of trauma scores and laboratory markers for mortality
The predictive ability of laboratory and trauma scores for mortality was evaluated using ROC curve analysis. The AUC, 95% confidence intervals (CI), and p-values for each variable were as follows: albumin (AUC: 0.699, CI: 0.488–0.911, p = 0.060), ALT (AUC: 0.736, CI: 0.507–0.965, p = 0.026), AST (AUC: 0.838, CI: 0.721–0.955, p = 0.001), total calcium (AUC: 0.811, CI: 0.673–0.948, p = 0.003), CK (AUC: 0.650, CI: 0.404–0.896, p = 0.158), CRP (AUC: 0.740, CI: 0.561–0.918, p = 0.024), potassium (AUC: 0.894, CI: 0.809–0.979, p < 0.001), LDH (AUC: 0.864, CI: 0.777–0.950, p = 0.001), urea (AUC: 0.830, CI: 0.726–0.933, p = 0.002), GCS (AUC: 0.790, CI: 0.581–0.998, p = 0.006), ISS (AUC: 0.935, CI: 0.853–1.000, p < 0.001), RTS (AUC: 0.677, CI: 0.444–0.911, p = 0.095), and TRISS (AUC: 0.913, CI: 0.811–1.000, p < 0.001) (Table 5, Fig. 2).
| Predictor | Optimal Cut Point | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | AUC |
| Albumin | 31.80 | 75.00 | 74.55 | 17.65 | 97.62 | 0.699 |
| ALT | 259.60 | 62.50 | 90.00 | 31.25 | 97.06 | 0.736 |
| AST | 192.50 | 87.50 | 70.91 | 17.95 | 98.73 | 0.838 |
| Ca | 8.37 | 87.50 | 70.00 | 17.50 | 98.72 | 0.811 |
| CK | 40,515.20 | 50.00 | 91.82 | 30.77 | 96.19 | 0.650 |
| CRP | 119.11 | 75.00 | 79.09 | 20.69 | 97.75 | 0.740 |
| K | 4.81 | 87.50 | 83.64 | 28.00 | 98.92 | 0.894 |
| LDH | 646.40 | 100.00 | 68.18 | 18.60 | 100.00 | 0.864 |
| Urea | 59.90 | 75.00 | 82.73 | 24.00 | 97.85 | 0.830 |
| GCS | 5.00 | 62.50 | 94.55 | 45.45 | 97.20 | 0.790 |
| ISS | 48.00 | 87.50 | 90.00 | 38.89 | 99.00 | 0.935 |
| RTS | 5.97 | 37.50 | 99.09 | 75.00 | 95.61 | 0.677 |
| TRISS | 84.70 | 87.50 | 86.36 | 31.82 | 98.96 | 0.913 |
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; AUC: area under the curve; Ca: calcium total; CK: creatine kinase; CRP: c-reactive protein; GCS: glasgow coma scale; ISS: injury severity score; K: potassium; LDH: lactate dehydrogenase; NPV: negative predictive value; PPV: positive predictive value; RTS: revised trauma score; TRISS: trauma score and injury severity score. |
Fig. 2.ROC curve of laboratory and trauma scores by mortality. (a) ROC curve of ALT, AST, CK, CRP, ISS, K, LDH, Urea. (b) ROC curve of Albumin, Ca, GCS, RTS, and TRISS. ROC: receiver operating characteristic; ALT: alanine aminotransferase; AST: aspartate aminotransferase; CK: creatine kinase; CRP: c-reactive protein; ISS: injury severity score; K: potassium; LDH: lactate dehydrogenase; Ca: calcium total; GCS: glasgow coma scale; RTS: revised trauma score; TRISS: trauma score and injury severity score.
Despite these findings, multivariate logistic regression analysis did not identify any of the studied variables as independent predictors of mortality (Table 6).
| Variables | Univariate | Multivariate | ||
| Odds ratio (95% CI) | p-value | Odds ratio (95% CI) | p-value | |
| Albumin | 1.105 (1.000–1.221) | 0.049 | ||
| ALT | 0.995 (0.992–0.998) | 0.003 | ||
| AST | 0.998 (0.997–0.999) | 0.003 | ||
| Ca | 1.821 (1.065–3.115) | 0.029 | ||
| CK | 1.000 (1.000–1.000) | 0.014 | ||
| CRP | 0.990 (0.980–1.000) | 0.041 | ||
| K | 0.318 (0.158–0.640) | 0.001 | ||
| LDH | 1.000 (0.999–1.000) | 0.006 | ||
| Urea | 0.992 (0.986–0.998) | 0.015 | ||
| GCS | 2.627 (1.316–5.245) | 0.006 | ||
| ISS | 0.871 (0.806–0.941) | <0.001 | ||
| RTS | 5.342 (1.953–14.614) | 0.001 | ||
| TRISS | 1.078 (1.039–1.118) | <0.001 | ||
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; Ca: calcium total; CK: creatine kinase; CI: confidence interval; CRP: c-reactive protein; GCS: glasgow coma scale; ISS: injury severity score; K: potassium; LDH: lactate dehydrogenase; RTS: revised trauma score; TRISS: trauma score and injury severity score. |
3.6 Predictive accuracy of trauma scores and laboratory markers for crush syndrome
The ability of laboratory markers and trauma scores to predict crush syndrome was assessed using ROC curve analysis. The AUC, 95% CI, and p-values for each parameter were as follows: albumin (AUC: 0.648, CI: 0.548–0.748, p = 0.007), ALT (AUC: 0.792, CI: 0.711–0.872, p < 0.001), AST (AUC: 0.850, CI: 0.779–0.921, p < 0.001), total calcium (AUC: 0.639, CI: 0.540–0.738, p = 0.011), CK (AUC: 0.835, CI: 0.759–0.911, p < 0.001), CRP (AUC: 0.812, CI: 0.735–0.889, p < 0.001), LDH (AUC: 0.822, CI: 0.744–0.901, p < 0.001), urea (AUC: 0.667, CI: 0.570–0.763, p = 0.002), WBC (AUC: 0.618, CI: 0.515–0.722, p = 0.029), GCS (AUC: 0.561, CI: 0.457–0.664, p = 0.265), ISS (AUC: 0.822, CI: 0.744–0.900, p < 0.001), and TRISS (AUC: 0.773, CI: 0.687–0.860, p < 0.001) (Table 7, Fig. 3).
| Predictor | Optimal Cut Point | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | AUC |
| Albumin | 31.90 | 40.00 | 85.42 | 80.00 | 49.40 | 0.648 |
| ALT | 76.90 | 54.29 | 95.83 | 95.00 | 58.97 | 0.792 |
| AST | 48.70 | 88.57 | 68.75 | 80.52 | 80.49 | 0.850 |
| Ca | 8.98 | 71.43 | 54.17 | 69.44 | 56.52 | 0.639 |
| CK | 395.60 | 97.14 | 60.42 | 78.16 | 93.55 | 0.835 |
| CRP | 70.96 | 65.71 | 87.50 | 88.46 | 63.64 | 0.812 |
| LDH | 411.40 | 77.14 | 77.08 | 83.08 | 69.81 | 0.822 |
| Urea | 45.90 | 45.71 | 89.58 | 86.49 | 53.09 | 0.667 |
| WBC | 12.01 | 62.86 | 62.50 | 70.97 | 53.57 | 0.618 |
| GCS | 5.00 | 14.29 | 97.92 | 90.91 | 43.93 | 0.561 |
| ISS | 18.00 | 87.14 | 64.58 | 78.21 | 77.50 | 0.822 |
| TRISS | 98.80 | 88.57 | 54.17 | 73.81 | 76.47 | 0.773 |
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; AUC: area under the curve; Ca: calcium total; CK: creatine kinase; CRP: c-reactive protein; GCS: glasgow coma scale; ISS: injury severity score; LDH: lactate dehydrogenase; NPV: negative predictive value; PPV: positive predictive value; TRISS: trauma score and injury severity score; WBC: white blood cell. |
Fig. 3.ROC curve of laboratory and trauma scores by crush syndrome. (a) ROC curve of ALT, AST, CK, CRP, ISS, K, LDH, Urea, WBC. (b) ROC curve of Albumin, Ca, GCS, TRISS. ROC: receiver operating characteristic; ALT: alanine aminotransferase; AST: aspartate aminotransferase; CK: creatine kinase; CRP: c-reactive protein; ISS: injury severity score; LDH: lactate dehydrogenase; WBC: white blood cell; Ca: calcium total; GCS: glasgow coma scale; TRISS: trauma score and injury severity score.
Multivariate logistic regression analysis demonstrated that CRP (OR: 1.018, 95% CI: 1.005–1.030, p = 0.006), LDH (OR: 0.997, 95% CI: 0.994–1.000, p = 0.030), and ISS (OR: 1.105, 95% CI: 1.019–1.197, p = 0.015) were independent predictors of crush syndrome (Table 8).
| Variables | Univariate | Multivariate | ||
| Odds ratio (95% CI) | p-value | Odds ratio (95% CI) | p-value | |
| Albumin | 0.924 (0.867–0.986) | 0.016 | ||
| ALT | 1.017 (1.008–1.027) | 0.001 | ||
| AST | 1.007 (1.003–1.011) | <0.001 | ||
| Ca | 0.639 (0.416–0.982) | 0.041 | ||
| CK | 1.000 (1.000–1.000) | 0.010 | ||
| CRP | 1.023 (1.014–1.033) | <0.001 | 1.018 (1.005–1.030) | 0.006 |
| LDH | 1.002 (1.001–1.003) | 0.001 | 0.997 (0.994–1.000) | 0.030 |
| Urea | 1.019 (1.005–1.034) | 0.009 | ||
| WBC | 1.046 (0.993–1.102) | 0.088 | ||
| GCS | 0.538 (0.239–1.215) | 0.136 | ||
| ISS | 1.108 (1.062–1.156) | <0.001 | 1.105 (1.019–1.197) | 0.015 |
| TRISS | 0.908 (0.847–0.974) | 0.007 | ||
Abbreviations: ALT: alanine aminotransferase; AST: aspartate aminotransferase; Ca: calcium total; CK: creatine kinase; CI: confidence interval; CRP: c-reactive protein; GCS: glasgow coma scale; ISS: injury severity score; LDH: lactate dehydrogenase; TRISS: trauma score and injury severity score; WBC: white blood cell. |
4. Discussion
Earthquakes are catastrophic natural disasters that often lead to severe injuries and complications, including crush syndrome, which plays a pivotal role in determining clinical outcomes and prognosis [1, 2, 3]. In light of these, our present study investigated the predictive value of trauma scores and laboratory findings for mortality and crush syndrome among patients affected by the Kahramanmaraş Pazarcık earthquake. For assessing mortality, the ISS and TRISS demonstrated the highest sensitivity among trauma scores, while potassium, LDH, and AST were the most sensitive laboratory markers. However, despite their strong predictive performance, multivariate analysis did not identify these variables as independent risk factors for mortality. In contrast, both ISS and TRISS were also the most sensitive trauma scores for predicting crush syndrome, while AST, CK, and LDH emerged as the most sensitive laboratory markers. Notably, multivariate analysis identified CRP, LDH, and ISS as independent risk factors associated with the development of crush syndrome.
Earthquake-related injuries frequently involve multiple organ systems, contributing to considerable morbidity and mortality [18], with extremity injuries being among the most common, followed by head trauma [18, 19]. In the study by Bicakcioglu et al. [20], only two cases involved injuries resulting from falls during an earthquake. Consistent with existing literature, extremity trauma was the most commonly observed injury type in the present study. Moreover, most injuries occurred due to entrapment beneath rubbles. Some patients involved in traffic accidents or falls from heights were also subsequently trapped. Notably, patients injured by falls or traffic accidents typically presented to the emergency department in the early hours following the earthquake.
Previous studies have compared ISS and TRISS in predicting mortality among trauma patients, although findings on their relative superiority have varied. Generally, however, TRISS has been reported to outperform ISS [21]. For instance, Sammour et al. [22] reported AUC values of 0.8547 and 0.963 for ISS and TRISS, respectively, in mortality prediction. Earthquake-related trauma has historically resulted in high mortality. In a study involving 1028 earthquake-related trauma patients, the mortality rate was 0.5%, with 38 patients (3.7%) exhibiting ISS scores above 25 [19]. Additionally, a previous investigation comparing ISS, TRISS, and New Injury Severity Score (NISS) in relation to hospitalization duration in earthquake trauma patients demonstrated that only NISS was associated with hospital stay. Mortality was not analyzed due to the low number of deaths (n = 2) [9]. In another study, Gao et al. [10] reported a mean ISS of 32.1 among patients who died following earthquake-related injuries. Similarly, research evaluating the prognostic value of RTS, ISS, and TRISS scores in mortality prediction found all three to be significant. Among the deceased, the median (IQR) values were RTS 7.84 (4–7.84), ISS 25 (17–33), and TRISS 86 (50–97) [11]. In agreement with previous findings, the present study identified ISS and TRISS as the most sensitive trauma scores for predicting mortality among earthquake-injured patients. However, unlike crush syndrome, none of these trauma scores or laboratory markers were independent predictors of mortality in multivariate analysis.
Complications such as crush injury, crush syndrome, compartment syndrome, hemorrhagic shock, AKI, and MOF frequently develop following earthquakes [3, 5, 18]. Additionally, interventions, including fasciotomy and amputation, are more commonly required in earthquake-related trauma [18, 23]. Early mortality is predominantly attributable to hypovolemia and hyperkalemia, whereas late mortality typically results from sepsis and MOF [18]. In the present study, complication rates were notably higher among patients who died. Late deaths were primarily caused by sepsis and MOF, while one early death was due to hemorrhagic shock.
Crush injury leads to cellular destruction and myonecrosis [5]. During revascularization, calcium, sodium, and water enter necrotic muscle tissue, whereas potassium, phosphorus, lactic acid, myoglobin, and intracellular enzymes such as CK, ALT, AST and LDH are released into the bloodstream [5, 18, 24]. Previous studies have explored the relationship between serum enzyme and electrolyte levels and mortality in earthquake-related injuries [25, 26]. Consistent with the literature, the present study demonstrated that potassium, LDH, and AST were the most sensitive laboratory markers associated with mortality. These findings suggest that elevated potassium, LDH, and AST levels during initial emergency department evaluation may serve as important indicators for early ICU admission, even in patients not requiring immediate surgical intervention.
Crush syndrome occurs in approximately 2–15% of patients following major earthquakes [27]. This syndrome is characterized by systemic complications, including AKI, rhabdomyolysis, sepsis, MOF, DIC, VTE, ARDS, and electrolyte imbalances [5]. In the present cohort, patients with crush syndrome demonstrated a higher incidence of complications, consistent with prior observations.
Extremity trauma is the most frequent injury pattern among patients with crush syndrome resulting from earthquake-related injuries [26]. Furthermore, in patients with multiple trauma, an ISS score of ≥25 has been associated with the development of AKI [28]. In this study, extremity injuries were most prevalent, and ISS was the most significant trauma score in predicting crush syndrome. Given that ISS is an anatomical scoring system, it may better reflect the extent of muscle damage, which likely explains its superior performance in predicting crush syndrome compared to other trauma scores.
Previous research has assessed the diagnostic utility of serum enzymes and electrolytes in predicting AKI, crush syndrome, fasciotomy, and amputation [13, 23, 29, 30]. Hu et al. [29] identified CK as the most effective predictor of AKI, while Omrani et al. [30] found CK, LDH, AST, and uric acid to be the most relevant biomarkers. Similarly, Feng et al. [26] reported that CK had the highest diagnostic value for crush injury and crush syndrome (AUC: 0.994; sensitivity: 99.4%; specificity: 100%). Consistent with these findings, the present study identified AST, CK, and LDH as the most sensitive laboratory markers for predicting crush syndrome.
This study has several limitations. First, it was conducted at a single center with a relatively small sample size, which may limit the generalizability of the findings, particularly to hospitals with different triage or trauma care protocols. Future multicenter investigations are necessary to validate and extend these results. Second, the retrospective design may introduce inherent biases. However, conducting prospective studies during natural disasters such as earthquakes poses considerable logistical and ethical challenges. Finally, long-term outcomes, including functional status and psychological sequelae, were not assessed in this study.
5. Conclusions
This cross-sectional study provides a comparative evaluation of trauma scores and laboratory markers in predicting mortality and crush syndrome among patients with earthquake-related injuries. The findings underscore the clinical relevance of these parameters in determining prognosis and guiding management decisions in such settings. Among the trauma scores, ISS and TRISS demonstrated superior performance in predicting both mortality and crush syndrome. Similarly, among serum biomarkers, AST, CK, and LDH were the most sensitive indicators. Furthermore, multivariate analysis identified CRP, LDH, and ISS as independent risk factors for crush syndrome. These results highlight the potential utility of trauma scores and laboratory markers as early triage tools to facilitate timely risk stratification in earthquake-induced trauma. In addition, the findings may provide a foundation for the development of advanced predictive models, including machine learning algorithms, to support real-time prognostic assessment and optimize clinical decision-making in disaster scenarios.
Availability of data and materials
The data generated and analysed during the study can be obtained from the corresponding author.
Author contributions
Mİ—writing–original draft, methodology, investigation, data curation, conceptualization, formal analysis, validation, supervision. TB—methodology, investigation, data curation, conceptualization, formal analysis.
Ethics approval and consent to participate
Permission to study was obtained from The Ethics Committee of Dicle University Faculty of Medicine with registration number 14.06.2023-205. Due to the retrospective nature of the study, informed consent was waived by The Ethics Committee of Dicle University Faculty of Medicine.
Acknowledgment
Not applicable.
Funding
This research received no external funding.
Conflict of interest
The authors declare no conflict of interest.
