Pediatric Wound Infections Following 2023 Kahramanmaras Earthquakes: Case Series

On February 6, 2023, 2 devastating earthquakes struck Kahramanmaras, Turkey. The earthquakes registered 7.7 and 7.6 magnitude on the moment scale. According to official data, 50 500 people lost their lives in the earthquake and more than 100 000 individuals were injured.¹ Crush syndrome, trauma-related extremity injuries, and wound site infections are the main causes of morbidity and mortality in earthquake-affected patients. Crush syndrome is a clinical entity characterized by severe muscle damage, myoglobinuria, and acute renal failure, which is frequently seen in patients with earthquake-related injuries. Infections that develop in crush syndrome are a significant complication that can result in death.²

Traumatic injuries can result in open wounds and closed injuries requiring fasciotomies and amputations. While these surgical procedures are life-saving interventions, they also expose tissues to external contaminants and increase the risk of wound infection. As a consequence of traumatic injuries, bacterial contamination develops in necrotic ischemic tissues, resulting in wound infections that can lead to delayed healing, prolonged hospitalization, and sepsis.³ 

Invasive procedures such as vascular catheters, urinary catheters, and long-term antibiotic use increase the risk of nosocomial infections.⁴ Clinical factors that increase the risk of wound infections are diabetes, immunodeficiency, hypoxia-inducing anemia, cardiac and respiratory system diseases, malignancies, obesity, and malnutrition. Contaminated surgical locations, long surgical procedures, delayed treatment after trauma, necrotic tissue, presence of a foreign body, involvement of a large and deep area, and anatomical proximity to a contaminated area (anal region) also increase the risk of infection.⁵

In the current case series, pediatric patients who were injured in the Kahramanmaras earthquakes and admitted to Marmara University Pendik Training and Research Hospital were evaluated. The aim was to investigate infective complications, causative microorganisms, treatments, and responses to these treatments.

Setting and patients

The authors analyzed the demographic, clinical, and laboratory data of 12 earthquake victims admitted to Marmara University Pendik Training and Research Hospital in Istanbul, Turkey, between February 9 and 24, 2023. 

Data collections and definitions

Wound sample cultures were obtained from patients with wound infections. At least 1 of the following 3 criteria had to be met for the diagnosis of a wound infection: (1) body temperature of 38°C and/or local findings such as pain, tenderness, edema, and erythema; (2) wound dehiscence; and (3) wound with purulent drainage.

Both aerobic and anaerobic bacteriological cultures were obtained from wound secretions and tissue samples collected during surgery. Cultures were incubated on blood agar plates, bacterial isolates were identified, and antibiotic resistance was determined. Complete blood count, C-reactive protein, blood culture, urine culture, and rectal swab samples were analyzed.

Demographic features, clinical data, and laboratory findings on admission were recorded. 

Statistical analysis

Patient data were analyzed with IBM SPSS Statistics 26.0. Demographic and descriptive continuous variables with normal distribution are reported as mean (SD), whereas non-normally distributed data are presented as median values (IQR). Categorical variables are expressed as percentages. 

Ethical compliance

A written informed consent form was obtained from the parents of the patients. The ethics committee approval was obtained from Marmara University Faculty of Medicine Research Ethics Committee (number: 07.2023.926).

Twelve earthquake-affected children were admitted to the authors’ hospital between February 9 and 24, 2023. Three patients were female (25%), and 9 were male (75%). The mean age of patients was 12.6 ± 3.7 years. One in 12 patients had an underlying disease, which was asthma. Prior to rescue, patients were trapped under debris for between 0.5 and 80 hours, with a median of 14 hours. Demographic, clinical, and trauma features of the patients are summarized in Table 1.

Four patients (33.3%) had fractures secondary to trauma. Two patients had lumbar spine fractures (L2, L5), 1 had bilateral femoral shaft and humerus fractures, and the other had clavicula fractures and femur neck and shaft fractures. Two of these 4 patients with fractures underwent open reduction internal fixation surgery; 1 underwent bilateral pedicle screw fixation. An external fixator was implanted in 1 patient with a femur fracture. 

Seven patients (58.3%) underwent a total of 11 fasciotomy operations: 5 on the cruris, 4 on the foot, and 2 on the thigh. In 4 patients, fasciotomies were performed in 2 or more regions. These patients underwent consecutive operations for wound debridement. Two patients had ecchymoses on bilateral extremities, and 1 patient had abrasions. The extremities of 2 patients were necrotic, with 1 patient presenting with necrosis in the left foot and the other bilaterally below the knee. Amputation was performed on these 2 patients, consisting of a single-limb above-knee amputation and bilateral above-knee amputation, respectively. The patient who underwent bilateral above-knee amputation is shown in Figure 1. 

Wound care involved the utilization of different treatment methods. Sterile paraffin tulle dressing coated with chlorhexidine acetate was used in 7 patients (58.3%). Additionally, NPWT was applied to 4 patients (33.3%). NPWT was conducted using hydrophobic polyurethane silver foam. The maximum pressure applied ranged from 100 mm Hg to 125 mm Hg, with intermittent intervals of 5 minutes on and 2 minutes off. The lowest pressure set in closed mode was 50 mm Hg. Applications were performed every 48 to 72 hours. Furthermore, HBOT was applied in 4 patients (33.3%) for 1 to 2 sessions daily and the number of sessions was determined by evaluating the patient’s overall condition and the wound’s status. The number of sessions varied between 6 and 18. 

Three patients (25%) presented with a fever upon admission, while 6 additional patients developed fever (body temperature >38°C) during hospitalization. The median number of febrile days was 1.5 days (IQR: 2.75), and the longest fever lasted 7 days. The patients’ laboratory findings are shown in Table 2. 

The main infectious complications were wound infections and urinary tract infections. Seven (58.3%) injured children experienced wound infections, all of whom underwent a fasciotomy operation. Microbiological cultures were taken from the wound sites of injured children.

The main organisms were Enterococcus spp in 5 (41.6%) cases, Acinetobacter spp in 4 (33.3%) cases, Pseudomonas spp in 2 (16.6%) cases, and Stenotrophomonas maltophilia (8.3%), Enterobacter cloacae (8.3%), Achromobacter xylosoxidans (8.3%), Escherichia coli (8.3%), Staphylococcus haemolyticus (8.3%), and Candida albicans (8.3%) were detected in 1 patient each (Table 3).

Co-infection with 2 or more pathogenic organisms was detected in 5 (41.6%) patients. In 4 patients (33.3%), the same organism was isolated in repeated cultures. Repeated growth of Acinetobacter spp was seen in 3 patients; Enterococcus spp in 2 patients; and S maltophilia and A xylosoxians in 2 patients each. Antibiotic resistance (percentages) of the detected bacterial microorganisms is shown in Table 4. However, the antifungal susceptibility of C albicans could not be evaluated.

A urinary tract infection was detected in 3 patients. The pathogenic organisms were C parapsilosis and C glabrata in 2 cases, and E coli in 1 case. Rectal swab samples of 9 patients were screened, and carbapenem-resistant Enterobacteriaceae were detected in 8 of them. None of the patients had positive blood cultures.

All earthquake victims had already received antibiotic therapy at the initial intervention center. Upon admission to the authors’ center, empirical antibiotic therapy was given to all patients. The empirical antibiotic therapy regimens were revised according to the infectious organisms detected following bacteriological cultures. Therapy regimens and their durations are shown in Table 1. The most commonly used antibiotics were glycopeptides in 12 patients (12/12, 100%), carbapenems in 11 patients (11/12, 91.6%), metronidazole in 6 patients (6/12, 50%), cephalosporins in 4 patients (4/12, 33.3%), aminoglycoside in 4 patients (4/12, 33.3%), and colistin (polymyxin E) in 4 patients (4/12, 33.3%). 

The median length of stay in the hospital was 69 days; the range was 15 to 152 hospital days (min-max). The mean duration of antibiotic treatment was 23 ± 11 days. Seven patients experienced crush syndrome, of whom 5 needed renal replacement therapy. All patients were discharged after recovery, and all needed physiotherapy after discharge. Two patients experienced limb loss. The patient who had a single-limb above-knee amputation was diagnosed with osteomyelitis, and antibiotic therapy was continued after discharge.

One of the most frequent complications that victims experience following an earthquake is infection, which raises the risk of morbidity and mortality. A prolonged stay under rubble increases the severity of trauma,⁴ and crush syndrome and compartment syndrome occur as a result of increased trauma severity. Open injuries and fasciotomies cause wound infections due to bacterial contamination.⁶ In the Kahramanmaras earthquakes, people were trapped under brick and concrete debris, and traumatic wounds were contaminated with soil and rubble. Contamination was also caused by environmental factors resulting from the lack of sterile conditions during emergency first aid administered at the disaster site. 

Following an earthquake, 19% to 35% of trauma victims develop an infection.⁷ Consistent with other literature,^4,7 wounds were the most susceptible sites for infection in the current case series. In open wounds and fasciotomies, the loss of the mucocutaneous barrier and the presence of necrotic tissues provide an environment for the invasion of microorganisms. High infection rates after fasciotomies have been shown in certain studies.^8,9 In the current case series, 7 of 12 patients (58.3%) were fasciotomized, and all of these patients developed infections. 

In this case series, 9 different organisms were found to cause wound infections. The major bacterial isolates from wounds were Enterococcus faecium, Acinetobacter baumanii, and Pseudomonas aeruginosa. In the 1999 Marmara earthquake in Turkey, the most common agents of wound infections were found to be Acinetobacter spp, Pseudomonas spp, and Klebsiella spp. The majority of infections that were documented originated nosocomially during intensive care unit therapy.⁴ In the 2008 Wenchuan earthquake in Sichuan, China, the main pathogens of pediatric wound infections were A baumannii, E cloacae, P aeruginosa, and Staphylococcus aureus.¹⁰

E faecium, which has been reported in previous studies among wound pathogens in earthquake victims,^11,12 was found to be the most common microorganism in the current case series. Three of the 5 Enterococcus strains were determined to be vancomycin resistant. Through a variety of adaptive mechanisms that boost bacterial virulence and resistance in a nosocomial setting, E faecium has developed into a global health care hazard.¹³ The presence of serious clinical factors such as crush syndrome, susceptibility of fasciotomy wounds to contamination, venous or urinary catheters, and length of hospital stay may have increased the risk of infection in patients.¹⁴

Similar to previous studies,^4,10Acinetobacter spp was one of the major pathogens in the current case series. A baumanii is present in soil, water, humans, and animals. As such, earthquakes may pose a risk for bacterial spread due to disaster victims being buried under rubble and contaminated environmental conditions.¹⁵ In addition, the first interventions and transfers of patients in contaminated conditions and the presence of invasive catheters may have increased the risk of colonization.¹⁶

In the current cases, all Acinetobacter strains were MDR, resistant to carbapenem and aminoglycoside group antibiotics, and sensitive to colistin. MDR pathogens have been previously reported following natural disasters, including earthquakes.^10-12 A high prevalence of antibiotic-resistant A baumannii has been seen in infections linked to significant earthquakes in 1999 in Turkey, 2005 in northern Pakistan, 2008 in China, and 2010 in Haiti.^10,11,17,18

The large number of patients and the occurrence of life-threatening injuries following disasters such as earthquakes often lead to neglect of wound sterilization and care, which can result in serious infections, sepsis, and even tissue and limb loss. In order to treat wounds effectively, a variety of care methods should be selected according to individual patient needs. NPWT, which was applied in this case series, has been shown in the literature to significantly accelerate the healing process of open wounds and promote complete healing with minimal scarring.^19,20 Following the Kahramanmaras earthquake, successful results following HBOT were recorded by Gökmen and Uluöz, especially in patients with musculoskeletal injuries.²¹ Although HBOT application yielded positive results for 4 patients in the current case series, further prospective studies are needed to fully evaluate the long-term efficacy of these treatment modalities in post-earthquake injuries.

A small number of patients were followed up because of the inclusion of patients admitted to a single center.

Major disasters such as earthquakes are rare but result in various severe injuries. Infection is the primary complication that increases morbidity and mortality among earthquake victims. Since MDR strains are common pathogens in these patients, initial appropriate treatment contributes to improved outcomes. Moreover, the application of appropriate wound care strategies is critical in determining the outcome for these patients. 

Authors: Seyhan Yilmaz, MD¹; Gulsen Akkoc, MD¹; Sevgi Aslan Tuncay, MD¹; Burcu Parlak, MD¹; Aylin Dizi Isik, MD¹; Pinar Canizli Erdemli, MD¹; Ahmet Hamdi Sakarya, MD²; Yucel Agirdil, MD³; Hayati Kart, MD³; Eda Kepenikli Kadayifci, MD¹

Affiliations: ¹Department of Pediatric Infectious Disease, Marmara University School of Medicine, Istanbul, Turkey; ²Department of Plastic, Reconstructive and Aesthetiic Surgery, Marmara University School of Medicine, Istanbul, Turkey; ³Department of Orthopedics and Traumatology, Marmara University School of Medicine, Istanbul, Turkey

Disclosure: The authors disclose no financial or other conflicts of interest.

Correspondence: Gulsen Akkoc, MD; Marmara University, Pendik Training and Research Hospital, Fevzi Cakmak Quarter, Muhsin Yazicioglu Street, number: 10, 34899 Pendik/Istanbul, Turkey; agulsenakkoc@gmail.com 

Manuscript Accepted: May 21, 2024