Abstract
The Israeli military invasion of 2023-2025 has decimated Gaza’s health system, constraining trauma capacity and disrupting prehospital and hospital care. We aimed to characterize emergency department (ED) orthopedic injury patterns. We conducted a multicenter, descriptive cross-sectional study across four referral hospitals in different governorates (September 1–16, 2024). A standardized interviewer-administered instrument and chart review captured demographics, mechanism, injury type/site, soft-tissue and neurovascular findings, acute interventions, disposition, perceived care quality, and time metrics (to first care, ED evaluation, surgery). Primary outcomes were admission and emergency surgery. Analyses included descriptive statistics and bivariable/multivariable logistic regression with hospital-clustered standard errors. Among 449 patients, the cohort was predominantly young, male, and displaced. Most presentations followed explosive violence. Fractures predominated, with a large share open; comminuted and segmental patterns were frequent, and neurovascular compromise was common. Many patients reached some form of care rapidly, yet few received first aid before ED arrival, indicating prehospital collapse amid unsafe transit. Median wait-time was short but highly variable between hospitals. Operative demand exceeded theatre capacity, with notable interhospital variation in time to surgery. Orthopedic trauma care in Gaza continues to be affected by widespread damage, though unevenly across the enclave. Immediate priorities include the entry of essential medical supplies and safe protection for hospitals and healthcare workers to operate.
Introduction
The Gaza Strip’s health system has been decimated since October 2023, with the Israeli military invasion resulting in repeated mass displacement events and crowding of people in hospital grounds, increasingly limited access for humanitarian agencies and actors, and a large burden of traumatic injury through mid-2025. United Nations (UN) situation updates through August 2025 describe the continuation of large-scale hostilities, recurrent evacuations of densely populated areas, and severe disruptions to essential health services that impede both pre-hospital evacuation and hospital-based emergency care.1,2
Within this context, the capacity for trauma care in Gaza has contracted sharply. By August 2025, only 18 of 36 hospitals were even partially functional, and numerous health facilities had been damaged, destroyed, or raided, with limited medical supplies being among the leading causes of service interruption.2–4 Hundreds of healthcare personnel, including trained orthopedic surgeons, have been killed, abducted, or displaced outside of Gaza.5,6 Shortage of beds due to the destruction and forced evacuation of hospitals, increased risk of the spread of infections due to crowded and unhygienic conditions, and the manufactured scarcity of surgical equipment (through its destruction or blockading of its entry) have forced trauma teams to prioritize life-saving amputations over limb salvage in many cases, with early discharges and high wound-infection burdens reported by front-line clinicians.7–9
World Health Organization (WHO) analyses drawing on emergency medical team data estimated more than 24,000 people with major life-altering injuries by September 2024, presaging a long horizon of rehabilitation needs that far exceeds the current in-strip surgical and post-acute care capacity.10,11 International emergency medical team deployments that sought to offset these deficits in 2025 operated amid stringent entry restrictions on personnel and supplies, continued attacks on health infrastructure, and unsafe corridors for movement, all of which undermined the continuity and quality of surgical and postoperative care. 12
Evidence from other areas of armed conflict showcases some of the consequences of rapidly saturated and delayed orthopedic services and care. Blast injuries and gunshot wounds often present with complex patterns of open fractures, neurovascular compromise, and contaminated soft-tissue defects that often require staged debridement, external fixation, timely use of antibiotics, and meticulous wound care.13,14 Delays in any of these steps may amplify the risks of infection, non-union, and amputation, and they impose sustained demands on rehabilitation systems that are rarely met in conflict-affected settings. 15 For example, a field hospital in Aleppo during the Syrian civil war managed open long-bone fractures with external fixation as the primary and definitive modality and reported union in approximately two-thirds of followed patients with an infection rate near 17%, showcasing a pragmatic, damage-control pathway when staged internal fixation and dependable follow-up care is not possible. 16 This is in line with what clinicians have reported across Gaza since late 2023 - that operating theatres, sterilization capacity, and supplies for fracture stabilization and advanced wound management have become increasingly limited. 17
Against this backdrop, a standardized characterization of orthopedic injury epidemiology in Gaza is urgently needed to inform triage decisions, surgical prioritization, and planning for rehabilitation of patients. Our multicenter, emergency department–based cross-sectional study was designed to address this gap by describing the demographic profile, mechanisms, anatomical distribution, and acute management of conflict-related orthopedic trauma across four referral hospitals in various governorates during a defined period in 2024. These data provide a needed empirical baseline for service planning in a health system that must simultaneously manage acute mass-casualty surges and a rising cohort of patients with long-term disability.
Methods
We conducted a hospital-based, descriptive cross-sectional study to characterize the demographic profile, mechanisms, anatomical distribution, and acute management of orthopedic injuries presenting to emergency departments (EDs) in the Gaza Strip during the Israeli military invasion (2023-ongoing). The study was conceived, implemented, and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 18 Data were collected at four major referral hospitals distributed across various governorates from September 1-September 16, 2024; this included the Al-Shifa Medical Complex (Gaza City), Al-Ahli Arab Hospital (Gaza City), Al-Aqsa Martyrs’ Hospital (Deir Al-Balah), and the Nasser Medical Complex (Khan Yunis). These hospitals were selected for the critical role they played in regional emergency and trauma care, and for their efforts to continue admitting patients with conflict-related casualties during the study window. This study was conducted in accordance with relevant guidelines stipulated under the Declaration of Helsinki and received approval from the Islamic University of Gaza.
Participants: Eligibility and sampling
The target population comprised all patients of any age or sex presenting to the participating EDs with a clinically confirmed orthopedic injury during the study period. Inclusion required presentation to one of the four EDs with musculoskeletal trauma, including but not limited to fractures, dislocations, sprains, and strains, and provision of informed consent by the patient or a guardian. To ensure the cohort reflected acute and clinically significant presentations, we excluded individuals managed exclusively in primary-care or outpatient settings without ED evaluation, as well as cases with insufficient clinical documentation for core variables or those declining participation. All consecutive, eligible patients during the study window were enrolled, yielding a final sample of 449 participants.
Data sources, instruments, and field procedures
Data were collected using a standardized, interviewer-administered questionnaire developed for this study with reference to established orthopedic injury surveillance frameworks and refined through expert consultation.19–21 The instrument captured demographics (age, sex, residence), injury context and mechanism (e.g., blast, crush, fall, road traffic, assault), injury type and anatomical site, ED triage information where available, acute interventions (e.g., analgesia, reduction, immobilization, irrigation and debridement, antibiotic administration), disposition (discharge, admission, transfer), and early operative management. An initial pilot was conducted with 10 patients to assess clarity, item flow, and feasibility, prompting some minor wording adjustments before full deployment.
Palestinian research staff, comprising final-year medical students and recently graduated physicians familiar with local emergency department protocols, participated in a 2-day training session that included standardized interviewer instruction, pilot role-playing, and calibration exercises to ensure consistency. They approached patients or their guardians only after the initial stabilization was accomplished, conducted bedside interviews in conjunction with the treating clinicians to prevent disruption of care, and recorded responses utilizing the standardized study instrument. Supervisors reviewed the forms daily and addressed discrepancies by revisiting records or consulting the treating team. Data were coded and entered into SPSS version 26 using a double-entry workflow with programmed range and logic checks.
Variables and definitions
Orthopedic injury categories followed clinicians’ diagnoses as recorded in the ED or orthopedic notes. Mechanisms of injury were recorded as reported by the patient or caregiver and corroborated by clinical documentation when available. Primary outcomes of interest were hospital admission and receipt of emergency surgical intervention during the index ED encounter or same-day admission. Pre-specified covariates for analysis included age, sex, mechanism, anatomical site, and hospital. ‘Bleeding’ in the text refers to vascular injury requiring medical intervention due to the presence of unstable vital signs. ‘Raw’ refers to an area of exposed, damaged tissue where the protective skin or mucosal surface has been lost.
Fracture morphology terms (e.g., comminuted, segmental, oblique, greenstick) reflect basic orthopedic descriptors documented by treating clinicians and/or radiology notes. We did not apply a formal AO/OTA classification because imaging and documentation were not uniformly available across hospitals. For clarity: a comminuted fracture indicates >2 bony fragments, a segmental fracture indicates ≥2 distinct fracture lines isolating an intermediate segment (typically in long bones), an oblique fracture indicates an angled fracture line relative to the bone axis, and a greenstick fracture indicates an incomplete fracture pattern typically seen in children.
Open versus closed status was recorded when documentation allowed. An ‘open’ injury indicates a skin breach with communication to deeper tissues (e.g., open fracture, open dislocation, or traumatic amputation), while ‘closed’ indicates no such breach. Bone-specific details (e.g., Gustilo grade or consistent bone-level openness) were not uniformly available so therefore, we report openness as an injury-level descriptor and do not stratify it by specific bone.
A ‘direct bombing’ indicates the patient was injured by an explosive event at the point of impact or immediate blast radius (e.g., strike on the same structure/area), including injury from blast effects, shrapnel, or collapse of the impacted structure. On the other hand, ‘indirect bombing’ indicates injury from a nearby explosive event not impacting the patient’s immediate location, such as being struck by falling debris or fragments from an adjacent strike or structural collapse in the vicinity.
Statistical analysis
We summarized categorical variables as counts and percentages, and continuous variables as means with standard deviations or medians with interquartile ranges, selected according to distributional properties. Group comparisons used the Chi-square test or Fisher’s exact test for categorical data and the independent-samples t test or one-way analysis of variance for approximately normal continuous variables; when normality appeared implausible, non-parametric alternatives were applied. We constructed bivariable logistic regression models to screen associations between candidate predictors and the primary outcomes (admission; emergency surgery). Multivariable logistic regression was then used to estimate adjusted odds ratios with 95% confidence intervals, including covariates selected a priori on clinical grounds and retained if they contributed meaningfully to model fit or confounding control.
To account for potential correlation by site, standard errors were calculated with hospital-level clustering. Statistical significance was defined as a two-sided p-value ≤0.05. Missing data were summarized by variable; analyses used available cases, and denominators are reported wherever missingness occurred. Records with missing outcome data were excluded from outcome-specific models only.
Results
Demographics
Demographics of study participants (n = 449).
Clinical characteristics
Clinical characteristics of patients.
Injury patterns
Injury distribution and classification.
The mechanism of injury was predominantly direct bombing (299/449, 66.6%), followed by indirect bombing (93/449, 20.7%), falls (47/449, 10.5%), and road-traffic crashes (10/449, 2.2%). Associated visceral injury was absent in 277/449 (61.7%); when present, the most frequent pattern was multiple visceral injuries with burns (104/449, 23.2%), with isolated head injury in 25/449 (5.6%), abdominal injury in 18/449 (4.0%), thoracic injury in 8/449 (1.8%), and multiple burns without specified visceral injury in 17/449 (3.8%). One limb was injured in 364/449 (81.1%), two limbs in 45/449 (10.0%), three limbs in 6/449 (1.3%), and four limbs in 2/449 (0.4%); pelvic injuries occurred in 23/449 (5.1%) and pelvis with other limb injuries in 9/449 (2.0%). Injuries most often occurred in the morning (207/449, 46.1%), followed by evening (141/449, 31.4%) and night (101/449, 22.5%).
Most patients reported reaching some form of medical care at the accident site within 1 hour (323/449, 71.9%); 100/449 (22.3%) arrived within 1–6 h, 11/449 (2.4%) in 6–24 h, and 15/449 (3.3%) after more than 24 h. The distribution of reported minutes to first care was highly right-skewed (mean 237 min, SD 2086; median 30; IQR 70; range 0–43,200). Only 124/449 (27.6%) reported receiving any first aid or treatment before reaching the emergency department. Emergency department waiting time before first medical attention was less than 1 hour for 338/449 (75.3%) and 1–6 h for 97/449 (21.6%); the distribution was also skewed (mean 58.9 min, SD 154.4; median 15; IQR 50; range 0–1500).
Perceived ED care quality, assessed through self-report, was rated as normal by 178/449 (39.6%), bad by 136/449 (30.3%), good by 103/449 (22.9%), excellent by 29/449 (6.5%), and unknown by 3/449 (0.7%). While in the ED, 154/449 (34.3%) reported being on the ground, 73/449 (16.3%) on a mattress, and 222/449 (49.4%) on a bed. Analgesics were used by 320/449 (71.3%). Regarding operative needs, 217/449 (48.3%) were told they required one operation, 128/449 (28.5%) multiple operations, 65/449 (14.5%) no operation, and 39/449 (8.7%) were unknown. Among those with a recorded estimate of time to surgery (n = 276), the distribution was skewed (mean 7.1 h, SD 15.9; median 2 h; IQR 5; range 0–150).
Bivariable associations
Age differed by fracture type (Kruskal–Wallis p = 0.001). Post hoc Mann–Whitney tests indicated patients with greenstick fractures were significantly younger than those with comminuted (p < 0.001), segmental (p < 0.001), oblique (p < 0.001), and “other” fractures (p = 0.002). Age also varied by perceived ED care quality (Kruskal–Wallis p = 0.009): patients rating care as normal were younger than those rating it bad (p = 0.017) or good (p = 0.002), and those rating care good were older than those rating it excellent (p = 0.047) (Table 4).
Time to first care at the accident site differed by location (Kruskal–Wallis p = 0.019); patients in Deir Al-Balah reported fewer minutes than those in Gaza City (p = 0.003) and North Gaza (p = 0.032). It also differed by which hospital they were received in (Kruskal–Wallis p = 0.010); patients linked to Al-Aqsa reported fewer minutes than those at Al-Ahli (p = 0.008) and Al-Shifa (p = 0.009). Emergency department waiting time differed by hospital (Kruskal–Wallis p < 0.001); patients at Al-Ahli reported longer waits than those at Al-Shifa (p = 0.050), Al-Aqsa (p < 0.001), and Nasser (p < 0.001), and waits at Al-Shifa exceeded those at Nasser (p < 0.001). Among patients given an estimate of time till operation, time to surgery differed by hospital (Kruskal–Wallis p < 0.001); estimates were shorter at Al-Ahli than at Al-Shifa (p < 0.001) and Al-Aqsa (p < 0.001).
Discussion
This multicenter, emergency department-based study describes the demographic profile, mechanisms, injury patterns, and acute management of orthopedic trauma presenting to major referral hospitals in Gaza during the 2023-2025 period of active hostilities with the aim of informing triage, surgical decisions, and health system planning under severe resource constraints. The findings are consistent with the high frequency of blast-related injury causes and of open and comminuted fractures observed in this cohort. Fractures accounted for nearly three quarters of case presentations, almost half of such documented fractures were open, comminution and segmental patterns predominated, and vascular or neurological compromise was recorded in roughly one quarter of patients. This is somewhat consistent with injuries in other modern conflicts in which extremity trauma, open wounds, contamination, and complex soft-tissue loss can drive early operative demand and long-term disability.13,22–25 Contemporary war-surgery guidance emphasizes staged debridement, prompt antibiotic therapy, and external fixation as the safest definitive strategy where sterile capacity, implants, and follow-up are unreliable, which is the situation repeatedly described by Gaza clinicians since late 2023. 26
Even beyond the injury pattern, the process of receiving care points to a system that is variable and changing across the Gaza Strip. Some patients reached some form of care relatively quickly, yet many still did not receive meaningful first-aid before arrival in a hospital. Although median emergency department waiting times were short overall, waiting times varied significantly across hospitals, indicating that summary statistics may not fully capture inter-hospital differences in patient flow and triage capacity. Bed scarcity also resulted in care spilling onto floors and improvised surfaces, and operative demand consistently outstripped theatre capacity, producing site-specific delays to surgery that are clinically consequential for open, contaminated injuries. In well-resourced trauma systems, there are often targets to ensure care for open extremity injuries is conducted as soon as possible, including the administration of rapid prophylactic antibiotics (ideally within ∼1 h of injury) and operative irrigation/debridement as soon as is feasible (typically aiming for <24 h when possible though timing can be dependent on contamination and vascular compromise).27–29
Bivariable associations and statistical tests.
The Gaza Strip’s orthopedic system’s capacity constraints are acute and system-wide, and they are essential to interpreting the care-process gradients observed between hospitals in our sample. By early August 2025, only half of Gaza’s 36 hospitals were even partially functional, with entire governorates showing zero fully functioning facilities.1,2 At the Al-Aqsa Martyrs’ Hospital, the only public hospital serving Gaza’s middle area (Deir Al-Balah), the orthopedic service reports having converted obstetrics/gynecology theaters and even former delivery rooms into operating suites, with roughly two-thirds of all operations becoming orthopedic, emergency care spilling into expanded floor space and tents, and a fluid green/red triage coping with approximately 300 orthopedic ED cases per day. 32 Inpatient capacity was pushed to more than triple its design limit with many patients in corridors, a three-team 24-h rotation was instituted to sustain basic services, and chronic shortages of sterile supplies and implants forced repeated delays of open fractures and other non-urgent cases, driving complications such as non-union, compartment syndrome, and wound infection. 32 To mitigate the backlog, clinicians established a “Dressing Under General Anesthesia” pathway that processed about 30 severe wound cases per day in repurposed rooms, yet mass-casualty surges, such as one of the incidents at Al-Nuseirat (∼1000 casualties within hours, including ∼100 severe orthopedic injuries and amputations), still required ad hoc referrals to MSF and other field hospitals and exposed persistent coordination challenges despite receiving support from international NGOs. 32 This potentially helps explain why a third of ED patients in our cohort were on the floor or a mattress rather than on a bed, why fewer than one in three received any first aid before reaching the ED, and why “time to surgery” estimates and waiting times varied markedly by site.
While this study was not designed to examine causal relationships between nutrition and orthopedic outcomes, indicators of nutritional and health vulnerability can at times lead to further complications in recovery.33,34 In our cohort, three-quarters reported an imbalanced diet, and over 40% reported anemia or malnutrition during the crisis. These exposures are plausible proximal causes for poorer health outcomes and are consistent with international rehabilitation assessments that estimate substantial post-trauma rehabilitation needs in Gaza.33–40
The patterns we report also map onto some of the operational realities of surgical delivery amid this blockade. Delays in pre-hospital evacuation, ambulance access, and safe corridors for staff movement and supply lines have been extensively documented. They are consistent with our finding that only 27.6% of patients received first aid before ED arrival, even though most reached “some form of medical care” within an hour. Targeting and obstruction of health services degrade triage capabilities, sterilization procedures, and infection-prevention systems, which in turn heighten the risk of fracture-related infection after external fixation or debridement. There is also a risk that the victims to experience psychological distress as a consequence of these injuries.41–43
Given the observational, cross-sectional design and short sampling window, the findings should be interpreted with appropriate caution. Several key variables, such as time to first care and perceived care quality were gathered from patient reports and are therefore potentially subject to limitations such as recall bias and measurement error. Time to surgery data were not available for all patients and, where recorded, reflect less reported estimates rather than more accurate operative start times. In addition, participating sites were major referral hospitals that were at least partially functional during the period of the study, and observed patterns may not fully represent conditions in other facilities that were non-operational, inaccessible, or functioning under different constraints at other points in the 2023-2025 period.
While the sample size is limited and tied to a short time frame, the findings provide a useful snapshot into the patterns and trends of orthopedic trauma in Gaza. In documenting injury patterns alongside variations in the administration of first aid, emergency department waiting times, and times to surgery between hospitals, the analysis reflects aspects of orthopedic trauma care in Gaza that are directly relevant to service organizations under similar ongoing constraints. Humanitarian agencies and clinical teams involved in trauma response and recovery planning in such settings can be informed by the observations made by this study. Examination of operative and early post-operative outcomes alongside this data may help clarify the clinical implications of the delays to care observed. Further work using similar emergency department-based data across longer time periods and additional hospitals would help clarify how observed patterns vary as access to care and its capacity continue to change.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.