The research, conducted at the Pediatric Emergency Department of Hacettepe University between January 2019 and December 2022, enrolled 559 patients aged 0 to 18 years who presented within one hour of injury and were observed for at least eight hours. The cohort had a median age of 91.4 months—roughly seven and a half years—and was nearly two-thirds male, a distribution that reflects the well-documented epidemiology of pediatric blunt trauma. Of the 559 children, 179, or 32 percent, ultimately required at least one life-saving intervention, giving the researchers a substantial group of positive cases against which to test the predictive power of each scoring system.

The study evaluated five widely used instruments. The Abbreviated Injury Scale, or AIS, grades the severity of injury in individual body regions on a six-point scale. The Injury Severity Score, or ISS, is derived from the AIS by summing the squares of the three most severely injured regions, producing a composite anatomical measure. The Pediatric Trauma Score, or PTS, integrates physiological and anatomical elements—airway status, blood pressure, level of consciousness, open wounds, skeletal injury, and weight—into a single number designed specifically for children. The Shock Index, the simple ratio of heart rate to systolic blood pressure, attempts to capture compensated shock through bedside hemodynamics. Finally, the BIG score—Base deficit, International Normalized Ratio, and Glasgow Coma Scale—combines metabolic, coagulation, and neurological markers into a laboratory-anchored risk estimate.

Because several of these scores are mathematically interdependent—the ISS is built directly from AIS values, for example—the researchers faced a statistical trap familiar to anyone modeling correlated predictors. Stuffing all five scores into one regression would produce unstable estimates distorted by multicollinearity. Their solution was a parsimonious multivariable model built from one representative variable per score family: the ISS for the anatomical domain, the PTS for the integrated physiological-anatomical domain, and the Shock Index for the hemodynamic domain. Each variable was entered into a multivariable logistic regression with life-saving intervention as the outcome, allowing the team to isolate the independent contribution of each domain while controlling for the others.

The results were striking. Higher ISS values independently predicted the need for a life-saving intervention, with an adjusted odds ratio of 1.145 per point (95 percent confidence interval 1.091 to 1.202), while lower PTS values were similarly predictive, with an adjusted odds ratio of 0.552 per point (95 percent confidence interval 0.474 to 0.642); both associations were highly significant at p less than 0.001. The Shock Index, by contrast, failed to reach independent significance (p equals 0.134), a finding that will resonate with pediatric emergency physicians who have long suspected that adult-derived hemodynamic thresholds translate poorly to children, whose robust physiological compensation can mask serious blood loss until decompensation is abrupt and late.

Discriminatory power—the ability of a score to separate children who need interventions from those who do not—was quantified with the area under the receiver operating characteristic curve, or AUC. Among individual scores, the Pediatric Trauma Score led the field with an AUC of 0.947, followed closely by the Glasgow Coma Scale at 0.920 and the ISS at 0.918. The BIG score achieved 0.897 and the AIS 0.878, while the Shock Index trailed dramatically at 0.601, barely better than a coin flip. The combined three-variable model, however, outperformed everything: it achieved an AUC of 0.961 (95 percent confidence interval 0.943 to 0.978), and a formal DeLong test confirmed that this improvement over the best individual score, the PTS, was statistically significant (p equals 0.004).

The metabolic story added an important nuance. In a separate adjusted analysis restricted to the subgroup of patients with complete laboratory data, the researchers examined base deficit and lactate—two markers of tissue hypoperfusion measured on arterial or venous blood gases. After adjustment for injury severity and neurological status, base deficit remained an independent metabolic predictor of intervention requirement, whereas lactate did not. This distinction matters operationally: base deficit reflects the cumulative metabolic debt from anaerobic metabolism and is less susceptible to confounding by liver function, clearance kinetics, and the stress response than lactate, which can rise for reasons unrelated to hemorrhage in children. Crucially, the authors note that these metabolic markers were not components of the combined ISS-PTS-Shock Index model, meaning the multiparametric framework they propose can function without waiting for laboratory results.

The clinical implications are considerable. Triage in pediatric trauma is a high-stakes balancing act: over-triage floods trauma teams and imaging suites with children who turn out to be fine, while under-triage sends seriously injured children to wards unequipped for sudden deterioration. A scoring framework that integrates anatomical burden (ISS), child-specific physiological reserve (PTS), and bedside hemodynamics (Shock Index)—with base deficit layered in once laboratory data become available—offers a staged approach that matches the natural rhythm of emergency care. The near-perfect discrimination of the combined model, if replicated prospectively, could support earlier activation of massive transfusion protocols, earlier involvement of pediatric surgeons, and more rational decisions about which children can safely be observed rather than subjected to whole-body computed tomography.

The study’s limitations deserve honest acknowledgment. As a retrospective single-center analysis at a Level I pediatric trauma center, its findings may not generalize to community hospitals or to health systems with different prehospital care patterns, and the requirement that patients present within one hour of injury and be observed for at least eight hours introduces selection considerations. Retrospective designs also cannot fully exclude the possibility that some scoring data were influenced by treatments already underway. The authors themselves frame the multiparametric model as a framework for early in-hospital risk stratification after initial diagnostic evaluation, not as a prehospital triage tool, and they emphasize that the Shock Index’s weak performance argues against relying on vital-sign ratios alone in children.

Still, the central message is likely to influence practice and future research alike: in pediatric trauma, the whole exceeds the sum of its parts. The Pediatric Trauma Score, the Glasgow Coma Scale, and the Injury Severity Score are each strong individual predictors, but a parsimonious model spanning anatomical, physiological, and hemodynamic domains significantly outperforms the best of them. As emergency departments worldwide grapple with rising pediatric injury volumes and pressure on critical care resources, the study offers a data-driven template for identifying, within the first hour, which injured child truly needs the team’s full attention—and which life-saving intervention should begin before the child has even left the resuscitation bay.

Subject of Research: Predictive performance of clinical, physiological, and trauma scoring systems for identifying life-saving intervention needs in pediatric trauma patients

Article Title: Predictive performance of clinical, physiological, and trauma scores for life-saving interventions in pediatric trauma: a retrospective cohort study

Article References: Gungor, E., Birbilen, A. Z., Yildiz, L. A., & Teksam, O. (2026). Predictive performance of clinical, physiological, and trauma scores for life-saving interventions in pediatric trauma: a retrospective cohort study. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07843-x

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07843-x

Keywords: pediatric trauma, life-saving interventions, Injury Severity Score, Pediatric Trauma Score, Shock Index, Glasgow Coma Scale, base deficit, emergency triage, trauma scores, predictive medicine, pediatric emergency medicine, retrospective cohort study

The biomechanics of a horse kick explain why such injuries deserve serious respect. Studies of equine impact forces have measured strikes reaching up to 7,000 newtons, roughly equivalent to the weight of a 700-kilogram mass, delivered through a small hoof contact area in a fraction of a second. That concentration of energy over a narrow surface is what makes horse kicks capable of tearing bowel, fracturing ribs, and rupturing solid organs. Yet, paradoxically, the external evidence of such a blow can be minimal. In modern societies, routine contact with horses has become less common, but riding clubs, farms, and stables keep the risk alive, and emergency physicians may see such cases rarely enough that the injury pattern is not immediately at the front of their minds.

On arrival, the patient was alert and her vital signs were reassuring: blood pressure of 148/79 mmHg and a heart rate of 62 beats per minute. Her abdomen, however, told a different story. Examination revealed generalized tenderness with rebound tenderness and muscular guarding, the classic signs of diffuse peritoneal irritation, indicating that the lining of the abdominal cavity was inflamed, often a signal of leaking bowel contents or blood. Most strikingly, the emergency team noted a faint, horseshoe-shaped contusion in the left lower quadrant, a subcutaneous hemorrhage whose curved shape mirrored the contour of the hoof that struck her. The bruise was photographed at presentation, and no other external injuries such as lacerations or fractures were found.

Laboratory results added to the diagnostic puzzle. Her white blood cell count was elevated at 13,400 per microliter, consistent with an inflammatory response, but hemoglobin, platelets, liver enzymes, renal function, and C-reactive protein were all within normal ranges. A focused assessment with sonography for trauma, the bedside ultrasound scan widely used to detect free fluid in injured patients, was negative. Arterial blood gas analysis showed a lactate of 2.4 mmol/L and a base excess of minus 0.9 mmol/L, values suggesting only mild physiological stress. Her overall Injury Severity Score was calculated at 9, a figure that on its own would not suggest a life-threatening injury. This is precisely the trap that trauma surgeons warn about: stable vital signs and modest laboratory abnormalities can coexist with a perforated hollow organ quietly leaking intestinal contents into the peritoneal cavity.

The decisive test was contrast-enhanced computed tomography of the abdomen. The scan demonstrated intraperitoneal free air, an unmistakable marker of bowel perforation, along with focal thickening of the small bowel wall and increased contrast enhancement located directly beneath the horseshoe-shaped contusion. Crucially, there was no injury to the liver, spleen, or any other solid organ. The anatomical correspondence between the skin bruise and the underlying bowel damage was remarkable: the hoof had compressed the jejunal loop against the spine and pelvic ring, crushing the bowel wall at exactly the point where the contusion marked its impact. In blunt abdominal trauma, solid organ injuries are far more common than hollow viscus injuries, so in a patient without hemorrhagic findings, clinical attention can easily drift toward the liver and spleen while a perforated intestine goes unnoticed until peritonitis or sepsis sets in.

With diffuse peritoneal irritation and clear CT evidence of free air, the surgical team made the call for emergency laparotomy, performed approximately six hours after the injury. Through a midline incision, they systematically explored the entire gastrointestinal tract. The culprit was found: a fingertip-sized perforation on the antimesenteric border of the jejunum, the segment of small bowel richest in blood supply, located about 70 centimeters distal to the ligament of Treitz. The injury was classified as Grade III on the American Association for the Surgery of Trauma bowel injury scale. Notably, the surrounding bowel wall appeared viable, without extensive contusion, ischemia, or mesenteric vascular compromise, and no other injuries were found anywhere in the abdomen.

The repair itself was elegantly straightforward, a fact that underscores how early detection changes outcomes. After irrigating the abdominal cavity with 10 liters of normal saline, the surgeons debrided the edges of the injured bowel wall and closed the perforation primarily with simple interrupted 4-0 polydioxanone sutures, an absorbable monofilament that retains tensile strength for weeks. The fascia was closed with heavy #1 polyglactin sutures, the subcutaneous wound was washed with a further 2 liters of saline, and the skin was closed with buried subcuticular stitches. Because the perforation was caught before extensive contamination developed, a bowel resection and stoma, procedures that carry far greater morbidity, were avoided entirely. The horseshoe bruise, in effect, had pointed the surgeons directly to the injury.

Recovery was not entirely smooth. The patient experienced a transient postoperative ileus, a temporary paralysis of bowel function that is a common complication after abdominal surgery, and she received intravenous piperacillin/tazobactam, a broad-spectrum antibiotic combination, for seven days to guard against infection. Oral intake resumed on postoperative day six, and after gradual dietary advancement and confirmation of stable bowel function, she was discharged on postoperative day eleven. At a one-month follow-up visit, she reported no abdominal symptoms, no pain, and no signs of bowel obstruction. For an injury that, if missed, could have progressed to fecal peritonitis, septic shock, and death, the outcome represents the best-case scenario that prompt imaging and timely surgery make possible.

The case also fuels an ongoing debate in trauma surgery about operative strategy. The team chose open laparotomy because of diffuse peritoneal irritation and concern for hollow viscus injury, the traditional and safest approach when bowel perforation is suspected. But the authors note that in hemodynamically stable patients in whom the injury can be reasonably localized, diagnostic and therapeutic laparoscopy may be a reasonable alternative, provided that a thorough exploration of the entire bowel can be achieved and the threshold for converting to an open procedure remains low. Minimally invasive exploration could reduce wound complications and hospital stay, but it demands surgical judgment and institutional capability, and the diffuse peritoneal signs in this patient justified the more definitive open approach.

The broader message for readers is one that emergency physicians and horse handlers alike should take to heart. Isolated small bowel perforation after blunt trauma can present with subtle or nonspecific early findings, and delayed recognition has been repeatedly associated with increased postoperative complications and mortality. Previous reports have described a range of small-bowel injuries from horse kicks, including jejunal perforation and complex injuries involving the mesentery and abdominal wall, but this case stands out for its isolated nature, the striking anatomical match between the horseshoe-shaped contusion and the perforation beneath it, and the early CT diagnosis that enabled simple primary repair. Localized external findings, however reassuring they may appear, do not reliably exclude significant internal injury when abdominal pain or peritoneal signs persist. After any high-energy localized impact, whether from a hoof, a handlebar, or a seatbelt, persistent abdominal pain should trigger early cross-sectional imaging, and when bowel injury is suspected, timely surgical intervention remains the difference between a straightforward suture repair and a catastrophic outcome.

Subject of Research: Isolated jejunal perforation following blunt abdominal trauma from a horse kick

Article Title: Jejunal Perforation Associated With a Horseshoe‐Shaped Abdominal Contusion After a Horse Kick: A Case Report

Article References: Yoshida, J., Hanaki, T., Sakano, Y., & Fujiwara, Y. (2026). Jejunal Perforation Associated With a Horseshoe‐Shaped Abdominal Contusion After a Horse Kick: A Case Report. Clinical Case Reports, 14(10), Article e73687. https://doi.org/10.1002/ccr3.73687

Image Credits: AI Generated

DOI: 10.1002/ccr3.73687

Keywords: horse kick, jejunal perforation, blunt abdominal trauma, hollow viscus injury, computed tomography, emergency laparotomy, peritonitis, trauma surgery, case report, bowel injury, abdominal contusion, injury severity score