Existing studies have shown that EF alone controls bleeding of venous origin by 70-85%7, but it has limited effect on persistent bleeding due to injury to the branches of the internal iliac arteries8. A study by Grilz et al.9 noted that untreated arterial hemorrhage can increase mortality by 3.2 times. In contrast, the AE technique, by selectively embolizing the bleeding artery, can theoretically achieve precise hemostasis, but there are potential risks such as delayed operation time and contrast nephropathy10. Of note, a systematic review by Cullinane et al.11 showed that combination therapy may shorten ICU stays, but the evidence for this is still limited to small sample cohort studies. The aim of this study is to clarify the effect of combined treatment in terms of transfusion requirements, ICU stay time and hemostatic success rate, as well as to systematically assess its safety by comparatively analyzing the clinical outcomes of AE + EF versus EF alone. The results of the study will provide a high-level evidence-based basis for optimizing the EF process in patients with unstable pelvic fracture with hemorrhagic shock.
This study utilized a retrospective cohort study design and followed the STROBE statement. Five hundred and sixty-two patients with pelvic fractures admitted to Yuyao People's Hospital from March 2017 to February 2025 were included, among which 102 cases of unstable pelvic fractures with hemodynamic instability were eligible, and 90 patients with unstable pelvic fracture with hemorrhagic shock (Tile C type/Young-Burgess APC III/LC III) were included after the exclusion criteria. They were categorized into AE + EF group (n = 45) and EF group (n = 45) according to the treatment modality, as shown in Fig. 1.The exclusion criteria included 1) Age ≤ 18 years or ≥ 75 years (n = 5), 2) Death in emergency surgery (n = 3), 3) Thoracic and abdominal surgery during treatment (n = 2), 4) Time from injury to admission exceeding 6 h (n = 1), 5) End-stage renal disease (n = 1).
Patients in the EF group underwent pelvic external fixation in accordance with the Eastern Association for the Surgery of Trauma guidelines, targeting a pelvic volume compression of ≥ 30%. Patients in the AE + EF group initially underwent arterial embolization under digital subtraction angiography (DSA) guidance. The procedural sequence encompassed puncturing the femoral artery on the healthy side or the less severely injured side, followed by catheter insertion. A 5F pigtail catheter (Terumo Corporation, Japan) was positioned superior to the abdominal aortic bifurcation for angiography, thereby imaging the bilateral internal iliac arteries. In cases where angiography revealed positive findings (such as contrast extravasation, arteriovenous fistula, or pseudoaneurysm), a superselective microcatheter was advanced to the bleeding vessel, and gelatin sponge particles (350-560 μm) manufactured by Hangzhou Alicon Pharmaceutical Co., Ltd. were employed to embolize the target site, with follow-up angiography confirming the resolution of bleeding signs; in instances of negative angiography, the corresponding branches of the internal iliac artery were directly embolized based on imaging localization of severe fracture regions. Subsequently, akin to the EF group, external fixation was performed, and the catheter sheath was removed after 24 h. The hemostatic approaches employed in this investigation were rigorously confined to external fixation (EF) alone or in conjunction with arterial embolization (AE). At our institution, the management of hemodynamically unstable pelvic fracture patients follows a multidisciplinary diagnostic and treatment protocol aligned with current guidelines. Arterial embolization (AE) is considered a first-line treatment option for patients in hemorrhagic shock with imaging evidence of active arterial bleeding. The decision to proceed directly with AE or to combine it with preperitoneal pelvic packing (PPP) and external fixation (EF) in the operating room is made collaboratively by trauma surgeons, interventional radiologists, and anesthesiologists, based on the patient's physiological condition, response to resuscitation, and the availability of a hybrid operating room.
Initial vital signs, ISS score, and shock index SI were recorded in both groups. Observational indicators: 24 h postoperative red blood cell transfusion (performed according to Massive Transfusion Protocol), hemostatic success rate (defined as stable SBP > 90 mmHg for 1 h without secondary intervention), ICU stay (meeting the criteria for transfer out: FiO2 < 40% and lactate < 2 mmol/L), Hospitalization mortality, Total operation time consumption/minutes. Additional recording of DSA findings, number of embolized vessels, and complications (classified by SIR) in the treatment group.
Data were analyzed using SPSS 26.0. Measurement data were expressed as median [IQR] using the Mann-Whitney U test; count data were expressed as rate (%) using the χ test or Fisher's exact probability method. P < 0.05 was considered a statistically significant difference.