Pediatric hepatic pseudoaneurysm formation following blunt and penetrating abdominal trauma: a 5-year retrospective analysis
Article information
Abstract
Purpose
Hepatic pseudoaneurysm (HPA) is a rare but potentially life-threatening sequela of liver trauma. This study aimed to evaluate the number of children diagnosed with HPAs following blunt and penetrating liver trauma at a major trauma center in London over a 5-year period.
Methods
A retrospective analysis was conducted of patients aged <18 years with liver injuries admitted to our institution between 2017 and 2022. Patients were identified through the local trauma registry. Data collected included patient demographics, imaging details (including ultrasound scanning and computed tomography [CT]), liver injury grade, and management plans.
Results
Thirty-four patients were identified (mean age, 12 years; range, 1–17 years). There were 26 men (76.5%) and 8 women (23.5%). Twenty-three patients (67.6%) sustained blunt injuries, and 11 (32.4%) sustained penetrating liver injuries. Among 30 patients (excluding 4 patients who died), 18 (60.0%) underwent repeat CT imaging. CT imaging identified four HPAs (13.3%), two of which were treated with embolization. Of the 12 patients who did not undergo repeat CT, 6 (50.0%) had follow-up ultrasound scanning.
Conclusions
Hepatic pseudoaneurysms following trauma are uncommon. Their management in the pediatric population presents challenges regarding follow-up imaging, particularly the use of CT. Although the absence of standardized protocols complicates surveillance strategies, an individualized approach informed by patient-specific factors is essential. Contrast-enhanced ultrasound may optimize HPA detection while minimizing radiation exposure.
INTRODUCTION
Background
Hepatic pseudoaneurysm (HPA) is an uncommon vascular complication of traumatic liver injury that is associated with considerable morbidity and mortality [1]. Patients with HPAs often remain asymptomatic unless rupture and bleeding occur, at which point they may present with acute abdominal pain and signs of hemorrhage [2]. In hemodynamically stable adults with traumatic liver injuries, follow-up contrast-enhanced computed tomography (CT) and emergent endovascular embolization constitute current standard practice. Durkin et al. [3] previously reported that approximately half of posttraumatic HPAs became symptomatic around one week after injury and that HPA formation may not occur within the first few days. Therefore, extending the interval before repeat imaging may increase detection rates. Navarro et al. [4] reported that asymptomatic, hemodynamically stable children with blunt liver trauma did not benefit from routine follow-up CT or ultrasound. However, data to guide management in patients younger than 18 years remain limited [5].
The American Association for the Surgery of Trauma (AAST) classifies traumatic liver injuries into five grades (I–V) based on the extent of hepatic surface involvement and vascular injury severity [6]. Following liver trauma, extravasated blood accumulates within adjacent tissue, forming a cavity lined by a fragile fibrin-platelet matrix. This structure lacks the structural integrity of a true arterial wall, thereby increasing the risk of rupture and potentially fatal hemorrhage [7]. Although standardized measures of injury severity, including the AAST grading system and the Injury Severity Score (ISS), have not been consistently correlated with subsequent HPA formation in adults, current evidence suggests that high-grade liver injuries are associated with an increased risk of HPA and may warrant closer surveillance compared with lower-grade injuries [2,5–8]. Experience from a level 1 major trauma center demonstrated a correlation between higher AAST grades and the need for embolization of HPAs [9]. Pseudoaneurysms identified during surveillance may be electively embolized by interventional radiologists to reduce the risk of spontaneous rupture and hemodynamic deterioration. However, no evidence-based consensus currently exists regarding how pseudoaneurysm size or location should influence management decisions in asymptomatic patients [10].
Due to the relative paucity of pediatric data, no published recommendations currently exist for HPA surveillance in children with traumatic liver injuries. Radiation exposure is a critical consideration in this population because of the increased lifetime risk of radiation-induced malignancy. Smith-Bindman et al. [11] reported a median radiation dose of 31 millisieverts per multiphase abdominal CT, equivalent to approximately 442 plain-film chest radiographs. Historically, Doppler ultrasound has been used for follow-up imaging in children with traumatic liver injuries [12]. In 2015, Durkin et al. [3] reported that contrast-enhanced ultrasound (CEUS) effectively identified HPAs <5 mm in any location within the liver or spleen. Deftereos et al. [13] proposed CEUS as an alternative to contrast-enhanced CT, reporting high sensitivity for detecting visceral injury in patients aged 4 to 14 years. However, no HPAs were identified among the 59 patients in that cohort, underscoring the low incidence of this complication [13]. Although CEUS represents a promising nonionizing alternative in pediatric populations, its diagnostic accuracy is operator-dependent and generally considered less objective than CT imaging. At present, the available evidence is insufficient to support routine CEUS use in this group.
Objectives
At our institution, a major trauma center in London, an algorithm has been implemented to guide follow-up imaging in adults with traumatic liver injuries. These patients undergo a repeat limited CT scan of the abdomen 48 to 72 hours after admission to assess for vascular complications, including HPAs and arteriovenous fistulae. The present study aimed to evaluate the number of children diagnosed with HPAs following blunt and penetrating liver trauma.
METHODS
Ethics statement
Ethical approval was granted by the local audit and trauma research committees (St Mary's Hospital, Imperial College Healthcare NHS Trust) to analyze data from the institutional trauma database. The requirement for informed consent was waived due to the use of deidentified data and the retrospective nature of the study.
Study design
A prospectively maintained database of trauma patients treated at a London major trauma center (level 1) was reviewed. A retrospective analysis was conducted of patients aged <18 years who were admitted with traumatic liver injuries between 2017 and 2022. Patients with blunt or penetrating liver injuries (AAST grades I–V) identified on their index trauma CT scan were included. Hemodynamically stable patients without evidence of arterial bleeding on the initial CT were managed nonoperatively. Two nonblinded radiologists independently determined the AAST grade; in cases of discrepancy, imaging was reviewed by a third nonblinded radiologist to confirm the final grade. Data regarding patient demographics, ISS, mechanism of injury, and clinical interventions were obtained from the trauma registry and the radiology information system.
Trauma pathway
At our major trauma center, separate pathways exist for the management of splenic trauma in adults and pediatric patients in order to minimize radiation exposure in children (Fig. 1). In contrast, liver trauma is managed similarly in both cohorts. Distinct management pathways are followed for blunt and penetrating liver injuries (Fig. 2).
Local trauma pathway for blunt and penetrating splenic trauma in (A) adults and (B) pediatric patients. IR, interventional radiologists; CT, computed tomography.
Local trauma pathway for (A) blunt liver injury and (B) penetrating liver injury. CT, computed tomography. a)After surgery if clinically indicated the patient should go straight to interventional radiology for hepatic angiogram with or without embolization.
Patients were managed in accordance with Advanced Trauma Life Support guidelines. A validated imaging protocol was used, whereby patients presenting with blunt injury underwent combi-CT (biphasic arterial and portal venous intravenous contrast injection with a single acquisition). Patients with penetrating injury underwent dual-phase CT (separate arterial and portal venous phase acquisitions) due to a high index of suspicion for vascular injury [14]. Clinically unstable children with penetrating liver injury underwent triple-phase abdominal CT on postoperative day 3, followed by liver ultrasound at 6 weeks.
Surgical management followed damage control principles, with either definitive liver repair or packing and planned relook laparotomy at 24 to 48 hours, depending on hemodynamic status. Nonoperative management consisted of clinical observation with serial examinations and endovascular embolization when clinically indicated.
Outcomes
The primary outcome was the incidence of HPAs identified on follow-up CT performed 48 to 72 hours after injury. Secondary outcomes included the timing of repeat CT imaging and the management of diagnosed HPAs.
Statistical analysis
Categorical variables were summarized as counts and percentages. Continuous variables were reported as mean±standard deviation with 95% confidence intervals. Medians with interquartile ranges (IQRs) were calculated to provide a more granular sense of the data distribution. The independent t-test was used due to the relatively small sample size (n=34). Data collection and statistical analyses were performed using Microsoft Excel (Microsoft Corp).
RESULTS
Between June 2017 and October 2022, 34 pediatric trauma patients aged <18 years (mean age, 12.0±5.4 years) presented with liver injury identified on initial CT. There were 26 men (76.5%) and 8 women (23.5%). Twenty-three patients (67.6%) sustained blunt liver injury, whereas 11 (32.4%) sustained penetrating liver injury (Table 1). Fourteen patients (41.2%) underwent primary surgical management following initial imaging. The median ISS for the entire cohort was 26 (IQR, 16–34). The blunt liver injury subgroup had a higher median ISS of 32 (IQR, 22–42), whereas the penetrating injury subgroup had a lower median ISS of 17.5 (IQR, 10.75–26.75). Initial CT-based AAST liver injury grades are presented in Table 2.
Summary of the mechanism of injury (n=34)
The AAST liver injury score on the CT scan performed upon initial admission CT scan
The overall mortality rate was 11.8% (n=4). Three patients died from catastrophic intra-abdominal hemorrhage with trauma-induced coagulopathy at trauma laparotomy, and one patient died due to concurrent internal decapitation. These patients were excluded from subsequent analysis. Of the 30 patients, 18 (60.0%) underwent repeat CT imaging. Of the 12 patients who did not undergo repeat CT, 6 (50.0%) had follow-up ultrasonography at 6 weeks. Nine patients (30.0%) underwent repeat CT within 48 to 72 hours of admission, whereas the other nine (30.0%) underwent imaging outside this interval. The mean time to follow-up CT was 83.8±48.5 hours. Four HPAs (13.3%) were identified on repeat imaging. Two HPAs (6.7%), associated with AAST grade III and IV injuries, respectively, were treated with embolization, whereas the remaining two (AAST grades III and IV) were managed conservatively (Table 3). One patient who underwent embolization had previously undergone trauma laparotomy for a grade IV liver laceration, with subsequent HPA identification on repeat CT prompting embolization. All 6-week follow-up ultrasound examinations were negative. Decisions regarding intervention were made following multidisciplinary discussion between the trauma and interventional radiologist teams. Hemodynamically stable children without hemoglobin decline were managed conservatively. No procedure-related complications were recorded following HPA embolization.
Characteristics of detected HPAs and associated management
DISCUSSION
Hepatic pseudoaneurysm formation represents a potentially serious complication of both adult and pediatric liver trauma and may lead to significant intra-abdominal hemorrhage. The pediatric population poses unique challenges in diagnosis and management, primarily due to concerns regarding radiation exposure and the absence of standardized surveillance protocols. Furthermore, clinical presentation in children may be nonspecific, and liver function tests lack adequate sensitivity and specificity for detecting vascular complications. Consequently, no consensus exists regarding optimal imaging strategies. The true incidence of HPA in children with liver trauma remains uncertain, partly due to variation in diagnostic practices and possible under- or over-reporting. Some studies report incidences of up to 25% [13], comparable to rates described in adult populations.
Although no international guidelines specifically address pediatric liver trauma surveillance, some authors suggest that routine repeat imaging is unnecessary and should be considered only when clinical deterioration occurs [15]. We propose that identification of HPAs using imaging modalities that minimize or eliminate radiation exposure may help prevent clinical deterioration while limiting long-term risk. CEUS has emerged as a promising modality, with several studies demonstrating high sensitivity and specificity, as well as real-time visualization of blood flow, with diagnostic accuracy approaching that of CT [16]. Durkin et al. [3] identified 14 HPAs among 57 patients with liver injury using CT and CEUS; 5 required embolization, and CEUS was predominantly used for follow-up imaging. CEUS reduces radiation exposure and may therefore be particularly advantageous in children. Safavi et al. [12] used Doppler ultrasound 5 to 7 days after blunt liver trauma and identified 3 HPAs among 176 patients. In contrast, Navarro et al. [4] did not identify any HPAs among 66 patients during follow-up periods ranging from 6 days to 14 months.
In the present study, all four patients in whom HPAs were identified had AAST grade III or IV injuries, supporting previous findings that HPA formation is more frequently observed in high-grade liver injuries [3,9,12]. However, other studies have reported HPA formation in lower-grade injuries [17]. Therefore, follow-up imaging cannot be restricted solely to a specific injury grade. These findings suggest that imaging modalities that avoid ionizing radiation, such as CEUS, may represent a valuable alternative to CT. In cases of isolated low-grade hepatic injuries with low clinical suspicion, CT angiography may reasonably be avoided. In our cohort, no HPAs were identified in patients with AAST grade I or II injuries; thus, the risks associated with CT surveillance may outweigh potential benefits in this subgroup. Additionally, no clear association was observed between abdominal symptoms and subsequent HPA detection, indicating that symptomatology alone may be insufficient to guide imaging decisions.
Another strategy to reduce radiation exposure involves modification of CT protocols. The standard protocol at our institution involves triple-phase liver CT, requiring three abdominal acquisitions. However, evidence suggests that dual-phase CT does not reduce sensitivity or diagnostic accuracy in detecting blunt splenic injury [18]. Further research is required to better characterize HPA pathophysiology in both blunt and penetrating trauma. Specifically, future studies should examine lesion size, natural history (including spontaneous resolution), indications for intervention, and the potential role of primary embolization at initial imaging. Not all trauma centers have established rescanning protocols, and comparative morbidity and mortality data remain limited. One experimental study developed a rabbit model (n=54) of traumatic femoral pseudoaneurysm and demonstrated that pseudoaneurysms formed within 5 to 11 days of injury, enlarging for up to 4 weeks. However, 7.4% of animals died from hemorrhage during the 12-week study period [19].
Limitations
There are several limitations to this study. First, the mean interval for repeat imaging (approximately 84 hours) exceeded the institution’s intended 48- to 72-hour timeframe. Second, follow-up data were incomplete, as some patients may have attended local hospitals for subsequent care; therefore, later clinical events may not be captured in our dataset. Third, the small sample size (n=34) reflects the relatively low incidence of pediatric traumatic liver injury and precluded meaningful regression analysis evaluating associations between ISS, AAST grade, and HPA formation. Finally, only four HPAs were identified; therefore, conclusions should be interpreted with caution.
Conclusions
In this cohort, the incidence of HPA following pediatric liver trauma was low. Although the absence of standardized surveillance protocols presents a challenge, repeat imaging may facilitate early detection and management of HPAs, potentially preventing clinical deterioration. CEUS may optimize HPA detection while eliminating radiation exposure; however, current evidence remains insufficient to support its routine implementation. Further research is needed to refine diagnostic and therapeutic strategies and to improve outcomes for children affected by this potentially life-threatening complication of abdominal trauma.
Notes
Author contributions
Conceptualization: NRP, MAW, EK, NA; Data curation: JG, MB, RH, CB; Formal analysis, JG, MB, RH, KR, SC; Methodology: JG, NRP, MAW; Project administration: JG, NRP; Supervision: NRP, MAW, EK, NA; Visualization: JG, CB; Writing–original draft: JG, CB, KR, SC; Writing–review & editing: all authors. All authors read and approved the final manuscript.
Conflicts of interest
The authors have no conflicts of interest to declare.
Funding
The authors received no financial support for this study.
Acknowledgments
The authors thank Christopher Lynch (Imperial College Healthcare NHS Trust) for providing TARN (Trauma Audit Research Network) database access.
Data availability
Data analyzed in this study are available from the corresponding author upon reasonable request.
Additional information
This study was presented at the 2024 Association of Surgeons of Great Britain and Ireland (ASGBI) International Surgical Congress on May 8–10, 2024, in Belfast, UK.