High incidence of surgical site infection may be related to suboptimal case selection for non-selective arterial embolization during resuscitation of patients with pelvic fractures: a retrospective study

Background In most institutions, arterial embolization (AE) remains a standard procedure to achieve hemostasis during the resuscitation of patients with pelvic fractures. However, the actual benefits of AE are controversial. In this study, we aimed to explore AE-related outcomes following resuscitation at our center and to assess the predictive value of contrast extravasation (CE) during computed tomography (CT) for patients with hemodynamically unstable closed pelvic fractures. Methods We retrospectively reviewed data from patients who were treated for closed pelvic fractures at a single center between 2014 and 2017. Data regarding the AE and clinical parameters were analyzed to determine whether poor outcomes could be predicted. Results During the study period, 545 patients were treated for closed pelvic fractures, including 131 patients who underwent angiography and 129 patients who underwent AE. Nonselective bilateral internal iliac artery embolization (nBIIAE) was the major AE strategy (74%). Relative to the non-AE group, the AE group had higher values for injury severity score, shock at hospital arrival, and unstable fracture patterns. The AE group was also more likely to require osteosynthesis and develop surgical site infections (SSIs). Fourteen patients (10.9%) experienced late complications following the AE intervention, including 3 men who had impotence at the 12-month follow-up visit and 11 patients who developed SSIs after undergoing AE and osteosynthesis (incidence of SSI: 11/75 patients, 14.7%). Nine of the 11 patients who developed SSI after AE had undergone nBIIAE. The positive predictive value of CE during CT was 29.6%, with a negative predictive value of 91.3%. Relative to patients with identifiable CE, patients without identifiable CE during CT had a higher mortality rate (30.0% vs. 11.0%, p = 0.03). Conclusion Performing AE for pelvic fracture-related hemorrhage may not be best practice for patients with no CE detected during CT or for unstable patients who do not respond to resuscitation after exclusion of other sources of hemorrhage. Given the high incidence of SSI following nBIIAE, this procedure should be selected with care. Given their high mortality rate, patients without CE during imaging might be considered for other hemostasis procedures, such as preperitoneal pelvic packing.


Background
Pelvic fractures can often result in potentially lethal hemorrhage and hypotension [1,2]. Thus, hemodynamically unstable patients with pelvic fractures present a major challenge for acute care and orthopedic surgeons, and such patients are best managed by a multi-disciplinary team [3]. However, the appropriate acute management of hemodynamically unstable patients with pelvic fractures remains controversial. Arterial embolization (AE) and preperitoneal pelvic packing (PPP) are common procedures in this setting, although there is no conclusive evidence for the superiority of either procedure [2,4,5].
Some reports have described AE as an effective procedure for managing pelvic fracture-related retroperitoneal arterial hemorrhage [6,7]. AE is a classical treatment and remains a standard hemostatic procedure in many institutions [7][8][9], including our institution, based on reports regarding its efficacy during the management of pelvic fracture-related hemorrhage [8,10]. However, AE is associated with various complications, including inadequate hemostasis, gluteal muscle necrosis, and surgical site infection (SSI) [11,12]. Therefore, as an observation of contrast extravasation (CE) during computed tomography (CT) [13] is unreliable as a sole basis for selecting AE, additional information is needed to help determine when AE can help provide optimal outcomes. This study aimed to explore AE-related outcomes following resuscitation at our center and to assess the predictive value of CE during CT for patients with hemodynamically unstable closed pelvic fractures.

Methods
At our institution, AE is a standard procedure for pelvic fracture-related hemorrhage and is performed when CE is detected during contrast-enhanced CT or for unstable patients who do not respond to resuscitation after the exclusion of other sources of hemorrhage. Angiography is initially performed to evaluate the abdominal aorta, lumbar arteries, bilateral common iliac arteries, bilateral external iliac arteries, and bilateral internal iliac arteries. When arterial bleeding is detected or suspected, the first embolization material is generally gelfoam (SPONGOSTAN, Ethicon Inc., Belgium), and bleeding is re-checked 5 min later. If an aneurysm persists or CE recurs, permanent embolization is performed using metal coils (Tornado Embolization Coils, Cook Inc., USA or VortX-18 Coils, Boston Scientific, USA). Selective or non-selective embolization is performed at the discretion of the interventional radiologist. After any AE, the patient is admitted to the intensive care unit or the ward, depending on the clinical requirements. Osteosynthesis is then performed for pelvic ring injuries after the patient's clinical condition has stabilized. During the study period, resuscitative endovascular balloon occlusion of the aorta was not available for managing these patients at our center.
We retrospectively reviewed trauma registry data, including clinical and imaging records, from all cases of pelvic fracture that were treated at our center between January 2014 and December 2017. The study's retrospective protocol was approved by our institutional review board (IRB: 201802139B0). Patients were excluded if they had an isolated acetabular fracture, were dead upon arrival, were diagnosed with a pelvic fracture without imaging, or had undergone AE as a hemostatic procedure targeting non-pelvic regions. All patients with pelvic fractures were resuscitated and managed according to our established protocol, which was based on the Advanced Trauma Life Support guidelines.
The patients' demographic characteristics, injury severity score (ISS), fracture patterns, AE details, and AErelated complications were recorded. Complications were defined as adverse responses or unexpected conditions that were likely related to the AE procedure. We defined an SSI as a post-osteosynthesis wound complication, which yielded a positive bacterial culture. Information regarding impotence in men was self-reported at the 12-month follow-up visit.
Data were analyzed using SPSS software (version 18.0; SPSS Inc., Chicago, IL). Continuous variables were compared using the t-test, and categorical variables were compared using the chi-squared test and Fisher's exact test. Predictive analyses were performed to determine whether the presence of CE during CT could be used to predict clinical outcomes.

Results
During the 48-month study period, we identified 545 patients with closed pelvic fractures who underwent resuscitation and treatment (Table 1). Traffic accident was the most common injury mechanism (70.2%). Initial clinical assessments revealed that 112 patients (20.6%) were in shock at the initial presentation and required resuscitation and blood transfusions. After resuscitation, 49 patients (43.8%) were identified as responders, and the remaining patients did not initially respond to resuscitation. The CT assessments revealed that 390 patients (71.5%) had partially stable or unstable ring patterns. Angiography was performed for 131 patients (24%) who were unresponsive to resuscitation or who had CE during the CT examination. There were no complications during or immediately after the angiography procedure, and AE was performed for 129 of the 131 patients (98.5%) who underwent angiography ( Table 2).
The average time from the order to the angiography procedure was 63.8 ± 44.7 min. Ninety-seven patients (74.0%) underwent non-selective bilateral internal iliac artery embolization (nBIIAE) using gelfoam as the filling material. Only 11 of the 129 patients (8.5%) underwent AE using metal coils. Fourteen patients (10.9%) experienced late complications after the AE intervention, including 3 men who had impotence at the 12-month follow-up visit and 11 patients who developed SSIs after the AE and osteosynthesis (Fig. 1). The incidence of SSIs among patents who underwent AE and osteosynthesis was 14.7% (11/75 patients). Among the 11 patients, 6 had Morel-Lavallée lesions, and 2 had open fractures   (Table 3). Table 4 shows comparisons between patients who did and did not undergo AE. Patients who underwent AE had a higher ISS (26.1 vs. 14.5, p = 0.01), a greater likelihood of being in shock upon arrival (68.2% vs. 5.3%, p = 0.001), a higher likelihood of having an unstable pelvic fracture pattern (55.9% vs. 19.9%, p = 0.001), and a higher likelihood of undergoing osteosynthesis (57.3% vs. 32.9%, p = 0.001). The AE group also had significantly higher rates of impotence in men (4.8% vs. 0.6%, p = 0.036) and SSIs (14.7% vs. 3.7%, p = 0.006).
Among the 129 patients who underwent AE, 109 patients (84.5%) had evidence of arterial bleeding at the time of the CT scan. Therefore, the value of AE was evaluated on the basis the characteristics of patients with and without identifiable CE during CT (Table 5). No significant inter-group differences were detected in initial shock status, response to resuscitation, age, sex, ISS, time to AE, fracture classification, volume of blood transfusion, or rate of repeat AE. However, patients without identifiable CE during CT or angiography had a significantly higher 1-week in-hospital mortality rate than patients with identifiable CE (30.0% vs. 11.0%, p = 0.03).
One of the indications for AE in our institution was the presence of CE during CT. Therefore, we examined the relationship between identifiable bleeding during CT or during angiography (the standard for comparison) to evaluate the effectiveness of CT for assessing arterial bleeding in patients with unstable pelvic fractures. We used the angiography results as a standard baseline to examine the predictive value of CT examination. All 131 patients who underwent angiography had undergone CT examination upon their emergency department arrival. The CT examination revealed retroperitoneal bleeding in 108/131 patients (82.4%), although CE during angiography was only identified in 34/131 patients (26%). Among the 34 patients with CE during angiography, 32 patients (94.2%) had CE during the CT examination. Predictive analysis revealed that CE during CT had a sensitivity of 94.1%, a specificity of 21.6%, a positive predictive value (PPV) of 29.6%, and a negative predictive value (NPV) of 91.3%.

Discussion
Pelvic fractures may be accompanied by hemorrhage, which can lead to hemodynamic instability and death. At our institution, AE intervention is a standard procedure in this setting and is associated with good bleeding control outcomes [8,10]. Furthermore, delayed angiography in these patients may be associated with poor outcomes [8,15,16]. Technological improvements and increased availability have led to an increasing number of vascular injuries being identified during CT examinations, although recent studies have indicated that imaging-based evidence of CE does not necessarily indicate a need for AE to control bleeding [4,[17][18][19]. Thus, Fig. 1 A case with surgical site infection following trans-arterial embolization and osteosynthesis for pelvic ring injury. a Computed tomography revealed an AO/OTA B3.1 pelvic ring injury. b Bilateral internal iliac artery trans-arterial embolization was performed using gelfoam (right side) and metal coil (left side). c Radiographic findings after the osteosynthesis. d Significant bilateral gluteus maximus necrosis was observed at 1 month after the trans-arterial embolization. e Final appearance after treatment for the soft-tissue defect hasty AE following pelvic fracture might not be an optimal step during the busy resuscitation process. Although AE can be promptly performed at our center (in a mean time of 60 min from order to procedure), the main diagnostic tool was CT, which had a low PPV, and the high incidence of SSIs following nBIIAE was suboptimal. For example, 82.4% of patients in the AE group exhibited signs of arterial bleeding during the CT examination, while only 26% of patients exhibited signs of CE during angiography. This corresponds to a PPV of only 29.6%  for the use of CT to detect arterial injury in patients with pelvic fractures, which is similar to the results from previous studies [13,20]. We suspect that it may be difficult to differentiate between bone marrow and arterioles as the source of contrast extravasation during CT examination soon after a pelvic fracture, and our results suggest that evidence of CE during CT may not be a suitable major indication for AE, despite this being previously recommended [7,21]. Dreizin et al. [22] have suggested that the use of AE should be guided instead by the hematoma volume when arterial bleeding is detected during CT. We believe that angiography and AE should be performed more selectively, perhaps on the basis of non-response to resuscitation, a continuing decrease in hemoglobin concentration without evidence of bleeding from other sources, and the absence of arterial blush during CT. Non-selective embolization is considered a hemostatic procedure for hemodynamically unstable patients with pelvic fractures [7,23], although recent studies have focused on selective or super-selective embolization to decrease the negative effects of non-selective embolization in cases of vasospasm or rich collateral arterial networks within the pelvis [24,25]. Thus, we evaluated the outcomes of different AE strategies, given the fact that AE interventions might be accompanied by a high complication rate after these patients undergo osteosynthesis to treat their unstable pelvis. The present study revealed that 74% of patients underwent nBIIAE as their AE intervention, and 9 of 11 patients who developed SSIs after AE and osteosynthesis had undergone nBIIAE. Local soft tissue injuries after high-energy trauma, such as Morel-Lavallée lesions, might have contributed to the incidence of SSIs, although it would also be prudent to consider the potential contribution of nBIIAE.
Contrast-enhanced CT scans have become routine at many institutions, including ours, to detect arterial bleeding within the pelvis [26]. However, relative to angiography as the gold standard, CT had a low PPV and a high NPV for detecting arterial bleeding. In addition, we observed that patients without signs of CE during imaging had a significantly higher mortality rate after AE, relative to patients with signs of CE during CT (30.0% vs. 11.0%, p = 0.03). Moreover, there were no significant differences between the two groups in age, sex, ISS, time to AE, fracture classification, volume of blood transfusion, or rate of repeat AE. We speculate that patients without signs of CE during CT would have been experiencing shock due to causes other than arterial injury, such as cancellous fracture surface bleeding or venous plexus bleeding. As a result, AE may have provided little benefit in these patients, while delaying the resuscitation procedure and potentially increasing the risk of mortality. Therefore, AE should not be a routine procedure for hemodynamically unstable patients with unstable pelvic fractures, and other hemostasis procedures, such as PPP, should be considered during the resuscitation process for these patients.
Although we made every effort to reduce the influence of bias in this study, it has several limitations. First, the retrospective design is inherently associated with risks of bias, and we observed that cases that underwent AE were more severe and had higher ISS scores than the non-AE cases. Thus, the higher complication rate in the AE group might reflect the severity of the patients' injuries, and strict selection criteria may be needed for the AE procedure to avoid further complications that are difficult to manage. Second, although our center uses an established protocol to identify patients who should undergo angiography, the choice of selective or nonselective embolization was at the radiologist's discretion. The interventional radiologists commonly noted that non-selective embolization was performed "to facilitate hemostasis," even when no arterial bleeding was identified during angiography. Third, AE has been performed at our center during the last two decades as part of a fixed protocol for hemodynamically unstable patients with pelvic fractures, and it may be difficult to encourage experienced emergency traumatologists to include PPP as an alternative strategy in our resuscitation protocol.

Conclusions
In conclusion, we believe that angioembolization is an effective, time-saving, and readily available procedure for pelvic fracture-related arterial hemorrhage. However, given the high rate of SSIs, non-selective AE may not be the best practice in this setting when there is no sign of CE during CT or when the patient does not respond to resuscitation after other sources of hemorrhage have been excluded. It may be appropriate to perform more selective or super-selective embolization procedures in carefully selected patients and to consider PPP as an alternative hemostasis strategy in the resuscitation protocol when there are no imaging-based signs of arterial bleeding.