Extra-parenchymal splenic abnormalities not vascular injury predict need for primary splenectomy.

PURPOSE: Radiographic indications for primary splenectomy (PS) in blunt splenic injury (BSI) after radiographic diagnosis are unknown. Improved understanding of radiographic characteristics of patients requiring splenectomy will help to appropriately triage patients to PS or non-operative management (NOM). METHODS: A retrospective, single-center review was performed of BSI diagnosed with computerized tomography (CT). Patients undergoing splenectomy prior to CT diagnosis were excluded. RESULTS: BSI was identified in 195 patients. On logistic regression, only subcapsular hematoma presence (OR 7.521, p = 0.002) and left upper quadrant hemoperitoneum (OR 6.146, p = 0.03) were associated with need for PS, while splenic laceration length, number of pseudoaneurysms (PSA), and active contrast extravasation (NS for all) were not. CONCLUSIONS: Need for PS is predicted by extra-parenchymal pathology in subcapsular hematoma and hemoperitoneum. Splenic vascular injuries through PSA and active contrast extravasation do not predict the need for PS and can be considered for NOM.

Natural history of splenic vascular abnormalities after blunt injury: A Western Trauma Association multicenter trial.

BACKGROUND: Following blunt splenic injury, there is conflicting evidence regarding the natural history and appropriate management of patients with vascular injuries of the spleen such as pseudoaneurysms or blushes. The purpose of this study was to describe the current management and outcomes of patients with pseudoaneurysm or blush. METHODS: Data were collected on adult (aged ≥18 years) patients with blunt splenic injury and a splenic vascular injury from 17 trauma centers. Demographic, physiologic, radiographic, and injury characteristics were gathered. Management and outcomes were collected. Univariate and multivariable analyses were used to determine factors associated with splenectomy. RESULTS: Two hundred patients with a vascular abnormality on computed tomography scan were enrolled. Of those, 14.5% were managed with early splenectomy. Of the remaining patients, 59% underwent angiography and embolization (ANGIO), and 26.5% were observed. Of those who underwent ANGIO, 5.9% had a repeat ANGIO, and 6.8% had splenectomy. Of those observed, 9.4% had a delayed ANGIO, and 7.6% underwent splenectomy. There were no statistically significant differences between those observed and those who underwent ANGIO. There were 111 computed tomography scans with splenic vascular injuries available for review by an expert trauma radiologist. The concordance between the original classification of the type of vascular abnormality and the expert radiologist's interpretation was 56.3%. Based on expert review, the presence of an actively bleeding vascular injury was associated with a 40.9% risk of splenectomy. This was significantly higher than those with a nonbleeding vascular injury. CONCLUSIONS: In this series, the vast majority of patients are managed with ANGIO and usually embolization, whereas splenectomy remains a rare event. However, patients with a bleeding vascular injury of the spleen are at high risk of nonoperative failure, no matter the strategy used for management. This group may warrant closer observation or an alternative management strategy. LEVEL OF EVIDENCE: Prognostic study, level III.

Indirect signs of blunt duodenal injury on computed tomography: Is non-operative management safe?

INTRODUCTION: Clear signs of duodenal injury (DI) such as pneumoperitoneum and/or oral contrast extravasation mandate laparotomy. Management when computed tomography (CT) reveals indirect evidence of DI namely duodenal hematoma or periduodenal fluid is unclear. We evaluated the utility of indirect signs to identify DI and the success of expected management, hypothesizing patients with indirect evidence of DI on CT can be safely managed non-operatively. METHODS: We retrospectively reviewed patients with a computed tomography (CT) scan with periduodenal hematoma or periduodenal fluid treated between January 2003 and January 2013 at a level 1 Trauma Center. Demographics, injury characteristics, laboratory values, injury severity scores (ISS), and outcome measures were recorded. Patients having immediate laparotomy were compared to those initially managed nonoperatively. RESULTS: We identified 74 patients with indirect signs of DI, with 35 patients (47%) undergoing immediate operative exploration and 39 (53%) initially managed non-operatively. Lactate (4.5 mg/dL, standard deviation (SD) 2.1 vs 3.1 mg/dL, SD 1.4, p<0.001), ISS (median (IQR) 34 (27-44) vs. 24 (17-34), p=0.002) and abdominal AIS (3 (3-4) vs 2 (2-3), p<0.001) were higher in those with immediate operation. The incidence of DI requiring operative repair was 11% (8 of 74). Six of 35 (17%) explored urgently had a DI requiring repair while 29 of 35 (83%) had no DI or minor injury not requiring surgical therapy. Of those managed non-operatively, 7 of 39 (18%) failed observation but only two (5%) required duodenal repair. There was no significant difference in intensive care unit (ICU) (10.2 days, standard error [SE] 2.1 vs 9.7 days, SE 4.8, p=0.93) or hospital (22.5 days, SE 3.8 vs 23.6 days, SE 8.5, p=0.91) length of stay between those operated on immediately and those that failed non-operative management when adjusted for age, sex, and ISS. There was no mortality in the non-operative group related to an intra-abdominal injury. CONCLUSION: Observation of patients with indirect sign of DI fails in about 20% of patients, but failure rate due to DI is low at 5%. Conservative management in appropriately selected patients is reasonable with close observation.

Disruptions in Resting State Functional Connectivity and Cerebral Blood Flow in Mild Traumatic Brain Injury Patients.

Mild traumatic brain injury (mTBI) is often occult to conventional imaging techniques. However, there is growing evidence that mTBI patients who lack evidence of structural intracranial injury may develop post-concussive syndrome (PCS). We investigated longitudinal alterations in resting state functional connectivity (rs-FC) in brain networks in a population of 28 patients compared to 28 matched control participants. Rs-FC and cerebral blood flow (CBF) within the nodes of the Default Mode Network (DMN) and Task Positive Network (TPN) were assessed at three time points including acute, sub-acute, and chronic stages following mTBI. Participants received the Automated Neuropsychological Assessment Metrics (ANAM) to assess cognitive performance. Main findings indicate that despite normalized cognitive performance, chronic mTBI patients demonstrate increased rs-FC between the DMN and regions associated with the salience network (SN) and TPN compared to the control populations, as well as reduced strength of rs-FC within the DMN at the acute stage of injury. In addition, chronic mTBI patients demonstrate an imbalance in the ratio of CBF between nodes of the DMN and TPN. Furthermore, preliminary exploratory analysis suggests that compared to those without chronic PCS, patients with chronic PCS reveal an imbalance in the ratio of CBF between the DMN nodes and TPN nodes across multiple stages of recovery. Findings suggest that the altered network perfusion with the associated changes in rs-FC may be a possible predictor of which mTBI patients will develop chronic PCS.

Predictors of intramedullary lesion expansion rate on MR images of patients with subaxial spinal cord injury.

OBJECT Studies of preclinical spinal cord injury (SCI) in rodents indicate that expansion of intramedullary lesions (IMLs) seen on MR images may be amenable to neuroprotection. In patients with subaxial SCI and motor-complete American Spinal Injury Association (ASIA) Impairment Scale (AIS) Grade A or B, IML expansion has been shown to be approximately 900 µm/hour. In this study, the authors investigated IML expansion in a cohort of patients with subaxial SCI and AIS Grade A, B, C, or D. METHODS Seventy-eight patients who had at least 2 MRI scans within 6 days of SCI were enrolled. Data were analyzed by regression analysis. RESULTS In this cohort, the mean age was 45.3 years (SD 18.3 years), 73 patients were injured in a motor vehicle crash, from a fall, or in sport activities, and 77% of them were men. The mean Injury Severity Score (ISS) was 26.7 (SD 16.7), and the AIS grade was A in 23 patients, B in 7, C in 7, and D in 41. The mechanism of injury was distraction in 26 patients, compression in 22, disc/osteophyte complex in 29, and Chance fracture in 1. The mean time between injury onset and the first MRI scan (Interval 1) was 10 hours (SD 8.7 hours), and the mean time to the second MRI scan (Interval 2) was 60 hours (SD 29.6 hours). The mean IML lengths of the first and second MR images were 38.8 mm (SD 20.4 mm) and 51 mm (SD 36.5 mm), respectively. The mean time from the first to the second MRI scan (Interval 3) was 49.9 hours (SD 28.4 hours), and the difference in IML lengths was 12.6 mm (SD 20.7 mm), reflecting an expansion rate of 366 µm/ hour (SD 710 µm/hour). IML expansion in patients with AIS Grades A and B was 918 µm/hour (SD 828 µm/hour), and for those with AIS Grades C and D, it was 21 µm/hour (SD 304 µm/hour). Univariate analysis indicated that AIS Grade A or B versus Grades C or D (p < 0.0001), traction (p= 0.0005), injury morphology (p < 0.005), the surgical approach (p= 0.009), vertebral artery injury (p= 0.02), age (p < 0.05), ISS (p < 0.05), ASIA motor score (p < 0.05), and time to decompression (p < 0.05) were all predictors of lesion expansion. In multiple regression analysis, however, the sole determinant of IML expansion was AIS grade (p < 0.005). CONCLUSIONS After traumatic subaxial cervical spine or spinal cord injury, patients with motor-complete injury (AIS Grade A or B) had a significantly higher rate of IML expansion than those with motor-incomplete injury (AIS Grade C or D).

Associations between interhemispheric functional connectivity and the Automated Neuropsychological Assessment Metrics (ANAM) in civilian mild TBI.

This study investigates cognitive deficits and alterations in resting state functional connectivity in civilian mild traumatic brain injury (mTBI) participants with high and low symptoms. Forty-one mTBI participants completed a resting state fMRI scan and the Automated Neuropsychological Assessment Metrics (ANAM) during initial testing (<10 days of injury) and a 1 month follow up. Data were compared to 30 healthy control subjects. Results from the ANAM demonstrate that mTBI participants performed significantly worse than controls on the code substitution delayed subtest (p = 0.032). [corrected]. Among the mTBI patients, high symptom mTBI participants performed worse than those with low symptoms on the code substitution delayed (p = 0.017), code substitution (p = 0.012), repeated simple reaction time (p = 0.031), and weighted throughput score (p = 0.019). [corrected]. Imaging results reveal that during the initial visit, low symptom mTBI participants had reduced interhemispheric functional connectivity (IH-FC) within the lateral parietal lobe (p = 0.020); however, during follow up, high symptom mTBI participants showed reduced IH-FC compared to the control group within the dorsolateral prefrontal cortex (DLPFC) (p = 0.013). Reduced IH-FC within the DLPFC during the follow-up was associated with reduced cognitive performance. Together, these findings suggest that reduced rs-FC may contribute to the subtle cognitive deficits noted in high symptom mTBI participants compared to control subjects and low symptom mTBI participants.

A longitudinal evaluation of diffusion kurtosis imaging in patients with mild traumatic brain injury.

PRIMARY OBJECTIVE: To investigate longitudinal diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI) changes in white and grey matter in patients with mild traumatic brain injury (mTBI). RESEARCH DESIGN: A prospective case-control study. METHODS AND PROCEDURES: DKI data was obtained from 24 patients with mTBI along with cognitive assessments within 10 days, 1 month and 6 months post-injury and compared with age-matched control (n» 24). Fractional anisotropy (FA), mean diffusivity (MD), radial diffusion (l(r)), mean kurtosis (MK) and radial kurtosis (Kr) were extracted from the thalamus, internal capsule and corpus callosum. MAIN OUTCOMES AND RESULTS: Results demonstrate reduced Kr and MK in the anterior internal capsule in patients with mTBI across the three visits, and reduced MK in the posterior internal capsule during the 10 day time point. Correlations were observed between the change in MK or Kr between 1­6 months and the improvements in cognition between the 1 and 6 month visits in the thalamus, internal capsule and corpus callosum. CONCLUSIONS: These data demonstrate that DKI may be sensitive in tracking pathophysiological changes associated with mTBI and may provide additional information to conventional DTI parameters in evaluating longitudinal changes following TBI.

Longitudinal and prognostic evaluation of mild traumatic brain injury: A 1H-magnetic resonance spectroscopy study.

In the majority of patients with mild traumatic brain injury (mTBI), brain tissue impairment is undetectable by computed tomography and/or structural magnetic resonance imaging. Even in confirmed cases of head injury, conventional neuroimaging methods lack sensitivity in predicting neuropsychological outcomes of patients. The objectives of this study were to (1) cross-sectionally determine deviations in the neurometabolic profile of patients with mTBI from healthy controls at different stages of mTBI using tightly controlled examination windows, and (2) determine associations between acute neurometabolic markers of mTBI and chronic neurocognitive performance. Patients were examined at the early subacute (n=43; 5.44 ± 3.15 days post-injury (DPI)), late subacute (n=33; 37.00 ± 12.26 DPI) and chronic (n=27; 195.30 ± 19.60 DPI) stages of mTBI. Twenty-one neurologically intact subjects were used as controls. Proton magnetic resonance spectroscopy imaging ((1)H-MRSI) was used to obtain metabolic measurements from different brain regions. The Automated Neuropsychological Assessment Metrics (ANAM) was used for cognitive evaluation of patients at the chronic stage of mTBI. Measurements in the thalamus and centrum semiovale (CSV) emerged as the most indicative of injury and were used to predict neurocognitive outcome. The major findings of this study are (1) decreases in Cho/Cre (choline-to-creatine ratio) measured in the thalamus (p=0.042) and CSV (p=0.017) at the late subacute stage of mTBI; (2) positive associations of early subacute Cre measurements in the CSV with chronic ANAM scores measuring performance in delayed (r=0.497, p=0.019) and immediate (r=0.391, p=0.072) code substitution. These findings show that metabolic measurements in the thalamus and CSV can potentially serve as diagnostic and prognostic markers of mTBI.

Comparative effectiveness of surgical versus nonoperative management of unilateral, nondisplaced, subaxial cervical spine facet fractures without evidence of spinal cord injury: clinical article.

OBJECT: Facet joints are major stabilizers of cervical motion allowing for effortless and pain-free multidimensional cervical spine movements without significant linear or rotational translation, thus minimizing any chance for spinal cord or nerve root impingement. Unilateral, nondisplaced subaxial facet fractures do not meet the conventional criteria for spinal instability under physiological loads. Limited evidence indicates that even with no or minimal displacement, 20%-80% of these fractures fail nonoperative management. The risk factors for instability in isolated nondisplaced subaxial facet fractures remain uncertain. In this retrospective study of prospectively collected data, the authors attempted to identify the predictors of failure in the management of isolated, nondisplaced subaxial facet fractures admitted to their Level I trauma center over a 10-year period. METHODS: Demographic, clinical, imaging, and follow-up data for 25 patients with unilateral nondisplaced subaxial facet fractures who were managed surgically (n = 10) or nonoperatively (n = 15) were statistically analyzed. RESULTS: The mean age of the patients was 38 years, 19 were male, and 21 of the fractures were the result of either motor vehicle accidents or falls. The mean motor score on the American Spinal Injury Association scale was 99.2, and the mean Subaxial Injury Classification (SLIC) severity score was 3 (operated 3.5, nonoperated 2.3). Allen mechanistic classification included 22 compressive-extension Stage 1 and 2 distractive-extension Stage 1 fractures. Subaxial facet fractures involved C-7 in 17 patients (68%), C-6 in 7 (28%), and C-3 in 1 (4%). The anatomical plane of fracture through the lateral mass was sagittal in 12 patients, axial in 8, and coronal in 3 patients. Nondisplaced floating lateral mass injuries were noted in 2 patients. The mean instability score, considering 7 components of the discoligamentous complex on MRI, was 3.2 (operated 3.6, nonoperated 3.0). Ten (40%) of 25 patients in this investigation did not have successful management, 9 in the nonoperated and 1 in the operated group (p = 0.018). Unsuccessful management was significantly greater in younger patients (p = 0.0008), possibly indicating selection bias (p = 0.07, Wilcoxon ranksum test). Fracture plane, instability, and SLIC scores did not play a significant role in treatment failure in this study. CONCLUSIONS: In this study, surgery was superior to nonoperative management of isolated, nondisplaced, or minimally displaced subaxial cervical spine facet fractures.

Default mode network interference in mild traumatic brain injury - a pilot resting state study.

In this study we investigated the functional connectivity in 23 Mild TBI (mTBI) patients with and without memory complaints using resting state fMRI in the sub-acute stage of injury as well as a group of control participants. Results indicate that mTBI patients with memory complaints performed significantly worse than patients without memory complaints on tests assessing memory from the Automated Neuropsychological Assessment Metrics (ANAM). Altered functional connectivity was observed between the three groups between the default mode network (DMN) and the nodes of the task positive network (TPN). Altered functional connectivity was also observed between both the TPN and DMN and nodes associated with the Salience Network (SN). Following mTBI there is a reduction in anti-correlated networks for both those with and without memory complaints for the DMN, but only a reduction in the anti-correlated network in mTBI patients with memory complaints for the TPN. Furthermore, an increased functional connectivity between the TPN and SN appears to be associated with reduced performance on memory assessments. Overall the results suggest that a disruption in the segregation of the DMN and the TPN at rest may be mediated through both a direct pathway of increased FC between various nodes of the TPN and DMN, and through an indirect pathway that links the TPN and DMN through nodes of the SN. This disruption between networks may cause a detrimental impact on memory functioning following mTBI, supporting the Default Mode Interference Hypothesis in the context of mTBI related memory deficits.

Intramedullary lesion expansion on magnetic resonance imaging in patients with motor complete cervical spinal cord injury.

OBJECT: The authors performed a study to determine if lesion expansion occurs in humans during the early hours after spinal cord injury (SCI), as has been established in rodent models of SCI, and to identify factors that might predict lesion expansion. METHODS: The authors studied 42 patients with acute cervical SCI and admission American Spinal Injury Association Impairment Scale Grades A (35 patients) and B (7 patients) in whom 2 consecutive MRI scans were obtained 3-134 hours after trauma. They recorded demographic data, clinical information, Injury Severity Score (ISS), admission MRI-documented spinal canal and cord characteristics, and management strategies. RESULTS: The characteristics of the cohort were as follows: male/female ratio 37:5; mean age, 34.6 years; and cause of injury, motor vehicle collision, falls, and sport injuries in 40 of 42 cases. The first MRI study was performed 6.8 ±2.7 hours (mean ± SD) after injury, and the second was performed 54.5 ± 32.3 hours after injury. The rostrocaudal intramedullary length of the lesion on the first MRI scan was 59.2 ± 16.1 mm, whereas its length on the second was 88.5 ± 31.9 mm. The principal factors associated with lesion length on the first MRI study were the time between injury and imaging (p = 0.05) and the time to decompression (p = 0.03). The lesion's rate of rostrocaudal intramedullary expansion in the interval between the first and second MRI was 0.9 ± 0.8 mm/hour. The principal factors associated with the rate of expansion were the maximum spinal cord compression (p = 0.03) and the mechanism of injury (p = 0.05). CONCLUSIONS: Spinal cord injury in humans is characterized by lesion expansion during the hours following trauma. Lesion expansion has a positive relationship with spinal cord compression and may be mitigated by early surgical decompression. Lesion expansion may be a novel surrogate measure by which to assess therapeutic effects in surgical or drug trials.

Prognostic value of diffusion tensor imaging parameters in severe traumatic brain injury.

Diffusion tensor imaging (DTI) has recently emerged as a useful tool for assessing traumatic brain injury (TBI). In this study, the prognostic value of the relationship between DTI measures and the clinical status of severe TBI patients, both at the time of magnetic resonance imaging (MRI), and their discharge to acute TBI rehabilitation, was assessed. Patients (n=59) admitted to the trauma center with severe closed head injuries were retrospectively evaluated after approval from the institution's institutional review board, to determine the prognostic value of DTI measures. The relationship of DTI measures, including apparent diffusion coefficient (ADC), fractional anisotropy (FA), axial (λ‖) and radial diffusivity (λ⊥) from the whole brain white matter, internal capsule, genu, splenium, and body of the corpus callosum, were compared with neurological status at MRI and at discharge to acute TBI rehabilitation. Whole brain white matter averages of ADC, λ‖, and λ⊥, and their coefficient of variation (CV) were significantly correlated with the Glasgow Coma Scale (GCS) score on the day of MRI. The average λ‖ was significantly correlated with GCS scores on the day of MRI in all measured brain regions. Outcomes were associated with whole brain white matter averages of ADC and λ‖, and the CVs of FA, ADC, λ‖, and λ⊥; and the averages and CVs of FA and λ‖ in all corpus callosum regions. The inclusion of regional and global DTI measures improved the accuracy of prognostic models, when adjusted for admission GCS score and age (p<0.05). Whole brain white matter and regional DTI measures are sensitive markers of TBI, and correlate with neurological status both at MRI and discharge to rehabilitation. The addition of DTI measures adjusted for age, gender, and admission GCS score significantly improved prognostic models.

Splenic trauma: what is new?

Evaluation for splenic injury is an important component of patient assessment after blunt abdominal trauma. Key imaging modalities include ultrasound, particularly for rapid identification of hemoperitoneum, and computed tomography (CT), which permits a more detailed and accurate determination of splenic integrity. Specific findings at contrast-enhanced multidetector CT (MDCT) should prompt the consideration of catheter angiography with arterial embolization as an adjunct to nonsurgical management. This article reviews the roles of imaging in the management of splenic trauma, illustrates the MDCT appearance of various splenic injuries, and discusses imaging-based indications for operative and angiographic intervention.

Accuracy of computed tomography (CT) scan in the detection of penetrating diaphragm injury.

BACKGROUND: The use of computed tomography (CT) to identify injury after penetrating torso trauma has become routine in the hemodynamically stable patient. The diaphragm has been a historically difficult structure to evaluate, however, and missed injuries to the diaphragm may result in significant morbidity. With the increasing use of multidetector row CT (MDCT), we hypothesized that CT would be an accurate detection modality to identify patients with diaphragm injuries. METHODS: We retrospectively reviewed the admission CT of consecutive patients admitted for penetrating injury to the torso during a 4-year period. The CT scans were reviewed and classified into three categories: positive (P), negative (N), or equivocal (Eq). Data from the medical records of these patients were abstracted to identify demographics, injury-specific data, length of stay, length of follow-up (LOFU), and operative findings. RESULTS: There were 803 patients who met inclusion criteria. Mechanism of injury was gunshot wound in 36% and stab wound in 64%. Mean length of stay was 4 days (+/-6.6) and mean length of follow-up was 43 days (+/-184). CT was read as P in 57, N in 710, and Eq in 36 patients. Diaphragm injury was detected in 67 patients overall and was excluded in 736. For the entire study population, sensitivity and specificity were calculated as 94.0% (95% CI = 88.4-99.7) and 95.9% (94.5-97.4) with an overall accuracy of 95.8% (94.4-97.2) if the CT scan was used to exclude diaphragm injury ([P and Eq] vs. N). Sensitivity and specificity were 82.1% (72.9-91.3) and 99.7% (99.4-100) if CT was used to detect diaphragm injury (P vs. [N and Eq]). One hundred and forty-eight patients underwent operative procedures in which the diaphragm was evaluated. Diaphragm injury was identified in 50 (38 P, 4 N, 8 Eq) and was surgically excluded in 104 patients (2 P, 93 N, 9 Eq). Three hundred and eighty-four patients were lost to follow-up; including 348 who had negative finding on CT. There were no known missed diaphragm injuries during the study period or in follow-up. CONCLUSIONS: Injuries to the diaphragm occur commonly after penetrating torso trauma. MDCT scan is an accurate test to detect diaphragm injury. When MDCT is equivocal, further investigation is required to evaluate the diaphragm.

Optimization of selection for nonoperative management of blunt splenic injury: comparison of MDCT grading systems.

OBJECTIVE: The purpose of this study was to compare the usefulness of two CT grading systems of blunt splenic trauma in predicting which patients need surgery or angioembolization. MATERIALS AND METHODS: Four hundred patients in hemodynamically stable condition admitted with blunt splenic injury were included in the study. All patients underwent contrast-enhanced MDCT. Grade of splenic injury was prospectively assigned according to the American Association for the Surgery of Trauma (AAST) splenic injury scale. Patients were treated with surgical intervention, splenic arteriography with or without embolization, or observation alone. All MDCT images were retrospectively reviewed and regraded according to a novel grading system that specifically incorporates the findings of active bleeding or splenic vascular injury, including pseudoaneurysm and arteriovenous fistula. Receiver operating characteristics curves were generated with both grading systems for all splenic interventions, and statistical analyses were performed. RESULTS: The area under the ROC curves for the new splenic grading system for splenic arteriography, surgery, and both interventions exceeded 80%. The area under the curve for the new splenic grading system was greater than that for the AAST injury scale for all interventions. Differences were found to be statistically significant for splenic arteriography (p = 0.0036) and the combination of arteriography and surgery (p = 0.0006). CONCLUSION: The proposed CT grading system is better than the AAST system for predicting which patients with blunt splenic trauma need arteriography or splenic intervention.