Abstract
Introduction
Nonoperative management of hemodynamically stable patients with blunt splenic and/or hepatic injury has been widely accepted in the pediatric population. However, variability exists in the utilization and timing of repeat imaging to assess for delayed complications during index hospitalization. Recent level-IV evidence suggests that repeat imaging in children should be performed based on a patient’s clinical status rather than on a routine basis. The aim of this study is to examine the rate of delayed complications and interventions in pediatric trauma patients with blunt splenic and/or hepatic injuries who undergo repeat imaging prompted either by a clinical change (CC) or non-clinical change (NCC).
Methods
A 9-year (2011-2019), retrospective, dual-institution study was performed of children (0-17 years) with blunt splenic and/or hepatic injuries. Patients were grouped based on reason for repeat imaging: CC or NCC. The rate of organ-specific delayed complications and interventions was examined by reason for scan.
Results
A total of 307 injuries were included in the study period (174 splenic, 113 hepatic, and 20 both). Of 194 splenic injuries, 30(15.5%) underwent repeat imaging (CC = 19; NCC = 11). Of 133 hepatic injuries, 27(20.3%) underwent repeat imaging (CC = 21; NCC = 6). There was no difference in the incidence of organ-specific delayed complications between the CC and NCC groups. Of the 4 patients with complications necessitating intervention, only one was identified based on NCC.
Conclusions
Our data suggest routine repeat imaging is unnecessary in children with blunt splenic and/or hepatic injuries; therefore, practitioners may rely on a patient’s clinical change.
Introduction
Nonoperative management (NOM) of hemodynamically stable children with blunt solid organ injury has been widely accepted as standard of care.1,2 Success of NOM in the pediatric population, defined as the lack of further procedures, has been reported to be as high as 90-95%, even for high grade injuries.3-5 Despite assurances that NOM is a safe option, and reports concluding that repeat imaging is often unnecessary,6,7 the management of pediatric blunt solid organ injury continues to vary regarding the utilization and timing of repeat imaging.
A recent review by the American Pediatric Surgical Association (APSA) Outcomes and Evidence Based Practice Committee has recommended that repeat imaging in pediatric patients with blunt solid organ injury should be primarily guided by a patient’s clinical status. 8 This recommendation is based on level-IV data derived from several studies demonstrating a low rate of delayed complications in asymptomatic patients with low grade injuries.9,10 Despite this recommendation, there are reports of delayed complication rates as high as 1.7-17%.3,11,12 In a study of 362 patients, Safavi et al found that 5.4% of children with blunt splenic injury and 1.7% of children with blunt hepatic injury developed traumatic pseudoaneurysm on repeat imaging, many of which required intervention. 11 Similarly, Durkin et al cites a pseudoaneurysm rate of 17%, with 64.7% of those being asymptomatic. 12 Both the clinical significance of delayed complications and if repeat imaging will assist in their earlier identification and intervention remains unclear.
The aim of this study was to examine the rate of delayed complications and subsequent interventions in pediatric trauma patients with blunt splenic and/or hepatic injuries who undergo repeat imaging, and to further characterize the reason for imaging. We aimed to provide a direct comparison between repeat imaging performed due to a patient’s clinical change (CC) and that prompted by a non-clinical change (NCC) including an institution’s standard protocol/recommendations or an individual physician practice pattern. We hypothesized that repeat imaging in children who sustain blunt trauma to the spleen and/or liver and could be safely guided by a patient’s clinical course and that routine imaging is unnecessary in this patient population.
Methods
Study Design and Patient Selection
This study was a retrospective chart review of all pediatric patients (age 0-17 years) who sustained blunt splenic and/or hepatic injury over a 9-year period (January 1, 2011-December 31, 2019). Following institutional review board approval at both institutions, patients were identified using the trauma registries at 2 Trauma Centers: one state verified Adult Level I (further mentioned as Center 1) and the other a state verified combined Adult and Pediatric Level I (further mentioned as Center 2). Patients with incomplete medical records or no documentation of blunt splenic and/or hepatic injuries were excluded. Patients were also excluded if they were age ≥18, sustained a penetrating mechanism of injury, or were transferred to another facility prior to hospital admission.
Variables
Electronic medical record review was performed to obtain patient information including demographic information, Injury Severity Score (ISS), Abbreviated Injury Scale (AIS) for all organ systems (head/neck, face, chest, abdomen/pelvis, extremity, and external), Trauma Injury Severity Score (TRISS), mechanism of injury, length of intensive care unit (ICU) stay, length of hospital admission, and mortality.
Information regarding the initial trauma resuscitation was collected including first vital signs, initial imaging findings [computed tomography (CT) or ultrasound (US)] if performed, and associated injuries. Total volume of blood products, anticoagulation administration, initial operative or nonoperative management, and timing of interventions were compiled. Additionally, findings of follow-up imaging, presence of delayed complications, and subsequent procedures were also reviewed. Both splenic and hepatic injuries were classified based on the American Association for the Surgery of Trauma (AAST) grading scale using imaging reads performed by the attending radiologist or operative notes if appropriate. Patients who sustained both splenic and hepatic injuries were included in both injury groupings for analysis.
Study Groups
Patients were identified based on whether repeat imaging was performed. Repeat imaging was classified as CT or US which occurred following the initial trauma evaluation but was still performed during the index hospital admission. Follow-up scans in the outpatient setting or which may have been obtained at another institution were not included for the purposes of this study. If the patient underwent multiple repeat scans during their hospital admission, only the first repeat scan was included.
Repeat imaging was categorized as performed by either clinical change (CC) or non-clinical change (NCC). Clinical change was defined as repeat imaging prompted by a change in patient status including vital sign changes, laboratory abnormalities, increased abdominal pain, and other patient symptoms. Non-clinical change was defined as imaging performed on a routine/scheduled basis including that following institutional protocols/recommendations and individual physician practice patterns. Imaging findings were recorded, focusing on the presence of organ-specific delayed complications. In patients with blunt splenic injuries, delayed complications included pseudoaneurysm, increased hemoperitoneum, injury progression, abscess, vascular thrombosis, pancreatic fistula, arteriovenous fistula, or delayed contrast extravasation. In patients who sustained blunt hepatic injuries, the delayed complications observed were pseudoaneurysm, increased hemoperitoneum, injury progression, bile leak/biloma, abscess, or vascular thrombosis.
Post-imaging interventions included both operative and interventional radiologic procedures. The timing of intervention performed in relation to repeat imaging was also noted.
Statistical Analysis
T-tests were used to analyze age, ISS, AIS, ICU length of stay, and hospital length of stay and Fisher’s exact tests were used to analyze sex, mechanism of injury, and mortality to determine if there were significant differences between patients who had repeat imaging via clinical change and patients who had repeat imaging via non-clinical change. Patients were divided by their AAST grade of injury, and Fisher’s exact tests were performed to determine if there were significant differences between patients who underwent repeat imaging via clinical change and patients who underwent repeat imaging via non-clinical change for all patients who underwent repeat imaging, patients who experienced complications identified, and patients who had interventions performed.
Results
In the period from January 2011 to December 2019, 307 pediatric patients with blunt splenic and/or hepatic injuries presented to the 2 institutions [Center 1, n = 57 (18.6%); Center 2, n = 250 (81.4%)]. Isolated splenic injuries were present in 174 patients, isolated hepatic injuries in 113 patients, and both injuries (splenic and hepatic) in 20 patients. Therefore, in total, 194 splenic and 133 hepatic injuries were included in the analysis.
Three patients who underwent repeat imaging expired during their index hospitalization. All 3 sustained significant neurologic injuries that contributed to their mortality. The first, a 17-year-old female, sustained both a grade I splenic injury and grade II hepatic injury which remained stable on repeat imaging. The second was a 9-year-old female with a grade II hepatic injury that demonstrated increased hemoperitoneum on repeat imaging. Finally, the third was a 4-year-old male who sustained a grade I splenic injury and grade III hepatic injury. This patient demonstrated increased hemoperitoneum on repeat imaging. All patients who underwent repeat imaging and had fatal injuries were treated at Center 2.
Splenic Injury
Of 194 patients with blunt splenic injury, 30 (15.5%) underwent repeat imaging (Figure 1). Nineteen were scanned due to clinical change and eleven due to non-clinical change. Median age for both groups was 16 years old with an interquartile range (IQR) of 10-17 years for CC and 15-17 years for NCC. The characteristics of the study populations were similar (Table 1). Repeat imaging in pediatric blunt splenic injury. Characteristics of Patients With Blunt Splenic Injury. Median age reported in years. Abbreviations: AIS = Abbreviated Injury Scale; ATV = all-terrain vehicle; ICU = intensive care unit; IQR = interquartile range; ISS = Injury Severity Score; LOS = length of stay; MVC = motor vehicle collision.
Mean time to repeat imaging in hours from admission was similar between the CC and NCC groups [CC: 101.5 ± 82.4 hours; NCC: 79 ± 73.9 hours; P = .480]. Delayed complications were identified in 7 CC patients [Center 1: n = 1; Center 2: n = 6] and 2 NCC patients [Center 1: n = 2; Center 2: n = 0] (P = .420). Two of the 7 patients with complications found by CC scan underwent intervention. These included angioembolization for pseudoaneurysm (Center 1) and exploratory laparotomy with washout and pelvic drain placement for abscess (Center 2). One of the two NCC patients with complications found underwent angioembolization for pseudoaneurysm. There was no significant difference in interventions performed by reason for scan (P = 1.000).
Blunt Splenic Injury Complications and Interventions Based on Clinical Change or Non-Clinical Change.
*Denotes significance. Repeat imaging P-values calculated based on other grades. P-values for complication and intervention based on yes/no present for that grade. N/a: not applicable.
Hepatic Injury
Of 133 patients with blunt hepatic injury, 27 (20.3%) underwent repeat imaging (Figure 2). Twenty-one were scanned due to CC and 6 due to NCC. Median age did not significantly differ between groups with CC at 12 years old (IQR 4-16 years) and NCC at 14 years old (IQR 13-15 years). The CC group had higher ISS and hospital length of stay (Table 3). Repeat imaging in pediatric blunt hepatic injury. Characteristics of Patients With Blunt Hepatic Injury. *Denotes significance. Median age reported in years. IQR = interquartile range. ISS = Injury Severity Score. AIS = Abbreviated Injury Scale. MVC = motor vehicle collision. ATV = all-terrain vehicle. ICU = intensive care unit. LOS = length of stay.
Blunt Hepatic Injury Complications and Interventions Based on Clinical Change or Non-Clinical Change.
*Denotes significance. Repeat imaging P-values calculated based on other grades. P-values for complication and intervention based on yes/no present for that grade.
Discussion
The majority of pediatric patients who sustain blunt splenic and/or hepatic injuries can be successfully managed nonoperatively.1-5 Given the lack of guidance for the role of repeat imaging in this patient population, we found it essential to better characterize the incidence of delayed complications and interventions. Our aim was to distinguish between repeat imaging performed due to clinical change and that performed on a routine basis (non-clinical change). Our data found no difference in the incidence of organ-specific delayed complications or interventions between the CC and NCC groups. This study supports earlier findings by the APSA Outcomes and Evidence Based Practice Committee which state blunt splenic and hepatic injuries can be managed according to clinical changes/physician discretion, as opposed to a standard reimaging protocol. 8
Delayed complications following blunt solid organ injury are rare in the pediatric population. In a study of long-term outcomes of nonoperative management in blunt splenic injury, Kristoffersen and Mooney 9 report the incidence of delayed complications to be .44%. However, some studies have found much higher rates of complications. Safavi et al 12 demonstrated a pseudoaneurysm rate of 5.4% in blunt splenic injuries and 1.7% in blunt hepatic injuries. Durkin et al 11 saw an even higher rate of pseudoaneurysm at 17% when contrast enhanced ultrasound (CEUS) was routinely used for screening in both splenic and hepatic injuries. The patients that went on to intervention in both studies were reported to be symptomatic, but the clinical significance of delayed complications remains unknown. In our study, 1 adolescent NCC patient with splenic injury was found to have a pseudoaneurysm and went on to receive intervention with angioembolization. It is unclear whether the patient would have demonstrated a clinical change had the repeat imaging not occurred on a routine basis. Additionally, given the patient’s age, it is possible that in a larger population, there would be differences between adolescents and younger children. This questions if adolescents should be managed more similarly to adults in regard to repeat imaging.
Grade of injury in both blunt splenic and hepatic trauma has assisted in determination of management strategies.1,2 Safavi et al 12 show pseudoaneurysm formation in only grade III and IV splenic injuries. Similarly, in pediatric blunt hepatic injury, Giss et al 3 demonstrated that though 3.8% of patients developed delayed complications (biloma, pseudoaneurysm, and necrotic gallbladder), these occurred only in grade III or higher. Our data did not demonstrate a difference in the rate of delayed complications or interventions when compared by grade for either blunt splenic or hepatic injuries. We did, however, see that grade III blunt splenic injuries were more likely to undergo repeat imaging due to a non-clinical change scan, suggesting that institutions may anticipate a greater likelihood of developing complications with higher grade injuries.
Children present a unique challenge to the trauma surgeon. There has been evidence to suggest that surgeons are more likely to pursue nonoperative management in pediatric patients with blunt splenic and/or hepatic injury when practicing at a dedicated pediatric center.13,14 However, the majority of pediatric patients are managed in an adult setting. 15 It is essential to more clearly define the clinical course of this patient population to assist in successful nonoperative management. Our study included patients from 2 Trauma Centers; one a combined Pediatric and Adult Level 1 and one an Adult Level 1. It is worth noting that patients presenting to Center 1 are primarily adolescents (age 15-17 years), whereas all ages (0-17 years) present to Center 2. Though not a primary aim of this study, there were recorded delayed complications and interventions at both institutions. Further research could be aimed at distinguishing between the imaging practices of adult and pediatric providers.
This study is limited by its retrospective nature as we were unable to standardize the imaging protocol or the strategies used in nonoperative management of the patients (ie, timing of repeat laboratory examinations, bedrest, activity restrictions, and ICU length of stay). This lack of standardization could be reflected in differences between the two institutions as well as the possibility for changing institutional protocols/recommendations over the 9-year time period. As a result, we defined repeat imaging performed based on clinical change as that which occurred following patient symptoms, vital sign changes, or laboratory abnormalities. In contrast, non-clinical change was defined as imaging obtained on a scheduled basis following initial stabilization and during the index hospital admission. This group included imaging prompted by an institutional protocol/recommendation and/or due to individual physician practice patterns. Moderate variation in the NCC cohort is reflected in the hours to splenic and hepatic reimaging at 79 ± 73.9 and 73.5 ± 81.2, respectively.
The Injury Severity Score (ISS), hospital length of stay, and ICU length of stay of our study population are elevated compared to prior studies.5,11 It is important to note that Center 2 serves as a tertiary referral center for pediatric trauma patients and therefore may represent a more severely injured population. Given the small sample size, patients were not excluded based on other organ systems injured. This is seen in the 3 mortalities which were attributed to neurologic injuries. Polytrauma could have contributed to higher ISS, hospital LOS, and ICU LOS.
Only the first repeat imaging performed during the index hospitalization was included in this study. This could have limited our ability to assess for delayed complications which occurred later in a patient’s course. Furthermore, a limitation of this data set is reflected in the inability to capture complications that occurred post-hospitalization. This differs from prior studies which looked to quantify complications post-discharge.16,17 Finally, despite this being a dual-institution study, the sample size of delayed complications is small and thus may not reflect true differences between imaging populations. Due to the lack of large numbers of complications, data from a larger multicenter study may be able to address the concerns of small sample size.
Conclusions
Though recent updated guidelines recommend the use of a patient’s clinical status, 8 institutional differences persist in the utilization and timing of repeat imaging in the management of pediatric blunt solid organ injury. A major critique of the prior recommendations is their derivation from small data sets and single institution experiences. This study looked to expand on prior data while providing a direct comparison between imaging based on clinical change and that performed on a routine, scheduled basis. We found no difference in the incidence of clinically significant delayed complications between CC and NCC groups. These data suggest that it is safe to rely on clinical change to prompt repeat imaging in pediatric blunt splenic and hepatic injuries. Further research is in process to garner a larger prospective study population in the hopes of more effectively answering this important question.
Footnotes
Acknowledgments
The authors thank Kellie Bresz, MS for assistance with the statistical plan and analysis.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.