Secondary Logo

Journal Logo

Computed tomography scans prior to transfer to a pediatric trauma center

Transfer time effects, neurosurgical interventions, and practice variability

Snyder, Christopher W. MD; Danielson, Paul D. MD; Gonzalez, Raquel MD; Chandler, Nicole M. MD

Journal of Trauma and Acute Care Surgery: October 2019 - Volume 87 - Issue 4 - p 808–812
doi: 10.1097/TA.0000000000002258
Editor's Choice

BACKGROUND Many nontrauma centers perform computed tomography (CT) on injured children prior to transfer to a pediatric trauma center (PTC), but the institutional variability and clinical impact of this practice is unclear. This study evaluated the association of pretransfer CT with transfer delays, the likelihood of emergent neurosurgical intervention among patients who underwent pretransfer head CT, and the effects of transfer distance on prevalence and regional variability of pretransfer CT.

METHODS All injured children transferred from outlying nontrauma centers to a single freestanding PTC from 2009 to 2017 were included. Patients were categorized by undergoing pretransfer CT head alone, CT of multiple/other areas, or no CT. Transfer time (referring hospital arrival to PTC arrival) was compared between CT groups, using multivariable modeling to adjust for covariates. Neurosurgical interventions were compared between patients with normal and abnormal Glasgow Coma Scale (GCS) scores. The prevalence of pretransfer CT among referring centers was compared, with stratification by transfer distance.

RESULTS Of 2,947 transfer patients, 1,225 (42%) underwent pretransfer CT (29%, head CT alone; 13%, other/multiple CT). Transfer times were significantly longer for patients who underwent pretransfer head CT or multiple CT (287 or 298 minutes vs. 260 minutes, p < 0.0001) after adjustment for baseline characteristics, injury severity, and transfer distance. Among patients with normal pretransfer GCS who received a pretransfer head CT, the likelihood of urgent neurosurgical intervention was 1.3%. Prevalence rates of pretransfer CT by referring center varied from 15% to 94%; prevalence increased with increasing transfer distance but demonstrated wide variability among centers of similar distance.

CONCLUSION Pretransfer CT, whether of the head alone or multiple areas, is associated with delays in transfer to definitive care. Among patients with pretransfer GCS 15, the risk of urgent neurosurgical intervention is very low. Wide variability in pretransfer CT use between referring centers suggests opportunity for development of standardized protocols.

LEVEL OF EVIDENCE Economic/decision, level III.

From the Division of Pediatric Surgery (C.W.D., P.D.D., R.G., N.M.C.), Johns Hopkins All Children's Hospital, St. Petersburg, Florida.

Submitted: February 4, 2019, Accepted: February 11, 2019, Published online: March 11, 2019.

Presented at the 5th Annual Meeting of the Pediatric Trauma Society, November 8–10, 2018, Houston, TX.

Address for reprints: Christopher W. Snyder, MD, MSPH, 601 5th St South, Suite 306, St. Petersburg, FL 33701; email:

Online date: March 9, 2019

Computed tomography (CT) is a valuable tool in the workup of trauma victims, but should be used judiciously in the pediatric population due to ionizing radiation.1 Injured children presenting to nontrauma centers with significant injuries should be expeditiously transferred to a pediatric trauma center (PTC) after initial stabilization.2 However, these children often undergo CT prior to transfer, which has been associated with transfer delays and higher radiation.3–6 Computed tomography scans frequently must be repeated at the PTC, further increasing radiation exposure.7 Because transfer can require relatively long transport distances in a regionalized pediatric trauma system,8 some advocate for a head CT before transfer to guide transport decisions and evaluate need for urgent neurosurgical intervention.4,9 The clinical impact and regional practice variability of pretransfer CT, especially head CT alone, remains unclear.

This study of pediatric trauma patients transferred to a freestanding PTC sought to answer the following three questions: (1) Is performance of pretransfer head CT alone associated with transfer delays? (2) What is the likelihood of emergent neurosurgical intervention among patients who underwent pretransfer head CT? (3) How does transfer distance affect the prevalence and regional variability of pretransfer CT?

Back to Top | Article Outline


This study was approved by the institutional review board (protocol 146703). All trauma patients transferred to our freestanding PTC between January 2009 and December 2017 were identified from the institutional trauma registry. Only patients transferred from centers that were not designated adult or PTCs were included; patients transferred from state-designated trauma centers were excluded to minimize selection bias. Designated adult trauma centers have varying levels of pediatric capability and may use CT as a triage tool to decide which patients require transfer to a PTC, whereas nontrauma centers would be expected to transfer all injured pediatric patients regardless of CT findings. Patients were categorized by pretransfer CT status: no CT, head CT alone, or multiple/other CT. Transfer time was calculated as the time from referring center arrival to PTC arrival. Patient characteristics, injury patterns, resource use, and outcomes were obtained from the registry. Categorical variables were compared across CT groups using χ2 or Fisher's exact test; continuous variables were compared using Wilcoxon rank-sum tests. Transfer time was compared between groups, using multivariable linear regression modeling to adjust for covariates found to be significant on bivariate analysis. Since the inherently longer transport times for patients being transferred greater distances is an important confounding factor, we included the transfer distance (categorized by transfer region, as described in detail below) in the model a priori.

The subgroup of patients who had a pretransfer head CT and a recorded pretransfer Glasgow Coma Scale (GCS) scores were analyzed for neurosurgical interventions. These patients were stratified by pretransfer GCS score of normal (15) versus abnormal (≤14). Details of neurosurgical interventions were obtained from the registry and focused chart review. Fisher's exact test was used to compare rates of neurosurgical intervention and death between GCS strata.

To evaluate the prevalence and inter-center practice variability of pretransfer CT imaging, rates of pretransfer CT were calculated for each referring center that transferred at least 50 patients to our PTC during the study period. Centers that transferred patients less frequently were excluded due to expectation of highly skewed prevalence rates. Centers were categorized by location: (1) within the same county as our PTC, (2) within one of the three counties adjacent to our county, or (3) within a county that is not adjacent to ours. Distributions of pretransfer CT prevalence rates, within a given region and between regions, were compared with χ2 and Wilcoxon rank-sum tests, respectively. Analysis was performed with SAS 9.4 (SAS Institute, Cary, NC).

Back to Top | Article Outline


A total of 2,947 trauma patients were transferred to our PTC from nontrauma centers during the study period. Of these, 1,225 (41.5%) underwent at least one pretransfer CT. Computed tomography of the head was performed in 1,076, abdomen/pelvis in 202, spine or other body part in 182, and chest in 73. Isolated CT head was performed in 841, whereas 384 had CT of multiple or other body areas. Characteristics, injuries, disposition, and outcomes of these patients are described in Tables 1 and 2.





Median transfer times for patients undergoing pretransfer head CT alone and multiple CT were 287 and 298 minutes, respectively. Both of these times were significantly longer than the median transfer time (260 minutes) for patients who received no pretransfer CT (p < 0.0001, Wilcoxon rank-sum test). On multivariable regression modeling to adjust for age, sex, Injury Severity Score, injury mechanism, GCS score, and transfer region, the pretransfer CT category remained a significant predictor of longer transfer times (F = 35.4, p < 0.0001).

Table 3 provides details for neurosurgical intervention among the 964 patients who had a pretransfer head CT and a recorded pretransfer GCS. Overall, 35 (3.6%) patients required neurosurgical intervention, 17 (1.8%) emergently within 6 hours of presentation to our center. Patients with normal pretransfer GCS were significantly less likely to require any (3.1% vs. 6.9%, p = 0.04) or emergent (1.3% vs. 4.6%, p = 0.02) intervention. The most common indication for intervention was epidural hematoma.



Prevalence rates of pretransfer CT for each nontrauma center with at least 50 transfers during the study period, grouped by transfer region, are shown in Figure 1. Rates of pretransfer CT increased with increasingly distant regions from the PTC, with median rates of 27%, 37%, and 44% for in-county, adjacent county, and nonadjacent county transfers, respectively (p = 0.01, Wilcoxon rank-sum test). Within each of the three regions, centers differed significantly from each other in prevalence rates of pretransfer CT (p < 0.05, χ2 test). The results were nearly identical for prevalence rates of head CT alone.

Figure 1

Figure 1

Back to Top | Article Outline


This study of injured children transferred to a PTC from nontrauma centers provides useful information on the clinical impact and regional practice variability of pretransfer CT, particularly head CT alone.

First, we found that performance of pretransfer CT, whether of the head alone or of multiple anatomic areas, was associated with delays in transfer. CT head alone was associated with a median delay of approximately half an hour, whereas multiple CT was associated with a delay of nearly 10 additional minutes. These delays persisted after adjustment for transfer distance and other clinically relevant covariates.

Our findings are consistent with a previous study of 748 injured children transferred in to a level I PTC, which found higher probability of late transfer (>2 hours) among children receiving pretransfer CT, independent of injury characteristics and transport distance or mode.6 That study did not attempt to specifically quantify the CT-associated delay or evaluate the effect of head CT alone. In another study at a freestanding PTC in Utah, Fenton et al.3 demonstrated that any CT imaging prior to transfer was associated with prolongation of median transfer time by approximately 67 minutes. Those authors did not evaluate head CT alone or adjust for potential confounding variables, such as transport distance. Fahy et al.5 found the pediatric blunt trauma patients undergoing pretransfer chest or abdominal CT had a mean time from injury to PTC arrival that was 112 minutes longer than those who did not undergo CT. Adjustment for covariates and evaluation of head CT was not attempted.

Second, among patients who underwent pretransfer head CT, the overall likelihood of emergent neurosurgical intervention within 6 hours of PTC arrival was low (1.8%) and was strongly influenced by pretransfer GCS. Only 11 (1.3%) of 833 patients with normal pretransfer GCS required emergent intervention, and 3 (27%) of these underwent repeat imaging at the PTC prior to operation. This study was unable to evaluate the appropriateness of pretransfer CT according to validated clinical decision rules such as those developed by Kuppermann et al.10 However, the high proportion of patients with normal GCS receiving head CT alone, along with the low rate of emergent neurosurgical intervention, suggests that the practice of pretransfer head CT alone is overutilized in patients with normal GCS.

Patients with abnormal pretransfer GCS had a significantly higher rate (4.6%) of emergent neurosurgical intervention. Although head CT was indicated for these patients,10 whether the initial imaging was best performed before or after PTC transfer is unclear. Sixty-seven percent of the abnormal GCS patients who required emergent operation had imaging repeated at the PTC prior to intervention, and adult studies suggest that head CT prior to transfer does not expedite craniotomy.11 While the nearly half-hour delay associated with pretransfer head CT may seem like a small amount of time, minutes matter in trauma. Longer transfer times have been associated with a higher rate of complications, such as pneumonia, pulmonary embolism, sepsis, seizures, and postoperative hemorrhage, in pediatric trauma victims.12 Prolonged transfer time is also predictive of worse head injury outcomes in teenagers and adults.13 We saw higher injury severity and mortality among patients who received pretransfer CT, reflecting their higher acuity and potentially greater benefit of expeditious transfer.

Third, we found that the prevalence of pretransfer CT increased with increasing distance from the PTC, suggesting that referring clinicians are obtaining CT with the intention of using the information for planning longer-distance transports. However, we saw nearly two to four-fold variation in prevalence rates for pretransfer CT among centers within the same region (i.e., centers with similar transport distances). This high variability suggests that pretransfer CT practice patterns are largely driven by other factors specific to the referring institution. Interestingly, the referring institution with the lowest rate of pretransfer CT (15%) was a center in an adjacent county where pediatric emergency medicine (EM) care is formally coordinated with our PTC and pediatric surgeons provide emergency department consultation. To our knowledge, this center is the only nontrauma center in our catchment area that formally coordinates pediatric EM care with a regional PTC.

Our findings must be interpreted with several limitations in mind. This was a single-center study in a US state with a unique population distribution and regionalized pediatric trauma system, so it may not be generalizable to PTCs in other regional systems. The primary data source was a trauma registry, which may contain missing or inaccurate data. A large proportion of the transfer patient population had isolated long bone fractures, which would not be expected to routinely require CT scan. While we attempted to control for confounding variables between CT and non-CT groups with multivariable modeling, residual or unmeasured confounding could still affect the results.

In summary, CT prior to transfer from a nontrauma center to a PTC is associated with transfer delays, whether the head alone or multiple anatomic areas are imaged. The likelihood of emergent neurosurgical intervention is very low among patients with a normal pretransfer GCS, suggesting head CT is overutilized in transfer patients with normal mental status. The prevalence of pretransfer CT increases with distance from the PTC, but its high variability among centers of similar transport distance suggests that referring institution preferences play a large role in the decision whether or not to scan before transfer.

Decisions regarding pretransfer CT should be made in a collaborative fashion between referring institutions and the regional PTC. Ideally, shared protocols should be developed by pediatric trauma surgery, EM, and neurosurgery that provide consistent messaging to referring institutions regarding when to perform pretransfer CT. A united voice of collaboration and education may help reduce arbitrary variability and optimize the delicate balance of risk and benefit for injured children requiring transfer.

Back to Top | Article Outline


C.W.S., N.M.C. participated in the literature search. C.W.S., R.G., P.D.D., N.M.C.participated in the study design. C.W.S.participated in the data collection. C.W.S., R.G. participated in the data analysis. C.W.S., R.G., P.D.D., N.M.C. participated in the data interpretation. C.W.S., R.G., P.D.D., N.M.C. participated in the writing. C.W.S., R.G., P.D.D., N.M.C. participated in the critical revision.

Back to Top | Article Outline


The authors declare no conflicts of interest.

Back to Top | Article Outline


1. Miglioretti DL, Johnson E, Williams A, Greenlee RT, Weinmann S, Solberg LI, Feigelson HS, Roblin D, Flynn MJ, Vanneman N, et al. The use of computed tomography in pediatrics and the associated radiation exposure and estimated cancer risk. JAMA Pediatr. 2013;167(8):700–707.
2. Brinkman AS, Gill KG, Leys CM, Gosain A. Computed tomography-related radiation exposure in children transferred to a level I pediatric trauma center. J Trauma Acute Care Surg. 2015;78(6):1134–1137.
3. Fenton SJ, Hansen KW, Meyers RL, Vargo DJ, White KS, Firth SD, Scaife ER. CT scan and the pediatric trauma patient—are we overdoing it? J Pediatr Surg. 2004;39(12):1877–1881.
4. Benedict LA, Paulus JK, Rideout L, Chwals WJ. Are CT scans obtained at referring institutions justified prior to transfer to a pediatric trauma center? J Pediatr Surg. 2014;49(1):184–187; discussion 187–188.
5. Fahy AS, Antiel RM, Polites SF, Ishitani MB, Moir CR, Zielinski MD. Pretransfer computed tomography delays arrival to definitive care without affecting pediatric trauma outcomes. J Pediatr Surg. 2016;51(2):323–325.
6. Chatoorgoon K, Huezo K, Rangel E, François N, Schweer L, Daugherty M, Koehn M, Ricketts C, Brown RL, Garcia VF, et al. Unnecessary imaging, not hospital distance, or transportation mode impacts delays in the transfer of injured children. Pediatr Emerg Care. 2010;26(7):481–486.
7. Chwals WJ, Robinson AV, Sivit CJ, Alaedeen D, Fitzenrider E, Cizmar L. Computed tomography before transfer to a level I pediatric trauma center risks duplication with associated increased radiation exposure. J Pediatr Surg. 2008;43(12):2268–2272.
8. Snyder CW, Chandler NM, Litz CN, Pracht EE, Danielson PD, Ciesla DJ. Immature patients in a mature system: regional analysis of Florida's pediatric trauma system. J Trauma Acute Care Surg. 2017;83(4):711–715.
9. Havrda JB. Pretransfer imaging decisions in rural trauma cases. Radiol Technol. 2014;86(2):187–198.
10. Kuppermann N, Holmes JF, Dayan PS, Hoyle JD, Atabaki SM, Holubkov R, Nadel FM, Monroe D, Stanley RM, Borgialli DA, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374(9696):1160–1170.
11. Tonui PM, Spilman SK, Pelaez CA, Mankins MR, Sidwell RA. Head CT before transfer does not decrease time to craniotomy for TBI patients. Am Surg. 2018;84(2):201–207.
12. Locke T, Rekman J, Brennan M, Nasr A. The impact of transfer on pediatric trauma outcomes. J Pediatr Surg. 2016;51(5):843–847.
13. Dinh MM, Bein K, Roncal S, Byrne CM, Petchell J, Brennan J. Redefining the golden hour for severe head injury in an urban setting: the effect of prehospital arrival times on patient outcomes. Injury. 2013;44(5):606–610.

Pediatric trauma; computed tomography; transfer; neurosurgery; practice variability

© 2019 Lippincott Williams & Wilkins, Inc.