Secondary Logo

Journal Logo


A comparison of adolescent penetrating trauma patients managed at pediatric versus adult trauma centers in a mature trauma system

Rogers, Frederick B. MD, MS, MA, FACS; Horst, Michael A. PhD; Morgan, Madison E. BS; Vernon, Tawnya M. BA; Gaines, Barbara A. MD; Rogers, Amelia T. MD; Gross, Brian W. BS; Cook, Alan D. MD, MS, FACS; Bradburn, Eric H. DO, MS, FACS

Author Information
Journal of Trauma and Acute Care Surgery: June 2020 - Volume 88 - Issue 6 - p 725-733
doi: 10.1097/TA.0000000000002643
  • Free
  • Editor's Choice


According to the Centers for Disease Control and Prevention, unintentional injury remains the leading cause of death among adolescents in the United States, with homicide and suicide encompassing the second and third most prevalent mechanisms, respectively.1 While the majority of these injuries present as blunt trauma, Cotton and Nance2 found that penetrating wounds account for up to 20% of all trauma cases in the pediatric/adolescent population. Although regionalization of trauma centers within organized trauma systems and advances in prehospital and critical care have resulted in improved outcomes for penetrating injuries, penetrating trauma continues to disproportionately contribute to overall trauma mortality in the United States—particularly among America's youth.1

In 1976, the American College of Surgeons Committee on Trauma published the “Resources for Optimal Care of the Injured Patient” which set forth guidelines for care of the injured patient and verification requirements for trauma center accreditation. The development of pediatric trauma centers (PTCs) was concurrent to that of adult trauma centers (ATCs), and began in the 1970s and 1980s.3 In recent years, care at PTCs has proven to reduce morbidity and death related to pediatric traumatic injuries.4–8 While there is little debate that PTCs are uniquely equipped to manage the pediatric trauma patient, the extent to which older adolescents benefit from treatment at these facilities remains undetermined.9,10

When tasked with managing the severely injured pediatric/adolescent trauma patient, both prehospital and in-hospital providers must decide whether these cases can be appropriately managed at an ATC, or whether triaging and/or transferring these patients to pediatric centers would result in improved care. Hospital volume is known to be inversely related to mortality. Multiple studies have found that mortality was lower and quality of care was higher at high-volume hospitals compared with low-volume hospitals.11–13

Pediatric trauma centers were established to provide the highest level of definitive care to the critically injured pediatric trauma patient. Although held to the same essential standards as their adult counterparts, PTCs must also meet additional pediatric-specific trauma criteria which include having a dedicated pediatric trauma service directed by a pediatric surgeon, pediatric specialists, and specified pediatric resuscitation equipment and resources.14–16

The purpose of this investigation was to elucidate differences in management approach and outcome between PTCs and ATCs for the adolescent penetrating trauma population. Due to differences in the types of facilities and cases, we hypothesized that improved mortality would be observed at ATCs for this subset of older adolescent pediatric patients.


Following review and approval by the institutional review board of Penn Medicine Lancaster General Health, all adolescent penetrating trauma admissions (age, 15–18 years) in the Pennsylvania Trauma Outcome Study (PTOS) database of the Pennsylvania Trauma Systems Foundation (PTSF) were retrospectively queried from 2003 to 2017. The Pennsylvania Trauma Systems Foundation is a private, nonprofit organization written into the Emergency Medical Services Act to accredit trauma centers within the Commonwealth of Pennsylvania. To maintain accreditation, trauma centers must submit deidentified patient data to PTOS for all patients meeting one or more of the following trauma criteria:

  • Intensive care unit admission,
  • Step-down unit admission,
  • Hospital length of stay (LOS) longer than 48 hours or hospital stay of 36 hours to 48 hours with an Injury Severity Score (ISS) of 9 or higher,
  • Transfer in/out status,
  • Trauma death (including dead on arrival patients).17

To investigate the impact of trauma center designation on adolescent penetrating trauma injury management approach and outcome, the study population was divided into two subgroups: patients treated at PTC and those managed at ATC. For the purpose of this investigation, PTCs were defined as all standalone PTCs in the state, as well as all adult trauma centers with pediatric affiliation. Adult trauma centers were classified as strictly standalone adult facilities. We chose to classify into two groups instead of three (PTC, ATC, ATC with pediatric affiliation) because our purpose was to evaluate the trauma system as it currently is, not to hone in on the intricacies of clinical care. To ensure the study population would encompass adolescent penetrating trauma patients receiving care exclusively at ATC or PTC, blunt trauma, transfer patients, and adolescents admitted to Levels III and IV trauma centers were excluded from analysis. Patients presenting dead on arrival (defined as death within 1 hour of arrival to the emergency department [ED]) were also excluded, as to compare only adolescent trauma cases with manageable injuries presenting to ATC or PTC within the Pennsylvania trauma network.

Univariate analysis in the form of two-sample t tests, nonparametric tests, and χ2 were used to assess unadjusted baseline demographic, management, and outcome differences between ATC and PTC adolescent penetrating trauma subgroups as well as factors associated with the four outcomes of interest. Mean annual cases were defined as the average number of cases each trauma center saw across the 13-year study period that were adolescent penetrating trauma. Compared variables included year of ED arrival, age, sex, systolic blood pressure (SBP) at admission, intubation on admission, ISS, Abbreviated Injury Scale (AIS), trauma center, number of International Classification of Diseases, Ninth/Tenth revision (ICD-9/ICD-10) complications, hospital LOS, surgical intervention, and in-hospital mortality. Systolic blood pressure was converted to a binary variable (SBP <90 vs. SBP ≥90) to designate shock. Maximum AIS classified as 3 or higher by region were included in the analysis based on the recent work of Brown et al.18,19 in pediatric trauma patients.

Multilevel mixed effects logistic regression models with trauma center as the clustering variable were used to assess the impact of PTC trauma designation on surgical intervention and in-hospital mortality, adjusted for relevant covariates. For analyses of LOS, complications, and surgery, we only modeled patients discharged alive due to competing risks (death is a competing event that prevents the outcomes of interest such as complications, surgery, and LOS from being observed) and the lack of specific time markers in the PTOS database for complications and surgery. Similarly, in the surgery and complication models, we did not include complications or surgery, respectively, because of the lack of precise time markers of these events to determine which may have preceded the other to establish a causal link. Length of stay was dichotomized by median LOS of 4 days or less and longer than 4 days. For all models, nonsignificant variables were retained at the expense of statistical efficiency to maintain consistency in being able to compare between the four outcomes (see Tables 2 and 2B and 3 and 3B).

Multilevel Logistic Models of Adolescent Penetrating Trauma Treated at PTC vs. ATC Mortality
Multilevel Logistic Models of Adolescent Penetrating Trauma Treated at PTC vs. ATC Complications
Multi-Level Logistic Models of Adolescent Penetrating Trauma Treated at PTC vs. ATC Any Surgery (Discharged Alive)
Multilevel Logistic Models of Adolescent Penetrating Trauma Treated at PTC vs. ATC LOS > 4 Days/Overall Median (Discharged Alive)

The area under the receiver operating characteristic curve and Bayesian information criterion (BIC) were calculated as a measure of performance for the multilevel regression models. The BIC is a commonly used metric used for model development and assesses how well the model fits, but is penalized based on the number of covariates included in the model (this provides better estimates of fit when higher numbers of covariates or large samples are used). We used network analysis in ArcGIS software to calculate the road distance from the patient home zip code to the nearest PTC to help assess differences in the access to PTCs across the state. The ICD E-codes were summarized to classify the penetrating injury cause into groups related to firearm, cutting, assaults, accidents and intentional groups. All data manipulation and statistical analyses were performed using Stata/IC, version 16.0. A p value less than 0.05 was considered statistically significant.


A total of 26,585 adolescent trauma patients aged 15 to 18 years presented to Pennsylvania's 36 accredited trauma centers between 2003 and 2017, of which 3,773 had a penetrating mechanism of injury. Within this penetrating subgroup, 897 cases were classified as transfer patients and 227 presented dead on arrival, and 19 were at a Level III or Level IV trauma center and were excluded. The final sample included 2,630 trauma cases with 428 (16.3%) cases presented to PTC (6 centers), and 2,202 cases (83.7%) were managed at ATC (30 centers) (Fig. 1). The adolescent penetrating trauma population under investigation was predominantly comprised of 18-year-old (39.4%; n = 1,036) male (88.3%; n = 2,320) trauma patients with a median (Q1–Q3) ISS of 10 (5–17). An overview of study population demographic data is presented in Table 1.

Figure 1
Figure 1:
Exclusion criteria.
Univariate Analysis of Adolescent Penetrating Trauma: All Patients, ATC, and PTC

Compared with ATCs, PTCs tended to treat a higher proportion of females (17.1% vs. 10.7%, p < 0.001) and those with a Hispanic ethnicity (15.6% vs. 7.7%, p < 0.001). Pediatric trauma centers treated a lower proportion of black patients (61.3% vs. 78.7%, p < 0.001) compared with ATCs. For key clinical variables, PTCs had a lower rate of patients with LOS longer than 4 days (p = 0.028), surgery rate (p < 0.001), complication rate (p = 0.012), and mortality rate (p = 0.001). The median (Q1–Q3) distance from the patient home zip code to the nearest PTC was no different (p = 0.982) for patients going to a PTC 5.0 (2.8–14.8) versus ATC 5.1 (2.7–14.8) (Table 1).

PTCs had a lower adjusted odds of mortality (adjusted odds ratio[AOR], 0.35; 95% confidence interval [CI], 0.17–0.74; p = 0.006; Table 2A) and a lower adjusted odds of surgery (AOR, 0.67; 95% CI, 0.48–0.93; p = 0.016; Table 2B) than their ATC counterparts. There were no differences in complication rates (AOR, 0.94; 95% CI, 0.57–1.55; p = 0.793; Table 2B) or LOS longer than 4 days (AOR, 0.95; 95% CI, 0.61–1.48; p = 0.812; Table 3B) between the PTC and ATC. Pediatric trauma centers saw a different distribution of penetrating injury causes (p < 0.001) (Table 4), most notably a lower percentage of firearm assault cases than the ATCs; 44.9% and 69.3%, respectively. Annual median (Q1–Q3) penetrating trauma volume during the 15-year period of the study was no different in the ATCs 5 (2–10) versus PTCs 4.5 (3–7) (p = 0.551) (Table 1).

Injury Type Treated at PTC vs. ATC vs. Total


There has been much debate about whether the adolescent trauma patient is best served at an ATC or PTC. The quandary is that in most cases, from a physiologic standpoint, these patients resemble adults; however, from an emotional and psychologic standpoint, these patients are children. This study sought to determine the outcome of older adolescents (15–18 years) with penetrating trauma of any kind when treated at an ATC versus PTC. We hypothesized that because there are differences in the facilities and the case mix between ATCs and PTCs, that there would be differences in outcomes. The results of this study refute our hypothesis. It appears in this statewide database over a 15-year period that PTCs are as good as ATCs at least in terms of mortality and the need for surgery. Of note, the LOS and complication analysis did not appear to show any difference between ATC and PTC. We deliberately choose to exclude transfer patients from this study because we felt that it would be difficult to adjudicate the results of care in the transferring hospital versus the receiving hospital. Further, we chose to exclude Level III and Level IV trauma centers from this analysis because their directives (especially as it relates to the pediatric patient) are more on stabilization and transfer and not definitive care (it is interesting to note there were only 19 adolescent patients taken care of in all the Level III and IVs. in the state over the entire 15-year study period). Finally, we chose to look at older adolescents (15–18 years) and not younger adolescents (13–14 years) because it is a specific mandate within our state that patients younger than 15 years be transferred to PTC if they meet certain criteria. The decision with regard to older adolescents is at the discretion of the physicians/Emergency Medical Services (EMS).

An interesting question arises as to why the PTC has improved outcome. The PTCs had a 2.86 times lower odds of dying compared with their adult counterparts (AOR, 0.35; 95% CI, 0.17–0.74; p = 0.006) in the adjusted model controlling for observed differences in the covariates included in the multivariate model. There certainly could be other unobserved factors that were not taken into account in the model that might explain the difference. There are only 6 PTCs versus 30 ATCs in the entire state, and we found no difference in the median (Q1–Q3) travel distances for patients going to a PTC of 5.0 miles (2.8–14.8 miles) versus an ATC 5.1 miles (2.7–14.8 miles) despite the fact that ATCs are serving regions with more limited access in the western part of the state. In PA, all but one of the PTCs are clustered in the eastern part of the state, and there are large regions, particularly in the central and western parts of the state, where the closest trauma centers are ATCs. We also sought to determine whether it was a surgical volume phenomenon at any one trauma center that could explain the improved outcome at PTC. Unfortunately, there were not enough surgical cases, distributed over 36 trauma centers over 15 years, to do any meaningful analysis of low-volume trauma centers versus high-volume trauma centers (the median number of cases per year was 5).

The literature surrounding PTCs or ATCs in adolescent trauma patients has mixed results. A study conducted by Webman et al.6 compared mortality of adolescents (15–19 years old) treated at PTCs versus ATCs using data from Level I and II trauma centers within the 2010 National Trauma Data Bank using multilevel remodeling evaluating center characteristics on mortality. In over 29,613 adolescents, the authors noted that mortality was higher among adolescents treated at ATCs (odd ratios, 4.19; 95% CI, 1.30–13.51) than those treated at PTCs. Rogers et al.20 examined if there was a difference in mortality between adolescent (12–17 years old) blunt severe polytrauma patients treated at ATCs versus PTCs using data extracted from the PTOS database from 2003 to 2015. A total of 1,606 patients met inclusion criteria, and the authors found no significant difference in mortality between PTCs and ATCs (AOR, 1.10; 95% CI, 0.54–2.24; p = 0.794). A study performed by Walther et al.21 examined the differences between adolescent (15–19 years old) trauma patient outcomes at PTCs versus ATCs using Level I trauma center data from the National Trauma Data Bank from 2007 to 2011. Of the 12,861 adolescents, shorter LOS (p = 0.03) and higher home discharge rates (p < 0.01) were seen at PTCs, but there was no significant difference found in mortality between adolescent trauma patients treated at PTCs versus ATCs. Miyata et al.4 examined outcomes of pediatric patients (age, ≤18 years) with penetrating trauma treated at PTCs versus ATCs using data from the National Trauma Data Bank from 2007 to 2011. There were 26,276 eligible patients (PTC, n = 3737; ATC, n = 22,539) and AORs for mortality showed no significant difference in survival outcomes (odds ratio, 0.592; p = 0.054) between treatment at and ATC or PTC.

Pediatric traumatic centers had a lower odds of surgical intervention (AOR, 0.67; 95% CI, 0.48–0.93; p = 0.016) compared with their ATC counterparts. Those who required surgery had significantly higher maximum AIS abdomen scores (p < 0.001) than those who did not require surgery (Table 3A). One potential reason for the decreased operative intervention in PTCs could be related to the practical philosophical approach to certain types of injuries. For instance, the pediatric trauma surgeon may take a more “watch and wait” attitude to a tangential gunshot wound (GSW) to the abdomen or an anterior abdominal stab wound (SW) than the adult trauma surgeon who may be inclined more toward immediate exploration. The pediatric surgeon decision making would be drawing from the immense success realized in the nonoperative management of solid organ injuries in the last 30 years (and which now adult trauma surgeons also use) (Table 3B). Of note, there was a greater percentage of SW versus GSW taken care of at the PTC (Table 4). The relative increase in SW may account for the decrease in surgery at PTCs, as these injuries are less likely to require surgery than GSWs, per se.

The first study regarding nonoperative treatment of traumatic injuries in children was first reported in 1968 at Sick Children's Hospital in Toronto and specifically looked at nonoperative management of splenic trauma.22 This study produced promising results of nonoperative management in pediatric trauma. The results led to much more focused investigations on nonoperative versus operative management of traumatic injuries. Many adult trauma centers began adopting a more conservative approach to treat specific types of pediatric traumatic injuries. Matsushima et al. found that in adolescent (13–18 years old) trauma patients with blunt abdominal solid organ injuries (SOI), operative intervention was more likely to be performed at an ATC versus a PTC.23 Walther et al.21 found imaging and invasive procedures were more common at ATCs (p < 0.01). Safavi et al.24 also found that operative management for blunt abdominal SOI in adolescents occurred more at ATCs. Most of the research surrounding nonoperative versus operative management in adolescents specifically focuses on blunt trauma, not penetrating as we analyzed in our study, making our study a novel concept.

While outcomes can be simply functional, psychosocial development is an important area to consider in adolescents. Due to this, trauma care given to adolescents should incorporate psychosocial development. Pediatric trauma centers are better equipped to handle psychological trauma through clinically trained child life specialists. On top of the original traumatic injury, invasive procedures have been found to evoke a stress response in pediatric patients. Many PTCs have specific programs and policies in place to aid in preventing or reducing traumatic stress, therefore improving outcomes in these young patients.16

This study is not without its limitations. As this study was retrospective and only used data from the PTOS database, the findings may be inapplicable to the greater trauma population. It may be advantageous to include multiple statewide trauma registries to achieve a comprehensive interpretation. The zip code of injury is not available in the PTOS database, only home zip code, which could have impacted the study results. Pennsylvania Trauma Outcome Study does not have the means to assess decision making surrounding nonoperative management, which may have created selection bias in our results. Prehospital bias with regard to transporting GSW to ATC over PTC was not examined. This could have impacted the number of GSW patients seen at ATC versus PTC. We also did not look at functional outcomes of adolescent trauma patients which may not produce an accurate representation of the function of the adolescent following the traumatic injury. Mortality in general is fortunately low in pediatric trauma victims, and hence not always the best measure of quality of care. This study, like most in the trauma literature, is limited by our lack of long term follow-up. Issues regarding recidivism, return to school, long-term disability, and quality of life are clearly important to assess in a young population.


Although the optimal management facility for adolescent trauma remains controversial, the results of this investigation suggest adolescent penetrating trauma patients treated at PTC have increased rates of nonoperative management with improved mortality compared with ATC counterparts. This is the first investigation to the authors' knowledge that has analyzed penetrating injury outcome differences between PTC and ATC in an adolescent population.


F.R. participated in the study design, editorial oversight, interpretation of data. M.H. participated in the data collection and analysis. M.M. participated in the article preparation. T.V. participated in the article preparation and data acquisition. B.G. participated in the editorial oversight, and interpretation of data. A.R. participated in the editorial oversight. B.G. participated in the article preparation and editorial oversight. A.C. participated in the data interpretation. E.B. participated in the editorial oversight.


The opinions and views expressed in this article are solely those of the author(s) and do not represent an endorsement by or position of the Pennsylvania Trauma Systems Foundation.

All authors have neither conflicts of interest nor disclosures of funding or proprietary interest to declare on the materials or subject matter discussed herein.


1. National Center for Health Statistics, Office of Analysis and Epidemiology. Leading causes of death and numbers of deaths, by age: United States, 1980 and 2016. Available at: Centers for Disease Control and Prevention. Atlanta, GA. Accessed June 15, 2019.
2. Cotton BA, Nance ML. Penetrating trauma in children. Semin Pediatr Surg. 2004;13(2):87–97.
3. American College of Surgeons Committee on Trauma. Part 1: A Brief History of Trauma Systems. Accessed June 17, 2019.
4. Miyata S, Cho J, Lebedevskiy O, Matsushima K, Bae E, Bliss DW. Trauma experts versus pediatric experts: comparison of outcomes in pediatric penetrating injuries. J Surg Res. 2017;208:173–179.
5. Notrica DM, Weiss J, Garcia-Filion P, Kuroiwa E, Clarke D, Harte M, Hill J, Moffat S. Pediatric trauma centers: correlation of ACS-verified trauma centers with CDC statewide pediatric mortality rates. J Trauma Acute Care Surg. 2012;73(3):566–570.
6. Webman RB, Carter EA, Mittal S, Wang J, Sathya C, Nathens AB, Nance ML, Madigan D, Burd RS. Association between trauma center type and mortality among injured adolescent patients. JAMA Pediatr. 2016;170(8):780–786.
7. Matsushima K, Schaefer EW, Won EJ, Nichols PA, Frankel HL. Injured adolescents, not just large children: difference in care and outcome between adult and pediatric trauma centers. Am Surg. 2013;79(3):267–273.
8. Walther AE, Pritts TA, Falcone RA, Hanseman DJ, Robinson BR. Teen trauma without the drama: outcomes of adolescents treated at Ohio adult versus pediatric trauma centers. J Trauma Acute Care Surg. 2014;77(1):109–116.
9. Gross BW, Edavettal MM, Cook AD, Rinehart CD, Lynch CA, Bradburn EH, Wu D, Rogers FB. Big children or little adults? A statewide analysis of adolescent isolated severe traumatic brain injury outcomes at pediatric versus adult trauma centers. J Trauma Acute Care Surg. 2017;82(2):368–373.
10. Mollberg NM, Tabachnick D, Lin FJ, Merlotti GJ, Varghese TK, Arensman RM, Massad MG. Age-associated impact on presentation and outcome for penetrating thoracic trauma in the adult and pediatric patient populations. J Trauma Acute Care Surg. 2014;76(2):273–277; discussion 277-8.
11. Brown JB, Rosengart MR, Kahn JM, Mohan D, Zuckerbraun BS, Billiar TR. Impact of volume change over time on trauma mortality in the United States. Ann Surg. 2017;266(1):173–178.
12. Pasquale MD, Peitzman AB, Bednarski J, Wasser TE. Outcome analysis of Pennsylvania trauma centers: factors predictive of nonsurvival in seriously injured patients. J Trauma. 2001;50(3):465–474.
13. Dudley RA, Johansen KL, Brand R, Rennie DJ, Milstein A. Selective referral to high-volume hospitals estimating potentially avoidable deaths. JAMA. 2000;283(9):1159–1166.
14. Wesson DE, Coselli JS. Pediatric trauma centers—coming of age. Tex Heart Inst J. 2012;39(6):871–873.
15. American Academy of Pediatrics Committee on Pediatric Emergency Medicine. Management of pediatric trauma. Pediatrics. 2016;138(2):e20161569.
16. Rokach A. Psychological, emotional and physical experiences of hospitalized children. Clin Case Rep Rev. 2016;2(4):399–401.
17. American College of Surgeons. Resources for Optimal Care of the Injured Patient: Committee on Trauma. Revised April 2014. Accessed November 2019.
18. Brown JB, Gestring ML, Leeper CM, Sperry JL, Peitzman AB, Billiar TR, Gaines BA. The value of the injury severity score in pediatric trauma: time for a new definition of severe injury? J Trauma Acute Care Surg. 2017;82(6):995–1001.
19. Brown JB, Gestring ML, Leeper CM, Sperry JL, Peitzman AB, Billiar TR, Gaines BA. Characterizing injury severity in nonaccidental trauma: does injury severity score miss the mark? J Trauma Acute Care Surg. 2018;85(4):668–673.
20. Rogers AT, Gross BW, Cook AD, Rinehart CD, Lynch CA, Bradburn EH, Heinle CC, Jammula S, Rogers FB. Outcome differences in adolescent blunt severe polytrauma patients managed at pediatric versus adult trauma centers. J Trauma Acute Care Surg. 2017;83(6):1082–1087.
21. Walther AE, Falcone RA, Pritts TA, Hanseman DJ, Robinson BR. Pediatric and adult trauma centers differ in evaluation, treatment, and outcomes for severely injured adolescents. J Pediatr Surg. 2016;51(8):1346–1350.
22. Upadhyaya P, Simpson JS. Splenic trauma in children. Surg Gynecol Obstet. 1968;126(4):781–790.
23. Matsushima K, Kulaylat AN, Won EJ, Stokes AL, Schaefer EW, Frankel HL. Variation in the management of adolescent patients with blunt abdominal solid organ injury between adult versus pediatric trauma centers: an analysis of a statewide trauma database. J Surg Res. 2013;183(2):808–813.
24. Safavi A, Skarsgard ED, Rhee P, Zangbar B, Kulvatunyou N, Tang A, O'Keeffe T, Friese RS, Joseph B. Trauma center variation in the management of pediatric patients with blunt abdominal solid organ injury: a national trauma data bank analysis. J Pediatr Surg. 2016;51(3):499–502.


KATHRYN D. BASS, M.D. (Buffalo, New York): I would like to thank the AAST for the privilege of previewing this well-written manuscript and discussing this presentation.

I congratulate my colleagues from Pennsylvania for this provocative paper on comparing the outcomes of adolescents treated for penetrating injuries at adult versus pediatric trauma centers in Pennsylvania.

The authors hypothesized that adult centers with higher volume of penetrating injuries overall will have a lower mortality when treating adolescents.

There are ample examples in the literature describing the volume quality curve; however, based on the interpretation of their data, the authors reach a surprisingly contrary conclusion.

Using the Pennsylvania Trauma Outcome Study Database and both univariate and multilevel logistic models, they concluded that pediatric trauma centers outperform adult trauma centers for adolescent penetrating trauma, with better survival, same length of stay, fewer operative interventions, and no difference in complications.

As a pediatric surgeon, this all sounds very validating to me, but I have a few comments and questions for the authors after a careful review of the data with our biostatistician.

Access to pediatric trauma centers across Pennsylvania varied, with 6 PTCs mainly clustered in the east, compared to the 33 ATCs. The mean distance traveled was three miles greater to the ATCs compared to PTCs, which may contribute to the survival benefit at the PTCs.

Of note, the number of penetrating injuries during the 15-year period declined in both adult and pediatric centers. Though this was not statistically significant, it was a trend year-over-year in the dataset.

This is in contrast to some other locations in the country, and I wondered whether you had any rationale for why this might be. Are there any changes in state policy that might explain?

The univariate analysis of the adult and pediatric centers demonstrated some statistical differences, with the ATCs treating an older population of adolescents.

And though the physiology of adolescents from 15-18 years old is unlikely different, this may describe the differences observed in the mechanism of injuries, with the greater number of assault firearm injuries in the ATCs, and thus the suggestion of higher risk of mortality in this group.

Also, there are more severely injured patients in the ATC group based on ISS. When examining injuries specific to body compartments – head, chest, and abdomen AIS - there were no statistical differences between groups.

However, when examining the AIS categories of head, chest, and abdomen, there were greater numbers of severe, critical, and un-survivable patients in the ATC group, again suggestive of a greater mortality risk at the ATC.

With regard to the models that were used to predict mortality and outcomes, I wonder why you did not include age or injury type as a covariate.

And, why did you selectively choose head AIS for predicting mortality, abdomen AIS for predicting complications in surgeries, and include the three AIS categories – head, chest, and abdomen – for predicting length of stay? Did this bias your conclusion?

In my region, I know from working with prehospital providers, that their decision-making regarding destination, biases where patients land. In other words, there is a prehospital bias towards bringing a gunshot wound to adult centers, although this is not part of the traditional field triage criteria. This bias appeared to be at play in your dataset as well, and I wonder whether you have a way to explore this.

Finally, I congratulate the authors again for their study. I support this research as it digs into the outcomes of adult and pediatric trauma centers, and I look forward to learning more about how we can best care for the injured child at both centers. Thank you.

PATRICK REILLY, M.D. (Philadelphia, Pennsylvania): Dr. Rogers, as always, great work. Pennsylvania has a number of centers that are credentialed as both adult and pediatric trauma centers, so how did you account for a patient who showed up at one of those facilities? And, when they showed up at those facilities do you know that they were cared for by a pediatric as opposed to an adult surgeon?

JOHN R. HALL, M.D. (Hilton Head Island, South Carolina): In 1986, we at Cook County Hospital did the first paper, over 2000 kids from various databases, showing the very significant improvement and outcome in blunt trauma in children.

However, in penetrating injuries, it was equal, with the adult centers having a slight non-significant increase in survival.

Since that time, the incidents of gunshot wounds have increased for most pediatric trauma centers, with some being over 50 percent, which, as Dr. Bass said, may give them more experience and more comfort in doing it.

However, I really think to make this paper valid, you need to eliminate the knife wounds. It's the gunshot wounds that make the big difference, and I would propose that the adult centers who see more adult gunshot wounds would have a better outcome.

CHRIS NEWTON, M.D. (Oakland, California): So, I want to applaud the effort for looking at this problem, because we struggle with this in almost every community of what to do., and in some ways, I look back – I want to echo some of the previous comments – I think there is one important aspect in your data, you're missing – the population is just different at most the adult only vs. and pediatric only hospitals.

Many of us would treat gunshot wounds the same way at either place; so, the difference winds up being hard to clarify. There’re different circumstances with kinds of gunshot wounds coming from rural or urban or whatever.

Can you get that kind of data from your Pennsylvania stuff to clarify a little bit better of the mechanism type and the parameters around the patient, rather than just the age?

MICHAEL L. NANCE, M.D., (Philadelphia, Pennsylvania): Can you clarify what DOA was? I think you said you excluded them, and I think when you said that, you said that it was a death within an hour. If that is the case, do you think there's a difference between somebody who arrives without vitals and then somebody who may be in the trauma center for an hour and might even make it to the operating room and undergo intervention?

FREDERICK B. ROGERS, M.D., M.S., F.A.C.S., (Lancaster, Pennsylvania): I’d like to thank Dr. Bass and all the other commenters for their very pertinent and insightful questions.

The first question was about the declining penetrating over the time course of the study. I don’t really have a good explanation for that. It was something that e noticed, but it wasn’t significant, but it definitely was a factor.

Another question was about the use of the maximal AIS score of head, chest, and abdomen as one of the variates in our univariate analysis, why did we choose those? We used those variables because there is some literature to support their use, specifically in the pediatric patient—they may be better predictors, even than ISS.


Pediatric trauma center; adult trauma center; penetrating trauma; nonoperative management; mortality

Copyright © 2020 American Association for the Surgery of Trauma.