A Re-Evaluation of the Effect of Trauma Center Verification Level on the Early Risk of Death in Hemodynamically Unstable Patients

Background Studies show increased early and overall mortality at level II compared to level I trauma centers in hemodynamically unstable patients. We hypothesize there is no mortality difference between level I and level II centers applying more contemporary data. Study design Utilizing the 2017 Trauma Quality Program Participant Use File (TQP-PUF), we identified adult patients (age >14 years) who presented to an American College of Surgeons (ACS) verified level I or II center with hypotension (systolic blood pressure [SBP] < 90 mmHg). Logistic regression was performed to identify adjusted associations with mortality. Results A total of 7,264 patients met the inclusion criteria, of whom most were males (4,924 [67.8%]) with blunt trauma (5,924 [81.6%]) being predominated. Mean admission SBP was 73.2 (±13.0) mmHg. There were 1,097 (15.1%) deaths. Level I admissions (4,931 (67.9%]) were more likely male (3,389 [68.7%] vs. 1,535 [65.8]; p=0.012), non-white (3,119 [63.3%] vs. 1,664 [71.3%]; p<0.001), a victim of penetrating trauma (933 [18.9%] vs. 385 [16.5%]; p=0.015), and more severely injured (mean Injury Severity Score: 19.3 [±15] vs. 16.7 [±13.7]; p<0.001). Level II admissions (2,333 [32.1%]) were older (46.8 [±18.5] vs. 50.3 [±20.1] years; p<0.001) with more co-morbidities (mean Charlson Comorbidity Index: 1.43 [±2] vs. 1.77 [±2.2]; p<0.001). Adjusted mortality between level I and II admissions was similar (766 [15.5%] vs. 331 [14.2%]; p=0.918). Early hourly mortality also did not differ. Conclusion There is no overall or hourly mortality discrepancy between ACS-verified level I and II centers for patients presenting with hypotension. This potentially relates to the use of more contemporary data gathered after implementation of updated verification requirements.


Introduction
Previous literature outlines significant outcome differences between level I and level II trauma centers. Superior outcomes are reported at level I centers for the severely injured [1,2], traumatic brain injury (TBI) patients [3,4], those with other specific injuries [5], and overall mortality [6,7]. Less prevalent are studies showing improved outcomes or equivalency at level II centers [8][9][10][11]. Of significance to our study, previous data show that trauma patients presenting with hemodynamic instability have significantly lower mortality in level I versus level II centers and that this discrepancy is sustained during the first hours of admission [2]. It was hypothesized that level II centers have access to inferior resources. However, during the time many of these investigations were being reported, there were differences in clinical requirements at level I versus level II trauma centers.
Since 1976, the American College of Surgeon Committee on Trauma (ACS-COT) has issued trauma center resource guidelines. "Resources for the Optimal Care of the Injured Patient" (resources manual) emphasizes the importance of a systems-based approach mandating escalating clinical resources at higher level trauma centers [12]. The 2014 update mandated equivalent clinical resources at level I and II centers so that, theoretically, outcomes would be similar. However, there are little recent data to support this. We hypothesize that more contemporary analysis would support improved outcomes at level II centers relative 1 2 1 1 2 to their level I counterparts in patients who present with hemodynamic instability [2].

Materials And Methods
Utilizing the 2017 Trauma Quality Program Participant Use File (TQP-PUF), we identified adult patients (age >14 years) who presented to an ACS-COT verified level I or II trauma center with hemodynamic instability (systolic blood pressure [SBP] < 90 mmHg) [2]. We excluded patients with isolated TBI and interfacility transfers. Isolated TBI was identified by an Abbreviated Injury Scale (AIS) score for head of ≥3 with an AIS score for all other body regions of <3 [3]. We extracted all pertinent demographic and injury variables. This included, but was not limited to, gender, race, E code mechanism (mechanism), admission Glasgow Coma Scale (GCS) score, Injury Severity Score (ISS), and the presence of medical co-morbidities (

Results
There were 7,264 patients meeting the inclusion criteria ( Figure 1).    On logistic regression ( Table 4), admission GCS < 9, ISS > 15, age > 60 years, hypoxia, payor group, mechanism, and the presence of a TBI were independently associated with in-hospital death, but admission to a level I versus a level II center was not (1.009 [0.851-1.196]; p=0.918). The hourly risk of death, similarly, was not statistically significant with admission to a level I versus II center (Figure 2).

Discussion
Literature is replete with studies addressing the outcome differences between level I and II trauma centers. Significant mortality increases at level II centers for specific subsets have been demonstrated. These include the severely injured [1,2], those with specific wounding [5], those with TBI [8,11], those transferred to a level II center after TBI [13], and overall populations [6,7]. Of specific interest to our study, those admitted with hemodynamic instability have been shown to have better mortality outcomes at a level I center that is sustained through early admission [2]. Of concern is that these reports may utilize obsolete data gathered prior to the 2014 "resources manual" update [12]. The previously mentioned study [2] on hypotensive trauma patients is an example of one of the more egregious confounding as it utilized data nearly over a decade old to support conclusions that were no longer valid at the time of publication.
The 2014 revision of the ACS-COT resources manual (Orange Book) specifically addresses clinical resourcing discrepancies between level I and II ACS-COT verified centers. "Level I and II criteria were revised to ensure that level I and II trauma centers are available to provide high quality definitive care" [12]. The assumption is that this would lead to reduction or elimination outcome discrepancies, though there are little data to support this. Our current investigation demonstrates this relative improvement and contradicts earlier reports. Furthermore, previous studies that demonstrate outcome equivalence or even improved outcomes at level II centers used locoregional data [8,14].
Our investigation shows no mortality difference between admission to a level I versus level II center in patients who present with hemodynamic instability. We hypothesize that this is associated with utilization of more contemporary datasets that were gathered after implementation of the 2014 ACS-COT "resources manual". Updated requirements include, but are not limited to, uninterrupted emergency medicine staffing, more stringent operating room resourcing with performance tracking, defined minimums for highest level activations, minimum registrar training requirements, dedicated injury prevention positions, equivalent surgical and non-surgical subspecialty services, and changes in guidance for consultant bedside presence. One of the more pertinent improvements for both level I and II centers relate to performance improvement mandating exacting identification and trending of important outcome and process metrics. In addition, participation in a risk-adjusted outcome benchmarking program (Trauma Quality Improvement Program [TQIP]) was a significant new mandate [15]. Elements for a level I center, not required of a level II center, are admission volume minimums, the presence of higher level surgical resident trainees, a surgically directed intensive care unit, and minimum research productivity.
Despite a reasonably valid clinical association between improved resources and better outcomes, our study is hindered by its retrospective design. Trauma care at level II centers could have simply improved over time with the introduction of new techniques and protocols, though this is doubtful relative to outcome improvements related to mandated improved resourcing. While the 2017 TQP-PUF is a powerful tool to assess the impact of sweeping administrative mandates, it lacks the granularity to define elements that may have had an impact on our findings or that could have contributed to confounding. There were significant differences between level I and level II admissions. Most notably, level I admissions were more severely injured and differed demographically and by mechanism. Despite significant differences, adjusted mortality that included these variables was similar between level I and II centers and sustained hourly through the first 24 hours of admission. Of note, co-morbidities were not associated with death. This could possibly be due to survivors having more opportunity for their care teams to identify and document co-morbidities.
Applicability of our study extends only to ACS-COT verified centers regardless of the designating authority. While the scope of this organization is broad and many local designating authorities model their verification efforts to mirror the ACS-COT process and requirements, not all designated trauma centers are verified by this committee. It may be reasonable to assume that level II trauma centers verified by their local designating authority or other non-ACS-COT construct also experience improved outcomes relative to level I; however, our study excluded these trauma centers.
Despite these shortcomings, our study is impactful since it demonstrates, in contradiction to older published data, that level II trauma centers achieve similar outcomes in patients who present with hemodynamic instability compared to level I centers. In this subgroup, immediate presence of trauma surgeons, competent consultants, and timely availability of interventions are critical [15]. This change likely relates, in part, to the updated requirements that these elements be in place at a level II center just as they are at the level I center. Additionally, our study demonstrates the potentially significant and widespread beneficial impact of ongoing process improvement, resource standardization, and involvement of a national verification program. It is important for the public and policy-makers since it supports that the significant investment in a level II can be expected to generate outcomes similar to those at a level I center.

Conclusions
As opposed to previous studies, level II trauma centers perform similar to level I centers for trauma patients who present with hemodynamic instability. This may relate to compliance with the ACS-COT resources manual, Resources for the Optimal Care of the Injured Patient, 2014 version. Further study would include examination of trauma centers that are not verified by the ACS-COT that could reveal more specific variables associated with improved outcome in these patients.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.