a b s t r a c t

Introduction: To identify the association between Skull fracture (SF) and in-hospital mortality in patients with se- vere Traumatic brain injury .

Materials and methods: This Multicenter cohort study included a retrospective analysis of data from the Japan Trauma Data Bank (JTDB). JTDB is a nationwide, prospective, observational trauma registry with data from 235 hospitals. Adult patients with severe TBI (Glasgow Coma Scale <9, head Abbreviated Injury Scale >= 3, and any other AIS < 3) who were registered in the JTDB between January 2004 and December 2017 were in- cluded in the study. Patients who (a) were < 16 years old, (b) developed cardiac arrest before or at hospital ar- rival, and (c) had burns and Penetrating injuries were excluded from the study. In-hospital mortality was the primary outcome assessed. Multivariable logistic regression analyses were performed to calculate the adjusted odds ratios (ORs) of SF and their 95% confidence intervals (CIs) for in-hospital mortality.

Results: A total of 9607 patients were enrolled [median age: 67 (interquartile range: 50-78) years] in the study. Among those patients, 3574 (37.2%) and 6033 (62.8%) were included in the SF and non-SF groups, respectively. The overall in-hospital mortality rate was 44.1% (4238/9607). A multivariate analysis of the association between SF and in-hospital mortality yielded a crude OR of 1.63 (95% CI: 1.47-1.80). A subgroup analysis of the association of skull vault fractures, skull base fractures, and both fractures together with in-hospital mortality yielded ad- justed ORs of 1.60 (95% CI: 1.42-1.98), 1.40 (95% CI: 1.16-1.70), and 2.14 (95% CI: 1.74-2.64), respectively, rel-

ative to the non-SF group.

Conclusions: This observational study showed that SF is associated with in-hospital mortality among patients with severe TBI. Furthermore, patients with both skull base and skull vault fractures were associated with higher in-hospital mortality than those with only one of these injuries.

(C) 2021

1. Introduction

traumatic brain injury is a critical condition complicated by multiple traumas, often resulting in traumatic death [1,2]. TBI is an over- whelming and major global Public health problem, and one of the most important causes of morbidity and mortality in both industrialized and Developing countries [3]. A skull fracture (SF) occurs in response to a

* Corresponding author at: 465 Kajii-cho, Kamigyo-ku, Kyoto-shi, Kyoto 565-0871, Japan.

E-mail addresses: [email protected] (G. Fujiwara), [email protected] (Y. Okada), [email protected] (W. Ishii), [email protected] (R. Iizuka), [email protected] (T. Sakakibara), [email protected] (T. Yamaki).

force that is strong enough to break the skull bone, which means SF can be an objective indicator of the extent of the impact of the head and primary brain injury. Therefore, we hypothesized that SF might be associated with the overall prognosis and mortality in patients with TBI. Previous basic studies of TBI in mice have reported that SF is associ- ated with the release of inflammatory cytokines and poor prognosis [4]. In addition, patients with SF have been reported to have an increased risk of intracranial lesions requiring surgical intervention [5,6]. Al- though there is a fairly widespread agreement that SF increases the risk of complications, there are only a few studies that have reported the association between SF and mortality [7-9], and the value of SF as a predictor of mortality is unclear. This study aimed to identify the asso- ciation between SF on admission and in-hospital mortality among pa- tients with severe TBI to improve the quality of trauma care provided.

https://doi.org/10.1016/j.ajem.2021.03.020

0735-6757/(C) 2021

1. Materials and methods

Our retrospective study was conducted in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [10].

• 1. Study design and setting

The data were obtained from the Japan Trauma Data Bank (JTDB) dataset, which contains information for 235 hospitals across Japan, be- tween January 2004 and December 2017. The JTDB is a nationwide, mul- ticenter, prospective, observational trauma registry, which was established in 2003 by the Trauma Registry Committee of the Japanese Association for the Surgery of Trauma and the Committee for Clinical Care Evaluation of the Japanese Association for Acute care medicine [11], and is managed by the Japan Trauma Care and Research (JTCR), which is a non-profit organization for trauma research. This registry was developed to improve the quality of trauma care through the collec- tion of prehospital information, clinical information during the hospital stay, data on trauma diagnoses based on the Abbreviated Injury Scale and Injury Severity Score (ISS), and data on mortality-related out- comes. These data are compiled by the administrators based on the in- hospital charts and prehospital records, which are routinely submitted by paramedics to the hospitals. Nearly all the institutions participating in the JTDB are government-certified tertiary emergency and critical care centers [11-13]. Details on the JTDB have been published previ- ously [14]. Anonymized data from the JTDB are available to institutional members of the JTCR for research, and we obtained these data from the JTCR.

• 1. Study participants

We included adult (>=16 years old) trauma patients with isolated se- vere head injury registered in the JTDB data set. An isolated severe head injury was defined as one with a Glasgow Coma Scale score of <9 on arrival at the hospital, head AIS score >= 3, and other AIS score < 3, based on the previous reports [15,16]. The exclusion criteria were:

(a) less than 16 years of age, (b) cardiac arrest (predefined as a heart rate [HR] of 0 and systolic blood pressure [SBP] of 0) at the injury scene, (c) main causes of injury were burns or any penetrating or un- known injuries, and (d) information on the primary outcome (in-hospi- tal mortality) was missing.

• 1. Data collection, variables, and potential bias

We collected and described the following clinical information from the JTDB: sex, age, mechanism of injury (e.g., bicycle, pedestrian fall), vital signs [SBP, HR, and respiratory rate (RR)], GCS score on arrival at the hospital, AIS score, ISS, and in-hospital mortality. Patients were stratified by age into the following groups: < 40, 40-49, 50-59, 60-69, 70-79, and >=80 years. They were additionally categorized by vital signs as follows: SBP, < 90, 90-139, 140-199, and >=200 mmHg; RR, < 10, 10-29, and >= 30 bpm; body temperature (BT), <36.0 and

>=36.0 degree Celsius; and GCS score, 3, 4-5, and 6-8, based on the disas- ter triage protocol [17]. The missing covariates were categorized as ‘Unknown.’

The AIS is an anatomically based, consensus-derived, and globally accepted severity scoring system used to classify the injuries by body re- gion (1: head, 2: face, 3: neck, and 4: thorax) and subsequently by the relative severity on a 6-point scale (1: minor to 6: maximal) [18,19]. Generally, an AIS score >= 3 is considered a serious or more severe injury. The ISS is another globally accepted trauma severity scale associated with mortality. ISS scores were calculated using AIS and ranged from 1 to 75 [19].

Patients enrolled in JTDB have an AIS code entered for each injury. The AIS is coded with seven digits; the first six digits indicate the trauma

classification followed by a severity code (1-6). The code for an SF is “150” in the upper three digits. The presence of an SF is indicated by in- cluding the code of “150” in the upper three digits of the entered AIS. For example, the upper four digits “1502” means a skull base fracture, and “1504” means a skull vault fracture.

• 1. Primary outcome

The primary outcome assessed in this study was in-hospital mortal- ity, defined as death at any time before hospital discharge, registered in the JTDB.

• 1. Statistical analysis

Based on the presence and absence of an SF, patients were catego- rized into SF and non-SF groups, respectively. For the two groups, we used SF as the explanatory variable and mortality as the objective vari- ables. We conducted logistic regression analyses to generate crude odds ratios (ORs) for the SF group with 95% confidence intervals (CIs) and compared them to those of the non-SF group. For potential confounders, sex, categorized age, mechanism of injury, categorized RR, HR, BT, GCS score, and ISS were designated as covariates. We conducted multivari- able logistic regression analyses to calculate the adjusted ORs for SF and non-SF groups with their 95% CIs.

Further, we performed a subgroup analysis to determine the associ- ation of skull vault fractures and skull base fractures with mortality. The statistical results were calculated as point estimates with 95% CIs. Statis- tical significance was defined as the absence of overlap of 95% CI with the null effect value (OR = 1). All statistical analyses were performed using JMP Pro(R) 14 software (SAS Institute Inc., Cary, NC, USA).

• 1. Sensitivity analysis

In addition, we performed a sensitivity analysis by adding the sever- ity of the Intracranial injury as a confounding factor. In the AIS codes, in- tracranial injury (e.g., contusion, subdural hematoma, epidural hematoma, subarachnoid hemorrhage) begins with 140 in the upper three digits. We extracted this code, and the AIS score for the severity of the intracranial injury was taken. For patients without AIS score for intracranial injury, it was considered as “< 3.”

1. Results
   1. Patient characteristics

Fig. 1 shows the study flowchart. Of 294,274 patients registered in the JTDB, 9607 were included in the analysis: 3574 (37.2%) in the SF group, and 6033 (62.8%) in the non-SF group. The characteristics and clinical data of the patients are summarized in Tables 1 and 2. More than two-thirds (69.2%) of the patients were men, and the median age was 67 [interquartile range (IQR), 50-78] years. The median GCS score and ISS of the patients were 5 (IQR, 3-7) and 25 (IQR, 16-25), respectively.

• 1. Outcomes

The overall in-hospital mortality rate was 44.1% (n = 4238). The dis- tribution of outcomes by each group is shown in Table 3. Univariate lo- gistic regression analyses of the association between SF and in-hospital mortality yielded a crude OR of 1.42 (95% CI: 1.31-1.55), relative to non- SF. In the multivariable logistic regression analysis, the corresponding adjusted OR was 1.63 (95% CI: 1.47-1.80) (Fig. 2A). The crude and ad- justed ORs for the other covariates are shown in the supplementary file.

Fig. 1. Flow chart of patient selection. JTDB: Japan Trauma Data Bank, GCS: Glasgow Coma Scale, TBI: traumatic brain injury.

Table 1

Characteristics of the study participants

Table 2

Clinical data of the study participants

Parameters

Total

SF

Non-SF

Parameters

Total

SF

Non-SF

N = 9607

N = 3574

N = 6033

N = 9607

N = 3574

N = 6033

Sex (male), n, (%)

6653(69.3)

2654(74.3)

3999(66.3)

Heart rate, n, (%) >=100 bpm

2903(30.2)

1077(30.1)

1826(30.3)

Age(years), n, (%)

60-99	5578(58.0)	2041(57.1)	3537(58.6)
<60	783(8.2)	343(9.6)	440(7.3)
Unknown	343(3.6)	113(3.2)	230(3.8)

50-59	11,101(11.6)	449(12.6))	661(11.0)	Systolic blood pressure, n, (%)
60-69	1888(19.7)	747(21.0)	1141(18.9)	>=140	5809(60.5)	2091(58.5)	3718(61.6)
70-79	2133(22.2)	769(21.5)	1364(22.6)	90-139	3071(32.0)	1170(32.7)	1901(31.5)
80-	2087(21.7)	541(15.1)	1546(25.6)	<90	594(6.2)	269(7.5)	325(5.4)
Unknown 133(1.4) 44(1.2)							89(1.5)
Bicycle 1151(12.0) 599(16.8) 552(9.2) Respiratory rate, n, (%)
Fall	1677(17.5)	776(21.7)	901(14.9)	<10	533(5.5)	217(6.1)	316(5.2)
Free fall	653(6.8)	354(9.9)	299(5.0)	11-29	7166(74.6)	2586(72.4)	4580(75.9)
Motor bike	855(8.9)	334(9.3)	521(8.6)	>=30	951(9.9)	406(11.4)	545(9.0)
Car	438(4.6)	145(4.1)	293(4.9)	Unknown	957(10.0)	365(10.2)	592(9.8)

-39	1673(17.4)	754(21.1)	919(15.2)
40-49	716(7.5)	314(8.8)	402(6.7)

Mechanism, n, (%)

Body temperature, n, (%)

Pedestrian	1174(12.2)	525(14.7)	649(10.8)
Slip	2505(26.0)	474(13.3)	2031(33.7)
Others	1154(12.0)	367(10.3)	787(13.0)

<36.0		4896(51.0)	1291(36.1)	1915(31.7)
>=36.0		3206(33.4)	1698(47.5)	3198(53.0)
SF: skull fracture. unknown		1505(15.7)	585(16.4)	920(15.2)
Glasgow Coma Scale, n, (%)
3.3. Subgroup analysis 3 3188(33.2) 1253(35.1)				1935(32.1)
4-5		1753(18.2)	585(16.4)	1168(19.4)
6-8		4666(48.6)	1736(48.6)	2930(48.6)
The results of the subgroup analysis for in-hospital mortality are shown in Fig. 2B. Multivariable logistic regression analyses of the asso- Intracranial lesion (AIS), n, (%)
ciation of skull vault fracture, skull base fracture, and both fractures to-	<3 3	273(2.8) 1260(13.1)	74(2.1) 382(10.7)	199(3.3) 878(14.6)
gether with in-hospital mortality yielded adjusted ORs of 1.60 (95% CI:	4	2746(28.6)	1109(31.0)	1637(27.1)
1.42-1.98), 1.40 (95% CI: 1.16-1.70), and 2.14 (95% CI: 1.74-2.64), re-	5	5309(55.3)	1994(55.8)	3315(54.9)
spectively, relative to the non-SF group (reference). In summary, each	6	19(0.2)	15(0.4)	4(0.1)
type of fracture was associated with in-hospital mortality, but the asso-	ISS, median, (IQR)
ciation was stronger in the presence of both types of fractures.		25(16-25)	25(16-26)	25(16-25)

3.4. Sensitivity analysis

We performed a sensitivity analysis by adding the severity of the in- tracranial injury as a confounding factor. The adjusted OR of SF with in- hospital mortality was 1.64 (95% CI: 1.48-1.82), relative to the non-SF group, suggesting that the patients with SF were associated with in-hospital mortality. Our findings, therefore, suggest that SF is associ- ated with mortality independent of the intracranial injury.

Table 3

Primary outcome between two groups

Parameters	Total		SF		Non-SF
	N = 9607		N = 3574		N = 6033
Mortality, n, (%)	4238(44.1)		1774(49.6)		2464(40.8)

Fig. 2. The forest-plot of adjusted odds ratio for in-hospital mortality in SF group (2A), and in both-SF, vault-SF and base-SF (2B). The odds ratio adjusted by sex, categorized age, mechanism of injury, heart rate, systolic blood pressure, respiratory rate, categorized Glasgow Coma Scale, and Injury Severity Score. SF: skull fracture. both-SF: merging skull vault fracture and skull base fracture. vault-SF: skull vault fracture, base-SF: skull base fracture.

1. Discussion
   1. Key observations

This Multicenter observational study, including 235 hospitals, showed an association between SF and in-hospital mortality in patients with severe TBI. Further, the subgroup analysis showed higher ORs for in-hospital mortality when skull base and skull vault fractures were taken together than when they were considered separately.

• 1. Strengths of the study

Our study has several strengths compared to the previous studies. First, our findings on the association of skull base and skull vault frac- tures with in-hospital mortality are novel. While most previous studies have reported associations of SF with intracranial hemorrhage, surgical intervention [5,20], and neuropsychological dysfunction [21], its inde- pendent association with the prognosis of in-hospital mortality has not yet been reported. Although a few studies have reported mortality [7-9], they did not evaluate the type of SF. Thus, to the best of our knowl- edge, this is the first study to show a clear association between the de- gree of SF and in-hospital mortality. Second, our study overcomes the limitations of the previous studies and can be validated more objec- tively. Previous studies either had small sample sizes or included partic- ipants from a single center [7-9]. Moreover, all studies included exploratory analyses of the risk factors for TBI, which we believe are not adequately adjusted for confounding factors methodologically

[7-9]. However, the present study using a nationwide multicenter dataset had a large sample size, and so we could adjust for the potential confounding factors. We, therefore, believe our findings are more likely to be generalizable to all areas of Japan and possibly even to the devel- oped countries. Third, this study provides an opportunity for future clin- ical applications in terms of severity prediction. In the previous studies, the association of SF with intracranial hemorrhage and surgical inter- vention [5,6,20] was evaluated based on the computed tomography findings. CT findings can show SF, intracranial hemorrhage, and the necessity of surgical intervention at the same time, allowing us to judge them simultaneously. However, we evaluated the potential of SF as a prognostic indicator of mortality, which will be of use in an actual clinical setting to predict the clinical course of the patients.

• 1. Interpretation of the findings

We assume the following potential mechanisms to explain our results. SF is an evidence of a direct physical impact on the head and the brain, which may be life-threatening. As shown in previous studies, SF may reflect the magnitude of the impact on the head and can be considered as an objective indicator of direct injury [22-25]. Such an impact can cause micro intracranial hemorrhage or brain tissue injury, including Axonal injury, which cannot be detected on a CT scan. Additionally, such an impact may also induce neural tissue injury involving more intense neuroinflammation when compared to a head injury without SF, and could lead to coagulopathy [26- 28]. In a basic study using mice TBI model, SF elicited inflammatory cytokines, including tissue inhibitor of metalloproteinases 1 and tumor necrosis factor-?, leading to poor neurological outcomes [4]. The results of our study suggest that neuroinflammation due to cytokine release in TBI [29] could have been affected and amplified by SF. We also hypothesized that in a TBI setting, acute traumatic coagulopathy [26] is induced by tissue injury, including neural tissue injury caused by the SF itself. Based on the above potential mechanisms and our results, the detection of SF on CT can be interpreted not only as an objective indicator of primary injury but also as a factor causing Secondary damage, including the release of neuroinflammatory cytokines and coagulopathy.

In the subgroup analysis, no significant differences were observed between skull base fractures and skull vault fractures. However, pa- tients with both fractures had a greater risk of mortality. Fracture com- plications are assumed to be strongly associated with mortality due to the greater impact, which means greater primary injury. Skull vault fracture has been associated with intracranial hemorrhage in cases of “talk and die syndrome” [30]. In contrast, skull base fractures have been reported to potentially cause shock in patients with brainstem in- juries [31,32]. We hypothesized that the two fractures together might have contributed to poor prognosis by synergistically causing coagulop- athy and expansion of hematoma, with extensive brain tissue damage indicating severe neuroinflammation.

• 1. Clinical implications

Our findings can help clinicians estimate the severity and prognosis of TBI. We believe that these results may be useful in predicting the emergence or increase of intracranial hemorrhage and the need for sur- gical intervention. They may also help in confirming coagulopathy asso- ciated with TBI and predicting neuroinflammation. Furthermore, these results highlight the importance of avoiding direct damage and provide evidence supporting the benefits of wearing a helmet or other head pro- tective equipment [33-35].

• 1. Limitations

Our study has several limitations. First, excluding patients with missing data might have introduced a selection bias. Second, regional

or inter-hospital differences in the environment and type of primary care, including the decision to perform CT, might be an unmeasured confounding factor. Third, although the data are based on standardized information, there is some concern regarding their quality and mea- surement bias. Fourth, regarding objective validity, these results may not be applied to other countries with patient structures, injury pat- terns, and treatment systems different from Japan.

1. Conclusions

Patients with an SF on hospital arrival were found to be associated with in-hospital mortality. Moreover, subgroup analyses showed higher ORs for in-hospital mortality in the presence of both skull base and skull vault fractures than in the presence of one of these injuries. Further re- search is necessary to eliminate the potential biases and to validate these results.

Ethics approval

Ethics approval was obtained from the committees of the Japanese Association for the Surgery of Trauma, as well as each participating in- stitution. Following the Ethical Guidelines for Medical and Health Re- search Involving Human Subjects published by the Ministry of Health, Labor, and Welfare of Japan, and the use of anonymized data from the Japan Trauma Data Bank (JTDB), informed consent was waived. The ap- proval document from the Japanese Association for the Surgery of Trauma and the representative institution (National Defense Medical College Research Institute) are available on the JTDB website (https:// www.jtcr-jatec.org/traumabank/dataroom/ethics2.htm) (approval ID No. 2548).

Funding

Nothing to disclose.

Author contributions

GF and YO were involved in study design and Data interpretation. WI and RI were involved in the data collection. GF and YO were involved in the data analysis. All authors critically revised the report, commented on drafts of the manuscript, and approved the final report.

Declaration of Competing Interest

Nothing to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2021.03.020.

References

1. Tiret L, Hausherr E, Thicoipe M, Garros B, Maurette P, Castel JP, et al. The epidemiol- ogy of head trauma in Aquitaine (France), 1986: a community-based study of hospi- tal admissions and deaths. Int J Epidemiol. 1990;19:133-40. https://doi.org/10.1093/ ije/19.1.133.
2. Kraus JF, Black MA, Hessol N, Ley P, Rokaw W, Sullivan C, et al. The incidence of acute brain injury and serious impairment in a defined population. Am J Epidemiol. 1984; 119:186-201. https://doi.org/10.1093/oxfordjournals.aje.a113737.
3. Sosin DM, Thurman Jes DJ. Incidence of mild and moderate brain injury in the United States, 1991. Brain Inj. 1996;10:47-54. https://doi.org/10.1080/ 026990596124719.
4. Zvejniece L, Stelfa G, Vavers E, Kupats E, Kuka J, Svalbe B, et al. Skull fractures induce neuroinflammation and worsen outcomes after closed head injury in mice. J Neurotrauma. 2020;37:295-304. https://doi.org/10.1089/neu.2019.6524.
5. Faried A, Halim D, Widjaya IA, Badri RF, Sulaiman SF, Arifin MZ. Correlation between the skull base fracture and the incidence of intracranial hemorrhage in patients with traumatic brain injury. Chin J Traumatol = Zhonghua Chuang Shang Za Zhi. 2019; 22:286-9. https://doi.org/10.1016/j.cjtee.2019.05.006.
6. Munoz-Sanchez MA, Murillo-Cabezas F, Cayuela-Dominguez A, Rincon-Ferrari MD, Amaya-Villar R, Leon-Carrion J. Skull fracture, with or without clinical signs, in mTBI is an independent risk marker forneurosurgicallyrelevantintracranial lesion: a cohort study. Brain Inj. 2009;23:39-44. https://doi.org/10.1080/02699050802590346.
7. Fabbri A, Servadei F, Marchesini G, Stein SC, Vandelli A. Early predictors of unfavourable outcome in subjects with moderate head injury in the emergency de- partment. J Neurol Neurosurg Psychiatry. 2008;79:567-73. https://doi.org/10.1136/ jnnp.2007.120162.
8. Hsiao KY, Hsiao CT, Weng HH, Chen KH, Lin LJ, Huang YM. Factors predicting mortal- ity in victims of blunt trauma brain injury in emergency department settings. Emerg Med J. 2008;25:670-3. https://doi.org/10.1136/emj.2007.053405.
9. Tseng WC, Shih HM, Su YC, Chen HW, Hsiao KY, Chen IC. The association between skull bone fractures and outcomes in patients with severe traumatic brain injury. J Trauma – Inj Infect Crit Care. 2011;71:1611-4. https://doi.org/10.1097/TA. 0b013e31823a8a60.
10. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007; 147:573-7. https://doi.org/10.7326/0003-4819-147-8-200710160-00010.
11. Hondo K, Shiraishi A, Fujie S, Saitoh D, Otomo Y. In-hospital trauma mortality has de- creased in Japan possibly due to trauma education. J Am Coll Surg. 2013;217:850-7. https://doi.org/10.1016/j.jamcollsurg.2013.05.026.
12. Endo A, Shiraishi A, Matsui H, Hondo K, Otomo Y. Assessment of progress in early trauma care in japan over the past decade: achievements and areas for future im- provement. J Am Coll Surg. 2017;224:191-8 e5 https://doi.org/10.1016/j. jamcollsurg.2016.10.051.
13. Japan T. Trauma Data Bank Report 2005-2009. Trauma. 2009:2017.
14. Shibahashi K, Sugiyama K, Tomio J, Hoda H, Morita A. In-hospital mortality and length of hospital stay with craniotomy versus craniectomy for acute subdural hem- orrhage: a multicenter, propensity score-matched analysis. J Neurosurg. 2019:1-10. https://doi.org/10.3171/2019.4.JNS182660.
15. Genet GF, Johansson PI, Meyer MAS, Solbeck S, Sorensen AM, Larsen CF, et al. Trauma-induced coagulopathy: standard coagulation tests, biomarkers of coagulop- athy, and endothelial damage in patients with traumatic brain injury. J Neurotrauma. 2013;30:301-6. https://doi.org/10.1089/neu.2012.2612.
16. Samuels JM, Moore EE, Silliman CC, Cohen MJ, Ghasabyan A, Chandler J, et al. HHS Public Access, 86; 2019; 686-93. https://doi.org/10.1097/TA.0000000000002173. Severe.
17. Benson M, Koenig KL, Schultz CH. disaster triage: START, then SAVE-a new method of dynamic triage for victims of a catastrophic earthquake. Prehosp Disaster Med. 1996;11:117-24. https://doi.org/10.1017/s1049023x0004276x.
18. Tohira H, Jacobs I, Mountain D, Gibson N, Yeo A. Comparisons of the outcome predic- tion performance of injury severity scoring tools using the abbreviated injury scale 90 update 98 (AIS 98) and 2005 update 2008 (AIS 2008). Ann Adv Automot Med.

2011;55:255-65.

1. Lecky F, Woodford M, Edwards A, Bouamra O, Coats T. Trauma scoring systems and databases. Br J Anaesth. 2014;113:286-94. https://doi.org/10.1093/bja/aeu242.
2. Chan KH, Mann KS, Yue CP, Fan YW, Cheung M. The significance of skull fracture in acute traumatic Intracranial hematomas in adolescents: a prospective study. J Neurosurg. 1990;72:189-94. https://doi.org/10.3171/jns.1990.72.2.0189.
3. Smith-Seemiller L, Lovell MR, Smith S, Markosian N, Townsend RN. Impact of skull fracture on neuropsychological functioning following closed head injury. Brain Inj. 1997;11:191-6. https://doi.org/10.1080/026990597123638.
4. Vander Vorst M, Stuhmiller J, Ho K, Yoganandan N, Pintar F. Statistically and biome- chanically based criterion for impact-induced skull fracture. Annu Proc Assoc Adv Automot Med. 2003;47:363-81.
5. Sahoo D, Deck C, Yoganandan N, Willinger R. Anisotropic composite human skull model and skull fracture validation against temporo-parietal skull fracture. J Mech Behav Biomed Mater. 2013;28:340-53. https://doi.org/10.1016/j.jmbbm.2013.08. 010.
6. Haut RC, Wei F. Biomechanical studies on patterns of cranial bone fracture using the immature porcine model. J Biomech Eng. 2017;139. https://doi.org/10.1115/1. 4034430.
7. Sahoo D, Deck C, Yoganandan N, Willinger R. Development of skull fracture criterion based on real-world head trauma simulations using finite element head model. J Mech Behav Biomed Mater. 2016;57:24-41. https://doi.org/10.1016/j.jmbbm.2015. 11.014.
8. Kushimoto S, Kudo D, Kawazoe Y. Acute traumatic coagulopathy and trauma- induced coagulopathy: an overview. J Intensive Care. 2017;5:1-7. https://doi.org/ 10.1186/s40560-016-0196-6.
9. Maegele M, Schochl H, Menovsky T, Marechal H, Marklund N, Buki A, et al. Coagu- lopathy and haemorrhagic progression in traumatic brain injury: advances in mech- anisms, diagnosis, and management. Lancet Neurol. 2017;16:630-47. https://doi. org/10.1016/S1474-4422(17)30197-7.
10. Hulka F, Mullins RJ, Frank EH. Blunt brain injuryactivates thecoagulation process. Arch Surg. 1996;131:923-8. https://doi.org/10.1001/archsurg.1996.01430210021004.
11. Schmidt OI, Heyde CE, Ertel W, Stahel PF. Closedheadinjury-aninflammatorydisease? Brain Res Brain Res Rev. 2005;48:388-99. https://doi.org/10.1016/j.brainresrev.2004. 12.028.
12. Shibahashi K, Sugiyama K, Okura Y, Hoda H, Hamabe Y. Multicenter retrospective cohort study of “talk and die” after traumatic brain injury. World Neurosurg. 2017;107:82-6. https://doi.org/10.1016/j.wneu.2017.07.117.
13. Yanagawa Y, Sakamoto T, Saitoh D, Terai C, Okada Y, Nawashiro H, et al. Significance of shock in head-injured patients with skull fracture. Neurol Med Chir (Tokyo). 2000;40:133-9. https://doi.org/10.2176/nmc.40.133.
14. Simpson DA, Blumbergs PC, Cooter RD, Kilminster M, McLean AJ, Scott G. Pontomedullary tears and other gross brainstem injuries after vehicular accidents. J Trauma. 1989;29:1519-25. https://doi.org/10.1097/00005373-198911000-00011.
15. Weisenbach CA, Logsdon K, Salzar RS, Chancey VC, Brozoski F. Preliminary investiga- tion of skull Fracture patterns using an impactor representative of helmet back-face deformation. Mil Med. 2018;183:287-93. https://doi.org/10.1093/milmed/usx210.
16. Liu B, Ivers R, Norton R, Blows S, Lo SK. Helmets for preventing injury in motorcycle riders. Cochrane Database Syst Rev. 2004:CD004333. https://doi.org/10.1002/ 14651858.CD004333.pub2.
17. Rughani AI, Lin C-T, Ares WJ, Cushing DA, Horgan MA, Tranmer BI, et al. helmet use and reduction in skull fractures in skiers and snowboarders admitted to the hospital. J Neurosurg Pediatr. 2011;7:268-71. https://doi.org/10.3171/2010.12.PEDS10415.