Comparison of emergency department outcomes between traumatic brain injury patients on oral anticoagulants and those on acetylsalicylic acid

Abstract

AimThis study aimed to compare the emergency department (ED) course and clinical outcomes of Traumatic Brain Injury (TBI) patients using acetylsalicylic acid (ASA) or a novel oral anticoagulant (NOAC). MethodsThis is a retrospective, single-center cohort study. Consecutive adult TBI patients using ASA or a NOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) at the time of admission were included. Demographic characteristics, comorbidities, cranial computed tomography (CT) findings, and ED management/clinical outcome variables were recorded. Multivariable logistic regression analysis was performed to identify factors independently associated with positive CT finding. ResultsA total of 844 patients were included in the study, of whom 429 (50.8%) were using ASA and 415 (49.2%) were using NOAC. The NOAC group was significantly older (82.0±8.0 vs. 78.6±9.4 years, P < .001) and had a markedly higher prevalence of atrial fibrillation (76.4% vs. 14.9%, P < .001). The rate of a positive finding on cranial CT did not differ significantly between the groups (ASA 9.1%, NOAC 7.2%; P = .389). No significant differences were found between the groups in terms of length of hospital stay, intensive care unit admission, need for neurosurgical intervention, delayed intracranial hemorrhage, or in-hospital mortality (all P > .05). On multivariable analysis, NOAC use was not an independent predictor of a positive cranial CT finding (adjusted OR 0.63, 95% CI 0.33–1.21, P = .170), whereas only advanced age was identified as an independent and significant risk factor (adjusted OR 1.04/year, 95% CI 1.01–1.07, P = .010). ConclusionTBI patients using ASA or NOAC were found to be similar in terms of ED course and clinical outcomes. Our findings suggest that patient age, rather than antithrombotic drug class, may be a more determinative factor in risk stratification following head trauma.

Keywords

Introduction

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide and is responsible for nearly half of all trauma-related deaths.¹ The emergency department (ED) is the point of first healthcare contact for the majority of these patients.² Recent epidemiological data indicate that ED presentations for TBI have increased in recent years, while among patients requiring hospital admission, the most notable differences are related to age, comorbidities, and the use of medications affecting coagulation.^3,4
Although many patient-related factors can be determinative in the management of trauma patients in the ED, the use of anticoagulant and antiplatelet agents is particularly prominent among patients with TBI. Any anticoagulant use is associated with an increased risk of bleeding, and bleeding complications can be life-threatening, while the use of antiplatelet and anticoagulant therapy is common among patients today.⁵
Over the past decade, a fundamental shift has occurred in oral anticoagulation practice: warfarin use has been progressively declining among patients with atrial fibrillation, while the use of novel oral anticoagulants (NOACs) has increased markedly.^6,7 In parallel, acetylsalicylic acid (ASA) use for primary prevention has shown a declining trend, particularly among older adults, following recent changes in clinical guidelines; however, ASA use for secondary prevention indications remains prevalent.⁸ These diverging usage trends for the two drug classes have led ED physicians to encounter TBI patients on anticoagulant or antiplatelet therapy with increasing frequency in daily practice.⁹
Current guidelines place greater clinical emphasis on oral anticoagulant use than on antiplatelet therapy in TBI patients and recommend closer observation accordingly;¹⁰ however, some studies have challenged this approach, suggesting that antiplatelet therapy may be associated with a higher risk of traumatic intracranial hemorrhage compared with oral anticoagulation.¹¹ Acetylsalicylic acid acts through inhibition of platelet aggregation, whereas NOACs act through inhibition of the coagulation cascade—distinct pharmacological mechanisms.^12,13 Therefore, their effects on bleeding risk after TBI cannot be assumed to be equivalent; nevertheless, existing literature directly comparing these two specific drug classes remains limited and conflicting. Accordingly, further studies focusing on antiplatelet therapy in the context of TBI have been recommended to improve the diagnostic accuracy of current guidelines.
The present study therefore aimed to compare the ED course and clinical outcomes—including cranial CT findings, length of hospital stay, ED observation and disposition decisions, ICU admission, need for neurosurgical intervention, delayed intracranial hemorrhage, and in-hospital mortality—of TBI patients presenting to the ED while on chronic ASA or NOAC therapy at the time of admission.

Materials and Methods

Study Design, Setting This was a retrospective, single-center cohort study conducted at the ED of İzmir City Hospital, comparing the ED course and clinical outcomes of TBI patients using ASA versus NOAC at the time of injury. This study included patients presenting to the ED of İzmir City Hospital between March 1, 2024, and March 1, 2025 (a one-year study period). This study was conducted and is reported in accordance with the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) Statement for cohort studies. Study Population Consecutive adult patients who presented to the ED with TBI while on chronic antithrombotic therapy — either ASA or a NOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) — were identified from the hospital information system and included in the analysis. The hospital information system was queried using ICD-10 codes S00–S09 (injuries of the head), with particular attention to the traumatic intracranial hemorrhage subcodes S06.4 (epidural hemorrhage), S06.5 (traumatic subdural hemorrhage), S06.6 (traumatic subarachnoid hemorrhage), and S06.8 (other intracranial injuries, including traumatic intracerebral and cerebellar hemorrhage), cross-referenced with pharmacy/medication records confirming chronic ASA or NOAC use at the time of admission. Patients were classified into two groups according to the antithrombotic agent used at the time of admission: the ASA group and the NOAC group. Inclusion and Exclusion Criteria Inclusion criteria were: (1) age ≥18 years; (2) presentation to the ED with acute TBI (ICD-10 codes S00–S09) during the study period; and (3) documented chronic use of ASA or a NOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) at the time of injury, confirmed by pharmacy/medication records. Exclusion criteria were: (1) age <18 years; (2) concurrent use of warfarin or another antiplatelet/anticoagulant agent in addition to, or instead of, ASA/NOAC monotherapy; (3) missing or unconfirmed antithrombotic medication data; (4) absence of an initial non-contrast cranial CT performed in the ED; (5) incomplete medical records precluding extraction of key study variables; and (6) duplicate ED presentations by the same patient within the study period, in which case only the first presentation was retained. Of 1042 patients initially identified through the ICD-10–based hospital information system query, 198 were excluded: 9 for age <18 years, 61 for missing or unconfirmed antithrombotic medication data, 34 for absence of an initial cranial CT, 52 for concurrent use of warfarin or another antiplatelet/anticoagulant agent, 27 for incomplete medical records, and 15 as duplicate presentations of the same patient. The remaining 844 patients constituted the final analysis cohort (429 in the ASA group and 415 in the NOAC group). Data Collection and Study Variables For each patient, demographic characteristics (age, sex), comorbidities (hypertension, diabetes mellitus, coronary artery disease, atrial fibrillation, chronic kidney disease), injury characteristics (mechanism of trauma, Glasgow Coma Scale [GCS] score on admission), and findings on the initial non-contrast cranial computed tomography (CT) were recorded. Cranial CT findings were categorized as any positive finding (traumatic intracranial hemorrhage of any type) and further sub-classified as subdural hematoma, subarachnoid hemorrhage, intracerebral hemorrhage, and epidural hematoma. ED management and clinical outcome variables included length of hospital stay, ED observation (versus direct discharge from the ED), performance of a repeat cranial CT, intensive care unit (ICU) admission, need for neurosurgical intervention, delayed intracranial hemorrhage, and in-hospital mortality. Data Extraction and Verification All study variables were extracted from the hospital information system and electronic medical records by the investigators using a predefined, standardized data collection form to ensure consistency across cases. To improve reproducibility and minimize transcription error, a random subset of the extracted records was cross-checked against the source medical records by a second investigator, and any discrepancies identified were resolved by consensus review of the original chart. All patient data were de-identified prior to analysis and stored in a password-protected database accessible only to the study investigators. Ethical Approval The study protocol was reviewed and approved by the İzmir City Hospital Non-Interventional Clinical Research Ethics Committee (Decision No: 2025/190, approval date: April 18, 2025). The study was conducted in accordance with the principles of the Declaration of Helsinki. Given its retrospective design based on de-identified hospital records, the requirement for individual informed consent was waived by the ethics committee. Sample Size Considerations This study was based on a retrospective, consecutive-sampling design in which all eligible patients presenting to the ED during a predefined one-year period were included, rather than a prospectively recruited sample drawn to a prespecified target size; therefore, a formal a priori sample size (power) calculation was not performed before data collection. To characterize the precision of our primary comparison, we report the 95% confidence interval for the absolute between-group difference in cranial CT positivity rather than relying on post hoc power, which is mathematically determined by the observed P value and is discouraged as a stand-alone metric of study adequacy. The observed difference in cranial CT positivity between groups (ASA 9.1% vs. NOAC 7.2%) was 1.9 percentage points (95% CI, −1.9 to 5.6 percentage points), and the corresponding adjusted odds ratio for NOAC use was 0.63 (95% CI, 0.33–1.21). As both intervals are narrow enough to exclude a strong effect in either direction while remaining compatible with a modest difference of a few percentage points, our sample size provided reasonable — though not unlimited — precision for the primary comparison; the findings should be interpreted as indicating no strong signal of differential risk between groups, while acknowledging that a small-to-moderate difference cannot be fully excluded. This limitation is more pronounced for the rarer secondary outcomes, for which very few events were observed Statistical Analysis Continuous variables were tested for normality of distribution using the Shapiro–Wilk test and are presented as mean ± standard deviation when normally distributed or as median (interquartile range, IQR) when not. Between-group comparisons of continuous variables were performed using the Student's t-test or the Mann–Whitney U test, as appropriate. Categorical variables are presented as frequencies and percentages and were compared using the chi-square test; Fisher's exact test was substituted whenever any expected cell count was below 5. A two-sided P value < 0.05 was considered statistically significant throughout. To identify factors independently associated with a positive finding on cranial CT, a multivariable logistic regression model was constructed. Candidate predictors were selected a priori on clinical grounds rather than through stepwise or data-driven selection, to avoid model overfitting: antithrombotic agent group (NOAC vs. ASA), age, sex, GCS score of 15 versus <15 on admission, ground-level fall as the trauma mechanism, and atrial fibrillation. Atrial fibrillation was included as a covariate independent of its association with the outcome because it is the principal clinical indication for NOAC prescription and therefore a potential confounder by indication (channeling bias) for the antithrombotic-group comparison. With 69 CT-positive events in the cohort and six predictor terms, the events-per-variable ratio was approximately 11.5, within the range generally considered adequate for stable logistic regression estimation. Results are reported as crude and adjusted odds ratios (OR) with 95% confidence intervals (CI). Model discrimination was quantified using the area under the receiver operating characteristic curve (C-statistic), and calibration was assessed with the Hosmer–Lemeshow goodness-of-fit test. Because the limited number of outcome events constrains the reliability of a single random train–test split for internal validation, bootstrap-based optimism correction (1000 resamples, Harrell's method) was used as the primary approach to internal validation, yielding an optimism-corrected C-statistic. As a sensitivity analysis, the cohort was additionally partitioned once into an 80% training set and a 20% test set, and the model refitted on the training set was evaluated on the held-out test set; given the small resulting number of test-set events, this result is reported as supportive rather than primary evidence of model performance. All analyses were performed in R (version 4.4; R Foundation for Statistical Computing, Vienna, Austria) using the dplyr, ggplot2, pROC, broom, and flextable packages.

Results

A total of 844 patients were included in the analysis, of whom 429 (50.8%) were using ASA and 415 (49.2%) were using a NOAC at the time of TBI (Table 1). Patients in the NOAC group were significantly older than those in the ASA group (82.0 ± 8.0 vs. 78.6 ± 9.4 years, P < .001). The prevalence of atrial fibrillation was markedly higher in the NOAC group (76.4% vs. 14.9%, P < .001), consistent with atrial fibrillation being the principal indication for NOAC prescription in this population, whereas coronary artery disease was more common in the ASA group (32.9% vs. 25.8%, P = .029). Sex distribution, GCS = 15 on admission, ground-level fall as the trauma mechanism, and the prevalence of hypertension, diabetes mellitus, and chronic kidney disease did not differ significantly between groups (all P > .05). The overall rate of positive finding on initial cranial CT was 9.1% in the ASA group and 7.2% in the NOAC group, a difference that was not statistically significant (P = .389) (Table 2). Similarly, no significant between-group differences were observed for any individual hemorrhage subtype, including subdural hematoma, subarachnoid hemorrhage, intracerebral hemorrhage, or epidural hematoma (all P > .05). Median length of hospital stay was identical between groups (3.0 [IQR 2.0–5.0] days for both). No statistically significant differences between-group differences were observed for ED observation, repeat cranial CT, ICU admission, neurosurgical intervention, delayed intracranial hemorrhage, or in-hospital mortality (all P > .05) (Table 3, Figure 1). Notably, the absolute number of events was very low for several of these outcomes (in-hospital mortality, n = 4 in total; delayed intracranial hemorrhage, n = 2; neurosurgical intervention, n = 9), so these comparisons are underpowered and should be interpreted with caution; for this reason, no multivariable modeling was attempted for these sparse outcomes, and only descriptive statistics with exact tests are reported. Presented as median (interquartile range); compared using the Mann–Whitney U test. All other variables were compared using the chi-square test, except neurosurgical intervention, delayed intracranial hemorrhage, and in-hospital mortality, for which Fisher's exact test was used owing to low expected cell counts. On univariable (crude) analysis, none of the candidate predictors, including antithrombotic agent group, reached statistical significance for association with a positive cranial CT finding (Table 4). After multivariable adjustment for age, sex, GCS = 15, atrial fibrillation, and trauma mechanism, NOAC use was not independently associated with a positive cranial CT finding compared with ASA use (adjusted OR 0.63, 95% CI 0.33–1.21, P = .170). Age was the only variable independently and significantly associated with a positive CT finding, with each additional year of age increasing the adjusted odds by 4% (adjusted OR 1.04, 95% CI 1.01–1.07, P = .010). Sex, GCS on admission, atrial fibrillation, and trauma mechanism were not independently associated with the outcome (Table 4, Figure 2). The multivariable model showed modest discriminative ability, with an apparent C-statistic (AUC) of 0.607 (95% CI 0.542–0.672). The Hosmer–Lemeshow test did not indicate significant lack of fit (χ² = 13.13, df = 8, P = .107), suggesting adequate calibration. Bootstrap-based internal validation (1000 resamples) yielded a mean optimism of 0.047, and the corresponding optimism-corrected C-statistic was 0.560, indicating that the apparent discrimination was modestly inflated by in-sample overfitting and that the model's true discriminative performance is limited. In the sensitivity analysis using a single 80/20 train–test split, the model achieved a comparable AUC of 0.571 in the held-out test set; however, this test set contained only 12 CT-positive events, limiting the precision of this estimate and supporting the bootstrap-corrected result as the more reliable measure of internal validity.

Discussion

In this study, a large cohort of patients presenting ED with TBI while on chronic ASA or NOAC therapy was evaluated, and no significant difference was found between the two antithrombotic agent classes with respect to ED course and clinical outcomes.
Age has a positive correlation with both bleeding and thromboembolism.¹⁴ When the New Orleans Criteria (NOC), the National Institute for Health and Care Excellence (NICE) guidelines, the Canadian CT Head Rule (CCHR), the Eastern Association for the Surgery of Trauma (EAST) practice management guidelines, and the Scandinavian Neurotrauma Committee (SNC) guidelines—all recommended for risk stratification and imaging decisions in TBI patients—are examined, each incorporates a baseline age threshold along with particular emphasis on anticoagulant therapy In our study, although the NOAC group was significantly older, consistent with the higher prevalence of atrial fibrillation, and theoretically had a more potent anticoagulant effect, it did not differ from the ASA group in terms of the rate of a positive finding on cranial CT (7.2% vs. 9.1%), need for ICU admission, need for neurosurgical intervention, or in-hospital mortality.
Several possible explanations exist for this finding. First, most studies reporting a higher risk with anticoagulant therapy compared with antiplatelet therapy have evaluated vitamin K antagonists and direct oral anticoagulants as a single combined group, whereas our study included only NOAC/DOAC agents (apixaban, rivaroxaban, dabigatran, and edoxaban). DOACs are known to have a more predictable pharmacokinetic profile and a substantially lower risk of intracranial hemorrhage compared with warfarin; therefore, comparisons with cohorts that include warfarin may not be directly comparable to our findings. However, a study conducted in a similar population of patients with fall-related TBI similarly argued that antithrombotic therapy has a negligible effect on patient clinical management, with comparable outcomes observed between patients on antiplatelet therapy and those on anticoagulant therapy.¹⁵
Among the risk stratification scoring systems used in the evaluation of head trauma, elderly patients are consistently identified as the common high-risk group, as they account for approximately 40% of TBI-related deaths.¹⁶ The physiological changes and additional comorbidities associated with age determine frailty in elderly patients. In our study's multivariable analysis, NOAC use did not emerge as an independent predictor of a positive cranial CT finding compared with ASA use, whereas advanced age alone was identified as the only variable independently and significantly associated with increased bleeding risk. These findings challenge the conventional approach of guidelines that attribute greater clinical importance to oral anticoagulant use than to antiplatelet therapy, and suggest that patient age, rather than drug class, may be a more determinative factor in risk stratification following head trauma.
Taken together, the findings of this study suggest that patient-specific factors—particularly age—rather than the distinction between drug classes (ASA or NOAC), may be more determinative in assessing bleeding risk after TBI. The fact that clinical outcomes were similar between the ASA and NOAC groups, despite NOAC patients being significantly older and theoretically having a more potent anticoagulant effect, suggests that "advanced age" and "anticoagulant use," which are listed as separate criteria in most current risk scoring systems, may in fact be interrelated rather than independent risk determinants. This finding is consistent with the growing body of literature questioning the guideline-based approach of categorizing anticoagulant therapy as conferring a separate and higher risk category than antiplatelet therapy in TBI patients, and offers a clinically applicable insight suggesting that patients on single-agent ASA and single-agent NOAC therapy may be evaluated using a similar observation and imaging strategy in the ED.

Limitations

The results of this study should be interpreted considering several important limitations. 1) Our study included only patients who were able to reach the ED; patients who died in the prehospital setting due to bleeding complications could not be captured in the sample. 2) Our study design did not include a control group of head trauma patients not using any antithrombotic agent. Therefore, our findings only allow a comparison between ASA and NOAC, and no inference can be drawn regarding whether either agent increases bleeding risk relative to patients not receiving antithrombotic therapy. 3) The NOAC group combined patients using apixaban, rivaroxaban, dabigatran, and edoxaban into a single category. Because the pharmacokinetic properties, renal elimination pathways, and bleeding profiles of these agents differ from one another, subgroup-level heterogeneity may be present; the sample size did not permit subgroup analysis. 4) A formal a priori sample size (power) calculation was not performed, as this was a retrospective study based on consecutive, routinely collected data over a fixed period rather than a prospectively targeted sample size; consequently, the study may have been underpowered to detect modest between-group differences, particularly for outcomes with few events.

Conclusion

In this study, no significant difference was found between TBI patients using ASA or NOAC therapy with respect to cranial CT findings, length of hospital stay, ICU admission, need for neurosurgical intervention, delayed intracranial hemorrhage, or in-hospital mortality. On multivariable analysis, antithrombotic agent class did not emerge as an independent predictor of a positive cranial CT finding, whereas advanced age alone was identified as an independent and significant risk factor.

Declarations

Abbreviations

References

1. Fu TS, Jing R, McFaull SR, Cusimano MD. Health and economic burden of traumatic brain injury in the emergency department. Can J Neurol Sci. 2016;43(2):238-247. doi:10.1017/cjn.2015.320
   Article · PubMed · Google Scholar
2. Daugherty J, Thomas K, Waltzman D, Chang D, Ballesteros MF. Comparison of US traumatic brain injury–related emergency department visits from 2 data sources: NEISS-AIP and HCUP-NEDS. Inj Prev. 2025;ip-2025-045769. doi:10.1136/ip-2025-045769
   Article · PubMed · Google Scholar
3. Hsia RY, Markowitz AJ, Lin F, Guo J, Madhok DY, Manley GT. Ten-year trends in traumatic brain injury: a retrospective cohort study of California emergency department and hospital revisits and readmissions. BMJ Open. 2018;8(9). doi:10.1136/bmjopen-2018-022297
   Article · PubMed · Google Scholar
4. Niklasson E, Svensson E, André L, Areskoug C, Forberg JL, Vedin T. Higher risk of traumatic intracranial hemorrhage with antiplatelet therapy compared to oral anticoagulation: a single-center experience. Eur J Trauma Emerg Surg. 2024;50(4):1237-1248. doi:10.1007/s00068-024-02493-z
   Article · PubMed · Google Scholar
5. Vehviläinen J, Virta JJ, Skrifvars MB, et al. Effect of antiplatelet and anticoagulant medication use on injury severity and mortality in patients with traumatic brain injury treated in the intensive care unit. Acta Neurochir (Wien). 2023;165(12):4003-4012. doi:10.1007/s00701-023-05850-w
   Article · PubMed · Google Scholar
6. Navar AM, Kolkailah AA, Overton R, et al. Trends in oral anticoagulant use among 436 864 patients with atrial fibrillation in community practice, 2011 to 2020. J Am Heart Assoc. 2022;11(22). doi:10.1161/jaha.122.026723
   Article · PubMed · Google Scholar
7. Karcioglu O, Zengin S, Ozkaya B, et al. Direct (new) oral anticoagulants (DOACs): drawbacks, bleeding, and reversal. Cardiovasc Hematol Agents Med Chem. 2022;20(2):103-113. doi:10.2174/1871525719666210914110750
   Article · PubMed · Google Scholar
8. Wilson LM, Anderson TS. Trends in preventive aspirin use by atherosclerotic cardiovascular risk. JAMA. 2025;333(10):904-907. doi:10.1001/jama.2024.27333
   Article · PubMed · Google Scholar
9. Bettschen D, Tsichlaki D, Chatzimichail E, et al. Epidemiology of maxillofacial trauma in elderly patients receiving oral anticoagulant or antithrombotic medication: a Swiss retrospective study. BMC Emerg Med. 2024;24:121. doi:10.1186/s12873-024-01039-1
   Article · PubMed · Google Scholar
10. Mower WR, Hoffman JR, Herbert M, Wolfson AB, Pollack CV, Zucker MI; NEXUS II Investigators. Developing a decision instrument to guide computed tomographic imaging of blunt head injury patients. J Trauma. 2005;59(4):954-959. doi:10.1097/01.ta.0000187813.79047.42
    Article · PubMed · Google Scholar
11. Vedin T, Lundager Forberg J, Anefjäll E, Lehtinen R, Faisal M, Edelhamre M. Antiplatelet therapy contributes to a higher risk of traumatic intracranial hemorrhage compared to anticoagulation therapy in ground-level falls: a single-center retrospective study. Eur J Trauma Emerg Surg. 2022;48(6):4909-4917. doi:10.1007/s00068-022-02016-8
    Article · PubMed · Google Scholar
12. Mogielnicki A. Pharmacological management of thrombosis: current state and future strategies. Int J Mol Sci. 2026;27(7):3151. doi:10.3390/ijms27073151
    Article · PubMed · Google Scholar
13. Karcioglu O, Yilmaz S, Afacan G, et al. An update of the efficacy and comparative characteristics of direct (new) oral anticoagulants (DOACs). Cardiovasc Hematol Agents Med Chem. 2022;20(2):114-124. doi:10.2174/1871525719666210914112402
    Article · PubMed · Google Scholar
14. Patti G, Lucerna M, Pecen L, et al. Thromboembolic risk, bleeding outcomes, and effect of different antithrombotic strategies in very elderly patients with atrial fibrillation: a subanalysis from the PREFER in AF (PREvention oF Thromboembolic Events–European Registry in Atrial Fibrillation). J Am Heart Assoc. 2017;6(7). doi:10.1161/jaha.117.005657
    Article · PubMed · Google Scholar
15. Fakhry SM, Morse JL, Garland JM, et al. Antiplatelet and anticoagulant agents have minimal impact on traumatic brain injury incidence, surgery, and mortality in geriatric ground-level falls: a multi-institutional analysis of 33 710 patients. J Trauma Acute Care Surg. 2021;90(2):215-223. doi:10.1097/ta.0000000000002985
    Article · PubMed · Google Scholar
16. Okrah AK, Tharrington S, Shin I, Wagoner A, Woodsmall KS, Jehu DA. Risk factors for fall-related mild traumatic brain injuries among older adults: a systematic review highlighting research gaps. Int J Environ Res Public Health. 2025;22(2):255. doi:10.3390/ijerph22020255
    Article · PubMed · Google Scholar

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions