The discrimination between falls and blows is an important task in forensic anthropology and pathology. This research aimed to test a discrimination method between falls and blows. This method was created from the quotation of 549 types of fractures for 57 bones and 12 anatomical regions. Different models were tested according to the sensibility of random forest parameters and their effects on model accuracies. The best model was based on binary coding of 12 anatomical regions or 28 bones with or without baseline (age and sex). We tested this new method in the distinction between falls and blows on post-mortem computerized tomography scans (PMCT). The sample was composed of 47 subjects with 36 falls and 11 blows, whose aetiologia was based on forensic reports.

Of the 47 bodies, 35 were complete, and 12 presented missing bones; 39 were estimated to be falls and 8 to be blows.

Of the 12 individuals with missing bones, 11 had a reasonable estimation of the etiology of fractures, i.e., 91.7%.

Methods showed excellent etiology estimation for fall cases (97.2%) but misclassified 36.4 % of blow cases. Our method misclassified 5 subjects (10.6%), more precisely, 4 blows and 1 fall. Overall, the reliability of the estimation of the etiology is substantial, with a Cohen’s k-values of 0.67. The method could be used in distinguishing between blows and falls and is also suitable for fragmented or missing bones. To ensure an easy and fast use of this method, we have developed a freely available online automated tool.

Résumé: This research aims to test a method of discrimination between falls and blows.

This method was created from the rating of 549 types of fractures, 57 bones and 12 anatomical regions. Different models were tested depending on the sensitivity of random forest parameters and their effects on model accuracy. The best model is based on binary coding of 12 anatomical regions or 28 bones with or without reference data (age and sex). We tested this new method of distinguishing between falls and blows on post-mortem CT scans (PMCT). The sample consists of 47 subjects, 36 falls and 11 blows, the etiology of which was based on medico-legal reports.

Of the 47 bodies, 35 were complete, and 12 had missing bones; 39 were judged to be falls and 8 to be hits.

Of the 12 individuals with missing bones, 11 had a good estimate, or 91.7%.

Our method misclassified 5 subjects (10.6%) more precisely, 4 hits and 1 fall.

The method showed excellent etiology estimation for fall cases (97.2%) but misclassified 36.4% of blow cases.

The reliability of the etiology estimate is substantial with a Cohen’s kappa of 0.67. The method could be used to distinguish blows from falls and would also be suitable in the case of fragmented or missing bones. To ensure easy and quick use of this method, we have developed an automated tool, free and accessible online.

• Forensic science

• Blunt force trauma

• Falls

• Blows

• Skeletal fractures

• CT scan

• Random forests

Henriques M, Bonhomme V, Piercecchi-Marti MD, Delteil C, Carballeira-Alvarez A, et al. (2024) Test of a New Method in the Distinction between Falls and Blows on A Post-Mortem CT-Scans Sample. J Fract Sprains 3(1): 1009.

One of the tasks of the forensic anthropologist is to analyze and interpret skeletal trauma to provide information about the circumstances surrounding the death of [1,2]. Blunt force trauma (BFT) is one of the most common injuries and is difficult to interpret due to its variability [3-5]. Etiology estimation of fractures is important in analyzing skeletal remains [1,2,6]. Some studies showed that a distinction between falls and blows could be made with a multi-criteria approach, but there is no reliable method to differentiate both etiologies [6-9].

In a precedent study, we proposed a “revised” method. We highlighted that falls and blows could be predicted with a probability between 77% and 83% by four models based on the quotation of fracture on 28 bones, 12 anatomical regions, and baseline (age/sex) [10]. The revised method was established on living CT scans. It scored the presence/absence of fractures on 28 bones and 12 anatomical regions, with sex and age entered. This study evaluated the method’s validity on a post-mortem CT scan sample. 

Sample

We carried out a retrospective descriptive study from November 2009 to November 2020. We included 47 anonymized patients aged between 20 and 49 who had undergone a forensic autopsy for a trauma clearly identified during the first survey data (fall regardless of the height or blow with or without an object). We had undergone post-mortem computerized tomography scans (PMCT). A forensic pathologist made the case selection. Thus, the analysis of the scans was carried out by an independent anthropologist and blinded to the lesion mode. The data collected on the autopsy report were made a posteriori and transmitted by the medical examiner.

PMCT records were extracted from the forensic report and the digital archiving systems of the Forensic Department of the Assistance Publique-Hôpitaux de Marseille (AP-HM, France). The CT devices used were a Siemens Somatom Definition (Siemens Healthineers headquarters: Erlangen, Germany) and a General Electrics Optima CT 660 (GE Healthcare headquarters: Chicago, IL, USA). Whenever possible, the body was in a supine position, arms resting along the body. Two acquisitions were made with 1mm sections: cervicocranial and from the cervical region to the feet. The scanner parameters were 120 Kv, 1000 mAs for the cervicocranial acquisition and 300 mAs for the whole-body acquisition.

Imaging analysis

We collected the following data on the autopsy reports: gender and age (later referred to as ‘baseline’), and whether the death was due to falls or blows. This last information was used only at the end of our study to perform blind tests. Some bones were absent in the case of charring and/or putrefaction, so the fracture was encoded as non-available.

The method was built on the quotation of 549 types of fractures for 57 bones (skull and trunk) using two classifications, AO/OTA and Galloway, Wedel, when possible [5,11-13]. For the fractures without classification, the quotation was made by localization and according to the absence and the presence of a fracture. We excluded 534 types and 29 bones to cope with absent and rare events (less than 5%). Various models were tested according to the sensibility of random forest parameters and their effects on model accuracies. The best ones were based on the binary coding of 12 anatomical regions or 28 bones with or without baseline (age and sex). The quotation of the anatomical regions and the bones were complementary; one is more general, and the other more refined.

We registered fractures (present or absent) on 28 bones (frontal, parietal, occipital, temporal, sphenoid, ethmoid, nasal, maxilla, zygomatic, mandible, scapula, 3rd to 10th rib, 12th thoracic vertebrae, 1st to 5th lumbar vertebrae, sacrum, coxal bone, and femur) and 12 anatomical regions (cranium, basicranium, cranial vault, face, mandible, ribs, scapula, thoracic vertebrae, lumbar vertebrae, coxal bone, sacrum, and femur).

The observations were performed using multiplanar reconstructions (MPRs), maximum intensity projection (MIP), and volume rendering (VR) reconstructions on Horos version 3.3.5®.

Statistical analyses

Observer agreement tests

We randomly selected 30 individuals from the sample to assess the repeatability and reproducibility. Repeatability was tested by the same observer repeating the protocol twice several weeks apart; for reproducibility, a second observer applied the protocol once (both forensic anthropologists). The intra- and inter-error observers were evaluated using the same statistical test (Cohen’s kappa).

Following Landis and Koch et al. [14], a kappa value of < 0.2 was considered poor agreement, 0.21-0.4 fair, 0.41-0.6 moderate, 0.61-0.8 substantial, and more than 0.8 near-complete agreement. 

Prediction method

The analyses were carried out in the R environment version 4.1.3 (R Foundation for Statistical Computing, 2022, Vienna, Austria) with RandomForest and KappaGUI R packages [15]. Random forests are an ensemble learning method that can be used for classification by constructing a multitude of decision trees at training time, and which avoids the bias of overfitting to the training set that is sometimes encountered with classical decision tree approaches.

We ran the best four models of prediction tested and presented in Henriques et al., using the random forest algorithm [9]. The four models were based on binary scoring (0: absence, 1: presence) of fracture on 28 bones with or without baseline (age and sex) and 12 anatomical regions with or without baseline. 

Fisher’s exact tests were used to identify the association between two qualitative variables, especially the correlation between the localization of fracture of our models and the sex or age group. Thanks to implementing a simple tool (an online app), it is possible to directly obtain the probabilities, and therefore the inferred etiology, according to the four models. The online tool is available at: http://www.fracture.cloud http://fracture.cloud

Cohen’s kappa was applied to assess the correct evaluations, that is the agreement of the answer of our discrimination method and the real aetiology based on the forensic report (conclusion of the autopsy which considers the findings of the investigation and the lesions found at the autopsy).

Observer agreement tests

The inter- and intra-observer errors were evaluated using Cohen’s kappa (Annex 1) [16,17].

Annex 1: Inter- and intra-observer errors of the assessment of the presence of fractures on studied bones and anatomical regions using Cohen's kappa (N=30)

The inter-observer reliability of the scoring was good or even excellent for all bones and anatomical regions, with values higher than 0.61, except for the 8th rib for which the value was 0.44, indicating moderate reliability.

The intra-observer reliability of the scoring was excellent for all quotations, with values higher than 0.84, except for fracture on the 5th lumbar vertebrae, where the value was 0.76, indicating good reliability.

Prediction method

The new method in the distinction between falls and blows is based on the anatomical regions and bones presented in Table 1.

Table 1: Presence of fractures by anatomical regions and bones, in both aetiologia (P value associated with the Fisher’s exact test; in bold: significant values) (V: vertebrae)

We observed that on all anatomical regions and bones fractured considered for the development of the new method, six elements did not present a significant difference between falls and blows (fractures on the basicranium, ethmoid, parietal, sphenoid, temporal and zygomatic bone).

Many anatomical regions and bones are more frequent in falls, except for fractures of the cranium, face, mandible, ethmoid, maxilla, nasal and zygomatic.

The prediction method tested on the postmortem CT-scans sample

Of the 47 bodies, 35 were complete, and 12 presented missing bones (NA in Table 2); 39 were estimated to be falls and 8 to be blows.

Table 2 Scoring of fractures on the forensic sample (IND: individual, RCT: real context, ECT: estimated context, FT: frontal, TP: temporal, PT: parietal, OC: occipital, MX: maxillary, ET: ethmoid, ZY: zygomatic, SP: Sphenoid, N: nasal, MD: mandible, SC: scapula, R3: Rib3, R4: Rib4, R5:rib5, R6: rib6, R7: rib7, R8: rib8, R9: rib9, R10: rib10, T12: thoracic vertebrae 12, L1:lumbar vertebrae 1, L2:lumbar vertebrae 2, L3:lumbar vertebrae 3, L4:lumbar vertebrae 4, L5:lumbar vertebrae 5, SA: sacrum, CO: coxal, FE: femur, BL: blow, FA: fall, F: female, M: male, 0: absence of fracture, 1: presence of fracture, NA: not available, TR: true, FL: false)

According to the forensic reports, the sample comprised 36 falls and 11 blows.

Our method misclassified 5 subjects and, more precisely: 4 blow cases and 1 fall. Of the 5 subjects, 1 has missing data (20%). The false predictions concerned 10.6 % of the sample: 4 males aged 33, 42, 43, and 49 years; and one female aged 49 [Table 3]

Table 3: Scoring of fractures on the misclassified cases by the method (IND: individual, RCT: real context, ECT: estimated context, FT: frontal, TP: temporal, PT: parietal, OC: occipital, MX: maxillary, ET: ethmoid, ZY: zygomatic, SP: Sphenoid, N: nasal, MD: mandible, SC: scapula, R3: Rib3, R4: Rib4, R5:rib5, R6: rib6, R7: rib7, R8: rib8, R9: rib9, R10: rib10, T12: thoracic vertebrae 12, L1:lumbar vertebrae 1, L2:lumbar vertebrae 2, L3:lumbar vertebrae 3, L4:lumbar vertebrae 4, L5:lumbar vertebrae 5, SA: sacrum, CO: coxal, FE: femur, BL: blow, FA: fall, F: female, M: male, 0: absence of fracture, 1: presence of fracture, NA: not available).

The 4 misclassified blow cases are the individuals 17, 21, 26, and 31, briefly discussed as follows:

- Individual 17 presented fractures on the mandible, scapula, and ribs (3rd to 9th).

- Individual 21 had temporal, parietal, maxilla, sphenoid, mandible, and 1st and 2nd lumbar vertebrae fractures.

- Individual 26 presented fractures on the maxilla, ethmoid, nasal, mandible, ribs (4th, 6th, 7th, 8th), 1st lumbar vertebrae, and coxal bone.

- Individual 31 had maxilla, nasal, and rib fractures (4th to 6th).

The fall case misclassified (individual 35) presents fractures on the frontal, temporal, parietal, maxilla, ethmoid, zygomatic, sphenoid, and the 4th ribs.

The reliability between the estimated etiology and those written on forensic reports was substantial, with a Cohen’s k-value of 0.67.

When confronted with blunt force trauma, forensic experts are often asked to determine if the trauma is related to a fall or a blow. This discrimination remains a challenge, mainly because of unreliable methods. In this study, we tested the application of four prediction models, allowing us to give the probability of belonging to one etiology or another (falls or blows).

An excellent level of reliability in the intra- and inter-observer tests was found for all criteria except fracture on the 8th rib. This error may be due to a ranking error.

Most anatomical regions and bones with fractures were more frequent in falls, except for fractures of the cranium, face, mandible, ethmoid, maxilla, nasal and zygomatic. Violence is the most frequent cause of craniofacial fractures, and our results are consistent with this [7,9,10,18-33]. Concerning the localization of fractures in falls, our results are consistent with the literature except for fractures on the cranial vault [5,34-49]. According to the literature, fractures on the cranial vault could not result from falls except under certain conditions (repeated falls, falls from a height, falls with an impact against an edge or a corner) [6-8,50- 54].

Cohen’s k-values of this study show that the presented method is relevant in distinguishing between falls and blows even if we must be cautious because of the small sample size. Models applied to an independent forensic sample showed excellent etiology estimation for fall cases (97.2%) but misclassified 36.4% of blow cases.

A lack of classification in the category of cases of blows can be explained by the fact that there are fewer cases of blows in the samples used for the method and its test. This lack impacted the prediction results in this type of etiology, but we believe that, nevertheless, the results are representative of a trend that requires stronger validation by a larger sample. Moreover, we support the idea that despite the constraints of sample size and ratio between falls and blows cases, the method’s substantial agreement with forensic reports, indicated by a Cohen’s kappa value of 0.67, demonstrates its reliability and supports its utility as an adjunct to traditional autopsy findings. The use of postmortem CT scans aligns with contemporary forensic practices, offering a non-invasive approach that can be seamlessly integrated into existing workflows.

A second explanation can be provided by the unbalanced sample. For the misclassified blows, the age range and most fractured bones tend towards a fall rather than a blow (scapula, ribs, temporal, parietal, sphenoid, lumbar vertebrae, and coxal bone) [5,9,34,35,38-41,55-60]. The wrong fall estimation for the individual 35 can be explained by the fact that many bones of the face were fractured and that males aged between 30 and 39 were more involved in blows than in falls [9,61-64].

Clearly, the question of fall height will need to be addressed in future studies. In this work, cases were retrieved retrospectively, and fall heights were not systematically mentioned in the medical reports associated with the imaging examinations. A protocol modification has been requested to enable us to collect this crucial information prospectively. We will be able to refine our results and, likely, make them even more reliable with this information.

In some forensic cases, there was no guarantee of full information. Some fractures may not be recorded, or some bones may just be missing, and difficult to assess whether they were broken or not when the person passed away. These elements were considered, quoted as not available (NA) and tested in this study. Individual 26 has missing data, but only for two bones (parietal and occipital). As fractures on these bones occur more frequently in case of falls and the estimated context was fall, we did not believe there was an impact on this estimation of the etiology of fractures. Within the 12 individuals with missing data, only 1 was misclassified, i.e., 8.3%. This outcome indicates that the method is also suitable for fragmented or missing bones.

This study tested a revised method to estimate the etiology of fractures from their presence on 28 bones and 12 anatomical regions. An app using the four statistically validated models makes it possible to discriminate between falls and blows easily and automatically, with probabilities for each model (www. fracture.cloud).

The lack of information did not systematically prevent the estimation of the etiology of fractures. The reliability of this method was good, with good predictions (89.4% of the overall sample). The method correctly predicted 97.2% of fall cases and 63.6 % of blow cases.

Although we know the limitations of the present study, namely the small sample, which precludes clear-cut inferences, this study provided promising results. We look forward to increasing the sample, particularly the one concerning blows. We plan in the near future to identify and test fracture patterns and morphologies between detailed falls (different heights) and blows (object or not).

Mélanie Henriques carried out the study and performed the statistical analysis and the original draft of the manuscript.

Vincent Bonhomme has ported the app to Shiny and revised the manuscript.

Marie-Dominique Piercecchi-Marti allowed access to the forensic sample and revised the manuscript.

Clémence Delteil helped access the forensic sample and revised the manuscript.

Ana Carballeira-Alvarez helped access the forensic sample.

Pascal Adalian participated in the study design and coordination and revised the manuscript.

Eugénia Cunha participed in the studydesign and coordination and revised the manuscript.

All authors contributed to the final text and approved it

The authors would like to thank the anonymous reviewers for their comments and suggestions, which helped improve the manuscript’s quality.

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Ethics Committee of Faculdade de Medicina da Universidade de Coimbra, Portugal (protocol code: 026-CE-2019 and date of approval: 25/03/2019).

Informed consent: For this study, it was not necessary to ask for an informed consent since our study is retrospective and based on anonymized tomographic images that were made for a clinical diagnostic purpose, and therefore in the patient’s best interest.

1. Kranioti E. Forensic investigation of cranial injuries due to blunt force trauma. Research and Reports in Forensic Medical Science. 2015; 5: 25-37.
2. Passalacqua NV, Fenton TW. Developments in Skeletal Trauma: Blunt-Force Trauma. In: A Companion to Forensic Anthropology. John Wiley & Sons, Ltd; 2012. 400-11.
3. Ambade VN, Godbole HV. Comparison of wound patterns in homicide by sharp and blunt force. Forensic Sci Int. 2006; 156: 166-170.
4. Sterzik V, Duckwitz D, Bohnert M. Accident or crime? About the meaning of face injuries inflicted by blunt force. Forensic Sci Res. 2016; 1: 14-21.
5. Wedel VL, Galloway A. Broken Bones: Anthropological Analysis of Blunt Force Trauma. 2nd edition. Springfield, Illinois: Charles C Thomas Pub Ltd; 2014. 504.
6. Kremer C, Sauvageau A. Discrimination of Falls and Blows in Blunt Head Trauma: Assessment of Predictability Through Combined Criteria. J Forensic Sci. 2009; 54: 923-926.
7. Guyomarc’h P, Campagna-Vaillancourt M, Kremer C, Sauvageau A. Discrimination of Falls and Blows in Blunt Head Trauma: A Multi- Criteria Approach. J Forensic Sci. 2010; 55: 423-427.
8. Kremer C, Racette S, Dionne CA, Sauvageau A. Discrimination of Falls and Blows in Blunt Head Trauma: Systematic Study of the Hat Brim Line Rule in Relation to Skull Fractures. J Forensic Sci. 2008; 53: 716–719.
9. Henriques M, Saliba-Serre B, Martrille, Blum A, Donato P, Campos N, et al. The Discrimination between falls and blows from the localization and the number of fractures on CT scans of the skull and the trunk. Forensic Sci Res. 2023; 8: 30-40.
10. Henriques M, Bonhomme  V, Cunha E,  Adalian P. Blows  or Falls? Distinction by Random Forest Classification. Biology. 2023; 12: 206.
11. Müller Maurice Edmond, Nazarian Serge, Koch Peter, Müller ME. AO classification of fractures. 1, Long bones. M.E. Müller, S. Nazarian, P. Koch; with the collab. by Urs Heim. Berlin Paris [etc: Springer-Verlag; 1987.
12. Marsh JL, Slongo TF, Agel J, Broderick JS, Creevey W, DeCoster TA, et al. Fracture and dislocation classification compendium - 2007: Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007; 21 : S1-133.
13. Meinberg EG, Agel J, Roberts CS, Karam MD, Kellam JF. Fracture and Dislocation Classification Compendium—2018. J Orthop Trauma. 2018; 32: S1-S170.
14. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977; 33: 159-174.
15. R: a language and environment for statistical computing . 2022.
16. Cohen J. A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement. 1960; 20: 37-46.
17. Cohen J. Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. Psychol Bull. 1968; 70: 213- 220.
18. Kahramansoy N, Erkol H, Kurt F, Gurbuz N, Bozgeyik M, Kiyan A. Analysis of trauma patients in a rural hospital in Turkey. TJTES. 2011; 17: 231-237.
19. Hussain K, Wijetunge DB, Grubnic S, Jackson IT. A comprehensive analysis of craniofacial trauma. J Trauma. 1994; 36: 34-47.
20. Brook IM, Wood N. Aetiology and incidence of facial fractures in adults. Int J Oral Surgery. 1983; 12: 293-298.
21. Ribeiro MFP, Marcenes W, Croucher R, Sheiham A. The prevalence and causes of maxillofacial fractures in patients attending Accident and Emergency Departments in Recife-Brazil. Int Denl Journal. 2004; 54: 47-51.
22. Beaumont E, Lownie JF, Cleaton-Jones PE, Newtyon NPD. An analysis of fractures of the facial skeleton in three populations in the Johannesburg urban area. J Den Assoc S Afr. 1985; 40: 633-638.
23. Starkhammar H, Olofsson J. Facial fractures: A review of 922 cases with special reference to incidence and aetiology. Clin OtolaryngolAllied Sci. 1982; 7: 405-409.
24. Erdmann D, Follmar KE, DeBruijn M, Bruno AD, Jung SH, Edelman D, et al. A Retrospective Analysis of Facial Fracture Etiologies. Ann Plast Surg. 2008; 60: 398-403.
25. Scherer M, Sullivan WG, Smith DJ, Phillips LG, Robson MC. A Analysis of 1,423 Facial Fractures in 788 Patients at an Urban Trauma Center. J Trauma; 1989; 29: 388-390.
26. Redfern RC. Injury and Trauma in Bioarchaeology: interpreting violence in past lives. Cambridge University Press, Cambridge. 2017; 329.
27. Walker PL. A Bioarchaeological Perspective on the History of Violence. Annu Rev Anthropol. 2001; 30: 573-596.
28. Sojat AJ, Meisami T, Sandor GK, Clokie CM. The epidemiology of mandibular fractures treated at the Toronto general hospital: A review of 246 cases. J Can Dent Assoc. 2001 ;67: 640-44.
29. Lee KH. Interpersonal violence and facial fractures. J Oral Maxillofac Surg. 2009; 67: 1878-1883.
30. Zaleckas L, Drobnys P, Rimkuvien? J. Incidence and etiology of mandibular fractures treated in Vilnius University Hospital Žalgiris clinic, Lithuania: a review of 1 508 cases. Acta medica Lituanica. 2013; 20: 53-60.
31. Ellis E, Moos KF, el-Attar A. Ten years of mandibular fractures: an analysis of 2,137 cases. Oral Surg Oral Med Oral Pathol. 1985; 59: 120-129.
32. Paza A, Abuabara A, Passeri L. Analysis of 115 mandibular angle fractures. J Oral Maxillofac Surg. 2008; 66: 73-76.
33. Silvennoinen U, Iizuka T, Lindqvist C, Oikarinen K. Different patterns of condylar fractures: an analysis of 382 patients in a 3-year period. J Oral Maxillofac Surg. 1992; 50: 1032-1037.
34. Burke MP. Forensic pathology of fractures and mechanisms of injury : postmortem CT scanning. Boca Raton (Fla.): CRC; 2012.
35. Court-Brown CM. Chapter 3: The Epidemiology of Fractures and Dislocations. In: Court-Brown CM, Heckman JD, McQueen MM, Ricci WM, Tornetta III P, McKee MD, editors. Rockwood and Green’s: Fractures in Adults. Wolters Kluwer; 2015; 59-108.
36. Rowbotham SK, Blau S, Hislop-Jambrich J, Francis V. An Anthropological Examination of the Types of Skeletal Fractures Resulting from Fatal High (?3 m) Free Falls. J Forensic Sci. 2018; 14.
37. Talbot BS, Gange CP, Chaturvedi A, Klionsky N, Hobbs SK, Chaturvedi A. Traumatic Rib Injury: Patterns, Imaging Pitfalls, Complications, and Treatment. RadioGraphics. 2017; 37: 628-651.
38. Cooper C, Dunham CM, Rodriguez A. Falls and major injuries are risk factors for thoracolumbar fractures: cognitive impairment and multiple injuries impede the detection of back pain and tenderness. J Trauma. 1995; 38: 692–696.
39. Hsu JM, Joseph T, Ellis AM. Thoracolumbar fracture in blunt trauma patients: guidelines for diagnosis and imaging. Injury. 2003; 34: 426- 433.
40. Meldon SW, Moettus LN. Thoracolumbar spine fractures: clinical presentation and the effect of altered sensorium and major injury. J Trauma. 1995; 39: 1110-1114.
41. Samuels LE, Kerstein MD. Routine Radiologic evaluation of the thoracolumbar spine in blunt trauma patients: a reappraisal. J Trauma. 1993; 34: 85-89.
42. Bydon M, De la Garza-Ramos R, Macki M, Desai A, Gokaslan AK, Bydon A. Incidence of Sacral Fractures and In-Hospital Postoperative Complications in the United States: An Analysis of 2002–2011 Data. Spine. 2014; 39: E1103-1109.
43. Denis F, Davis S, Comfort T. Sacral fractures: an important problem. Retrospective analysis of 236 cases. Clin Orthop Relat Res. 1988; 227: 67-81.
44. Meredith DS, Taher F, Cammisa FP, Girardi FP. Incidence, diagnosis, and management of sacral fractures following multilevel spinal arthrodesis. Spine J. 2013; 13: 1464-1469.
45. Rodrigues-Pinto R, Kurd MF, Schroeder GD, Kepler CK, Krieg JC, Holstein JH, et al. Sacral Fractures and Associated Injuries. Global Spine J. 2017; 7: 609-616.
46. Sabiston CP, Wing PC. Sacral Fractures: Classification and Neurologic Implications: The Journal of Trauma: Injury, Infection, and Critical Care. 1986; 26: 1113-1115.
47. Hollis AC, Ebbs SR, Mandari FN. The epidemiology and treatment of femur fractures at a northern tanzanian referral centre. Pan Afri Med J. 2015; 22: 338.
48. Kratter J. Textbook of forensic medicine based on German and Austrian legislation and its reorganization. In: Commentary on: Kremer C, Racette S, Dionne CA, Sauvageau A Discrimination of falls and blows in blunt head trauma: systematic study of the hat brim rule in relation to skull fractures J Forensic Sci 2008; 53:716-719. Journal of Forensic Sciences, 2011; 1921. 1662.
49. Rogers LF. Radiology of Skeletal Trauma. Vol 2. Churchill Livingstone; 1992.
50. Ehrlich E, Maxeiner H. External injury marks (wounds) on the head in different types of blunt trauma in an autopsy series. Med Law. 2002; 21: 773-782.
51. Fracasso T, Schmidt S, Schmeling A. Commentary on: Kremer C, Racette S, Dionne CA, Sauvageau A. Discrimination of falls and blows in blunt head trauma: systematic study of the hat brim rule in relation to skull fractures. J Forensic Sci 2008 May; 53(3):716–9. J Forensic Sci. 2011; 56: 1662.
52. Geserick G, Krocker K, Wirth I. Walcher’s hat brim line rule--a literature review. Arch Kriminol. 2014; 234: 73-90.
53. Maxeiner H, Ehrlich E. Site, number and depth of wounds of the scalp in falls and blows--a contribution to the validity of the so-called hat brim rule. Arch Kriminol. 2000; 205: 82-91.
54. Walcher K. Ber ‘“stumpfe”’ Kopfverletzungen. Dtsch Z Gesamte Gerichtl Med. 1931;17:22–9.
55. Talbot LA, Musiol RJ, Witham EK, Metter EJ. Falls in young, middle- aged and older community dwelling adults: perceived cause, environmental factors and injury. BMC Public Health. 2005; 5: 86.
56. Kani KK, Mulcahy H, Porrino JA, Chew FS. Thoracic cage injuries. Eur J Radiol. 2019; 110: 225-232.
57. Senekjian L, Nirula R. Rib Fracture Fixation. Crit Care Clin. 2017; 33: 153-165.
58. Restrepo CS, Martinez S, Lemos DF, Washington   L,   McAdams HP, Vargas D, et al. Imaging Appearances of the Sternum and Sternoclavicular Joints. Radiographics. 2009; 29: 839-859.
59. Cooper C, O’Neill T, Silman A. The epidemiology of vertebral fractures. European Vertebral steoporosis Study Group. Bone. 1993; 14: 89-97.
60. Richter D, Hahn MP, Ostermann PAW, Ekkernkamp A, Muhr G.Vertical deceleration injuries: a comparative study of the injury patterns of 101 patients after accidental and intentional high falls. Injury. 1996; 27: 655-659.
61. Wedel VL, Galloway A. Broken Bones: Anthropological Analysis of Blunt Force Trauma. 1st edition. Springfield, Illinois: Charles C Thomas Pub Ltd; 1999; 504.
62. Henn V, Lignitz E. Kicking and Trampling to Death. In: Forensic Pathology Reviews. Totowa, NJ: Humana Press. 2004; 31–50. (Forensic Pathology Reviews).
63. Shepherd JP, Al-Kotany MY, Subadan C, Scully C. Assault and facial soft tissue injuries. Br J Plastic Surgery. 1987; 40: 614-619.
64. Strauch H, Wirth I, Taymoorian U, Geserick G. Kicking to death - forensic and criminological aspects. Forensic Sci Int. 2001; 123: 165- 171.