- Email: [email protected]
Biomechanical analysis of biphasic distribution of skull injury in falls from height
Accepted Manuscript Title: Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height Author: Sungji Park Jang Gyu Cha Youngseok Lee Insoo Seo Bongwoo Lee Youngshik Choi Woongchul Choi Kyungmoo Yang PII: DOI: Reference:
S0379-0738(15)00244-3 http://dx.doi.org/doi:10.1016/j.forsciint.2015.06.009 FSI 8034
To appear in:
FSI
Received date: Revised date: Accepted date:
6-1-2015 27-5-2015 8-6-2015
Please cite this article as: S. Park, J.G. Cha, Y. Lee, I. Seo, B. Lee, Y.C. ,Woongchul Choi, K. Yang, Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height, Forensic Science International (2015), http://dx.doi.org/10.1016/j.forsciint.2015.06.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ip t
Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height
Sungji Parka, Jang Gyu Chab, Youngseok Leea, Insoo Seoa, Bongwoo Leea, Youngshik
us
cr
Choia ,Woongchul Choic, Kyungmoo Yanga,*
National Forensic Service, Wonju city, South Korea
b
Department of radiology, Soonchunhyang university, Bucheon city, South Korea
c
Department of Automotive Engineering, Kookmin University, Seoul City, South Korea
te
d
M
an
a
* Corresponding author. Tel.: +82 10 9228 8606 ; fax: +82 2 2600 4828
Ac ce p
E-mail address:[email protected]
Page 1 of 13
ip t
Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height
Highlights
cr
Skull injury patterns of Falls from Height were investigated through biomechanical analysis.
Biphasic distribution of severe head injuries related to the height of the fall was studied and explained in
us
detail
Various research findings from other investigators were well correlated with the current research results.
an
Current research revealed the strength of the biomechanical analysis that would explain the injury
M
mechanism of FFH cases.
Abstract
d
Fall from height is one of the most common ways of suicide in Korea. Skull fractures are typically
te
accompanied with these cases, but several autopsy cases show absence of skull fracture even with serious body injuries including sternal and vertebral fracture. The mechanism of this pattern of injury can be
Ac ce p
explained by impact of facial part on chest or abdomen when the back of the body touches the ground first. We tried to figure out the relevance of this pattern of injury to the height of fall using a computer simulation tool (MADYMO 7.5). For this experiment, a condition of initial pose was limited to leaning forward. The simulation showed that when the body rotated forward, the body parts which got injured by the ground depended on the height of fall. For relatively lower height, head got injured, but as the height was set higher, the point of first impact area changed to the back, hip and then legs. When the body struck first around hip area on supine position, the impact made forceful flexion of lumbar, thoracic and cervical vertebrae, leading to folding the body in two, which resulted in collision between the part of face and the anterior part of body. Through the current investigation, it was explained that the biphasic distribution of the number of head injury cases versus the height distribution was attributed to the forward rotation of the body during the fall.
Page 2 of 13
cr
ip t
Key words: skull fracture, fall from height, biphasic, simulation, MADYMO
1. Introduction
us
When human body is impacted by the collision of fall from height, wide range of injuries including skull fractures are observed. Usually, it is assumed that the affected skull damages, organ damages and fractures are more extensive as the height of fall is higher. However, several cases of fall from height
an
showed absence of skull fracture even with serious body injuries including sternal and vertebral fractures and various soft and skeletal injuries on their back. By careful autopsy examination, this pattern of injuries
M
could be explained by hyper-flexed vertebral folding and impact of facial part on chest or abdomen when their back side of body touched the ground first. Based on this observation, we postulated that this phenomena could be a factor of explaining somewhat mysterious ‘biphasic distribution of skull injury’
d
with regard to the falling height or impacting energy [1, 2]. Thus, the fall from height cases without skull
te
fracture were further reviewed and the biomechanical simulations for the simple situation of fall from
Ac ce p
various heights were carried out with a multi-body dynamic simulation software, MADYMO 7.5.
2. Material and methods
12 falls from height autopsy cases showing absence of skull fracture with serious injuries at their
back were reviewed. The cases were collected from the autopsy data of National Forensic Service in South Korea within the recent 5 years (2010 to 2014). As an effort to systematically understand the injuries on human body by falls from height, widely
used multi-body dynamic simulation software called, MADYMO 7.5 was adopted in the current study. For a human body model, a typical pedestrian body model was used provided by MADYMO 7.5 program. This pedestrian body model consisted of 75 ellipsoids and the height and weight of the simulated body was 174 cm and 75 kg, respectively. Posture before the fall was assumed to be straight up and forward falling
Page 3 of 13
as an initial investigation since it seemed rather common while actual situations before the fall might vary
ip t
widely depending on the occasions case by case.
3. Results
cr
3-1 Autopsy results
us
Total of 12 autopsy cases of FFH where no head injuries were analyzed. In Table 1, in all cases, the damages on the posterior of the body including back, waist, buttock and posterior thigh were noticed by
an
compressed abrasions and soft tissue injuries.
M
Table 1. Distribution of injuries in falling from height autopsy cases showing no skull fracture
Back impaction
d
Height
Chin area Injuries: abrasion (A) bruise (B), mandibular fracture (MFx)
1
Ac ce p
te
Cases
Vertebral fracture site: cervical (C), thoracic (T), lumbar (L)
Chest injuries: sternal fracture (SFx), multiple anterior rib fracture (ARFx), bruise on sternal area (B)
Scalp injuries: abrasion (A), subgaleal hemorrhage (SGH) Site: occipital (O), parietal (P)
Cerebral injuries: focal subarachnoid hemorrhage (SAH)
10m
+
T11
-
-
A (O)
-
~10m
+
C3,4, T7,8, L2,3
ARFx, B
A
SGH (O)
-
10m
+
T7
B
B
SGH (P)
-
10m
+
C1, T12, L1
SFx, B
B
SGH (O)
SAH
12.5m
+
T6,9, L1
ARFx
-
-
-
~20m
+
C3,4, L1-4
SFx, B
B
SGH (P)
SAH
20m
+
C1, T4,7,12
SFx, B
A, B, MFx
-
SAH
8
22.5m
+
C4,5
ARFx, B
B
-
SAH
9
22.5m
+
C7
ARFx
-
-
SAH
10
30m
+
L1
ARFx
-
A (O)
-
2 3 4 5 6 7
Page 4 of 13
32.5m
+
C6, T4
ARFx, B
A
-
SAH
12
32.5m
+
C2,3, T1, T3
SFx, B
-
SGH (O)
SAH
100% (12/12)
C: 58.3% (7/12) T: 75% (9/12) L: 1.7% (5/12)
91.7% (11/12)
58.3% (7/12)
58.3% (7/12)
58.3% (7/12)
cr
Notes
ip t
11
us
Also, all cases showed vertebral fractures. Most common hyper-flexion fractures were found around thoracic area. Multiple posterior rib fractures including thoracic vertebral fracture site were also found in most of cases. 11 cases revealed sternal (4 cases) or multiple anterior rib fractures (6 cases) or bruise (1
an
case). Among these cases, 7 cases (Case 2, 3, 4, 6, 7, 8 and 11) disclosed chin area injuries including abrasions, bruises and mandibular fracture as shown Fig. 1 and the remaining 5 cases showed no evidence
Ac ce p
te
d
M
of chin area injury.
(a)
(b)
Figure 1: (a) Abrasion and bruise on chin area (ellipsoid) and bruise on sternal area (rectangle) (b) The deceased’s neck was able to be flexed so the chin area could approach to the sternal bruise area (case 7)
Typically patterned intradermal hemorrhages caused by clothes on the chin area and on the chest were noticed correspondingly in two cases as shown in Fig. 2 and 3. All cases showed the absence of skull fracture but the evidence of head injuries including small abrasion, contusion on scalp or focal
Page 5 of 13
subarachnoid hemorrhage on cerebral hemisphere. Based on all these observation and investigation, it was determined that the correlation was strong between the phenomena where the damages around the back
ip t
area, vertebral hyper-flexion fractures, and the damages around the mandible and the anterior part of chest were noticed and the phenomena where the skull fractures were NOT found. For the reviewed cases,
cr
falling height range is 10 to 32.5 meter. The one floor was transformed as 2.5 meter because exact falling
d
M
an
us
height were not measured in all cases by a ruler.
te
Figure 2: Shirt marks (circle: similar intradermal hemorrhage caused by the collar part of cloth) were
Ac ce p
impressed symmetrically on mandible and upper chest (case 6).
Page 6 of 13
Figure 3: Another case (case 4) also showed symmetrically impressed shirt marks (circle area: similar
ip t
intradermal hemorrhage caused by folded thin cloth) on the mandible and the upper chest.
cr
3-2 Computer simulation results
First simulation result of human body dynamics for the case of fall from 1 m height is illustrated in
us
Fig. 4. In the current simulation of human body dynamics during the fall, arbitrary action and/or reaction by the human was completely ignored due to the computer simulation capability and the arbitrary nature of the human actions and/or reactions. As shown in Fig. 4, overall body shapes changes starting from the
an
straight up position to the folded around waist line naturally and at the end, the top of the head of the human body model hits the ground first. For the case of forward falls, rotational motion of the human body
Ac ce p
te
d
M
model occurs consistently over various height situations.
Figure 4: Posture changes during the fall at 1 m height
As illustrated in Fig. 5 (a), for the case of fall from 2 m height, the back of the head (occiput) contacts
the ground first and for the case of fall from 6 m height (Fig. 5 (b)), the occipital region and the back of the shoulder makes the initial contacts to the ground absorbing all the impact from the fall. Additional simulations were repeated with height increases all the way up to 30 m case. As the height was raised, the rotation of the body model continued. For the case of fall from 14 m height, the haunch of the body model
Page 7 of 13
hits the ground first. Especially for the case of haunch first impact from forward falling case, hyper-flexion around the head and the upper body was noticed and as shown in Fig. 5 (f), the mandible of the body
ip t
model made a collision against the chest. Furthermore, it was also observed that the thoracic vertebra area was folded. For the fall from 20 m height, leg first impact on the ground was estimated and for the fall
cr
from 30 m height, the frontal body of the model made the direct contact to the ground as the body
(b) 3 m
(c) 6 m
(d) 10 m
(g) 20 m
(h) 30 m
Ac ce p
te
d
(a) 2 m
M
an
us
continued to rotate with the height increase.
(e) 13 m
(f) 14 m
Figure 5: Posture changes at impact for falls from various heights
The phenomena of the collision between the mandible and the chest of the model and the hyperflexion around the thoracic vertebra appeared consistently in the cases of falls from 13 m ~ 18 m height. In other words, when the legs hit the ground first in an oblique angle, the haunch area followed to impact the
Page 8 of 13
ground. Due the strong impact on the haunch, the hyper-flexion around the thoracic vertebra occurred and eventually the mandible and the chest collided against each other. This phenomenon was observed
ip t
consistently with the height increase until the legs of the body model hit the ground with near straight up position. In this case, the head of the body model did not make any direct contact with the ground however,
cr
the head made the final contact to the ground from the secondary motion after the rebound from the initial contact to the ground. With more simulations for higher altitude cases, it was clearly observed that the
us
body model continued to rotate and eventually caused the frontal body impact. For these cases, the direct collision of the head resulting into severe head injuries started to reappear again.
an
4. Discussion
Fall from height (FFH) is one of the most common ways of suicide and is becoming one of the increasing causes of common accidental deaths in South Korea. With the solid tendency of current
M
urbanization and manhattanization (high rise buildings), cases of falls from height is fully expected to increase steadily. Nevertheless, not all these cases come to autopsies for detailed forensic inspection and
d
naturally the understanding of damages on human bodies are quite limited to a certain extent. With a lack of solid evidence for the case of FFH like incident, series of questions are always surfacing such as 1) is
te
this FFH case?, 2) was he/she alive during the fall or dead prior to the major impact?, 3) is there any suspicious or other kind of injuries that are not compatible with FFH?, 4) how come there is no bleeding
Ac ce p
and/or skull injury even after the fall from high floors?. Especially for suspected murder cases, these questions become even more critical to support crime or incident scene inspection of law enforcement group because most of clues for possible criminal actions are left in the scene itself. In regards to the relation between the height and the damage on human body, Gilbert Lau analyzed
416 cases of FFH where the height information and the body age were known and reported the existence of correlation between the ages of human and the heights of the fall [3]. S. M. Gupta reported interesting observation in his paper such that the skull damages seemed to decrease as the heights of fall seemed to increase [4]. Riordain utilized a numerical simulation technique to simulate damages on the human body from a case of fall from a chair. With the impact simulation results and the damage reports from related autopsies, he explained the sequence of falls and resulting damages on the human bodies [5]. Elizabeth E.
Page 9 of 13
Tǖrk reported another interesting observation such that the head injuries from the falls from height of less than 10 m high or more than 25 m high were found to be severe while barely any injuries were noticed
ip t
from FFH for the cases between the 10 m and the 25 m high [6]. A. Tatjana et al. studied for 660 cases of FFH and reported similar observation that head injuries decreased as the height increased up to 15 m and
cr
remained relatively low for the cases from 15 m to 36 m high and then increased again for higher heights
M
an
us
as shown in Fig. 6 [1].
te
d
Figure 6: Percentage of cases with head injuries related to the height of fall [1]
Y. Weilemann studied for 20 cases with CT and compared the fractures and their distributions with
Ac ce p
the impact energy showing the similar trends noticed in the other observations discussed earlier. Cranial fracture frequency displayed a biphasic distribution with regard to the impact energy. The cranial fractures were more frequent in energies of less than 10kJ, and more than 20 kJ, but less frequent in the intermediate energy range around 10~20kJ [2]. Just as discussed earlier, the tendency of decreasing head injuries with increasing heights up to a certain level and re-occurrence of increasing head injuries along with further increasing heights has been reported in many research papers. However, not many papers are found that discuss about the reasons for this biphasic distribution systematically. Therefore, in our current research, efforts have been made to understand the reasons of rare head injuries for increasing fall heights to a certain level and to analyze the characteristics of damages on the human bodies shown in these cases. Based on the analysis of the multi-body dynamics utilizing MADYMO 7.5 for the cases of forward
Page 10 of 13
falls from various heights, the following observations were made. For the relatively low height cases (less than 10 m), the head first collisions of the body model were noticed since the body made face forward
ip t
rotation as it fell from the height. However, for the relatively moderate height cases (10 m ~ 18 m), the haunch first or the leg first contacts of the body model were observed as the body made more than 270
cr
degree rotation. In these cases, the direct head collisions of the body model to the ground were not estimated. For the cases of falls from more than 20 m height, it was determined that the frontal body of the
te
d
M
an
us
model made direct hit to the ground causing the direct head collision against the ground.
Ac ce p
Figure 7: Acceleration at the head prior to the impact vs height of fall
In Fig. 7, the acceleration values at the head of the body model right prior to the ground impact were
plotted against the height changes. It may not be possible to correlate the acceleration values at the head and the skull fracture cases in a quantitative manner. However, it is perfectly reasonable to connect these two parameters in a qualitative manner especially for the relative merit analysis. As depicted in Fig. 7, the acceleration at the head of the body model started to decrease at the height of 11 m and stayed reasonably low and then for the height of 30 m and higher, the acceleration started to increase sharply. Very interesting enough, this tendency showed the biphasic distribution as illustrated in Fig. 6 earlier. In other words, for the cases of relatively moderate heights (10 ~ 18 m), even with the height increase, the reasonable low
Page 11 of 13
acceleration values at the head were estimated indirectly proving that the direct head collision did not occur as the forward body rotation made the haunch first and/or legs first contacts to the ground as depicted in
ip t
Fig. 5 (d), (e), (f) and (g). However, with yet further increase of the fall heights (higher than 20 m), sharply increasing acceleration values were estimated secondarily implying the head collision against the ground as
cr
the continued forward body rotation caused the frontal impact of the body model as illustrated in Fig. 5 (h). It is quite noteworthy that the back or the legs of the body model hits the ground first at an oblique
us
angle just as shown in Fig. 5 (d), (e) and (f), impact in the haunch area, hyper-flexion in the vertebrae and the collision on the mandible and the chest seemed to occur in sequence. This is very similar to the autopsy
an
findings.
M
6. Conclusion
With the multi-body dynamic simulation analysis for the number of cases of falls from various heights and the reports from the autopsy findings, the reason for the phenomena that the cases of head
d
injuries decreased as the height increased to a certain level and after that the cases of head injuries
te
increased as the height increased further up beyond the certain level, was explained. For the cases of forward falls from relatively low heights, the head made the direct impact to the ground resulting in the
Ac ce p
skull fractures and related injuries easily. However, for the cases of forward falls from relatively moderate heights, forward rotation of the body continued and ended up causing other portion of the body, typically the back, legs, haunch and more, to contact the ground first. Therefore, the initial main impact on the back, legs and haunch usually triggered the folding of upper body resulting into the forward motion of the head and avoiding the direct head impact against the ground. In this case, mainly due to the hyper-flexion in the cervical vertebrae, it was determined that severe collision could be occurred between the mandible and the chest to cause the skin injuries, sternal fractures and more. In this research, it was found and explained that the biphasic distribution of the number of head
injury cases versus the height distribution was mainly due to the forward rotation of the body during the fall. Also, for these cases, it was understood that the trace of damages on the mandible and the chest were occasionally found in response to the collision to the ground with the back of the body first.
Page 12 of 13
ip t
References
[1] Tatjana C. Atanasijevic, Slobodan N. Savic, Slobodan D. Nikolic, Vesna M. Djokic, Frequency and
cr
severity of injuries in correlation with the height of fall, Journal of forensic science, Vol. 5, May 2005.
us
[2] Y. Weilemann, M.J. Thali, B.P. Kneubuehl, S.A. Bolliger, Correlation between skeletal trauma and energy in falls from great height detected by post-mortem multislice computed tomography (MSCT), Forensic Science International, 2008. Vol. 180, p81-85
an
[3] Gilbert Lau, Peng Lim Ooi, Bernadette Phoon, Fatal falls from a height: the use of mathematical models to estimate the height of fall from the injuries sustained, Forensic Science International 93 (1998)
M
33-44.
[4] S. M. Gupta, J. Chadra, T. D. Dogra, Blunt force lesions related to the heights of a fall, The American
d
Journal of Forensic Medicine and Pathology, Vol. 3, No. 1, March 1982. [5] K O’Riordain, P.M. Thomas, J.P. Phillips, M.D. Gilchrist, Reconstruction of real world head injury
te
accidents resulting from falls using multibody dynamics, Clinical Biomechanics 18 (2003) 590–600. [6] Elizabeth E. Tǖrk, Michael Tsokos, Pathologic features of fatal falls from height, The American
Ac ce p
Journal of Forensic Medicine and Pathology, Vol. 25, No. 3, September 2004.
Page 13 of 13
S0379-0738(15)00244-3 http://dx.doi.org/doi:10.1016/j.forsciint.2015.06.009 FSI 8034
To appear in:
FSI
Received date: Revised date: Accepted date:
6-1-2015 27-5-2015 8-6-2015
Please cite this article as: S. Park, J.G. Cha, Y. Lee, I. Seo, B. Lee, Y.C. ,Woongchul Choi, K. Yang, Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height, Forensic Science International (2015), http://dx.doi.org/10.1016/j.forsciint.2015.06.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ip t
Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height
Sungji Parka, Jang Gyu Chab, Youngseok Leea, Insoo Seoa, Bongwoo Leea, Youngshik
us
cr
Choia ,Woongchul Choic, Kyungmoo Yanga,*
National Forensic Service, Wonju city, South Korea
b
Department of radiology, Soonchunhyang university, Bucheon city, South Korea
c
Department of Automotive Engineering, Kookmin University, Seoul City, South Korea
te
d
M
an
a
* Corresponding author. Tel.: +82 10 9228 8606 ; fax: +82 2 2600 4828
Ac ce p
E-mail address:[email protected]
Page 1 of 13
ip t
Biomechanical Analysis of Biphasic Distribution of Skull Injury in Falls from Height
Highlights
cr
Skull injury patterns of Falls from Height were investigated through biomechanical analysis.
Biphasic distribution of severe head injuries related to the height of the fall was studied and explained in
us
detail
Various research findings from other investigators were well correlated with the current research results.
an
Current research revealed the strength of the biomechanical analysis that would explain the injury
M
mechanism of FFH cases.
Abstract
d
Fall from height is one of the most common ways of suicide in Korea. Skull fractures are typically
te
accompanied with these cases, but several autopsy cases show absence of skull fracture even with serious body injuries including sternal and vertebral fracture. The mechanism of this pattern of injury can be
Ac ce p
explained by impact of facial part on chest or abdomen when the back of the body touches the ground first. We tried to figure out the relevance of this pattern of injury to the height of fall using a computer simulation tool (MADYMO 7.5). For this experiment, a condition of initial pose was limited to leaning forward. The simulation showed that when the body rotated forward, the body parts which got injured by the ground depended on the height of fall. For relatively lower height, head got injured, but as the height was set higher, the point of first impact area changed to the back, hip and then legs. When the body struck first around hip area on supine position, the impact made forceful flexion of lumbar, thoracic and cervical vertebrae, leading to folding the body in two, which resulted in collision between the part of face and the anterior part of body. Through the current investigation, it was explained that the biphasic distribution of the number of head injury cases versus the height distribution was attributed to the forward rotation of the body during the fall.
Page 2 of 13
cr
ip t
Key words: skull fracture, fall from height, biphasic, simulation, MADYMO
1. Introduction
us
When human body is impacted by the collision of fall from height, wide range of injuries including skull fractures are observed. Usually, it is assumed that the affected skull damages, organ damages and fractures are more extensive as the height of fall is higher. However, several cases of fall from height
an
showed absence of skull fracture even with serious body injuries including sternal and vertebral fractures and various soft and skeletal injuries on their back. By careful autopsy examination, this pattern of injuries
M
could be explained by hyper-flexed vertebral folding and impact of facial part on chest or abdomen when their back side of body touched the ground first. Based on this observation, we postulated that this phenomena could be a factor of explaining somewhat mysterious ‘biphasic distribution of skull injury’
d
with regard to the falling height or impacting energy [1, 2]. Thus, the fall from height cases without skull
te
fracture were further reviewed and the biomechanical simulations for the simple situation of fall from
Ac ce p
various heights were carried out with a multi-body dynamic simulation software, MADYMO 7.5.
2. Material and methods
12 falls from height autopsy cases showing absence of skull fracture with serious injuries at their
back were reviewed. The cases were collected from the autopsy data of National Forensic Service in South Korea within the recent 5 years (2010 to 2014). As an effort to systematically understand the injuries on human body by falls from height, widely
used multi-body dynamic simulation software called, MADYMO 7.5 was adopted in the current study. For a human body model, a typical pedestrian body model was used provided by MADYMO 7.5 program. This pedestrian body model consisted of 75 ellipsoids and the height and weight of the simulated body was 174 cm and 75 kg, respectively. Posture before the fall was assumed to be straight up and forward falling
Page 3 of 13
as an initial investigation since it seemed rather common while actual situations before the fall might vary
ip t
widely depending on the occasions case by case.
3. Results
cr
3-1 Autopsy results
us
Total of 12 autopsy cases of FFH where no head injuries were analyzed. In Table 1, in all cases, the damages on the posterior of the body including back, waist, buttock and posterior thigh were noticed by
an
compressed abrasions and soft tissue injuries.
M
Table 1. Distribution of injuries in falling from height autopsy cases showing no skull fracture
Back impaction
d
Height
Chin area Injuries: abrasion (A) bruise (B), mandibular fracture (MFx)
1
Ac ce p
te
Cases
Vertebral fracture site: cervical (C), thoracic (T), lumbar (L)
Chest injuries: sternal fracture (SFx), multiple anterior rib fracture (ARFx), bruise on sternal area (B)
Scalp injuries: abrasion (A), subgaleal hemorrhage (SGH) Site: occipital (O), parietal (P)
Cerebral injuries: focal subarachnoid hemorrhage (SAH)
10m
+
T11
-
-
A (O)
-
~10m
+
C3,4, T7,8, L2,3
ARFx, B
A
SGH (O)
-
10m
+
T7
B
B
SGH (P)
-
10m
+
C1, T12, L1
SFx, B
B
SGH (O)
SAH
12.5m
+
T6,9, L1
ARFx
-
-
-
~20m
+
C3,4, L1-4
SFx, B
B
SGH (P)
SAH
20m
+
C1, T4,7,12
SFx, B
A, B, MFx
-
SAH
8
22.5m
+
C4,5
ARFx, B
B
-
SAH
9
22.5m
+
C7
ARFx
-
-
SAH
10
30m
+
L1
ARFx
-
A (O)
-
2 3 4 5 6 7
Page 4 of 13
32.5m
+
C6, T4
ARFx, B
A
-
SAH
12
32.5m
+
C2,3, T1, T3
SFx, B
-
SGH (O)
SAH
100% (12/12)
C: 58.3% (7/12) T: 75% (9/12) L: 1.7% (5/12)
91.7% (11/12)
58.3% (7/12)
58.3% (7/12)
58.3% (7/12)
cr
Notes
ip t
11
us
Also, all cases showed vertebral fractures. Most common hyper-flexion fractures were found around thoracic area. Multiple posterior rib fractures including thoracic vertebral fracture site were also found in most of cases. 11 cases revealed sternal (4 cases) or multiple anterior rib fractures (6 cases) or bruise (1
an
case). Among these cases, 7 cases (Case 2, 3, 4, 6, 7, 8 and 11) disclosed chin area injuries including abrasions, bruises and mandibular fracture as shown Fig. 1 and the remaining 5 cases showed no evidence
Ac ce p
te
d
M
of chin area injury.
(a)
(b)
Figure 1: (a) Abrasion and bruise on chin area (ellipsoid) and bruise on sternal area (rectangle) (b) The deceased’s neck was able to be flexed so the chin area could approach to the sternal bruise area (case 7)
Typically patterned intradermal hemorrhages caused by clothes on the chin area and on the chest were noticed correspondingly in two cases as shown in Fig. 2 and 3. All cases showed the absence of skull fracture but the evidence of head injuries including small abrasion, contusion on scalp or focal
Page 5 of 13
subarachnoid hemorrhage on cerebral hemisphere. Based on all these observation and investigation, it was determined that the correlation was strong between the phenomena where the damages around the back
ip t
area, vertebral hyper-flexion fractures, and the damages around the mandible and the anterior part of chest were noticed and the phenomena where the skull fractures were NOT found. For the reviewed cases,
cr
falling height range is 10 to 32.5 meter. The one floor was transformed as 2.5 meter because exact falling
d
M
an
us
height were not measured in all cases by a ruler.
te
Figure 2: Shirt marks (circle: similar intradermal hemorrhage caused by the collar part of cloth) were
Ac ce p
impressed symmetrically on mandible and upper chest (case 6).
Page 6 of 13
Figure 3: Another case (case 4) also showed symmetrically impressed shirt marks (circle area: similar
ip t
intradermal hemorrhage caused by folded thin cloth) on the mandible and the upper chest.
cr
3-2 Computer simulation results
First simulation result of human body dynamics for the case of fall from 1 m height is illustrated in
us
Fig. 4. In the current simulation of human body dynamics during the fall, arbitrary action and/or reaction by the human was completely ignored due to the computer simulation capability and the arbitrary nature of the human actions and/or reactions. As shown in Fig. 4, overall body shapes changes starting from the
an
straight up position to the folded around waist line naturally and at the end, the top of the head of the human body model hits the ground first. For the case of forward falls, rotational motion of the human body
Ac ce p
te
d
M
model occurs consistently over various height situations.
Figure 4: Posture changes during the fall at 1 m height
As illustrated in Fig. 5 (a), for the case of fall from 2 m height, the back of the head (occiput) contacts
the ground first and for the case of fall from 6 m height (Fig. 5 (b)), the occipital region and the back of the shoulder makes the initial contacts to the ground absorbing all the impact from the fall. Additional simulations were repeated with height increases all the way up to 30 m case. As the height was raised, the rotation of the body model continued. For the case of fall from 14 m height, the haunch of the body model
Page 7 of 13
hits the ground first. Especially for the case of haunch first impact from forward falling case, hyper-flexion around the head and the upper body was noticed and as shown in Fig. 5 (f), the mandible of the body
ip t
model made a collision against the chest. Furthermore, it was also observed that the thoracic vertebra area was folded. For the fall from 20 m height, leg first impact on the ground was estimated and for the fall
cr
from 30 m height, the frontal body of the model made the direct contact to the ground as the body
(b) 3 m
(c) 6 m
(d) 10 m
(g) 20 m
(h) 30 m
Ac ce p
te
d
(a) 2 m
M
an
us
continued to rotate with the height increase.
(e) 13 m
(f) 14 m
Figure 5: Posture changes at impact for falls from various heights
The phenomena of the collision between the mandible and the chest of the model and the hyperflexion around the thoracic vertebra appeared consistently in the cases of falls from 13 m ~ 18 m height. In other words, when the legs hit the ground first in an oblique angle, the haunch area followed to impact the
Page 8 of 13
ground. Due the strong impact on the haunch, the hyper-flexion around the thoracic vertebra occurred and eventually the mandible and the chest collided against each other. This phenomenon was observed
ip t
consistently with the height increase until the legs of the body model hit the ground with near straight up position. In this case, the head of the body model did not make any direct contact with the ground however,
cr
the head made the final contact to the ground from the secondary motion after the rebound from the initial contact to the ground. With more simulations for higher altitude cases, it was clearly observed that the
us
body model continued to rotate and eventually caused the frontal body impact. For these cases, the direct collision of the head resulting into severe head injuries started to reappear again.
an
4. Discussion
Fall from height (FFH) is one of the most common ways of suicide and is becoming one of the increasing causes of common accidental deaths in South Korea. With the solid tendency of current
M
urbanization and manhattanization (high rise buildings), cases of falls from height is fully expected to increase steadily. Nevertheless, not all these cases come to autopsies for detailed forensic inspection and
d
naturally the understanding of damages on human bodies are quite limited to a certain extent. With a lack of solid evidence for the case of FFH like incident, series of questions are always surfacing such as 1) is
te
this FFH case?, 2) was he/she alive during the fall or dead prior to the major impact?, 3) is there any suspicious or other kind of injuries that are not compatible with FFH?, 4) how come there is no bleeding
Ac ce p
and/or skull injury even after the fall from high floors?. Especially for suspected murder cases, these questions become even more critical to support crime or incident scene inspection of law enforcement group because most of clues for possible criminal actions are left in the scene itself. In regards to the relation between the height and the damage on human body, Gilbert Lau analyzed
416 cases of FFH where the height information and the body age were known and reported the existence of correlation between the ages of human and the heights of the fall [3]. S. M. Gupta reported interesting observation in his paper such that the skull damages seemed to decrease as the heights of fall seemed to increase [4]. Riordain utilized a numerical simulation technique to simulate damages on the human body from a case of fall from a chair. With the impact simulation results and the damage reports from related autopsies, he explained the sequence of falls and resulting damages on the human bodies [5]. Elizabeth E.
Page 9 of 13
Tǖrk reported another interesting observation such that the head injuries from the falls from height of less than 10 m high or more than 25 m high were found to be severe while barely any injuries were noticed
ip t
from FFH for the cases between the 10 m and the 25 m high [6]. A. Tatjana et al. studied for 660 cases of FFH and reported similar observation that head injuries decreased as the height increased up to 15 m and
cr
remained relatively low for the cases from 15 m to 36 m high and then increased again for higher heights
M
an
us
as shown in Fig. 6 [1].
te
d
Figure 6: Percentage of cases with head injuries related to the height of fall [1]
Y. Weilemann studied for 20 cases with CT and compared the fractures and their distributions with
Ac ce p
the impact energy showing the similar trends noticed in the other observations discussed earlier. Cranial fracture frequency displayed a biphasic distribution with regard to the impact energy. The cranial fractures were more frequent in energies of less than 10kJ, and more than 20 kJ, but less frequent in the intermediate energy range around 10~20kJ [2]. Just as discussed earlier, the tendency of decreasing head injuries with increasing heights up to a certain level and re-occurrence of increasing head injuries along with further increasing heights has been reported in many research papers. However, not many papers are found that discuss about the reasons for this biphasic distribution systematically. Therefore, in our current research, efforts have been made to understand the reasons of rare head injuries for increasing fall heights to a certain level and to analyze the characteristics of damages on the human bodies shown in these cases. Based on the analysis of the multi-body dynamics utilizing MADYMO 7.5 for the cases of forward
Page 10 of 13
falls from various heights, the following observations were made. For the relatively low height cases (less than 10 m), the head first collisions of the body model were noticed since the body made face forward
ip t
rotation as it fell from the height. However, for the relatively moderate height cases (10 m ~ 18 m), the haunch first or the leg first contacts of the body model were observed as the body made more than 270
cr
degree rotation. In these cases, the direct head collisions of the body model to the ground were not estimated. For the cases of falls from more than 20 m height, it was determined that the frontal body of the
te
d
M
an
us
model made direct hit to the ground causing the direct head collision against the ground.
Ac ce p
Figure 7: Acceleration at the head prior to the impact vs height of fall
In Fig. 7, the acceleration values at the head of the body model right prior to the ground impact were
plotted against the height changes. It may not be possible to correlate the acceleration values at the head and the skull fracture cases in a quantitative manner. However, it is perfectly reasonable to connect these two parameters in a qualitative manner especially for the relative merit analysis. As depicted in Fig. 7, the acceleration at the head of the body model started to decrease at the height of 11 m and stayed reasonably low and then for the height of 30 m and higher, the acceleration started to increase sharply. Very interesting enough, this tendency showed the biphasic distribution as illustrated in Fig. 6 earlier. In other words, for the cases of relatively moderate heights (10 ~ 18 m), even with the height increase, the reasonable low
Page 11 of 13
acceleration values at the head were estimated indirectly proving that the direct head collision did not occur as the forward body rotation made the haunch first and/or legs first contacts to the ground as depicted in
ip t
Fig. 5 (d), (e), (f) and (g). However, with yet further increase of the fall heights (higher than 20 m), sharply increasing acceleration values were estimated secondarily implying the head collision against the ground as
cr
the continued forward body rotation caused the frontal impact of the body model as illustrated in Fig. 5 (h). It is quite noteworthy that the back or the legs of the body model hits the ground first at an oblique
us
angle just as shown in Fig. 5 (d), (e) and (f), impact in the haunch area, hyper-flexion in the vertebrae and the collision on the mandible and the chest seemed to occur in sequence. This is very similar to the autopsy
an
findings.
M
6. Conclusion
With the multi-body dynamic simulation analysis for the number of cases of falls from various heights and the reports from the autopsy findings, the reason for the phenomena that the cases of head
d
injuries decreased as the height increased to a certain level and after that the cases of head injuries
te
increased as the height increased further up beyond the certain level, was explained. For the cases of forward falls from relatively low heights, the head made the direct impact to the ground resulting in the
Ac ce p
skull fractures and related injuries easily. However, for the cases of forward falls from relatively moderate heights, forward rotation of the body continued and ended up causing other portion of the body, typically the back, legs, haunch and more, to contact the ground first. Therefore, the initial main impact on the back, legs and haunch usually triggered the folding of upper body resulting into the forward motion of the head and avoiding the direct head impact against the ground. In this case, mainly due to the hyper-flexion in the cervical vertebrae, it was determined that severe collision could be occurred between the mandible and the chest to cause the skin injuries, sternal fractures and more. In this research, it was found and explained that the biphasic distribution of the number of head
injury cases versus the height distribution was mainly due to the forward rotation of the body during the fall. Also, for these cases, it was understood that the trace of damages on the mandible and the chest were occasionally found in response to the collision to the ground with the back of the body first.
Page 12 of 13
ip t
References
[1] Tatjana C. Atanasijevic, Slobodan N. Savic, Slobodan D. Nikolic, Vesna M. Djokic, Frequency and
cr
severity of injuries in correlation with the height of fall, Journal of forensic science, Vol. 5, May 2005.
us
[2] Y. Weilemann, M.J. Thali, B.P. Kneubuehl, S.A. Bolliger, Correlation between skeletal trauma and energy in falls from great height detected by post-mortem multislice computed tomography (MSCT), Forensic Science International, 2008. Vol. 180, p81-85
an
[3] Gilbert Lau, Peng Lim Ooi, Bernadette Phoon, Fatal falls from a height: the use of mathematical models to estimate the height of fall from the injuries sustained, Forensic Science International 93 (1998)
M
33-44.
[4] S. M. Gupta, J. Chadra, T. D. Dogra, Blunt force lesions related to the heights of a fall, The American
d
Journal of Forensic Medicine and Pathology, Vol. 3, No. 1, March 1982. [5] K O’Riordain, P.M. Thomas, J.P. Phillips, M.D. Gilchrist, Reconstruction of real world head injury
te
accidents resulting from falls using multibody dynamics, Clinical Biomechanics 18 (2003) 590–600. [6] Elizabeth E. Tǖrk, Michael Tsokos, Pathologic features of fatal falls from height, The American
Ac ce p
Journal of Forensic Medicine and Pathology, Vol. 25, No. 3, September 2004.
Page 13 of 13