Accepted Manuscript

Review

The shape and mobility of the thoracic spine in asymptomatic adults - A sys-
tematic review of in vivo studies

Fumin Pan, Ali Firouzabadi, Sandra Reitmaier, Thomas Zander, Hendrik

Schmidt

PII: S0021-9290(18)30647-X
DOI: https://doi.org/10.1016/j.jbiomech.2018.07.041
Reference: BM 8804

To appear in: Journal of Biomechanics

Received Date: 17 May 2018

Revised Date: 27 July 2018
Accepted Date: 31 July 2018

Please cite this article as: F. Pan, A. Firouzabadi, S. Reitmaier, T. Zander, H. Schmidt, The shape and mobility of
the thoracic spine in asymptomatic adults - A systematic review of in vivo studies, Journal of Biomechanics (2018),
doi: https://doi.org/10.1016/j.jbiomech.2018.07.041

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.
 Review Article

The shape and mobility of the thoracic spine in asymptomatic adults - A systematic

review of in vivo studies

Fumin Pan a, Ali Firouzabadi a, Sandra Reitmaier a, Thomas Zander a, Hendrik Schmidt a,*

a
Julius Wolff Institute, Charité – Universitätsmedizin Berlin, corporate member of Freie
Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Germany

Corresponding author
*
Hendrik Schmidt

Julius Wolff Institute

Charité – Universitätsmedizin Berlin

Augustenburger Platz 1

13353 Berlin

Germany

Phone: +49 30 2093 46016

Fax: +49 2093 46001

Email: hendrik.schmidt@charite.de

Word count:

Abstract: 262

Introduction - Acknowledgments: 4595

1
Abstract

A comprehensive knowledge of the thoracic shape and kinematics is essential for effective

risk prevention, diagnose and proper management of thoracic disorders and assessment of

treatment or rehabilitation strategies as well as for in silico and in vitro models for realistic

applications of boundary conditions.

After an extensive search of the existing literature, this study summarizes 45 studies on in

vivo thoracic kyphosis and kinematics and creates a systematic and detailed database. The

thoracic kyphosis over T1–12 determined using non-radiological devices (34°) was relatively

less than measured using radiological devices (40°) during standing. The majority of

kinematical measurements are based on non-radiological devices. The thoracic range of

motion (RoM) was greatest during axial rotation (40°), followed by lateral bending (26°), and

flexion (21°) when determined using non-radiological devices during standing. The smallest

RoM was identified during extension (13°). The lower thoracic level (T8–12) contributed

more to the RoM than the upper (T1–4) and middle (T4–8) levels during flexion and lateral

bending. During axial rotation and extension, the middle level (T4–8) contributed the most.

Coupled motion was evident, mostly during lateral bending and axial rotation. With aging, the

thoracic kyphosis increased by about 3° per decade, whereas the RoM decreased by about 5°

per decade for all load directions. These changes with aging mainly occurred in the lower

region (T6–12). The influence of sex on thoracic kyphosis and the RoM has been described as

partly contradictory. Obesity was found to decrease the thoracic RoM. Studies comparing

standing, sitting and lying reported the effect of posture as significant.

Keywords: Thoracic spine, Kyphosis, Kinematics, Range of motion, In vivo, Review

2
Contents

1. Introduction

2. Methods

3. Results

3.1. Search results

3.2. Thoracic kyphosis

3.3. Thoracic Range of Motion (RoM)

3.3.1. Range of Flexion (RoF)

3.3.2. Range of Extension (RoE)

3.3.3. Range of Flexion plus Extension (RoFE)

3.3.4. Range of Lateral Bending (RoLB)

3.3.5. Range of Axial Rotation (RoAR)

3.3.6. Segmental RoM

3.3.7. Coupled motion

3.4. Influencing factors

3.4.1. Age

3.4.2. Sex

3.4.3. Posture

3.4.4. Other factors

4. Discussion

Conflict of interest

Acknowledgments

References

3
1. Introduction

The one-year prevalence of thoracic spine pain ranges from 3–55% among the occupational

population (Briggs et al., 2009; Fouquet et al., 2015). Particularly in the elderly population,

hyper-kyphosis and hypo-mobility of the thoracic spine occur frequently, and are associated

with poorer physical performance and impaired pulmonary function (Brigatto et al., 2014;

Jang et al., 2015; Kado et al., 2004). Thoracic fractures caused by high-force trauma during

accidents or osteoporosis display a prevalence of 1–35%, mainly in the lower segments

(Bakhsheshian et al., 2014; Cummings and Melton, 2002; Tong et al., 2016). Because of the

relatively narrow spinal canal, this is frequently accompanied by neurological complications

(Magerl et al., 1994). To rule out these disorders, a comprehensive knowledge of the

asymptomatic thoracic shape and kinematics is generally required.

An understanding of the asymptomatic thoracic shape and kinematics is not only important to

generate reference values for diagnosis, but is essential for optimal restoration of injured

thoracic spinal structures and for treatment assessment. Following thoracic fracture or spinal-

cord injury, the main aim is to restore thoracic curvature and regain sagittal balance (Lee et al.,

2016). Complex osteoporotic fractures or spinal metastases are also indications for

reconstructive and stabilizing surgeries in the thoracic spine (Aebi, 2003; Fleming, 1992). A

correction of hyper-kyphosis and the maintenance of thoracic curvature or mobility by

surgery or conservative treatment can help to improve the quality of life and ameliorate

respiratory function (Aboutorabi et al., 2018; Jang et al., 2015; Katzman et al., 2015).

In addition to its clinical relevance, knowledge of in vivo thoracic kyphosis and kinematics is

also necessary for in silico and in vitro models for realistic applications of boundary

conditions. Because of the stiffness caused by the rib cage, many musculoskeletal models

consider the thoracic region as rigid (Angelini et al., 2018; Arjmand et al., 2009; Arshad et al.,

4
2018; Granata and Wilson, 2001), which is in contrast to the real in vivo conditions, indicating

a high mobility of the thoracic spine (Madinei and Arjmand, 2018; Morita et al., 2014;

O'Gorman and Jull, 1987; Willems et al., 1996). In a computational model study, Ignasiak et

al. (2016) indicated out the importance of considering thoracic kinematics and demonstrated

that a rigid simulation of the thoracic spine would underestimate the lumbar loads. The in vivo

kinematical data can also be applied directly on cadaveric models with or without the rib cage

to investigate the kinetics of the thoracic spine (Volkheimer et al., 2015), or used as reference

to verify whether the applied loads are appropriate (Anderson et al., 2018; Brasiliense et al.,

2011; Mannen et al., 2015; 2017; 2018; Sis et al., 2016; Volkheimer et al., 2015).

Though with significant importance both in the clinical and biomechanical fields, fewer

studies were focused on the thoracic kinematics compared to the cervical and lumbar regions

(Brismee et al., 2006; Intolo et al., 2009; Pan et al., 2018; Wilke et al., 2017; Wood et al.,

1995). Therefore, in this study, we conducted a systematic review on in vivo asymptomatic

thoracic kyphosis and kinematics to fill a gap in the literature.

5
2. Methods

A literature search of PubMed (National Library of Medicine and National Institute of Health,

USA) was performed from inception to April 2018 according to the Preferred Reporting Items

for Systematic Reviews and Meta-Analyses (PRISMA) statements (Moher et al., 2010).

The following keywords were applied for the systematic search: thoracic spine, range of

motion, mobility, motion, kinematics, coupled motion, reposition error, flexion, extension,

rotation, bending. Only in vivo studies with adult participants (>18 years old) without pain in

the thoracic spine were included and irrelevant studies were excluded. A manual search of

references was performed to include possible missing studies. The STROBE (Strengthening

the Reporting of Observational Studies in Epidemiology) statement was used to assess the

quality of studies, and the authors’ “appraisal of risk of bias”(Vandenbroucke et al., 2007).

6
3. Results

3.1. Search results

The PubMed search yielded 893 hits. Based on the titles and abstracts, 804 references were

excluded, leaving 89 papers for eligibility assessment. Going through the full-texts, 41 studies

met the eligibility criteria. Together with four studies found by manual search, 45 studies

were included in this systematic review. The entire body of included studies with PubMed

links to abstracts is provided in Appendix 1. The total STROBE quality score (maximum = 34)

ranged from 11 to 29 (mean = 19) (Appendix 2). Additionally, the authors’ appraisal of risk of

bias varied considerably. Three studies were prospective while 42 were cross-sectional.

Twenty-five studies reported the validity or reliability of the results while 20 studies did not.

3.2. Thoracic kyphosis

Thoracic kyphosis was determined in seven studies using radiological devices, in thirteen

studies with non-radiological devices and in three studies using both (Table 1).

When determining thoracic kyphosis by X-ray, a standing posture is normally adopted with

arms flexed forward (Hasegawa et al., 2016; Iyer et al., 2016; Madinei and Arjmand, 2018) or

elevated (Edmondston et al., 2012a; 2012b). The thoracic kyphosis is normally determined

from T3 or 4 to 11 or 12, because thoracic vertebral bodies above T3 and below T11 could not

be consistently visualized in the radiographs of all participants (Edmondston et al., 2012b;

Horton et al., 2005; Vialle et al., 2005). However, the EOS device (EOS Imaging SA., Paris,

France) is now widely used to evaluate the sagittal balance with less radiation and better

exposure of the entire spine (Hasegawa et al., 2016; Iyer et al., 2016; Weitkunat et al., 2016).

Zemp et al. (2013) utilized MRI to assess thoracic kyphosis during sitting over T3–11 and

Lorbergs et al. (2017) utilized CT during supine lying for T4–12.

7
Because of the ethical problem of radiation exposure, non-radiological devices have been

used in many studies, including goniometer, inclinometer, photographs and videos,

accelerometer, ultrasound and optoelectronic devices. Thoracic kyphosis is determined during

standing, sitting or prone lying with arms hanging down or forward flexed (Edmondston et al.,

2012a; 2012b), and usually covers the whole thoracic spine from C7 or T1 to T12 or L1.

The mean thoracic kyphosis of T1–12 ranged from 29 to 45° (mean=34°, Fig.1) when

assessed using non-radiological devices during standing (Alderighi et al., 2016; Edmondston

et al., 2012a; 2012b; Mannion et al., 2004; Park et al., 2015; Schroeder et al., 2015) and about

40° using radiological devices (Hasegawa et al., 2016; Madinei and Arjmand, 2018). Over

T3–12, the thoracic kyphosis was 32° during standing, consisting of 14° for T3–6, 14° for

T6–9 and 4° for T9–12 (Schinkel-Ivy et al., 2014). During sitting, the kyphosis over T1–12

was 39° using a non-radiological device (Cloud et al., 2014). During supine lying, the

kyphosis was 33° over T4–12 by CT (Lorbergs et al., 2017). During prone lying, it was 32°

from C7 to the changing point (kyphosis to lordosis) (Salem et al., 2015).

3.3. Thoracic Range of Motion (RoM)

3.3.1. Range of Flexion (RoF)

The thoracic RoF was determined in one study using a radiological device, in 14 studies with

non-radiological devices, and in one study using both (Table 2).

During standing, the RoF of the whole thoracic spine (from C7 or T1 to T12 or L1) ranged

from 12° to 28° with a mean of 21° (Fig.2a) using both radiological and non-radiological

devices, which was slightly smaller than that of 28° when determined during sitting for C7–

L1 (Folsch et al., 2012). Over T3–12, the thoracic RoF was 18° during standing (Schinkel-Ivy

et al., 2014).

8
Regarding the contribution of different thoracic levels to the RoF, Madinei and Arjmand

(2018) found that the lower level (T6–12) contributed 60% to the whole thoracic RoF

measured by X-ray. Hajibozorgi and Arjmand (2016) and Narimani and Arjmand (2018)

using an inertial device also showed that the upper level (T1–5, 25%) contributed much less

than the lower level (T5–12, 75%) during standing. Alqhtani et al. (2015) in standing and

Willems et al. (1996) in sitting determined the RoF in three different areas: upper (T1–4),

middle (T4–8) and lower (T8–12) using non-radiological devices. Both studies demonstrated

that the lower level contributed more to the RoF than the upper and middle levels (67% vs. 17%

and 16% during standing, 40% vs. 27% and 33% during sitting), which are similar to the

measurements of Hajibozorgi and Arjmand (2016) and Narimani and Arjmand (2018).

Zemp et al. (2014) found that during sitting the RoF determined by photography was

relatively greater than by MRI, which was possibly caused by soft-tissue artefacts.

3.3.2. Range of Extension (RoE)

The thoracic RoE was determined in nine studies using non-radiological devices and in three

studies with both radiological and non-radiological devices (Table 3).

During standing, the mean RoE of the whole thoracic spine (T1–12 or C7–T12) ranged from

11° to 17°, with a mean of 13° (Fig.2b) using both radiological and non-radiological devices.

During sitting, the RoE determined by an ultrasonic device of the whole thoracic (C7–L1) was

as high as 26° (Folsch et al., 2012). During prone lying, Park et al. (2015) reported a thoracic

RoE of 7° for T1–12 using an inclinometer.

Narimani and Arjmand (2018) reported that the upper thoracic spine (T1–5, 65%) contributes

more to the whole RoE during standing than the lower level (T5–12, 35%). When measured in

three levels, the upper thoracic spine (T1–4) contributed 27%, the middle level (T4–8)

approximately 43% and lower level (T8–12) approximately 30% (Alqhtani et al., 2015).
9
During sitting, the middle (35%) and lower (34%) levels contributed almost equally to the

thoracic RoE and were slightly greater than the upper level (31%) (Willems et al., 1996).

Edmondston et al. (2012a; 2012b) compared the thoracic RoE between X-ray and

photographic measurements for T1–12 during standing and found no significant difference.

By contrast, Zemp et al. (2014) showed a greater RoE when measured by photographs than by

MRI during sitting for T3–11 because of soft-tissue artefacts.

3.3.3. Range of Flexion plus Extension (RoFE)

The thoracic RoFE was determined in one study using a radiological device and in four

studies with non-radiological devices (Table 4).

During standing, the RoFE was 24° for T1–12 (Edmondston et al., 2011; Mannion et al.,

2004), which is much smaller than 70° determined by Troke et al. (1998) using the CA 6000

system. Furness et al. (2016) reported a 78° RoFE using an inclinometer at T1/2 during sitting.

Morita et al. (2014) found a thoracic RoFE of 32° by CT over T1–L1 during supine lying.

3.3.4. Range of Lateral Bending (RoLB)

The thoracic RoLB was determined in one study using a radiological device, and in eight

studies with non-radiological devices (Table 5).

The thoracic RoLB was symmetric between both sides (Fig.2c). The whole thoracic (C7 or T1

to T12) RoLB to one side was 26° (Fig.2c) during standing when measured using non-

radiological devices (Hsu et al., 2008; Narimani and Arjmand, 2018). Over T3–12, it was 19°

during standing when using an optoelectronic system (Schinkel-Ivy et al., 2014). Troke et al.

(1998) reported a two-sided thoracic RoLB of 75° for T1–12 during standing using an

electronic goniometer system. During supine lying, it was 16° to one side by CT over T1–L1

(Fujimori et al., 2014).

10
The lower thoracic spine (T8–12) contributed more to the RoLB than the upper (T1–4) and

middle (T4–8) levels both during standing (48% vs. 23% and 29%) (Alqhtani et al., 2015) and

sitting (47% vs. 23% and 30%) (Willems et al., 1996). Similar results were reported by

Schinkel-Ivy et al. (2014) (T3–6: 25%, T6–9: 26%,T9–12: 49%) and Narimani and Arjmand

(2018) (T1–5: 25%, T5–12: 75%) during standing.

3.3.5. Range of Axial Rotation (RoAR)

The thoracic RoAR was determined in one study using a radiological device and in thirteen

studies with non-radiological devices (Table 6).

The thoracic RoAR was symmetric between both sides (Fig.2d). During standing, it was 40°

to one side of the whole thoracic spine (C7 or T1 to T12) when determined using non-

radiological devices (Hsu et al., 2008; Narimani and Arjmand, 2018). Over T3–12, it was 35°

to one side using an optoelectronic system (Schinkel-Ivy et al., 2014) and 64° to both sides for

T1–12 using the CA 6000 system (Troke et al., 1998). During sitting, it was 45° to one-side

for T1–12 measured using an inclinometer (Johnson et al., 2012) and 40° over T2–10 using an

optical system (Edmondston et al., 2007) During supine lying, it was 25° to one side when

assessed by CT for T1–L1 (Fujimori et al., 2012). With a lumbar locked position, the thoracic

RoAR to one side was 40° using an inclinometer at the T1/2 point (Furness et al., 2016).

In standing, the middle level (T4–8) contributed the most to the RoAR, and the upper level

(T1–4) rotated to the contralateral side compared to the middle (T4–8) and lower (T8–12)

levels (Alqhtani et al., 2015). The T6–9 level also contributed more than the T3–6 and T6–12

levels but all rotated in the ipsilateral direction (Schinkel-Ivy et al., 2014). During sitting, the

middle level (T4–8, 51%) also contributed the most to the RoAR, followed by the upper level

(T1–4, 29%) and finally the lower level (T8–12, 20%), and all three levels rotated to the

ipsilateral side (Willems et al., 1996). Narimani and Arjmand (2018) reported a 64%

11
contribution of the lower level (T5–12) to the whole RoAR and of 36% by the upper level

(T1–5).

3.3.6. Segmental RoM

The segmental RoM was determined in three studies by CT (Fig.3). The segmental RoFE

tended to decrease from T1–2 to T4–5 and increase from T4–5 to T12–L1 (Morita et al.,

2014). For the segmental RoLB, Fujimori et al. (2014) concluded that 23% occurred between

T1–2 and T3–4, 16% between T4–5 and T6–7, 25% between T7–8 and T9–10, and 35%

between T10–11 and T12–L1. The greatest RoAR was measured between T5–6 and T10–11

and the lowest between T11–12 and T12–L1 (Fujimori et al., 2012). All these three CT

studies were performed during supine lying and with small cohorts in Japan.

3.3.7. Coupled motion

Coupled motion was determined in two studies using radiological devices (Fujimori et al.,

2012; 2014) and in four studies with non-radiological devices (Edmondston et al., 2007;

Gregersen and Lucas, 1967; Narimani and Arjmand, 2018; Willems et al., 1996).

During primary movement in the sagittal plane (RoFE), subjects mostly displayed slight axial

rotation or lateral bending (Narimani and Arjmand, 2018; Willems et al., 1996). Subjects

performed the largest coupled motion during lateral bending and axial rotation (Edmondston

et al., 2007; Fujimori et al., 2012; 2014; Narimani and Arjmand, 2018; Willems et al., 1996).

Fujimori et al. (2014) demonstrated that lateral bending during supine lying was associated

with an 6° of axial rotation to the ipsilateral side and by Narimani and Arjmand (2018) as 5°

during standing. Gregersen and Lucas (1967) similarly demonstrated that coupled axial

rotation occurred in the ipsilateral direction as lateral bending. Fujimori et al. (2012) further

found that during axial rotation in the supine lying position, approximately 8° coupled lateral

12
bending was observed in the ipsilateral direction, and 11° in standing (Narimani and Arjmand,

2018).

Different thoracic levels displayed different coupled motion patterns. During sitting, lateral

bending was accompanied by the greatest axial rotation in the middle level (T4–8), followed

by the lower (T8–12) and upper (T1–4) levels (Willems et al., 1996). When axial rotation was

the primary motion, coupled lateral bending was greatest in the middle (T4–8) and lower (T8–

12) levels and smallest in the upper level (T1–4) (Willems et al., 1996). Narimani and

Arjmand (2018) demonstrated that in standing both during lateral bending and axial rotation,

greater coupled RoM occurred in the lower thoracic spine (T5–12) and that the direction of

coupled motion was inconsistent among subjects.

3.4. Influencing factors

3.4.1. Age

Age is one of the most important factors that influences thoracic geometry and kinematics.

Kuo et al. (2009) reported that older subjects (60–83 years old) presented with greater

thoracic kyphosis than younger subjects (17–27 years old) in both standing and sitting

positions. Lorbergs et al. (2017) reported that subjects ≥ 65 years displayed a larger thoracic

kyphosis than subjects < 65 years. As presented in Fig.4a, with a one-decade age increase,

thoracic kyphosis significantly increased by about 3° both in radiological and non-

radiological studies (Fon et al., 1980; Iyer et al., 2016; O'Gorman and Jull, 1987). Kyphosis of

the upper thoracic level (T2–5) displayed no difference with age, in contrast, the lower level

(T5–12) showed an increasing tendency (Iyer et al., 2016). Morita et al. (2014) demonstrated

that subjects ≥ 55 years displayed a larger segmental kyphosis in the lower region (below

T10/11) than subjects <55 years.

13
O'Gorman and Jull (1987) reported that the thoracic RoM decreased about 5° per decade in all

three planes during sitting (Fig.4b). Lee et al. (2013) and Ignasiak et al. (2017) also

demonstrated a decreased thoracic RoF as well as RoLB in elderly subjects compared to

young subjects. Morita et al. (2014) and Ignasiak et al. (2017) both found a significant effect

of age on the sagittal RoM in the lower level (T6–12), but not in the upper level (T1–6).

3.4.2. Sex

There is considerable debate over the influence of sex on geometry and kinematics. While

Edmondston et al. (2011) reported no difference in thoracic kyphosis between sexes, Lorbergs

et al. (2017) demonstrated that females displayed a slightly greater thoracic kyphosis than

males. Fon et al. (1980) found no difference in kyphosis between males and females under the

age of 40, however, when aged ≥ 40 years, females displayed a greater kyphosis than males.

Morita et al. (2014) demonstrated that females displayed greater segmental kyphosis in the

lower region than males.

Regarding the effect of sex on the thoracic RoM, Willems et al. (1996) found only non-

significant differences in all three anatomical planes between males and females. By contrast,

Edmondston et al. (2011) reported that males displayed a significantly greater RoE both

during standing and sitting than females. The segmental RoFE were larger in females than in

males from T1–2 to T12–L1 excluding T4–5 and T8–9 (Morita et al., 2014).

3.4.3. Posture

Different postures can induce differences both in thoracic geometry and kinematics. Salem et

al. (2015) showed that thoracic kyphosis was greater during neutral standing than during

prone lying. Kuo et al. (2009) demonstrated that kyphosis during neutral standing was

relatively greater than during neutral sitting. Edmondston et al. (2011) found a significant

interaction of the thoracic RoE among different postures: 4-point kneeling position induced
14
the greatest RoE, followed by the prone lying position and the sitting position. The standing

position induced the least RoE. Edmondston et al. (2007) also reported that during a neutral

sitting posture, the thoracic RoAR was less than during a flexed sitting position. They also

found that during axial rotation, different sitting postures led to differences in coupled lateral

bending: a flexed sitting posture led to the greatest coupled lateral bending, followed by an

extended sitting posture and finally a neutral sitting posture.

3.4.4. Other factors

Obesity can influence the thoracic RoM. Gilleard and Smith (2007) demonstrated that the

thoracic RoF was significantly decreased in the obese group both during sitting and standing.

Schinkel-Ivy and Drake (2016) reported that an increasing breast size was associated with a

greater thoracic RoF.

15
4. Discussion

This systematic review summarizes the current state of knowledge on the shape and

kinematics of the thoracic spine in asymptomatic adults, with special focus on individuals’

characteristics (e.g., age and sex), initial measurement postures (e.g., standing and sitting) and

measurement devices (e.g., radiologic and non-radiologic). Forty-five articles published from

1967 to April 2018 were included.

We identified 11 studies that used radiological devices, 31 studies that used non-radiological

devices and three studies that used both. During standing, a mean thoracic kyphosis (T1–12)

of 40° and 34° was measured using radiological and non-radiological devices, respectively.

This difference might be caused by the artefacts of the skin and sub-cutaneous tissue. The

majority of kinematical measurements are based on non-radiological devices. During standing,

the thoracic RoM was greatest during axial rotation, approximately 40°, followed by lateral

bending with approximately 26° and flexion with approximately 21°. The smallest RoM, with

approximately 13°, was identified during extension. These data can be used as physiological

reference values in the clinical setting for the diagnosis of thoracic disorders, for optimal

restoration of pathological thoracic spinal structures and for the evaluation of treatment

success.

The contributions of different spinal levels to the thoracic RoM appear to be dependent on the

performed choreography (load-direction dependent). While the lower level (T8–12) behaves

as the lumbar spine and exhibits a greater RoF and RoLB than the upper (T1–4) and middle

(T4–8) levels, the latter two levels behave as the cervical spine and exhibit a greater RoE and

RoAR than the lower level (White and Panjabi, 1990). These results allow the thoracic levels

with altered mobility to be specifically identified and can be used for targeted treatment,

because changes in the relative contribution of each level to mobility are likely to alter the

16
movement and loading behavior of the thoracic spine (Alqhtani et al., 2015; Narimani and

Arjmand, 2018; Willems et al., 1996).

During the primary motion towards one anatomical plane, subjects performed evident coupled

motion in the other two planes, which was most obvious during lateral bending and axial

rotation. Exact knowledge about the coupled motion of the thoracic spine can help to diagnose

thoracic pathologies, for example, a more obvious coupled behavior during lateral bending

and axial rotation existing in scoliosis patients (Pope et al., 1984; White and Panjabi, 1990).

However, a large variability was found both in the direction and level of coupled motion

among studies, which may be a result of subtle anatomical variations (e.g., asymmetries in

soft-tissue tension), different initial postures (e.g., standing, sitting and lying) or slightly

different measurement orientations (Fujimori et al., 2012; Narimani and Arjmand, 2018;

Willems et al., 1996). Therefore, future study is required to confirm the variability sources

and to compare the difference of coupled motion behaviors between the physiological and

different pathological thoracic spines.

With aging, the thoracic kyphosis increased by approximately 3° per decade, whereas the

RoM decreased by approximately 5° per decade. These results emphasize that hyper-kyphosis

and hypo-mobility of the thoracic spine in symptomatic subjects with thoracic diseases partly

consists of a natural adaptive process during aging, which also occurs in asymptomatic

subjects. The changes in thoracic shape and kinematics with aging mainly occurred in the

lower level (T6–12), but not in the upper level (T1–6). These facts could help to understand

the mechanical challenges that the lower thoracic spine has to withstand, which may have

important implications for the relatively higher fracture rate in the lower thoracic spine during

injury (Bakhsheshian et al., 2014; Tong et al., 2016). Because the shape and kinematics

change differently for certain levels of the thoracic spine with aging, an age- and thoracic

level-specific treatment may be important for long-term patient satisfaction.

17
The influence of sex on thoracic kyphosis and the RoM was inconclusive among the studies.

Because Fon et al. (1980) suggested that the difference in kyphosis between males and

females was small and varied between different age groups, this might partly explain why

studies with heterogeneous cohort sizes and different mean ages produced contradictory

results. Obesity decreased the thoracic RoM, which might be caused by the physical

obstruction of soft tissue (Gilleard and Smith, 2007). Thoracic kyphosis and the RoM are

significantly influenced by the initial body posture. This is quite important in the clinical

setting, because radiological examinations are normally performed in a standing posture,

which would display a greater kyphosis than a prone lying posture during surgery (Salem et

al., 2015). This difference between postures should be considered for optimal correction of

thoracic deformity. It is also important to note that for functional examination of the thoracic

spine, results can only be compared when the subjects are measured in the same posture

(Edmondston et al., 2011).

In addition to its clinical importance, results from the current study are also essential for in

vitro studies in defining realistic boundary conditions. In their cadaveric thoracic spinal

models with intact rib cage, Mannen et al. (2015) first applied a ±5 Nm pure moment in three

anatomical directions on T1 and reported a flexion of 7.7°, extension of 9.6°, lateral bending

of 23.3° and axial rotation of 26.3°, which were relatively smaller than the results reported in

the current study. Subsequently, a 400 N compressive follower load together with a ±5 Nm

dynamic moment on T1 was applied and they found the least mobility in the middle thoracic

level (T4-8) in the sagittal and coronal planes (Anderson et al., 2018; Mannen et al., 2018; Sis

et al., 2016), which was also different from the results in the current study that the middle

level contributed the most to the RoE and RoAR. This might be due to the much more

complicated in vivo conditions compared to the in vitro conditions, which includes the skin

artefacts, the rib cage, the muscles, the intra-abdominal and intra-thoracic pressure and the

18
interaction with the cervical and lumbar regions. All these characteristics result in the loads

under in vivo conditions being quite different from those under the in vitro conditions, and

thus difficult to replicate.

Several limitations should be noted in this study. Inevitably, we cannot confirm a complete

literature search nor the avoidance of publication bias, particularly for thoracic kyphosis.

Most of the studies included were performed with a small sample size. Furthermore, because

of the large heterogeneity among studies, a meta-analysis could not be performed, in contrast

to our recent review and meta-analysis on the cervical RoM (Pan et al., 2018). The

heterogeneity is reflected among others in a lack of a clear consensus on the measurement

devices and measured levels or postures.

From such limitations, some considerations for the design of future studies can be derived.

Further multi-center work with a larger sample size needs to be conducted. A homogeneous

protocol still needs to be created to make the results more comparable. To investigate the

effect of age, a longitudinal study within the same cohort with a long-term follow-up is

considered to be more powerful (Adams et al., 1999). While we only included studies

assessing static postures or the RoM, future work should also aim to evaluate thoracic

kinematics during functional tasks (e.g., walking, respiration, sit-to-stand and lifting) or in

daily life under a more natural environment as performed in our previous study for the lumbar

shape and kinematics (Dreischarf et al., 2016), which might provide a closer reference to the

thoracic pathological status.

In conclusion, this study systematically reviewed the literature on thoracic shape and

kinematics in asymptomatic adults and its moderating parameters. The results can have

implications in the clinical setting and in the biomechanical field. However, large

19
heterogeneity of the measurement methodologies existed among studies and a homogeneous

protocol (e.g., measurement device, segment and posture) is preferable in the future.

Conflict of interest

The authors declare that they have no conflict of interest.

Acknowledgments

This study was financed by the Federal Ministry of Education and Research (BMBF), Bonn,

Germany (MEDITHENA, https://www.technik-zum-menschen-bringen.de/projekte/medithen

a ) , a n d b y t h e C h i n a S c ho l a r s h i p C o u n c i l ( C S C , N o . 2 0 1 7 0 8 0 8 0 0 9 0 ) .

20
References

Aboutorabi, A., Arazpour, M., Ahmadi Bani, M., Keshtkar, A.A., 2018. Effect of spinal
orthoses and postural taping on balance, gait and quality of life in older people with
thoracic hyperkyphosis: protocol for a systematic review and meta-analysis. BMJ open 8,
e015813.
Adams, M.A., Mannion, A.F., Dolan, P., 1999. Personal risk factors for first-time low back
pain. Spine 24, 2497-2505.
Aebi, M., 2003. Spinal metastasis in the elderly. European spine journal : official publication
of the European Spine Society, the European Spinal Deformity Society, and the European
Section of the Cervical Spine Research Society 12 Suppl 2, S202-213.
Alderighi, M., Ferrari, R., Maghini, I., Del Felice, A., Masiero, S., 2016. Intra and interrater
reliability of spinal sagittal curves and mobility using pocket goniometer IncliMed(R) in
healthy subjects. Journal of back and musculoskeletal rehabilitation 29, 873-880.
Alqhtani, R.S., Jones, M.D., Theobald, P.S., Williams, J.M., 2015. Reliability of an
accelerometer-based system for quantifying multiregional spinal range of motion. Journal
of manipulative and physiological therapeutics 38, 275-281.
Anderson, D.E., Mannen, E.M., Tromp, R., Wong, B.M., Sis, H.L., Cadel, E.S., Friis, E.A.,
Bouxsein, M.L., 2018. The rib cage reduces intervertebral disc pressures in cadaveric
thoracic spines by sharing loading under applied dynamic moments. Journal of
biomechanics 70, 262-266.
Angelini, L., Damm, P., Zander, T., Arshad, R., Di Puccio, F., Schmidt, H., 2018. Effect of
arm swinging on lumbar spine and hip joint forces. Journal of biomechanics 70, 185-195.
Arjmand, N., Gagnon, D., Plamondon, A., Shirazi-Adl, A., Larivière, C., 2009. Comparison
of trunk muscle forces and spinal loads estimated by two biomechanical models. Clinical
Biomechanics 24, 533-541.
Arshad, R., Angelini, L., Zander, T., Di Puccio, F., El-Rich, M., Schmidt, H., 2018. Spinal
loads and trunk muscles forces during level walking - A combined in vivo and in silico
study on six subjects. Journal of biomechanics 70, 113-123.
Bakhsheshian, J., Dahdaleh, N.S., Fakurnejad, S., Scheer, J.K., Smith, Z.A., 2014. Evidence-
based management of traumatic thoracolumbar burst fractures: a systematic review of
nonoperative management. Neurosurgical focus 37, E1.
Brasiliense, L.B., Lazaro, B.C., Reyes, P.M., Dogan, S., Theodore, N., Crawford, N.R., 2011.
Biomechanical contribution of the rib cage to thoracic stability. Spine 36, E1686-1693.
Brigatto, P., Carbinatto, J.C., Costa, C.M., Montebelo, M.I., Rasera-Junior, I., Pazzianotto-
Forti, E.M., 2014. Application of positive airway pressure in restoring pulmonary function
and thoracic mobility in the postoperative period of bariatric surgery: a randomized clinical
trial. Brazilian journal of physical therapy 18, 553-562.

21
Briggs, A.M., Bragge, P., Smith, A.J., Govil, D., Straker, L.M., 2009. Prevalence and
associated factors for thoracic spine pain in the adult working population: a literature
review. Journal of occupational health 51, 177-192.
Brismee, J.M., Gipson, D., Ivie, D., Lopez, A., Moore, M., Matthijs, O., Phelps, V., Sawyer,
S., Sizer, P., 2006. Interrater reliability of a passive physiological intervertebral motion test
in the mid-thoracic spine. Journal of manipulative and physiological therapeutics 29, 368-
373.
Cloud, B.A., Zhao, K.D., Breighner, R., Giambini, H., An, K.N., 2014. Agreement between
fiber optic and optoelectronic systems for quantifying sagittal plane spinal curvature in
sitting. Gait & posture 40, 369-374.
Cummings, S.R., Melton, L.J., 2002. Epidemiology and outcomes of osteoporotic fractures.
Lancet (London, England) 359, 1761-1767.
Dreischarf, M., Pries, E., Bashkuev, M., Putzier, M., Schmidt, H., 2016. Differences between
clinical "snap-shot" and "real-life" assessments of lumbar spine alignment and motion -
What is the "real" lumbar lordosis of a human being? Journal of biomechanics 49, 638-644.
Edmondston, S.J., Aggerholm, M., Elfving, S., Flores, N., Ng, C., Smith, R., Netto, K., 2007.
Influence of posture on the range of axial rotation and coupled lateral flexion of the
thoracic spine. Journal of manipulative and physiological therapeutics 30, 193-199.
Edmondston, S.J., Christensen, M.M., Keller, S., Steigen, L.B., Barclay, L., 2012a. Functional
radiographic analysis of thoracic spine extension motion in asymptomatic men. Journal of
manipulative and physiological therapeutics 35, 203-208.
Edmondston, S.J., Ferguson, A., Ippersiel, P., Ronningen, L., Sodeland, S., Barclay, L., 2012b.
Clinical and radiological investigation of thoracic spine extension motion during bilateral
arm elevation. The Journal of orthopaedic and sports physical therapy 42, 861-869.
Edmondston, S.J., Waller, R., Vallin, P., Holthe, A., Noebauer, A., King, E., 2011. Thoracic
spine extension mobility in young adults: influence of subject position and spinal curvature.
The Journal of orthopaedic and sports physical therapy 41, 266-273.
Fleming, L.A., 1992. Osteoporosis: clinical features, prevention, and treatment. Journal of
general internal medicine 7, 554-562.
Folsch, C., Schlogel, S., Lakemeier, S., Wolf, U., Timmesfeld, N., Skwara, A., 2012. Test-
retest reliability of 3D ultrasound measurements of the thoracic spine. PM & R : the
journal of injury, function, and rehabilitation 4, 335-341.
Fon, G.T., Pitt, M.J., Thies, A.C., Jr., 1980. Thoracic kyphosis: range in normal subjects. AJR.
American journal of roentgenology 134, 979-983.
Fouquet, N., Bodin, J., Descatha, A., Petit, A., Ramond, A., Ha, C., Roquelaure, Y., 2015.
Prevalence of thoracic spine pain in a surveillance network. Occupational medicine
(Oxford, England) 65, 122-125.
Fujimori, T., Iwasaki, M., Nagamoto, Y., Ishii, T., Kashii, M., Murase, T., Sugiura, T.,
Matsuo, Y., Sugamoto, K., Yoshikawa, H., 2012. Kinematics of the thoracic spine in trunk
rotation: in vivo 3-dimensional analysis. Spine 37, E1318-1328.
22
Fujimori, T., Iwasaki, M., Nagamoto, Y., Matsuo, Y., Ishii, T., Sugiura, T., Kashii, M.,
Murase, T., Sugamoto, K., Yoshikawa, H., 2014. Kinematics of the thoracic spine in trunk
lateral bending: in vivo three-dimensional analysis. The spine journal : official journal of
the North American Spine Society 14, 1991-1999.
Furness, J., Climstein, M., Sheppard, J.M., Abbott, A., Hing, W., 2016. Clinical methods to
quantify trunk mobility in an elite male surfing population. Physical therapy in sport :
official journal of the Association of Chartered Physiotherapists in Sports Medicine 19, 28-
35.
Gilleard, W., Smith, T., 2007. Effect of obesity on posture and hip joint moments during a
standing task, and trunk forward flexion motion. International journal of obesity (2005) 31,
267-271.
Granata, K.P., Wilson, S.E., 2001. Trunk posture and spinal stability. Clinical Biomechanics
16, 650-659.
Gregersen, G.G., Lucas, D.B., 1967. An in vivo study of the axial rotation of the human
thoracolumbar spine. The Journal of bone and joint surgery. American volume 49, 247-262.
Hajibozorgi, M., Arjmand, N., 2016. Sagittal range of motion of the thoracic spine using
inertial tracking device and effect of measurement errors on model predictions. Journal of
biomechanics 49, 913-918.
Hasegawa, K., Okamoto, M., Hatsushikano, S., Shimoda, H., Ono, M., Watanabe, K., 2016.
Normative values of spino-pelvic sagittal alignment, balance, age, and health-related
quality of life in a cohort of healthy adult subjects. European spine journal : official
publication of the European Spine Society, the European Spinal Deformity Society, and the
European Section of the Cervical Spine Research Society 25, 3675-3686.
Horton, W.C., Brown, C.W., Bridwell, K.H., Glassman, S.D., Suk, S.I., Cha, C.W., 2005. Is
there an optimal patient stance for obtaining a lateral 36" radiograph? A critical
comparison of three techniques. Spine 30, 427-433.
Hsu, C.J., Chang, Y.W., Chou, W.Y., Chiou, C.P., Chang, W.N., Wong, C.Y., 2008.
Measurement of spinal range of motion in healthy individuals using an electromagnetic
tracking device. Journal of neurosurgery. Spine 8, 135-142.
Ignasiak, D., Ferguson, S.J., Arjmand, N., 2016. A rigid thorax assumption affects model
loading predictions at the upper but not lower lumbar levels. Journal of biomechanics 49,
3074-3078.
Ignasiak, D., Rüeger, A., Ferguson, S.J., 2017. Multi-segmental thoracic spine kinematics
measured dynamically in the young and elderly during flexion. Human movement science
54, 230-239.
Intolo, P., Milosavljevic, S., Baxter, D.G., Carman, A.B., Pal, P., Munn, J., 2009. The effect
of age on lumbar range of motion: a systematic review. Manual therapy 14, 596-604.
Iyer, S., Lenke, L.G., Nemani, V.M., Albert, T.J., Sides, B.A., Metz, L.N., Cunningham, M.E.,
Kim, H.J., 2016. Variations in Sagittal Alignment Parameters Based on Age: A

23
 Prospective Study of Asymptomatic Volunteers Using Full-Body Radiographs. Spine 41,
1826-1836.
Jang, H.J., Kim, M.J., Kim, S.Y., 2015. Effect of thorax correction exercises on flexed posture
and chest function in older women with age-related hyperkyphosis. J Phys Ther Sci 27,
1161-1164.
Johnson, K.D., Kim, K.M., Yu, B.K., Saliba, S.A., Grindstaff, T.L., 2012. Reliability of
thoracic spine rotation range-of-motion measurements in healthy adults. Journal of athletic
training 47, 52-60.
Kado, D.M., Huang, M.H., Karlamangla, A.S., Barrett-Connor, E., Greendale, G.A., 2004.
Hyperkyphotic posture predicts mortality in older community-dwelling men and women: a
prospective study. Journal of the American Geriatrics Society 52, 1662-1667.
Katzman, W.B., Harrison, S.L., Fink, H.A., Marshall, L.M., Orwoll, E., Barrett-Connor, E.,
Cawthon, P.M., Kado, D.M., 2015. Physical function in older men with hyperkyphosis.
The journals of gerontology. Series A, Biological sciences and medical sciences 70, 635-
640.
Kuo, Y.L., Tully, E.A., Galea, M.P., 2009. Video analysis of sagittal spinal posture in healthy
young and older adults. Journal of manipulative and physiological therapeutics 32, 210-
215.
Lee, B.A., Leiby, B.E., Marino, R.J., 2016. Neurological and functional recovery after
thoracic spinal cord injury. The journal of spinal cord medicine 39, 67-76.
Lee, N.-G., Lee, J.-A., Kim, J.-B., 2013. A Comparison of Three-dimensional Spine
Kinematics during Multidirectional Trunk Movement between Elderly Subjects with
Degenerative Spine Disease and Healthy Young Adults. Journal of Physical Therapy
Science 25, 21-26.
Lorbergs, A.L., Murabito, J.M., Jarraya, M., Guermazi, A., Allaire, B.T., Yang, L., Kiel, D.P.,
Cupples, L.A., Bouxsein, M.L., Travison, T.G., Samelson, E.J., 2017. Thoracic Kyphosis
and Physical Function: The Framingham Study. Journal of the American Geriatrics Society
65, 2257-2264.
Madinei, S.S., Arjmand, N., 2018. Sagittal Range of Motion of the Thoracic Spine Using
Standing Digital Radiography; A throughout Comparison with Non-Radiographic Data
Reviewed from the Literature. Scientia Iranica. 10.24200/sci.2018.20503
Magerl, F., Aebi, M., Gertzbein, S.D., Harms, J., Nazarian, S., 1994. A comprehensive
classification of thoracic and lumbar injuries. European spine journal : official publication
of the European Spine Society, the European Spinal Deformity Society, and the European
Section of the Cervical Spine Research Society 3, 184-201.
Mannen, E.M., Anderson, J.T., Arnold, P.M., Friis, E.A., 2015. Mechanical analysis of the
human cadaveric thoracic spine with intact rib cage. Journal of biomechanics 48, 2060-
2066.
Mannen, E.M., Arnold, P.M., Anderson, J.T., Friis, E.A., 2017. Influence of Sequential Ponte
Osteotomies on the Human Thoracic Spine With a Rib Cage. Spine deformity 5, 91-96.
24
Mannen, E.M., Friis, E.A., Sis, H.L., Wong, B.M., Cadel, E.S., Anderson, D.E., 2018. The rib
cage stiffens the thoracic spine in a cadaveric model with body weight load under dynamic
moments. Journal of the mechanical behavior of biomedical materials 84, 258-264.
Mannion, A.F., Knecht, K., Balaban, G., Dvorak, J., Grob, D., 2004. A new skin-surface
device for measuring the curvature and global and segmental ranges of motion of the spine:
reliability of measurements and comparison with data reviewed from the literature.
European spine journal : official publication of the European Spine Society, the European
Spinal Deformity Society, and the European Section of the Cervical Spine Research
Society 13, 122-136.
Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., 2010. Preferred reporting items for
systematic reviews and meta-analyses: the PRISMA statement. International journal of
surgery (London, England) 8, 336-341.
Morita, D., Yukawa, Y., Nakashima, H., Ito, K., Yoshida, G., Machino, M., Kanbara, S.,
Iwase, T., Kato, F., 2014. Range of motion of thoracic spine in sagittal plane. European
spine journal : official publication of the European Spine Society, the European Spinal
Deformity Society, and the European Section of the Cervical Spine Research Society 23,
673-678.
Narimani, M., Arjmand, N., 2018. Three-dimensional primary and coupled range of motions
and movement coordination of the pelvis, lumbar and thoracic spine in standing posture
using inertial tracking device. Journal of biomechanics 69, 169-174.
O'Gorman, H., Jull, G., 1987. Thoracic kyphosis and mobility: the effect of age.
Physiotherapy Practice 3, 154-162.
Pan, F., Arshad, R., Zander, T., Reitmaier, S., Schroll, A., Schmidt, H., 2018. The effect of
age and sex on the cervical range of motion - A systematic review and meta-analysis.
Journal of biomechanics 75, 13-27.
Park, K.H., Oh, J.S., An, D.H., Yoo, W.G., Kim, J.M., Kim, T.H., Kang, M.H., 2015.
Difference in selective muscle activity of thoracic erector spinae during prone trunk
extension exercise in subjects with slouched thoracic posture. PM & R : the journal of
injury, function, and rehabilitation 7, 479-484.
Pope, M.H., Stokes, I.A., Moreland, M., 1984. The biomechanics of scoliosis. Critical reviews
in biomedical engineering 11, 157-188.
Salem, W., Coomans, Y., Brismee, J.M., Klein, P., Sobczak, S., Dugailly, P.M., 2015. Sagittal
Thoracic and Lumbar Spine Profiles in Upright Standing and Lying Prone Positions
Among Healthy Subjects: Influence of Various Biometric Features. Spine 40, E900-908.
Schinkel-Ivy, A., Drake, J.D., 2016. Breast size impacts spine motion and postural muscle
activation. Journal of back and musculoskeletal rehabilitation 29, 741-748.
Schinkel-Ivy, A., Pardisnia, S., Drake, J.D., 2014. Head and arm positions that elicit maximal
voluntary trunk range-of-motion measures. Journal of applied biomechanics 30, 689-696.
Schroeder, J., Reer, R., Braumann, K.M., 2015. Video raster stereography back shape
reconstruction: a reliability study for sagittal, frontal, and transversal plane parameters.
25
 European spine journal : official publication of the European Spine Society, the European
Spinal Deformity Society, and the European Section of the Cervical Spine Research
Society 24, 262-269.
Sis, H.L., Mannen, E.M., Wong, B.M., Cadel, E.S., Bouxsein, M.L., Anderson, D.E., Friis,
E.A., 2016. Effect of follower load on motion and stiffness of the human thoracic spine
with intact rib cage. Journal of biomechanics 49, 3252-3259.
Tong, J., Zhou, R., Liu, S., Zhu, Q., 2016. Management of fracture and lateral dislocation of
the thoracic spine without any neurological deficits: three case reports and review of the
literature. Irish journal of medical science 185, 949-954.
Troke, M., Moore, A.P., Cheek, E., 1998. Reliability of the OSI CA 6000 Spine Motion
Analyzer with a new skin fixation system when used on the thoracic spine. Manual therapy
3, 27-33.
Vandenbroucke, J.P., von Elm, E., Altman, D.G., Gotzsche, P.C., Mulrow, C.D., Pocock, S.J.,
Poole, C., Schlesselman, J.J., Egger, M., 2007. Strengthening the Reporting of
Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS
medicine 4, e297.
Vialle, R., Levassor, N., Rillardon, L., Templier, A., Skalli, W., Guigui, P., 2005.
Radiographic analysis of the sagittal alignment and balance of the spine in asymptomatic
subjects. The Journal of bone and joint surgery. American volume 87, 260-267.
Volkheimer, D., Malakoutian, M., Oxland, T.R., Wilke, H.J., 2015. Limitations of current in
vitro test protocols for investigation of instrumented adjacent segment biomechanics:
critical analysis of the literature. European spine journal : official publication of the
European Spine Society, the European Spinal Deformity Society, and the European
Section of the Cervical Spine Research Society 24, 1882-1892.
Weitkunat, T., Buck, F.M., Jentzsch, T., Simmen, H.P., Werner, C.M., Osterhoff, G., 2016.
Influence of high-heeled shoes on the sagittal balance of the spine and the whole body.
European spine journal : official publication of the European Spine Society, the European
Spinal Deformity Society, and the European Section of the Cervical Spine Research
Society 25, 3658-3665.
White, A.A., 3rd, Panjabi, M.M., 1990. Clinical biomechanics of the spine (2nd ed.).
Lippincott, Philadelphia.
Wilke, H.J., Herkommer, A., Werner, K., Liebsch, C., 2017. In vitro analysis of the segmental
flexibility of the thoracic spine. PloS one 12, e0177823.
Willems, J.M., Jull, G.A., J, K.F., 1996. An in vivo study of the primary and coupled rotations
of the thoracic spine. Clinical biomechanics (Bristol, Avon) 11, 311-316.
Wood, K.B., Garvey, T.A., Gundry, C., Heithoff, K.B., 1995. Magnetic resonance imaging of
the thoracic spine. Evaluation of asymptomatic individuals. The Journal of bone and joint
surgery. American volume 77, 1631-1638.

26
Zemp, R., List, R., Gulay, T., Elsig, J.P., Naxera, J., Taylor, W.R., Lorenzetti, S., 2014. Soft
tissue artefacts of the human back: comparison of the sagittal curvature of the spine
measured using skin markers and an open upright MRI. PloS one 9, e95426.
Zemp, R., Taylor, W.R., Lorenzetti, S., 2013. In vivo spinal posture during upright and
reclined sitting in an office chair. BioMed research international 2013, 916045.

27
Figure and Table Legends

Figure 1 Mean ± standard deviation of thoracic kyphosis measured over T1–12. The bars are

organized first by the used measurement device (radiological and non-radiological), second

by posture (standing and sitting). Therein, the bars are ordered by published years from the

latest to the oldest. n: number of subjects; ♂: males; ♀: females.

Figure 2 Mean ± standard deviation of thoracic range of motion (RoM). a. Thoracic range of

flexion (RoF), b. Thoracic range of extension (RoE), c. Thoracic range of lateral bending

(RoLB), d. Thoracic range of axial rotation (RoAR). The bars are organized first by thoracic

spinal region (whole, upper, middle and lower), second by the used measurement device

(radiological and non-radiological), third by posture (standing, sitting, prone and supine).

Therein, the bars are ordered by published years from the latest to the oldest. n: number of

subjects; ♂: males; ♀: females.

Figure 3 Mean ± standard deviation of segmental range of motion (RoM) determined by CT

during supine lying. The bars are organized by RoM in different directions. n: number of

subjects; ♂: males.

Figure 4 a. The effect of age on thoracic kyphosis (mean ± standard deviation), the bars are

organized first by thoracic spinal region (whole, upper, middle and lower), second by the used

measurement device (radiological and non-radiological), third by posture (standing and

sitting). b. The effect of age on the thoracic range of motion (RoM, mean ± standard

deviation), the bars are organized by thoracic RoM in different directions. n: number of

subjects; ♂: males; ♀: females.

Table 1 Mean ± standard deviation of thoracic kyphosis.

Table 2 Mean ± standard deviation of thoracic range of flexion (RoF).

28
Table 3 Mean ± standard deviation of thoracic range of extension (RoE).

Table 4 Mean ± standard deviation of thoracic range of flexion plus extension (RoFE).

Table 5 Mean ± standard deviation of thoracic range of lateral bending (RoLB).

Table 6 Mean ± standard deviation of thoracic range of axial rotation (RoAR).

Appendix

Appendix 1 The entire body of included studies with links to abstracts.

Appendix 2 Quality assessment results of included studies.

29
30
31
32
Table 1
Thoracic kyphosis.
Age Weight Height BMI PubMed
Posture Author(Year) Country Numb, Sex (ys) (kg) (cm) (kg/m2) Device Level Kyphosis (°) Reliability
Link

Radiological studies
Madinei and Arjmand
Iran 8m 28.5 ± 5.7 78.2 ± 12.9 176.7 ± 9.1 - Radiographs T1-12 38.7 ± 6.1 Y >>>
(2018)

Weitkunat et al. (2016) Switzerland 23 f 29.1 ± 6.3 65.5 ± 11.9 170 ± 10 23.9 ± 4.0 EOS T5-12 32.9 ±7.9 - >>>

15 f 5 m 21 - 30 77.3 ± 23.6 170.0 ± 7.1 26.5 ± 6.6 T2-12: 41.2 ± 13.3; 43.2 ± 11.4; 44.5 ± 13.2;
12 f 6 m 31 - 40 78.9 ± 20.4 167.1 ± 8.4 28.2 ± 7.0 50.9 ± 11.2; 47.7 ± 13.6; 49.8 ± 16.1;
T2-12
14 f 3 m 41 - 50 88.3 ± 20.5 166.4 ± 8.2 31.4 ± 7.8 T2-5: 15.1 ± 6.9; 16.3 ± 4.2; 14.7 ± 6.4; 17.2
Iyer et al. (2016) USA EOS T2-5 Y >>>
1 3f 3 m 51 - 60 77.2 ± 15.4 166.1 ± 8.5 28.2 ± 6.2 ± 6.3; 15.6 ± 8; 13.8 ± 7.6;
T5-12
16 f 11 m 61 - 70 82.1 ± 16.0 169.8 ± 11.0 27.5 ± 2.7 T5-12: 28.9 ± 11; 31.2 ± 9.3; 32.7 ± 12.8;
8f9m >70 77.1 ± 16.9 168.2 ± 9.4 26.8 ± 4.0 38.3 ± 12.2; 36.4 ± 13.3; 38.3 ± 15.2
30 m T1-12 T1-12: 41.5 ± 9.9
Standing

Hasegawa et al. (2016) Japan 39.4 ± 11.3 - - 21.1 ± 2.4 EOS - >>>
96 f T4-12 T4-12: 29.6 ± 9.2
Edmondston et al.
Australia 21 m 22.6 ± 3.2 72.6 ± 8.1 177 ± 7.0 ≤25 Radiographs T3-11 31.5 ± 12.1 - >>>
(2012b)
Edmondston et al.
Australia 14 m 30.2 ± 7 74.7 175 - Radiographs T3-11 42.1 ± 13.4 Y >>>
(2012a)
110 f
Vialle et al. (2005) France 35.4 ± 12 - - 23.5 ± 3 Radiographs T4-12 40.6 ± 10 - >>>
190 m

37 m 24 f 20 - 29
26 m 26 f 30 - 39 m: 26.27 ± 8.12; 29.04 ± 7.93; 29.75 ± 6.93;
20 m 32 f 40 - 49 33 ± 6.46; 34.67 ± 5.12; 40.67 ± 7.57;
Fon et al. (1980) USA - - - Radiographs T2-12 - >>>
10 m 17 f 50 - 59 f: 26.83 ± 7.98; 28.42 ± 8.63; 32.66 ± 6.72;
9m7f 60 - 69 40.71 ± 9.88; 44.86 ± 7.80;41.67 ± 9.00
3m6f 70 - 79
Supine Sitting

3f 36.0 ± 12.4
Zemp et al. (2014) Switzerland 29 71 174 MRI T3-11 - >>>
4m 40.7 ± 10.0

604 f f <65 ys: 32.6 ± 8.7; ≥65 ys: 36.8 ± 10.9;

Lorbergs et al. (2017) USA 61 - - - CT T4-12 - >>>
496 m m<65 ys: 31.5 ± 10.9; ≥65 ys: 34.8 ± 9.1

Non-radiological studies
Schinkel-Ivy and Drake
Canada 15 f 22.8 ± 2.7 59.12 ± 6.38 166 ± 5 - Vicon C7-T12 43.1 ± 7.6 - >>>
(2016)
Boulet et al. (2016) France 21 m 25 ± 2 72 ± 11 176 ± 8 - Vicon T4-L1 30.9 ± 5.9 - >>>
Standing

Alderighi et al. (2016) Italy 34 f 19.17 ± 4.52 - 164 ± 8 21 ± 2.6 Goniometer T1-12 37.96 ± 7.17 - >>>
9f
Schroeder et al. (2015) Germany 25.4 ± 5.5 69.6 ± 12.8 174 ± 9.0 22.8 ± 2.7 Stereography T1-12 45.3 ± 9.4 Y >>>
11 m
38 m C7-changing
Salem et al. (2015) Belgium 22.6 ± 4.3 66 ± 10.9 172 ± 8 22.2 ± 2.5 3D digitizer 45.5 ± 10.3 Y >>>
37 f point
10 m
Park et al. (2015) South Korea 26.15 ± 5.34 61.92 ± 9.17 168.25 ± 7.46 - Inclinometer T1-12 30.05 ± 4.72 - >>>
12 f

33
 T3-12 31.7 ± 10.1
12 m T3-6 14.1 ± 5.6
Schinkel-Ivy et al. (2014) Canada 25.3 ± 3.9 79.97 ± 9.38 180 ± 5 - Vicon - >>>
12 f T6-9 14 ± 6.2
T9-12 3.7 ± 7.3
Edmondston et al.
Australia 21 m 22.6 ± 3.2 72.6 ± 8.1 177 ± 7.0 ≤25 Photographs T1-12 28.7 ± 7.6 - >>>
(2012b)
Edmondston et al.
Australia 14 m 30.2 ± 7 74.7 175 - Photographs T1-12 32.9 ± 10.6 Y >>>
(2012a)
20 m 24 ± 3.6
Edmondston et al. (2011) Australia - - - Photographs T1-12 21.6± 5.6 - >>>
20 f 21.6 ± 2.8

15 f 9 m 17 - 27 62.6 ± 8.9 170.2 ± 9.1 21.5 ± 1.9 38.3 ± 8.2

Kuo et al. (2009) Taiwan Video T1-L1 - >>>
14 f 8 m 60 - 83 69.3 ± 12.1 163.9 ± 8.4 25.8 ± 4.0 42.8 ± 11.3
9m 45.4 ± 7.7 87.6 ± 16.8 181 ± 9 26.5 ± 5.2
Mannion et al. (2004) Switzerland Spinal Mouse T1-12 44.7 ± 9.2 Y >>>
11 f 38.2 ± 7.6 63.8 ± 7.6 167 ± 10 22.9 ± 5.7

3f
Zemp et al. (2014) Switzerland 29 71 174 - Photographs T3-11 40.7 ± 10.0 - >>>
4m
13 m
Cloud et al. (2014) USA 28.2 ± 4.7 74.7 ± 12.6 172.4 ± 9.1 25.1 ± 3.5 Optoelectronic T1-12 39.37 ± 9.16 - >>>
13 f

14 f
Folsch et al. (2012) Germany 33 ± 14.8 73.9 ± 15.1 1.74 ± 0.11 - Zebris CMS 20 C7-T12 45.8 ± 16.2 Y >>>
14 m
Sitting

15 f 9 m 17 - 27 62.6 ± 8.9 170.2 ± 9.1 21.5 ± 1.9 33.3 ± 8.1

Kuo et al. (2009) Taiwan Video T1-L1 - >>>
14 f 8 m 60 - 83 69.3 ± 12.1 163.9 ± 8.4 25.8 ± 4.0 40.2 ± 11.2

20 f 22 - 29 40.68 ± 8.77
20 f 30 - 39 42.05 ± 5.94
O'Gorman and Jull 20 f 40 - 49 45.23 ± 7.66
Australia - - - Inclinometer T1-L1 Y >>>
(1987) 20 f 50 - 59 54.43 ± 9.61
20 f 60 - 69 55.33 ± 11.88
20 f >70 65.73 ± 10.30
Prone

38 m C7-changing
Salem et al. (2015) Belgium 22.6 ± 4.3 66 ± 10.9 172 ± 8 22.2 ± 2.5 3D digitizer 32.2 ± 8.2 Y >>>
37 f point

Table 2
Thoracic range of flexion (RoF).
Posture Age Weight Height BMI PubMed
Author (Year) Country Numb Sex Device Level RoF (°) Reliability
(ys) (kg) (cm) (kg/m2) Link

Posture Radiological studies

34
Sitting Standing
T1-12 22.5 ± 4.1
Madinei and Arjmand Iran Y
8m 28.5 ± 5.7 78.2 ± 12.9 176.7 ± 9.1 - Radiographs T1-6 7.1 ± 1.8 >>>
(2018)
T6-12 15.5 ± 3.1

Zemp et al. (2014) Switzerland 3f 29 71 174 - MRI T3-11 4.6 ± 2.6 -

>>>
4m

Non-radiological studies

T1-12 19.9 ± 6.4

Narimani and Arjmand
Iran 22 m 24.8 ± 1.0 71.5 ± 10.0 178.0 ± 5.4 22.5 ± 2.7 Inertial device T1-5 4.2 ± 3.4 - >>>
(2018)
T5-12 14.9 ± 5.2

C7-L1 20.3; 17.8

C7-T3 -11.8; -11.2
T3-5 -10.7; -8.7
8 m 13 f Young: 18 – 35 68.33 ± 13.73 173.29 ± 9.5 22.56 ± 2.90
Ignasiak et al. (2017) Switzerland Vicon T5-7 0.6; 3.1 - >>>
8 m 13 f Old: ≥65 67.48 ± 11.31 168.52 ± 8.85 23.72 ± 3.15
T7-9 9.5; 9.3
T9-11 13.5; 11.5
T11-L1 17.1; 9.8

Schinkel-Ivy and Drake

Canada 15 f 22.8 ± 2.7 59.12 ± 6.38 166 ± 5 - Vicon C7-T12 27.6 ± 13.1 - >>>
(2016)
T1-12 20.5 ± 6.5
Hajibozorgi and Arjmand
Iran 40 m 22.5 ± 1.8 75.5 ± 10.8 177.8 ± 7.3 23.9 ± 3.1 Inertial device T1-5 5.8 ± 3.1 - >>>
Standing

(2016)
T5-12 14.8 ± 5.4
T1-4 3.9 ± 4
Alqhtani et al. (2015) UK 18 m 30.6 ± 7.4 76.6 ± 7.4 171 ± 5 - Accelerometer T4-8 3.5 ± 4 Y >>>
T8-12 15 ± 8
T3-12 18.1
12 m 25.3 ± 3.9 79.97 ± 9.38 180 ± 5 T3-6 9.6
Schinkel-Ivy et al. (2014) Canada - Vicon - >>>
12 f 23.1 ± 2.9 60.03 ± 6.56 166 ± 6 T6-9 6.1
T9-12 8.6
20 m 24 ± 3.6
Edmondston et al. (2011) Australia - - - Photographs T1-12 11.5 ± 3.7 - >>>
20 f 21.6 ± 2.8
Hsu et al. (2008) Taiwan 18 m 31 79 171 - Electromagnetic C7-12 26 ± 9 - >>>

9m 45.4 ± 7.7 87.6 ± 16.8 181 ± 9 26.5 ± 5.2

Mannion et al. (2004) Switzerland Spinal Mouse T1-12 25.3 ± 9.9 Y >>>
11 f 38.2 ± 7.6 63.8 ± 7.6 167 ± 10 22.9 ± 5.7
13 m
Tully and Stillman (1997) Australia 26.3 ± 13.1 - 170 ± 10 - Videotape T1-12 17.8 ± 8.6 - >>>
9f

3f
Zemp et al. (2014) Switzerland 29 71 174 - Photographs T3-11 6.2 ± 3.5 - >>>
4m
C7-T3
South 10 m 15 f 71.7 ± 4 63.1 ± 8.6 158.7 ± 6.6 T3-6 Elder: 4.8, 3.3, 4.1, 4.9
Lee et al. (2013) - Vicon - >>>
Korea 9m2f 26.5 ± 1.8 65.4 ± 3.5 177.2 ± 2.8 T6-9 Young: 7.2, 6.7, 7.7, 11.6
T9-12
14 f
Folsch et al. (2012) Germany 33 ± 14.8 73.9 ± 15.1 1.74 ± 0.11 - Zebris CMS 20 C7-L1 27.6 ± 14.1 Y >>>
Sitting

14 m
T1-4
30 m m: 7.8 ± 5.0,10 ± 3.3, 12.5 ± 2.8:
Willems et al. (1996) Australia 18-24 - - - Fastrak T4-8 - >>>
30 f f: 9.5 ± 4.9, 11.4 ± 4.3,12.8 ± 3.9
T8-12
20 f 22 - 29 32.98 ± 9.02
20 f 30 - 39 24.58 ± 5.93
20 f 40 - 49 22.83 ± 7.56
O'Gorman and Jull (1987) Australia - - - Inclinometer T1-L1 Y >>>
20 f 50 - 59 15.98 ± 8.03
20 f 60 - 69 17.25 ± 9.60
20 f >70 15.85 ± 8.29

35
Table 3
Thoracic range of extension (RoE).
Author Age Weight Height BMI PubMed
Posture Country Numb, Sex Device Level RoE (°) Reliability
(Year) (ys) (kg) (cm) (kg/m2) Link

Radiological studies
arms elevated

≤25
Standing

Edmondston et al. (2012b) Australia 21 m 22.6 ± 3.2 72.6 ± 8.1 177 ± 7.0 Radiographs T1-12 12.8 ± 7.6 - >>>

Edmondston et al. (2012a) Australia 14 m 30.2 ± 7 74.7 175 - Radiographs T1-12 12 ± 8.9 Y >>>
Sitting

3f
Zemp et al. (2014) Switzerland 29 71 174 - MRI T3-11 7.4 ± 4.4 - >>>
4m

Non-radiological studies

T1-12 11.7 ± 3.4

Narimani and Arjmand (2018) Iran 22 m 24.8 ± 1.0 71.5 ± 10.0 178.0 ± 5.4 22.5 ± 2.7 Inertial device T1-5 10.1 ± 6.7 - >>>
T5-12 5.4 ± 4.5

T1-4 7.1 ± 4
Standing

Alqhtani et al. (2015) UK 18 m 30.6 ± 7.4 76.6 ± 7.4 171 ± 5 - Accelerometer T4-8 11.2 ± 4 Y >>>
T8-12 7.9 ± 6

20 m 24 ± 3.6 11.3 ± 4.2

Edmondston et al. (2011) Australia - - - Photographs T1-12 Y >>>
20 f 21.6 ± 2.8 7.7 ± 3.9

Hsu et al. (2008) Taiwan 18 m 31 79 171 - Electromagnetic C7-T12 17 ± 10 - >>>

9m 45.4 ± 7.7 87.6 ± 16.8 181 ± 9 26.5 ± 5.2

Mannion et al. (2004) Switzerland Spinal Mouse T1-12 13.5 ± 10.7 Y >>>
11 f 38.2 ± 7.6 63.8 ± 7.6 167 ± 10 22.9 ± 5.7
arms elevated

Edmondston et al. (2012b) Australia 21 m 22.6 ± 3.2 72.6 ± 8.1 177 ± 7.0 ≤25 Photographs T1-12 10.5 ± 4.4 - >>>
Standing

Edmondston et al. (2012a) Australia 14 m 30.2 ± 7 74.7 175 - Photographs T1-12 12.4 ± 4.1 Y >>>

36
 3f
Zemp et al. (2014) Switzerland 29 71 174 Photographs T3-11 12.7 ± 5.8 - >>>
4m

14 f
Folsch et al. (2012) Germany 33 ± 14.8 73.9 ± 15.1 174 ± 11 - Zebris CMS 20 C7-L1 26.1 ± 11.3 Y >>>
14 m

20 m 24 ± 3.6 12.8 ± 4.0

Edmondston et al. (2011) Australia - - - Photographs T1-12 Y >>>
20 f 21.6 ± 2.8 8.8 ± 3.7
Sitting

T1-4
30 m m: 9.0 ± 6.4; 9.9 ± 5.8; 7.8 ± 4.6;
Willems et al. (1996) Australia 18 - 24 - - - Fastrak T4-8 - >>>
30 f f: 7.1 ± 7.8; 7.9 ± 4.6; 9.7 ± 5.2
T8-12

20 f 22 - 29 37.43 ± 10.52
20 f 30 - 39 28.68 ± 12.42
20 f 40 - 49 20.75 ± 5.90
O'Gorman and Jull (1987) Australia - - - Inclinometer T1-L1 Y >>>
20 f 50 - 59 19.73 ± 7.06
20 f 60 - 69 16.68 ± 6.68
20 f >70 11.85 ± 7.04

10 m
Park et al. (2015) South Korea 26.15 ± 5.34 61.92 ± 9.17 168.25 ± 7.46 - Inclinometer T1-12 7.21 ± 9.9 - >>>
Prone

12 f

20 m 24 ± 3.6 14.6 ± 3.4

Edmondston et al. (2011) Australia - - - Photographs T1-12 Y >>>
20 f 21.6 ± 2.8 13.8 ± 4.1
kneeling
4-point

20 m 24 ± 3.6 14.8 ± 4.3

Edmondston et al. (2011) Australia - - - Photographs T1-12 Y >>>
20 f 21.6 ± 2.8 15.4 ± 3.9

Table 4
Thoracic range of flexion plus extension (RoFE).
Author Age Weight Height BMI PubMed
Posture Country Numb, Sex Device RoFE (°) Results Reliability
(Year) (ys) (kg) (cm) (kg/m2) Link

Radiological study

37
 T1 - L1 31.7 ± 11.3
T1/2 3.9 ± 2.8
T2/3 2.6 ± 3.3
T3/4 1.2 ± 2.5
T4/5 0.9 ± 3.0
Supine

T5/6 1.5 ± 2.9

34 m
Morita et al. (2014) Japan 55.4 ± 14.7 - - - CT T6/7 2.1 ± 2.5 - >>>
16 f
T7/8 1.9 ± 2.6
T8/9 2.7 ± 2.6
T9/10 3.3 ± 2.5
T10/11 3.6 ± 2.2
T11/12 3.8 ± 2.5
T12/L1 4.2 ± 2.1

Non-radiological studies

20 m 24 ± 3.6
Edmondston et al. (2011) Australia - - - Photographs T1-12 20.2 ± 6.6 - >>>
Standing

20 f 21.6 ± 2.8
9m 45.4 ± 7.7 87.6 ± 16.8 181 ± 9 26.5 ± 5.2
Mannion et al. (2004) Switzerland Spinal Mouse T1-12 27.4 ± 13.3 Y >>>
11 f 38.2 ± 7.6 63.8 ± 7.6 167 ± 10 22.9 ± 5.7
6m
Troke et al. (1998) UK 26 ± 6.3 73 ± 10.79 172 ± 7.91 - CA 6000 T1-12 70 ± 16.23 Y >>>
5f
Sitting

12 m
Furness et al. (2016) Australia 30.83 ± 10.96 - - - Inclinometer T1/2 78.09 ± 15.24 Y >>>
15 f

Table 5
Thoracic range of lateral bending (RoLB).
Author Numb Age Weight Height BMI PubMed
Posture Country Device Level RoLB (°) Reliability
(Year) Sex (ys) (kg) (cm) (kg/m2) Link

Radiological devices

38
 T1-L1 15.6 ± 6.3
T1/2 1.4 ± 1.3
T2/3 1.3 ± 1.2
T3/4 1.4 ± 1.3
T4/5 0.9 ± 0.9
Supine

T5/6 0.8 ± 1.0

Fujimori et al. (2014) Japan 15 m 32.8 ± 1.6 70.9 ± 8.4 173.1 ± 5.7 23.6 ± 2.3 CT T6/7 1.1 ± 1.1 - >>>
T7/8 1.7 ± 1.2
T8/9 1.3 ± 1.2
T9/10 1.6 ± 0.7
T10/11 1.8 ± 0.8
T11/12 2.3 ± 1.0
T12/L1 2.2 ± 0.8

Non-radiological devices

T1-12
Narimani and Arjmand R: 26.5 ± 6.1; 6.6 ± 4.9; 21 ± 5.4
Iran 22 m 24.8 ± 1.0 71.5 ± 10.0 178.0 ± 5.4 22.5 ± 2.7 Inertial device T1-5 - >>>
(2018) L: 24.5 ± 7.4; 6.3 ± 3.2;18.8 ± 6.1
T5-12

T1-4
Alqhtani et al. (2015) UK 18 m 30.6 ±7.4 76.6 ± 7.4 171 ± 5 - Accelerometer T4-8 R: 6.5 ± 3; 7.8 ± 2; 12.1 ± 3 Y >>>
T8-12 L: 5.4 ± 4; 7.1 ± 3; 12.4 ± 4
Standing

T3-12 19.4
12 m 25.3 ± 3.9 79.97 ± 9.38 180 ± 5 T3-6 4.6
Schinkel-Ivy et al. (2014) Canada - Vicon - >>>
12 f 23.1 ± 2.9 60.03 ± 6.56 166 ± 6 T6-9 4.9
T9-12 9.2

R : 27 ± 6
Hsu et al. (2008) Taiwan 18 m 31 79 171 - Electromagnetic C7-T12 - >>>
L :26 ± 4

6m
Troke et al. (1998) UK 26 ± 6.3 73 ± 10.79 172 ± 7.91 - CA 6000 T1-12 Two-sides: 75 ± 12.21 Y >>>
5f
C7-T3
South 10 m 15 f 71.7 ± 4 63.1 ± 8.6 158.7 ± 6.6 T3-6 R: Elder: 3, 2.2, 6.2, 4.9; Young: 5.4, 4.9, 10.7, 9.5;
Lee et al. (2013) - Vicon - >>>
Korea 9m2f 26.5 ± 1.8 65.4 ± 3.5 177.2 ± 2.8 T6-9 L: Elder: 3, 2.3, 5.3, 4.4; Young: 5.4, 5.6, 8.6, 10.5
T9-12
R: m: 6.2 ± 2.7; 7.7 ± 2.3; 12.4 ± 2.9; f: 6.2 ± 2.1; 8.5 ±
T1-4
30 m 2.3; 11.9 ± 2.3;
Sitting

Willems et al. (1996) Australia 18 - 24 - - - Fastrak T4-8 - >>>

30 f L: m: 5.6 ± 2.5; 8.1 ± 2.5; 13.2 ± 2.4; f: 6.1 ± 2.9; 7.9 ±
T8-12
2.1; 12 ± 2.3
20 f 22 - 29
20 f 30 - 39 R: 37.45 ± 8.86; 32.60 ± 9.83; 25.85 ± 5.08; 22.25 ±
20 f 40 - 49 4.67; 21.95 ± 5.22; 20.45 ± 6.3;
O'Gorman and Jull (1987) Australia - - - Inclinometer T1-L1 Y >>>
20 f 50 - 59 L: 37.05 ± 6.6; 30.55 ± 8.96; 25.8 ± 5.83; 22 ± 5; 22.2
20 f 60 - 69 5± 4.95; 19.15 ± 5.92
20 f >70

Table 6

39
Thoracic range of axial rotation (RoAR).
Author Numb Age Weight Height BMI PubMed
Posture Country Device Level RoAR (°) Reliability
(Year) Sex (ys) (kg) (cm) (kg/m2) Link

Radiological study

T1-L1 24.9 ± 4.9

T1-4 13.6 ± 4.8
T4-8 23.3 ± 5.9
T8-12 9.1 ± 5.2
T1/2 1.2 ± 0.8
T2/3 1.6 ± 0.7
T3/4 1.4 ± 0.9
Supine

T4/5 1.6 ± 0.8

Fujimori et al. (2012) Japan 13 m 33.2 70.5 171.7 23.9 CT - >>>
T5/6 1.8 ± 0.7
T6/7 1.9 ± 0.6
T7/8 2.3 ± 0.7
T8/9 2.5 ± 0.8
T9/10 2.7 ± 0.6
T10/11 2.6 ± 0.8
T11/12 1.3 ± 0.7
T12/L1 0.5 ± 0.4

Non-radiological studies

T1-12
Narimani and Arjmand R: 37.1 ± 11.7; 17.2 ± 6.6; 19.5 ± 9.7
Iran 22 m 24.8 ± 1.0 71.5 ± 10.0 178.0 ± 5.4 22.5 ± 2.7 Inertial device T1-5 - >>>
(2018) L: 33.5 ± 10; 8.8 ± 5.3;24.8 ± 8.8
T5-12

T1-4
R: -14.9 ± 16; 34.8 ± 18; 21.4 ± 9
Alqhtani et al. (2015) UK 18 m 30.6 ± 7.4 76.6 ± 7.4 171 ± 5 - Accelerometer T4-8 Y >>>
L: -11.3 ± 21; 29.7 ± 18; 22.6 ± 13
T8-12

T3-12
Standing

12 m 25.3 ± 3.9 79.97 ± 9.38 180 ± 5 T3-6 m: 37.3; 10.8; 19.5; 12.1
Schinkel-Ivy et al. (2014) Canada - Vicon - >>>
12 f 23.1 ± 2.9 60.03 ± 6.56 166 ± 6 T6-9 f: 33.5; 14.3; 15.6; 8.4
T9-12

R: 49 ± 9
Hsu et al. (2008) Taiwan 18 m 31 79 171 - Electromagnetic C7-T12 - >>>
L: 42 ± 9

6m
Troke et al. (1998) UK 26 ± 6.3 73 ± 10.79 172 ± 7.91 - CA 6000 T1-12 Two-sides: 64 ± 10.79 Y >>>
5f

Gregersen and Lucas

USA 7m 22.86 ± 1.86 71.3 ± 9.2 178.8 ± 5.7 22.27 ± 2.31 Pin insertion T1-12 Two-sides: 74 - >>>
(1967)

15 m R: 45.4 ± 10
Johnson et al. (2012) Omaha 23.6 ± 4.3 71.4 ± 16.7 171.0 ± 9.6 - Inclinometer T1/2 Y >>>
31 f L: 47.2 ± 9.1

9m
Heneghan et al. (2009) UK 24.96 ± 2.6 70.8 ± 14.35 170.2 ± 8.7 - Ultrasound - 85.15 ± 14.8 Y >>>
15 f
Sitting

25 m L: 40.0 ± 7.9
Edmondston et al. (2007) Australia 23.2 68.5 172.3 - Optical system T2-10 - >>>
27 f R: 41.8 ± 7.0

R: m:15.9 ± 4.9; 25.3 ± 6.7; 8.3 ± 6.1; f: 14.4 ±

T1-4
30 m 4.6; 24.3 ± 6.1; 7.9 ± 5.5;
Willems et al. (1996) Australia 18 - 24 - - - Fastrak T4-8 - >>>
30 f L: m: 12.1 ± 5.1; 22.1 ± 6.3; 11.8 ±4.9; f: 11.8 ±
T8-12
4.4; 21.5 ± 4.6; 8.3 ± 4.2

40
 20 f 22 - 29
20 f 30 - 39 R: 60.28 ± 12.59; 51.85 ± 10.39; 47.55 ± 8.35;
O'Gorman and Jull 20 f 40 - 49 45.25 ± 7.63; 42.15 ± 7.44; 26 ± 5.4;
Australia - - - Inclinometer T1-L1 Y >>>
(1987) 20 f 50 - 59 L: 59.3 ± 11; 51.55 ± 10.77; 46.7 ± 7.81; 44.05 ±
20 f 60 - 69 6.75; 40.95 ± 8.05; 25.85 ± 6.05
20 f >70
locked position

10 m L:63.5 ± 11.4;
Furness et al. (2018) Australia 29.8 ± 8.9 67.9 ± 10.1 167.8 ± 8.9 - Goniometer T1/2 Y >>>
20 f R:62.5 ± 12.2
Lumbar

L: 40.33 ± 11.9;
12 m
Furness et al. (2016) Australia 30.83 ± 10.96 - - - Inclinometer T1/2 R: 41.5 ± 10.77; Y >>>
15 f
Total: 81.33 ± 21.1

41