Thoracic kyphosis comparisons in adolescent female competitive Field
Hockey players and untrained controls
Reza Rajabi PhD1, L. Mobarakabadi Msc1 , H.M .Alizadhen PhD1, P. Hendrick 2
1Physical Education and Sport Sciences Department, Health and Sport Medicine
Group, University of Tehran, Iran,
2Centre for Physiotherapy, University of Otago, Dunedin New Zealand
Corresponding author: R. Rajabi, PhD, Physical Education and Sport Sciences
Department, Health and Sport Medicine Group, University of Tehran, Iran, P.O.B;
1439813117, Tel; +98 9121772997 (e-mail: firstname.lastname@example.org)
Statements and declarations
The authors declared that they had no conflicts of interest in their authorship and
publication of this contribution.
Background: Field hockey requires athletes to adopt regular semi-crouched posture during
training and competition. However, it is not known to what extent field hockey training
regimes might impact on postural features particularly in thoracic area.
Study design: Cross-sectional study.
Purpose: To compare the thoracic kyphosis angles of female field hockey players and non-
athletes and to examine the relationship between the thoracic kyphosis and training regimes
Methods: Seventy four female participants including 37 field hockey players (mean age
19.03 ±1.24 years) and 37 non-athletes (mean age 18.21± 1.22) were recruited. The hockey
players met a minimum criterion of 3 years of experience in the Iranian first division female
hockey league. The thoracic kyphosis degree was measured between T1 -T12 using a non-
invasive flexible ruler.
Results: A significant difference in the thoracic kyphosis degree for athletes (M = 41. 71,
SD = 5.38) and non-athletes (M = 36.72, SD =6.01); t (72) = 3.76, p= 0.001 was revealed.
The magnitude of the differences in the means was very large (eta squared = .016). There was
a medium positive correlation between the athletic history and thoracic kyphosis, r =0.36, n =
37, P = 0.031. However, the relationship between the cumulative training exposure and the
degree of thoracic kyphosis was not significant.
Conclusion: A significant increase in thoracic curvature of adolescent female field hockey
players was found and this was associated with the cumulative number of years of hockey.
These results suggest a possible association between the semi-crouched posture in field
hockey and thoracic kyphosis in adolescent (immature) athletes who have been engaged in
this sport at a higher level, thus implicating the field hockey training regime as a possible
determinant of kyphosis.
Keywords: Field hockey, female, adolescent population, thoracic kyphosis
It has been claimed that during specific physical activities, the body adopts various subtle
postural deviations that are well suited to the physical effort and, to the required demands of
any given activity.14, 32 As a result, athletes frequently present with postures that are generally
specific to their sport. Studies have shown that athletes are more prone to developing postural
deviations than non-athletic individuals. 31 One such postural deviation is increased thoracic
kyphosis, which has been previously reported in bilateral elite athletes such as; cyclists23,
young Kayakers15, wrestlers 22and in school children participating in rigorous sports such as
track, volleyball, football, swimming, gymnastic, ice hockey and wrestling. 34 It is suggested
that the increased kyphosis recorded in these sports results from the intensive sport-specific
training programme 3, 34and the specific postural demands imposed .32
Field hockey has a long history, dating back to 2050 BC 25 and currently is one of the most
popular team sports in the world, second only to soccer. 19 The popularity of this game
amongst women is high. 33 A unique aspect of the sport is the ability to run, at pace, in the
bent forward position whilst dribbling the ball and moving quickly in a semi-crouched
posture. 24 Therefore the position of the spinal column in hockey players during training or
competition differs from the anatomical standing position. Professional hockey players
sustain this position for a considerable amount of time per day over a period of many years.
During the hitting of the ball, there is an additional inclination to the trunk, involving slight
torsion. 12 It could be argued that the typical sagittal anatomical curves may be accentuated in
athletes that take up hockey or may be increased as a consequence of prolonged training or
competition, as has been suggested in other sports.1, 22, 23, 34
Although injury rates in field hockey have been extensively studied 6, 7, 20; the sagittal
thoracic posture of hockey players and the interrelationships between posture and hockey
have received little or no attention in the research literature. The natural history of thoracic
postural changes in hockey players is not clear and it is not known as to what extent the
hockey training regime impacts on thoracic spine posture. Therefore the aims of this study
was first to compare the thoracic kyphosis among female competitive field hockey players
and secondly, to examine the association between thoracic kyphosis and cumulative training
exposures in terms of hours and years of hockey participation. To the best of our knowledge,
this is the first study to focus on the thoracic posture of adolescent (skeletally immature)
female competitive field hockey players and its potential interrelationship with training
MATERIALS AND METHODS
Thirty seven athletes volunteered to take part in the study from the 14 teams in the Iranian
female hockey league (n =224). A group (n=37) of age and gender matched non-athletic
participants were recruited from some local high schools as a control group. The entry
criterion for the athletic group was competing in the Iranian national hockey league. The
athletic group was also required to have a minimum of 3 years hockey experience at club
level (confirming that when the athletes started playing hockey at the higher level, they were
skeletally immature). Purposive sampling (criterion-based sampling) was used to recruit the
subjects. Each athlete completed a questionnaire regarding personal demographic data,
sports participation history, musculoskeletal injuries, and self-reported training programs.
The inclusion criteria for the non-athlete (control) group were self-report of nonparticipation
in any competitive sport or recreational physical training program. Exclusion criteria for both
groups were a history of a traumatic back injury, a radiographically documented scoliosis
exceeding 20° (Cobb angle)34, or a documented congenital spine abnormality. The athletes
were measured during the Iranian female hockey league season (mid season) at the Azadi
Sports Complex. This research was approved by the University of Tehran Institutional
Review Board and participants provided written informed consent prior to being involved
with the study.
Personal demographic data, sports participation history, musculoskeletal injuries and self-
reported training programs were collected prior to undertaking the physical measurements of
thoracic kyphosis, height (cm) and weight (kg). Thoracic kyphosis degree was measured
using a 50 cm non-invasive flexible ruler (flexicurve) tool (Rumold, UK) between the skin
overlying the spinous processes of T1 to T12. 10 The validity and reliability of the flexible
ruler for the measurement of thoracic kyphosis has previously been established both in
healthy and kyphotic participants.13 Furthermore, the intra-class correlation coefficient (ICC)
between radiographic measurements and the flexible ruler method has been shown to be
excellent 0.90.30 Participants were asked to remove their upper body clothing so that the
surface landmarks of T1 and T12 spinous processes could be identified by palpation. All
measurements were performed with participants in their usual upright posture and without
attempting to stand unusually straight, and with their weight evenly distributed. Participants
were asked to look straight ahead and breathe normally with their arms by their side. The
flexible ruler was then placed over the T1 to T12 spinous processes of the thoracic spine and
shaped to fit the contours of the sagittal spinal curve. The thoracic kyphosis measurement
procedure and kyphosis calculation in this study was in keeping with previous studies. 26, 30
The normality distribution of the two groups’ raw data was confirmed using the
Kolmogorov–Smirnov Test. Descriptive statistics were used to profile the anthropometric
and thoracic kyphosis angles for the two groups. An independent-samples t-test was used to
compare the means of thoracic kyphosis between the two groups. Baseline comparability of
the two groups in demographic variables (age, height, weight) was confirmed prior to
analysis (using t-test). Associations between hockey playing exposure variables and thoracic
kyphosis were examined using the Pearson product-moment correlation coefficient (r). Effect
size (ES) was calculated using the formula of; Eta squared=
t2+ (N1+N2−2) . Cohen
classification; 01 as a small effect, .06 as a medium effect and .14 as a large effect was used
to judge the results of ES. 21 All statistical analyses were performed using the statistical
package of SPSS for Windows (version 16). The probability value (alpha) of statistical
significance was set at P≤.05.
Sports related specification of athletes is detailed in table 1. Using an independent-samples t-
test, there was no significant difference between the anthropometric characteristics (Table 2)
of the two groups in terms of age, height and weight (P>0.05). There was a significant
difference in the degrees of thoracic kyphosis between the athletic group (M = 41. 71, SD =
5.38) and the non-athletes (M = 36.72, SD =6.01); t (72) = 3.76, p= 0.001 (two tailed). The
magnitude of the difference in means (mean difference = 4.99, 95% CI: 2.34 to 7.63) was
very large (eta squared = .016). There was a medium 21, positive correlation between a
longer athletic history (years) and increased thoracic kyphosis, r =0.36, n = 37, P = 0.031.
However, the relationship between the cumulative training exposure (hours) and the degree
of thoracic kyphosis was not significant r =0.14, n = 37, P = 0.388.
This study found that there was a significant difference (P≤.001) in the mean thoracic
kyphosis between female hockey players (41.71˚± 5.38°) and an age-matched non-athletic
(36.72˚± 6.01°) control group. From the descriptive data it was found that 78.4% of the
hockey players exceeded the mean thoracic kyphosis angle of 36.72° of the non-athletic
group. The normality distribution in thoracic kyphosis for both groups confirms that these
significant group differences could not be attributed to significant outliers in either group.
Considering the non-significant differences in age, height and weight among the two groups,
the difference in the degree of thoracic kyphosis between the hockey players and non-athletes
is arguably related to the long term specific training and sport related posture of the athletes.
There are no previous studies on the thoracic posture of field hockey players to compare our
findings, however, a radiographic study of 189 ice hockey players also reported that the
degree of thoracic kyphosis among children between the ages of 8 and 18 years was higher
(38.1±11.7) than an age-matched control group(16.1 ±10.4). However, the standard deviation
for thoracic kyphosis was noted to be high and the values of kyphosis were not reported for
boys and girls separately in the study.
In our study, increased thoracic kyphosis in the athletes was proportional to the number of
years each athlete was engaged in playing hockey. This result indicates that the longevity of
an athletes’ sports history is associated with the extent of thoracic kyphosis. This result is
consistent with findings in other sports such as; football, gymnastics, ice hockey, swimming,
wrestling, track and volleyball. 34 However, the relationship between the cumulative training
exposure (hours) and the degree of thoracic kyphosis was not significant in our study. A
previous study reported that an increase in training hours of 300% was associated with just a
5° increase in spinal curvature of athletes (including ice hockey) between the ages of 8-18
years. 34 Therefore, our results along with above mentioned study 34 suggest that the years of
involvement in sport has a greater impact on increased thoracic posture than the cumulative
hours in the sport.
In comparison with the usual (anatomical) standing position of the body, the spine in hockey
players adopts an unusually flexed position while moving the ball (dribbling, hitting, pushing
or flicking) during training and competition. Theoretically this position seems to have an
adverse effect on normal spine posture and might be an important factor associated with the
increased thoracic kyphosis found in this group of athletes. The influence of mechanical
loading (stress) on the thoracic vertebrae in the flexed position and its association with a
morphological adaptation of the thoracic elements and an accompanying increase in kyphosis
is well established. 4, 9; 28 Apophyseal rings, cartilaginous secondary growth centers are
mostly highly stressed when axial compression loads are combined with a bending moment.
As the anterior aspect of the thoracic vertebra is narrow, the most ventral region of the
apophyseal ring will experience the highest compressive stress under a flexion bending
moment. 2 The vertebral bodies account for most of the variability in thoracic kyphosis 9 27
and this may affect the vertebral shape in the region and thereby, potentially the thoracic
curvature as whole.
A number of theories as cited in 16, 17 29 have been proposed to explain the body’s adaption of
an increased thoracic kyphosis amongst athletes with semi-flexed position. It is claimed that
bone growth disturbances can occur as a result of asymmetric vertebral loading 4 and these
have the potential to affect vertebral structure and thereby sagittal curvature. This claim
proposes that the increased pressure on the anterior portion of vertebrae, as a result of a
flexed kyphotic posture, produces a "remodeling" or "reforming" of the bone (vertebrae),
resulting in the appearance of anterior compression. The low compression stress in the
posterior part of the vertebral column consequently results in an increase in vertebral body
growth in this area compared with the anterior portion of the vertebral body which is under
higher compression stresses. This theory has been used as a possible justification for the
increased kyphosis found in cyclists who adopt flexed posture while cycling 23: 11 and
freestyle wrestling.22 Although there is no experimental evidence to support this theory, it can
be verified by Volkmann’s law in respect of vertebrae 18. Volkmann’s law indicates that
pressure on an epiphysis delays the speed of growth in the affected area whilst tension in the
posterior portion of the vertebral column increases the amount of the growth. It has been
suggested that bending forward greatly increases the intradiscal pressure, causing fracture of
the normal vertebral endplate 11. From a biomechanical perspective, both impact forces and
large thoracic extensor muscle forces, associated with lengthening contractions as a result of
torso flexion, could cause excessive stress on the anterior thoracic apophyses.2 Similar semi-
crouched postures to that required 24 during a game of hockey are associated with greater
pressure on the anterior portion in comparison to the posterior portion of the vertebrae. If the
athlete is young, this pressure may act on an epiphysis of the anterior portion of the vertebrae,
delaying the speed of growth in this area. As cited in 2 the length of the thoracic spine grows
most rapidly between the ages of 10 and 16 years. It may be during this time that the
columnar arrangements of cells in the ring apophyses are most vulnerable to excessive
loading and repetitive loading. Thus immature bones are more sensitive to loading and
postural adaptation than mature bones. 5 It is also suggested that strenuous physical activity
has the potential to cause abnormalities in the growth of the immature spine. 34 Therefore, the
above mentioned physiological and biomechanical mechanisms might be associated with the
increased kyphosis found amongst our adolescent female hockey players in the current study.
However, there is no evidence to indicate that the observed increase in thoracic kyphosis of
the hockey players in this current study represents a true pathological change.
Thoracic kyphosis has been proposed as a surrogate indicator of bone mass density of the
lumbar spine with moderate associations found between the two parameters 8. What is not
known is whether the increased kyphosis observed in the current study is due to an
accompanying loss of bone mass or alternatively other structural changes within the spine
such as changes within the vertebral–endplate 11, 17, 29. Further prospective research is
required to evaluate whether the finding of an increased thoracic kyphosis angle predisposes
hockey players to back problems in years to come.
LIMITATIONS AND RESEARCH DIRECTIONS
Limitations of the methods used in this study include the use of a convenience sample of
female field hockey players in lieu of a randomized sampling technique, and the cross-
sectional study design – this is really a limitation in terms of looking at causation rather than
association between the variables. From a clinical perspective, the findings are of relevance
where knowledge of the potential adverse effects of prolonged training in semi-crouched
postures on the thoracic spine and when advising those at risk with underlying spinal
pathology who may be entertaining the idea of entering the field hockey sport at competitive
Hockey players were selected as the sample for this study because of the unusual anatomical
position of the spine during competition and training. Of primary importance was our finding
that a significant increase in spinal curvature in the adolescent female field hockey player was
found (compared to an age-matched control group) and this was associated with cumulative
number of years in the sport. These results suggest a possible association between field
hockey semi-crouched posture and thoracic kyphosis in athletes who have been engaged in
this sport at a higher level and implicate the field hockey training regime as a determinant of
These results suggest the adolescent female participating in rigorous (competitive level) field
hockey sport had a higher risk of becoming hyperkyphotic than those not participating in any
form of sports. An association between increased thoracic kyphosis angle and years involved
in field hockey in higher level was revealed.
SUGGESTION FOR FUTURE STUDIES;
There is a need to establish the presence and extent of a causal relationship between kyphosis
and training in sports where spinal flexion stress is evident. Youth sports programs should
endeavor to monitor the degree of kyphosis over time in the same athlete and correlate these
outcomes with training dose.
For future research, investigation on lateral curvature of spine (scoliosis) provides more
detailed information about possible postural adaptation to spinal torsion involvement in this
1. Archer J, Hamley E. J, E. RH. The Fitness Assessment of Competitive Cyclists. Bull
Br Assoc Sport Med. 1965;2(3-4):90–100.
Ashton-Miller JA. Thoracic hyperkyphosis in the young athlete: a review of the
biomechanical issues. Current Sports Medicine Reports. Feb 2004;3(1):47-52.
Bloomfield J. Applied Anatomy and Biomechanics in Sport. 2 ed. Melbourne: Blackwell
Scientific Publication; 1994.
Bradford DS. Juvenile Kyphosis. Clinical Orthopaedics and Related Research.
Carter DR. Mechanical Loading Histories and Cortical Bone Remodeling. Calcified
Tissue International. 1984;36:S19-S24.
Dick R, Hootman JM, Agel J, et al. Descriptive epidemiology of collegiate women's
field hockey injuries: National Collegiate Athletic Association Injury Surveillance
System, 1988-1989 through 2002-2003. Journal of Athletic Training. Apr-Jun
Elliott AJ, Jones D. Major ocular trauma: a disturbing trend in field hockey injuries.
British Medical Journal Clinical Research Ed. Jul 7 1984;289(6436):21-22.
Goh S, Price RI, Leedman PJ, Singer KP. The relative influence of vertebral body and
intervertebral disc shape on thoracic kyphosis. Clinical Biomechanics. Aug
Goh S, Price RI, Leedman PJ, Singer KP. A comparison of three methods for
measuring thoracic kyphosis: implications for clinical studies. Rheumatology. Mar
Hinman MR. Comparison of thoracic kyphosis and postural stiffness in younger and
older women. Spine J. Jul-Aug 2004;4(4):413-417.
Jayson MI, Herbert CM, Barks JS. Intervertebral discs: nuclear morphology and
bursting pressures. Ann Rheum Dis. Jul 1973;32(4):308-315.
Junghanns H. Clinical implications of normal biomechanical stresses on spinal
function,. Maryland: Aspen Publishers Inc; 1990.
Kahalkhali M, Parnianpour M, Hossein. K, Mobinin B, A. K. The validity and reliabilty fo
measurement of thoracic kyphosis using Flexble ruler in postural hyper-kyphotic patients.
Journal of Rehabilitaiton. 2003;4(Fall-Winter)(14-15):18-23.
Kyle CR. Energy and Aerodynamics in Bicycling. Clinics in Sports Medicine. Jan
Lopez-Minarro PA, Muyor JM, Alacid F. Sagittal spinal curvatures and pelvic tilt in
elite young kayakers. Medicina Dello Sport. Dec 2010;63(4):509-519.
Mellion MB. Neck and Back Pain in Bicycling. Clinics in Sports Medicine. Jan
Micheli LJ. Back Injuries in Gymnastics. Clinics in Sports Medicine. 1985;4(1):85-
Modi HN, Suh SW, Song HR, Yang JH, Kim HJ, Modi CH. Differential wedging of
vertebral body and intervertebral disc in thoracic and lumbar spine in adolescent
idiopathic scoliosis - A cross sectional study in 150 patients. Scoliosis. 2008;3:11.
Murtaugh K. Injury patterns among female field hockey players. Medicine and
Science in Sports and Exercise. Feb 2001;33(2):201-207.
Murtaugh K, Murtaugh K. Field hockey injuries. Current Sports Medicine Reports.
21. Pallant J. Survival Manual; A Step by Step Guide to Data Analysis using SPSS for
Windows. England: Open University Press; 2007.
Rajabi R, Doherty P, Goodarzi M, Hemayattalab R. Comparison of thoracic kyphosis
in two groups of elite Greco-Roman and freestyle wrestlers and a group of non-
athletic participants. Br J Sports Med. Mar 2008;42(3):229-232; discussion 232.
Rajabi R, Freemnt T, Doherty P. The investigation of thoracic kyphosis in cyclists and
non-cyclists. Kinanthropometry VIII: Proceedings of the 8th International Conference
of the International Society for the Advancement of Kinanthropometry (ISAK).
illustrated ed: Routledge; 2003:263-273.
Reilly T, Seaton A. Physiological Strain Unique to Field Hockey. Journal of Sports
Medicine and Physical Fitness. Jun 1990;30(2):142-146.
Rishiraj N, Taunton JE, Niven B. Injury profile of elite under-21 age female field
hockey players. Journal of Sports Medicine and Physical Fitness. Mar 2009;49(1):71-
Seidi F, Rajbi R, Ebrahimi TI, Tavaniei AR, Moussavi SJ. The Iraninan Flexible ruler
reliability and validity in lumbar lordosis measurements. World Journal of Sport
Sizer PS, Jr., Phelps V, Azevedo E. Disc related and non-disc related disorders of the
thoracic spine. Pain Practice. Jun 2001;1(2):136-149.
Stokes IAF. Mechanical effects on skeletal growth. J Musculoskel Neuron Interact; .
Sward L, Hellstrom M, Jacobsson B, Peterson L. Back pain and radiologic changes in
the thoraco-lumbar spine of athletes. Spine (Phila Pa 1976). Feb 1990;15(2):124-129.
Teixeira FA, Carnalho GA. Reliability and validity of thoracic kyphosis
measurements using flexicurve method. Revista Brasileira De Fisioterapia. May-Jun
Thurzova E, Kutlik D. Funkcne a svalove zmeny u mladych rychlostnych kanoistov a
ich ovplyvnenie v treningovom procese. (Muscle function changes in young speed
canoeists and their influencing in training program.). Acta Facultatis Educationis
Physicae.Universitatis Comenianae. 1997; 38:137-143.
Usabiaga J, Crespo R, Iza I, Aramendi J, Terrados N, Poza JJ. Adaptation of the
lumbar spine to different positions in bicycle racing. Spine. Sep 1 1997;22(17):1965-
Wassmer DJ, Mookerjee S. A descriptive profile of elite U.S. women's collegiate field
hockey players. Journal of Sports Medicine and Physical Fitness. Jun
Wojtys EM, Ashton-Miller JA, Huston LJ, Moga PJ. The association between athletic
training time and the sagittal curvature of the immature spine. American Journal of
Sports Medicine. Jul-Aug 2000;28(4):490-498.
TABLE 1. Activity related characteristics of hockey players(n=37)
Age starting hockey as a main sport 14.70 (±1.59)
15 Download full-text
Athletic history 4.36 (± 1.37)
Weekly training hour 6.24 (±.88)
Yearly training hours
Values are presented as mean (SD)
TABLE 2. Demographic characteristics of study participants (n =74).
Group N Age Weight (Kg) Height
Hockey players 37 19.03(±1.24) 57.67 (±6.91) 163.90 (±5.89)
Non-athletes 37 18.21 (±1.22) 56.62 (±6.90) 162.32 (±3.94)
Values are presented as mean (SD)