THE ASSOCIATION BETWEEN ATHLETIC TRAINING TIME AND THE SAGITTAL CURVATURE OF THE IMMATURE

SPINE
+*Wojtys, E; **Ashton-Miller, J; *Huston, L; ***Moga, P
+*MedSport; Section of Orthopaedic Surgery; University of Michigan, Ann Arbor, MI. MedSport; 24 Frank Lloyd Wright Dr.; Ann Arbor, MI 48106, 734-998-7410,
Fax: 734-998-7993, edwojtys@umich.edu

INTRODUCTION: The normal development of the spine’s sagittal found in the thoracic and lumbar spinal curves when adolescents were
curvature depends upon a complex interaction between heritable growth stratified by their primary sport (Table 1). The non-athletic controls had
factors and the mechanical environment in which the spine grows. Sagittal significantly smaller thoracic and lumbar spine curves compared with each of
spine curvature is known to be modifiable by altering the loads applied during the athletic groups (p<0.001).
growth. Studies of the immature rat tail vertebrae have demonstrated that disc
height can be modulated by mechanical loading history (1). Excessive spinal
compressive loads have also been found to affect normal human spinal
development at the apophyses (2). The research question addressed in this
study is whether exposure to long hours of athletic training is associated with
increases in the sagittal curvature of the immature spine. The primary null
hypothesis (Ho) tested was that larger angles of thoracic kyphosis or lumbar
lordosis are not associated with increased exposure to athletic training as
quantified by the number of annual training hours. Secondary null hypotheses
were that the angles of kyphosis or lordosis would not differ by the primary
sport of choice, age, or gender.
MATERIALS AND METHODS: 2,270 children (407 females) between the
ages of 8 and 18 were studied after obtaining consent from the Institutional
Review Board as well as written permission from their parent/guardian.
These individuals attended local schools or volunteered from our University’s
annual summer athletic training sports camps between 1989 and 1996. The
athletic subject population was recruited by the primary sport declared by the Figure 1
child. An age-matched group of children reporting no sports participation or
experience of manual labor was recruited and used as a ‘non-athletic’ control Table 1. Mean (SD) training exposure, thoracic kyphosis, and lumbar lordosis
group. angles stratified by sport.
Testing Apparatus: This method used to measure spine curvature was similar Sport Training Exp. Thoracic Angle Lumbar Angle
to that developed by Drerup and Hierholzer (3,4). A tripod-mounted custom- (hours/year) (°) (°)
made 33mm projector, a motor drive Canon F1 35 mm film camera, and a Control Group 0 16.1 (10.4) 17.6 (15.6)
custom-built platform for standardizing subject position and orientation were Track 198 29.5 (10.1)* 33.5 (17.0)*
utilized. Fixed onto the platform were two vertical stanchions, each Volleyball 150 29.9 (13.9)* 37.4 (21.7)*
supporting a fixed array of four light-emitting diodes and a trochanter pad.
Hockey 198 38.1 (11.7)* 44.5 (14.8)*
The LEDs provided fixed points in 3D space for the calibration procedure and
Wrestling 228 39.5 (12.2)* 42.4 (16.2)*
thereafter served as permanent fiducial reference points in every image.
Adjustable, bilateral cupped trochanter pads served to fix the pelvis within the Football 282 39.8 (16.2)* 46.2 (20.2)*
photographic reference plane. Swimming 379 40.8 (13.0)* 44.0 (16.5)*
Testing Protocol: Each subject was interviewed about their sports Gymnastics 439 42.4 (13.4)* 52.1 (16.7)*
participation history, work activities, musculoskeletal injuries, and self- Overall Avg. 38.5 (13.6) 42.4 (17.5)
reported training programs. Anthropometric and flexibility data was also * significant with the control group at p<0.0001
collected. Photographs of the backs of each subject were obtained to measure
the degree of thoracic and lumbar curvature. Surface markers were placed at DISCUSSION: The novel finding in this study was the significant increase
T1, T10, L2 and S1 landmarks. Films were taken after instructing the subject in spine curvature associated with cumulative training exposure in this sample
to stand in a relaxed, upright position. Each 35 mm photograph image was of children. The finding of significantly smaller thoracic and lumbar angles in
converted to bitmap files and then processed using a sequence of custom the non-athletic controls suggests that a certain amount of physical activity
computer programs written to quantify the angles of thoracic kyphosis and may be required for the development of normal curves. We believe from a
lumbar lordosis of each subject. musculoskeletal developmental standpoint that physical activity is beneficial
Statistics: Univariate analyses were conducted to provide summary statistics for proper bone growth and maturation. Strenuous exercise, however, can
of continuous and discrete variables. Since data followed a Gaussian have a deleterious effect on immature bone morphology and its mechanical
distribution, ANOVA (for primary hypothesis) with post-hoc t tests with integrity. In highly competitive sports, the spine of the growing athlete
Bonferonni’s correction factor (for secondary hypothesis) were used to appears to be vulnerable. Long-term exposure to such forces appears to
determine differences in thoracic and lumbar spine angles between sports. correlate with altered spinal development.
Nonlinear regression was used to explore the relationship between the number REFERENCES: 1) Stokes IA et al., J Spinal Disorders, 11:261-5, 1998.
of training hours and angle of kyphosis and lordosis. 2) Ippolito E et al., J Ped Ortho, 5: 315-322, 1985.
RESULTS: No significant differences were found between thoracic and 3) Drerup B et al., JOB, 20:961-970, 1987.
lumbar spine sagittal angles and either gender or age. Both thoracic and 4) Drerup B et al. JOB, 20: 971-977, 1987.
lumbar angles of curvature increased significantly with the number of annual
training hours (p<0.05, Table 1). A nonlinear regression of curvature angle on ACKNOWLEDGMENTS: This work was supported by a grant from the
training time demonstrated coefficients of determination of 0.72 for thoracic Orthopaedic Research and Education Foundation.
kyphosis and 0.69 for lumbar lordosis (Figure 1). A significant difference was
**Department of Mechanical Engineering and Applied Mechanics; University
of Michigan, Ann Arbor, MI.
***Department of Kinesiology; University of Michigan, Ann Arbor, MI.

Poster Session - The Spine - VALENCIA FOYER

46th Annual Meeting, Orthopaedic Research Society, March 12-15, 2000, Orlando, Florida 0377