Oct 19, 1992 - Corps. We report a prospective study of783 male Israeli recruits aged from 17 to 26 years. The risk of stress fracture was inversely proportional.
YOUTH
IS A
A STUDY
C.
OF
MILGROM,
From
RISK
783
A.
FACTOR
INFANTRY
STRESS
N. SHLAMKOVITCH,
University
Hospital,
N.
Jerusalem
RAND,
and
Surg
Received
[Br]
1994;
76-B:20-2.
1992; Accepted
19 October
after
revision
1993
2 June
B.
the Israel
We report a prospective study of783 male Israeli recruits aged from 17 to 26 years. The risk of stress fracture was inversely proportional to age on both univariate and multivariate analysis. Each year of increase of age above 17 years reduced the risk of stress fracture by 28%. J BoneJoint
FRACTURE
RECRUITS
FINESTONE,
the Hadassah
FOR
LEV,
A.
Defence
of the
epidemiology
of stress
fractures
derives from mass, density, three-dimensional and the quality of the component material but
role
of these
strains
the
which
lead
Study
is made
ofbone Wall,
more
architecture Chatterji The basic
factors to stress
in resisting fracture
difficult (Lindahl
yielded
(Giladi et al that strength
arrangement (Heaney 1992), the repetitive
in bone
by the strong
is not
1967;
and Jeffery 1979; Cohn et al 1982). training of Israeli Army infantry
Garn
1970;
offers
an
almost laboratory-like environment for the creation and study of stress fractures (Lancet 1986), and previous studies have identified some of the risk factors (Giladi et al 1991). We have hypothesised that bone age may be important, and as part of our ongoing research we have prospectively
evaluated stress
the effect of the age of recruits
fracture
among
on the incidence
of
them.
Before
AND
In a prospective recruits during
not complete
training
for reasons
796 male elite infantry training. Of these 13 did
unrelated
starting
to stress fracture,
training
all recruits
N. Rand,
MD,
MD, Associate Professor MD, Orthopaedic Resident Orthopaedic
Resident
A. Simkin, PhD, Biomechanical Engineer Department of Orthopaedics, Hadassah P0 Box 12000, Jerusalem, Israel. N. Shlamkovitch, MD, Major B. Lev, MD, Vice-Surgeon General M. Wiener, MD, Surgeon General Israel Defence Forces Medical Corps,
Correspondence
should
be sent
©1994 British Editorial Society 030l-620X194/1684 $2.00
20
University
Hospital,
Em
Kerem,
were
all gave
interviewed
sports activity and any history suggestive They all had an orthopaedic examination
distance,
external
quadriceps
fitness
was assessed
possible
number
rotation
et al length,
of
at 85#{176} knee
strength
the
1993). tibial
hip,
flexion.
and
the number
of sit-ups
one minute. The date and place of birth of each recruit recorded, and the country of origin of each parent. immigrants, the date of arrival in Israel was recorded. The
recruits
and
Physical
by the time of a 2 km run, the maximum of chin-ups,
were
followed
during
training
in were For
by a team
of
army doctors and an orthopaedic surgeon, and had free access to the medical staff. Examinations for stress fractures were made every two to three weeks. Suspected cases, with symptoms
suggesting
surgeon,
had appropriate
ten days
of the orthopaedic
intravenous with imaging Israel). Spot
stress
fracture
to
radiographs
the
orthopaedic
and a bone
examination.
This
scan
within
involved
an
dose of 20 mCi of ‘“Tc-methyldiphosphonate by an Elscint Dynmax camera (Elscint, Haifa, views of the pelvis, femora, tibiae and feet were
taken 120 minutes after injection and evaluated on a grading system of 1 to 4 (Zwas, Elkanovitch and Frank 1987). All data were recorded on a specially designed the
form and processed Israel
North
using
Defence
through
Medical
the Statistical
Carolina),
with
the computer Corps.
Analysis
standard
System
univariate
facilities
Statistics
were
(SAS, analysis
Cary, using
unpaired t-tests for comparison of the means of groups with and without stress fractures. For comparison of rates of morbidity we used the chi-squared or the Fisher exact test. To determine the interdependence analysis we used multiple logistic
C. Milgrom, A. Finestone,
783 recruits
to reveal stress fractures (Milgrom were made of weight, height, tibial
calculated
study we followed 14 weeks of basic
Corps
designed Records
of
METhODS
Medical
previous fracture.
evaluation MATERIALS
WIENER
regarding of stress
high known.
age-dependency
and Lindgren
Forces
intercondylar
has
some insight into the causes of these injuries 1991). In engineering structures it is clear
M.
or had incomplete data. The remaining their informed consent to our study.
isometric Study
SIMKIN,
of the results of univariate regression models of the
variables that were significant on univariate analysis. The results for these models were summarised by odds ratios for the factors of interest, including 95% confidence intervals.
RESULTS Military
to Professor of Bone
P0
Box
02191,
C. Milgrom. and Joint
Surgery
Israel.
The Only basic
mean
age of the 783
26 were training.
symptoms
over A
suggestive
recruits
19 years total of
was
18.6
years
(17 to 26).
of age or older at the start of 35 1 recruits presented with
of stress THE JOURNAL
fractures OF BONE
and all had scintiAND JOINT
SURGERY
YOUTH Table
1. Variables
related
IS A RISK
to incidence
of stress
18.58
Height
(mm)
Tibial
175.4
length
(mm)
(n
Table stress
hip rotation
(degrees)
6.9 (p
II. Summary of multiple fractures at all sites
logistic
±
176.6 0.027)
±
38.9 0.049)
±
=
12.0 60.9 (p = 0.059)
±
regression
analysis
Age (yr)
38.4
of age.
The
most
physeal
plates.
shown
to
incidence
12.0
results
summarised
±
Table III. tibial stress
for
12.0
Summary fractures
60.7
±
176.4
(p
=
0.588)
2.3 (p
=
0.804)
38.7
12.0 60.5 (p = 0.626)
of multiple
logistic
±
6.6
±
2.5
±
12.0
regression
analysis
0.958
to 1.045
1.013
0.999
to 1.028
External
1.018
1.002
to 1.034
0.986
0.881
to 1.103
Tibial
0.905
0.802
to 1.022
0.716
0.531
to 0.965
Age (yr)
0.754
0.541
to 1.050
limits
the tibia,
the
followed
None evidence
results
for
significant
the
of the of open
logistic
fractures
regression
in Table
fractures in Table III. For each age from 17 to 26 years the risk
hip
rotation
length
(mm)
are
II and for tibial
year of increase for stress fracture
in at
all sites rotation
decreased by 28%. For each 1#{176} increase of the hip, the risk for tibial stress fracture
in external increased
by 2%.
The
showed
(degrees)
limits
geometry which governs their ability to withstand and torsional forces, and the mechanical properties bone tissue (Heaney in the age dependency Milgrom
to
studies
(mm)
confidence
for
during infantry training. In this context, the biomechanics of long bones can be viewed as a combination of their overall
variables
relationship
analysis.
for all stress
12.0 61.0 (p = 0.026)
±
7.2
0.69
±
1.000
the metatarsals. taken showed
of multiple
±
38.8
2.4
±
0.43 18.68 (p=O.l26)
Height
site was
I gives
63.4
175.9
6.6
±
to 1.015
a statistically
by univariate
The stress recruit
Table
have
38.9 0.009)
±
=
2.1
±
18.58
0.938
common
by the femur and then scintigraphs or radiographs
176.6 0.054)
±
=
(p
graphy. This was never performed on asymptomatic recruits. On the basis of these results 190 (24%) were diagnosed as having stress fractures, but none was seen in those over 19 years
0.8 (p
±
0.72
±
to 999.000
0.976
(mm)
175.4
0.44 18.69 (p=O.04l)
None (n = 723)
0.147
4.839
length
2.4
±
Fracture (n = 60)
643)
=
629.533
999.000
Tibial
6.6
18.59
(n
Intercept
Intercept
(degrees)
Femur None
to 999.000
confidence
hip rotation
by the I-test)
95%
95%
External
21
Odds ratio
Odds ratio
(mm)
significance
SD,
±
FRACTURE
Variable
Variable
Height
0.74
±
=
2.2
±
62.8
STRESS
Fracture (n = 140)
593)
=
0.44 18.70 (p=O.009)
±
(p External
(mean
None
±
38.5
FOR
Tibia
Fracture (n = 190)
Age (yr)
fracture
sites
All
Variable
FACTOR
tibial in the in this inertia, radius moment
1992). Both factors which we found.
et al (1989)
showed
may
that stress
bending of the
be implicated fractures
of the
diaphysis are usually due to repetitive bending forces mediolateral plane. The bending strength of the tibia plane is related to the cross-sectional area moment of which is proportional to the fourth power of the of the tibia. A narrow tibia with a low cross-sectional of inertia in the mediolateral plane was shown to be
a risk factor; this accords with Bone density is a measure
engineering of the mass
no
related
(Heaney
of
shown, however, to be a relatively poor predictor of fractures in the elderly and is only a risk factor (Ott 1992). This may
Within our study group 87% ofthe recruits were Israeliand 13% were born overseas, including 18 (2.3%) of Ethiopian origin. Ethiopian recruits had no stress fractures,
be because bone porosity, rather than density, is directly related to bone fragility (Schaffler and Burr 1988). In Israeli infantry recruits bone density does not correlate with stress
giving
fracture
risk
density
in this population
other
multiple
factors
femoral,
logistic
with
tibial,
regression
significant
or all-site
effects
stress
models on
the
incidence
fractures.
born
a statistically
compared p
=
with There
0.006).
Israeli-born
significant
Israeli-born was
reduction
recruits
no
significant
and non-Israeli-non-Ethiopian
in their
(Fisher’s difference
incidence
exact
test;
between
recruits.
results
indicate
for the likelihood VOL.
76-B,
No. 1. JANUARY
that
age
is a strong
of the development 1994
predicting
of stress
Bone
material
strength
(Leichter quality
et al 1989), also
perhaps
is within contributes
1992).
because
the normal to the
which
It has
is
been
the bone
range.
strength
of bone
(Grynpas 1992; Heaney 1992) and factors in this include its micro-architecture, matrix, mineralisation and the influence of microfatigue damage. We could find no studies of the relationship between bone quality and stress fracture.
DISCUSSION Our
to
principles. density,
variable
fractures
We found of 24%,
an incidence
corresponding
of scintigraphic
to that
in previous
stress
fracture
Israeli
studies,
C. MILGROM,
with
most
in the
All
tibia.
351
recruits
A. FINESTONE,
who
(45%)
N. SHLAMKOVITCH,
had
LET AL
It is unknown
whether
the
in risk for stress 1 7 to 26 years
fracture
bones
(Giladi et al 1987) but our measurement of another factor, high external rotation of the hip, confirmed the results of
studied
were
mature,
the
related to the fact that bones not reached full structural 25 000
radiographs
and
decreasing
of younger maturity. showed
risk
may
be
recruits had probably Garn (1970) studied
that
there
cortical bone width et al (1986) showed
well into the third that a 14-week period
infantry Leichter
training increased bone mineral content by 8%, and et al (1989) also reported an increase in bone
density during infantry training. (1986) showed that stress fractures of North American thoroughbreds
younger animals. Butterweck and demonstrate racehorses
of
these
Giladi
decade. of basic
higher
in younger
did
know
no
of
Ethiopian generally
recruit shorter,
not
cross-sectional or a combination
measure
tibial
bone
that Ethiopian recruits than their Israeli-born
width
recorded
case
of
stress
had a lower counterparts. fracture
to be protective
of these
clarify
other
benefits
racial
factors
in any
in
in the past four years. These Ethiopians lighter, able to run 2 km in less time,
lower upper-body strength than Israeli-born also have less external rotation of the hip,
been shown
No
aged
et al (1991).
We
study
were
We
indicates fracture
In a subsequent study, Nunamaker, Provost (1990) used strain gauges to strains
factors.
risk
have They
for recruits
bone density, increased changes in bone quality
Our study for stress
In horses, Nunamaker of the third metacarpal were more common in
that compressive during
is due to increased moment of inertia,
is a continued
gain in Margulies
of 28% per annum
decrease
scintigraphy had closed tibial and femoral epiphyses. In this population, both univariate and multivariate analysis showed that age was an important risk factor, with a decrease in incidence of 28% per added year of age. Although the long
form
for tibial stress
differences related have
may
to stress been
recruits. which has
fractures.
help
an are but
Further
to identify
and
fracture.
received
or will
be received
from
a
running. commercial
party
related
directly
or indirectly
to the subject
of this article.
REFERENCES Cohn
Garn
SH, Aloia JF, Vaswani AN, et al. Age bone mass measured by neutron activation. Makin M, Steinberg R, eds. Osteoporosis. Sons, 1982:33-43. SM.
The earlier gain perspective. Springfield,
and
the later Ill: Charles
and sex related changes in In: Menczel J, Robin GC, Chichester, etc: Wiley and
loss ofcortical C. Thomas,
bone: 1970.
in nutritional
Giladi
M, Milgrom C, Simkin A, et al. Stress fractures and tibial width: a risk factor. J BoneJoint Surg [Br] 1987; 69-B:326-9.
Giladi
M, Milgrom C, Simkin A, Danon Y. Stress risk factors. Am J Sports Med 1991; 19:647-52.
Grynpas MD. Workshop Bethesda,
Stress
fractures:
editorial.
Leichter I, Simkin A, Margulies following strenuous physical Lindahl
identifiable
0, Lindgren and density with
Lancet
1986;ii:727.
JY, et activity.
al.
Nunamaker DM. The bucked shin complex. Convention of AAEP, 1986; 457-60.
of the amount 38:133-40.
DM, Provost with
intense limbs
physical of young
of the
Proc
of the
MT.
Fatigue fractures peak bone strain,
age,
32nd
Annual in and
SM. When bone mass fails to predict bone failure: evidence for differences in bone quality. Workshop on Aging and Bone Quality. National Institute of Health, Bethseda, Maryland, 1992:15.
Schaffier MB, Burr DB. Stiffness of compact density. J Biomech 1988; 21:13-6.
bone:
effects changes bone.
Wall
JC, Chatterji 5K, and tensile strength 1979; 27: 105-8.
Jeffery JW. Age-related of human femoral cortical
Zwas
ST, bone
R, Frank G. Interpretation findings in stress fractures.
Gain in mass density of bone J Orthop Res 1989; 7:86-90.
AG. Cortical bone in man: 1 . Variation age and sex. Acta Orthop Scand 1967;
DM, Butterweck
thoroughbred racehorses: relationships training.JOrthopRes 1990; 8:604-11.
Ott Worshop Bethseda,
I, et al. Effect of content in the lower 1986; 68-A: 1090-3.
Milgrom C, Finestone A, Shlamkovitch N, et al. The clinical assessment of femoral stress fractures: a comparison of two methods. Military Medicine 1993; 158:190-2.
of bone. In: of Health,
in fragility fractures? Institute of Health,
JY, Simkin A, Leichter on the bone-mineral J BoneJoint Surg [Am/
Milgrom C, Giladi M, Simkin A, et al. The area moment of inertia tibia: a risk factor for stress fractures. J Biomech 1989; 22:1243-8.
Nunamaker
Age and disease related changes in the mineral on Aging and Bone Quality, National Institute Maryland, 1992:22.
Heaney RP. Is there a role for bone quality on aging and bone quality. National Maryland, 1992:14. Lancet.
fractures:
bone
Margulies activity adults.
Elkanovitch scintigraphic
of porosity
and
in the density Ca/cit Tiss Jnt/
and classification J NucI Med
of 1987;
28:452-7.
ThE
JOURNAL
OF BONE
AND
JOINT
SURGERY
