Journal of Clinical Densitometry, vol. 3, no.

4, 383–389, Winter 2000

© Copyright 2000 by Humana Press Inc.
All rights of any nature whatsoever reserved.
0169-4194/99/2:383-389/$11.75

Original Article

Change in Bone Mass After Colles’ Fracture

A Case Report on Unique Data Collection and Long-Term Implications

Jasminka Z. Ilich, PHD,RD,1 Michael Zito, MS,PT,1 Rhonda A. Brownbill, RD,1

and Michael E. Joyce, MD2
1School of Allied Health, University of Connecticut, Storrs, CT; and 2University of Connecticut Health Center,
Farmington and Connecticut Sports Medicine and Orthopaedic Center, Storrs, CT

Abstract
The purpose of this case report was to describe changes in bone mass in different skeletal sites triggered
by Colles’ fracture. The case is unique regarding the existence of baseline measurements taken just a few days
before the fracture on all measurable skeletal sites, including the fractured radius. The patient was a healthy,
premenopausal woman, who fractured her non-dominant wrist after a fall. The arm was immobilized to the
elbow for 5 weeks. Bone mass was measured at baseline, 5, 10, 13, 21, and 52 weeks post-injury. Upon plas-
ter removal, there was a notable increase in bone mass in all sites of ulna and radius of the injured forearm
(10%–73%), followed by the apparent decline to or below the baseline at all follow-up measurements. Other
skeletal sites were measured at 10 weeks when substantial decrease in spine and hip bone mass was noticed,
most notably in L3–L4, Ward’s, and femoral neck (2%–8%), and remained such after 1 year. Although this
patient had no previous osteopenia, the trauma caused long-standing bone loss in fracture-prone areas. The
changes in bone mass after wrist fracture should be monitored carefully and more elaborate treatment might
be needed to prevent any potential risk for later fractures.

Key Words: Colles’ fracture; bone mineral density; immobilization; dual X-ray absorptiometry.

Introduction other parts of the fractured bone as well as in the

It is generally accepted that cast immobilization
of a fractured limb results in a reduction in bone
mass, which usually rebounds several months to a
year after recovery (1). The loss of bone is typically
attributed to a combination of disuse of the affected
limb and increased rate of bone turnover in response
to fracture. However, it seems that skeleton responds
to fracture with an increased rate of bone turnover in
Received 05/06/00; Revised 08/07/00; Accepted 08/11/00.
Address correspondence to Jasminka Ilich, University of
Connecticut, School of Allied Health, 358 Mansfield Road U-
101, Storrs, CT 06269. E-mail: ernst@uconnvm.uconn.edu
adjacent bones. Such changes were first demon-
strated in rats (2) and men after tibial fractures (3).
The net result of these changes shortly after immo-
bilization is usually a loss of bone mass, and it is
questionable whether the initial bone mass is
restored and what the long-term consequences for
other skeletal sites are, especially when a middle age
or older patient is involved.
Several studies have reported lower overall bone
mass in patients with Colles’ or other wrist fractures
(4–8) and tibial fractures (9,10) compared to their
own contralateral limb or controls. However, the
bone mass measurements in injured or noninjured

383
384
skeletal sites are usually performed after the injury.
Therefore, it is not always clear whether the patients
with fractures had lower bone mass to begin with, or
whether the trauma of fracture caused some bone
loss when presented at 4–6 wk after injury; the time
measurements are typically taken. In view of this,
the changes in bone status after a fracture and possi-
ble long-term or even permanent loss of bone mass
in the affected or unaffected skeletal sites need to be
elucidated.
The purpose of this study was to prospectively
describe changes in bone mass in different skeletal
sites triggered by the fracture of the distal radius.
The case is unique regarding the existence of base-
line measurements taken just a few days before the
fracture in all measurable skeletal sites, including the
fractured bone. Therefore, it was also possible to
determine whether the injury caused long-term bone
loss in the affected and unaffected skeletal sites.
Materials and Methods
We present a case of a healthy, Caucasian, pre-
menopausal woman, in her late forties, who frac-
tured the distal radius of her left, nondominant arm
(Colles’ fracture) after a low-energy impact fall on
ice from a standing height (classified as low trauma).
The fracture was confirmed by X-ray and clinical
symptoms as minimally comminuted and nondis-
placed. The patient had no previous disease, injury,
or medications that might have affected her skeletal
status. This case is unique because just a few days
before the fracture incident, complete bone mass
measurements were performed and all measurable
skeletal sites were evaluated, providing for the accu-
rate and timely baseline data. The subject was then
followed for an entire year. Informed consent was
obtained.
On onset of fracture, the arm and the metacarpals
were immobilized to the elbow for 5 wk. The patient
took calcium (Ca) with vitamin D supplements (630
mg/d of Ca and 400 IU/d of vitamin D) (Citracal®;
Mission Pharmacal, San Antonio, TX) from the time
of injury. Her total Ca intake, including dietary
(assessed by the food frequency questionnaire for
calcium [11]) and supplemental, was ~1500 mg/d.
However, the patient stopped the Ca supplements
several weeks after the fracture and remained on her
Journal of Clinical Densitometry
Ilich et al.
usual dietary Ca intake of about 900 mg/d. Activity
level (sedentary) did not change during the entire
period and was not different from that several
months before the fracture. Note that the patient was
gradually gaining weight, resulting in 11.3 kg gain
after 1 yr. Her menstrual cycle remained regular and
unchanged throughout.
Approximately 6 wk after injury the subject was
referred to physical therapy services. However,
owing to an ongoing inflammatory response of soft
tissues and their hypertrophic scarring, pain man-
agement, rather than therapeutic progression of the
subject’s motion impairments, was the goal over
the course of therapy. The full benefits attributable
to a progressive physical activity regimen were
therefore not obtainable, and physical therapy was
discontinued.
Bone mass measurements were performed with a
Lunar densitometer (DPX-MD, version 4.6e; Lunar,
Madison, WI) with specialized software for hand,
forearm, hip, spine, and total body. The instrument
was calibrated daily. Long-term stability was deter-
mined by measuring an aluminum spine phantom on
a weekly basis, with resulting coefficient of variation
(CV) of 0.49%. The %CV of bone mineral density
(BMD) for different skeletal sites in vivo was calcu-
lated from the repeated measurements (10 times
each) on three normal individuals. The %CV for the
BMD of different sections of the forearm (radius and
ulna pooled together) was as follows: ultradistal, 1.2;
5-mm separation distance, 2.7; 10% distance from
styloid process, 2.5; 33% distance from styloid
process, 0.7. The %CV for other skeletal sites was
comparable with the manufacturer’s published val-
ues for precision, as well as with our previous mea-
surements (12,13) and other published data (14),
indicating acceptable precision of our instrument.
The %CV for BMD of selected sites was as follows:
hand, 0.7; lumbar spine (L2–L4), 1.0; femoral neck,
1.5; Ward’s triangle, 1.5; femoral shaft, 0.7;
trochanter, 0.9; and total hip, 0.6.
The baseline measurements were taken a few
days before injury in both forearms (radius and
ulna at ultradistal, 5-mm separation distance, 10
and 33% of the distance from styloid process), non-
dominant hand, femur (Ward’s triangle, trochanter,
neck, shaft, and total), anteroposterior lumbar
spine, and whole body. The subsequent measure-
Volume 3, 2000
Change in Bone Mass After Forearm Fracture 385

Fig. 1. Change in BMD of the fractured radius, presented as a percentage difference from baseline (calculated by
subtracting the baseline value from each subsequent measurement and dividing the difference by the baseline value and
multiplying by 100). RUD, radius ultradistal; R5mm, radius 5-mm separation distance; R10, radius 10% distance from
styloid process; R33, radius 33% distance from styloid process.

ments of the left forearm and hand were taken
immediately on removal of the plaster, 5 wk after
the injury, and then at 10, 13, 21, and 52 wk. Other
skeletal sites were measured at baseline and 10 and
52 wk postinjury. The analysis of each subsequent
scan for spine and hip was performed by superim-
posing it to the corresponding one from baseline
measurement, providing for the minimal analysis
error. The data are presented as a percentage differ-
ence from baseline (calculated by subtracting the
baseline value from each subsequent measurement
and dividing the difference by the baseline value
and multiplying by 100).
Results
Normal BMD values at baseline for all skeletal
sites (including the subsequently injured arm) were
detected, indicating very little or no osteopenic
changes. At the 5-wk measurement (on removal of
plaster), there was a notable increase in BMD and
bone mineral content (BMC) in all sites of the
injured forearm, whereas at 10 wk most of the other
skeletal sites revealed a substantial decrease. Figure
1–4 present some of the most apparent changes. The
BMD of the radius at ultradistal and 5-mm sites
increased from 0.361 to 0.554 g/cm2 and from 0.356
to 0.579 g/cm2 (53.5 and 62.6%), respectively, prob-
ably owing to the callus formation. Moreover, the
increase for the same sites at the ulna was from
0.238 to 0.397 g/cm2 and from 0.194 to 0.335 g/cm2
(66.8 and 72.7%), respectively, although the ulna
was not fractured. Likewise, the BMD of the radius
and ulna at 33% distance increased from 0.677 to
0.778 g/cm2 and from 0.767 to 0.839 g/cm2 (14.9
and 9.4%), respectively (Figs. 1 and 2). At the time
of further follow-up (10, 13, and 21 wk), there was
an apparent drop in BMD and BMC in the injured
forearm to or below the baseline values. For exam-
ple, at 21 wk BMD of the 10% distance and total
radius remained 14.2 and 5.0%, respectively, below
the baseline values. The same locations for ulna
showed a similar tendency although not of such a
magnitude. The BMD of the left hand was between
2 and 3% below the baseline at each measurement
and remained such after 1 yr. The right forearm,

Journal of Clinical Densitometry Volume 3, 2000

386 Ilich et al.

Fig. 2. Change in BMD of the ulna (not fractured), presented as a percentage difference from baseline (calculated by
subtracting the baseline value from each subsequent measurement and dividing the difference by the baseline value and
multiplying by 100). UUD, ulna ultradistal; U5mm, ulna 5-mm separation distance; U10, ulna 10% distance from styloid
process; U33, ulna 33% distance from styloid process.

however, showed some increase in BMC and BMD,
compared to baseline (e.g., from 5 to 10% [BMD] in
some regions of radius [except in 33% distance],
while the ulna remained the same or slightly
decreased at 10 wk. The same trends remained evi-
dent at 52 wk (Fig. 3).
Other skeletal sites were measured at the 10-wk
interval, at which time the BMD and BMC in some
of the hip regions and lumbar spine were substan-
tially below the baseline and remained such
throughout (Fig. 4). Because spine and hip were not
measured at 5 wk, we have no way of knowing if
the drop in BMD and BMC in these regions was at
its bottom at 10 wk or had already begun ascend-
ing. The biggest loss occurred in L3–L4 BMD and
remained after 1 yr (5.8%). In the hip region, BMD
decreased the most in the Ward’s triangle and neck
and the resulting deficit remained evident after 1 yr,
whereas the shaft and total hip recovered after the
initial drop and increased slightly above the base-
line (Fig. 4). Other regions and whole-body mea-
surements (not presented) had less variations or
remained the same.
Discussion
We report a unique case regarding the existence of
accurate and timely baseline data enabling the com-
parison of all measured skeletal sites after the fracture
to the same ones before the fracture. Most of the stud-
ies evaluating the effect of fractures are retrospective,
and the bone density before the injury in affected or
nonaffected skeletal sites is not known. Usually, the
conclusions about the changes in bone mass in affected
limb or other skeletal sites are based on comparison
with the nonaffected limb or with an age-matched ref-
erence population (15). Therefore, it is uncertain
whether the difference in bone mass before and after
the fracture already existed or was created subse-
quently from the atrophy of the affected limb or the
hypertrophy of the other limb or both. Furthermore, it
is not possible to know about the changes in bone mass

Journal of Clinical Densitometry Volume 3, 2000

Change in Bone Mass After Forearm Fracture 387

Fig. 3. Change in BMD of the nonfractured radius, presented as a percentage difference from baseline (calculated by
subtracting the baseline value from each subsequent measurement and dividing the difference by the baseline value and
multiplying by 100). RUD, radius ultradistal; R5mm, radius 5-mm separation distance; R10, radius 10% distance from
styloid process; R33, radius 33% distance from styloid process.

Fig. 4. Change in BMD of the different regions of hip and lumbar spine (L2–L4), presented as a percentage differ-
ence from baseline (calculated by subtracting the baseline value from each subsequent measurement and dividing the dif-
ference by the baseline value and multiplying by 100).

in other, noninjured skeletal regions and whether they
were affected by the trauma of fracture.
Studies have shown that the incidence of Colles’
fracture is higher among women with lower bone
mass (8,16) and that the fracture per se is accom-
panied by increased risk for future hip fractures
(17). Interestingly, in our case, the patient’s base-
line values in the affected arm or in any other skele-
tal site could not be defined as osteoporotic and not
even osteopenic (per World Health Organization
definition) (18), and that is particularly true for the
site of the fracture (distal radius, with T-score at
about –0.5). Nonetheless, the fracture occurred
even with what is defined as minimal trauma. This
supports the notion that other factors besides BMD,
such as bone quality, including microarchitectural
structure, have an important role in fracture risk.
In our study, we clearly demonstrated a very high
transitional increase in BMD and BMC in the
injured radius and adjacent ulna, varying from 10 to
73% in different regions. This increase was noticed
5 wk after the injury or immediately on removal of

Journal of Clinical Densitometry Volume 3, 2000

388
plaster and then followed by a rapid decline.
Furthermore, it was not localized only in the callus
region of the radius but was present in the whole
forearm including both the radius and ulna.
At all subsequent measurements (10, 13, 21, and
52 wk), there was an apparent decline in BMD and
BMC to or below the baseline level in the injured
arm, as well as in some other skeletal sites, except in
the contralateral arm. Most notable was the decline
in the regions of the hip (Ward’s triangle and neck)
and spine, reflecting mostly trabecular bone (Fig. 4),
presumably owing to the healing process in the frac-
tured site. The remodeling activity in trabecular bone
is higher than in cortical bone and this loss of bone
mineral in trabecular bone could have been a
response to injury. It would not have been possible to
notice this decrease in bone mass (from 2 to 10%)
without the existence of the baseline measurements.
Other studies suggest similar but more extreme
patterns. Westlin (19) could not demonstrate any
restoration of the bone mineral lost from the radius
and ulna 1 yr after the fracture, except as a very slow
process in the ensuing years. Comparison with the
contralateral arm showed 18% loss in BMD. Nilsson
and Westlin (20) noted (also based on the side-to-
side comparison) the maximum bone loss of 20% in
the radius of peri- and postmenopausal women who
sustained a Colles’ fracture. Mallmin et al. (21)
examined 74 patients with Colles’ fracture 2 mo after
injury and found an average decrease of 11% in bone
density of the injured arm compared with matched
controls. Restoration of bone mass was not noticed
in any of these studies.
Another factor that needs to be considered in eval-
uating bone loss after fracture is the extent of immo-
bilization and timing of the full use of a limb after
immobilization. A study by Houda et al. (22) sug-
gests that immobilization-induced bone loss contin-
ues for weeks after the cessation of immobilization;
therefore, it may cause an impairment in the bone
cell function that lasts longer than the time the limb
is actually in the cast.
It is evident that our study patient developed
some level of posttraumatic osteopenia, not only in
the injured arm but in the hip and spine. In many
regions bone mass was not restored to the baseline
level 1 yr postinjury, despite a weight gain of 11.3
kg. This is in accordance with some other studies
Journal of Clinical Densitometry
Ilich et al.
documenting (although indirectly by comparison
with reference population) that early response to
trauma leads to an increased bone turnover and loss
of bone mass not only at the site of fracture but at
other adjacent skeletal sites (23–25). A recent
analysis of the literature on the association between
prior and subsequent fractures showed that patients
with a history of any fracture have a higher risk for
developing new fractures (26). In addition, the
recent National Institutes of Health Consensus
Development Conference on Osteoporosis recom-
mends that any fracture in adults be evaluated for
assessment of possible osteoporosis in order to pre-
vent future risks and complications (27).
The changes in bone after injury and posttrau-
matic osteoporosis should be monitored and inter-
preted carefully, particularly when dealing with
middle age or older patients, because the conse-
quences of fracture could lead to the permanent loss
of bone mass. In our case, although the patient had a
normal bone mineral status and no osteopenia
detected before fracture, the trauma of fracture
caused long-standing bone loss, particularly in frac-
ture-prone areas—hip and spine. Because about 70%
of the variability in bone strength is explained by its
mineral density (28), this patient might be at
increased risk for fracture later in life. Therefore,
Colles’ fractures should be treated more aggressively
with respect to the patient’s follow-up, evaluation of
other skeletal sites, recommended exercises, and
prevention of future fractures.
Acknowledgments
This study was funded in part by the Donaghue
Medical Research Foundation (DF98-056) and the
University of Connecticut Office for Sponsored
Programs.
References
1. Hangartner TN. 1995 Osteoporosis due to disuse. In:
Physical Medicine and Rehabilitation, Clinics of North
America. Matkovic V, ed. WB Saunders, Philadelphia,
579–594.
2. Bohr H, Sorensen AH. 1950 Study of fracture healing by
means of radioactive tracers. J Bone Joint Surg 32A:567–570.
3. Andersson SM, Nilsson BE. 1979 Changes in bone mineral
content following tibia shaft fractures. Clin Orthop Related
Res 144:226–229.
Volume 3, 2000
Change in Bone Mass After Forearm Fracture
4. Mallmin H, Ljunghall S. 1994 Distal radius fracture is an
early sign of osteoporosis: bone mass measurements in a
population-based study. Osteoporos Int 4:357–361.
5. Harma M, Karjalainen P. 1986 Trabecular osteopenia in
Colles’ fracture. Acta Orthop Scand 57:38–40.
6. Krolner B, Tondevold E, Toft B, Berthelsen B, Nielsen SP.
1982 Bone mass of the axial and the appendicular skeleton
in women with Colles’ fracture: its relation to physical
activity. Clin Physiol 2:147–157.
7. Varenna M, Binelli L, Zucchi F, Beltrametti P, Gallazzi M,
Sinigaglia L. 1997 Is the metatarsal fracture in post-
menopausal women an osteoporotic fracture? A cross-sec-
tional study on 113 cases. Osteoporos Int 7:558–563.
8. Earnshaw SA, Cawte SA, Worley A, Hosking DJ. 1998
Colles’ fracture of the wrist as an indicator of underlying
osteoporosis in postmenopausal women: a prospective study
of bone mineral density and bone turnover rate. Osteoporos
Int 8:53–60.
9. Kannus P, Jarvinen M, Sievanen H, Oja P, Vuori I. 1994
Osteoporosis in men with a history of tibial fracture. J Bone
Miner Res 9:423–429.
10. Cattermole HC, Cook JE, Fordham JN, Muckle DS,
Cunningham JL. 1997 Bone mineral changes during tibial
fracture healing. Clin Orthop Related Res 339:190–196.
11. Ilich JZ, Shugor M, Hangartner T, Baoshe A, Matkovic V.
1998 Relation of nutrition, body composition, and
physical activity to skeletal development: a cross-
sectional study in preadolescent females. J Am Coll Nutr
17:136–147.
12. Ilich JZ, Hangartner TN, Skugor M, Roche AF, Goel PK,
Matkovic V. 1996 Skeletal age as a determinant of bone
mass in preadolescent females. Skel Radiol 25:431–439.
13. Ilich JZ, Hsieh LC, Tzagournis MA, Wright JK, Saracoglu
M, Barden HS, Matkovic V. 1994 A comparison of single
photon and dual x-ray absorptiometry of the forearm in chil-
dren and adults. Bone 15:187–191.
14. Augat P, Fuerst T, Genant HK. 1998 Quantitative bone
mineral assessment at the forearm: a review. Osteoporos Int
8:299–310.
15. Jarvinen M, Kannus P, Finland T. 1997 Current concepts
review: injury of an extremity as a risk factor for
the development of osteoporosis. J Bone Joint Surg
79:263–276.
Journal of Clinical Densitometry
389
16. Clement EP, Patka P, Vandormael K, Haarman H, Lips P.
2000 The effect of alendronate on bone mass after distal
forearm fracture. J Bone Miner Res 15:586–593.
17. Owen RA, Melton LJ, Ilstrup DM, Johnson KA, Riggs BL.
1982 Colles’ fractures and subsequent hip fracture risk. Clin
Orthop 171:37–43.
18. World Health Organization. 1994 Assessment of fracture
risk and its application to screening for postmenopausal
osteoporosis. Report of WHO Study Group Technical
Services, No. 843, Geneva, Switzerland.
19. Westlin NE. 1974 Loss of bone mineral after Colles’ frac-
ture. Clin Orthop 102:194–198.
20. Nilsson BE, Westlin NE. 1975 Long term observations on
the loss of bone mineral following Colles’ fracture. Acta
Orthop Scand 46:61–64.
21. Mallmin H, Ljunghall S, Naesson T. 1992 Colles’ fracture
associated with reduced bone mineral content: photon den-
sitometry in 74 patients with matched controls. Acta Orthop
Scand 63:552–554.
22. Houda J, Schulz LA, Morgan WJ, Breen T, Warhold L,
Crane GK, Baran DT. 1995 Bone mineral changes in the
forearm after immobilization. Clin Orthop 317:199–205.
23. Anderson SM, Nilsson BE. 1979 Changes in bone mineral
content following tibia shaft fractures. Clin Orthop Related
Res 144:226–229.
24. Eyers KS, Kanis JA. 1995 Bone loss after tibial fracture:
evaluated by dual energy x-ray absorptiometry. J Bone Joint
Surg 77(B):473–478.
25. Ulivieri FM, Bossi E, Azzoni R, Ronzani C, Trevisan C,
Montesano A, Ortolani S. 1990 Quantification by dual pho-
ton absorptiometry of local bone loss after fracture. Clin
Orthop 250:291–296.
26. Klotzbuecher CM, Ross PD, Landsman PB, Abbot TA III,
Berger M. 2000 Patients with prior fractures have an
increased risk of future fractures: a summary of the litera-
ture and statistical synthesis. J Bone Miner Res 15:721–739.
27. National Institutes of Health. 2000 Consensus Development
Conference Statement. Osteoporosis prevention, diagnosis,
and therapy. Draft document, March 2000 (http://consen-
sus.nih.gov).
28. Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook
PN, Eberl S. 1987 Genetic determinants of bone mass in
adults: a twin study. J Clin Invest 80:706–710.
Volume 3, 2000