Osteoporosis Int (1992) 2:201-204 © 1992 European Foundation for Osteoporosis
J"~kA
0
osteoporosls International
Original Article Earlier Wrist Fracture: A Confounding Factor in Distal Forearm Bone Screening K. A k e s s o n , P. Gfirdsell, I. Sernbo, O. Johnell and K. J. O b r a n t Department of Orthopaedics, Malta6 General Hospital, Lund University, Malm6, Sweden
Abstract. Thirty-six women aged 60, 70 or 80 years who had fractured one of their distal radii (Colles' fracture) 0 to 35 (median 10) years earlier were examined in this population-based study. Single photon measurements (SPA) were performed on both arms 1 cm and 6 cm proximal to the sty!oid process of the ulna. All women were questioned about earlier wrist fractures and which, if any, side had been affected. Information about the type and site of the fracture was also gathered from the hospital records. The bone mineral content (BMC) was found to be increased by almost 20% in the oncefractured radius at the distal measuring site (1 cm) when compared with the unaffected side. This difference did not seem to diminish with time. At the proximal measuring site there was no difference between the once-fractured and the non-fractured side. There was a progressive loss of bone mineral in the once-fractured arm during the first years after the fracture however. This appeared to be reversed after about 10 years to a relative gain, even though the correlation was weak. Six of the 36 women could not remember which side had been fractured and five could not remember having had such a fracture at all. Determination of osteoporosis by measuring forearm BMC with single photon densitometry is therefore of limited value in some women of the oldest age groups. Keywords: Bone; Distal radius; Fracture; Single photon absorptiometry
Introduction Single photon absorptiometry (SPA) using either 1251 [1] or ~ l A m [2] has, since the beginning of the 1960s, produced an enormous amount of data on the bone mineral content (BMC) in the distal radius. These data have been used for screening for osteoporosis [3,4] and for prediction of patients susceptible to future fragility fractures [5,6]. Fracture of the distal radius is one of the fragility fractures [7] more common in women and its occurrence increases with age [8]. This fracture should therefore occur more frequently in osteoporotic women than in healthy controls. Our clinical impression is that women who have sustained this type of fracture many years ago have often either forgotten which side was affected or totally forgotten having had a fracture of the wrist. The typical fracture of the distal radius, i.e. Colles' fracture, affects the most distal few centimetres of the bone. The normal fracture healing processes with callus formation and altered normal macro- and micro-anatomy of the trabecular bone at this location may thus influence the reliability of SPA measurements at this level when screening for osteoporosis. The present population-based study was therefore designed to relate bilateral SPA measurements of the mineral content at the 1 cm and 6 cm levels of the distal forearm to ascertained information about previously sustained wrist fractures. This information was made possible by the unique health care system in Malm6, which includes saving of all radiographs.
Patients and Methods Correspondence and offprint requests to: Dr K. Akesson, Clinical
Tutor, Department of Orthopaedics, Malm6 General Hospital, S-214 01 Maim6, Sweden.
Women aged 60, 70 or 80 years who were living in their own homes were randomly selected from the city
202
register and invited to participate in the study. Sixty of 83, 85 of 120 and 48 of 96 women participated from each age group, respectively, resulting in a mean age of 73 years. BMC was measured by a single photon apparatus using 241Am as the energy source [2]. Transverse scans were made of both arms, radius and ulna, 1 cm and 6 cm proximal to the tip of the styloid process of the ulna, thus giving a mainly trabecular value at the 1 cm level and a mainly cortical value at the 6 cm level. The BMC was calculated and expressed as the average thickness of bone mineral in the pathway of the beam (mg/cm2) [9]. The precision from our SPA apparatus is around 4%5% in relatively osteoporotic subjects 60-80 years old. The women were asked in advance about previous fracture history by means of a questionnaire. Later they were asked specifically whether they had had a previous wrist fracture and, if so, which side had been affected. The referrals and the reports from the Department of Diagnostic Radiology at Malta6 General Hospital were searched for information on the study participants. This was possible because everyone in Sweden has a personal identification number. Also the health care system in Malta6 is unique as all radiographs since the beginning of the century have been saved. Malm6 has only one department of orthopaedics and virtually only one department of diagnostic radiology, both of which are at Maim6 General Hospital. At the Department of Diagnostic Radiology all radiographs from 1950 onwards have been coded according to fracture type. This system, unchanged today, enabled us to identify the type and side of the wrist fracture in these women. From the three age groups respectively, 7 of 60, 17 of 85 and 17 of 48 women had had a fracture of the wrist according to the radiological records. Five of these were excluded for further investigation: two cases of bilateral fractures, one who was a child at the time of the fracture, one who had had a fracture of the styloid process as the only fracture, and one case in which the side of the fracture could not be judged with certainty. Two of the 36 remaining patients had had repeated fractures but on the same side and were therefore included in the follow-up. In a few patients one measuring site is missing for technical reasons. For statistical evaluation the Student's t-test and a simple regression equation were used.
K. ~ k e s s o n et al.
The time elapsed between the fracture and the followup was 0-35 years (mean 11 years), and there was no difference in terms of time elapsed from fracture between those who later remembered or not. The right/left distribution of the fractures was 17/19. The right hand was dominant in all the women. The BMC of the distal radius at the 1 cm level was significantly higher in the once-fractured arm than in the non-fractured arm (228 + 65 mg/cm2 versus 192 + 52 mg/cm2, p = 0.0003), while at the 6 cm level no difference was seen between the once-fractured and non-fractured sides (381 + 58 mg/cm2 versus 382 + 59 mgc'cm2). When relating the time elapsed since fracture to the difference in BMC between once-fractured and nonfractured side no correlation was found at the i cm level of the radius (Fig. 1). Thus, several decades after fracture the once-fractured bone had, at the i cm level, a higher BMC than the non-fractured side. When comparing the fractured and non-fractured side at the 6 cm level of the radius a lower BMC was noted up to one BMC difference mg/cm 2
I cm l e v e l
iii1
•
® 60 years 0
°°5o~t v
O"
t"
-I0@
.
.
®
O@ o
o 13 (
.
.
0
.
.
.
I0
5
15
,
20
.
.
,
.
,
.
,
25 30 35 40 Years after fracture
BMC difference mg/cm2
level
6 cm
I O0.
o
0
'
0
50., 25. '
Eleven of the 36 women could not adequately remember a prior wrist fracture even after being questioned on two separate occasions. Five of these could not recall a fracture at all, while six remembered the fracture but not which side had been affected. The distribution by age group of those who had forgotten was 2/5 60-yearolds, 4/16 70-year-olds and 5/15 80-year-olds. At the time of fracture the mean age was 61 years (range 3677), and there was no difference in terms of age between those who later remembered or not.
.
Fig. 1. The difference between the bone mineral content of the fractured and non-fractured arm at the 1 cm level. T h e values are in relation to time elapsed after the wrist fracture. T h e single smallfilled circle represents two patients; the pair ofsmallfilled circles represents two patients aged 70 and 80 years.
75-
Results
0 80 years
0 ®
ok -50
• • 0
~
-
• 70 years
•
0
~
)
~
® 60 years
0
~ • 70 years
0
O®
0 80 years -25- • -50o ,e® -75.
0 ~1~
0
• o
-100 . . . . . . . . . . . . . . . . . 5 10 15 20 25 30 35 40 Years a f t e r fracture
Fig. 2. T h e difference between the bone mineral content of the fractured and non-fractured a r m at the 6 cm level. T h e values are in relation to time elapsed after the wrist fracture.
Fracture and Bone Mineral Content
decade after fracture (p = 0.029, r = 0.38) (Fig. 2). After this time a relative increase was seen. Pooling the data from all patients there was no significant difference between the BMC of the right and the left sides at either level. The difference in BMC at the 1 cm and 6 cm levels was not age-related, although there was a tendency, even if it was not statistically significant, to a decreased BMC with age at both levels and in both arms. There was also a correlation of the BMC at both the 1 cm and the 6 cm levels, respectively, when comparing fractured and nonfractured side (p = 0.111, r = 0.43).
Discussion Fracture of the distal radius, i.e. Colles' fracture, is a very common fracture and may precede other fragility fractures such as hip fractures [10]. The annual incidence of this fracture is well over 1% for women over 60 years of age [8]. It was therefore not surprising that so many women in this population-based study had sustained this type of fracture earlier in their lives. However, it was surprising that 30% of the women could not remember either having sustained this type of fracture or which side had been fractured. This inability to remember was not explained by an age difference either at the time of fracture or at follow-up between those who remembered and those who did not. All women were still living in their own homes and participation in the study required them to arrange for a visit to the hospital; thus we cannot assume any difference between them in cognitive function. These findings lead us to conclude that questioning alone is not sufficient for the identification of patients who have had an earlier forearm fracture. The apparently higher BMC (about 20% at the 1 cm level) in women aged between 60 and 80 years who have had a previous fracture corresponds to a 'bone age' more than a decade younger. As the 1 cm level in most cases is almost identical to the fracture site, the reason for the higher BMC there should be the callus formation and reorganization of the cortical and trabecular bone which evidently persists many years after fracture. Thus without knowledge of prior wrist fractures the BMC values obtained during screening procedures in this population may not accurately identify osteoporosis in individual women. At the 6 cm level the B M C was the same in both arms. However, an initial loss of bone mineral was seen immediately after fracture. This is in agreement with earlier findings of post-traumatic osteopenia in a oncefractured extremity [11,12] that is particularly pronounced in the once-fractured bone [13]. Westlin [14] has shown that the loss of bone mineral at the proximal measuring site during the first year after Colles' fracture amounts to 18%. Evidently this negative difference between the two sides persists for many years although a reversal to a positive difference seems to take place after a decade, even though the correlation is weak. This may
203
be explained by a greater loss with time due to a higher content on the non-fractured side. Nilsson and Westlin [15] presented similar data, finding a reversion of the post-traumatic osteopenia within 10 years. Their study was, however, limited to a time span of one decade after the fracture. We did not find a significantly higher BMC in the dominant arm and half of the women (n = 17) had a fracture of the dominant arm and half (n = 19) of the non-dominant arm. We therefore do not think that dominancy of one hand is of any relevance to the outcome of the study. It is not clear why there seems to be a higher BMC distally in the fractured arm as compared with the nonfractured arm many years after fracture, but it could be due to a persistent alteration of the biomechanical loading and rebuilding of bone according to Wolff's law [16]. We conclude that measurement of the BMC by single photon densitometry may be of doubtful value when screening for osteoporosis in specific individuals, particulary if measured on only one side in elderly women. The reason being that a large percentage of elderly women do not remember either having sustained a wrist fracture or which side was once fractured, and this forgetfulness cannot be accounted for only by age or time elapsed from fracture. This study points to the fact that a previous wrist fracture clearly affects the local BMC even in the long term and thus should not be disregarded as a confounding factor in SPA measurement of the distal forearm. When screening populations, especially of younger subjects, for osteoporosis for research purposes, though, the effect of previous fractures should not be of significant importance.
References 1. Cameron JR, Sorenson J. Measurement of bone mineral in vivo: an improved method. Science 1963;142:230-2. 2. Naucl6r LOW, Nilsson BE, Westlin NE. An apparatus for gamma absorptiometry of bone: technical data. Opuscula Medico-Technica Lundensia 1974:12. 3. Christiansen C, Rodbro P. Estimation of total body calcium from the bone mineral content of the forearm. Scand J Clin Lab Invest 1977;35:425-37. 4. Ott SM, Kilcoyne RF, Chestnut CH. Ability of four different techniques of measuring bone mass to diagnose vertebral fractures in postmenopausal women. J Bone Miner Res 1987;2:201. 5. Wasnich RD, Ross PD, Heilbrun LK, Vogel JM. Prediction of postmenopausal fracture risk with use of bone mineral measurements. Am J Obstet Gynecol 1985;153:745-51. 6. G~irdsell P, Johnell O, Nilsson BE. Predicting fractures in women by using forearm bone densitometry. Calcif Tissue Int 1989;44:235-42. 7. Obrant KJ, Bengn6r U, Johnell O, Nilsson BE, Sernbo I. Increasing age-adjusted risk of fragility fractures: a sign of increasing osteoporosis in successive generations? Calcif Tissue Int 1989;44:157-67. 8. Bengn6r U, Johnell O. Increasing incidence °ff°rearm fractures" A comparison of epidemiologic patterns 25 years apart. Acta Orthop Scand 1985;56:15860. 9. Nilsson BE, Johnell O, Petersson C. In vivo bone-mineral measurement. How and why: a review. Acta Orthop Scand 1990;61:275-81.
204 10. AlfframP-A. An epidemiologicstudyofcervicalandtrochanteric fractures of the femur in an urban population. Analysis of 1664 cases with special reference to etiologic factors. Acta Orthop Scand 1964;Suppl 65. 11. Nilsson BE. Post-traumatic osteopenia. A quantitative study of the bone mineral mass in the femur following fracture of the tibia in man using americium-241 as a photon source. Acta Orthop Scand 1966; Suppl 91. 12. Nilsson BE, Obrant K. Post-fracture changes of the femur cortex. Acta Orthop Scand 1983;54:862-4.
K. Akesson et al. 13. Obrant KJ, Nilsson BE. Histomorphologic changes in the tibial epiphysis after diaphyseal fracture. Clin Orthop 1984;185: 270-5. 14. Westlin NE. Loss of bone mineral after Coltes' fracture. Clin Orthop 1974;102:194-9. 15. Nilsson BE, Westlin NE. Long term observations on the loss of bone mineral following Colles~ fracture. Acta Orthop Scand 1975;46:61-6. 16. Wolff J. Das Gesetz der Transformation der Knochen. Berlin: A Hirschwald, 1982.
Received for publication 28 October 199l Accepted in revisedJbrm 20 January 1992
Book Review Textbook of Internal Medicine, Volume 1. Edited by William Kelley. J P L i p p i n c o t t , P h i l a d e l p h i a , P A . P a g e s : 2512. Price: $99.00 US. I S B N 0 - 3 9 7 - 5 1 0 4 8 - 9 . Reviewed by Robert Lindsay, MD, PhD, Chief, Internal Medicine, Helen Hayes Hospital, New York, USA. In this second edition, large textbook of medicine, presented as a single volume, each system is dealt with in three basic sections. Firstly, the scientific bases of the problems in each system are reviewed. In the second section the specific disorders are described. In the third section the specific approaches to the diagnosis and management of disorders, appropriate to that system, are reviewed. In each case the editor has assembled an impressive array of highly qualified clinicians and this is especially true in the areas of calcium and phosphorus metabolism and osteoporosis. Osteoporosis in the elderly is recognized to be a sufficiently large clinical problem to warrant its own chapter in the geriatric section. Dr. Kenneth Lyles, the author of the chapter, has done an excellent job of summarizing the current status of osteoporosis presentation, differential diagnosis, and management in an effective way and includes a brief but clinically useful section on the use of bone mass measurement in the management of the condition. Specific medical interventions are reviewed in reasonable detail. However, doses are not given, either for calcitonin or estrogen, and advice about the management of patients on either of these medications is omitted. I particularly liked the sections on rehabilitation of the
elderly and the role of falls in the elderly in fractures, which provide insights into osteoporotic fracture that are not commonly seen in general textbook of medicine. All in all, this is an extremely welt written textbook full of sound clinical advice about patient management - although it has the same problems as all major textbooks of medicine in that its size and the necessity to cover what has now become such broad topic results in a somewhat superficial discussion of some areas. As far as learning about osteoporosis is concerned, I would heartily recommend the chapter on osteoporosis in this textbook to the medical student, but for the resident and fellow in medicine or endocrinology who wishes for further information on osteoporosis and metabolic bone disease, it would be necessary to utilize further reading materials such as the primer on metabolic bone disease prepared by the American Society for Bone and Mineral Research. In the fast-moving world of internal medicine this textbook represents an authoritative view of a broad field with several contributions that can be clearly labeled as outstanding. It is amazing that the book, which first appeared in 1989, has in only two years been updated to the extent that 20% of the material is entirely new or has been totally rewritten. The addition of a cross-referencing system, allowing the reader of one chapter to find easily in other areas of the textbook situations in which the same topic or a closely related topic is dealt with, is a useful addition that is to be heartily commended. Under the tutorship of William Kelley this textbook is rapidly becoming the standard in printed volumes on internal medicine for the medical student.
