Longitudinal Study of Cortical Bone Loss in Patients with Inflammatory Bowel Disease D. CLEMENTS, R. J. MOTLEY, W. D. EVANS, A. D. HARRIES, J. RHODES, R. J . COLES & J . E. COMPSTON @ep$ of Medici4and Medical Physics,cniversity Hospital of Wales, Cardiff, U . 3
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Clements D, Motley RJ, Evans WD, Harries AD, Rhodes J, Coles RJ, Compston JE. Longitudinal study of cortical bone loss in patients with inflammatory bowel disease. Scand J Gastroenterol 1992;27:1055-1060. c o n e mineral density of the radius was measured by single-photon absorptiometry in 50 patients with inflammatory bowel disease. Thirty-three had Crohn’s disease and 17 ulcerative colitis; 25 were women. The mean age was 45 years (range, 18-70 years). Measurements were repeated in 39 of them after a mean follow-up period of 7.9 years (range, 7.1-8.2 years). In female patients the mean (95% confidence interval) annual change in radial bone mineral density was -0.74% (-1.34% to -0.14%) ( P = 0.022), the greatest bone loss occurring in postmenopausal women (mean, -1.16% (-2.01% to -0.30%)). In male patients the mean annual rate of bone loss was -0.07% (-0.41% to 0.28%) ( P = NS). Patients with abnormally low values at the first measurement remained osteopenic at the second measurement, whilst some others with aormal values initially showed increased rates of bone loss and had a subnormal bone mineral density after the follow-up period. These results show increased rates of cortical bone loss in some patients with inflammatory bowel disease and emphasize the need to monitor bone mass in these patients so that prophylactic measures can be instituted. 4
Key words: Bone loss; cortical bone; inflammatory bowel disease; osteoporosis David Clements, Senior Registrar, University Hospital of Wales, Heath Park, Cardiff, U.K . CF4 4XW
Patients with inflammatory bowel disease have an increased prevalence of osteoporosis, with severe clinical manifestations in some who are relatively young (1-5). This increase in prevalence could be due to a low peak bone mass and/or an increased rate of bone loss. However, although increased rates of spinal trabecular bone loss have been demonstrated in some patients with inflammatory bowel disease (6), there have been no longitudinal studies of cortical bone loss in these patients. Between 1981 and 1982, 50 patients with inflammatory bowel disease had measurement of proximal forearm bone density at a site where bone is predominantly cortical. These measurements were repeated after a mean follow-up of 7.9 years. ,
SUBJECTS AND METHODS Patient details (Table I) Fifty outpatients, 25 women, with inflammatory bowel disease were studied between 1981 and 1982. Thirty-three had Crohn’s disease and 17 ulcerative colitis. In the 33 patients with Crohn’s disease the distribution of disease was as follows: diffuse small bowel, 4; terminal ileum, 17; ileum colon, 6; colonic, 4; and anorectal, 2. Nineteen of these patients had undergone one or more small-bowel resections. Of the 17 patients with ulcerative colitis, 4 had total colitis, 8 distal colitis, and 5 rectal disease only. Their age ranged from 17.8 to 69.8 years (mean, 45 years) at the
+
time of the first measurements. All the patients from the initial study were traced; 11 were unavailable for follow-up measurements. Eight had died (six men), and three (two men) declined t o return for another measurement. The mean follow-up period in the remaining 39 patients was 7.9 years (range, 7.1-8.2 years). Fourteen of the 25 women were postmenopausal; all of these were more than 47 years old, and only 1 patient (aged 60 years at the time of the second measurement) received hormone replacement therapy, which was given continuously for 6 years during the study. One patient was receiving calcium and vitamin D tablets, and 19 patients were receiving corticosteroid therapy. No patient had clinical or radiologic evidence of osteoporotic fracture. Single-photon absorptiometry Bone mineral content (BMC) was measured at the junction of the distal one-third and proximal two-thirds of the left radius, a site containing predominantly cortical bone, using an in-house single-photon absorptiometer based on the original design of Sorensen & Cameron (7). It incorporates a 1251 radiation source and is calibrated against a commercial instrument (Novo Densitometer GT35) so that BMC is expressed in units of g/cm, and the ratio of BMC to bone width (BMC/BW) in g/cm2. Long-term precision (coefficient of variation) was estimated from repeat measurements ( n = 41) of BMC/BW over a period of 5.2 years. In vitro precision for a n aluminium/perspex phanton was 2.1%,
1056
D. Clements et al.
Table 1. Details of patients at the time of the first measurement ~~
~
Women (n = 25)
Men (n = 25)
Age (years)
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Height (m) Weight (kg) BMC/BW (g/cm') BMC/BW (2-score)
Mean
Range
Mean
Range
43.7 1.69 64.5 0.78 -0.11
17.8 to 68.6 1.56 to 1.84 42.3 to 82.4 0.49 to 0.92 -3.60 to 2.40
46.5 1.59 61 .S 0.64 -0.37
18.0 to 69.8 1.48 to 1.85 37.8 to 98.8 0.42 to 0.80 -3.50 to 2.20
whilst in vivo precision was 2.7% for both a normal male (age 37 years) and female volunteer (age 29 years). The same instrument and technique were used for the original and the subsequent reading.
Data unalysis Changes in height, weight, and BMC/BW were calculated, and their statistical significance assessed by paired t test. The annual percentage change in BMC/BW was calculated for each patient. Each patient's result for BMC/BW was also converted to a Z-score (value with mean/standard deviation) using the data of Ringe et al. (8), who studied 773 normal
German volunteers, using a similar technique. These data are similar t o those of normal volunteers studied in Cardiff (9); the mean Z-score (SD) for Cardiff men ( a = 66) was 0.53 (1.41) and for women ( n = 63) +0.45 (1.61). The age and first Z-score of patients with inflammatory bowel disease were compared with Cardiff normals by means of the MannWhitney test; a similar comparison was made between male and female patients. RESULTS Figs. 1 and 2 show initial and follow-up BMC/BW values for
0.9.
0.80.7.
-
(u
0.6-
E
f
0
\
9 0.5-
3 !?
0
z
0.4-
Males
0.3-
Crohn's disease o Ulcerative colitis 0
0.2. 0.1.
-
0
20
40
60
80
160
Age (years) Fig. 1. Initial and follow-up values for bone mineral density of the radius in 25 male patients with inflammatory bowel disease plotted against the normal range of Ringe e t al. (8), shown as two standard deviations above and below the mean. BMC/BW = bone mineral content/bone width (g/cm').
Cortical Bone Loss in IBD
1057
0.E 0.E
f
0.7
-
cu
..
ox
\
5
.
\
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
9
0.5
3
e 5
0
0.4 \ \
Females
0.3
0
..
Crohn's disease Ulcerative colitis
0.2
0.1
0
0
20
40
60
80
100
Age (years) Fig. 2. Initial and follow-up values for bone mineral density of the radius in 25 female patients with inflammatory bowel disease plotted against the normal range of Ringe e t al. (8). shown as two standard deviations above and below the mean. BMC/BW = bone mineral content/bone width (g/cm2).
each subject, superimposed on the reference ranges of Ringe et al. (8). There was no significant difference in the first Z-scores between male and female patients. A t the time of the first measurement the mean Z-score (range) was -0.11 (-3.60 to 2.40) in the men and -0.37 (-3.50 to 2.20) in the women (Table I). There was no significant difference in age between the patients and Cardiff controls (9). The BMC/BW Z-score for men was not significantly different from that for controls ( P = 0.25), whilst for women it was significantly lower (P = 0.01). Altogether five patients (one woman) had Z-scores below -2.00 (Figs. 1 and 2); of these, four had Crohn's disease and one had ulcerative colitis. After a mean follow-up period of 7.9 years (7.1-8.2 years), the mean (range) Z-scores were -0.41 (-5.00 to 2.67) and -0.58 (-2.71 to 2.83) in men and women, respectively. Five women and four men had Z-scores below -2.00 at follow-up (Figs. 1 and 2 ) ; these included the woman and three of the four men with initially abnormal values, the fourth man having died before the second measurement. Of the five other patients with low follow-up values, two had ulcerative colitis and three Crohn's disease.
The mean (95% confidence interval) change in BMC/BW over the course of the study was -0.04 g/cm2 (-0.07 to 0.01) in women and -0.01 g/cm2 (-0.02 to 0.01) in men, representing annual percentage changes of -0.74 (-1.34 to -0.14)and -0.07(-0.41 to0.28),respectively.Thisachieved statistical significance for women ( P = 0.022) but not for men. Figs. 3 and 4 show the cumulative absolute changes in BMC/ BW in men and women after ranking the patients by age. The mean annual decrease in the postmenopausal women ( n = 14)was -1.16% (-2.01% to -0.30%).Themean(95%confidence limits) fall in Z-scores was -0.25 (-0.70 to 0.21) for women and -0.27 (-0.74 to 0.19) for men. Of the three women with only an initial measurement, two died and one declined follow-up. All had Crohn'sdisease, and their ages at the time of the first measurement were 30.5,49.6, and 59.2 years. The BMC/BW was within 1SD of the normal mean value in all three cases. Eight men were unavailable for follow-up (six died and two declined further measurements). Four had ulcerative colitis and four Crohn's disease: their mean age was 56.4 years (range, 3 1 4 9 years). O n e of these had a Z-score below -2.00 at the initial measurement, and in two the Z-score was exactly -2.00.
1058
D.Clements et al. BMCIBW
g/cm2
r
0.2
26
26
29
31
36
38
\../-.--
39
41
42
-
-
-
46
60
62
67
68
61
68
76
-
-0.2
-0.4 -
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
-0.6 -
-0.8 -1'
0
I
I
I
I
I
I
I
20
40
60
80
100
120
140
cumulative years of follow up
-BMC/BW Fig. 3. Rates of bone loss in individual male patients ( n = 17), plotted as the cumulative absolute changes in BMC/BW. The age of each patient (in years) at the time of the second measurement is shown above each point.
BMC/BW g/cm2 0.2 r 26 2 6
28 32 3 4 3 8
43
0-l/-0.2
-
-0.4
-
-0.6
-
-0.8
-
69 7 1 7 3
-1
74
I
I
I
I
_c
I
I
I
I
77
I
BMC/BW
Fig. 4. Rates of bone loss in individual female patients ( n = 22), plotted as the cumulative absolute changes in BMC/BW. The age of each patient (in years) at the time of the second measurement is shown above each point.
DISCUSSION This study provides the first longitudinal data o n rates of bone loss from the radius in patients with inflammatory bowel disease and demonstrates significant losses in women, but not in men, over the period of the study. The annual
loss rate in women, 0.74%, is derived from changes in all age groups and underestimates the rate of loss in the postmenopausal women, which was 1.16%. This figure is a t the upper end of the range reported for cortical bone in normal postmenopausal women (10, 11).In men the rate of
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Cortical Bone Loss in IBD
bone loss was considerably lower, although again the mean figure of 0.07% includes all age groups, higher rates of bone loss being counterbalanced to some extent by gains in bone mass in the youngest subjects (Fig. 2). Even in elderly men the rate of bone loss was well within reported ranges (10, 11); however, it should be noted that, for both men and women, these have been derived predominantly from cross-sectional studies (12). The long period between measurements in this study (mean, 7.9 years) is easily sufficient to observe significant changes in bone mass, given the precision of the technique and the number of patients studied. However, the long time interval between measurements also has limitations imposed by the death of some patients during the study; this was especially evident in the men, in whom omission of some older patients from the second measurement is likely to result in underestimation of rates of bone loss, since these are higher in older than in younger subjects. In addition, those who died may have had more severe inflammatory bowel disease, with a greater likelihood of steroid therapy, malnutrition, and other factors that accelerate bone loss. Another disadvantage of the length of follow-up in our study was that we were unable to assess accurately the possible contribution of factors such as steroid therapy and severity of disease to rates of bone loss. In normal populations several studies have demonstrated that single measurements of bone mass in the 7th or 8th decades of life are predictive of future fracture risk (13-16). Whilst the biologic variability of physiologic bone loss is considerable, initial bone mass rather than subsequent rates of bone loss is probably the main determinant of future bone mass and fracture risk in the middle years of life (17). In keeping with this, we found that all patients with abnormally low BMC/BW at the beginning of the study remained abnormal at the second measurement; however, several patients with normal values initially were found to be abnormal at the second measurement, indicating that the relationship between initial and future bone mass is less predictable in patients with inflammatory bowel disease than in the normal population. This is likely to be due to factors such as corticosteroid therapy (18), weight loss (19), malnutrition, and sex hormone deficiency (20,21), which are frequently associated with active inflammatory bowel disease and may accelerate bone loss during disease relapse. Thus in an earlier study we observed very high rates of spinal trabecular bone loss over the course of 1 year in some patients with inflammatory bowel disease (7); subsequent follow-up has shown that these rates were not maintained over the following 2 years. Those data, together with the findings of the present study, are consistent with a greater variance in rates of bone loss in patients with inflammatory bowel disease than in-the normal population. The clinical significance of osteoporosis associated with inflammatory bowel disease requires further study. Whilst spinal fractures undoubtedly occur in some premenopausal
1059
patients, it has not been established whether the prevalence of these or other fractures is increased in this population. The results of the present study indicate that the rate of bone loss from the radius at a predominantly cortical site is within normal limits in most patients, although a small proportion exhibit high rates of bone loss. The persistence of abnormally low values over nearly 8 years indicates that, in those patients with low values, prophylactic measures should be considered to prevent further bone loss. Finally, the higher variance in rates of bone loss in this patient group lessens the predictive value of a single measurement of bone mass and indicates that bone densitometry should be repeated at intervals when feasible, especially in patients with strong risk factors such as corticosteroid therapy, malnutrition and amenorrhoea. The question of which site of measurement best predicts fracture risk in the spine and femur is controversial (22-24), but in practice the technique used will be dictated by the availability of resources. Since bone loss in osteoporosis may be heterogeneous, our earlier observation that the prevalence of spinal osteoporosis is increased in patients with inflammatory bowel disease indicates that, when possible, spinal bone mass should be assessed. In those patients who are considered to be at risk, hormone replacement therapy (25) should be given when appropriate, whilst etidronate (26,27) or sodium fluoride (28) should be considered in men. Calcium supplementation should also be given to those with low dietary intakes (29), corticosteroid therapy kept to a minimum, and risk factors such as cigarette smoking and excessive alcohol consumption avoided. ACKNOWLEDGEMENTS We are grateful to Dr. T. G. Newcombe for statistical advice. Dr. J. Compston is supported by the Wellcome Trust. REFERENCES 1. Salvesen HA, Boe J. Osteomalacia in sprue. Acta Med Scand 1953;146:290-9. 2. Nordin BEC. Effects of malabsorption syndrome on calcium metabolism. Proc R Soc Med 1961;54:497-500. 3. Genant HK, Mall JC, Wagonfield JB, Horst JV, Lanzl LH. Skeletal demineralization and growth retardation in inflammatory bowel disease. Invest Radiol 1976;11:541-9. 4. Hylander E, Ladefoged K, Madsen S. Calcium balance and bone mineral content following small intestinal resection. Scand J Gastroenterol 1981;16:167-76. 5. Compston JE, Judd D, Crawley EO, et al. Osteoporosis in patients with inflammatory bowel disease. Gut 1987;28:410-5. 6 . Motley RJ, Crawley EO, Evans C, Rhodes J, Compston JE. Increased rate of spinal trabecular bone loss in patients with inflammatory bowel disease. Gut 1Y88;2Y:1332-6. 7. Sorensen JA, Cameron JR. A reliable in vivo measurement of bone mineral content measurements. Scand J Clin Lab Invest 1967 ;49A: 481-9 7. 8. Ringe JD, Rehpenning W, Kuhlencordt F. Physiologische Anderung des Mineralgehalts von Radius und Ulna in Abhangigkeit von lebensalter under Geschlecht. ROEFO 1977;126:37680. 9. Compston JE, Evans WD, Crawley EO, Evans C. Bone mineral content in normal UK subjects. Br J Radiol 1988;61:631-6.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
1060
D. Clements et al.
10. Mazess RB. On ageing and bone loss. Clin Orthop 1982;165: 239-52. 11. Sowers M, Clark K , Wallace R, Jannausch M, Lemke J. Prospective study of radial bone mineral density in a geographically defined population of postmenopausal Caucasian women. Calcif Tissue Int 1991;48:232-9. 12. Ross PD, Davis JW, Vogel JM, Wasnich RD. A critical review of bone mass and the risk of fractures in osteoporosis. Calcif Tissue Int 1990;46:149-61. 13. Wasnich RD, Ross PD, Heilbrun LK, Vogen JM. Prediction of postmenopausal fracture risk with bone mineral measurements. Am J Obstet Gynecol 1985;153:745-51. 14. Hui SL, Slemenda CS, Johnston CC. Baseline measurement of bone mass predicts fracture in white women. Ann Intern Med 1989;11 11355-61. 15. Gadsell P, Johnell 0, Nilsson B. Predicting fractures in women by using forearm bone densitometry. Calif Tissue Int 1989;44: 235-42. 16. Cummings SR, Black DM, Nevitt MC. Appendicular bone density and age predict hip fractures in women. JAMA 1990; 2631665-8, 17. Hui SL, Slemenda CW, Johnston CC. The contribution of bone loss to postmenopausal osteoporosis. Osteoporosis Int 1990;1:30-4. 18. Hahn TJ, Boisseau C , Avioli LV. Effect of chronic corticosteroid administration on diaphyseal and metaphyseal bone mass. J Clin Endocrinol Metab 1974;39:274-82. 19. Rigotti NA, Nussbaum SR, Herzog DB, Neer RM. Osteoporosis in women with anorexia nervosa. N Engl J Med 1984:311:1601-6. Received 14 January 1992 Accepted 10 July 1992
20. Lindsay R, Aitken JM, Anderson JB, Hart DM, McDonald EB, Clark AC. Long-term prevention of postmenopausal osteoporosis by oestrogen. Lancet 1976;1:1038-41. 21. Lichtarowicz A, Normal C, Calcraft B, Morris JS, Rhodes J , Mayberry J . A study of menopause, smoking, and contraception in women with Crohn’s disease. Q J Med 1989;267:623-31. 22. Wasnich RD, Ross PD, Heilbrun LK, Vogel JM. Selection of the optimal skeletal site for fracture risk prediction. Clin Orthop 1987;216:262-8. 23. Mazess RB, Barden H, Ettinger M, Schultz E. Bone density of the radius, spine and proximal femur in osteoporosis. J Bone Min Res 1988;3:13-8. 24. Need AG, Nordin BEC. Which bone to measure? Osteoporosis Int 1990;1:3-6. 25. Lindsay R , Hart DM, Forrest C, Baird C. Prevention of spinal osteoporosis in oophorectomised women. Lancet 1980;2:11513. 26. Storm T, Thamsborg G, Steiniche T, Genant HK, Sorensen OH. Effects of intermittent cyclical etidronate therapy on bone mass and fracture rate in women with postmenopausal osteoporosis. N Engl J Med 1990;322:1265-71. 27. Watts NB, Harris ST, Genant HK, et al. Intermittent cyclical etidronate treatment of postmenopausal osteoporosis. N Engl J Med 1990;323:73-9. 28. Mamelle N, Meunier PJ, Dusan R, et al. Risk-benefit ratio of sodium fluoride treatment in primary vertebral osteoporosis. Lancet 1988;2:361-5. 29. Cumming RG. Calcium intake and bone mass: a quantitative review of the evidence. Calcif Tissue Int 1990;47:194-201.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Clements D, Motley RJ, Evans WD, Harries AD, Rhodes J, Coles RJ, Compston JE. Longitudinal study of cortical bone loss in patients with inflammatory bowel disease. Scand J Gastroenterol 1992;27:1055-1060. c o n e mineral density of the radius was measured by single-photon absorptiometry in 50 patients with inflammatory bowel disease. Thirty-three had Crohn’s disease and 17 ulcerative colitis; 25 were women. The mean age was 45 years (range, 18-70 years). Measurements were repeated in 39 of them after a mean follow-up period of 7.9 years (range, 7.1-8.2 years). In female patients the mean (95% confidence interval) annual change in radial bone mineral density was -0.74% (-1.34% to -0.14%) ( P = 0.022), the greatest bone loss occurring in postmenopausal women (mean, -1.16% (-2.01% to -0.30%)). In male patients the mean annual rate of bone loss was -0.07% (-0.41% to 0.28%) ( P = NS). Patients with abnormally low values at the first measurement remained osteopenic at the second measurement, whilst some others with aormal values initially showed increased rates of bone loss and had a subnormal bone mineral density after the follow-up period. These results show increased rates of cortical bone loss in some patients with inflammatory bowel disease and emphasize the need to monitor bone mass in these patients so that prophylactic measures can be instituted. 4
Key words: Bone loss; cortical bone; inflammatory bowel disease; osteoporosis David Clements, Senior Registrar, University Hospital of Wales, Heath Park, Cardiff, U.K . CF4 4XW
Patients with inflammatory bowel disease have an increased prevalence of osteoporosis, with severe clinical manifestations in some who are relatively young (1-5). This increase in prevalence could be due to a low peak bone mass and/or an increased rate of bone loss. However, although increased rates of spinal trabecular bone loss have been demonstrated in some patients with inflammatory bowel disease (6), there have been no longitudinal studies of cortical bone loss in these patients. Between 1981 and 1982, 50 patients with inflammatory bowel disease had measurement of proximal forearm bone density at a site where bone is predominantly cortical. These measurements were repeated after a mean follow-up of 7.9 years. ,
SUBJECTS AND METHODS Patient details (Table I) Fifty outpatients, 25 women, with inflammatory bowel disease were studied between 1981 and 1982. Thirty-three had Crohn’s disease and 17 ulcerative colitis. In the 33 patients with Crohn’s disease the distribution of disease was as follows: diffuse small bowel, 4; terminal ileum, 17; ileum colon, 6; colonic, 4; and anorectal, 2. Nineteen of these patients had undergone one or more small-bowel resections. Of the 17 patients with ulcerative colitis, 4 had total colitis, 8 distal colitis, and 5 rectal disease only. Their age ranged from 17.8 to 69.8 years (mean, 45 years) at the
+
time of the first measurements. All the patients from the initial study were traced; 11 were unavailable for follow-up measurements. Eight had died (six men), and three (two men) declined t o return for another measurement. The mean follow-up period in the remaining 39 patients was 7.9 years (range, 7.1-8.2 years). Fourteen of the 25 women were postmenopausal; all of these were more than 47 years old, and only 1 patient (aged 60 years at the time of the second measurement) received hormone replacement therapy, which was given continuously for 6 years during the study. One patient was receiving calcium and vitamin D tablets, and 19 patients were receiving corticosteroid therapy. No patient had clinical or radiologic evidence of osteoporotic fracture. Single-photon absorptiometry Bone mineral content (BMC) was measured at the junction of the distal one-third and proximal two-thirds of the left radius, a site containing predominantly cortical bone, using an in-house single-photon absorptiometer based on the original design of Sorensen & Cameron (7). It incorporates a 1251 radiation source and is calibrated against a commercial instrument (Novo Densitometer GT35) so that BMC is expressed in units of g/cm, and the ratio of BMC to bone width (BMC/BW) in g/cm2. Long-term precision (coefficient of variation) was estimated from repeat measurements ( n = 41) of BMC/BW over a period of 5.2 years. In vitro precision for a n aluminium/perspex phanton was 2.1%,
1056
D. Clements et al.
Table 1. Details of patients at the time of the first measurement ~~
~
Women (n = 25)
Men (n = 25)
Age (years)
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Height (m) Weight (kg) BMC/BW (g/cm') BMC/BW (2-score)
Mean
Range
Mean
Range
43.7 1.69 64.5 0.78 -0.11
17.8 to 68.6 1.56 to 1.84 42.3 to 82.4 0.49 to 0.92 -3.60 to 2.40
46.5 1.59 61 .S 0.64 -0.37
18.0 to 69.8 1.48 to 1.85 37.8 to 98.8 0.42 to 0.80 -3.50 to 2.20
whilst in vivo precision was 2.7% for both a normal male (age 37 years) and female volunteer (age 29 years). The same instrument and technique were used for the original and the subsequent reading.
Data unalysis Changes in height, weight, and BMC/BW were calculated, and their statistical significance assessed by paired t test. The annual percentage change in BMC/BW was calculated for each patient. Each patient's result for BMC/BW was also converted to a Z-score (value with mean/standard deviation) using the data of Ringe et al. (8), who studied 773 normal
German volunteers, using a similar technique. These data are similar t o those of normal volunteers studied in Cardiff (9); the mean Z-score (SD) for Cardiff men ( a = 66) was 0.53 (1.41) and for women ( n = 63) +0.45 (1.61). The age and first Z-score of patients with inflammatory bowel disease were compared with Cardiff normals by means of the MannWhitney test; a similar comparison was made between male and female patients. RESULTS Figs. 1 and 2 show initial and follow-up BMC/BW values for
0.9.
0.80.7.
-
(u
0.6-
E
f
0
\
9 0.5-
3 !?
0
z
0.4-
Males
0.3-
Crohn's disease o Ulcerative colitis 0
0.2. 0.1.
-
0
20
40
60
80
160
Age (years) Fig. 1. Initial and follow-up values for bone mineral density of the radius in 25 male patients with inflammatory bowel disease plotted against the normal range of Ringe e t al. (8), shown as two standard deviations above and below the mean. BMC/BW = bone mineral content/bone width (g/cm').
Cortical Bone Loss in IBD
1057
0.E 0.E
f
0.7
-
cu
..
ox
\
5
.
\
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
9
0.5
3
e 5
0
0.4 \ \
Females
0.3
0
..
Crohn's disease Ulcerative colitis
0.2
0.1
0
0
20
40
60
80
100
Age (years) Fig. 2. Initial and follow-up values for bone mineral density of the radius in 25 female patients with inflammatory bowel disease plotted against the normal range of Ringe e t al. (8). shown as two standard deviations above and below the mean. BMC/BW = bone mineral content/bone width (g/cm2).
each subject, superimposed on the reference ranges of Ringe et al. (8). There was no significant difference in the first Z-scores between male and female patients. A t the time of the first measurement the mean Z-score (range) was -0.11 (-3.60 to 2.40) in the men and -0.37 (-3.50 to 2.20) in the women (Table I). There was no significant difference in age between the patients and Cardiff controls (9). The BMC/BW Z-score for men was not significantly different from that for controls ( P = 0.25), whilst for women it was significantly lower (P = 0.01). Altogether five patients (one woman) had Z-scores below -2.00 (Figs. 1 and 2); of these, four had Crohn's disease and one had ulcerative colitis. After a mean follow-up period of 7.9 years (7.1-8.2 years), the mean (range) Z-scores were -0.41 (-5.00 to 2.67) and -0.58 (-2.71 to 2.83) in men and women, respectively. Five women and four men had Z-scores below -2.00 at follow-up (Figs. 1 and 2 ) ; these included the woman and three of the four men with initially abnormal values, the fourth man having died before the second measurement. Of the five other patients with low follow-up values, two had ulcerative colitis and three Crohn's disease.
The mean (95% confidence interval) change in BMC/BW over the course of the study was -0.04 g/cm2 (-0.07 to 0.01) in women and -0.01 g/cm2 (-0.02 to 0.01) in men, representing annual percentage changes of -0.74 (-1.34 to -0.14)and -0.07(-0.41 to0.28),respectively.Thisachieved statistical significance for women ( P = 0.022) but not for men. Figs. 3 and 4 show the cumulative absolute changes in BMC/ BW in men and women after ranking the patients by age. The mean annual decrease in the postmenopausal women ( n = 14)was -1.16% (-2.01% to -0.30%).Themean(95%confidence limits) fall in Z-scores was -0.25 (-0.70 to 0.21) for women and -0.27 (-0.74 to 0.19) for men. Of the three women with only an initial measurement, two died and one declined follow-up. All had Crohn'sdisease, and their ages at the time of the first measurement were 30.5,49.6, and 59.2 years. The BMC/BW was within 1SD of the normal mean value in all three cases. Eight men were unavailable for follow-up (six died and two declined further measurements). Four had ulcerative colitis and four Crohn's disease: their mean age was 56.4 years (range, 3 1 4 9 years). O n e of these had a Z-score below -2.00 at the initial measurement, and in two the Z-score was exactly -2.00.
1058
D.Clements et al. BMCIBW
g/cm2
r
0.2
26
26
29
31
36
38
\../-.--
39
41
42
-
-
-
46
60
62
67
68
61
68
76
-
-0.2
-0.4 -
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
-0.6 -
-0.8 -1'
0
I
I
I
I
I
I
I
20
40
60
80
100
120
140
cumulative years of follow up
-BMC/BW Fig. 3. Rates of bone loss in individual male patients ( n = 17), plotted as the cumulative absolute changes in BMC/BW. The age of each patient (in years) at the time of the second measurement is shown above each point.
BMC/BW g/cm2 0.2 r 26 2 6
28 32 3 4 3 8
43
0-l/-0.2
-
-0.4
-
-0.6
-
-0.8
-
69 7 1 7 3
-1
74
I
I
I
I
_c
I
I
I
I
77
I
BMC/BW
Fig. 4. Rates of bone loss in individual female patients ( n = 22), plotted as the cumulative absolute changes in BMC/BW. The age of each patient (in years) at the time of the second measurement is shown above each point.
DISCUSSION This study provides the first longitudinal data o n rates of bone loss from the radius in patients with inflammatory bowel disease and demonstrates significant losses in women, but not in men, over the period of the study. The annual
loss rate in women, 0.74%, is derived from changes in all age groups and underestimates the rate of loss in the postmenopausal women, which was 1.16%. This figure is a t the upper end of the range reported for cortical bone in normal postmenopausal women (10, 11).In men the rate of
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
Cortical Bone Loss in IBD
bone loss was considerably lower, although again the mean figure of 0.07% includes all age groups, higher rates of bone loss being counterbalanced to some extent by gains in bone mass in the youngest subjects (Fig. 2). Even in elderly men the rate of bone loss was well within reported ranges (10, 11); however, it should be noted that, for both men and women, these have been derived predominantly from cross-sectional studies (12). The long period between measurements in this study (mean, 7.9 years) is easily sufficient to observe significant changes in bone mass, given the precision of the technique and the number of patients studied. However, the long time interval between measurements also has limitations imposed by the death of some patients during the study; this was especially evident in the men, in whom omission of some older patients from the second measurement is likely to result in underestimation of rates of bone loss, since these are higher in older than in younger subjects. In addition, those who died may have had more severe inflammatory bowel disease, with a greater likelihood of steroid therapy, malnutrition, and other factors that accelerate bone loss. Another disadvantage of the length of follow-up in our study was that we were unable to assess accurately the possible contribution of factors such as steroid therapy and severity of disease to rates of bone loss. In normal populations several studies have demonstrated that single measurements of bone mass in the 7th or 8th decades of life are predictive of future fracture risk (13-16). Whilst the biologic variability of physiologic bone loss is considerable, initial bone mass rather than subsequent rates of bone loss is probably the main determinant of future bone mass and fracture risk in the middle years of life (17). In keeping with this, we found that all patients with abnormally low BMC/BW at the beginning of the study remained abnormal at the second measurement; however, several patients with normal values initially were found to be abnormal at the second measurement, indicating that the relationship between initial and future bone mass is less predictable in patients with inflammatory bowel disease than in the normal population. This is likely to be due to factors such as corticosteroid therapy (18), weight loss (19), malnutrition, and sex hormone deficiency (20,21), which are frequently associated with active inflammatory bowel disease and may accelerate bone loss during disease relapse. Thus in an earlier study we observed very high rates of spinal trabecular bone loss over the course of 1 year in some patients with inflammatory bowel disease (7); subsequent follow-up has shown that these rates were not maintained over the following 2 years. Those data, together with the findings of the present study, are consistent with a greater variance in rates of bone loss in patients with inflammatory bowel disease than in-the normal population. The clinical significance of osteoporosis associated with inflammatory bowel disease requires further study. Whilst spinal fractures undoubtedly occur in some premenopausal
1059
patients, it has not been established whether the prevalence of these or other fractures is increased in this population. The results of the present study indicate that the rate of bone loss from the radius at a predominantly cortical site is within normal limits in most patients, although a small proportion exhibit high rates of bone loss. The persistence of abnormally low values over nearly 8 years indicates that, in those patients with low values, prophylactic measures should be considered to prevent further bone loss. Finally, the higher variance in rates of bone loss in this patient group lessens the predictive value of a single measurement of bone mass and indicates that bone densitometry should be repeated at intervals when feasible, especially in patients with strong risk factors such as corticosteroid therapy, malnutrition and amenorrhoea. The question of which site of measurement best predicts fracture risk in the spine and femur is controversial (22-24), but in practice the technique used will be dictated by the availability of resources. Since bone loss in osteoporosis may be heterogeneous, our earlier observation that the prevalence of spinal osteoporosis is increased in patients with inflammatory bowel disease indicates that, when possible, spinal bone mass should be assessed. In those patients who are considered to be at risk, hormone replacement therapy (25) should be given when appropriate, whilst etidronate (26,27) or sodium fluoride (28) should be considered in men. Calcium supplementation should also be given to those with low dietary intakes (29), corticosteroid therapy kept to a minimum, and risk factors such as cigarette smoking and excessive alcohol consumption avoided. ACKNOWLEDGEMENTS We are grateful to Dr. T. G. Newcombe for statistical advice. Dr. J. Compston is supported by the Wellcome Trust. REFERENCES 1. Salvesen HA, Boe J. Osteomalacia in sprue. Acta Med Scand 1953;146:290-9. 2. Nordin BEC. Effects of malabsorption syndrome on calcium metabolism. Proc R Soc Med 1961;54:497-500. 3. Genant HK, Mall JC, Wagonfield JB, Horst JV, Lanzl LH. Skeletal demineralization and growth retardation in inflammatory bowel disease. Invest Radiol 1976;11:541-9. 4. Hylander E, Ladefoged K, Madsen S. Calcium balance and bone mineral content following small intestinal resection. Scand J Gastroenterol 1981;16:167-76. 5. Compston JE, Judd D, Crawley EO, et al. Osteoporosis in patients with inflammatory bowel disease. Gut 1987;28:410-5. 6 . Motley RJ, Crawley EO, Evans C, Rhodes J, Compston JE. Increased rate of spinal trabecular bone loss in patients with inflammatory bowel disease. Gut 1Y88;2Y:1332-6. 7. Sorensen JA, Cameron JR. A reliable in vivo measurement of bone mineral content measurements. Scand J Clin Lab Invest 1967 ;49A: 481-9 7. 8. Ringe JD, Rehpenning W, Kuhlencordt F. Physiologische Anderung des Mineralgehalts von Radius und Ulna in Abhangigkeit von lebensalter under Geschlecht. ROEFO 1977;126:37680. 9. Compston JE, Evans WD, Crawley EO, Evans C. Bone mineral content in normal UK subjects. Br J Radiol 1988;61:631-6.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Sydney on 12/31/14 For personal use only.
1060
D. Clements et al.
10. Mazess RB. On ageing and bone loss. Clin Orthop 1982;165: 239-52. 11. Sowers M, Clark K , Wallace R, Jannausch M, Lemke J. Prospective study of radial bone mineral density in a geographically defined population of postmenopausal Caucasian women. Calcif Tissue Int 1991;48:232-9. 12. Ross PD, Davis JW, Vogel JM, Wasnich RD. A critical review of bone mass and the risk of fractures in osteoporosis. Calcif Tissue Int 1990;46:149-61. 13. Wasnich RD, Ross PD, Heilbrun LK, Vogen JM. Prediction of postmenopausal fracture risk with bone mineral measurements. Am J Obstet Gynecol 1985;153:745-51. 14. Hui SL, Slemenda CS, Johnston CC. Baseline measurement of bone mass predicts fracture in white women. Ann Intern Med 1989;11 11355-61. 15. Gadsell P, Johnell 0, Nilsson B. Predicting fractures in women by using forearm bone densitometry. Calif Tissue Int 1989;44: 235-42. 16. Cummings SR, Black DM, Nevitt MC. Appendicular bone density and age predict hip fractures in women. JAMA 1990; 2631665-8, 17. Hui SL, Slemenda CW, Johnston CC. The contribution of bone loss to postmenopausal osteoporosis. Osteoporosis Int 1990;1:30-4. 18. Hahn TJ, Boisseau C , Avioli LV. Effect of chronic corticosteroid administration on diaphyseal and metaphyseal bone mass. J Clin Endocrinol Metab 1974;39:274-82. 19. Rigotti NA, Nussbaum SR, Herzog DB, Neer RM. Osteoporosis in women with anorexia nervosa. N Engl J Med 1984:311:1601-6. Received 14 January 1992 Accepted 10 July 1992
20. Lindsay R, Aitken JM, Anderson JB, Hart DM, McDonald EB, Clark AC. Long-term prevention of postmenopausal osteoporosis by oestrogen. Lancet 1976;1:1038-41. 21. Lichtarowicz A, Normal C, Calcraft B, Morris JS, Rhodes J , Mayberry J . A study of menopause, smoking, and contraception in women with Crohn’s disease. Q J Med 1989;267:623-31. 22. Wasnich RD, Ross PD, Heilbrun LK, Vogel JM. Selection of the optimal skeletal site for fracture risk prediction. Clin Orthop 1987;216:262-8. 23. Mazess RB, Barden H, Ettinger M, Schultz E. Bone density of the radius, spine and proximal femur in osteoporosis. J Bone Min Res 1988;3:13-8. 24. Need AG, Nordin BEC. Which bone to measure? Osteoporosis Int 1990;1:3-6. 25. Lindsay R , Hart DM, Forrest C, Baird C. Prevention of spinal osteoporosis in oophorectomised women. Lancet 1980;2:11513. 26. Storm T, Thamsborg G, Steiniche T, Genant HK, Sorensen OH. Effects of intermittent cyclical etidronate therapy on bone mass and fracture rate in women with postmenopausal osteoporosis. N Engl J Med 1990;322:1265-71. 27. Watts NB, Harris ST, Genant HK, et al. Intermittent cyclical etidronate treatment of postmenopausal osteoporosis. N Engl J Med 1990;323:73-9. 28. Mamelle N, Meunier PJ, Dusan R, et al. Risk-benefit ratio of sodium fluoride treatment in primary vertebral osteoporosis. Lancet 1988;2:361-5. 29. Cumming RG. Calcium intake and bone mass: a quantitative review of the evidence. Calcif Tissue Int 1990;47:194-201.
