Changes in the cortical and trabecular bone compartments with different types of menopause measured by peripheral quantitative computed tomography

JOURNALOFTHE CuMI\cTERIcI POSTMENOPAUSE

Maturitas 23 (1996) 23-29

Changes in the cortical and trabecular bone compartments with different types of menopause measured by peripheral quantitative computed tomography E.R. Hernhdez”,

C. Seco”, M. Revilla”, L.F. Villa”, J. CortCsb, H. Rico”,*

aDepartment of Medicine, Alcahi de Henares University, 28801 Madrid, Spain bDepartment of Obstetrics, Gynecology Service, Principe de Asturias University Hospital, Alcali de Henares University, 28801 Madrid, Spain

Received 21 February 1995; revised 1 September 1995: accepted 3 October 1995

Abstract

Our objective was to study the changesin the bone mineral density of the cortical and trabecular compartments with different types of menopause.A total of 153 normal postmenopausalwomen (mean age 48 + 5 years) were studied.The women were divided’into three groupsbasedon meanageat menopause:early menopause(menopause before 43 years), normal menopause(menopauseat 44-52 years), and late menopause(menopauseafter 52 years). The number of years sincemenopausewas similar in all three groups ( f 5 years). Cortical and trabecular bone mineral density was determinedin all the women using peripheral quantitative computed tomography. Our results show that only the trabecular bone mineral density differed significantly among the groups (Kruskal-Wallis: P = 0.0029).The womenwith early menopausebad a lower trabecular bone densitythan the womenwith normal and late menopause(P = 0.0019 and P < 0.0001, respectively). Among the women with early menopause,22 had experiencedmenopausebeforethe ageof 40 and 25 after the ageof 40; there were significantdifferencesin trabecular bone mineral density betweenthesetwo subgroups(P < 0.05). Trabecular bone mineral density, the only variable studied that varied among the groups, correlated significantly with the duration of reproductive life (simplelinear regression:r = 0.340, P < 0.0001). In conclusion, thesefindings emphasizethe importance of the duration of reproductive life as a determinant of bone massin women. Peripheral quantitative computed tomography; Bone mass;Postmenopausalbone mass;Bone massand reproductive years; Menopause

Keywords:

1. Introduction The influence of age at menopause on postmenopausal bone loss is unclear, but early

* Corresponding author.

menopause is considered widely as a risk factor for osteoporosis [ 11. Bone densitometry studies, mainly of the spine and forearm, have found that early menopause, whether natural, surgical or chemical, is accompanied by a lower bone massin normal postmenopausal and osteoporotic post-

0378-5122/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved SSDI 0378-5122(95)00948-K

‘4

E.R.

Hem&de;

et al. I Maturitas

menopausal women [2-61. Although Pouilles et al. [l] and Genant and Cann [7], among the first authors of studies addressing this problem, note that existent data are confusing: in ovariectomized women these authors observed significant changes in trabecular bone mass using quantitative computed tomography and in the radial diaphysis, which measures mainly cortical bone, using single-photon absorptiometry; in contrast, no significant changes were found with radiogrammetry, which evaluates mainly cortical bone. In ovariectomized women, Horsman et al. [8] confirm the existence of significant cortical bone loss, and Fiore et al. [9] cortical and trabecular bone loss. Peripheral quantitative computed tomography has good precision [lo] and enables separate measurement of cortical bone and trabecular bone [1 11. In our opinion, this is an important advantage because these two types of bone have different temporal remodeling characteristics [ 121, structure [13], and metabolic activity [14]. Moreover, cortical and trabecular bone mass determined in the upper limbs with peripheral quantitative computed tomography differ significantly in the dominant limb [15], showing that the cortical and trabecular bone compartments respond differently to physical stimulation, As no studies have been made of changes in the bone mass of the cortical and trabecular compartments measured by peripheral quantitative computed tomography in relation to type of menopause, and the influence of age at menopause is unclear, we studied the bone mass of postmenopausal women using peripheral quantitative computed tomography to define cortical and trabecular bone mineral density in relation to early, normal, and late-occurring menopause.

2. Material

and methods

2.1. Subjects

A total of 153 normal postmenopausal women (mean age 48 -t 5 years) were selected. All the potential candidates were women from the health district of the Prlncipe de Asturias University Hospital (Alcala de Henares, Madrid, Spain) who

23 (1996)

23-29

had visited the clinic of the Rheumatology Department for nonspecific pain for which no organic cause was found. All candidates underwent a complete medical history and physical examination. The study was approved by the Office for Protection from Research Risks of the Alcala de Henares Medical School and all subjects gave written informed consent. Postmenopausal status was defined as the absence of menstruation for at least 12 months [16]. Depending on age at menopause the women were divided into three study groups: early menopause (menopause at 43 years or earlier), normal menopause (menopause between 44 and 52 years), and late menopause (menopause at 53 years or older). The number of years since menopause was similar in all three groups ( ) 5 years). The characteristics of each group are shown in Table 1. Only seven of the 47 (15%) women with early menopause had a menopause considered precocious because it was induced by ovariectomy. The remainder women were natural postmenopausal. Table 1 Number and characteristics of three groups of normal postmenopausal women Early

Normal

Later

n

41

Age (years) Age at menopause Reproductive years Height (m) Weight (kg) BMI (w/m*) Cortical (mg/cm’) Trabecular

43.8 f. 3.6 38.5 + 3.8 26.5 + 1.8” 1.59 + 0.06 63.9 +_9.7 25.2 & 3.4 444_+56 118k31“

12 53.9 + 4.7 48.0 + 2.3 37.0 i 2.3b 1.56 + 0.05 62.3 f 6.6 25.6 + 3.0 450 * 91 139 * 35

34 58.7 + 4.8 54.0 * 1.3 42.0 k 1.3 1.56 + 0.04 63.7 + 9.3 25.9 f. 3.8 477 _+72 152&46

(m&m31 Reproductive years = age at menopause minus age at menarthe. BMI = body mass index. Cortical and trabecular = cortical and trabecular bone mineral density by peripheral quantitative computed tomodensitometry. “P < 0.0001 vs normal and late, according to one-way analysis of variance. bP i 0.0001 vs late, according to one-way analysis of variante. ‘P = 0.0029 between groups according to Kruskall-Wallis analysis. dP = 0.0019 vs normal and P < 0.0001 vs late, according to one-way analysis of variance.

E.R. Hernrindez

et al. 1 Maturitas

All the women were considered normal on the basis of an interview and biochemical measurements of blood glucose, transaminases, creatinine, calcium, phosphorus, total proteins, bilirubin, alkaline phosphatase, tartrate-resistant acid phosphatase gonadal hormones, (TRW, gonadotropins, and a coagulation study. In all cases calcium was corrected for proteins. A biochemical study was made of 24-h urine to confirm the normality of calcium excretion and tubular phosphate resorption. Radiologic study of the thoracic and lumbar spine excluded vertebral deformities, defined as the loss of more than 25% of the height of the anterior, middle, or posterior vertebral body, compared with normal reference values matched for age and sex previously published by our group [ 171. None of the women smoked or was taking hormonal replacement therapy. Assessment of dietetic intake over a 7-day period revealed that the calcium intake of all subjects was more than 800 mg per day. None of them practiced sports assiduously. The subjects’ economic status was similar and they all came from a mid-sized industrial city, Alcala de Henares (Madrid, Spain). Height was measured using a Harpenden stadiometer. Weight was measured using a precision biomedical balance. 2.2. Peripheral studies

quantitative

computed tomography

23 (1996)

2.3. Analytical

25

studies

No coffee, tea, or alcohol intake, smoking, or exercise was permitted for 24 h before the day of investigation. Venous blood and urinary samples were collected at 0800 h, after an overnight fast, for hematologic and biochemical studies. Blood samples were centrifuged and the serum was stored at - 20/deg/C until analyzed. The biochemical measurements were all made in serum using a BM/ Hitachi automated analyzer system 717 (Boehringer, Mannheim, Germany). TRAP was quantitated in serum in the Hitachi automated analyzer as the substrate cr-naphthyl phosphate, using a reagent from Boehringer Laboratories (Boehringer, Mannheim, Germany) that reacts specifically with isoenzyme 5b synthesized by the osteoclast [18]. The 24-hour urinary calcium excretion was determined by atomic absorption spectroscopy using a spectrophotometer model 5000 (Perkin Elmer, Norfolk, CT, USA). All biological samples from every subject were analyzed in the same assay to eliminate interassay variation. Assay reproducibility was determined by assaying four samples five times in five different runs at two laboratories. The CV between runs and between laboratories were determined by components of variance [19], which give a statistical estimate of the variation of replicates of one for multiple assay runs. In every case, CV was less than 6%. 2.4. Statistical

Determinations were made using the Stratec XCT 900 (GmBH, Birkefeld, Germany) on the dominant forearm. After scanning the distal part of the arm, the lower part of the cubital base was marked and a slice at a distance corresponding to 4% of cubital length was scanned automatically, thus clearly differentiating the radius from the cubitus. Radial measurements then were made automatically. The instruments computer program gave measurements in mgjcm’. Our coefficient of variation (CV) for this method is l.l%, as calculated from five measurements on five individuals made at intervals of 2-3 weeks. The XCT 900 was calibrated daily with a calibrator (bone simulation standard) supplied by GmBH.

23-29

studies

One-way analysis of variance and KruskallWallis analysis were used as appropriate to compare differences between groups. Single regression analysis were used as appropriate to examine relations between continuous variables. All studies were made with the StatView 4.02 program (Abacus Concepts, Inc. Berkeley, CA, USA) for Macintosh computers.

3. Results

Only trabecular bone density differed significantly between groups (Kruskal-Wallis: P =

26

E.R. Hernbndez

et al. / Maturitas

CW1ICRI

23 (1996)

23-29

Trabecular

Fig. 1. Cortical and trabecular bone mineral density (BMD, mg/cm3) measured by peripheral quantitative computed tomodensitometry in 47 normal postmenopausal women with menopause at 143 years (early), in 72 normal postmenopausal women with menopause at 44-52 years (normal), and in 34 normal postmenopausal women with menopause after 52 years (late). *P = 0.0019 vs. normal and P < 0.0001 vs. late, according one-way analysis of variance.

O-0029), the differences in cortical bone density not being significant (P = 0.1193) (Table 1; values expressed as mean + S.D.). The women with early menopause had a lower trabecular bone density than the women with normal and late menopause (analysis of variance: 118 + 3 1 mg/cm3 vs. 139 F 35 mg/cm3 and 152 + 46 mg/cm3, P = 0.0019 and P < 0.0001, respectively) (Fig. 1). There were no differences in trabecular bone density between the women with normal and late menopause. There were no differences between groups in cortical bone mineral density, being 444 &- 56 mg/cm3 for early menopause, 450 + 91 mg/cm3 for normal menopause, and 477 + 72 mg/cm3 for late menopause. Among the women with early menopause, there were differences between the 22 women who experienced menopause at I 40 years and the 25 who experienced menopause at 40-43 years. There were marginal, but significant, differences in trabecular bone mass between the women with menopause at I 40 years and those with menopause at 40-43 years (108 + 26 mg/cm3 vs. 126 k 32 mg/cm3, respectively, P < 0.05). Likewise, the two subgroups showed significant differences in age at menopause (35.8 ) 3.9 years vs. 41.0 &- 1.0 years, P < 0.0001) and in duration of reproductive life (23.8 f 3.9 years vs. 29.0 f 1.0 years, respectively, both P < 0.0001). In all trabecular bone mineral density, the only bone

mass measurement that differed between groups, correlated significantly with duration of reproductive life (linear regression analysis: Y = 0.340, P < 0.0001) and significantly, but negatively, with years since menopause (r = - 0.322, P < 0.0001) (Fig. 2 and Fig. 3, respectively). No changes were observed in any of the analytic parameters studied.

4. Discussion Peripheral quantitative computed tomography is known to be a low-risk procedure that yields precise, separate bone density measurements of trabecular bone and compact bone [l 11. The high spatial resolution provided by peripheral quantitative computed tomography permits ‘compartmental’ analysis of bone structure [20]. Other advantages of peripheral quantitative computed tomography are the low radiation dose it delivers ( < 5 mRem) [21] and the volumetric bone mass measurements (mg/cm3) it yields [22]. Our findings obtained with peripheral quantitative computed tomography showed that the women with early menopause had a lower trabecular bone density but no change in cortical bone density. Only seven of the 47 (15%) women with early menopause had a menop.ause considered precocious because it was induced by ovariectomy at an early age; all of these women had a reduced

E.R. Herruindez

et al. / Maturitas

23 (1996)

27

23-29

275 250

Fig. 2. Simple lineal regression between reproductive years (age at menopause minus age at menarche) and trabecular bone mineral density (Trabecular, mg(cm3) in 153 normal postmenopausal women.

found a rapid loss of trabecular bone and increased bone remodeling in women with early menopause due to ovariectomy. Similar bone changes have been observed in women with artificial menopause; in women treated with Gn-RH agonists, Devogelaert et al. [30] found a loss of trabecular bone in spine and distal radius, but no loss of cortical bone measured in the middle radius, as has been observed by us and other authors {4,31]. The changes in bone mass are considered to be more dependent on age at menopause than on chronological age [32,33]. For instance, in the first 5 years of menopause no loss of cortical bone is reported [8,34]; the fact that our three groups of women had a mean duration of menopause of

trabecular bone mass. Moreover, there were marginal, but significant, differences in trabecular bone mass, and significant differences in age at menopause between the women with early menopause at < 40 years and those with early menopause at 40-43 years. Other authors have reported that ovariectomy produces more intense loss of bone mass [23-251 and that bone remodeling is more active in women ovariectomized at an early age than in women with normal menopause [26]. We and Fogelman et al. [27] found biochemical evidence in the first years after ovariectomy of a high rate of bone remodeling, but no loss of cortical bone. Genant et al. [28] also observed a preferential loss of trabecular bone and Devogelaert et al. [29] 275 250 225 “E5’

200

E

175

Ji e g e B

150 125 100 75 50 0

5

10

15

20 YSM

25

30

35

40

Fig. 3. Simple lineal regression between years since menopause (YSM) and trabecular bone mineral density (Trabecular, mg/cm)) in 153 normal postmenopausal women.

28

E.R. Hermindez

et al. I Maturitas

+ 5 years and loss of trabecular bone mass but not of cortical bone mass confirms this observation. Our findings also indicate that women with early menopause have a greater loss of trabecular bone than those who have a normal or late menopause; similar results are reported by Gambacciani et al. [35] in a wide study in which they only evaluate spine BMD. Likewise, the duration of reproductive life was significantly shorter in the group of women with early menopause and their trabecular bone density was marginally, but significantly, smaller. These observations concur with those of Vito et al. [36] that the number of reproductive years is a strong predictor of bone mineral density in the cancellous compartment of the tibia and spine. Kritzsilverstein and BarrettConnor [37] found that the number of reproductive years explained the variance in bone mineral density better than either age at menarche or age at menopause. Our findings and those cited highlight the importance of the duration of reproductive life and of estrogens on bone mass in women. References [1] Pouilles JM, Tremollieres F, Bonneu M, Rihot C. Influence of early age at menopause on vertebral bone mass. J Bone Miner Res 1994; 9: 311-315. 121 Cann ChE, Gennat HK, Ettinger B, Gordan GS. Spinal mineral loss in oophorectomized women. J Am Med Assoc 1980; 244: 2056-2059. [3] Manolagas SC, Anderson DC. Adrenal steroids and the development of osteoporosis in oophorectomised women. Lancet 1979; 2: 5977600. [4] Rico H, Arnanz F, Revilla M, Perera S, Iritia M, Villa LF, Arribas I. Total and regional bone mineral content in women treated with GnRH agonists. Calcif Tissue Int 1993; 52: 354-357. [5] Surrey ES, Foumet N, Voigt B, Judd HL. Effects of sodium etidronate in combination with low-dose norethindrone in patients administered a long-acting GnRH agonist - a preliminary report. Obstet Gynecol 1993; 81: 581-586. [6] Vega EM, Egea MA, Mautalen CA. Influence of the menopausal age on the severity of osteoporosis in women with vertebral fractures. Maturitas 1994; 19: 117-124. [7] Genant E HK, Cann CE. Vertebral mineral determination using quantitative computed tomography. In: DeLuca HF, Frost HM, Jee WSS, Johston CC, Par&t AM, eds. Osteoporosis, Recent Advances in Pathogenesis and Treatment. Baltimore: University Park Press, 1981; 37---4-f.

23 (1996)

23-29

[8] Horsman A, Simpson M, Kirby PA, Nordin BEC. Nonlinear bone loss in oophorectomized women. Br J Radio1 1977; 50: 504507. [9] Fiore CE, Falcidia E, Foti R, Caschetto S, Grimaldi DR. Postoophorectomy bone loss is associated with reduced bone Gla protein serum levels: a possible effect of osteoblastic insufficiency. Calcif Tissue Int 1987: 41: 303 306. [lo] Blake GM, Fogelman I. Recent advances in bone densitometry. Eur J Nucl Med 1993; 20: 735-737. [ll] Rico H, Hemandez ER, Seco C, Villa LF, Fernandez S. Quantitative peripheral computed tomodensitometric study of cortical and trabecular bone mass in relation with menopause. Maturitas 1994; 18: 183- 189. [12] Eriksen EF, Vesterby A, Kassem M, Melsen F, Mosekilde L. Bone remodeling and bone structure. In: Mundy CR, Martin TJ, eds. Physiology and Pharmacology of Bone. Berlin: Springer-Verlag, 1993; 67- 109. [13] Ninomiya JT, Tracy RP, Calore JD, Gendreau MA, Kelm RJ, Mann KG. Heterogeneity of human bone. J Bone Miner Res 1990; 5: 933-938. [14] Vaughan JM. The Physiology of Bone. Oxford: Clarendon Press, 1975. [15] Rico H, Gonzalez-Riola J, Revilla M, LF Villa, GomezCastresana F, Escribano J. Cortical versus trabecular bone mass: Influence of activity on both bone components. Calcif Tissue Int 1994, 54: 470-472. 1161 Hui SL, Slemenda CW, Johnston CC, Appledom CR. Effects of age and menopause on vertebral bone density. Bone Miner 1987; 2: 141-146. [17] Fraile E, Revilla M, Hemandez ER, Aguado F, Martin FJ, Rico H. Morfometrla vertebral y metacarpiana, por radiologia digitalizada, en mujeres postmenopausicas sanas. Rev Esp Enf Metb Oseas 1994; 3: 160-162. [18] Rico H, Villa LF. Serum tartrate-resistant acid phosphatase (TRAP) as a biochemical marker of bone remodeling. Calcif Tissue Int 1993; 52: 149-150. [19] Descos L, Andre C, Beorghia S, Vincent C, Revillard JP. Serum levels of b,-microglobulin - a new marker of activity in Crohn’s disease. N Engl J Med 1979;302: 440. [20] Riiegsegger P. The use of peripheral QCT in the evalation of bone remodelling. Endocrinologist 1994; 4: 1677 176. [21] Kaissi W, Keller H, Schaffner B, Riiegsegger P. Local dose and equivalence dose of 3D-pQCT. Bone Miner 1994; 25 (suppl 2): S20. [22] Ott SM. Bone mass measurements, reasons to be cautious. Br Med J 1994; 308: 931-932. [23] Dalen N, Lamke B, Wallgren A. Bone-mineral losses in oophorectomized women. J Bone Jt Surg 1974; 56A: 1235-1238. 1241 Richelson LS, Wabner HW, Melton LJ, Riggs BL. Relative contributions of aging and estrogen deficiency to postmenopausal bone loss. N Eng J Med 1984; 311: 1273-1275. [25] Smith ML, Fogelman I, Hart DM, Scott E, Bevan J, Leggate I. Effect of etidronate disodium on bone turnover

E.R.

[26] [27]

[28]

[29]

[30]

Hermindez

et al. / Maturitas

following surgical menopause. Calcif Tissue Int 1989; 44: 14-19. Young MM, Nordin BEC. Effects of natural and artificial menopause on plasma and urinary calcium and phosphorus. Lancet 1967; 2: 118-120. Fogelman I, Poser JW, Smith ML, Hart DM, Bevan JA. Alterations in skeletal metabolism following oophorectomy. In: Christiansen C, Arnaud CD, Nordin BEC, Par&t AM, Peck WA, Riggs BL, eds. Osteoporosis. Aalborg, Denmark: Stiftsbogtrykkeri, 1984; 5199521. Genant HK, Cann CE, Ettinger B, Gordan GS. Quantitative computed tomography of vertebral spongiosa: a sensitive method for detecting early bone loss after oophorectomy. Ann Intern Med 1982; 97: 699-705. Devogelaert JP, Huaux JP, Nagant de Deuxchaisnes C, Lecart J, Donnez J, Thomas K. Densitometric, biochemical, and hormonal changes induced by oophorectomy. In: Christiansen C, Johansen JS, Riis BJ, eds. Osteoporosis 1987. Kobenhavn. Denmark: Osteopress ApS, 1987; 615 617. Devogelaert JP, Huaux JP, Thomas K, Nagant de Deuxchaisnes C. Differential effect of the Gn-RH agonist buserelin on the axial and the appendicular skeleton at different scanning sites. Preferential loss of trabecular bone. In: Christiansen C, Johansen JS, Riis BJ, eds. Osteoporosis 1987. Kobenhavn, Denmark: Osteopress ApS, 1987: 5888591.

23 (1996)

23-29

29

[31] Johansen JS, Riis BJ, Hassager C, Moen M, Jacobson J, Christiansen C. The effect of a gonadotropin-release hormone agonist analog (nafarelin) on bone metabolism. J Clin Endocrinol Metab 1988; 67: 701-706. [32] Lindquist 0, Bengtsson C, Hansson T. Roos B. Age at menopause and its relation to osteoporosis. Maturitas 1979; 1: 175-181. [33] Rico H, Revilla M, Alvarez de1 Buergo Ma, Villa LF. Age-related changes in total and regional bone mass: a cross-sectional study with DEXA in 429 normal women. Osteoporosis Int 1993; 3: 154-159. [34] Rico H, Hemandez ER, Revilla M, Villa LF, Alvarez de Buergo Ma, Cuende E. Bone changes in postmenopausal Spanish women. Calcif Tissue Int 1993; 52: 103- 106. [35] Gambacciani M, Spinetti A, Desimone L, Cappagli B, Maffei S, Taponeco T, Fioretti P. The relative contributions of menopause and aging to postmenopausal vertebral osteopenia. J Clin Endocrinol Metab 1993; 77: 1148-1151. [36] Vito L, Prallet B, Chappard D, Pallot-Prades B. Pupier R, Alexandre C. Contributions of chronological age, age at menarche and menopause and of anthropometric parameters to axial and peripheral densities. Osteoporosis Int 1992; 2: 153-158. [37] Kritzsilverstein D, Barrett-Connor E. Early menopause. number of reproductive years, and bone mineral density in postmenopausal women. Am J Public Health 1993: 83: 983-988.