THE NEW BIOCHEMICAL MARKERS of bone turnover, most of

Transcription

1 JOURNAL OF BONE AND MINERAL RESEARCH Volume 13, Number 4, 1998 Blackwell Science, Inc American Society for Bone and Mineral Research Collagen-Related Markers of Bone Turnover Reflect the Severity of Liver Fibrosis in Patients with Primary Biliary Cirrhosis NURIA GUAÑABENS,1 ALBERT PARÉS,2 LUISA ALVAREZ, 3 a M JESÚS MARTÍNEZ DE OSABA,1 ANA MONEGAL, 1 PILAR PERIS, 1 ANTONIO M. BALLESTA, 3 and JOAN RODÉS2 ABSTRACT The influence of a nonskeletal disease with increased connective tissue synthesis or degradation in the collagenrelated markers of bone turnover has been evaluated in 34 women with primary biliary cirrhosis (PBC; age range years), a disease with increased hepatic fibrosis, often associated with osteoporosis. Serum osteocalcin (BGP), and carboxy-terminal (PICP) and amino-terminal (PINP) propeptides of type I collagen were assessed as indexes of bone formation, whereas serum tartrate-resistant acid phosphatase (TRAP), and cross-linked carboxyterminal telopeptide of type I collagen (ICTP), and urinary hydroxyproline (HYP), pyridinoline (PYR), deoxypyridinoline (DPYR), and type I collagen cross-linked N- (NTX) and C-telopeptide (CTX) were measured as markers of bone resorption. The histologic stage of the disease and serum amino-terminal propeptide of type III collagen (PIIINP) as an index of liver fibrogenesis were also evaluated. BGP levels were significantly lower, whereas PICP and PINP levels were higher in patients than in controls. Among the bone resorption markers, serum ICTP and urinary PYR, DPYR, HYP, NTX, and CTX levels were significantly higher in patients than in controls. Serum PIIINP levels were also increased in PBC patients. BGP did not correlate with PICP and PINP, but these markers of bone formation as well as ICTP, PYR, DPYR, and NTX correlated with serum PIIINP levels. Serum TRAP did not correlate with collagen-related markers of bone resorption. Moreover, patients with PIIINP and bilirubin above normal levels had higher PICP, PINP, ICTP PYR, DPYR, CTX, and NTX. These markers correlated with the histologic stage of the disease, but not with osteopenia measured by densitometric procedures in 22 patients. In conclusion, collagen-related markers of bone turnover do not reflect bone remodeling in PBC. The close association of these markers with PIIINP and the clinical and histologic stage of the liver disease suggests that they are influenced by liver collagen metabolism. (J Bone Miner Res 1998;13: ) INTRODUCTION THE NEW BIOCHEMICAL MARKERS of bone turnover, most of them based on products of collagen synthesis or degradation, improve the usefulness of classic markers, serum total alkaline phosphatase, and urinary hydroxyproline (HYP) for assessing bone metabolism disturbances. (1) Few studies have focused on the influence of a nonskeletal disease with increased connective tissue synthesis or degradation in the collagen-related markers of bone turnover. This point is critical for chronic liver diseases such as primary biliary cirrhosis (PBC), which is often associated with metabolic bone disorders, particularly osteoporosis. (2 5) Indeed, most chronic liver diseases result in increased fibrosis as a consequence of an imbalance between collagen synthesis and degradation. Thus, whereas in normal liver 80% of the total collagen consists of similar amounts of collagen types I and III, in liver fibrosis there is a marked increase of 1 Metabolic Bone Diseases Unit, Hospital Clínic i Provincial, University of Barcelona, Barcelona, Spain. 2 Liver Unit, Hospital Clínic i Provincial, University of Barcelona, Barcelona, Spain. 3 Laboratory of Clinical Biochemistry, Hospital Clínic i Provincial, University of Barcelona, Barcelona, Spain. 731

2 732 GUAÑABENS ET AL. all collagens, particularly type I, which is 4-fold higher than type III. (6,7) The theoretic advantages of measuring propeptides of type I procollagen and collagen cross-linking metabolites for monitoring bone turnover in liver diseases should depend on their bone specificity. In this respect, since collagen type I is the most abundant protein of bone, markers derived from synthesis and degradation of type I collagen have been considered specific of bone connective tissue remodeling. However, hepatic collagen type I is also increased in chronic liver diseases, and consequently liver fibrosis could influence the levels of these collagen-related markers. Therefore, the current study evaluates the usefulness of the new biochemical collagen-related markers of bone turnover for assessing bone remodeling in patients with PBC. Patients MATERIALS AND METHODS We studied 34 nonselected women (mean age SD ; range years) diagnosed with PBC using clinical, biochemical, immunologic, and histologic criteria. Twenty-seven were postmenopausal. All patients had normal serum calcium levels and renal function, and none had been treated with sodium fluoride, biphosphonates, estrogens, or other agents that could influence bone metabolism except that all patients received oral calcium supplements and 25-hydroxyvitamin D (20 g/day). Patients consented to participate in the investigation, and the study was approved by the Ethics Committee of the Hospital Clínic. A group of age-matched healthy females with no evidence of liver dysfunction and without disturbances of calcium metabolism was studied in order to obtain reference values. Sample collection and assays After an overnight fast, blood and 2 h urine samples were obtained between 8 and 10 a.m. No previous gelatin-free diet was consumed by patients or controls. Aliquots of serum and urine were kept frozen at 20 C until analysis. Markers of bone formation Serum osteocalcin (BGP) was assayed using an immunoradiometric method (IRMA; Elsa-Osteo, CIS Biointernational, Gif-sur-Yvette, France) in 30 patients. Serum carboxy-terminal propeptide of type I procollagen (PICP) and amino-terminal propeptide of type I procollagen (PINP) determinations were made by radiometric methods (RIA; Orion Diagnostica, Espoo, Finland) in 34 and 22 patients, respectively. Markers of bone resorption Serum tartrate-resistant acid phosphatase (TRAP) as a nonspecific estimation of bone resorption was measured in 32 patients by the modified Hillmann method using a kit from BioMerieux (Marcy-l Etoile, France) in a Cobas Mira S analyzer. Serum carboxy-terminal telopeptide of type I collagen (ICTP) levels were measured by a RIA (Orion Diagnostica) in 34 patients. Urinary free pyridinoline (PYR) and free deoxypyridinoline (DPYR) were measured by enzyme immunoassays (ELISA; Pyrilinks and Pyrilinks-D, Metra Biosystems, Mountain View, CA, U.S.A.) in 29 and 27 patients, respectively. Urinary crosslinked N-telopeptide of type I collagen (NTX) and crosslinked C-telopeptide of type I collagen (CTX) were measured in 25 patients by ELISA (Osteomark, Ostex International, Seattle, WA, U.S.A.; CIS Biointernational). Urinary HYP was measured by high performance liquid chromatography in 33 patients. Urine determinations were expressed in relation to creatinine excretion, which was measured in a Cobas Mira S analyzer using a modified Jaffe method (Hoffman-La Roche Diagnostics, Basel, Switzerland). The intra-assay coefficients of variation were as follows: BGP, 3.5%; PICP, 4%; PINP, 4.1%; TRAP, 1.5%; ICTP, 4.2%; HYP, 3%; PYR, 5%; DPYR, 5.3%; CTX, 5.2%; NTX, 5.5%; and serum amino-terminal propeptide of type III collagen (PIIINP), 7%. The interassay coefficients of variation for each of these assays are as follows: BGP, 4.5%; PICP, 6%; PINP, 6,3%; TRAP, 3%; ICTP, 6,5%; HYP, 6%; PYR, 6.5%; DPYR, 7%; CTX, 8%; NTX, 7.8%; and PII- INP, 9%. Liver function and histologic assessment In addition to standard liver function tests, PIIINP was measured as an index of liver fibrogenesis by an IRMA method (RIAGnost PIIIIP IRMA, Beringhwerke, Germany) in 28 patients. Percutaneous liver biopsy samples within 12 months of the laboratory determinations were taken in 31 patients. The liver biopsy was used for classifying patients according to the four progressive stages of the liver disease following Ludwig s criteria, (8) from stage 1 without fibrosis to stage 4 with cirrhosis. Bone mass assessment Bone mineral density (BMD) of the lumbar spine (L2 L4) was measured by dual-energy X-ray absorptiometry (Lunar DPX-L, Lunar Radiation Corporation, Madison, WI, U.S.A.) within 6 months of the laboratory measurements in 22 patients. The coefficients of variation in lumbar spine for healthy volunteers and for patients with osteoporosis were 0.8% and 1.3%, respectively. Statistical analysis Results are expressed as mean standard error of the mean. The Chi-square test was used to analyze differences in noncontinuous variables, and the Student s t-test or Mann Whitney test was used to analyze differences in continuous variables. A one-way analysis of variance or a Kruskal Wallis test was also used to find differences among groups. Associations between variables were calculated by Pearson s correlation. A two-tailed p value 0.05 was considered to indicate a significant difference.

3 BONE TURNOVER MARKERS IN BILIARY CIRRHOSIS 733 TABLE 1. CLINICAL, BIOCHEMICAL, AND HISTOLOGICAL FEATURES OF PATIENTS WITH PRIMARY BILIARY CIRRHOSIS Healthy controls (normal range) Patients n 34 mean (range) Age (years) (40 81) Postmenopausal status (%) Bilirubin (mg/dl) ( ) Aspartate aminotransferase (u/l) (38 167) Alanine aminotransferase (u/l) (35 268) Alkaline phosphatase (u/l) ( ) Gamma Glutamyl transferase (u/l) (51 750) Albumin (g/l) (25 47) Prothrombin index (%) (75 100) Histological stage (n) I 6 II 9 III 11 IV 5 TABLE 2. MARKERS OF BONE REMODELING IN PATIENTS WITH PRIMARY BILIARY CIRRHOSIS AND HEALTHY AGE-MATCHED CONTROLS Patients Controls p Bone formation Osteocalcin (ng/ml) (30) (20) 0.02 PICP (ng/ml) (34) (20) 0.01 PINP (ng/ml) (22) (20) 0.05 Bone resorption TRAP (u/l) (32) (30) NS Hydroxyproline (nm/mg) (33) (50) Pyridinoline (nm/mm) (29) (22) Deoxypyridinoline (nm/mm) (27) (19) ICTP (ng/ml) (34) (21) 0.05 NTX (nm BCE/mM) (25) (18) CTX ( g/mm) (25) (18) 0.03 Figures within parentheses indicate cases assessed. NS, not significant. RESULTS The clinical, biochemical, and histologic data of patients are summarized in Table 1. Serum BGP levels were significantly lower in PBC patients than in controls, whereas PICP and PINP levels were higher in patients than in controls (Table 2). Among the bone resorption markers, serum ICTP and urinary PYR, DPYR, HYP, NTX, and CTX levels were significantly higher in PBC patients than in controls (Table 2). Serum PIIINP levels were also significantly higher in PBC patients ( u/ml) than in controls ( u/ml) ( 0.01). BGP did not correlate with PICP (Fig. 1) and PINP, but these markers of bone formation as well as ICTP, PYR, DPYR, and NTX correlated with serum PIIINP levels (Table 3, Figs. 1 and 2). Serum TRAP did not correlate with any marker of bone resorption (Table 3), although an inverse correlation was observed between TRAP and ICTP (Fig. 2). Ten patients (29%) had bilirubin levels higher than 1.5 mg/dl, and 20 patients (71%) had PIIINP levels above the normal values. When patients were analyzed according to the serum levels of PIIINP, as an index of liver fibrogenesis, patients with increased PIIINP levels showed higher levels of all collagen-related markers of bone turnover. However, significant differences between patients with increased and normal PIIINP values were observed only for serum ICTP and urinary excretion of NTX, PYR, and DPYR (Table 4). Moreover, patients with hyperbilirubinemia had increased mean values of all collagen-related markers of bone formation and resorption (Table 5). Serum BGP levels were similar in patients with elevated or normal PIIINP levels as well as in patients with high or normal bilirubin levels. A significant linear correlation was observed between PIIINP and bilirubin (r 0.58, p 0.01). PIIINP levels increased with the progression of histologic damage (Table 6). All collagen-related markers of bone

4 734 GUAÑABENS ET AL. FIG. 1. Serum PICP correlated with PIIINP (A) but not with osteocalcin (BGP) (B) in patients with primary biliary cirrhosis. turnover also increased with the progression of the histologic stage of PBC. Thus, serum PINP and PICP values increased in parallel with the histologic stage of the disease. However, BGP levels did not change among the four histologic stages. Among bone resorption markers, serum ICTP and the urinary levels of PYR, DPYR, CTX, and NTX showed a sustained increase with the progression of the histologic stage of the disease. Serum TRAP was similar among patients with stages I, II, and III, but decreased significantly in patients with stage IV, which were those with the highest levels of collagen-related markers of bone formation and bone resorption. Among the 22 patients in whom bone mass was assessed, 14 (64%) fulfilled densitometric criteria for osteopenia (lumbar BMD between 1 and 2.5 SD below the young adult mean value) (9) and 6 (27%) the criteria for osteoporosis (lumbar BMD at least 2.5 SD below the young adult mean value). (9) Markers of bone formation and resorption were similar in patients with and without osteopenia, as well as in those with or without osteoporosis. DISCUSSION The results of this study strongly suggest that the collagen-related markers of bone turnover do not reflect bone remodeling in patients with PBC since they are influenced by the severity and progression of the liver disease but not by the severity of osteopenia. The close association between these markers and serum PIIINP levels, which reflect liver fibrogenesis, indicates the lack of bone specificity. Thus, PIIINP is increased in fibrotic liver diseases (10 13) including PBC, (14 16) and correlates with the amount of liver fibrosis as well as the hepatic activity of prolyl hydroxylase, a key enzyme in collagen synthesis. (13) However, the increased levels of collagen degradation markers observed in these patients could be explained by the fact that in chronic liver diseases liver fibrogenesis is also associated with increased collagen degradation. (17,18) The relationships among collagen-related markers and serum bilirubin levels which reflect disease severity, as well as the influence of the histologic stage of PBC, also indicate the lack of bone specificity of the collagen-related markers of bone remodeling in patients with abnormalities of soft connective tissue such as liver diseases. Since PBC patients have increased values of PINP and PICP, it could be argued that bone formation is increased in this liver disease and that these markers reflect an increased osteoblastic activity. However, serum osteocalcin, one of the currently most convincing markers of osteoblastic activity, was significantly decreased in PBC patients as compared with age-matched controls, and its values were not influenced by PIIINP and bilirubin levels or by the histologic stage. Furthermore, based on histologic criteria, several authors have reported a low bone formation state in PBC. (4,19,20) Instead of reflecting increased bone formation, the high levels of PINP and PICP in patients with PBC should indicate enhanced liver fibrogenesis. In this respect, the levels of both markers increase in parallel with the histologic stage of the disease, that is with the amount of liver fibrosis, and also direct correlations were observed between PICP, PINP, and PIIINP, the latter being a reliable marker of liver fibrosis and fibrogenesis in patients with PBC and other chronic liver diseases. In addition, serum PINP and PICP levels could also be partially increased as a result of alterations in their clearance from the circulation, which is primarily via hepatic pathways. Thus, the degradation of PICP is via mannose receptor mediated endocytosis, (21) and PINP is endocytosed by a scavengerreceptor, (22) both occurring in the liver endothelial cells. Although, this hypothesis cannot be excluded, it seems unlikely, since other collagen-related markers cleared by the kidney are also increased in PBC. Previous studies have found that PICP levels are elevated in patients with liver disease. (23,24) All markers of bone resorption based on collagen degradation were increased in PBC patients. Conversely, serum TRAP activity, which has been associated with bone resorption rates, (25,26) was not elevated in patients with PBC. This could merely indicate a lack of increased bone resorption in these patients but also may result from an inhibition of TRAP activity by factors related to the severity of liver damage. Moreover, the TRAP assay used in this study was not specific for bone since the enzyme may be released from cells other than osteoclasts. (27) Therefore, more specific bone TRAP assays should be used for monitoring bone turnover in fibrotic liver disorders. The apparent association between both the severity of liver damage and PIIIP levels with the collagen-related markers of bone resorption suggest that these markers do not reflect degradation of bone connective tissue but they indeed are elevated as the result of increased collagenolysis in the liver. In this respect, the classic biochemical marker of bone resorption, urinary HYP, may be influenced by nonosseous connective metabolism and is subject to interference from systemic disorders. (1) Moreover, we have found in this study that all the newer collagen cross-linking metabolites are also increased in PBC patients. Results of previous studies assessing the bone resorption rate in bone biopsies of PBC patients are

5 BONE TURNOVER MARKERS IN BILIARY CIRRHOSIS 735 TABLE 3. PEARSON S COEFFICIENTS OF CORRELATION AMONG ALBUMIN, BILIRUBIN, AND COLLAGEN-RELATED MARKERS Bilirubin PIIINP BGP TRAP PICP PINP HYP ICTP PYR DPYR NTX CTX Albumin Bilirubin PIIINP BGP TRAP PICP PINP HYP ICTP PYR DPYR NTX 0.85 Bold characters indicate p FIG. 2. Serum ICTP directly correlated with PIIINP (A) and inversely with TRAP (B) in patients with primary biliary cirrhosis. conflicting. Some authors have found an increased bone resorption as an early feature of the bone disease which complicates PBC (28) or an increased bone turnover influenced by the severity of hepatic disease and cholestasis, (29) whereas others have shown that bone resorption was similar to that of normal control subjects. (4) Since bone biopsies were not performed in the present series, we are unable to provide a histologic assessment of bone resorption in PBC in order to compare with the biochemical resorption markers. Although PYR and DPYR, in particular, have proven to be useful indicators of bone resorption activity, (30 32) at present awareness about lower bone specificity than previously anticipated is emerging. (33) Indeed, an interesting finding in our study was the increased levels of urinary PYR and DPYR to a similar extent in PBC patients. Furthermore, both markers correlated with PIIINP values and their levels clearly increased with the progression of the stage of TABLE 4. MARKERS OF BONE REMODELING IN PATIENTS WITH PRIMARY BILIARY CIRRHOSIS ACCORDING TO THE SERUM PIIINP LEVELS PIIINP 0.8 u/ml PIIINP 0.8 u/ml p Bone formation Osteocalcin (ng/ml) (7) (17) NS PICP (ng/ml) (8) (20) NS PINP (ng/ml) (6) (15) NS Bone resorption TRAP (u/l) (8) (19) NS Hydroxyproline (nm/mg) (8) (19) 0.09 Pyridinoline (nm/mm) (8) (16) 0.03 Deoxypyridinoline (nm/mm) (8) (15) ICTP (ng/ml) (8) (20) NTX (nm BCE/mM) (7) (15) 0.04 CTX ( g/mm) (7) (15) 0.07 Figures within parentheses indicate cases assessed. NS, not significant.

6 736 GUAÑABENS ET AL. TABLE 5. MARKERS OF BONE REMODELING IN PATIENTS WITH PRIMARY BILIARY CIRRHOSIS ACCORDING TO THE SERUM BILIRUBIN Bilirubin 1.5 mg/dl Bilirubin 1.5 mg/dl p Bone formation Osteocalcin (ng/ml) (22) (8) NS PICP (ng/ml) (24) (10) 0.03 PINP (ng/ml) (14) (8) 0.05 Bone resorption TRAP (u/l) (24) (8) Hydroxyproline (nm/mg) (23) (10) Pyridinoline (nm/mm) (20) (9) 0.01 Deoxypyridinoline (nm/mm) (18) (9) 0.01 ICTP (ng/ml) (24) (10) 0.01 NTX (nm BCE/mM) (16) (9) CTX ( g/mm) (16) (9) 0.09 Figures within parenthesis indicate cases assessed. NS, not significant. TABLE 6. SERUM BILIRUBIN AND PIIINP LEVELS AND MARKERS OF BONE FORMATION AND RESORPTION IN PATIENTS WITH PRIMARY BILIARY CIRRHOSIS ACCORDING TO THE HISTOLOGIC STAGE OF THE DISEASE Stage I Stage II Stage III Stage IV p* Bilirubin (mg/dl) (6) (9) (11) (5) PIIINP (u/ml) (5) (7) (10) (4) Bone formation Osteocalcin (ng/ml) (5) (9) (10) (3) NS PICP (ng/ml) (6) (9) (11) (5) 0.03 PINP (ng/ml) (4) (6) (8) (3) Bone resorption TRAP (u/l) (6) (9) (10) (4) Hydroxyproline (nm/mg) (6) (9) (10) (5) 0.01 Pyridinoline (nm/mm) (6) (7) (8) (5) 0.01 Deoxypyridinoline (nm/mm) (6) (6) (8) (4) NS ICTP (ng/ml) (6) (9) (11) (5) 0.02 NTX (nm BCE/mM) (4) (6) (8) (4) CTX ( g/mm) (4) (6) (8) (4) 0.02 *After Kruskal Wallis test. Figures within parentheses indicate cases assessed. NS, not significant. the liver disease, thus indicating that the increased levels result in part from increased liver collagen degradation. In this regard, high PYR content has been observed in the liver of patients with several fibrotic diseases, particularly alcoholic cirrhosis (34) and echinococcosis. (35) Indeed, it is known that PYRs act as mature cross-links in types I, II, III, and IX collagen of most connective tissues other than skin, and an increased content of PYR has been reported in the liver samples of viral liver fibrosis and cirrhosis, as well as in alcoholic cirrhosis. In one of these studies, DPYR was identified in some liver biopsy samples, although its quantity was small. (34) Unfortunately, in none of these studies was the contribution to urinary PYR and DPYR from the liver as a nonbone source determined. In agreement with our results, Seibel et al. found a significant increase of urinary PYR and DPYR levels in patients with fibrotic liver disease and alcoholic cirrhosis compared with normal subjects. (36) The cross-linked telopeptides, NTX and CTX, levels are also increased in PBC. It has been described that NTX may have greater specificity for bone than free pyridinolines. (37) Reasons for this affirmation are that the peptide antigen NTX becomes recognizable by the antibody only when bone collagen is degraded to small peptides, (37) and that this immunoreactive analyte is generated by osteoclasts at bone surfaces as demonstrated in vitro. (38) Furthermore, NTX levels show a greater degree of suppression than free PYRs by antireapsortive agents such as biphosphonates. (39) Urinary CTX has also proven to be a sensitive and specific index of bone resorption in patients with postmenopausal osteoporosis and other metabolic bone diseases, (40,41) and parallel measurement on urine and serum samples per-

7 BONE TURNOVER MARKERS IN BILIARY CIRRHOSIS 737 formed with a new enzyme-linked immunosorbent assay was highly correlated. (42) However, the results of the current study showing a clear-cut association between both NTX and CTX levels and the severity of liver fibrosis provide evidence suggesting that soft tissues may account for the increased CTX and particularly NTX levels. Despite the fact that collagen-derived markers of bone metabolism are markedly influenced by liver connective tissue metabolism, these markers can be helpful for monitoring the effects of different agents, such as sodium fluoride or biphosphonates for the treatment of osteopenia in patients with liver diseases. Indeed, cyclical etidronate treatment is associated with a marked decrease of urinary HYP, (43) thus indicating that the levels of this amino acid are in part derived from bone connective tissue in patients with PBC. In conclusion, increased collagen-related markers of bone remodeling do not reflect the level of bone turnover in patients with abnormalities of connective tissue metabolism other than bone. Therefore, the presence of fibrotic or chronic diseases with known abnormalities in collagen metabolism should always be taken into account when assessing bone turnover using the new collagen-related markers. REFERENCES 1. Calvo MS, Eyre DR, Gundberg CM 1996 Molecular basis and clinical application of biological markers of bone turnover. Endocr Rev 17: Mattloff DS, Kaplan MM, Neer RM, Goldberg MJ, Bitman W, Wolfe HJ 1982 Osteoporosis in primary biliary cirrhosis: Effects of 25-hydroxyvitamin D 3 treatment. Gastroenterology 83: Herlong HF, Recker RR, Maddrey WC 1982 Bone disease in primary biliary cirrhosis: Histologic features and response to 25-hydroxyvitamin D. Gastroenterology 83: Hodgson SF, Dickson ER, Wahner HW, Johnson KA, Mann KG, Riggs BL 1985 Bone loss and reduced osteoblast function in primary biliary cirrhosis. Ann Intern Med 103: Lindor KD 1993 Management of osteopenia of liver disease with special emphasis on primary biliary cirrhosis. Semin Liver Dis 13: Schuppan D 1990 Structure of the extracellular matrix in normal and fibrotic liver: Collagens and glycoproteins. Semin Liver Dis 10: Seyer JM, Hutcheson ET, Kang AH 1977 Collagen polymorphism in normal and cirrhotic human liver. J Clin Invest 59: Ludwig J, Dickson ER, McDonald GSA 1978 Staging of chronic non-suppurative destructive cholangitis (syndrome of primary biliary cirrhosis). Virchows Arch A Path Anat Histol 379: Kanis JA, Delmas P, Burckhardt P, Cooper C, Torgerson D 1997 Guidelines for diagnosis and management of osteoporosis. Osteoporos Int 7: Savolainen ER, Goldberg B, Leo MA, Velez M, Lieber CS 1984 Diagnostic value of serum procollagen peptide measurements in alcoholic liver disease. Alcohol: Clin Exp Res 8: Galambos MR, Collins DC, Galambos JT 1985 A radioimmunoassay procedure for type III procollagen: Its use in the detection of hepatic fibrosis. Hepatology 5: Torres M, Parés A, Caballería J, Bruguera M, Jiménez W, Heredia D, Ballesta AM, Rodés J 1986 El péptido aminoterminal del procolágeno tipo III como índice de fibrosis hepática en los alcohólicos crónicos. Gastroenterol Hepatol 9: Torres M, Parés A, Caballería J, Jiménez W, Heredia D, Bruguera M, Rodés J 1986 Serum procollagen type III peptide as a marker of hepatic fibrogenesis in alcoholic hepatitis. Gastroenterology 90: Ericksson S, Zettervall O 1986 The N-terminal propeptide of collagen type III in serum as a prognostic indicator in primary biliary cirrhosis. J Hepatol 2: Babbs C, Hunt LP, Haboubi NY, Smith H, Rowan BP, Warnes TW 1988 Type III procollagen peptide: a marker of disease activity and prognosis in primary biliary cirrhosis. Lancet 1: Salmerón JM, Caballería J, Parés A, Caballería L, Bruguera M, Rodés J 1989 El péptido aminoterminal del procolágeno tipo III en la cirrosis biliar primaria. Significado clínico y valor pronóstico. Gastroenterol Hepatol 12: Maruyama K, Feinman L, Fainsilber Z, Nakano M, Okazaki I, Lieber CS 1982 Mammalian collagenase increases in early alcoholic liver disease and decreases with cirrhosis. Life Sci 30: Arthur MJP 1995 Collagenases and liver fibrosis. J Hepatol 22 (Suppl 2): Guañabens N, Parés A, Mariñoso L, Brancós MA, Piera C, Serrano S, Rivera F, Rodés J 1990 Factors influencing the development of metabolic bone disease in primary biliary cirrhosis. Am J Gastroenterol 85: Stellon AJ, Webb A, Compston J, Williams R 1987 Low bone turnover state in primary biliary cirrhosis. Hepatology 7: Smedsrod B, Melkko J, Risteli L, Risteli J 1990 Circulating C-terminal propeptide of type I procollagen is cleared mainly via the mannose receptor in liver endothelial cells. Biochem J 271: Melkko J, Hellevik T, Risteli L, Risteli J, Smedsrod B 1994 Clearance of NH2-terminal propeptides of types I and III procollagen is a physiological function of the scavenger receptor in liver endothelial cells. J Exp Med 179: Hartmann DJ, Trinchet JC, Ricard-Blum S, Beaugrand R, Callard P, Ville G 1990 Radioimmunoassays of type I collagen that mainly detects degradation products in serum: Application to patients with liver diseases. Clin Chem 36: Gallouni A, Plebani M, Pontisso P, Chemello L, Masiero M, Mantovani G, Alberti A 1994 Serum markers of hepatic fibrogenesis in chronic hepatitis C treated with alfa-2 interferon. Liver 14: Lau KHW, Onishi T, Wergedal JE, Singer FR, Baylink DJ 1987Characterization and assay of tartrate-resistant acid phosphatase activity in serum: Potential use to assess bone resorption. Clin Chem 33: Rico H, Villa LF 1993 Serum tartrate-resistant acid phosphatase (TRAP) as biochemical marker of bone remodeling. Calcif Tissue Int 52: Halleen J, Hentunen TA, Hellman J, Vaananen HK 1996 Tartrate-resistant acid phosphatase from human bone: Purification and development of an immunoassay. J Bone Miner Res 11: Cuthbert JA, Pak CYC, Zerwekh JE, Glass KD, Combes B 1984 Bone disease in primary biliary cirrhosis: Increased bone resorption and turnover in the absence of osteoporosis or osteomalacia. Hepatology 4: Hodgson SF, Dickson ER, Eastell R, Eriksen EF, Bryant SC, Riggs BL 1993 Rates of cancellous bone remodeling and turnover in osteopenia associated with primary biliary cirrhosis. Bone 14: Uebelhart D, Gineyts E, Chapuy MC, Delmas PD 1990 Urinary excretion of pyridinium crosslinks: a new marker of bone resorption in metabolic bone disease. Bone Miner 8: Delmas PD, Schlemmer A, Gineyts E, Riis B, Christiansen C 1991 Urinary excretion of pyridinoline correlates with bone turnover measured on iliac crest biopsy in patients with vertebral osteoporosis. J Bone Miner Res 6: Eastell R, Colwell A, Hampton L, Reeve J 1997 Biochemical markers of bone resorption compared with estimates of bone

8 738 GUAÑABENS ET AL. resorption from radiotracer kinetic studies in osteoporosis. J Bone Miner Res 12: Eyre DR 1995 The specificity of collagen cross-links as markers of bone and connective tissue degradation. Acta Orthop Scand 66 (Suppl 266): Hayasaka A, Ilda S, Suzuki N, Kondo F, Miyazaki M, Yonemitsu H 1996 Pyridinoline collagen cross-links in patients with chronic viral hepatitis and cirrhosis. J Hepatol 24: Ricard-Blum S, Bresson-Hadni S, Vuitton DA, Ville G, Grimaud JA 1992 Hydroxypyridium collagen cross-links in human liver fibrosis: Study of alveolar echinococcosis. Hepatology 15: Seibel MJ, Schmidt-Gayk H, Stracke H, Burckhardt P 1993 Biochemical markers of bone metabolism are affected by nonskeletal diseases involving collagen turnover. J Bone Miner Res 8 (Suppl 1):S Hanson DA, Weis MA, Bollen AM, Maslan SL, Singer FR, Eyre DR 1992 A specific immunoassay for monitoring bone resorption: Quantitation of type I collagen cross-linked N- telopeptides in urine. J Bone Miner Res 7: Apone S, Fevold K, Lee M, Eyre D 1994 A rapid method for quantifying osteoclast activity in vitro. J Bone Miner Res 9 (Suppl 1):S Garnero P, Shih WJ, Gineyts E, Karpf DB, Delmas PD 1994 Comparison of new biochemical markers of bone turnover in late postmenopausal osteoporotic women in response to alendronate treatment. J Clin Endocrinol Metab 79: Garnero P, Ginetys E, Riou JP, Delmas PD 1994 Assessment of bone resorption with a new marker of collagen degradation in patients with metabolic bone disease. J Clin Endocrinol Metab 79: Bonde M, Qvist P, Fledelius C, Riis BJ, Christiansen C 1995 Applications of an enzyme immunoassay for a new marker of bone resorption (CrossLaps): Follow-up on hormone replacement therapy and osteoporosis risk assessment. J Clin Endocrinol Metab 80: Bonde M, Garnero P, Fledelius C, Qvist P, Delmas PD, Christiansen C 1997 Measurement of bone degradation products in serum using antibodies reactive with an isomerized form of an 8 amino acid sequence of the C-telopeptide of type I collagen. J Bone Miner Res 12: Guañabens N, Parés A, Monegal A, Peris P, Pons F, Alvarez L, Martínez de Osaba MJ, Roca M, Torra M, Rodés J 1997 Etidronate versus fluoride for treatment of osteopenia in primary biliary cirrhosis: Preliminary results after two years. Gastroenterology 113: Address reprint requests to: Nuria Guañabens, M.D. Service of Rheumatology Hospital Clínic i Provincial C/. Villarroel, Barcelona, Spain Received in original form June 5, 1997; in revised form October 24, 1997; accepted Novmber 11, 1997.