A variation in Bone Alkaline Phosphatase levels that correlates positively with bone loss and normal levels of aminoterminal propeptide of collagen I in girls with anorexia nervosa.

Anorexia nervosa (AN) is a very extended pathology among adolescent girls nowadays. These patients show a high degree of osteopenia; hence, study of their bone remodelling is of great interest. Serum bone alkaline phosphatase (bAP) and aminoterminal propeptide of procollagen I (PINP) provide good sensitivity in the analysis of bone alterations in postmenopausal osteoporosis. The aim of this study was to compare the usefulness of bAP and PINP in the study of bone remodelling in AN, and their possible correlation with the degree of osteopenia in this pathology. In order to help in the interpretation of the results, levels of the beta-isomer of urinary carboxyterminal propeptide of collagen I (beta-CTX) have also been included. Serum bAP (IRMA) Tandem R-Ostase, Hybritech), PINP (RIA, Orion Diagnostica) and CTX (CrossLaps ELISA, Osteometer) were determined in 41 girls with AN, aged 18.5+/-2.2 years (mean+/-SD) and in 31 healthy control women, aged 19+/-2.3 years. Bone mineral density (BMD) in lumbar spine was measured by DEXA in the AN group. We found that 41 of the 43 patients had BMD z-scores under -2. No significant differences were found in the levels of serum bAP nor in PINP and beta-CTX levels between controls and patients, although values in the AN group were highly variable. All the BMD z-score values were negative, and their absolute value correlates positively with bAP (P = 0.0279) and almost with beta-CTX (P = 0.0921) but not with PINP (P = 0.4627). Bone AP correlates with PINP in control girls (P = 0.017), but not in the AN group (P = 0.3573). Patients with AN were divided into three groups according to their levels of bAP: low (I), normal (II) or high (III). Patients with the highest bAP levels also presented the highest increase in bone resorption, according to their beta-CTX levels, and the highest degree of osteopenia. However, values of PINP were similar in the three groups of patients. The bAP/beta-CTX ratios in subgroups I, II and III of AN patients were 0.035, 0.065 and 0.073, a finding that suggests that bAP is not indicating the real degree of bone mineralization in these patients, because it is a contradiction that the formation/resorption ratio should be higher in the patients who have the highest bone loss. These results could suggest that bone loss in AN is produced by an increase in bone resorption (beta-CTX), without variations in bone matrix formation (PINP); bAP levels are a good marker in the follow-up of osteopenia degree, but not a real indicator of bone mineralization, a similar situation to that of osteomalacia.