2. The association between lumbar bone mineral density and advanced glycation end products derived from confocal fluorescence microscopy: a prospective investigation of bone biopsies in patients undergoing lumbar spinal fusion surgery

Abstract

BACKGROUND CONTEXT Quantitative computed tomography (QCT) is considered a standard reportable metric for the measurement of bone mineral density (BMD) of the lumbar spine. However, the sole determination of BMD is insufficient for the evaluation of overall bone health. Consequently, bone quality, apart from BMD, is increasingly recognized for its importance in the assessment of bone. Collagen cross-linking is a relevant factor for determining the tensile strength of bone. Non-enzymatic collagen cross-linking with the accumulation of advanced glycation end products (AGEs) stiffens and embrittles collagen fibers, thus increasing bone fragility and fracture risk. Despite the crucial role of AGEs in bone ageing, the relationship between AGEs and BMD is poorly understood. PURPOSE We hypothesized that an accumulation of AGEs, a marker of impaired bone quality, is related to decreased BMD. STUDY DESIGN/SETTING Prospective observational study. PATIENT SAMPLE A total of 107 patients undergoing open posterior lumbar fusion at a single academic institution. OUTCOME MEASURES Standardized lumbar QCT measurements were performed on preoperative CT scans using the L1/L2 average. Intraoperative bone biopsies from the posterior superior iliac spine were obtained and analyzed with confocal fluorescence microscopy for fluorescent advanced glycation end products (fAGEs), both trabecular and cortical, representing non-enzymatic collagen cross-linking. METHODS Pearson's correlation coefficients were calculated to examine crude relationships between BMD at L1/L2 and fAGEs, stratified by biological sex. Multivariable linear regression analysis with adjustments for age, sex, body mass index (BMI), diabetes and HbA1c was used to investigate associations between BMD at L1/L2 and fAGEs. Statistical significance was defined as p<0.05. RESULTS A total of 107 patients (51.2% female) with a mean age of 61.2 years and a BMI of 29.9 kg/m2 were included in the final analysis, excluding patients with anti-osteoporotic drug therapy. In the univariate analysis, cortical fAGEs increased with decreasing BMD at L1/L2 (r = -0.32; p = 0.021), but only in men. In the multivariate analysis, trabecular fAGEs increased with decreasing BMD at L1/L2 after adjusting for age, sex, BMI, diabetes mellitus, and HbA1c (β = 0.99; p = 0.04). CONCLUSIONS This is the first study to assess BMD and in-vivo non-enzymatic collagen cross-linking as a marker of bone quality in lumbar fusion patients. QCT-derived BMD measurements were found to be significantly associated with collagen properties in bone, such as fAGEs, which embrittle bone tissue. Our results enhance our understanding of bone health by suggesting that spine surgery patients with deficits in bone quantity may also have deficits in bone quality. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs. Quantitative computed tomography (QCT) is considered a standard reportable metric for the measurement of bone mineral density (BMD) of the lumbar spine. However, the sole determination of BMD is insufficient for the evaluation of overall bone health. Consequently, bone quality, apart from BMD, is increasingly recognized for its importance in the assessment of bone. Collagen cross-linking is a relevant factor for determining the tensile strength of bone. Non-enzymatic collagen cross-linking with the accumulation of advanced glycation end products (AGEs) stiffens and embrittles collagen fibers, thus increasing bone fragility and fracture risk. Despite the crucial role of AGEs in bone ageing, the relationship between AGEs and BMD is poorly understood. We hypothesized that an accumulation of AGEs, a marker of impaired bone quality, is related to decreased BMD. Prospective observational study. A total of 107 patients undergoing open posterior lumbar fusion at a single academic institution. Standardized lumbar QCT measurements were performed on preoperative CT scans using the L1/L2 average. Intraoperative bone biopsies from the posterior superior iliac spine were obtained and analyzed with confocal fluorescence microscopy for fluorescent advanced glycation end products (fAGEs), both trabecular and cortical, representing non-enzymatic collagen cross-linking. Pearson's correlation coefficients were calculated to examine crude relationships between BMD at L1/L2 and fAGEs, stratified by biological sex. Multivariable linear regression analysis with adjustments for age, sex, body mass index (BMI), diabetes and HbA1c was used to investigate associations between BMD at L1/L2 and fAGEs. Statistical significance was defined as p<0.05. A total of 107 patients (51.2% female) with a mean age of 61.2 years and a BMI of 29.9 kg/m2 were included in the final analysis, excluding patients with anti-osteoporotic drug therapy. In the univariate analysis, cortical fAGEs increased with decreasing BMD at L1/L2 (r = -0.32; p = 0.021), but only in men. In the multivariate analysis, trabecular fAGEs increased with decreasing BMD at L1/L2 after adjusting for age, sex, BMI, diabetes mellitus, and HbA1c (β = 0.99; p = 0.04). This is the first study to assess BMD and in-vivo non-enzymatic collagen cross-linking as a marker of bone quality in lumbar fusion patients. QCT-derived BMD measurements were found to be significantly associated with collagen properties in bone, such as fAGEs, which embrittle bone tissue. Our results enhance our understanding of bone health by suggesting that spine surgery patients with deficits in bone quantity may also have deficits in bone quality.