Clinical Endocrinology (2014) 81, 350–355
doi: 10.1111/cen.12472
ORIGINAL ARTICLE
Relationship between insulin sensitivity and bone mineral density in primary hyperparathyroidism Laura Gianotti*, Francesco Tassone*, Claudia Baffoni*, Micaela Pellegrino*, Sara Cassibba*, Elena Castellano*, Gianpaolo Magro*, Flora Cesario*, Gianluca Visconti† and Giorgio Borretta* *Division of Endocrinology, Department of Internal Medicine, S.Croce & Carle Hospital, and †Division of Clinical Pathology, S.Croce & Carle Hospital, Cuneo, Italy
Introduction Summary Introduction Evidence of crosstalk between bone and insulin metabolism has been identified. In primary hyperparathyroidism (PHPT), scant data exist on this relationship. Aim To evaluate the relationship between insulin levels or sensitivity and bone mineral density (BMD) in PHPT. Subject and Methods Two hundred and sixty-seven patients with PHPT without known diabetes mellitus were studied. Fasting blood glucose and insulin levels as well as BMD at lumbar spine, femoral neck and forearm were measured. Insulin sensitivity was assessed using Quantitative Insulin Sensitivity Check Index (QUICKI). The same parameters were measured 2 years (interquartile range 28 years) after surgery (PTX) in a subgroup of patients (n = 51). Results In univariate analysis, a positive relationship between insulin levels and BMD (R = 017, P < 003) or T-score (R = 020, P < 0005) was found at femoral neck level. Consequently, a negative relationship between QUICKI and femoral BMD (R = 020, P < 001) or T-score (R = 021, P < 0004) was found. In multivariate analysis, when femoral BMD was the dependent variable, age (beta = 035, P < 0000004), BMI (beta = 039, P < 000001), PTH (beta = 018, P < 005) and QUICKI (R = 015, P < 005) had an independent effect (R2 = 029). Insulin levels and QUICKI did not change after PTX. No relationship was found between QUICKI or insulin levels at the time of diagnosis and change in BMD at any site at follow-up. Conclusions Our data show a weak relationship between insulin levels and/or insulin sensitivity and BMD in PHPT. However, the insulin state does not influence change in bone density after PTX in PHPT. (Received 27 December 2013; returned for revision 20 January 2014; finally revised 29 March 2014; accepted 14 April 2014)
Primary hyperparathyroidism (PHPT) is a common endocrine disorder, leading mostly to bone damage and osteoporosis, even in asymptomatic patients.1,2 PHPT is also associated with derangement of insulin or glucose metabolism.3–5 Both hypercalcaemia and parathyroid hormone (PTH) excess can impair insulin sensitivity.3–6 In fact, a significant relationship between hypercalcaemia and insulin sensitivity has been found, and serum calcium has been reported to predict insulin sensitivity in PHPT.6 However, it is still unclear whether glucose tolerance or insulin sensitivity normalizes after parathyroidectomy (PTX).7,8 In the last decade, an interplay between bone and glucose or insulin metabolism has been identified. Insulin has been reported to have an anabolic effect on bone.9,10 Insulin receptors have been found on osteoblasts, and insulin signalling in these cells enhances osteocalcin (OC) activity, which ultimately increases insulin sensitivity.9 Thus, a circular relationship between bone and glucose/insulin metabolism is emerging.9,10 In humans, some findings indicate that the increase in bone mineral density (BMD), which is found in obesity, metabolic syndrome and type 2 diabetes mellitus (DM), can be connected to high insulin levels and insulin resistance.11–14 In PHPT, very few data exist on the relationship between insulin and bone metabolism. It has been found that fasting blood glucose, but not insulin, was positively related to BMD in menopausal women with PHPT.15 Based on these premises, the aim of our study was to investigate the relationship between insulin levels and insulin sensitivity and BMD in a large group of adult patients with PHPT without known DM. A subgroup of patients was also studied after PTX, to verify whether insulin sensitivity could influence changes in BMD after PTX.
Methods Study Correspondence: Laura Gianotti, Division of Endocrinology & Metabolism, S. Croce & Carle Hospital, Via Michele Coppino 26, 12100 Cuneo, Italy. Tel.: +39 0171 642030; Fax: +39 0171 642075; E-mail: [email protected]
350
This is a retrospective cross-section study with a subsequent analysis of a subgroup of patients who underwent PTX. The patients were studied at the Division of Endocrinology and © 2014 John Wiley & Sons Ltd
Insulin and bone in primary hyperparathyroidism 351 Metabolism of S. Croce and Carle Hospital, Cuneo, Italy, from 1996 through 2010. All evaluations were performed during a single hospitalization period. At diagnosis, the patients underwent a screening that included complete family and personal medical history, physical examination (including body weight, height and blood pressure) and a biochemical evaluation of general blood and urinary parameters, including total and ionized serum calcium, serum phosphate, serum creatinine, serum PTH, total alkaline phosphatase (ALP), serum 25OHD3, 24-h urinary calcium, fasting blood glucose levels as well as fasting insulin levels. Furthermore, patients underwent neck ultrasound, parathyroid scintigraphy, X-ray evaluation of the phalanges and skull, and BMD measurement through dual-energy X-ray absorptiometry (DXA) at lumbar spine, femoral neck and forearm. Biochemical and BMD measurements were also routinely performed at 12–36 months after PTX. Data obtained in a subgroup of patients (n = 51) who underwent PTX were collected at short-time follow-up after PTX (time since PTX, median: 2 years; interquartile range 28 years). All patients gave their informed consent both at diagnosis and at follow-up. Subjects Two hundred and sixty-seven consecutive patients with PHPT were studied (Table 1). Patients with PHPT associated with MEN or FHH were excluded. None of the patients had severe liver disease or overt renal failure (as creatinine-derived eGFR
doi: 10.1111/cen.12472
ORIGINAL ARTICLE
Relationship between insulin sensitivity and bone mineral density in primary hyperparathyroidism Laura Gianotti*, Francesco Tassone*, Claudia Baffoni*, Micaela Pellegrino*, Sara Cassibba*, Elena Castellano*, Gianpaolo Magro*, Flora Cesario*, Gianluca Visconti† and Giorgio Borretta* *Division of Endocrinology, Department of Internal Medicine, S.Croce & Carle Hospital, and †Division of Clinical Pathology, S.Croce & Carle Hospital, Cuneo, Italy
Introduction Summary Introduction Evidence of crosstalk between bone and insulin metabolism has been identified. In primary hyperparathyroidism (PHPT), scant data exist on this relationship. Aim To evaluate the relationship between insulin levels or sensitivity and bone mineral density (BMD) in PHPT. Subject and Methods Two hundred and sixty-seven patients with PHPT without known diabetes mellitus were studied. Fasting blood glucose and insulin levels as well as BMD at lumbar spine, femoral neck and forearm were measured. Insulin sensitivity was assessed using Quantitative Insulin Sensitivity Check Index (QUICKI). The same parameters were measured 2 years (interquartile range 28 years) after surgery (PTX) in a subgroup of patients (n = 51). Results In univariate analysis, a positive relationship between insulin levels and BMD (R = 017, P < 003) or T-score (R = 020, P < 0005) was found at femoral neck level. Consequently, a negative relationship between QUICKI and femoral BMD (R = 020, P < 001) or T-score (R = 021, P < 0004) was found. In multivariate analysis, when femoral BMD was the dependent variable, age (beta = 035, P < 0000004), BMI (beta = 039, P < 000001), PTH (beta = 018, P < 005) and QUICKI (R = 015, P < 005) had an independent effect (R2 = 029). Insulin levels and QUICKI did not change after PTX. No relationship was found between QUICKI or insulin levels at the time of diagnosis and change in BMD at any site at follow-up. Conclusions Our data show a weak relationship between insulin levels and/or insulin sensitivity and BMD in PHPT. However, the insulin state does not influence change in bone density after PTX in PHPT. (Received 27 December 2013; returned for revision 20 January 2014; finally revised 29 March 2014; accepted 14 April 2014)
Primary hyperparathyroidism (PHPT) is a common endocrine disorder, leading mostly to bone damage and osteoporosis, even in asymptomatic patients.1,2 PHPT is also associated with derangement of insulin or glucose metabolism.3–5 Both hypercalcaemia and parathyroid hormone (PTH) excess can impair insulin sensitivity.3–6 In fact, a significant relationship between hypercalcaemia and insulin sensitivity has been found, and serum calcium has been reported to predict insulin sensitivity in PHPT.6 However, it is still unclear whether glucose tolerance or insulin sensitivity normalizes after parathyroidectomy (PTX).7,8 In the last decade, an interplay between bone and glucose or insulin metabolism has been identified. Insulin has been reported to have an anabolic effect on bone.9,10 Insulin receptors have been found on osteoblasts, and insulin signalling in these cells enhances osteocalcin (OC) activity, which ultimately increases insulin sensitivity.9 Thus, a circular relationship between bone and glucose/insulin metabolism is emerging.9,10 In humans, some findings indicate that the increase in bone mineral density (BMD), which is found in obesity, metabolic syndrome and type 2 diabetes mellitus (DM), can be connected to high insulin levels and insulin resistance.11–14 In PHPT, very few data exist on the relationship between insulin and bone metabolism. It has been found that fasting blood glucose, but not insulin, was positively related to BMD in menopausal women with PHPT.15 Based on these premises, the aim of our study was to investigate the relationship between insulin levels and insulin sensitivity and BMD in a large group of adult patients with PHPT without known DM. A subgroup of patients was also studied after PTX, to verify whether insulin sensitivity could influence changes in BMD after PTX.
Methods Study Correspondence: Laura Gianotti, Division of Endocrinology & Metabolism, S. Croce & Carle Hospital, Via Michele Coppino 26, 12100 Cuneo, Italy. Tel.: +39 0171 642030; Fax: +39 0171 642075; E-mail: [email protected]
350
This is a retrospective cross-section study with a subsequent analysis of a subgroup of patients who underwent PTX. The patients were studied at the Division of Endocrinology and © 2014 John Wiley & Sons Ltd
Insulin and bone in primary hyperparathyroidism 351 Metabolism of S. Croce and Carle Hospital, Cuneo, Italy, from 1996 through 2010. All evaluations were performed during a single hospitalization period. At diagnosis, the patients underwent a screening that included complete family and personal medical history, physical examination (including body weight, height and blood pressure) and a biochemical evaluation of general blood and urinary parameters, including total and ionized serum calcium, serum phosphate, serum creatinine, serum PTH, total alkaline phosphatase (ALP), serum 25OHD3, 24-h urinary calcium, fasting blood glucose levels as well as fasting insulin levels. Furthermore, patients underwent neck ultrasound, parathyroid scintigraphy, X-ray evaluation of the phalanges and skull, and BMD measurement through dual-energy X-ray absorptiometry (DXA) at lumbar spine, femoral neck and forearm. Biochemical and BMD measurements were also routinely performed at 12–36 months after PTX. Data obtained in a subgroup of patients (n = 51) who underwent PTX were collected at short-time follow-up after PTX (time since PTX, median: 2 years; interquartile range 28 years). All patients gave their informed consent both at diagnosis and at follow-up. Subjects Two hundred and sixty-seven consecutive patients with PHPT were studied (Table 1). Patients with PHPT associated with MEN or FHH were excluded. None of the patients had severe liver disease or overt renal failure (as creatinine-derived eGFR
