BON-10631; No. of pages: 10; 4C: Bone xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Bone journal homepage: www.elsevier.com/locate/bone

Review

Biochemical bone turnover markers in diabetes mellitus — A systematic review Jakob Starup-Linde a,b,⁎, Peter Vestergaard b,c a b c

Department of Endocrinology and Internal Medicine, Aarhus University Hospital THG, Aarhus, Denmark Department of Clinical Medicine, Aalborg University Hospital, Aalborg, Denmark Department of Endocrinology, Aalborg University Hospital, Aalborg, Denmark

a r t i c l e

i n f o

Article history: Received 18 November 2014 Revised 13 February 2015 Accepted 14 February 2015 Available online xxxx Keywords: Diabetes mellitus Bone turnover Bone turnover markers Fracture Bone Diabetes treatment

a b s t r a c t Background: Diabetes mellitus is associated with an increased risk of fractures, which is not explained by bone mineral density. Other markers as bone turnover markers (BTMs) may be useful. Aim: To assess the relationship between BTMs, diabetes, and fractures. Methods: A systematic literature search was conducted in August 2014. The databases searched were Medline at Pubmed and Embase. Medline at Pubmed was searched by “Diabetes Mellitus” (MESH) and “bone turnover markers” and Embase was searched using the Emtree by “Diabetes Mellitus” and “bone turnover”, resulting in 611 studies. The eligibility criteria for the studies were to assess BTM in either type 1 diabetes (T1D) or type 2 diabetes (T2D) patients. Results: Of the 611 eligible studies, removal of duplicates and screening by title and abstract lead to 114 potential studies for full-text review. All these studies were full-text screened for eligibility and 45 studies were included. Two additional studies were added from other sources. Among the 47 studies included there were 1 metaanalysis, 29 cross-sectional studies, 13 randomized controlled trials, and 4 longitudinal studies. Both T1D and T2D were studied. Most studies reported fasting BTM and excluded renal disease. Conclusion: Markers of bone resorption and formation seem to be lower in diabetes patients. Bone specific alkaline phosphatase is normal or increased, which suggests that the matrix becomes hypermineralized in diabetes patients. The BTMs: C-terminal cross-link of collagen, insulin-like growth factor-1, and sclerostin may potentially predict fractures, but longitudinal trials are needed. This article is part of a Special Issue entitled Bone and diabetes. © 2015 Elsevier Inc. All rights reserved.

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . Material and methods . . . . . . . . . . . . . 2.1. Data sources, searches, and eligibility criteria 2.2. Data extraction and quality assessment . . Results . . . . . . . . . . . . . . . . . . . . 3.1. Search results . . . . . . . . . . . . . . 3.2. Qualitative assessment . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . 4.1. BTM and diabetes . . . . . . . . . . . . 4.2. BTM in type 1 diabetes . . . . . . . . . 4.3. BTM in type 2 diabetes . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

0 0 0 0 0 0 0 0 0 0 0

Abbreviations: 25 OHD, 25 hydroxy vitamin D; BMD, bone mineral density; BAP, bone specific alkaline phosphatase; BTM, bone turnover marker; PICP, carboxy-terminal propeptide of type I procollagen; CICP, collagen type 1C propeptide; CTX, C-terminal cross-link of collagen; DPD, deoxypyridinoline; HP, hydroxyproline; IDDM, insulin dependent diabetes mellitus; IGF-1, insulin-like growth factor-1; IVGTT, intravenous glucose tolerance test; NTX, N-terminal propeptide type 1 collagen; OGTT, oral glucose tolerance test; OC, osteocalcin; OPG, osteoprotegerin; P1NP, procollagen type 1 N-terminal propeptide; RANKL, Receptor Activator of Nuclear factor Kappa beta Ligand; TRAP, tartrate resistant acid phosphatase; ICTP, type I collagen crosslinked carboxy-terminal telopeptide; T1D, type 1 diabetes; T2D, type 2 diabetes. ⁎ Corresponding author at: Department of Endocrinology and Internal Medicine (MEA), Aarhus University Hospital, Tage-Hansens Gade 2, 8000 Aarhus C, Denmark. Fax: +45 78467684. E-mail address: [email protected] (J. Starup-Linde).

http://dx.doi.org/10.1016/j.bone.2015.02.019 8756-3282/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

2

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

4.4. 4.5. 4.6.

BTM and fractures . . . . . . . . . . . Nephropathy and BTM . . . . . . . . . Diabetes treatment . . . . . . . . . . 4.6.1. Metformin, bone, and BTM . . . 4.6.2. DPP-IV inhibitors, bone, and BTM 4.6.3. Glitazones, bone, and BTM . . . 4.6.4. SGLT-2, bone, and BTM . . . . . 4.6.5. Statin use and BTM . . . . . . 4.6.6. Recommendations . . . . . . . 4.7. Antiresorptives and BTM . . . . . . . . 4.8. BTM and glucose . . . . . . . . . . . 4.9. BTM and bone biomechanical competence 4.10. Future usage of BTM . . . . . . . . . 5. Conclusion . . . . . . . . . . . . . . . . . . Disclosures . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

1. Introduction Diabetes mellitus is associated with an increased risk of fractures, which is not explained by bone mineral density (BMD) — at least measured by Dual energy X-ray absorptiometry (DXA) [1]. Furthermore, in the Fracture Risk Assessment Tool (FRAX) model, common risk factors and BMD underestimated the fracture risk in type 2 diabetes patients [2]. Although, many studies have investigated biochemical bone turnover markers (BTMs) in diabetes patients, no definite inferences can be made [3]. A meta-analysis reported significant heterogeneity among the biochemical markers of bone turnover as well as a dissociative pattern in formative and resorptive markers [4]. BTMs are chemical compounds whose presence can be detected in serum, plasma, or urine, and who ideally reflect bone turnover, i.e. resorption, formation or combinations of both [5]. These compounds may reflect 1) the mineralized matrix (hydroxyapatite, i.e. calcium and phosphate), 2) the non-mineralized matrix (collagen, osteocalcin (OC), matrix metalloproteinases, osteopontin, osteonectin etc.), and 3) the cellular matrix (osteoclasts, osteoblasts, and osteocytes). The compounds may either be a part of the matrix (OC, osteonectin, and osteopontin), precursors or degradation products of the matrix (pro-collagen or cross-links of collagen), enzymes (alkaline phosphatase, tartrate resistant acid phosphatase (TRAP)), or signaling substances (OC, sclerostin). Some compounds may have several roles (OC is a part of the unmineralized matrix, but also a signaling compound i.e. has hormonal properties, alkaline phosphatase is both an enzyme which initializes mineralization, and a marker of osteoblast function). Some compounds may thus both represent cellular function (alkaline phosphatase) and be an enzyme in the matrix. In diabetes patients, bone and bone markers may be affected through different pathways. The marker C-terminal cross-link of collagen (CTX) was decreased by food intake [6], however the mechanism was unknown. Both an oral glucose tolerance test (OGTT) and intravenous glucose tolerance test (IVGTT) decreased CTX and OC in healthy postmenopausal women; however the OGTT induced a significantly larger decrease in CTX than the IVGTT [6]. Furthermore, it is unknown whether the differences observed between OGTT and IVGTT were related to the administration form or dose of glucose [6]. Healthy obese subjects have reduced BTM compared to controls [7]. Following OGTT the BTM decreased in both obese and controls, however the decrease in osteocalcin (OC) was significantly more pronounced in controls [7]. Different and changing medication use may also affect bone in diabetes patients and glitazones were related to an increased fracture risk in women with type 2 diabetes (T2D) [8]. The growth factors; growth hormone and insulin-like growth factor-1 (IGF-1) may be

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

altered in diabetes and were both related to bone turnover [9]. Furthermore, comorbidities and complications to diabetes may alter BTM and modulate fracture risk. To further explore BTM in diabetes we conducted a systematic literature search. 2. Material and methods We used the PRISMA guidelines [10]. 2.1. Data sources, searches, and eligibility criteria A systematic literature search was conducted in August 2014. The databases searched were Medline at Pubmed and Embase. Medline at Pubmed was searched by using the key words “Diabetes Mellitus” (MESH) and “bone turnover markers” leading to 378 potential studies. Embase was searched using the Emtree with the key words “Diabetes Mellitus” and “bone turnover” limited to human studies revealing 233 potential studies. In total 611 studies were gathered. Additionally two papers [11,12], were too recently published to be indexed at the time of the literature search, but the authors found these papers by handsearching in their respective journals. The eligibility criteria for the studies were to assess bone turnover markers in either type 1 diabetes (T1D) or T2D patients. The following BTMs were included: CTX, N-terminal propeptide type 1 collagen (NTX), TRAP, deoxypyridinoline (DPD), hydroxyproline (HP), OC, 25 hydroxy vitamin D (25 OHD), procollagen type 1 N-terminal propeptide (P1NP), collagen type 1C propeptide (CICP), bone specific alkaline phosphatase (BAP), parathyroid hormone (PTH), sclerostin, osteoprotegerin (OPG), Receptor Activator of Nuclear factor Kappa beta Ligand (RANKL), IGF-1, inflammatory markers, and pentosidine. If several studies used the same population and BTM, the studies rated as poorest were excluded. Studies included in the recent meta-analysis [4] were not included in the study, as the meta-analysis itself was included. 2.2. Data extraction and quality assessment The data extracted from the studies were study design, randomization and intervention (if applicable), HbA1c, duration of the study (if applicable), number of participants, diabetes type, how diabetes was determined, which BTMs were measured, fasting status of the time of blood sample collection, and if the participants had renal disease. Study quality was not rated due to differences in design ranging from meta-analysis, randomized controlled trials, cohort studies, to crosssectional studies. All the different study types had their own strengths and limitations, and were not found comparable by the authors.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

3. Results 3.1. Search results Fig. 1 is a flowchart of the process of study selection. Of the 611 potential studies we removed duplicates and screened by title and abstract, which lead to 114 potential studies. All these studies were full-text screened for eligibility and 45 studies were included. Two additional studies were added [11,12] as these were in the scope of the systematic review but not part of the systematic literature search. Table 1 gives an overview of the studies and data collected. 3.2. Qualitative assessment 47 studies were included among these 1 meta-analysis, 29 crosssectional studies, 13 randomized controlled trials, and 4 longitudinal studies. Both T1D and T2D were studied. Study size varied from 24 [13] to 1605 [14]. In general diabetes ascertainment was good although some did not mention their source of diagnosis or diagnostic criteria [13,15–17]. HbA1c values were provided in several studies and ranged from 5.96 [18] to 11.3 [19]. Most studies reported fasting BTM values although 5 studies either did not report it, or samples were taken after a standardized meal, or only had fasting values on a subsample [17,20–23]. Almost all studies assessing BTM had renal disease as an exclusion criterion although, few studies did not give this information, and one study was conducted in a hemodialysis population [21]. 4. Discussion The included studies display large heterogeneity on study design and sample size. HbA1c was – if reported – quite high but varied much. The validity of the diabetes diagnosis was good. Only few studies did not use fasting values or excluded patients with renal impairment. Thus, the studies are compatible. Studies using non-fasting samples and individuals with renal impairment were included to assess the diversity in the population. Results in the studies using non-fasting

3

samples and patients with renal impairment [17,20–23] did not differ from the other studies. Although, it should be noted that Bunck et al. [17] evaluated BTM after a standardized meal, Bilezikian et al. [22] did an open label randomized trial to metformin or rosiglitazone, and Keegan et al. [23] was a trial randomized to alendronate or placebo and therefore the study results are difficult to compare to non-intervention trials. 4.1. BTM and diabetes A recent meta-analysis evaluating BTMs in both T1D and T2D based on 22 studies reported increased levels of alkaline phosphate in diabetes patients and decreased OC, CTX, and 25 OHD levels compared to controls [4]. Neither P1NP, NTX, calcium, DPD, CICP, BAP nor PTH differed statistically significantly from controls, but in general the BTM levels tended to be decreased in diabetes patients with the exception of urinary NTX (u-NTX) [4]. Heterogeneity between the studies such as patient characteristics and the use of different assays to measure BTMs may have influenced the results [24]. Inflammatory processes in diabetes may also contribute to bone alterations as few studies have addressed. Berberoglu et al. showed a significant negative correlation between BAP and TNF-α, whereas correlations between OC and inflammation markers all were insignificantly negative [18]. Another study yielded contradictive results; OC was negatively correlated with the inflammatory markers interleukin-6 and high-sensitivity CRP (hs-CRP) [25]. In T2D adiposity may give rise to a low grade inflammation that may lower BTMs. Further studies are needed to assess the possible effect of inflammation on BTM. 4.2. BTM in type 1 diabetes The meta-analysis reported decreased OC levels in T1D compared to controls, whereas other specific formation or resorption markers were not available in sufficient numbers for detailed analysis [4]. Recent studies add to this, as OC and TRAP levels were lower and CTX seemed to be lower at onset of T1D, but all markers normalized after 3 months

Fig. 1. Flowchart of study selection.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

4

Study Meta-analyses Starup-Linde et al. [4]

Participants

How to determine diabetes?

Comments

HbA1c

BTM measured

Samples taken in fasting condition

Renal disease

22 studies from a systematic literature search and additional 18 studies included in the metaregression

Different from study to study

Meta-analysis and metaregression available

–

OC, BAP, CTX, 25 OHD, u-NTX, 25 OHD, CICP, DPD

In some studies

In most studies

From outpatient clinic

8.1% in T1D, 9.1% in T2D

PTH, 25 OHD, 1,25 OHD, BAP, OC, P1CP, ICTP, IGF-1

No

Kidney failure excluded

–

OC, P1NP, u-NTX,

Yes

No other chronic diseases

8.8% in T1D

CTX, OPG, OC

Yes

All had normal eGFR

Cross-sectional studies Jehle et al. [20] 27 T1D, 25 T2D, 100 controls

Karaguzel et al. [26] Alexopoulou et al. [27] Lumachi et al. [15]

58 T1D, 44 controls

ADA guidelines

42 male T1D, 24 male controls 18 IDDM, 21 controls

From diabetes clinic

Age, BMI and HbA1c were significantly higher in T2D than T1D. Diabetes duration was longer in T1D than T2D. Both boys and girls. Mean age 11.7 years for T1D 11 T1D with microalbuminuria

Confirmed — but no mentioning of how

Mix of genders. BMI 22.5 in IDDM, thus likely to be T1D

–

PTH, OC, BAP

Yes

Akin et al. [30]

57 T2D PM, 20 controls PM

Outpatient Clinic

BMI significantly lower in controls

OC, u-NTX, BAP

Yes

Reyes-Garcia et al. [31]

78 T2D, 55 controls

ADA guidelines

BAP, OC, TRAP, CTX, PTH, 25 OHD

Yes

No renal disease

Yamamoto et al. [32] Jiajue et al. [11]

255 T2D, 240 controls

Outpatient clinic Self-report and a fasting glucose ≥7.0 mmol/l

9.1% in T2D, 5.6/5.7% in controls –

PTH, OC, P1NP, CTX, 25 OHD CTX, P1NP, 25 OHD

Yes

236 T2D PM, 1055 controls PM

Yes

Excluded if serum creatinine was higher than normal range Renal disease excluded

Farr et al. [33]

30 T2D PM, 30 controls PM

ADA guidelines

Both men and women T2D with significantly higher age and BMI Vertebral fractures in 27.7% of T2D and 21.7% of controls Mix of genders, females were PM. VF in 90 T2D VF and non-VF fractures assessed in the population. Controls were significantly younger than T2D. BMI significantly lower in controls. Performs microindentation Performs bone biopsies on 9 subjects

9.76% in T2D 5.10% in controls 8.01% in T2D

Serum creatinine borderline significantly higher in IDDM than controls Chronic diseases excluded

P1NP, CTX, 25 OHD

Yes Yes Yes

Renal disease excluded

BMI significantly lower in controls VF in 20% of T2D and 7.5% of controls

7.27% in T2D, 4.94% in controls

PTH, 25 OHD, P1NP, BAP, OC, TRAP-5b, CTX PTH, 25 OHD, OC, CTX, P1NP 25 OHD, BAP, CTX, Sclerostin, Dkk-1, β-catenin PTH, IGF-1, 25 OHD, sclerostin, OC, P1NP, CTX, u-NTX P1NP, CTX, PTH, 25 OHD

Stage 4 and 5 chronic kidney diseases excluded eGFR b60 ml/min excluded

Evaluating the FRAX algorithm

7.7% in T2D, 5.4% in controls 8.4% in T2D, 5.8% in controls 6.9/7.3% in T2D

Yes

Renal bone disease excluded

Yes

Renal bone disease excluded

Yes

Manavalan 18 T2D PM, 27 controls PM et al. [34] Bhattoa et al. [35] 68 male T2D, 68 male controls Gaudio et al. [36] 40 T2D PM, 40 controls PM

At least 1 year use of antiglycemic medication Department of Internal Medicine ADA guidelines

Ardawi et al. [37]

482 T2D PM, 482 controls PM

ADA guidelines

VF in 24.5% of T2D and none in controls 9.6% in T2D, 4.71% in controls

Hernandez et al. [58]

2431 subjects of these 45 T2D

Fasting glucose N 6.9 mmol/l

PM females and older men

–

WHO criteria

T2D was newly diagnosed.

–

OC, interleukin-6, Hs-CRP

Yes

ADA guidelines

The diabetes group is a subgroup of the total population. BMI significantly higher in NIDDM

–

OPG, RANKL, OC, CTX

Yes

Coexisting medical disorder that might affect bone metabolism was excluded. Other systemic disorders excluded Renal osteodystrophy excluded

–

OC, TRAP, HP, PTH

Yes

No renal disorders

Investigates both cardiovascular and bone disease. 32.6% of T2D with VF and 20% of controls with VF Assigns a diabetic nephropathy group based on a serum creatinine higher

8.1% in T2D

OPG

Yes

–

–

OC, BAP, HP

Yes

No history of metabolic bone disease

Sarkar and 108 T2D, 50 controls Choudhury [25] Movahed 382 PM of these 102 T2D et al. [38] Sosa et al. [40] 47 female NIDDM, 252 female controls Rozas-Moreno 43 male T2D, 25 male controls et al. [41] Chen et al. [45]

55 T2D, 27 controls

National Diabetes Data Group criteria ADA guidelines

WHO criteria

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

Table 1 Characteristics of the included studies.

OC, BAP u-NTX

Yes

No renal dysfunction

9.1%

IGF-, OC, u-NTX

Yes

VFs in 33.2% of females and 34.7% of males Mix of genders. VF in 47 women and 72 men Mix of genders. All women premenopausal. Prevalent fractures in 24 individuals VF in 10 T2D patients and 11 non-T2D patients

8.5% in females and 8.7% in males 9.38% for men and 9.22 for women 7.7%

u-NTX, OC, IGF-1

Yes

None with renal dysfunction or macro-albuminuria None with renal dysfunction

BAP, PTH, u-NTX, Sclerostin OC, CTX, pentosidine

Yes

6.39% in T2D

PTH

VF in 28 males and 20 females. Non-VF fractures in 25 individuals All premenopausal. No description of control selection in the methods section All African American. Both genders represented Investigates renal function and BTM

9.1%

BAP, u-NTX, pentosidine

Yes

8% in T1D

Yes

8.1%

PTH, OC, BAP, HP, pyridinoline, DPD, sclerostin Sclerostin

Excluded if serum creatinine was higher than normal range Chronic renal disease excluded

Yes

Mean creatinine 1.0 mg/dl

9.9% in T1D, 9.2% in T2D

HP

–

–

12 week randomization to rosiglitazone and diet or diet alone Randomized to metformin or sitagliptin for 12 weeks Randomized to rosiglitazone, metformin or glyburide treatment. Evaluated after 12 months. Randomized to vildagliptin or placebo for 1 year. Only included if HbA1c b 7.6%

6.34% treated T2D, 5.96 in non-treated T2D 8.2 and 8.0% depending on treatment group Between 7.33 and 7.37% depending on treatment arm

BAP, OC, DPD, interleukins Yes 1 and 6, TNF-α OC, DPD Yes

Renal disorders excluded

PTH, 25 OHD, CTX, P1NP, BAP

Yes

Renal impairment was excluded.

6.0%

CTX

Outpatient clinics

Randomized to rosiglitazone or placebo. All with prior CVD. 6 months duration of the trial

7.6% in both groups

P1NP, PTH, 25 OHD, CTX, OPG

No mention Standardized meal and markers measured after this Yes No mention of exclusion

T2D and prior antidiabetic therapy with diet and exercise alone (drug-naive) or prior monotherapy (non-glitazone) for more than 2 weeks in the past 12 weeks From the PRAMID study

Randomized to rosiglitazone or metformin for 52 weeks and 24 weeks open label

6.8%

Vitamin D, PTH, CTX, P1NP, BAP

No information

Significant renal disease excluded

Sclerostin, CTX, P1NP

Yes

All had normal renal function

OPG

Yes

Impaired renal function excluded

838 T2D

University hospital

Kanazawa et al. [46] Kanazawa et al. [47] Yamamoto et al. [48] Neumann et al. [12]

131 T2D PM

University hospital

479 T2D men, 334 T2D PM 321 T2D

University hospital

128 T1D

Clinical criteria and GAD65 positive

Inaba et al. [21]

31 T2D HD patients, 83 non-T2D HD patients

Yamamoto et al. [50] Catalano et al. [43]

153 T2D

Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus Outpatient clinic

69 T1D, 10 controls

Outpatient clinic

450 T2D

Self report

73 T1D, 67 T2D, 75 controls

Diabetes clinic

Randomized controlled trials Berberoglu 56 obese T2D PM, et al. [18] 26 controls PM Hegazy [52] 40 diabetic PM Zinman et al. [14] 1605 T2D (689 women, 916 men) Bunck et al. [17]

59 T2D

Gruntmanis et al. [54]

111 T2D, 56 in rosiglitazone treatment, 55 in placebo treatment 225 T2D PM

Bilezikian et al. [22]

van Lierop et al. [39]

71 male T2D, 30 male controls

Nybo et al. [55]

371 T2D

Outpatient clinic

OGTT University hospital From the ADOPT study. FPG between 126 and 180 mg/dl –

Hospital centers

24 week randomization to pioglitazone – or metformin. Baseline compared with controls Randomized to eight treatment – groups based NPH insulin or Aspart insulin in combinations with rosiglitazone and metformin. 2 year duration of the trial

Yes

Excluded if serum creatinine was higher than normal range eGFR b30 ml/min excluded

All in hemodialysis

Renal diseases excluded

5

(continued on next page)

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

8.8% in men, 8.7% in women

Kanazawa et al. [49]

Register et al. [44] Selby et al. [51]

than 120 μmol/l or u-albumin excretion higher than 15 μg/ml All women PM. VF in 269 individuals. Investigate the association of antidiabetics and VF VF in 33 individuals

6

Study

Participants

Randomized controlled trials Ljunggren 180 T2D et al. [57] Braatvedt 25 T2D et al. [59]

Ikeda et al. [13]

24 T2D PM

Keegan et al. [23]

6458 PM of these 297 T2D

Mori et al. [61]

144 T2D PM

Longitudinal studies Okazaki et al. [16] 33 T2D

How to determine diabetes?

Comments

HbA1c

BTM measured

Samples taken in fasting condition

Renal disease

Randomized to dapagliflozin or placebo. 50 week analysis Crossover trial. Randomized to 12 weeks of placebo or atorvastatin 8 week washout and switch to placebo or active treatment Not given Randomized to alendronate or alfacalcidol for 12 months. u-NTX had to be more than 40 nmol/mmol creatinine. Data from the fracture intervention Self-report and use trial. Post hoc analysis of antidiabetic treatment or an elevated random plasma randomized controlled trial. Randomization to alendronate or glucose value placebo for 3 years Department of Medicine Randomization to no medication, alfacalcidol or raloxifene

7.16 and 7.19 in treatment groups –

P1NP, CTX

Yes

CTX, BAP, OC

Yes

No mention (1 with nephropathy) No mention

8.0 in alendronate group and 8.1 in control group

u-NTX,

Yes

Renal failure excluded

–

CTX, NTX, BAP

20% had fasting specimens

Factors that affect bone turnover were excluded

Ranging from 6.8 to 7.2

BAP, NTX, pentosidine

Yes

No information

No mention of source

4 troglitazone week treatment. Mix of genders Follow-up after 5 years

–

DPD, u-NTX, BAP, OC

Yes

No mention of exclusion

Yes

Baseline creatinine 104 and 84 μmol/l

Mix of genders. All children. Follow from T1D onset and 12 months forward Pioglitazone added to therapy for 3 months. Mix of pre- and postmenopausal among women

11.3% at T1D onset 7.1% at 12 months after T1D onset. 5.6/5.5% in controls 8.2%

OC, CTX, TRAP5b

Yes

No other chronic diseases

BAP, P1NP, OC, CTX

Yes

Serum creatinine N136 excluded

HbA1c 6.5–8–5%, FPG b 13.3 mmol/l Established T2D

Hamilton et al. [28] longitudinal Pater et al. [19]

26 T1D, 27 T2D

From the Fremantle diabetes study

17 T1D, 17 controls

From pediatrics department

Xiao et al. [56]

70 T2D

FPG N 7.0 b 13.1

Baseline 7.9 T1D and 7.4 T2D CTX, OC, PTH

Postmenopausal (PM), type 2 diabetes (T2D), type 1 diabetes (T1D), American Diabetes Association (ADA), estimated glomerular filtration rate (eGFR), vertebral fracture (VF), Non-insulin dependent diabetes mellitus (NIDDM), insulin dependent diabetes mellitus (IDDM), hemodialysis (HD), cardiovascular disease (CVD), fasting plasma glucose (FPG), oral glucose tolerance test (OGTT), 25 hydroxy vitamin D (25 OHD), bone mineral density (BMD), bone specific alkaline phosphatase (BAP), collagen type 1C propeptide (CICP), C-terminal cross-link of collagen (CTX), deoxypyridinoline (DPD), hydroxyproline (HP), insulin-like growth factor-1 (IGF-1), N-terminal propeptide type 1 collagen (NTX), osteocalcin (OC), osteoprotegerin (OPG), procollagen type 1 N-terminal propeptide (P1NP), Receptor Activator of Nuclear factor Kappa beta Ligand (RANKL), tartrate resistant acid phosphatase (TRAP), type I collagen cross-linked carboxy-terminal telopeptide (ICTP).

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

Table 1 (continued)

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

compared to controls. However, HbA1c fell from 11.3% (100 mmol/mol) to 5.9% (41 mmol/mol) in the same period [19], which may point at an effect of better diabetes control. In children the levels of OC, P1NP, and PTH were lower when compared with controls, whereas u-NTX did not differ [26]. Another study found no difference in OC, OPG or CTX levels in adult, male T1D compared with controls [27]. Seemingly children with T1D have a decreased bone turnover which may normalize in adulthood. However, adults with insulin dependent diabetes mellitus (IDDM) and normal BMI had lower OC and BAP levels than controls [15]. Bone markers did not seem to change in adult T1D as a 5 year longitudinal study found no changes in OC or CTX; however CTX showed a trend towards an increase [28]. The results of the longitudinal study may either reflect normal bone marker levels in adult T1D with a bone resorption increase over time, or a general suppression with an increase in bone resorption over time not complemented by a bone formation increase. T1D patients tend to have lower IGF-1 levels than T2D [20], therefore the mechanism of the lower bone turnover may be mediated by IGF-1 deficiency in T1D. Even so, T1D and T2D do not seem to differ in regard to BTMs, as neither 25 OHD, carboxy-terminal propeptide of type I procollagen (PICP), type I collagen cross-linked carboxy-terminal telopeptide (ICTP), BAP, nor OC was different between the groups [20]. T1D seems to have decreased bone turnover when evaluated by BTM, however a human histomorphometric study showed no difference in bone metabolism in T1D compared to non-diabetics [29]. Thus, evidence is conflicting. Low bone turnover in T1D may be related to different diabetes stages, complications, and metabolic regulation or to a specific age group, gender, or treatment. 4.3. BTM in type 2 diabetes The meta-analysis found borderline significantly decreased OC levels (P = 0.06) and increased levels of alkaline phosphate in T2D compared to non-diabetics [4]. Newer evidence suggested BTMs to be decreased, as CTX, OC, P1NP, TRAP, u-NTX, and PTH were lower in T2D than controls [11,25,30–38]. Also, IGF-1, sclerostin, osteocalcin, u-NTX, and BAP were all decreased in T2D [36,37]. Results are still conflicting, as studies have reported no differences in the BTMs or even increased levels in T2D. P1NP [35,39], OC, TRAP, alkaline phosphatase, OC, DPD, and HP were not different when comparing T2D with controls [18,31,40]. A longitudinal study found no differences in OC after 5 years, but an increase in CTX in T2D [28]. The differences between studies may be explained by differences in metabolic status, diabetes duration, and medication at the time of the measurement. OPG levels have been found to be higher in T2D than controls [38,41]. Thus OPG/RANKL signaling may be involved in the disturbed bone turnover in T2D. However, RANKL did not differ between T2D and controls [38]. In this context it should be noted that RANKL may be difficult to measure [42]. The effect on BTMs may be caused by alterations in WNT-signaling as sclerostin levels were elevated in T2D compared to controls in several studies [36,39], but not in T1D [43]. Furthermore, sclerostin was found to be positively correlated with BMD [44], which may explain previously found differences in BMD between T1D and T2D [1]. A single study reported rather variable levels of BTM in T2D compared to controls, which may reflect changes in different stages of the bone turnover cycle. In this study by Chen et al. [45], OC was decreased and HP was increased among T2D compared to non-diabetics, whereas BAP did not differ. The differences in BTMs may explain the paradox of low bone strength and increased BMD, as markers of bone resorption and formation seem to be lower, whereas the mineralization reflected by BAP is normal. Thus, the lower bone turnover may not be coupled to a similar lowering of mineralization, and the matrix becomes hypermineralized. This hypermineralization may lead to a frail “osteopetrotic” bone. The hypermineralization theory is supported by Manavalan et al. [34] who found increased BAP and otherwise lower

7

or normal BTMs, and a reduced bone formation by histomorphometric analysis on a small subgroup. Likewise BAP was not different when comparing T2D with controls [30,31] in studies that found that other BTMs decreased. 4.4. BTM and fractures Cross-sectional studies examining fracture risk and BTM have been conducted with different outcomes. Decreased IGF-1, decreased OC levels, and increased sclerostin levels are all evidence of decreased bone turnover, and are associated with vertebral fractures in T2D. Decreased IGF-1 levels were associated with vertebral fractures in T2D [37,46]. Furthermore, IGF-1 was inversely associated with the number of vertebral fracture in postmenopausal women with T2D, but not in men with T2D [47]. Caution should be made when interpreting the results as IGF-1 levels may be altered due to insulin treatment and diabetes severity. Decreased OC levels were also found as a marker of three or more fractures among postmenopausal women with T2D [47], likewise increased sclerostin levels were associated with vertebral fractures in T2D [37,48]. The association of the combination of low P1NP and high CTX with fractures in T2D further highlights the relationship between low bone turnover and fractures in T2D [11]. Furthermore, u-NTX was increased in T2D postmenopausal women with a vertebral fracture compared to those without fracture [49]. However, this relation was not seen in men [49]. It is uncertain whether OC is useful as a marker of vertebral fracture as it and the other commonly used markers of bone turnover PTH, BAP, CTX, and u-NTX were not associated with vertebral fractures in T2D [12,21,46,49,50]. The Advanced Glycation End-product marker pentosidine also shows conflicting evidence in terms of association with fractures: Pentosidine was positively associated with vertebral fractures in postmenopausal T2D women but not men [50]. Another study evaluating any fracture in T1D found a slight increase of fracture related to increased pentosidine levels, however the 95% CI stretches from 1.01 to 1.03 and was practically 1, whereas HbA1c had a more clear relationship (odds ratio 1.93, 95% CI 1.16–3.20) [12]. 4.5. Nephropathy and BTM Nephropathy may change bone turnover and alter fracture risk in diabetes. Increased HP excretion was found to be related to microalbuminuria in T1D and T2D [51]. Likewise higher levels of OC and HP were found among T2D with nephropathy than in patients without nephropathy [45]. Nephropathy may modulate the fracture risk, however little is known on the effect on BTM and further studies are needed. 4.6. Diabetes treatment 4.6.1. Metformin, bone, and BTM Only a few trials investigated BTM and metformin use, although metformin may reduce bone formation as evaluated by P1NP. Diabetic women randomized to metformin for 12 weeks did not change DPD or OC [52]. Metformin decreased P1NP compared to pioglitazone [39]. P1NP decreased more in metformin treated women than in rosiglitazone treated women, whereas BAP did not differ between treatment groups [14]. 4.6.2. DPP-IV inhibitors, bone, and BTM Dipeptidyl peptidase-4 (DPP-IV) inhibitors was associated with a decreased fracture risk in a meta-analysis, although caution should be made due to short trial periods (mean trial duration 35 weeks) [53]. Little is known of the effects of DPP-IV on BTM. 12 weeks of sitagliptin therapy decreased DPD levels, but did not change OC in postmenopausal diabetic women as well as OC and DPD levels were lower in sitagliptin

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

8

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

users than metformin users [52]. However, 12 month vildagliptin treatment did not change the bone resorption marker CTX in T2D, when evaluated after a standardized meal [17]. The results are thus conflicting, and BTM levels may either not change or decrease. 4.6.3. Glitazones, bone, and BTM Randomized studies have evaluated the effect of glitazones on BTM in T2D and reported a decrease in CTX levels: Glitazones were shown to increase CTX in T2D [14,22,39,54]. However, Zinman et al. could not find the association among men [14]. Three of these studies compared the results to patients randomized to metformin treatment [14,22,39, 54]. Pioglitazone also seemed to increase sclerostin levels compared to metformin [39]. The increase in sclerostin may increase the bone resorption. Further evidence of a decrease in bone turnover is backed by the fact that 12 weeks of rosiglitazone therapy decreased BAP [18], and that BAP decreased during 52 weeks of glitazone therapy compared to metformin [22]. Also, troglitazone was shown to decrease DPD, u-CTX, BAP, and OC [16]. However, data from randomized trials on BMTs are conflicting, as P1NP increased in rosiglitazone treated women compared to metformin users during 52 weeks of therapy [22]. OPG levels were found to be decreased after rosiglitazone treatment in another study [55]. Some studies also reported unchanged levels; P1NP and OPG did not change after glitazone treatment [54] like OC and DPD did not change [18]. A prospective study showed a decreased in P1NP, and BAP, whereas OC and CTX were unchanged after glitazone treatment [56]. 4.6.4. SGLT-2, bone, and BTM The Sodium–glucose co-transporter 2 (SGLT-2) inhibitors have recently been added to the list of oral antidiabetics and are a second line therapy against T2D. It leads to increased dieresis, and it has been speculated that the drug may affect the skeleton by an increased calcium excretion. However, the SGLT-2 inhibitor dapagliflozin did not change P1NP or CTX compared to placebo in T2D [57]. 4.6.5. Statin use and BTM Statin treatment may be life-long for some diabetes patients. It is unsure whether statin treatment may affect BTMs. P1NP and CTX have been reported to be lower in T2D statin users than non-T2D statin users, which were not observed for non-T2D [58]. Furthermore, statin use for more than 3 years was associated with lower CTX levels than among those receiving statin treatment for 1–3 years. In contrast, a randomized placebo controlled crossover trial with 12 weeks of atorvastin treatment found no differences in BAP, OC, or CTX [59]. The different results in these studies [58,59] may be due to differences in study design, lower number in the randomized trial, and too short treatment duration. Further studies are needed to evaluate long term effects of statins on BTM. 4.6.6. Recommendations Based on the current knowledge, caution should be made when applying glitazones to diabetes treatment due to negative effects on bone health [8]. No definite recommendations can be made on other antidiabetic drugs or statins. Statins may – due to its CTX lowering effect – have a protective effect on fractures, which has been hypothesized previously in non-diabetic patients [60]. Also, DPP-IV inhibitors potentially decrease fracture risk, however the results must be interpreted cautiously [53]. Head to head randomized controlled trials are needed to determine fracture risk and possible preventive effects of antidiabetic drugs. Only a single study reported results of sulfonylureas [14]. This study found no differences in CTX and BAP when comparing glyburide treatment with rosiglitazone treatment. P1NP was lower in men treated with rosiglitazone than those in SU treatment [14]. The relation between insulin treatment and BTMs has not been evaluated. BTMs need to be evaluated in diabetes treatment trials to determine the effect of antidiabetics on bone, as little is known.

4.7. Antiresorptives and BTM In targeting the diabetic bone disease antiresorptives may be used. Randomization to alendronate decreased u-NTX significantly compared to vitamin D in treated T2D subjects [13], and decreased CTX, NTX, and BAP compared to placebo in T2D [23]. Furthermore, randomization to raloxifene treatment decreased NTX, but not BAP in T2D and pentosidine levels were unaffected [61]. A registry based nationwide Danish cohort study concluded that antiresorptive treatment was as effective in preventing fractures in diabetics as in non-diabetics [62]. Thus, the suppressed bone resorption seems beneficial in fracture prevention in diabetes patients. 4.8. BTM and glucose The included studies display large heterogeneity in HbA1c levels. The HbA1c may not be as important as the glucose value at the time of blood sampling, as OGTT decreases BTM with a subsequent rise [6,63]. HbA1c reflects the general glycemic state in the previous 3 months, but plasma glucose may at times during these 3 months be normal, which is especially likely during fasting, which was seen in most studies included. The plasma glucose levels may affect BTM directly or through mechanisms of insulin, incretins or other gastrointestinal hormones. Octreotide, an inhibitor of insulin, incretins, and gastrointestinal hormones, abolished the effect of the OGTT [63], and the effect may thus be secondary to glucose. However, it may also imply, that glucose uptake in osteoblasts and osteoclasts, mediated by the combination of available insulin and glucose, decreases BTM. Insulin signaling and IGF signaling in osteoblasts are suggested to play an important factor for bone quality as shown in animal models [64] and highlight the need to distinguish the effects of insulin and glucose in experimental human studies. Based on prior studies, insulin increases OC secretion in osteoblasts [65], which may increase insulin sensitivity [66] and increase pancreatic insulin production [64]. A bone–glucose axis, mediated by osteocalcin and insulin, with bi-directional effects may exist, however further studies are needed to confirm and elaborate these effects. In vitro studies suggest, that high glucose levels (30 mM or more) per se act by decreasing bone resorption (reflected by pit formation in osteoclasts), increasing production of an unmineralized matrix (reflected by collagen I production in osteoblasts), and decreasing mineralization (reflected by alkaline phosphatase) [67,68]. These glucose levels are non-physiological for a diabetes patient, and studies using glucose concentrations of 10–15 mM glucose are needed. However, it underlines the effect of glucose on bone turnover, but does not explain the hypermineralization and increased BMD in diabetes patients. Diabetes patients may be in a state of cyclic varying bone turnover, where bone turnover is decreased at high glucose levels and normal at normal glucose levels. The mechanisms of glucose induced decreases of BTM, and the effects of glucose on formation of hydroxyapatite are not fully understood and further research is needed. 4.9. BTM and bone biomechanical competence BTM may yield a predictive value of fractures and longitudinal studies with fracture as primary endpoint are needed. Furthermore, BTMs have to some extent been related to bone biomechanical competence. Farr et al. [33] found both lower bone material strength, evaluated by microindentation, and lower BTMs in T2D. Microindentation and BTMs may thus be related. Histomorphometry in diabetes patients has been performed in few studies, and the only study also reporting BTMs only has biopsies from five T2D patients [34]. This study found a general decrease in both BTMs and bone formation evaluated by histomorphometry, with the exception of BAP, which was increased [34]. Thus, further research into metabolism and compression of bone biopsies, and microindentation, and the relation to BTMs is needed.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

4.10. Future usage of BTM BTMs are poorly related to fracture risk in diabetes patients as bone marker levels differ from study to study. At present, no single marker can evaluate bone biochemical competence. As CTX is a commonly used bone marker, it may also hold the potential as a fracture predictor. Based on current knowledge, antiresorptive treatment and glitazones have opposite effects in terms of bone health and fractures, and the therapies display opposite patterns for CTX in diabetes patients [14,22,23,39,54]. Thus, CTX may be a predictor of fracture risk. A substantial study-to-study variation is seen for CTX levels, which could be caused by differences in patient characteristics. Besides the potential use of CTX; IGF-1 [37,46] and sclerostin [37,48] may hold the potential to predict fractures. However, longitudinal studies evaluating these markers are needed. Determination of bone turnover in diabetes may require bone biopsies to reveal the microscopic changes, which may allow correlations between BTM and chemical investigations of collagen glycosylation. The heterogeneity of diabetes patients, in terms of complications, diabetes duration, medication, obesity, comorbidities, and HbA1c may complicate future investigations. Registry based studies may help determine if particular diabetic characteristics are associated with fractures and thus allow investigation of the group at risk. 5. Conclusion BTMs have been widely investigated in diabetes patients. BTMs seem to be lower in diabetes patients, but with large heterogeneity between studies. Markers of bone resorption and formation seem to be lower whereas BAP, an enzyme of mineralization, is normal to increased, and suggests that the matrix becomes hypermineralized in diabetes patients. This may explain the paradox of low bone strength and increased BMD. Cross-sectional trials have evaluated the association between BTMs and fracture. CTX, IGF-1, and sclerostin may potentially predict fractures, but longitudinal trials are needed. Currently, of the antidiabetics only glitazones are known to have negative effects on bone and BTMs by increasing CTX, whereas others appear to have neutral effects. Alendronate is associated with decreased CTX levels in diabetes patients, and has the same fracture preventing effect in diabetes patients as in non-diabetes patients. Little is known of the effects of other antiresorptive or anabolic therapies. Research is needed, to determine the effects of glucose on bone turnover and BTMs, as a clear relationship is present, but not well understood or examined. In conclusion, some BTMs may yield the potential to predict fractures in diabetes patients. However, little is known of the mechanisms affecting bone. Further investigation of the effect of glucose on bone in diabetes patients is needed. Disclosures None. Acknowledgments None. References [1] Vestergaard P. Discrepancies in bone mineral density and fracture risk in patients with type 1 and type 2 diabetes — a meta-analysis. Osteoporos Int 2007;18:427–44. [2] Giangregorio LM, Leslie WD, Lix LM, Johansson H, Oden A, McCloskey E, et al. FRAX underestimates fracture risk in patients with diabetes. J Bone Miner Res 2012;27: 301–8. [3] Starup-Linde J. Diabetes, biochemical markers of bone turnover, diabetes control, and bone. Front Endocrinol 2013;4.

9

[4] Starup-Linde J, Eriksen SA, Lykkeboe S, Handberg A, Vestergaard P. Biochemical markers of bone turnover in diabetes patients — a meta-analysis, and a methodological study on the effects of glucose on bone markers. Osteoporosis Int 2014;25: 1697–708. [5] P. Szulc, D.C. Bauer, R. Eastell, eds., Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism chapter 35, biochemical markers of bone turnover in osteoporosis (pages 297–306). Eighth Edition, Editor(s): Clifford J. Rosen ed., Print ISBN: 9781118453889, Online ISBN: 9781118453926, DOI: 10.1002/9781118453926, 19 JUL 2013. [6] Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, Henriksen DB, Christiansen C. Mechanism of circadian variation in bone resorption. Bone 2002; 30:307–13. [7] Viljakainen H, Ivaska KK, Paldanius P, Lipsanen-Nyman M, Saukkonen T, Pietilainen KH, et al. Suppressed bone turnover in obesity: a link to energy metabolism? A case– control study. J Clin Endocrinol Metab 2014;99:2155–63. [8] Bazelier MT, de Vries F, Vestergaard P, Herings RM, Gallagher AM, Leufkens HG, et al. Risk of fracture with thiazolidinediones: an individual patient data meta-analysis. Front Endocrinol (Lausanne) 2013;4:11. [9] Locatelli V, Bianchi VE. Effect of GH/IGF-1 on bone metabolism and osteoporosis. Int J Endocrinol 2014;2014:235060. [10] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009; 339:b2700. [11] Jiajue R, Jiang Y, Wang O, Li M, Xing X, Cui L, et al. Suppressed bone turnover was associated with increased osteoporotic fracture risks in non-obese postmenopausal Chinese women with type 2 diabetes mellitus. Osteoporos Int 2014;25:1999–2005. [12] Neumann T, Lodes S, Kastner B, Franke S, Kiehntopf M, Lehmann T, et al. High serum pentosidine but not esRAGE is associated with prevalent fractures in type 1 diabetes independent of bone mineral density and glycaemic control. Osteoporos Int 2014; 25:1527–33. [13] Ikeda T, Manabe H, Iwata K. Clinical significance of alendronate in postmenopausal type 2 diabetes mellitus. Diabetes Metab 2004;30:355–8. [14] Zinman B, Haffner SM, Herman WH, Holman RR, Lachin JM, Kravitz BG, et al, ADOPT Study Group. Effect of rosiglitazone, metformin, and glyburide on bone biomarkers in patients with type 2 diabetes. J Clin Endocrinol Metab 2010;95:134–42. [15] Lumachi F, Camozzi V, Tombolan V, Luisetto G. Bone mineral density, osteocalcin, and bone-specific alkaline phosphatase in patients with insulin-dependent diabetes mellitus. Ann N Y Acad Sci 2009;1173(Suppl. 1):E64–7. [16] Okazaki R, Miura M, Toriumi M, Taguchi M, Hirota Y, Fukumoto S, et al. Short-term treatment with troglitazone decreases bone turnover in patients with type 2 diabetes mellitus. Endocr J 1999;46:795–801. [17] Bunck MC, Poelma M, Eekhoff EM, Schweizer A, Heine RJ, Nijpels G, et al. Effects of vildagliptin on postprandial markers of bone resorption and calcium homeostasis in recently diagnosed, well-controlled type 2 diabetes patients. J Diabetes 2012;4: 181–5. [18] Berberoglu Z, Gursoy A, Bayraktar N, Yazici AC, Tutuncu NB, Demirag NG. Rosiglitazone decreases serum bone-specific alkaline phosphatase activity in postmenopausal diabetic women. J Clin Endocrinol Metab 2007;92:3523–30. [19] Pater A, Sypniewska G, Pilecki O. Biochemical markers of bone cell activity in children with type 1 diabetes mellitus. J Pediatr Endocrinol Metab 2010;23:81–6. [20] Jehle PM, Jehle DR, Mohan S, Bohm BO. Serum levels of insulin-like growth factor system components and relationship to bone metabolism in type 1 and type 2 diabetes mellitus patients. J Endocrinol 1998;159:297–306. [21] Inaba M, Okuno S, Kumeda Y, Yamakawa T, Ishimura E, Nishizawa Y. Increased incidence of vertebral fracture in older female hemodialyzed patients with type 2 diabetes mellitus. Calcif Tissue Int 2005;76:256–60. [22] Bilezikian JP, Josse RG, Eastell R, Lewiecki EM, Miller CG, Wooddell M, et al. Rosiglitazone decreases bone mineral density and increases bone turnover in postmenopausal women with type 2 diabetes mellitus. J Clin Endocrinol Metab 2013;98: 1519–28. [23] Keegan TH, Schwartz AV, Bauer DC, Sellmeyer DE, Kelsey JL, fracture intervention trial. Effect of alendronate on bone mineral density and biochemical markers of bone turnover in type 2 diabetic women: the fracture intervention trial. Diabetes Care 2004;27:1547–53. [24] Bauer D, Krege J, Lane N, Leary E, Libanati C, Miller P, et al. National bone health alliance bone turnover marker project: current practices and the need for US harmonization, standardization, and common reference ranges. Osteoporos Int 2012;23: 2425–33. [25] Sarkar PD, Choudhury AB. Relationships between serum osteocalcin levels versus blood glucose, insulin resistance and markers of systemic inflammation in central Indian type 2 diabetic patients. Eur Rev Med Pharmacol Sci 2013;17:1631–5. [26] Karaguzel G, Akcurin S, Ozdem S, Boz A, Bircan I. Bone mineral density and alterations of bone metabolism in children and adolescents with type 1 diabetes mellitus. J Pediatr Endocrinol Metab 2006;19:805–14. [27] Alexopoulou O, Jamart J, Devogelaer JP, Brichard S, de Nayer P, Buysschaert M. Bone density and markers of bone remodeling in type 1 male diabetic patients. Diabetes Metab 2006;32:453–8. [28] Hamilton EJ, Rakic V, Davis WA, Paul Chubb SA, Kamber N, Prince RL, et al. A fiveyear prospective study of bone mineral density in men and women with diabetes: the Fremantle Diabetes Study. Acta Diabetol 2012;49:153–8. [29] Armas LA, Akhter MP, Drincic A, Recker RR. Trabecular bone histomorphometry in humans with type 1 diabetes mellitus. Bone 2012;50:91–6. [30] Akin O, Gol K, Akturk M, Erkaya S. Evaluation of bone turnover in postmenopausal patients with type 2 diabetes mellitus using biochemical markers and bone mineral density measurements. Gynecol Endocrinol 2003;17:19–29.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019

10

J. Starup-Linde, P. Vestergaard / Bone xxx (2015) xxx–xxx

[31] Reyes-Garcia R, Rozas-Moreno P, Lopez-Gallardo G, Garcia-Martin A, Varsavsky M, Aviles-Perez MD, et al. Serum levels of bone resorption markers are decreased in patients with type 2 diabetes. Acta Diabetol 2013;50:47–52. [32] Yamamoto M, Yamaguchi T, Nawata K, Yamauchi M, Sugimoto T. Decreased PTH levels accompanied by low bone formation are associated with vertebral fractures in postmenopausal women with type 2 diabetes. J Clin Endocrinol Metab 2012;97: 1277–84. [33] Farr JN, Drake MT, Amin S, Melton 3rd LJ, McCready LK, Khosla S. In vivo assessment of bone quality in postmenopausal women with type 2 diabetes. J Bone Miner Res 2014;29:787–95. [34] Manavalan JS, Cremers S, Dempster DW, Zhou H, Dworakowski E, Kode A, et al. Circulating osteogenic precursor cells in type 2 diabetes mellitus. J Clin Endocrinol Metab 2012;97:3240–50. [35] Bhattoa HP, Onyeka U, Kalina E, Balogh A, Paragh G, Antal-Szalmas P, et al. Bone metabolism and the 10-year probability of hip fracture and a major osteoporotic fracture using the country-specific FRAX algorithm in men over 50 years of age with type 2 diabetes mellitus: a case–control study. Clin Rheumatol 2013;32:1161–7. [36] Gaudio A, Privitera F, Battaglia K, Torrisi V, Sidoti MH, Pulvirenti I, et al. Sclerostin levels associated with inhibition of the Wnt/beta-catenin signaling and reduced bone turnover in type 2 diabetes mellitus. J Clin Endocrinol Metab 2012;97: 3744–50. [37] Ardawi MS, Akhbar DH, Alshaikh A, Ahmed MM, Qari MH, Rouzi AA, et al. Increased serum sclerostin and decreased serum IGF-1 are associated with vertebral fractures among postmenopausal women with type-2 diabetes. Bone 2013;56:355–62. [38] Movahed A, Larijani B, Nabipour I, Kalantarhormozi M, Asadipooya K, Vahdat K, et al. Reduced serum osteocalcin concentrations are associated with type 2 diabetes mellitus and the metabolic syndrome components in postmenopausal women: the crosstalk between bone and energy metabolism. J Bone Miner Metab 2012;30: 683–91. [39] van Lierop AH, Hamdy NA, van der Meer RW, Jonker JT, Lamb HJ, Rijzewijk LJ, et al. Distinct effects of pioglitazone and metformin on circulating sclerostin and biochemical markers of bone turnover in men with type 2 diabetes mellitus. Eur J Endocrinol 2012;166:711–6. [40] Sosa M, Dominguez M, Navarro MC, Segarra MC, Hernandez D, de Pablos P, et al. Bone mineral metabolism is normal in non-insulin-dependent diabetes mellitus. J Diabetes Complications 1996;10:201–5. [41] Rozas Moreno P, Reyes Garcia R, Garcia-Martin A, Varsavsky M, Garcia-Salcedo JA, Munoz-Torres M. Serum osteoprotegerin: bone or cardiovascular marker in type 2 diabetes males? J Endocrinol Invest 2013;36:16–20. [42] Jorgensen L, Vik A, Emaus N, Brox J, Hansen JB, Mathiesen E, et al. Bone loss in relation to serum levels of osteoprotegerin and nuclear factor-kappaB ligand: the Tromso Study. Osteoporos Int 2010;21:931–8. [43] Catalano A, Pintaudi B, Morabito N, Di Vieste G, Giunta L, Bruno ML, et al. Gender differences in sclerostin and clinical characteristics in type 1 diabetes mellitus. Eur J Endocrinol 2014;171:293–300. [44] Register TC, Hruska KA, Divers J, Bowden DW, Palmer ND, Carr JJ, et al. Sclerostin is positively associated with bone mineral density in men and women and negatively associated with carotid calcified atherosclerotic plaque in men from the African American-Diabetes Heart Study. J Clin Endocrinol Metab 2014;99:315–21. [45] Chen H, Li X, Yue R, Ren X, Zhang X, Ni A. The effects of diabetes mellitus and diabetic nephropathy on bone and mineral metabolism in T2DM patients. Diabetes Res Clin Pract 2013;100:272–6. [46] Kanazawa I, Yamaguchi T, Yamamoto M, Yamauchi M, Yano S, Sugimoto T. Serum insulin-like growth factor-I level is associated with the presence of vertebral fractures in postmenopausal women with type 2 diabetes mellitus. Osteoporos Int 2007;18:1675–81. [47] Kanazawa I, Yamaguchi T, Sugimoto T. Serum insulin-like growth factor-I is a marker for assessing the severity of vertebral fractures in postmenopausal women with type 2 diabetes mellitus. Osteoporos Int 2011;22:1191–8. [48] Yamamoto M, Yamauchi M, Sugimoto T. Elevated sclerostin levels are associated with vertebral fractures in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab 2013;98:4030–7.

[49] Kanazawa I, Yamaguchi T, Yamamoto M, Sugimoto T. Relationship between treatments with insulin and oral hypoglycemic agents versus the presence of vertebral fractures in type 2 diabetes mellitus. J Bone Miner Metab 2010;28:554–60. [50] Yamamoto M, Yamaguchi T, Yamauchi M, Yano S, Sugimoto T. Serum pentosidine levels are positively associated with the presence of vertebral fractures in postmenopausal women with type 2 diabetes. J Clin Endocrinol Metab 2008;93:1013–9. [51] Selby PL, Shearing PA, Marshall SM. Hydroxyproline excretion is increased in diabetes mellitus and related to the presence of microalbuminuria. Diabet Med 1995;12: 240–3. [52] Hegazy SK. Evaluation of the anti-osteoporotic effects of metformin and sitagliptin in postmenopausal diabetic women. J Bone Miner Metab 2014;33(2):207–12. [53] Monami M, Dicembrini I, Antenore A, Mannucci E. Dipeptidyl peptidase-4 inhibitors and bone fractures: a meta-analysis of randomized clinical trials. Diabetes Care 2011; 34:2474–6. [54] Gruntmanis U, Fordan S, Ghayee HK, Abdullah SM, See R, Ayers CR, et al. The peroxisome proliferator-activated receptor-gamma agonist rosiglitazone increases bone resorption in women with type 2 diabetes: a randomized, controlled trial. Calcif Tissue Int 2010;86:343–9. [55] Nybo M, Preil SR, Juhl HF, Olesen M, Yderstraede K, Gram J, et al. Rosiglitazone decreases plasma levels of osteoprotegerin in a randomized clinical trial with type 2 diabetes patients. Basic Clin Pharmacol Toxicol 2011;109:481–5. [56] Xiao WH, Wang YR, Hou WF, Xie C, Wang HN, Hong TP, et al. The effects of pioglitazone on biochemical markers of bone turnover in the patients with type 2 diabetes. Int J Endocrinol 2013;2013:290734. [57] Ljunggren O, Bolinder J, Johansson L, Wilding J, Langkilde AM, Sjostrom CD, et al. Dapagliflozin has no effect on markers of bone formation and resorption or bone mineral density in patients with inadequately controlled type 2 diabetes mellitus on metformin. Diabetes Obes Metab 2012;14:990–9. [58] Hernandez JL, Olmos JM, Romana G, Martinez J, Castillo J, Yezerska I, et al. Bone turnover markers in statin users: a population-based analysis from the Camargo Cohort Study. Maturitas 2013;75:67–73. [59] Braatvedt GD, Bagg W, Gamble G, Davidson J, Reid IR. The effect of atorvastatin on markers of bone turnover in patients with type 2 diabetes. Bone 2004;35:766–70. [60] Rejnmark L, Vestergaard P, Mosekilde L. Statin but not non-statin lipid-lowering drugs decrease fracture risk: a nation-wide case–control study. Calcif Tissue Int 2006;79:27–36. [61] Mori H, Okada Y, Kishikawa H, Inokuchi N, Sugimoto H, Tanaka Y. Effects of raloxifene on lipid and bone metabolism in postmenopausal women with type 2 diabetes. J Bone Miner Metab 2013;31:89–95. [62] Vestergaard P, Rejnmark L, Mosekilde L. Are antiresorptive drugs effective against fractures in patients with diabetes? Calcif Tissue Int 2011;88:209–14. [63] Clowes JA, Allen HC, Prentis DM, Eastell R, Blumsohn A. Octreotide abolishes the acute decrease in bone turnover in response to oral glucose. J Clin Endocrinol Metab 2003;88: 4867–73. [64] Pramojanee SN, Phimphilai M, Chattipakorn N, Chattipakorn SC. Possible roles of insulin signaling in osteoblasts. Endocr Res 2014;39(4):144–51. [65] Ferron M, Wei J, Yoshizawa T, Del Fattore A, DePinho RA, Teti A, et al. Insulin signaling in osteoblasts integrates bone remodeling and energy metabolism. Cell 2010; 142:296–308. [66] Ferron M, McKee MD, Levine RL, Ducy P, Karsenty G. Intermittent injections of osteocalcin improve glucose metabolism and prevent type 2 diabetes in mice. Bone 2012;50:568–75. [67] Cunha JS, Ferreira VM, Maquigussa E, Naves MA, Boim MA. Effects of high glucose and high insulin concentrations on osteoblast function in vitro. Cell Tissue Res 2014;358(1):249–56. [68] Xu F, Ye YP, Dong YH, Guo FJ, Chen AM, Huang SL. Inhibitory effects of high glucose/ insulin environment on osteoclast formation and resorption in vitro. J Huazhong Univ Sci Technolog Med Sci 2013;33:244–9.

Please cite this article as: Starup-Linde J, Vestergaard P, Biochemical bone turnover markers in diabetes mellitus — A systematic review, Bone (2015), http://dx.doi.org/10.1016/j.bone.2015.02.019