The added value of trabecular bone score over bone mineral density for identication of the bone architecture in patients with Inammatory Bowel Disease

Background and aims: Patients with Inammatory Bowel Disease (IBD) suffer from different level of bone diseases in density and architectures. This study has been designed to measure Bone Mineral Density (BMD) and Trabecular Bone Score (TBS) in IBD patients, Golestan province, northeast of Iran. Materials and methods: In this cross-sectional study during January 2018 to August 2018, 69 patients with IBD (UC=57 and Crohn’s disease=10; male: female 0.74; mean age, 39 years) recruited from the IBD registry in Golestan Research Center of Gastroentrology and Hepatology. After taking the informed consent, all patients have been interviewed by a trained physician and a questionnaire has been completed. Trabecular bone score (TBS) was performed using lumbar spine dual X-ray absorptiometry (DXA) images. After entering data into the SPSS-16 software, analysis was done using X2 and non-parametric tests. Results: Normal


Introduction
In ammatory bowel diseases (IBDs) are the chronic in ammatory diseases of the gastrointestinal tract which consists of Crohn's disease (CD) and ulcerative colitis (UC) (1). Many patients have extraintestinal symptoms in addition to intestinal complaints (2)(3)(4). Extraintestinal manifestations most commonly affect the joints, skin and eyes, and some other organs. There are two kinds of extraintestinal complications, some are caused by the disease itself and some are caused by treatments (3,5).
Previous studies have shown decreased bone mineral density in patients with IBD (6-8). Calcium malabsorption, altered sex hormonal status, malnutrition and vitamin D de ciency are possible factors for osteoporosis associated with IBD (7,9). Bone density measurement with Dual Energy X-ray Absorptiometry (DXA) is currently the best means of detecting osteoporosis. This method can measure the amount of bone minerals which is an important factor in bone strength. But there is remarkable overlap in BMD values between individuals who develop fractures and those who do not (10,11). There are other factors involved in bone strength and fracture risk such as macro geometry of cortical bone, the microarchitecture of trabecular bone, bone microdamage, mineralization, and turnover (12,13).
The trabecular bone score (TBS) is a new texture measurement that can be applied to any X-ray images. Signi cant correlations have been identi ed between TBS and 3D parameters of bone microarchitecture, independent of any correlation between TBS and BMD (14,15). Also, TBS could predict fractures almost as well as BMD but the combination was reported superior to either measurement alone (10,15). The aim of the study was to evaluate the added value of TBS over BMD for identi cation of the bone architecture in patients with in ammatory bowel disease.

Study population
The study population included patients with biopsy-proven in ammatory bowel disease (IBD) registered in the Golestan Research Center of Gastroentrology and Hepatology (GRCGH) in Gorgan, Northeast of Iran; during January 2018 to August 2018. All registered patients with IBD (N=100) have been invited through phone calls to participate in the study. The exclusion criteria included cases with severe disease activity, hospital admission at the time of study and during the last month, history of trauma after the diagnosis of IBD, pregnant women and not willing to have an x-ray or children younger than 18 years.
Patient's demographic information (age, sex, ethnicity), colonoscopy and biopsy results from patient records, as well as information on vitamin D and calcium supplements and menopause of female patients were recorded.

Imaging
All patients were referred to the imaging center for evaluation of bone mineral density (BMD) and trabecular bone score (TBS). The device Hologic Horizon WI Bone Densitometer was used in the center. BMD was measured in the lumbar spine and femoral neck using DXA method. In accordance with international standards, T-score less than one standard deviation (SD) compared to maximal adult test was considered as bone mass reduction and less than 2.5 standard deviation as osteoporosis.
Bone density was measured in the lumbar spine (L2-L4) and proximal femur (femoral neck, trochanter and Ward) using a dual-energy X-ray absorptiometry (Lunar). TBS (absolute value) was obtained after re-analyzing DXA images from lumbar spine scan using TBS software and Horizon device. TBS was calculated as the mean of individual measurements for L1-L4 vertebrae (table 1).
Fracture Risk Assessment Tool (FRAX) was also calculated based on TBS, to measure the risk of fracture in 10-years in all subjects.

Statistics
After entering the data in SPSS version 16, the data were analyzed using mean, standard deviation, frequency and percentage. Chi-square test was used to compare the qualitative variables.
As seen in table 3, comparing the results of BMD and TBS in IBD patients showed some discrepancies between the results, and in some cases (N=3) TBS showed partially degraded bone while BMD showed them normal. On the other hand, 22 has normal report in TBS but low density or osteoporosis in BMD (Table 3). Sex and IBD type had shown no signi cant differences in TBS, BMD and FRAX, although females had higher risk of fracture and lower TBS or BMD (Table 4).
Age had a signi cant correlation with the three measurements, but BMI, waist circumference and abdominal circumference had been correlated signi cantly with TBS. The duration of disease was also correlated signi cantly with FRAX ( Figure 1). FRAX-adjusted-MOF was signi cantly correlated with TBS (r=-0.528, P<0.001) and BMD results (r=-0.411, P<0.001). There was also a signi cant positive correlation between TBS and BMD (r=0.414, P<0.001).

Discussion
In the present study, our results indicated stronger correlation between FRAX and TBS than FRAX and BMD, similarly to the previous studies (16), that raises the question whether TBS is a more accurate tool for predicting bone involvement than BMD. It has been shown that FRAX adjusted with TBS has a better predictive role than FRAX alone (16).
Correlation between the spine BMD and TBS was signi cant in the present study, similar to the previous studies that suggested enhanced gradients of risks when the results of TBS and BMD are combined together compared to the TBS or FRAX alone (17). According to the Hans et al, when BMD levels are close to the intervention threshold, TBS can be more useful in predicting the risk of future fracture risk (4). Similarly, our results showed that according to the BMD results, six patients considered to have osteoporosis, while TBS results showed that two of them does not need any intervention and classi ed as the normal group.
Our results suggested that the correlations between TBS and WC, AC, and BMI are stronger than BMD which raises the question whether obesity affects bone microarchitecture more than bone densit. Rates of obesity are rising in patients with IBD, as in the general population; 15-40% of adults with IBD are obese, and 20-40% are overweight (18,19). Previous studies showed that the higher adiposity in early life in pediatrics with IBD and also in newly diagnosed IBD patients is signi cantly associated with reduced bone mass, as shown by BMD z-score (20,21).
Obesity or fat accumulation affects bone in different ways. The fact that mesenchymal stem cells generates both osteoblasts and adipocytes is the cornerstone of the connections between their metabolisms. Obesity has positive effects on bone formation by decreasing apoptosis and increasing proliferation and differentiation of osteoblasts and osteocytes (22)(23)(24).
BMI or bodyweight is also showed a positive correlation with bone mineral density or bone mass. But there are evidences that shows negative consequences of fat accumulation on total bone mineral density and content; for instance, increased adiposity is a risk factor for bone fracture (25-26).
As the source of these two types of the cells is the same, increase adipocytes differentiation may lead to decrease osteoblast differentiation (27). Also, elevation of proin ammatory cytokines such as including TNF-a, IL-1, and IL-6 due to obesity increase bone resorption and bone loss in IBD patients (28).
On the other hand, high-fat diet decreases intestinal absorption of calcium (29). These evidences show that the effects of obesity and its related condition on bone are two sided and yet these interactions are not clearly de ned. Considering these facts, obesity effects could have different faces in BMD and TBS.
Introduction of TBS to the routine methods of bone densitometry, showed better prediction of the fraction risk. In a cohort study in Korea (2001-2014) 4000 Chinese men and women aged 65 years old or above were recruited; It showed that the combination of TBS to the BMD could better predict MOF compared to BMD alone, especially in older men (30). On the other hand, Nassar et al (2013, France) reported that TBS and BMD could detect patients with and without vertebral fractures, but when BMD is in the nonosteoporotic range, TBS would add more information to BMD lumbar spine alone and could better show spine deterioration (31).
According to our results, BMD was reduced in both UC and CD patients while some studies demonstrated that the reduction of BMD only appears in CD patients, not in UC patients (2). This con ict can lead us to further investigations.
The major limitation in this cross-sectional study was that we evaluated the density and trabecular bone mass in IBD patients at a time point with no follow up.

Conclusions
Age, BMI, waist circumference and abdominal circumference had been correlated signi cantly with bone weakness in IBD patients. The duration of disease was also correlated signi cantly with risk of major osteoporotic fracture in 10-years (FRAX-adjusted-MOF). So, it should be noti ed that early diagnosis of patients with chronic diseases that are at higher risk of fractures would have more economical bene ts for the community.   The correlation between TBS, FRAX and BMD with demographic variables in IBD patients

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.