Altered cortical bone strength and lean mass in young women with long‑duration (19 years) type 1 diabetes

 Daniel Novak 1,2 , Gun Forsander 1,2 , Eva Kristiansen1,2 , Anna Svedlund 1,2 , Per Magnusson 3*  & Diana Swolin‑Eide 1,2 

1Department of Paediatrics, Institute for Clinical Sciences, Sahlgrenska Academy, University of Gothenburg,

Gothenburg, Sweden. 2Region Västra Götaland, Sahlgrenska University Hospital, Department of Paediatrics,

Gothenburg, Sweden. 3Department of Clinical Chemistry, and Department of Biomedical and Clinical Sciences,

Linköping University, 581 85 Linköping, 



To investigate bone health and body composition in young women with long‑duration type 1 diabetes (T1D) in relation to matched controls. Twenty‑three Swedish women, age 19.2–27.9 years, with a T1D duration of 10 years or more were recruited from the Swedish National Diabetes Registry (NDR). An age‑, gender‑ and geography‑matched control group was recruited. Bone mass and body composition were assessed by dual‑energy X‑ray absorptiometry and peripheral quantitative computedtomography. Data was retrieved from the NDR and SWEDIABKIDS registries.

T1D individuals had a mean diabetes duration of 19 years. T1D individuals had reduced lean mass (40.0 ± 6.1 kg vs.43.9 ± 4.9 kg) and were shorter (1.66 ± 0.06 m vs. 1.71 ± 0.06 m) although comparable BMI. Subjects with T1D had lower muscle area (P = 0.0045).

No differences were observed for fractures; physical activity; total, lumbar spine or femur areal bone mineral density.

The cortical bone strength strain index was lower for TD1 patients (1875 ± 399 mm3 vs. 2277 ± 332 mm3). In conclusion, young women with long‑term diabetes duration showed reduced cortical bone strength, decreased periosteal circumference, endosteal circumference and altered body composition. These factors contribute to the health burden of TD1, which warrants further attention for advancing bone health in women with T1D.

Läs artikeln i sin helhet pdf utan lösenord pdf

From the article


An increase in the incidence of type 1 diabetes (T1D) has been observed worldwide in recent decades1. It is understood that T1D is a disease that involves environmental triggers acting with genetic susceptibility to initiate a destruction of pancreatic β-cells, in which an increasing role for environmental factors has been suggested for the disease aetiology2. Approximately 10% of all T1D cases are due to familial aggregation, but more than 20% when accounting for the extended family history3.

T1D is the second most common chronic disease amongst children and is associated with multiple secondary complications4. The risk of cardiovascular complications accelerates during adolescence, especially amongst women5. Female adolescents have a higher risk for diabetes ketoacidosis, dyslipidaemia and weight problems, and as a group, women smoke more than men6,7. Body composition of women with T1D compared with that of controls has shown a higher fat mass in the former8.

The treatment of children with T1D in Sweden has been regulated since 1982 by national guidelines, which thereafter have been regularly updated. The last version, from 2018, coincides with the guidelines of the International Society of Paediatric and Adolescent Diabetes ( The original Swedish guidelines stated that the treatment should aim for normoglycaemia, with the ultimate goal of preserving a high quality of life and preventing acute and future vascular health complications. Diabetes is still, however, a complex disease. Besides social and psychological consequences, metabolic disturbances can affect multiple organs and body tissues9.

The importance of bone health is increasingly recognised; however, numerous aspects have to be considered, as bone mass is influenced by genetics, mechanical loading, physical activity, growth, vitamin D and hormones, as well as nutritional factors. Peak bone mass is achieved during early adulthood and serves as the “bone bank” for the remainder of life. The interpretation of bone mineral measurements is more complex in children than in adults since children are growing individuals10. Many diseases and treatments during childhood and adolescence affect the acquisition of bone mineral that contributes to optimal peak bone mass. The risk of fractures is increased in adult patients with both T1D and type 2 diabetes (T2D)11,12. The underlying pathophysiological mechanisms are complex13. Reports, comprising measurements by dual-energy X-ray absorptiometry (DXA) and peripheral quantitative computed tomography (pQCT), have demonstrated reduced bone mass in children with diabetes14,15,16. Disease duration is an important factor that contributes to negative effects on bone mass13,17. However, information is limited, especially regarding young women, on background factors influencing the elevated risk for reduced bone mass and the elevated risk for fractures.

Female patients with T1D have a higher risk for poor metabolic control and early micro- and macrovascular complications in comparison with age-matched men. Moreover, long disease duration contributes to the risk of low bone mass. We hypothesised that young women with long diabetes duration, but still without a history of fractures, would have lower bone mass and changed bone parameters compared with healthy controls.

The aim of the present study was to investigate bone health and body composition in young Swedish women patients with long-duration T1D of more than 10 years, with an age-, gender- and geography-matched healthy control group.


In general, young women with T1D have a less favourable metabolic control than men and have, as a group, a higher risk of future complications5,23. This study was designed to investigate bone health and body composition in young women with long-duration T1D (on average 19 years) in relation to well-matched controls. The main findings were that T1D individuals had changed body composition with reduced lean mass although comparable BMI. This study also demonstrated decreased bone strength index of cortical bone, i.e., SSI, and decreased periosteal circumference by pQCT. These findings contribute to the future health burden and warrant further attention in terms of efforts to improve bone health.

The recommended HbA1c value in Sweden for young individuals is currently ≤≤48 mmol/mol (6.5%). The metabolic control was highly suboptimal in our study group, with a final average HbA1c value of 68 mmol/mol (8.4%). This could be compared with the current average HbA1c of 60 mmol/mol (7.6%) for all women in that age group with T1D in Sweden in 2018 ( Their suboptimal or poor metabolic control was reflected in the prevalence of vascular complications. Smoking is associated with lower BMD24 and is also associated with increased cardiovascular risk, which is why it is important to discuss smoking habits amongst young individuals with T1D.

In our study, women with T1D were shorter than controls, which could be explained either by suboptimal glycaemic control during the vulnerable adolescent growth spurt period or by chance because of unmatched participants regarding height in this relatively small study. Growth retardation is a well-known complication in patients with T1D25. Insulin deficiency may affect the quality of bone, especially if there is a lack of insulin during the intensive stages of bone growth. Unfortunately, the women with T1D in our study had suboptimal metabolic control during their intensive growth period (Table 3), which may have affected their final height. The HbA1c increased steadily with increasing age, which is a pattern still seen today (SWEDIABKIDS) and which portrays the difficulties in maintaining good metabolic control during a period with increased hormonal production and increased growth velocity. This is also a sensitive period during which the gain in bone mass is most rapid26.

In the study of a population-based cohort with data from 1994 to 2012, in which 30,394 individuals (aged 0–89 years) with T1D participated and were compared with a matched control group, it was concluded that T1D was associated with increased risk of fracture incidence and that individuals with T1D sustained a greater number of lower-extremity fractures27. Less is known about whether the degree of glycaemic control specifically plays a role in the fracture risk associated with diabetes. Blood glucose control appears to play a role in the known risk of fracture in diabetes; however, the effect is significant only in T1D cases, suggesting insulin deficiency and poor glycaemic control earlier in life could take a toll on bone mass amongst these patients28. A large, recent meta-analysis of genome-wide association studies found that decreased BMD has a profound effect on future fracture risk, which implies that investigating BMD and bone health in patients with diabetes would be of great interest29.

A paper by Liu et al.30 indicated that women with T1D show differences in BMD early in life, with significant differences present in females 20–37 years of age. It has recently been demonstrated that elevated blood glucose has a negative effect on bone before adulthood in patients with T1D, although no signs of osteoporosis were identified by DXA31. Lower hip BMD in these young women may explain, in part, the higher incidence of hip fracture experienced in postmenopausal women with T1D. In contrast, results from the current study do not indicate a difference in total aBMD, femur or lumbar spine aBMD for patients with T1D. Total BMC was reduced in T1D but not significantly different when adjusted for BMI, physical activity and height, which reflects the height difference in this group.

Reduced periosteal and endosteal circumferences were observed for the T1D group; and the smaller periosteal circumference was also evident after adjustments for physical activity, BMI, height, and Hochberg multiple comparison. In theory, smaller bone circumferences, despite similar cortical thickness and cortical area after adjustments for height, might in part explain why individuals with diabetes have an elevated risk for fractures. Bone mass that is further displaced from the central bending axis (i.e., wider bone width), results in an increased cross-sectional moment of inertia, which confers greater resistance to bending32. The presented results are in line with Saha et al.33 who recently showed cortical bone size deficits (assessed by pQCT) in postmenopausal women with T1D onset before the age of 20 years. Weber et al.34 investigated skeletal outcomes within the first year of T1D diagnosis in children (7 to 17 years) and found decreased tibia cortical volumetric BMD in addition to lower rates of bone accrual, which was associated with poor glycemic control.

A study by Saha et al.35 found that T1D is associated with decreased BMC which could affect cross-sectional size and cortical rigidity; the study also found that male adolescents were more prone to these changes. The estimate of bone strength index of cortical bone, SSI, was reduced in patients with T1D, which suggests that the skeletal microstructure is altered in an unfavourable way that could contribute to the increased fracture risk at a greater age. In a study by Nilsson et al.36 of elderly women with T2D, a more favourable bone microarchitecture was observed. No difference in adjusted aBMD was observed in the population compared with healthy controls. As in our study, Nilsson et al.36 found a reduced bone material strength index; they hypothesized that this, together with impaired physical function, may explain the increased fracture risk in women with diabetes. The pathophysiological mechanisms of diabetes-induced bone fragility are complex. On the molecular level, it has been demonstrated that accumulation of advanced glycation end products compromise collagen properties, which leads to abnormal biomechanical properties associated with reduced bone strength13.

Our study confirmed earlier findings with a changed body composition in individuals with T1D regarding lower lean mass. A Swedish study evaluating body composition by DXA of adolescent women with T1D compared with controls showed higher total fat mass amongst women with T1D than amongst the controls8. The increased fat mass was associated with a higher insulin requirement, poor metabolic control and increased blood lipids. In the present study, there was no difference in physical exercise or sedentary hours that could explain the difference in lean mass. In our study, however, the women with T1D had poor metabolic control, which is associated with higher insulin resistance and a higher insulin requirement; this may have affected their body composition.

The current study has a number of strengths, such as a well-matched control group, validated data from national paediatric and adult diabetes registers, and the long follow-up period, on average 19 years, from diabetes diagnosis. Furthermore, two- and three-dimensional bone measurement techniques by DXA and pQCT were used. However, the study also possesses some limitations that warrant consideration when interpreting the data, such as the relatively low number of patients. The use of accelerometers to monitor physical activity would have been preferable to the IPAQ questionnaire that was used, even though the IPAQ is a validated self-assessment form used frequently to assess the amount of physical activity.

In conclusion, young women with long-term T1D duration showed altered body composition and decreased cortical bone strength in comparison with controls. A changed body composition and affected bone parameters at this young age increase the health burden in T1D. Preventive interventions to improve metabolic control and bone health are of great importance for future healthcare, aiming at optimal health-related quality of life.

Published Dec 20 2020


www red DiabetologNytt