The doubled burden of diabetic bone disease: hip fracture and post-hip fracture mortality

Abstract

Objective: Patients with diabetes have an increased risk of osteoporosis and shorter life expectancy. Hip fracture (HF) is the most serious consequence of osteoporosis and is associated with increased mortality risk. We aimed to assess the association of antidiabetic medications with HF and the post-hip fracture mortality risk among diabetic patients ≥50 years. Design: In this nationwide case-control study 53,992 HF cases and 112,144 age-, sex- and region-matched non-hip fracture controls were analyzed. A cohort of hip-fractured diabetic patients were followed-up for an all-cause mortality. Methods: We defined three groups of diabetic patients based on a prescription of antidiabetic medications: group 1 treated with insulin monotherapy (G1DM), group 2 (G2DM) treated with blood glucose lowering drugs (BGLD) only, group 3 on a combined BGLD and insulin therapy (G3DM). We applied logistic regression and Cox regression. Results: We identified 2757 G1DM patients, 15,310 G2DM patients, 3775 G3DM patients and 144,294 patients without any antidiabetic treatment. All three groups of diabetic patients had increased odds of HF compared to controls. G1DM patients aged 50-64 years (aOR 4.80, 95%CI 3.22-7.17) and G3DM patients (aOR 1.39, 95%CI 1.02-1.88) showed the highest HF odds, whereas G2DM patients had 18% decrease in HF odds than their non-diabetic controls (aOR 0.82, 95%CI 0.69-0.99). All diabetic patients had increased post-hip fracture mortality risk compared to non-diabetic controls. The highest mortality hazard was observed in G1DM patients, being greater for women than men (HR 1.71, 95%CI 1.55-1.89 and HR 1.44, 95%CI 1.27-1.64, respectively). Conclusions: Antidiabetic medications increase probability of HF. Diabetic patients, who sustained HF have higher mortality risk than non-diabetic patients.

Impact of diabetes mellitus on risk of major complications after hip fracture: a systematic review and meta-analysis

“…Thirty-three studies were selected for full-text analysis. Seventeen studies were excluded with reasons and a total of 16 studies were included in this review [ 12 – 15 , 18 – 29 ]. Details of included studies are presented in Table 1 .…”

“…The number of patients in the DM group ranged from 61 to 92,181 while that of the control group ranged from 161 to 141,889. In five studies, the number of patients undergoing surgical intervention was not clear [ 12 , 18 , 21 , 24 , 25 ]. All patients were treated surgically in eight studies [ 13 – 15 , 19 , 20 , 22 , 26 , 27 ] while in the remaining studies the number of patients treated surgically was > 78%.…”

“…Risk of bias analysis is presented in Table 3 . Six studies [ 12 , 15 , 23 , 26 – 28 ] recorded a score of 5–6 on the NOS scale are were deemed to have a high risk of bias. All remaining studies were of moderate risk of bias.…”

“…These results concur with the generalized higher risk of mortality noted in men after hip fracture [ 33 , 34 ]. Considering the type of drug used for DM control, Behanova et al [ 12 ] and Lee et al [ 22 ] have noted significantly higher mortality in hip fracture patients on insulin as compared to those on oral hypoglycemic drugs. Contrastingly, Golinvaux et al [ 14 ] have noted no such difference in outcomes.…”

“…However, there have been controversial evidence on the risk of mortality and other major systemic complications between DM and non-DM patients with hip fracture. Some studies have reported increased risk of mortality and systemic complications amongst diabetics [ 12 , 13 ] while others have reported no such difference [ 14 , 15 ]. Such variability in results could be due to several reasons like underpowered study, bias in reporting, etc.…”