Home / Resources / Articles / Adverse Effects of Short-Term Corticosteroid Use in Diabetes Mellitus

Adverse Effects of Short-Term Corticosteroid Use in Diabetes Mellitus

Jan 27, 2018

Is preventive medication the key to reducing risk of fractures, VTEs, and sepsis?

It is reported that approximately one-fifth of the U.S. working population has been placed on some form of short-term corticosteroid regimen. When given to diabetes patients, it is important to educate and understand that this population requires additional monitoring. When oral corticosteroids are given, insulin secretion and sensitivity are depressed, putting this group of patients at further risk for hyperglycemia. Generally, short-term use has been considered safe but there are some side effects other than hyperglycemia to monitor for. Previous trials tend to have low statistical power, disparity in patient safety reporting, and under reporting risk assessments. It has been shown that 89% of randomized controlled trials do not report venous thromboembolisms (VTE).

This longitudinal study looked at the frequency of short-term oral corticosteroid use in working adults with diabetes to assess the occurrence of fractures, VTE, and hospitalizations for corticosteroid-induced sepsis. Data was collected from January 1st 2012- December 31st 2014.  Adults ages 18-64 years of age, who held 3 years of continuous private healthcare coverage, were analyzed for adverse effects (AE) associated with short-term oral corticosteroid use. Short-term use was considered less than 30 days. The drugs that were up for review included betamethasone, dexamethasone, methylprednisolone, triamcinolone, prednisone, prednisolone, hydrocortisone, and cortisone (tablets, capsules, and elixirs only).  People with type 1 and type 2 diabetes were divided into different group based on a minimum of two processed ICD-9-CM codes on different dates. Adverse effect outcomes were also assessed using ICD-9-CM code processing.

Using a tri-phase analysis: Phase one reviewed the frequency of corticosteroid use and divided results by type 1 diabetes (T1DM) vs. type 2 diabetes (T2DM), gender, and age categories. Phase one was measured by the Pearson’s Chi-squared test. Phase two assessed the incidence of adverse effects: fractures, VTEs, and hospitalizations for sepsis. This phase was also separated by T1DM, T2DM, and corticosteroid users vs. non-corticosteroid users. Results were calculated using 95% CIs and incidence rate ratios. The synergy index was also used to show how AE may be caused by interactions between steroid use and diabetes. Heterogeneity for AEs and the type of antidiabetic medication was also analyzed using the Cochran’s Q test and the I2 statistic test. In phase three, conditional Poisson regression model was used to evaluate the association between corticosteroid short-term use and AEs. A final fixed Poisson regression analysis was conducted to observe if the use of preventive medication, prior to corticosteroid use, made a difference. Vitamin D was added to corticosteroid therapy to prevent fractures, while a statin was added to corticosteroid therapy in hopes of preventing VTEs and sepsis.

Results showed that the patient profile of this study was 8.1% of adults had diabetes.  87.4% of the 8.1% of adults had T2DM, and 12.6% had T1DM. More men had diabetes in comparison to women. Those with T2DM were more likely to use short-term corticosteroids (~6day duration) than individuals with T1DM or without diabetes (P<0.001). Participants with T1DM were more likely to have the listed AE while using corticosteroids. 12.5% endured fractures, 4.6% VTEs and 4.4% were hospitalized for corticosteroid induced sepsis. The synergy index showed a positive interaction with all outcomes. Differentiating antidiabetic medications and their relationship to AE showed no significance. Adding 50,000 IU of ergocalciferol weekly for individuals taking oral corticosteroids showed an incidence rate ratio (IRR) of 1.13 in comparison to those on corticosteroids without Vit D, IRR=2.09. Adding statin therapy had two different effects: the risk of VTE remained elevated regardless, but the IRR of not including statin therapy was significantly higher at 4.91 when assessing sepsis (P=0.013).

In conclusion, patients with diabetes were at a higher risk of having fractures, VTEs, and hospitalizations for sepsis with using short-term oral corticosteroids. Using preventable measures have proven to provide better outcomes when used concomitantly with oral corticosteroids. A limitation of this study is the sensitivity of ICD-9-CM codes when comparing T1DM & T2DM. Another limitation is the inability to distinguish other illnesses at doctor visits. A strength of this study is the use of self-controlled case series, which allowed you to follow each individual. More trials should be conducted on preventative measures with corticosteroid use.

Practice Pearls:

  • VTE, fractures, and sepsis aren’t common complications of oral corticosteroid use, but the risk is there.
  • Monitoring diabetes patients while of steroid therapy is key, especially those with T1DM.
  • Preventative therapy has shown some benefit at reducing the risk of AEs associated with oral corticosteroid use.


Rogers, Mary A.m., et al. “Longitudinal study of short-Term corticosteroid use by working-Age adults with diabetes mellitus: Risks and mitigating factors.” Journal of Diabetes, 5 Jan. 2018, doi:10.1111/1753-0407.12631.

Adrianna Jackson, Doctor of Pharmacy Candidate: Class of 2018; LECOM College of Pharmacy