Over 10% of hospital inpatients receive treatment with glucocorticoids1 and yet the incidence of steroid diabetes (SD), and therefore monitoring requirements, in these patients is unknown. Acute illness can also induce hyperglycaemia through neuroendocrine and inflammatory responses,2 which may be exacerbated by use of glucocorticoids.
We determined the incidence of SD in acute medical admissions and whether SD corresponds to illness severity rather than to the dose of glucocorticoid administered.
This was a retrospective case-note review of inpatients who received treatment with glucocorticoids (dose ≥10 mg prednisolone once daily, or equivalent)3 for ≥24 hours in an acute medical admissions unit. We excluded patients using glucocorticoid therapy immediately prior to hospitalisation (other than inhaler or topical therapy); those with accident & emergency (A&E) department triage glucose ≥11.1 mmol/L, length of stay <24 hours, and diabetes mellitus (any of medical history, glycated haemoglobin ≥6.5% before or up to 3 months after admission, any glucose lowering therapy). A consecutive series was sought until n=100 in the study cohort.
New hyperglycaemia (capillary glucose ≥11 mmol/L after initiation of glucocorticoid therapy) was considered SD.2,4 Glucose values were evaluated for up to 7 days after admission. The National Early Warning Score (NEWS) was used to determine illness severity. Glucocorticoids were converted to equivalent potencies to allow dose comparison; 1 steroid unit ≡ 5 mg prednisolone.3 Data are mean (standard deviation) unless described. Local Research & Innovation approval was sought and we were advised that ethical approval was not required.
Between July 2015 and April 2017, 498 patients with an acute medical admission received glucocorticoids of ≥10 mg prednisolone per day (or equivalent) for at least 24 hours. We excluded n=268 without glucose monitoring; a further n=130 with pre-existing diabetes were also excluded, leaving 100 patients for analysis. Exacerbation of pulmonary disease constituted the reason for admission for over half the cohort. The incidence of SD was 14%. Median duration of admission before SD was identified was 2.5 days (interquartile range [IQR] 1–4). Patients with SD had higher NEWS at admission and higher peak NEWS during their hospitalisation (Table 1) but no difference in type of steroid, nor mean steroid dose. Glucose concentration from the A&E department did not predict likelihood of SD (Table 1).
Table 1.
Characteristics of steroid diabetes and normal glucose group
| Parameter | Steroid diabetes n=14 | Normal glucose n=86 | p value | |
|---|---|---|---|---|
| Age, years# | 64.9 (16.0) | 61.3 (18.0) | 0.498 | |
| Gender | Male | 6 (43%) | 40 (47%) | |
| Female | 8 (57%) | 46 (53%) | 0.799 | |
| Type of steroid | Prednisolone | 11 (79%) | 51 (59%) | |
| Hydrocortisone | 1 (7%) | 26 (30%) | 0.257 | |
| Dexamethasone | 1 (7%) | 7 (8%) | ||
| Methylprednisolone | 1 (7%) | 2 (2%) | ||
| Relative daily steroid dose, steroid unit* | 6.0 (6.0–10.9) | 7.3 (6.0–8.7) | 0.303 | |
| Admission NEWS* | 6 (3–9) | 4 (2–6) | 0.041 | |
| Maximum NEWS* | 8 (5–9) | 6 (3–8) | 0.048 | |
| A&E capillary glucose, mmol/L* | 6.1 (5.3–10.0) | 6.0 (5.3–7.0) | 0.195 | |
| Length of stay, days* | 7.0 (4.8–16.5) | 5.0 (3.0–8.0) | 0.113 | |
| Inpatient mortality | 2 (14%) | 5 (6%) | 0.249 |
# = mean (standard deviation); * = median (interquartile range); NEWS = National Early Warning Score
Patients developing SD following acute hospitalisation had greater illness severity than those maintaining normoglycaemia, whereas the steroid dose (equivalent to 35 mg prednisolone daily) was no different. The incidence of SD that we found (14%) is similar to Umpierrez et al5 who reported hyperglycaemia in 12% of a general medical inpatient population (with no history of diabetes), although in that series the use of glucocorticoids was not reported. We hypothesise that exogenous glucocorticoid therapy contributes little to the pathogenesis of hyperglycaemia with acute hospitalisation. Use of resource intensive glucose monitoring may be better directed to those of greater illness severity than to all hospitalised patients treated with glucocorticoids. This requires evaluation.
References
- 1.Narwani V, Swafe L, Stavraka C, et al. How frequently are bedside glucose levels measured in hospital inpatients on glucocorticoid treatment? Clin Med 2014;14:327–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet 2009;373:1798–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Roberts A, James J, Dhatariya K. Management of hyperglycaemia and steroid (glucocorticoid) therapy: a guideline from the Joint British Diabetes Societies (JBDS) for Inpatient Care group. Diabet Med 2014;35:1011–7. [DOI] [PubMed] [Google Scholar]
- 4.Plummer MP, Bellomo R, Cousins CE, et al. Dysglycaemia in the critically ill and the interaction of chronic and acute glycaemia with mortality. Intensive Care Med 2014;40:973–80. [DOI] [PubMed] [Google Scholar]
- 5.Umpierrez GE, Isaacs SD, Bazargan N, et al. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87:978–82. [DOI] [PubMed] [Google Scholar]