A literature search for this narrative review was conducted using PubMed to identify relevant studies published in English from inception up to June 2025. The following keywords/terms were used: “steroid-induced diabetes mellitus” OR “steroid-induced hyperglycemia” OR “glucocorticoid-induced diabetes mellitus” OR “glucocorticoid-induced hyperglycaemia” OR “corticosteroid therapy” OR “glucocorticoid therapy” OR “SIDM” OR “GIDM” AND “autoimmune rheumatic diseases” OR “ARDs” OR “immune-mediated inflammatory diseases” OR “rheumatic diseases” OR “inflammatory rheumatic diseases” OR “rheumatoid arthritis” OR “RA” OR “systemic lupus erythematosus” OR “SLE” OR “antiphospholipid syndrome” OR “APS” OR “vasculitis” OR “large-vessel vasculitis” OR “giant-cell arteritis” OR “GCA” OR “ANCA-associated vasculitis” OR “spondyloarthritis” OR “SpA” OR “psoriatic arthritis” OR “PsA” OR “axial spondyloarthritis” OR “AxSpA” OR “polymyalgia rheumatica” OR “PMR” OR “connective tissue diseases” OR “systemic sclerosis” OR “SSc” OR “Sjögren’s syndrome” OR “pSS” OR “idiopathic inflammatory myopathies” OR “myositis” OR “dermatomyositis”.

Studies were initially screened based on their titles and abstracts to determine relevance, followed by full-text review. Additionally, the reference lists of the selected articles were examined to identify further studies that may have been missed in the primary search.

To be eligible for this review, studies had to investigate GCs use in relation to glucose metabolism or DM presentation, specifically within individuals having ARDs. Original research articles, randomized controlled trials, or meta-analyses published up to June 2025 were included. Studies were excluded if the full text was not accessible or focused on populations having diseases other than ARDs, or on non-steroid-related causes of DM.

Glucocorticoid-induced diabetes mellitus

GC use is associated with significant metabolic complications, including impaired glucose tolerance and decreased insulin sensitivity often leading to glucocorticoid-induced hyperglycemia (GIH) and GIDM [4]. GIH refers to worsening glycemic control in the context of pre-existing DM associated with exogenous GC therapy, whereas GIDM describes new-onset DM as a direct consequence of GC exposure in individuals without a previous diagnosis of DM [5].

GIDM is diagnosed using standard DM criteria, including fasting plasma glucose ≥ 126 mg/dL (≥ 7 mmol/L), random plasma glucose ≥ 200 mg/dL (≥ 11.1 mmol/L), or a 2-h plasma glucose ≥ 200 mg/dL following an oral glucose tolerance test (OGTT), provided these irregularities occur after the initiation of GC therapy [6]. However, testing should account for the diurnal glycemic pattern induced by GC use (especially intermediate-acting GC), which disproportionately increases postprandial and evening glucose [7]. As a result, relying completely on morning fasting glucose measurement as a diagnostic criterion may underestimate the true incidence of GIH or GIDM [7]. Similarly, OGTT may not be optimal for GIDM diagnosis due to its fasting requirement and potential underestimation of glucose elevation occurring primarily in the evening. Yet, in high-risk, non-diabetic individuals, blood glucose measurement and OGTT should be performed early on during the monitoring of the patients to help detect underlying DM susceptibility [8].

GIDM most commonly emerges between the 2nd and 4th week of GC therapy, supporting continuous monitoring, especially for individuals experiencing hyperglycemia or requiring antidiabetic medications [9]. Individuals with pre-existing DM or risk factors for GIDM should be screened even when low GC doses are prescribed [10]. Typically, GIDM improves following dose tapering or GC discontinuation, but in some cases, it may persist [6].

Pathophysiology

Multiple tissue-specific effects of GCs have been described contributing to the development of GIDM, with most of these pathways mediated by the GC-receptor, binding to specific target genes and altering their expression [7]. Thus, changes in liver, skeletal muscle and adipose tissue primarily mediate GIDM and GIH pathophysiology, whereas other organs such as the pancreas, the hypothalamus and the bones further disrupt the glucose regulation [7].

Liver

Following GC therapy, elevated hepatic glucose production, insulin resistance and hepatic steatosis have been described [7]. GCs induce hepatic gluconeogenesis by upregulating enzymes of de novo glucose synthesis, which is further amplified by the uptake of gluconeogenic precursors from the GC-mediated protein catabolism in skeletal muscles and lipolysis in adipose tissue [7, 11, 12]. Hepatic insulin resistance is also developed, as GCs impair the phosphorylation of downstream second messengers of the insulin cascade, leading to unopposed gluconeogenesis [13,14,15]. Despite that, the pro-lipogenic effect of insulin is preserved [16], which in the setting of compensatory hyperinsulinemia (due to GIH), increases de novo hepatic lipogenesis leading to steatosis and exacerbation of hepatic insulin resistance [17, 18]. Other factors, such as the elevated liver uptake of circulating non-esterified fatty acids (NEFAs) and the increased ceramide synthesis, are also associated with hepatic dysregulation leading to GIDM [17,18,19].

Skeletal muscle

GCs decrease skeletal muscle insulin sensitivity and glucose uptake, by inhibiting the insulin-dependent upregulation of glucose transporter 4 (GLUT4) to the cell surface of skeletal muscles [20]. Moreover, long-term GC therapy leads to myopathy [21], by promoting proteolysis and diminishing protein synthesis [11]. The elevated amino acids indirectly impede insulin-induced glucose transport and glycogen synthesis in muscle tissue [22], whereas the also act as substrate for hepatic gluconeogenesis [7].

Adipose tissue

Adipose tissue actively participates in GIDM pathogenesis, via decreased glucose uptake, increased triglyceride uptake and synthesis, and increased lipolysis providing substrates for hepatic gluconeogenesis [7]. Similarly to endogenous hypercortisolemia [23], exogenous GCs lead to lipid redistribution from peripheral to abdominal adipose depots [24]. This depot-dependent adipose tissue breakdown and expansion results in central obesity, which is linked to insulin resistance and T2DM, due to perihepatic white adipose tissue (WAT) hypertrophy and direct drainage of NEFAs and pro-inflammatory factors into the portal circulation [7, 9, 25,26,27]. Moreover, despite promoting adipose expansion (via insulin dependent lipogenesis, triglyceride synthesis and WAT hypertrophy [28, 29]), GCs concurrently increase lipolysis by upregulating the hormone-sensitive lipase and monoacylglycerol lipase [29, 30]. This increases the serum levels of NEFAs and glycerol [31, 32], leading to hyperglycemia due to hepatic gluconeogenesis and impairment of liver and muscle insulin sensitivity from ectopic lipid accumulation [7]. Furthermore, GCs indirectly affect both adipose tissue and whole-body insulin sensitivity through the regulation of adipokine levels secreted by WAT [4, 7, 28].

Pancreas, Bone tissue and central nervous system

Pancreatic β-cell dysfunction has been previously linked to GIDM [7]. In susceptible individuals, short-term GC administration is associated with decreased β-cell insulin secretion and whole-body glucose disposal [33,34,35]. Normally, glucose uptake by β-cells causes an increase in intracellular ATP levels, leading to closure of the ATP-sensitive potassium channels and eventually depolarization of the cell membrane. Thus, voltage-gated channels are activated resulting in rapid calcium influx and eventually insulin release from the β-cells [7, 36]. GCs reduce insulin secretion by disrupting this process in multiple stages, such as glucose uptake, membrane de- and re-polarization and α2-adrenergic signaling [37,38,39]. Additionally, other mechanisms of pancreatic β-cell dysfunction have been described, such as direct inhibition of insulin synthesis and GC-induced cytotoxicity via reactive oxygen species leading to β-cell apoptosis [7, 40, 41].

When it comes to bone tissue, GC-osteoporosis is associated indirectly with GIDM development [7]. Osteoporosis following GC therapy is caused by suppression of the bone forming osteoblasts and activation of the bone-resorbing osteoclasts [42]. In addition, GCs also contribute to GIDM by increasing leptin secretion from adipose tissue, which in turn suppresses the expression of osteocalcin, thereby indirectly inhibiting insulin secretion [7, 43].

Finally, the effects of GCs on the central nervous system (CNS) should be also considered in GIDM pathogenesis. Under physiologic conditions, appetite is mediated by orexigenic peptides, such as neuropeptide Y (NPY) and agouti-related peptide (AgRP), which are released by neurons in the arcuate nucleus of the hypothalamus [7]. Leptin is one of the hormones responsible for the appetite regulation by suppressing NPY–AgRP neurons and thus, the release of orexigenic peptides [44]. GCs upregulate the expression and release of these peptides, whereas they also synergistically antagonize the action of leptin in the hypothalamus by directly reducing leptin-dependent JAK–STAT signaling [7, 45,46,47]. Despite that, GCs shift nutritional preference towards high-fat and/or high-sugar foods, known as “comfort food”. Therefore, GCs promote a hypercaloric diet, which indirectly results in obesity and DM [7]. The pathophysiological aspects of GIDM are further depicted in Fig. 1.

Fig. 1

Multi-organ pathophysiological aspects of GIDM. Schematic overview of GC effects across liver, skeletal muscle, adipose tissue, pancreatic β-cells, bone, and hypothalamus. The GC-receptor complex mediates transcriptional messages that: (i) increase hepatic gluconeogenesis and de novo lipogenesis, leading to hepatic insulin resistance and steatosis; (ii) reduce GLUT4-dependent glucose uptake and enhance proteolysis, while decreasing protein synthesis in skeletal muscle; (iii) promote visceral adiposity with concurrent lipogenesis and lipolysis in adipose tissue that aggravate systemic insulin resistance; (iv) decrease β-cell insulin synthesis and secretion, and promote β-cell apoptosis; (v) contribute to osteoporosis and reduced osteocalcin signaling; and (vi) increase leptin resistance and appetite. Collectively, these pathways shift glucose homeostasis toward hyperglycemia and GIDM. GC Glucocorticoids, GIDM Glucocorticoid-induced diabetes mellitus, GLUT4 Glucose transporter type 4

GCs nomenclature and route of administration

Accurately estimating the risk of GIH or GIDM when administering GCs remains challenging due to the heterogeneity of dosing regimens, pharmacologic potency, routes of administration and treatment duration used across different clinical conditions [7].

GCs are commonly classified according to their duration of action and relative potency [48]. Short-acting agents, like hydrocortisone and cortisone, have a half-life of less than 12 h [7, 10, 48, 49]. Intermediate-acting GCs, such as prednisone, prednisolone and methylprednisolone demonstrate half-lives of 16–36 h, whereas long-acting agents, including betamethasone and dexamethasone, have half-lives ranging from 36 to 54 h [7, 10, 48, 49]. Among these, the intermediate-acting GCs are the most frequently used, being 4–5 times more efficacious than the short-acting, while long-acting agents may reach up to 25-fold greater potency [7, 10, 48]. In terms of dosage, low-dose GCs are defined as prednisolone equivalent dose (PED) less than 7.5 mg/day, medium-dose regimens range from 7.5 mg to 30 mg/day, and high-dose from 30 to 100 mg PED/day [50, 51]. When the GCs exceed 100 mg PED/day they are considered very high dose [51]. For pulse therapy, the therapeutic regimen consists of at least 250 mg PED/day for one or a few days [51].

GCs can be administered via multiple routes. Although all common routes of GC administration, have been associated with a higher GIH risk, when given in high doses [7, 52,53,54,55], oral GCs pose the greatest risk, regardless of the dosing regimen [7, 56]. Intra-articular (IA) and intramuscular (IM) injections, which are commonly used in ARDs, have also been correlated with GIH development [57, 58], but factors such as differences in GC injection release profile and previously established good glycemic control appear to influence glucose homeostasis following these regimens [58,59,60]. In a post-hoc analysis of a phase 2 study enrolling 33 patients with knee osteoarthritis and T2DM, individuals receiving extended-release triamcinolone acetonide IA injections (n = 18) versus immediate-release triamcinolone acetonide (n = 15) were compared [61]. The former had lower median change from baseline in maximum glucose level, less median time with a glucose level > 250 mg/dL, a smaller proportion of patients with a maximum glucose level of > 250 mg/dL, and a greater percentage of time in the target glucose range [61]. In another study, evaluating the effects of IA injection of betamethasone acetate/betamethasone sodium phosphate on blood glucose levels in controlled DM individuals (glycated hemoglobin HbA1c < 7) with symptomatic knee osteoarthritis, no significant effect on serum fructosamine levels were observed (measured just prior and 2 weeks following the injection), whereas only transient increases in blood glucose levels were reported [58].

Epidemiology of GIDM in non-ARDs

GIDM is a clinically important complication of GC therapy affecting approximately 15–52% of individuals treated with GCs [9, 26, 27, 62,63,64,65,66,67]. The occurrence of GIDM depends on various factors, such as advanced age, higher BMI, abdominal obesity, hypertriglyceridemia and family history [7, 67], as well as the disease itself. In a retrospective study of individuals with respiratory diseases treated with prednisolone (GC equivalent > 20 mg/day), 15% developed GIDM [64]. Likewise, a study, enrolling 90 patients on GCs as a part of their cancer treatment, reported GIDM in 19% [