Major side effects of systemic glucocorticoids

INTRODUCTION — Glucocorticoids are important in the treatment of many inflammatory, allergic, immunologic, and malignant disorders, and the toxicity of glucocorticoids is one of the commonest causes of iatrogenic illness associated with chronic inflammatory disease. Recognition of these toxicities, many of which are similar to the findings in spontaneous (endogenous) Cushing's syndrome, is of value in their prevention and management. (See "Epidemiology and clinical manifestations of Cushing's syndrome".)

The major adverse effects seen with systemic (oral and parenteral) glucocorticoid therapy will be reviewed here. This topic will focus particularly on the side effects of systemic prednisone, prednisolone, and methylprednisolone, as there are more data for these compounds. The adverse effects resulting from the use of inhaled, topical, and intraarticular glucocorticoids; glucocorticoid withdrawal; use during pregnancy; and the clinical manifestations of endogenous glucocorticoid excess are discussed in detail separately. (See "Major side effects of inhaled glucocorticoids" and "Topical corticosteroids: Use and adverse effects", section on 'Adverse effects' and "Joint aspiration or injection in adults: Complications" and "Glucocorticoid withdrawal" and "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Glucocorticoids' and "Epidemiology and clinical manifestations of Cushing's syndrome".)

FACTORS RELATED TO GLUCOCORTICOID TOXICITY — Numerous toxicities, or adverse effects, have been attributed to glucocorticoids (table 1). However, the attribution of causality to the glucocorticoids alone cannot always be clearly established. Other factors that may contribute to such adverse effects include the nature and severity of the underlying disease being treated and the other medications being concurrently administered. Estimates of the frequency and severity of adverse effects, as well as the respective dose and duration of therapy that may result in such adverse effects, are also limited by the modest number of prospective trials that address this question.

Mechanisms of adverse effects — Glucocorticoids used in chronic disease (eg, prednisone or prednisolone) do not have significant mineralocorticoid, androgenic, or estrogenic activity; thus, their major adverse effects result from inhibition of hypothalamic-pituitary-adrenal function and the development of iatrogenic Cushing's syndrome. (See "Pharmacologic use of glucocorticoids", section on 'Complications of chronic use' and "Epidemiology and clinical manifestations of Cushing's syndrome".)

The effects of glucocorticoids are mediated by differences in bioavailability, and mediated by effects upon receptor activation by phosphorylation, translocation, and repression/activation of gene expression. This leaves multiple sites where differences among glucocorticoids and/or differences among individuals can result in differences in efficacy and toxicity. It helps explain the differential resistance to therapy and toxicities among patients [1-4]. By contrast, many of the clinically desirable effects appear to result primarily from transrepression, which results in the decreased production of proinflammatory proteins. However, the mechanisms of glucocorticoid transactivation and transrepression are poorly understood. Nongenomic effects of glucocorticoids include rapid, nonspecific interactions of glucocorticoids with cellular membranes, nongenomic effects medicated by cytosolic glucocorticoid receptors, and specific interactions with membrane-bound glucocorticoid receptors [2].

Dose-related effects — Several retrospective reviews have shown that long-term glucocorticoid use, even in low doses, is a significant independent predictor of numerous adverse effects and that the risk is both dose- and duration-dependent [5-8]. As an example, a study of 112 patients with rheumatoid arthritis (RA) found that the average daily prednisone dose was the strongest predictor of an adverse effect potentially attributable to glucocorticoid therapy (odds ratio [OR] = 4.5 for 5 to 10 mg and 32.3 for 10 to 15 mg) (figure 1) [6]. The preselected adverse effects included in the analysis were fracture, serious infections, gastrointestinal bleed or ulcer, and cataracts. The glucocorticoid-adverse effect association persisted after statistical adjustment for significant disease severity factors, such as the presence of rheumatoid nodules and bony erosions. There are also some data that have suggested that very low doses of glucocorticoids (eg, prednisone <5 mg/day) are associated with fewer adverse effects [9,10]. Interestingly, a small phase 2 trial found that patients on continuous prednisone for inflammatory diseases who were randomized to receive metformin demonstrated improvements in measures of metabolic activity as well as other clinical markers of disease activity compared with those receiving placebo [11].

A linear and threshold pattern of dose-related adverse effects have been described for different adverse effects. A large retrospective analysis, which included data from 1066 patients with RA on glucocorticoids for more than six months, observed a "linear" pattern with increasing dose for Cushingoid phenotype, ecchymosis, leg edema, mycosis, parchment-like skin, shortness of breath, and sleep disturbance [5]. A "threshold" pattern, which described an elevated frequency of events beyond a certain threshold value, occurred at doses of prednisone greater than 7.5 mg/day in which there was an increased frequency of glaucoma, depression, and elevated blood pressure. Also, at doses of prednisone greater than 5 mg/day, there was an increased frequency of weight gain and epistaxis. Cataracts were an exception, as cataracts were more common among patients on glucocorticoids, even when the prednisone dose was less than 5 mg/day [5].

Shorter-term use may also be associated with serious adverse effects, particularly with higher doses. A retrospective cohort study and self-controlled case series that used a nationwide dataset of private insurance claims in the United States assessed the risk of three adverse effects (sepsis, venous thromboembolism [VTE], and fracture) in 327,452 adults younger than 65 years of age who received at least one short-term (<30 days) outpatient prescription over a 3-year period [12]. Within 30 days of drug initiation, there was an increase in the rates of sepsis (incidence rate ratio [IRR] 5.30, 95% CI 3.80-7.41), VTE (IRR 3.33, 95% CI 2.78-3.00), and fracture (IRR 1.887, 95% CI 1.69-2.07), which then decreased over the subsequent 31 to 90 days. Limitations to the study include an inability to control for other time-varying factors that could have been differentially distributed between the risk and baseline periods. Further, the absolute risks of each of the adverse effects was low. Of note, only 6.3 percent of the prescriptions were for a prednisone-equivalent dose of less than 17.5 mg a day, suggesting that most of the exposures were to relatively higher doses. Nearly half (47 percent) of glucocorticoid prescriptions were prescribed as prepackaged methylprednisolone "dosepaks," which are convenient to prescribe but are costly and deliver a high glucocorticoid dose over a short period of time (ie, 105 mg prednisone equivalent over six days). Another large case-controlled case series that also used a nationwide dataset of private insurance claims included 2,623,327 persons aged 20 to 64 years who had received a single prednisone burst for 14 days or fewer [13]. Within the 5 to 30 days after glucocorticoid initiation, rates of gastrointestinal bleeding (IRR 1.80, CI 1.75-1.84), sepsis (IRR 1.99, CI 1.70-2.32), and heart failure (IRR 2.37, CI 2.13-2.63) increased, but then the risk attenuated during the subsequent 31 to 90 days. Limitations to both of these studies include potential misclassification of diagnoses in large databases and lack of accounting for comorbidities and concomitant medications.

ORGAN-BASED TOXICITY OF SYSTEMIC GLUCOCORTICOIDS — Glucocorticoids have adverse effects on many organ systems (table 1). Adverse effects range from those that are not necessarily serious but are displeasing to patients (eg, Cushingoid appearance) to those that are life-threatening (eg, serious infections). Some adverse effects, such as accelerated reductions in bone mineral density or early cataracts, may be largely asymptomatic until later manifestations develop that require medical attention (eg, acute vertebral collapse, cataract requiring surgical extraction). Estimates of the frequency of particular adverse effects vary between studies and depend in part on the specific condition being treated and on the glucocorticoid treatment regimen being used [14]. With the exception of cataracts, a potential acceleration in atherosclerotic vascular disease, and bone effects (osteoporosis and osteonecrosis), all glucocorticoid toxicity is at least partially reversible over time with glucocorticoid discontinuation. (See 'General treatment considerations and monitoring' below.)

Dermatologic effects and appearance — Many clinically relevant adverse effects of glucocorticoids on the skin and appearance that have been observed even at lower doses include skin thinning and ecchymoses, Cushingoid appearance, acne, weight gain, mild hirsutism, facial erythema, and striae. We discuss some of the major dermatologic effects in more detail below.

●Skin thinning and ecchymoses – Among the most common toxicities attributable to glucocorticoids are skin thinning and ecchymoses, even at low doses [9]. A large observational study with patient-reported data from patients with rheumatoid arthritis (RA) on glucocorticoids for at least six months reported rates of parchment-like skin and ecchymoses in 10 and 17 percent, respectively [5]. The prevalence of these effects also increased with higher doses of prednisone. In a prospective study with 80 patients, skin changes were observed in 46 percent of patients treated for three months with prednisone at doses greater than 20 mg/day [15]. The ecchymoses or purpura associated with glucocorticoid use often affects the sun-exposed areas of the dorsum of the hand and forearm and is not accompanied by palpable swelling. (See "Photoaging", section on 'Photoaging in individuals with phototypes I to IV'.)

●Cushingoid features – The development of Cushingoid features (redistribution of body fat with truncal obesity, buffalo hump, and moon face) and weight gain are dose- and duration-dependent and can develop within the first two months of therapy. A Cushingoid appearance can be quite troubling to patients and can develop even with low-dose therapy; however, it is uncommon at doses below the physiologic glucocorticoid-replacement range. Factors that may also contribute to increased weight include an increased appetite, a common side effect of glucocorticoid therapy, and an increase in food intake for symptomatic relief in patients with gastropathy or peptic ulcer disease [16]. (See 'Gastrointestinal effects' below and "Epidemiology and clinical manifestations of Cushing's syndrome", section on 'Progressive obesity'.)

●Weight gain – Several observational studies have assessed the frequency of Cushingoid features and weight gain. In an observational study of 779 patients with RA, weight gain was more frequent in patients treated with at least 5 mg/day of prednisone or equivalent for at least six months, compared with those who had not received any for at least 12 months (22.4 versus 9.5 percent) [5]. However, there was a threshold effect, such that the rate of weight gain was not increased in those who took less than 5 mg/day (8.7 percent) or further increased at doses greater than 7.5 mg/day (21.3 percent) [5]. By contrast, Cushingoid features showed a linear increase in frequency with dose rather than a threshold effect. Among patients who had received <5, 5 to 7.5, and >7.5 mg/day of prednisone or equivalent, Cushingoid features were observed in 4.3, 15.8, and 24.6 percent of patients, respectively. In another study with survey data from 2167 long-term users of glucocorticoids (mean prednisone equivalent dose ± standard deviation [SD] of 16±14 mg/day for ≥60 days), weight gain was the most common self-reported adverse effect (70 percent of patients) [8]. In those on ≤7.5 mg/day of prednisone or equivalent, increasing duration of use was significantly associated with weight gain. Also, in an analysis of four prospective trials of glucocorticoids in RA, the use of 5 to 10 mg/day of prednisone or equivalent over two years was associated with an increase of mean body weight of 4 to 8 percent [9].

Ophthalmologic effects — The risk of both cataracts and glaucoma is increased in patients on glucocorticoids and is dose-related [9]. (See "Cataract in adults" and "Open-angle glaucoma: Epidemiology, clinical presentation, and diagnosis" and "Angle-closure glaucoma".)

●Cataracts – Cataracts commonly occur after prolonged glucocorticoid use. They are usually bilateral and develop slowly. They typically occur in a posterior subcapsular location and can usually be distinguished from senile cataracts.

There may be no minimal safe dose with respect to the risk of cataract formation, although risk is dose- and time-dependent and is more common with prednisone doses greater than 10 mg/day or with medications that have been administered for more than one year [17-19]. In a study that evaluated 122 patients with RA taking a mean dose of prednisone of 8 mg/day for an average of 6.9 years, 29 percent developed cataracts compared with 18 percent of matched controls [7]. Another study of patients with RA receiving a mean dose of 6 mg of prednisone daily for a mean of six years also found cataracts more common in patients on glucocorticoids than in patients not using prednisone (15 versus 4.5 percent) [6]. Another study indicated that even prednisone doses <5 mg/day were associated with cataracts [5].

●Increased intraocular pressure – Glucocorticoids can also increase intraocular pressure. This form of glaucoma occurs most commonly in patients who use glucocorticoid eye drops, although it has been observed in chronic and, to a lesser extent, acute systemic glucocorticoid use [20]. This risk of glaucoma associated with glucocorticoid use is discussed in more detail separately. (See "Open-angle glaucoma: Treatment", section on 'Glucocorticoids'.)

●Exophthalmos – Exophthalmos and swelling of the lids and ocular muscles are rare ophthalmologic complications of glucocorticoids [21,22].

A rare adverse effect of systemic, local, or even topical use of glucocorticoids is central serous chorioretinopathy [23-25]. This type of chorioretinopathy is associated with edema formation that can separate the retina from the choroid. Reduction of glucocorticoid dose is the most important element of treatment if it can be done without causing a dangerous exacerbation of the disease that is being treated with the glucocorticoids [26].

Cardiovascular effects — Glucocorticoid use has been associated with a variety of adverse cardiovascular effects including fluid retention, premature atherosclerotic disease, and arrhythmias. Cardiovascular disease risk is dose-dependent and may be low or absent in patients on low-dose glucocorticoid therapy [27].

●Fluid retention and hypertension – Higher-dose glucocorticoids may promote fluid retention, which is of particular concern to patients with underlying heart or kidney disease. This is not a problem in normal subjects because of the phenomenon of mineralocorticoid escape that prevents progressive fluid overload (see "Pathophysiology and clinical features of primary aldosteronism"). Hypertension is a known adverse effect of glucocorticoids and has been observed in up to 20 percent of patients with iatrogenic Cushing's syndrome [28]; however, it is a dose-related adverse effect and is unlikely to occur at lower doses of glucocorticoids [5,29]. Some studies have shown that long-term use of prednisone in patients with RA can be associated with a higher risk of hypertension, although the data are inconsistent when low doses of glucocorticoids are used [30-32]. In patients receiving low doses of glucocorticoids (eg, 10 mg/day of prednisone), significant hypertension may be better explained by age and initial blood pressure than by the glucocorticoids themselves [33]. The mechanism by which glucocorticoid therapy can raise the blood pressure is not well understood [34,35]. (See "Epidemiology and clinical manifestations of Cushing's syndrome", section on 'Hypertension'.)

●Premature atherosclerotic disease – Glucocorticoid use has been associated with increased rates of myocardial infarction, stroke, heart failure, and all-cause mortality [27,36]. Increasing attention is being given to the increased risk of premature atherosclerosis in RA and other systemic rheumatic disease. Thus, the relative risks of cardiovascular events associated with glucocorticoid use may be confounded by the role of the underlying inflammatory disease of the vascular endothelium. In a large population-based study comparing 68,781 glucocorticoid users with 82,303 nonusers who had never been hospitalized for cardiovascular disease, patients who received glucocorticoid doses ≥7.5 mg/day were more than 2.5 times more likely than patients who did not receive glucocorticoids to experience a cardiovascular event (defined as myocardial infarction, angina, coronary revascularization, hospitalization for heart failure, transient ischemic attack [TIA], or stroke) [27]. In a systematic review including studies of patients taking <10 mg daily of prednisone, there did not appear to be an association of low-dose prednisone use and cardiovascular risk factors (including atherosclerosis and lipid profiles), but there was a trend towards an increased risk of major cardiovascular events [37].

Studies in which the patients were limited to systemic rheumatic diseases found more variable results in terms of glucocorticoid use and cardiovascular disease risk. As an example, a prospective cohort study with 364 patients with polymyalgia rheumatica (PMR) followed for a median of 7.6 years found that there was no increased risk of cardiovascular outcomes in patients treated with glucocorticoids compared with the group of patients who just received nonsteroidal antiinflammatory drugs (NSAIDs) [38]. By contrast, a large population-based study of 8384 patients with incident RA found that glucocorticoid use was associated with a 68 percent increased risk of myocardial infarction [39]. The risk of myocardial infarction increased by 13 percent for each 5 mg/day increase in glucocorticoid dose. However, the potential for unmeasured confounders and possible misclassification using administrative data limit these findings. Also, a randomized trial with 223 patients with early RA followed prospectively for 10 years found an increased risk of cerebrovascular events in patients treated with prednisolone compared with those not receiving prednisolone (hazard ratio [HR] 3.7 [1.2-11.4]) [32]. However, there were no significant differences between the two treatment groups in composite cardiovascular events (which included an acute myocardial infarction or an ischemic stroke) or a first coronary event. Another population-based cohort study in the United Kingdom of 87,794 patients with inflammatory diseases including PMR and/or giant cell arteritis, inflammatory bowel disease, RA, systemic lupus erythematosus (SLE), and vasculitis reported increased cardiovascular risks associated with long-term prednisone use at doses of less than 5 mg per day [40].

The development of iatrogenic Cushing's syndrome may be a marker for patients at a higher risk of cardiovascular disease. A cohort of 547 patients diagnosed with iatrogenic Cushing's syndrome in a large general practice database had a significantly greater risk of a cardiovascular event compared with patients who were receiving similar doses of glucocorticoids but who did not have a diagnosis of Cushing's syndrome and patients who did not receive glucocorticoids [41]. (See "Epidemiology and clinical manifestations of Cushing's syndrome", section on 'Cardiovascular risk'.)

●Arrhythmias – An association of glucocorticoid use with risk of atrial fibrillation and flutter has been reported several studies [42-44]. In a population-based case-control study, current glucocorticoid use was more common among 20,221 patients with atrial fibrillation or flutter compared with the 202,130 population controls (6.4 versus 2.6 percent) [44]. Currently using glucocorticoids was associated with a significant increased risk of atrial fibrillation or flutter, compared with never having used glucocorticoids (adjusted OR 1.9). Risk was increased for new users and long-term users, but not for former users (ORs 3.6, 1.7, and 1.0, respectively) and was unrelated to whether or not pulmonary or cardiovascular disease was present.

Serious adverse cardiovascular toxicities, including sudden death, have been reported in occasional patients who have been given pulse infusions of glucocorticoids (eg, 1 g/day of methylprednisolone for multiple infusions) [45]. In many of these cases, however, it was difficult to determine whether this adverse effect was more likely attributable to glucocorticoids or to the underlying disorder necessitating the therapy. Thus, cardiac monitoring is indicated in patients with significant cardiac disease who are treated with pulse glucocorticoid therapy, especially those on diuretics, the use of which may also be associated with electrolyte disturbances such as hypokalemia.

●Pulmonary emboli and venous thrombotic events – Limited observational data suggest that glucocorticoids may be associated with an increased risk of pulmonary emboli and venous thrombotic events. In a retrospective study in the Netherlands that compared 648 patients using chronic glucocorticoids for asthma with the general population, both severe asthma and oral glucocorticoids appeared to be independent risk factors for pulmonary emboli (HR 3.33, 95% CI 1.16-9.93 and 2.82, 1.09-7.30, respectively) [46]. Another retrospective cohort study and self-controlled case series that used a nationwide dataset of private insurance claims in the United States assessed the risk of venous thromboembolism (VTE; among other adverse effects) in 27,452 adults using short-term (ie, less than 30 days) glucocorticoid use over 3 years. Within 30 days of glucocorticoid use, there was an increased rate of VTE (incidence rate ratio [IRR] 3.33, 95% CI 2.78-3.00), which then decreased over the subsequent 31 to 90 days [12]. Another national database analysis of 94,620 neurosurgical patients found that glucocorticoids were associated with an increased risk of pulmonary emboli (odds ratio [OR] 1.47, 1.13-1.90) and deep vein thrombosis (OR 1.55, 1.28-1.87) within 30 days of surgery [47]. There was no control over other time-varying factors, and only 6.3 percent of the prescriptions were for a prednisone-equivalent dose of less than 17.5 mg a day, suggesting that most of the exposures were to relatively higher doses of glucocorticoids. Limitations to both of the national database studies include potential misclassification of diagnoses in large databases and lack of accounting for comorbidities and concomitant medications; thus, causal relationships cannot be made.

●Possible hyperlipidemia – The effect of glucocorticoids on atherosclerotic vascular disease is thought to be mediated in part by elevated nonfunctional lipoprotein levels. However, studies analyzing this issue have had mixed results, and beneficial effects of glucocorticoids on dyslipidemia have also been observed. In one report, moderate- to low-dose prednisone (20 mg/day tapered to 5 mg/day over three months) had no significant adverse effect on lipoprotein levels if other risk factors were taken into account [48]. Another observational study concluded that glucocorticoid use was associated with a more favorable lipid profile in older adults (≥60 years) [49]. Studies in patients with SLE have indicated that the adverse effects of glucocorticoids on lipid profiles are dose-dependent, occurring only at prednisone doses greater than 10 mg/day [50-52]. Interpretation of these data is complicated by the difficulty of distinguishing effects due to disease activity from effects directly related to the medications themselves [53]. A rodent study found that the combination of prednisone and atorvastatin improves the lipid profile better than either drug alone; however, we are unaware of direct data in humans.

Glucocorticoids may act by leading sequentially to peripheral insulin resistance, hyperinsulinemia, and increased hepatic very low-density lipoprotein (VLDL) synthesis. However, glucocorticoid-induced reduction in corticotropin (ACTH) release also contributes to the lipid changes. In one report, for example, the administration of ACTH for three weeks to nine hyperlipidemic, glucocorticoid-treated patients (five of whom were transplant recipients) led to substantial reductions in total and low-density lipoprotein (LDL) cholesterol and triglycerides and to an increase in high-density lipoprotein (HDL) cholesterol [54]. ACTH may act, in part, by upregulating LDL-receptor activity.

Gastrointestinal effects — Glucocorticoids increase the risk for adverse gastrointestinal effects, such as gastritis, ulcer formation, and gastrointestinal bleeding. The estimated relative risks of glucocorticoids alone for gastrointestinal adverse effects vary from 1.1 (not significant) to 1.5 (marginally significant) [55,56]. However, the combination of glucocorticoids and NSAIDs results in a synergistic increase in the incidence of gastrointestinal events as shown by the following findings from two meta-analyses:

●Glucocorticoid use is associated with a nearly twofold increased risk of a gastrointestinal adverse effect among patients also taking NSAIDs when compared with those who use NSAIDs alone [57,58].

●The use of NSAIDs and glucocorticoids is associated with a fourfold increased risk of a gastrointestinal adverse effect compared with nonuse of either drug [56].

Whether substitution of a selective cyclooxygenase (COX)-2 inhibitor for a nonselective NSAID would lower this risk is unclear.

In addition to upper gastrointestinal morbidity, other complications associated with glucocorticoid use include visceral perforation [59-61] and hepatic steatosis (fatty liver) that can rarely lead to systemic fat embolism or cirrhosis [62,63]. Although glucocorticoids have also been implicated in causing acute pancreatitis [64-66], other studies, particularly in patients with SLE, have shown that the disease is causative, rather than the drugs, and that the drugs may be beneficial therapeutically [67,68]. However, the role of glucocorticoids in causing acute pancreatitis remains uncertain. (See "Etiology of acute pancreatitis".)

Given the apparent association of tocilizumab and tofacitinib with lower gastrointestinal tract perforations, the question has been raised as to whether the concomitant use of glucocorticoids can further increase this risk. In one study using registry data of patients on biologic agents for RA, the risk of lower intestinal perforation increased by 1.28 (95% CI 1.18-1.38) for each 5 mg increase in prednisone dose per day when added to tocilizumab [58]. In another cohort study that included 167,113 patients with RA, the use of tocilizumab, and possibly tofacitinib, was associated with an increased risk of gastrointestinal perforation [69]. The risk of gastrointestinal perforation appeared to be further increased with prednisone use of more than 7.5 mg daily (HR 2.29, 95% CI 1.39-3.78).

Glucocorticoids may mask the symptoms of serious gastrointestinal disease, an effect that may account, in part, for the increased risk of perforated sigmoid diverticular abscess associated with their use [61].

Bone and muscle effects

●Osteoporosis – Osteoporosis is a well-known adverse effect of glucocorticoid use and is discussed in detail elsewhere. (See "Clinical features and evaluation of glucocorticoid-induced osteoporosis" and "Prevention and treatment of glucocorticoid-induced osteoporosis".)

●Osteonecrosis – Osteonecrosis (avascular or ischemic necrosis of bone) has also been associated with glucocorticoid use, particularly with high doses of glucocorticoids. Glucocorticoid-induced osteonecrosis is discussed separately.

●Myopathy – Myopathy is an infrequent complication of glucocorticoid therapy. It presents as painless proximal motor weakness in both the upper and lower extremities. Glucocorticoid-induced myopathy is discussed in detail separately. (See "Glucocorticoid-induced myopathy".)

More acute and severe weakness noted in critically ill patients has been attributed at least in part to the use of glucocorticoids. Variously referred to as critical illness myopathy and acute quadriplegic myopathy, it has also been suspected to be due to an interaction between glucocorticoids and neuromuscular blocking agents. This is discussed in more detail elsewhere. (See "Neuromuscular weakness related to critical illness", section on 'Critical illness myopathy'.)

Neuropsychiatric effects — Glucocorticoids induce a range of psychiatric and cognitive symptoms, which depend upon dose and duration of therapy [70]. In most patients, these symptoms are mild and reversible, but emotional lability, hypomania, mania, depression, psychosis, delirium, confusion, or disorientation (which are more common in older patients), and cognitive changes including memory deficits may occur [71,72]. Disturbances in sleep are reported, especially with split doses or evening dosing that may interfere with the normal pattern of diurnal cortisol production. Akathisia (motor restlessness) is a common glucocorticoid side effect. The risk of developing a given neuropsychiatric disorder following glucocorticoid therapy may be increased among patients with a past history of that condition [72]. Older patients may be at higher risk for depression, mania, delirium, confusion, or disorientation [72]. We describe several conditions in more depth below:

●Mood disorders – Patients receiving glucocorticoids often experience an improved sense of well-being within several days of starting the medications; mild euphoria or anxiety may also be seen [71,73,74]. Hypomanic reactions and activated states are more common early in therapy than is depression, but the prevalence of depression is greater in patients on more longstanding therapy, even on low to moderate doses [71,73,75]. Patients with a family history of depression or alcoholism are at increased risk for affective diseases when given glucocorticoids [76].

More severe psychiatric symptoms can occur within a few days in patients receiving high doses of glucocorticoids. As an example, in one prospective but uncontrolled study of 50 patients receiving over 75 to 100 mg of prednisone or equivalent for greater than a week for various ophthalmologic indications, hypomanic symptoms were induced in approximately 30 percent, and depressive symptoms in approximately 10 percent, of patients by the end of one week [77]. No patient became overtly psychotic, developed dementia, or became delirious.

●Psychosis – Psychosis can occur but does so almost exclusively at doses of prednisone above 20 mg/day given for a prolonged period [9,78,79]. Approximately 10 percent of patients have persistent symptoms that may require treatment despite reduction of glucocorticoid dose [80]. Response to antipsychotic drug treatment is typically complete and occurs within two weeks of initiation of neuroleptics. Hypoalbuminemia may be a risk factor for glucocorticoid-induced psychosis in patients with SLE [81]. Patients with SLE who are on higher glucocorticoid doses present a particular problem since it is often difficult to differentiate psychosis due to prednisone from neuropsychiatric lupus, which may require high-dose glucocorticoid treatment. (See "Neurologic and neuropsychiatric manifestations of systemic lupus erythematosus" and "Psychosis in adults: Epidemiology, clinical manifestations, and diagnostic evaluation", section on 'Medical and substance/medication-related causes'.)

●Memory impairment – Memory impairment has been associated with glucocorticoid use. As an example, a cohort study of 115 patients with RA found that glucocorticoid use was a predictor of poor cognition when controlling for depression, disease severity and duration, and C-reactive protein (CRP) level [82]. Another report found that patients treated with prednisone doses of 5 to 40 mg/day for at least one year had a partial loss of explicit memory; older patients were more susceptible to memory impairment with less protracted treatment [70,83]. The effect on memory began as early as three months after the initiation of therapy. Approximately 1 percent of patients may be affected by more severe and persisting cognitive disturbances beginning during glucocorticoid treatment; this has been termed steroid dementia. In some patients, this condition may not remit for between 1 and 11 months after discontinuing the medication.

●Other symptoms – In a retrospective analysis involving 372,696 patients in general practices in the United Kingdom, there was a five- to sevenfold increased risk of completed or attempted suicide among patients receiving glucocorticoids, compared with patients with the same diagnoses who were not receiving such medications; however, the absolute risk was extremely low, approximating 0.1 cases per 100 patient-years of therapy [72]. Younger patients were at higher risk. The observational nature of the study and potential for unknown confounding variables limit study interpretation.

Rare cases of pseudotumor cerebri have been associated with glucocorticoid use, although higher doses are often effective in alleviating this generally self-limiting disorder [84]. Akathisia can occur even in patients taking low doses. An increased risk of panic disorder may be present [72].

Metabolic and endocrine effects

●Hyperglycemia – Systemic glucocorticoids cause a dose-dependent, usually mild, increase in fasting glucose levels and a greater increase in postprandial values in patients without preexisting diabetes mellitus, but the development of de novo diabetes in a patient with initially normal glucose tolerance is uncommon [85]. In a case-control study of Medicaid recipients, the relative risk of developing hyperglycemia requiring glucose-lowering therapy increased progressively with increasing glucocorticoid dose [86]. The relative risk increased from 1.8 in patients treated with the equivalent of less than 10 mg/day of prednisone to 10.3 in those with the equivalent of more than 30 mg/day of prednisone [86]. Risk factors for new-onset hyperglycemia during glucocorticoid therapy are thought to be the same as those for other patients, including a family history of diabetes, increased age, obesity, and a history of gestational diabetes [87]. Transient hyperglycemia can also occur after intraarticular glucocorticoid therapy. (See "Comorbidities that impact management of osteoarthritis", section on 'Diabetes mellitus'.)

Patients with diabetes mellitus or glucose intolerance exhibit higher blood glucose levels while taking glucocorticoids, leading to increased difficulty with glycemic control. In addition, new-onset hyperglycemia or, rarely, a nonketotic hyperosmolar state or diabetic ketoacidosis develops without warning in patients with early subclinical diabetes or glucose intolerance [31,85,86].

The mechanism by which glucocorticoids cause hyperglycemia is multifactorial, including augmentation of hepatic gluconeogenesis, inhibition of glucose uptake in adipose tissue, and alteration of receptor and postreceptor functions [1,88-90]. It is also possible that some underlying disorders for which glucocorticoids are used, such as RA, may independently predispose to a higher rate of glucose intolerance [91].

●Hypothalamic-pituitary-adrenal axis suppression – Administration of exogenous glucocorticoids can suppress the hypothalamic-pituitary-adrenal (HPA) axis. Abrupt cessation or rapid withdrawal of glucocorticoids in such patients may cause symptoms of adrenal insufficiency. The approach to withdrawal of glucocorticoids, HPA axis suppression, and the clinical manifestations of adrenal insufficiency are presented separately [92]. (See "Glucocorticoid withdrawal" and "Pharmacologic use of glucocorticoids", section on 'HPA axis suppression' and "Clinical manifestations of adrenal insufficiency in adults".)

Immune system effects — Systemic glucocorticoids have many effects upon innate and acquired immunity that predispose to infection, resulting in a dose-dependent increase in the risk of infection, especially with common bacterial, viral, and fungal pathogens. Glucocorticoids may be associated with a greater infection risk among patients with RA, compared with other antirheumatic medications, such as the anti-tumor necrosis factor (TNF)-alpha agents [93]. In one large study of patients with RA, current and recent doses were most strongly associated with such risk, but the data also suggested a cumulative risk effect from doses taken during the preceding two to three years [94]. In another large observational study with patients with RA, doses as low as 5 mg or less of prednisone were associated with a small but clinically significant risk of infection [95]. The risk of infection with glucocorticoid therapy and the mechanisms underlying such risk are discussed in more detail elsewhere. (See "Glucocorticoid effects on the immune system".)

In addition to glucocorticoid dose, factors influencing infection risk include the underlying disorder, the presence of concomitant immunosuppressive therapies, hospitalizations, lymphopenia, and diabetes mellitus [93,96]. Older patients and those with lower functional status are also at higher risk for infection. In addition, patients taking glucocorticoids may not manifest signs and symptoms of infection as clearly, due to the inhibition of cytokine release and associated reduction in inflammatory and febrile responses. This can impair early recognition of infection. (See "Glucocorticoid effects on the immune system", section on 'Infection risk' and "Glucocorticoid effects on the immune system".)

Inhaled and topical glucocorticoids are usually not implicated in increased risk of systemic infections, in contrast to the effects seen with systemic agents. The side effects of inhaled and topical glucocorticoids are reviewed elsewhere. (See "Major side effects of inhaled glucocorticoids" and "Topical corticosteroids: Use and adverse effects", section on 'Adverse effects'.)

Hematologic effects — Pharmacologic doses of glucocorticoids often result in an increased white blood cell count (leukocytosis) that is due primarily to an increase in neutrophils (neutrophilia). This phenomenon is due to a decreased proportion of neutrophils that are adhering to the endothelium. This effect of glucocorticoids is discussed in detail elsewhere. (See "Approach to the patient with neutrophilia", section on 'Medications'.)

SPECIAL POPULATIONS

Young children — Growth impairment is commonly seen in children receiving glucocorticoids. The effect is most pronounced with daily therapy, may be less with an alternate-day regimen, and can occur with inhaled glucocorticoids. The effects of glucocorticoid therapy on growth in children is discussed in detail separately. (See "Causes of short stature", section on 'Glucocorticoid therapy'.)

Children are more susceptible to cataract formation compared with adults. (See "Cataract in children", section on 'Glucocorticoids'.)

Pregnancy — The risks associated with glucocorticoid use during pregnancy and lactation are discussed elsewhere. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Glucocorticoids'.)

GENERAL TREATMENT CONSIDERATIONS AND MONITORING — We try to limit the adverse effects of glucocorticoids by taking the following steps:

●Use of the lowest dose of glucocorticoids for the shortest period of time needed to achieve the treatment goals

●Management of preexisting comorbid conditions that may increase risk when glucocorticoids are required

●Monitoring of patients under treatment for adverse effects who may benefit from additional intervention

Assess for risk factors for complications — Preexisting conditions that should be assessed for and treated when glucocorticoids are to be instituted include [97]:

●Diabetes mellitus

●Poorly controlled hypertension

●Heart failure and peripheral edema

●Cataract or glaucoma

●Peptic ulcer disease

●Presence of infection

●Low bone density or osteoporosis

Patients who also require concomitant treatment with nonsteroidal antiinflammatory drugs (NSAIDs) or anticoagulants may require prophylaxis to prevent gastroduodenal toxicity. (See "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity".)

Immunization requirements — Patients who require an extended course of glucocorticoids should receive appropriate immunizations prior to the institution of therapy (table 2).

The safety of live virus vaccines in specific patient groups is described briefly below but is reviewed in detail separately. (See "Measles, mumps, and rubella immunization in adults" and "Vaccination for the prevention of chickenpox (primary varicella infection)" and "Vaccination for the prevention of shingles (herpes zoster)" and "Immunizations in hematopoietic cell transplant candidates and recipients" and "Immunizations in solid organ transplant candidates and recipients" and "Immunizations in adults with cancer" and "Immunizations in autoimmune inflammatory rheumatic disease in adults".)

In general, live virus vaccines may be administered to patients who have taken:

●Prednisone or its equivalent in doses of less than 20 mg/day for 14 days or less

●Glucocorticoids used for long-term physiologic replacement

●Glucocorticoids administered topically, by aerosol, or by intraarticular or bursal injection, provided that there is no clinical or laboratory evidence of immunosuppression

If higher doses of glucocorticoids are taken, the combination measles-mumps-rubella (MMR) vaccine and other live virus vaccines should not be administered for one month after the cessation of glucocorticoid therapy. The immune response to other vaccines may be compromised by glucocorticoid administration in doses in excess of those noted above.

Prevention of opportunistic infection — As described above, it is well known that glucocorticoids are associated with an increased risk of infection (see 'Immune system effects' above). In some circumstances, the use of prophylaxis for an opportunistic infection with Pneumocystis jirovecii pneumonia (PCP) is recommended. However, this varies with the dose and duration of glucocorticoid therapy, as well as for the underlying disease being treated (see appropriate topic reviews).

Prevention of osteoporosis — Depending on the length of the course of glucocorticoid treatment, osteoporosis prevention may be pursued at the start of therapy. Adequate dietary calcium and vitamin D intake should be encouraged. Guidance for the prevention and treatment of glucocorticoid-induced osteoporosis is discussed in detail separately. (See "Prevention and treatment of glucocorticoid-induced osteoporosis".)

Monitoring for adverse effects — Patients should be routinely asked about adverse effects related to glucocorticoids. During treatment with glucocorticoids and depending upon individual risk factors such as dose and duration of glucocorticoid usage, other medications being used, and comorbidities, particular attention should be given to [97]:

●Osteoporosis

●Infection

●Diabetes or glucose intolerance

●Cataracts or glaucoma

Patients on prolonged moderate- to high-dose therapy should be examined periodically by an ophthalmologist to promote early detection of cataracts and glaucoma.

Symptomatic diabetes mellitus or asymptomatic but clinically significant hyperglycemia that are glucocorticoid-induced are generally treated pharmacologically in the same way that they are in patients with diabetes mellitus or glucose intolerance in the absence of glucocorticoid therapy. Glucocorticoid-induced hyperglycemia improves with reduction in the dose of glucocorticoid and usually reverses when the medication is stopped, although some patients develop persistent diabetes [98,99]. (See "Management of type 2 diabetes mellitus in children and adolescents", section on 'Pharmacologic agents' and "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Glycemic management'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Side effects of anti-inflammatory and anti-rheumatic drugs".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Oral steroid medicines (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Factors related to glucocorticoid toxicity – The adverse effects of glucocorticoids are both dose- and duration-dependent. The effects of glucocorticoids are mediated by cytosolic glucocorticoid receptors and result from both genomic and nongenomic mechanisms that also have a role in the therapeutic effects of these agents. Genetic polymorphisms in the glucocorticoid receptor and in glucocorticoid metabolism may explain heterogeneity in glucocorticoid toxicities observed. (See 'Mechanisms of adverse effects' above and 'Dose-related effects' above.)

●Adverse effects – Glucocorticoids have adverse effects on many organ systems (table 1). Adverse effects range from those that are not necessarily serious but are displeasing to patients (eg, Cushingoid appearance) to those that are life-threatening (eg, serious infections). Some adverse effects, such as accelerated reductions in bone mineral density or early cataracts, may be largely asymptomatic until later manifestations develop that require medical attention (eg, acute vertebral collapse, cataract requiring surgical extraction). The following organ systems can be affected by systemic glucocorticoids to varying degrees (see 'Organ-based toxicity of systemic glucocorticoids' above):

•Dermatologic effects and appearance (see 'Dermatologic effects and appearance' above)

•Ophthalmologic effects (see 'Ophthalmologic effects' above)

•Cardiovascular effects (see 'Cardiovascular effects' above)

•Gastrointestinal effects (see 'Gastrointestinal effects' above)

•Bone and muscle effects (see 'Bone and muscle effects' above)

•Neuropsychiatric effects (see 'Neuropsychiatric effects' above)

•Metabolic and endocrine effects (see 'Metabolic and endocrine effects' above)

•Immune system effects (see 'Immune system effects' above)

•Hematologic effects (see 'Hematologic effects' above)

●General treatment considerations and monitoring – We try to limit the adverse effects of glucocorticoids by taking the following steps (see 'General treatment considerations and monitoring' above):

•Use of the lowest dose of glucocorticoids for the shortest period of time needed to achieve the treatment goals

•Management of preexisting comorbid conditions that may increase risk when glucocorticoids are required

•Monitoring of patients under treatment for adverse effects who may benefit from additional intervention

Additional considerations prior to longer-term therapy with systemic glucocorticoids include addressing appropriate immunization requirements, prevention of opportunistic infections, and prevention of osteoporosis. (See 'Immunization requirements' above and 'Prevention of opportunistic infection' above and 'Prevention of osteoporosis' above.)