Glucocorticoid therapy is a risk factor for cardiovascular diseases

Introduction

Medicines are an important component of healthcare, they are able to minimize symptoms, as well as to cure or control the course of diseases, while improving the quality and life expectancy of patients. The downside of using such medicines is abnormal reactions that can cause iatrogenic diseases with fatal outcomes. The immediate task of the physician is to compare the risks and minimize the negative consequences of the use of medicines.

Synthetic glucocorticoids (GCs) are a vivid illustration of this problem. GCs have anti-inflammatory and immunosuppressive effects, due to which they are widely used in the treatment of many abnormal conditions, such as bronchial asthma, rheumatological diseases, inflammatory bowel diseases, in transplantology, etc.

The prevalence of the systemic (injectable or oral) use of GCs (according to a population study conducted in Denmark) increased from 3% to 6.7–7.7% in people aged 60–79 years and to 9.7–11% in patients older than 80 years [1]. In France, oral GCs are used by 14.7% to 17.1% [2]. Obtaining accurate data on the statistics of the use of GCs in Russia is a difficult task, especially taking into account the unjustified or independent GC administration against the background of the spread of the new coronavirus infection COVID-19.

GCs exert their effect by binding to glucocorticoid receptors (GRs), which are present in all cells of the body containing nuclei, as well as by interacting with transcription factors. In addition to their genomic effects, GCs also have non-genomic effects and affect many different signaling pathways [3][4].

Despite the therapeutic effects, the use of GCs is limited by two main disadvantages. First, these are side effects, especially pronounced with prolonged use and high doses. Hyperglycemia, cardiovascular diseases (CVDs), osteoporosis, and infectious complications are considered to be especially serious ones [5]. The development of side effects can affect the tactics of therapy or increase the risk of negative consequences, up to fatal ones. A recent population study involving more than 87 thousand people with autoimmune diseases showed a dose-dependent increase in the risk of CVD with oral administration of GCs [6]. It is worth noting that long-term use of even low doses of GCs (less than 5 mg per day in terms of prednisone), previously considered safe, increased the risk of CVD. A similar conclusion was reached by a group of French scientists Roubille et al. [7]. A ten-year follow-up of patients with rheumatoid arthritis showed an increase in the risk of CVD and other side effects among those receiving GCs at low doses (1.9 mg/day (IQR 0.6–4.2)) over time. The risks increased with a follow-up duration of more than six years and reached a maximum by the tenth year.

The second disadvantage of GC therapy may be considered to be generalized or acquired resistance. It can be caused by initial changes in the GR NR3C1 gene or can be secondary. These changes are reduced to a decrease in the pool of the active α-isoform of the receptor, and may also be associated with the presence of polymorphisms of the NR3C1 gene. Patients with resistance to GCs often require higher doses for a long time to achieve the effect of therapy, which leads to a greater likelihood of adverse side effects and may even exacerbate relative resistance to GCs [8].

This review considers the effect of GCs on the state of the cardiovascular system (CVS), as well as the side effects of GCs that increase the risk of CVD and the mechanisms of their development.

Cardiovascular system and glucocorticoids

GCs play a critical role in the development of the CVS. A study on animal models revealed that the effect of GCs on mouse fetal myocardiocytes increased the activity of mitochondria, improved cell contractility, and led to the appearance of mature myofibrils [9]. A mouse model with no expression of the GR gene in cardiomyocytes showed that such individuals were born phenotypically healthy, but after six months, myocardial hypertrophy developed, followed by dilated cardiomyopathy and premature death of animals [10]. Another study of the cardiac tissue of mice without GR expression showed a violation of contractility, as well as a gender difference: the development of myocardial hypertrophy was noted only in male mice [11].

Nevertheless, speaking about the effects of GC excess, it is impossible to unambiguously assess the positive impact on the development of the CVS. In an experimental model, Peng et al. [12] administered dexamethasone to pregnant rats, after which the state of the CVS of adult male offspring was evaluated. There was a decrease in the ejection fraction of the left ventricle, an increased size of myocardial infarction, and apoptosis of cardiomyocytes compared to offspring not exposed to GCs.

The use of GCs during pregnancy or in premature infants in order to accelerate the development of the pulmonary system or to reduce mortality may have a negative side. It is likely that exposure to GCs may have different consequences for the proliferation and function of cardiomyocytes, depending on the gestational age and sex of the infant. Taking into account the complex GC effect on hemodynamics and heart development, it would be surprising if this did not affect the subsequent growth of the heart and cardiovascular events in the future [13].

It has been repeatedly noted in studies that activation of the hypothalamic-pituitary-adrenal axis of the mother as a result of stress can lead to cardio-metabolic disorders in children in later life, such as hyperglycemia, diabetes mellitus, obesity, CVD, decreased fetal height and weight, as well as weight loss of the adrenal glands and pancreas [14]. For example, according to a meta-analysis by Burgueño et al. [15], in an animal model in rabbits, prenatal maternal stress (and, as a consequence, an increase in endogenous GCs) was associated with an increase in the body mass index (BMI) of their offspring. Lamichhane et al. [16] also reported links between psychological prenatal maternal stress, overweight and obesity in children. Other reviews have reported the key role of GCs in fetal development and programming the development of diseases at a later stage of adulthood [17].

The level of endogenous GCs in adult life is not less important for the CVS. The influence of the basal concentration and peaks of circadian cortisol secretion on the outcomes of CVD in people without hypercorticism syndrome is emphasized by many authors [18–20]. High cortisol levels in response to stressful exposure, as a rule, predicted an increased risk of an angina attack, but better survival, while increased basal and circadian production were associated with disease progression, worse prognosis and mortality. However, higher serum cortisol levels immediately after a heart attack were associated with better survival, probably related to adaptive mechanisms. In people who underwent surgery, a high basal level and more flattened circadian rhythms of cortisol secretion predicted worse results of surgical intervention [21].

The effect of stress and, as a consequence, altered endogenous GC levels in patients without Cushing's disease or syndrome was also confirmed as a result of a 27-year follow-up of more than 136 thousand Swedish population. It was revealed that the frequency of CVD in study participants with diagnosed stress-related conditions was 1.64 times higher than in people who did not face a stressful event, including their brothers and sisters [22].

Thus, a higher level of endogenous GCs before stress exposure and their exogenous intake in the prenatal period are associated with impaired CVS function and a worse prognosis of cardio-metabolic risks during life.

Endogenous glucocorticoids, substitution therapy and CVD

The most striking example of the consequences for the CVS of prolonged exposure to excess GCs is patients with Cushing's disease or syndrome. The increased endogenous cortisol production in such patients leads to the development of many metabolic disorders, including dyslipidemia, obesity, and disorders of carbohydrate metabolism, which are risk factors for CVD [23]. Statistical data confirm the higher risks of cardiovascular events, such as myocardial infarction, stroke, coronary heart disease, and chronic heart failure, characteristic of patients with endogenous hypercorticism syndrome [24][25], and mortality from CVD is 5.5 times higher than the average in the population [26].

A curious fact is that even with successful treatment and the onset of remission of endogenous hypercorticism, the increased cardiovascular risk continues to persist both after a year and after 5 years [27][28]. This is probably due to the consequences of metabolic changes (obesity, hypertension, dyslipidemia) that occurred during the activity of the disease. It was noted that even during the remission of the disease, elevated levels of proinflammatory and prothrombotic substances continue to persist, causing an increased risk of CVD [27][29]. The frequent need for the use of GC replacement therapy in patients after surgical or other treatment of endogenous hypercorticism should be borne in mind as well, which may also affect outcomes.

Also, despite the achievement of normocortisolemia, a violation of the ultradian GC concentration can have its effect. In an animal model, it was shown that with prolonged hyperstimulation of GRs, a significant flattening of rhythmic changes in the concentration of cortisol in the plasma of rats occurred [30].

No less intriguing are the results of a population study by Skov et al. [31]. In patients with primary autoimmune adrenal insufficiency, the risk of CVD development was higher than in the control group of healthy people. At the same time, there was an increased risk of coronary heart disease, but not cerebrovascular diseases and only among women. The risk of CVD correlated with the amount of replacement doses of GCs and mineralocorticoids. The selection of a replacement GC dose based on clinical data and the inability to use more objective criteria for assessing the adequacy of doses may probably lead to supraphysiological levels of GCs that are not yet clinically manifested and therefore remain unrecognized. The reasons for the gender difference in this situation remain undetermined. However, it can be assumed that the lower levels of cortisol [32] and aldosterone [33] characteristic of women normally cause a greater probability of excess hormone replacement dose in adrenal insufficiency.

Thus, despite the achievement of normocortisolemia, the available methods of treatment of Cushing's syndrome or disease, as well as the selection of doses of hormone replacement therapy, apparently do not make it possible to completely neutralize the consequences of the disease.

The effect of glucocorticoids on the pathogenetic aspects of CVD development

The effect of GCs on atherosclerosis and hemodynamics. It is believed that the increased risk of CVD and mortality in patients with endogenous hypercorticism is a consequence of the activity of atherosclerotic processes [34]. According to the results of a meta-analysis by Lupoli et al., it is noted that the thickness of intima-media and the frequency of occurrence of atherosclerotic plaques in Cushing's syndrome or disease were significantly higher than in healthy participants of the control group. The development of atherosclerosis occurs in such situations, apparently, independently of other risk factors (smoking, glycemia, BMI, etc.) [35].

It is known that prolonged exposure to supraphysiological levels of GCs is associated with the development of irreversible atherosclerotic changes [36]; nevertheless, in early experimental models, it was shown that GCs had protective properties. Low doses of GCs (0.125 mg/day. dexamethasone) protected against the progression of atherosclerosis and reduced the number of macrophages and the formation of foam cells in a rabbit model with a cholesterol-rich diet [37]. Recent studies also confirm that exposure to GCs contributed to a reduction in the size of damage during plaque formation in a mouse model [38], and also reduced the accumulation of cholesterol by macrophages [39].

The contradictory data were obtained on other animal models. In mice with a genetically determined absence of the enzyme 11ß-hydroxysteroid dehydrogenase (type 2), which inactivates GCs, a faster progression of atherosclerosis was observed [40]. At the same time, mice without the enzyme 11ß-hydroxysteroid dehydrogenase (type 1), which leads to the transition of GCs to the active form, had a smaller size of the atherosclerotic plaque [41]. Probably, GCs can have a twofold effect on cells, contributing to or preventing the development of atherosclerosis.

In early experimental models, it was shown that GCs could induce endothelial dysfunction due to changes in vasodilation and increased vasoconstriction, which can contribute to the development of atherosclerosis [42]. However, these data are also ambiguous. As was shown by Hafezi-Moghadam et al., exposure to high GC doses increased the activity of vasodilators [43]. Some effects are difficult to attribute to beneficial or harmful, such as, for example, inhibition of smooth muscle cell migration, which can reduce the size of the damage, but increase the likelihood of plaque rupture [44].

GCs are also vital hormones involved in the regulation of blood pressure [45][56]. The mechanism of GC action in the development of hypertension is multicomponent: mineralocorticoid GC activity, activation of the renin-angiotensin-aldosterone system, an increase in vasoconstrictors (endothelin-1), a decrease in vasodilators (NO, etc.), increased resistance of small arteries due to the effects of growth factors, angiogenesis factors and increased vascular wall thickness, increased sensitivity to catecholamines, metabolic changes such as visceral obesity, impaired secretion of cytokines and adipokines [47].

Thus, it is unlikely that an unambiguous answer can be given to the question of whether GCs are a negative factor in the progression of atherosclerosis or have protective properties [48]. However, they contribute to increased pressure, which is an independent risk factor for many CVDs. Therefore, GCs have a multifactorial effect; the determination of target cells in the CVS and the comparison of the effects of endogenous hormones and synthetic GCs on them with systemic use, as well as an understanding of the mechanisms regulating the response of target cells, can be useful for resolving the described paradox.

Metabolic changes under the action of GCs. The use of GCs is associated with a violation of carbohydrate metabolism, hyperglycemia, and the development of steroid diabetes, which accounts for 2% of the total number of diabetes mellitus [49]. Long-term metabolic changes caused by GCs include weight gain, redistribution of adipose tissue and increased circulation of free fatty acids, decreased muscle mass, enhanced gluconeogenesis and increased endogenous glucose, bone loss, and a higher risk of fractures. Most of them are caused directly by the negative effects of GCs on the endocrine function of the pancreas and peripheral insulin sensitivity [50].

It is noted that the development of Cushing's signs depends on the dose and duration of treatment. The study involved more than 2 thousand participants who took an average of 15 mg/day for more than 60 days. Prednisone: 80% had an increase in body weight, 10% had hyperglycemia at a higher dose [51].

There is a paradox in the response of beta cells of the pancreas to the GC action in vitro and in vivo. Most in vitro studies have shown a negative GC effect on the proliferation, survival of β-cells, and insulin secretion [52–54]. A recent study [55] showed that there was no violation of insulin secretion or glucose response when exposed to GCs, despite the violation of the ion current in beta cells. Therefore, Fine et al. noted an increase in cyclic AMP under the GC action, which supported insulin secretion. However, this effect is suppressed due to lipotoxicity caused by the GC action. In this regard, it is necessary to consider the use of complex systems that take into account lipolysis and dyslipidemia at supraphysiological levels of GCs, for a better understanding of the effect of GCs on the pancreatic function.

Inter-organ interaction in vivo determines a wide heterogeneity in the response of the endocrine function of the pancreas to GCs, depending on the model, dose, and duration of the interaction. Also, GC therapy in adult animal models can inhibit the release of insulin in response to glucose [56][57], but other studies have demonstrated the inclusion of adaptive mechanisms to maintain or increase the mass of beta cells through proliferation or neogenesis [58], which leads to improved insulin secretion and better control of glycemia. Exposure to GCs at the prenatal stage reduces the number of beta cells in the adult body and ultimately leads to impaired insulin secretion later in adult rodents and humans [59].

Equally important in the formation of carbohydrate metabolism disorders is the development of insulin resistance and the effects of GCs on adipose tissue. In humans and rodents, chronic use of GCs leads to insulin resistance of adipose tissue, an increase in the number of macrophages in adipose tissue [60], the amount of white adipose tissue, a decrease in subcutaneous adipose tissue, and an increase in lipolysis, characterized by an increase in free fatty acids in the bloodstream and fat accumulation in the liver, skeletal muscles, and pancreas [61–64]. The negative effects of GCs are mediated through both GC and mineralocorticoid receptors [65]. Thus, in vivo studies have revealed a complex relationship between insulin resistance and cardio-metabolic disorders mediated by GC signaling pathways [66].

Another target of GCs is the liver. GCs play an important role in the transition from anabolism to catabolism during fasting. Their excessive amount or introduction from the outside distorts the physiological effects, which can lead to excessive deposition of lipids (mainly triglycerides) and non-alcoholic fatty liver disease (NAFLD).

A GC-induced increase in lipid deposition in the liver is mediated by a variety of mechanisms [67], including increased food intake, stimulation of gluconeogenesis, and synthesis of fatty acids in the liver [68] due to high levels of glucose, insulin, and GCs, as well as increased release of free fatty acids from adipose tissue and their absorption and deposition in the liver [69]. At the same time, the effect of GCs on an increase in the level of triglycerides in blood plasma is largely mediated by the inhibition of the activity of plasma lipoprotein lipase. Excess of GCs also leads to the inhibition of β-oxidation of fatty acids, which causes a further increase in the level of triglycerides in the liver [70]. These changes explain the rapid development of fatty hepatosis in experimental animals even after several days of GC therapy.

NAFLD is now recognized as a risk factor for adverse cardiovascular outcomes, regardless of the presence of metabolic syndrome [71][72]. The question remains whether the prognostic value of NAFLD in the development of CVD is related to steatohepatitis or simple steatohepatosis, and therefore additional research is needed to understand the pathophysiology linking NAFLD with CVD.

Inter-organ interaction and the effect of glucocorticoids. Muscle tissue is known as one of the main participants in the exchange and maintenance of blood glucose levels [73]. Muscles accumulate glucose in the form of glycogen and are also a source of amino acids that the liver uses during gluconeogenesis. The action of GCs is aimed at catabolism and mobilization of resources to maintain homeostasis, which is mediated by a decrease in the absorption of glucose from the blood by muscles, as well as their destruction and release of amino acids — the substrate of gluconeogenesis. Confirmation of a significant relationship between the effect of GCs and insulin sensitivity was the work that showed the normalization of the expression of the GR in muscle tissue in the treatment of type 2 diabetes mellitus and a decrease in insulin resistance [74][75].

The use of GCs is accompanied by the development of GC-induced myopathy, a decrease in muscle mass, and weakness [76]. The use of even low doses, regardless of the duration and method of administration, can cause myopathy [77]. The effect of GCs on muscles is mediated by a decrease in anabolic processes [signal transmission by insulin-like growth factor-1/PI3K/Akt (protein kinase B) is disrupted, ubiquitination and degradation of MyoD (the main transcriptional switch of muscle development and regeneration) occur, the transport of amino acids into the cell decreases] and activation of catabolism [degradation of muscle proteins occurs due to the activation of muscle-specific E3 ubiquitin ligase atrogin-1/MAFbx (F-box Protein 32), activation of calpains and cathepsins leads to dissociation of actin and myosin, the cascade of caspases and cytochrome C is also involved] [78][79].

The influence of muscle mass on the quality of life and adverse outcomes of various diseases has recently been actively emphasized by scientists. Studies have shown that low muscle mass is a predictor of mortality during hospitalization for chronic heart failure [80], as well as chronic kidney disease, cirrhosis, and other conditions [81][82]. Among the participants in the ten-year follow-up, people with greater muscle mass had an 81% (95% CI from 0.04 to 0.85) lower risk of developing CVD compared to participants with the lowest muscle mass [83].

Thus, muscle tissue is one of the most important participants in pathogenetic processes in various conditions, and a decrease in muscle mass and muscle strength, including under the influence of GCs, is considered one of the predictors of adverse outcomes, risks of cardio-metabolic disorders and CVD [84][85].

The equally important is the inter-organ communication of skeletal bones and internal organs. Excessive exposure of exogenous or endogenous GCs to bone tissue is associated with the development of osteopenia and glucocorticoid-induced osteoporosis. This condition is accompanied by a decrease in the activity of osteoblasts, a decrease in osteocalcin, and a change in the concentration of other mediators. Osteocalcin has direct antidiabetogenic effects and affects energy metabolism [86]. A decrease in osteocalcin is accompanied by a decrease in adiponectin and glucose uptake by adipocytes, an increase in fat accumulation by the liver, and a decrease in glucose uptake and the number of mitochondria in muscles, which in turn leads to an increase in insulin secretion by the pancreas and an aggravation of insulin resistance [87].

Despite the clinical significance of muscle dysfunction and osteoporosis, these changes remain unnoticed for a long time for patients. At this time, one of the most frequently mentioned and correspondingly important side effects of GCs in patients' lives is not only weight gain, which is often associated with increased food intake [88][89], but also insomnia, as shown by a recent analysis [90]. Sleep disorders are considered to be directly caused by the action of GCs [91], while the role of insomnia and impaired melatonin production has been repeatedly emphasized as an independent risk factor for metabolic disorders and CVD risk [92][93].

The effect of glucocorticoids on hemostasis

A particularly dramatic risk factor against the background of the use of GCs at the present time is a tendency to hypercoagulation. During the spread of the SARS-CoV-2 virus, GCs were able to be used in inpatient conditions in patients requiring respiratory support, according to the RECOVERY study [94]. The anti-inflammatory effects of GCs and their adaptive role are accompanied by an increase in the risk of thrombosis [95]. As has been shown, even in healthy people with the use of GCs in comparison with a placebo, there was a tendency to hypercoagulation. Taking into account the observed tendency to deep vein thrombosis and pulmonary embolism against the background of COVID-19, the use of GCs becomes riskier in this situation and requires careful monitoring, choice of doses and duration of therapy [96]. At the same time, it is believed that both the underlying disease and the fact of the use of GC therapy contribute to thrombosis [97]. In this regard, it is necessary to inform doctors and patients about the need for a balanced approach to the appointment of GCs and the inadmissibility of the independent use of this group of drugs, including at the outpatient stage.

Conclusion

Summing up, it is important to emphasize that GCs are vital hormones that cause adaptation and regulate metabolic processes in all human organs. It must be remembered that their supraphysiological amount, even in the case of small doses, may be accompanied by a violation of carbohydrate and lipid metabolism, mental changes (including sleep disorders and hyperphagia), changes in hemodynamics, hypercoagulation, effects on the bone, muscle, adipose tissue, and liver metabolism. All in all, these phenomena lead to the formation of atherosclerotic changes and increased cardiovascular risks in the future, even with the use of GCs in the perinatal period. Due to the abundance of side effects, GCs can be called a “double-edged sword”. The specialist needs to weigh all the possible risks and benefits of using GCs, to give preference to short-term use, as well as to inform the patient about possible side effects and monitor them in a timely manner.