Adrenal Insufficiency: Outcomes With Glucocorticoid Therapy
Adrenal Insufficiency: Outcomes With Glucocorticoid Therapy
Adrenal crisis, which is a factor in the increased mortality of individuals with AI, has been estimated to occur at an incidence of approximately 6·3 per 100 patient-years, based on a study of 883 patients in Germany. This number is higher than the incidence of approximately 3·3 per 100 patient-years reported previously in a smaller German study of 53 individuals with AI, but is broadly consistent with a frequency of 8% per year reported in the UK. In a retrospective study of 137 Japanese patients with AI, 40 (29%) had experienced at least one adrenal crisis. Some 7·4% of German patients with SAI and 10·6% of those with PAI had experienced an adrenal crisis on at least four or more occasions, indicating that there is a subgroup of patients at high risk.
Adrenal crises are mainly precipitated by gastrointestinal infection and fever (45%), but also by other stressful events, such as major pain, surgery, emotional distress, heat and pregnancy. Patients with comorbidities are at higher risk than those without. In those with SAI, female sex and diabetes insipidus were associated with an increased frequency of adrenal crisis. The incidence of adrenal crisis in the German study by Hahner et al. was not influenced by educational status, body mass index, glucocorticoid dose, dehydroepiandrosterone treatment, age at diagnosis, hypogonadism, hypothyroidism or GH deficiency. In a retrospective Japanese study, however, steroid replacement therapy for more than 4 years was the largest single risk factor for an adrenal crisis, followed by the presence of a mental disorder and sex steroid deficiency. In patients with SAI (n = 115), sex steroid deficiency was the greatest risk factor for an adrenal crisis. Individuals with untreated hypogonadism had a significantly higher relative risk (3·70; 95% CI, 1·71–7·98) than those without hypogonadism or with treated hypogonadism. Furthermore, among patients with hypogonadism aged younger than 50 years, individuals treated with sex hormones (5/51) experienced adrenal crisis less frequently than those not receiving sex steroid replacement (7/11; P = 0·0004). In a Canadian study of children with PAI (n = 102) and SAI (n = 34) conducted between 1973 and 2007, the use of stress glucocorticoid doses by parents increased significantly after 1997 (P < 0·05). The proportion of patients with signs or symptoms of acute adrenal failure therefore decreased to 27% after 1997 (P < 0·01). A major reason for hospitalizations in this study was suboptimal adherence to glucocorticoid stress protocols, indicating the need for continuous patient and carer education.
A detailed cross-sectional retrospective study by means of a disease-specific questionnaire and patient chart examination was conducted in individuals with congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency. The questionnaire was administered to 122 adult participants (50 men and 72 women; median age, 35 years; range, 18–69 years) and the results revealed 257 adrenal crises in 4456 patient-years (incidence, 5·8 crises per 100 patient-years). Charts from 67 patients (32 men and 35 women; median age, 31 years; range, 20–66 years) were examined from the time of diagnosis to the last follow-up appointment, with regard to the frequency and cause of adrenal crisis. This revealed 106 adrenal crises in 2181 patient-years (incidence, 4·9 crises per 100 patient-years). Gastrointestinal infections (29%) and saltwasting (18%) were the main causes of adrenal crisis. There was no difference in the overall incidence of adrenal crisis in men and women. Most adrenal crises occurred during childhood, with more than 70% in the first 10 years of life and one-third in the first year of life; 20% were observed in adults (>18 years of age), demonstrating the significant lifetime risk of adrenal crisis in patients with 21-hydroxylase deficiency CAH. Data on the occurrence of adrenal crisis support epidemiological data, showing an increased risk of death from infections; in addition, some of the deaths attributed to vascular causes may be due to adrenal crisis.
All corticosteroids increase calcium excretion and reduce the rate of bone remodelling. Reduction in the rate of bone remodelling during therapy with supraphysiological doses of glucocorticoids is associated with a decrease in markers of both bone formation and resorption.
No reductions in bone mineral density (BMD) were observed in patients receiving long-term replacement therapy for AI in two small studies (n = 15, PAI or SAI and n = 25, PAI). In another two studies conducted in individuals with PAI (n = 30 and n = 33), BMD was reduced only in postmenopausal women. Another small study reported reduced BMD in women with PAI but not in men with PAI (n = 14). However, a further study conducted in 29 patients with PAI found that BMD was inversely correlated with the cumulative glucocorticoid dose per kilogram body weight and with the duration of treatment, although reduced BMD values were observed mainly in men with low levels of testosterone. In a study by Koetz et al., patients with PAI who received lower daily doses of hydrocortisone than normally used as replacement therapy (mean, 12·0 ± 2·7 mg/m) had BMD values within the normal reference range, whereas those receiving prednisone had a lower BMD. Zelissen et al. found that BMD decreased linearly with increasing doses of hydrocortisone per kilogram body weight in men. In contrast, reduced spinal BMD (found in 71·4% of patients) did not correlate with the weight-adjusted hydrocortisone dose in a retrospective cohort study, which involved 28 participants with PAI for whom there were available dual-energy X-ray absorptiometry (DXA) data. In a two-cohort study (n = 292), both male and female patients with PAI on long-term glucocorticoid replacement therapy were shown to have reduced BMD compared with a control population, although the patients in this study were receiving, on average, more than 30 mg hydrocortisone daily. This study also showed that BMD decreased with increasing daily glucocorticoid dose (calculated as hydrocortisone equivalent/kg body weight) in one of the two cohorts (from Norway).
The risk of hip fracture was analysed in a population-based cohort study of 3219 patients who had received a diagnosis of PAI when they were at least 30 years of age and who had not had a previous hip fracture. Compared with age- and sex-matched controls, the overall hazard ratio for hip fracture in these individuals was 1·8 (95% CI, 1·6–2·1; P < 0·001), with the highest risk in women who were aged 50 years or younger when PAI was diagnosed (odds ratio, 2·7; 95% CI, 1·6–4·5). In patients with PAI, the absolute risk of hip fracture was 784/100 000 person-years, corresponding to an excess risk of 350/100 000 person-years. The same study also found that the relative risk of hip fracture was highest in the first year after diagnosis of PAI, indicating that the increased fracture risk may not be attributable only to long-term high-dose glucocorticoid replacement therapy.
In patients with SAI, reduced BMD has been reported in those with adult-onset GH deficiency in conjunction with deficiency of other pituitary hormones. A cross-sectional study of 365 patients with hypopituitarism showed that glucocorticoid replacement is independently associated with reduced BMD and an increased frequency of osteopenia in women with ACTH insufficiency. An earlier study, however, of 23 patients with PAI and 23 with SAI concluded that BMD in patients with AI is generally normal and does not require routine follow-up. Fracture frequency in patients with hypopituitarism has been shown to be increased in some studies, particularly in women with childhood-onset disease, but this has not been associated with the presence or absence of AI.
In summary, it appears from the available data that BMD may be reduced in patients with AI, particularly in patients with SAI and those receiving supraphysiological doses of glucocorticoids over an extended period of time. The limited data available suggest that fracture frequency is increased in patients with PAI and those with hypopituitarism, but its association with AI and glucocorticoid replacement has not been established. A recent study suggests that BMD is normal in patients with more physiological glucocorticoid replacement therapy. It will be interesting to undertake long-term follow up of changes in BMD to determine whether lower and more physiological glucocorticoid replacement has any clinically deleterious effect on bone physiology.
Glucocorticoids are known to exert cardiovascular effects, probably through a permissive enhancement of other vasoactive agents, such as angiotensin II and catecholamines. High doses of hydrocortisone can cause hypertension, salt and water retention, and increased excretion of potassium owing to saturation of the 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD-2) isoenzyme in the kidney, allowing access of hydrocortisone to the mineralocorticoid receptor. Changes in body weight and blood pressure due to high doses of hydrocortisone may contribute to the increased number of premature cardiovascular deaths in patients with AI.
Glucocorticoid excess has been suggested to induce cardiovascular risk factors and to promote progression of atheromatous vascular disease. In a study of 68 781 individuals taking high-dose glucocorticoids for a variety of conditions and 82 202 not receiving glucocorticoids, the relative risk for a cardiovascular event after adjustment for known covariates was 2·56 (95% CI, 2·18–2·99) in those receiving glucocorticoids. It should be emphasized, however, that the majority of glucocorticoid use is for inflammatory diseases that may themselves be cardiovascular risk factors. The prevalence of atherosclerosis was also significantly increased in patients with hypopituitarism (n = 34) compared with that of matched controls in a cross-sectional study, although there is no evidence that this was related to glucocorticoid use as opposed to untreated GH deficiency.
Glucocorticoids are known to affect glucose metabolism. They increase hepatic gluconeogenesis, inhibit peripheral glucose utilization and promote hepatic glycogen synthesis by making substrates available for an acute stress response. Elevated levels of glucocorticoids lead to protein catabolism and lipolysis, providing amino acids and free fatty acids as substrates for hepatic gluconeogenesis. Animal data have shown that high afternoon and evening levels of cortisol lead to a worsening of lipid and glucose homeostasis.
Patients with hypopituitarism are known to be at increased risk of developing signs and symptoms of the metabolic syndrome (a combination of cardiovascular risk factors, including hypertension, dyslipidaemia, central obesity and insulin insensitivity). Filipsson et al. found that daily hydrocortisone-equivalent doses of 20 mg or more were associated with an unfavourable metabolic profile in a large cohort of adults with hypopituitarism; increased serum levels of total cholesterol, triglycerides and low-density lipoprotein cholesterol were associated with higher daily doses of glucocorticoid.
Abnormal glucose tolerance is more common in patients receiving long-term glucocorticoid replacement therapy than in healthy individuals. In addition, type 1 diabetes mellitus is prevalent in patients with PAI owing to its common APS type 2 aetiology. Given that cortisol strongly influences glucose metabolism, insulin requirements are often higher in patients with PAI and type 1 diabetes mellitus, particularly in the afternoon, than in those with type 1 diabetes mellitus alone.
Standard therapies for type 1 diabetes mellitus and AI do not precisely mimic normal physiology. There is therefore a risk that either condition may decompensate, with a vicious cycle of clinical deterioration. Conventional oral hydrocortisone replacement is associated with supraphysiological glucocorticoid exposure in the early evening, which may have asymptomatic but adverse metabolic effects. Insufficient glucocorticoid therapy, however, may result in acute decompensation, and it is useful to consider the impact of primary derangements in each condition on clinical status.
In a study of European patients with GH deficiency receiving hydrocortisone (n = 1168), cortisone acetate (n = 487) or prednisone/dexamethasone (n = 52), a positive relationship was found between the glucocorticoid replacement dose and body mass index after adjustment for sex and age. Patients treated with glucocorticoids in the same study had a greater waist circumference and waist:hip ratio than those without ACTH insufficiency (n = 717).
Reductions in mean total body fat of 7·1 kg over 6 months were found in patients with SAI after reducing the dose of hydrocortisone by 50% (to 10–15 mg daily), although no changes in body weight were demonstrated in two previous studies after a dose reduction from 30 to 15 mg daily, or from 30 to 20 mg daily. In a recent open-label randomized study, patients received the same daily dose of once-daily oral dual-release hydrocortisone and thrice-daily immediate-release hydrocortisone. Although the daily doses were the same in the two groups, the pharmacokinetic properties of once-daily hydrocortisone meant that there was, on average, 20% less bioavailable hydrocortisone in patients receiving once-daily treatment. After 12 weeks of treatment, the mean reduction in body weight was significantly greater in the patients on once-daily treatment than in those on thrice-daily treatment. It is not clear whether this reduction in weight was due to the 20% reduced bioavailability of dual-release hydrocortisone compared with immediate-release hydrocortisone or whether it was due to the markedly different exposure pattern produced by the dual-release formulation.
Current hydrocortisone replacement regimens fail to normalize health-related quality of life (QoL) in patients with AI. Perceived health status and vitality were reduced in individuals with PAI (n = 79) in Norway who were receiving conventional replacement therapy. Analysis of 989 individuals with PAI or SAI from Denmark showed a significantly higher rate of affective disorders (2·68 times greater) and depressive disorders (2·12 times greater) than a control group of patients with osteoarthritis. Subjective health status was significantly impaired (P < 0·001 compared with sex- and age-matched controls) in patients with PAI (n = 132) and SAI (n = 78) from Germany, irrespective of concomitant disease. In this study, occupational changes due to AI were reported by 40% of participants; 18·3% reported that they were out of work and receiving a disability pension compared with 4·1% of the general population. Similarly, Bleicken et al. found that a large number of patients were out of work (25%) and receiving a disability pension. These results are supported by a Norwegian study in which 30% of 425 individuals with PAI were receiving disability benefits. Additionally, patients with AI had an impaired subjective health status compared with controls, but those who received a correct diagnosis within 3 months of the onset of symptoms had a significantly better subjective health status than those for whom diagnosis was delayed.
A worldwide survey of 1245 patients with AI (84% PAI) reported that 64% of individuals had a compromised subjective health status, necessitating changes to physical activity or social, work or family life, 40% reported absence from work or school in the past 3 months, 76% were concerned about the long-term side effects of therapy and 38% had been hospitalized in the past year. The findings from this large study are supported by a recent study of 54 patients with PAI on stable treatment compared with 54 healthy matched controls (P < 0·01). The patients with PAI reported psychological morbidity and impaired QoL compared with controls P < 0·01). There was also a strong relationship between the mean daily hydrocortisone dose and both reduced QoL and the prevalence of maladaptive personality traits (P < 0·05). The impaired QoL and maladaptive personality traits were associated with higher doses of hydrocortisone. The association between dose of hydrocortisone and QoL has also been demonstrated recently in a large cohort of patients with SAI. In 1750 hypopituitary patients with ACTH deficiency, QoL decreased with increasing doses of glucocorticoid; the highest QoL was achieved with hydrocortisone equivalent doses of <10 mg and the worst QoL with doses of more than 25 mg (P = 0·006 for trend). The apparent negative relationship between QoL and hydrocortisone dose may, however, have been due to an empirical increment in hydrocortisone dose in patients with non-specific symptoms.
It thus appears that conventional glucocorticoid replacement therapy does not restore QoL in patients with AI, and that higher replacement doses are associated with a negative effect on QoL.
Clinical Outcomes: Morbidity
Intercurrent Illness and Adrenal Crisis
Adrenal crisis, which is a factor in the increased mortality of individuals with AI, has been estimated to occur at an incidence of approximately 6·3 per 100 patient-years, based on a study of 883 patients in Germany. This number is higher than the incidence of approximately 3·3 per 100 patient-years reported previously in a smaller German study of 53 individuals with AI, but is broadly consistent with a frequency of 8% per year reported in the UK. In a retrospective study of 137 Japanese patients with AI, 40 (29%) had experienced at least one adrenal crisis. Some 7·4% of German patients with SAI and 10·6% of those with PAI had experienced an adrenal crisis on at least four or more occasions, indicating that there is a subgroup of patients at high risk.
Adrenal crises are mainly precipitated by gastrointestinal infection and fever (45%), but also by other stressful events, such as major pain, surgery, emotional distress, heat and pregnancy. Patients with comorbidities are at higher risk than those without. In those with SAI, female sex and diabetes insipidus were associated with an increased frequency of adrenal crisis. The incidence of adrenal crisis in the German study by Hahner et al. was not influenced by educational status, body mass index, glucocorticoid dose, dehydroepiandrosterone treatment, age at diagnosis, hypogonadism, hypothyroidism or GH deficiency. In a retrospective Japanese study, however, steroid replacement therapy for more than 4 years was the largest single risk factor for an adrenal crisis, followed by the presence of a mental disorder and sex steroid deficiency. In patients with SAI (n = 115), sex steroid deficiency was the greatest risk factor for an adrenal crisis. Individuals with untreated hypogonadism had a significantly higher relative risk (3·70; 95% CI, 1·71–7·98) than those without hypogonadism or with treated hypogonadism. Furthermore, among patients with hypogonadism aged younger than 50 years, individuals treated with sex hormones (5/51) experienced adrenal crisis less frequently than those not receiving sex steroid replacement (7/11; P = 0·0004). In a Canadian study of children with PAI (n = 102) and SAI (n = 34) conducted between 1973 and 2007, the use of stress glucocorticoid doses by parents increased significantly after 1997 (P < 0·05). The proportion of patients with signs or symptoms of acute adrenal failure therefore decreased to 27% after 1997 (P < 0·01). A major reason for hospitalizations in this study was suboptimal adherence to glucocorticoid stress protocols, indicating the need for continuous patient and carer education.
A detailed cross-sectional retrospective study by means of a disease-specific questionnaire and patient chart examination was conducted in individuals with congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency. The questionnaire was administered to 122 adult participants (50 men and 72 women; median age, 35 years; range, 18–69 years) and the results revealed 257 adrenal crises in 4456 patient-years (incidence, 5·8 crises per 100 patient-years). Charts from 67 patients (32 men and 35 women; median age, 31 years; range, 20–66 years) were examined from the time of diagnosis to the last follow-up appointment, with regard to the frequency and cause of adrenal crisis. This revealed 106 adrenal crises in 2181 patient-years (incidence, 4·9 crises per 100 patient-years). Gastrointestinal infections (29%) and saltwasting (18%) were the main causes of adrenal crisis. There was no difference in the overall incidence of adrenal crisis in men and women. Most adrenal crises occurred during childhood, with more than 70% in the first 10 years of life and one-third in the first year of life; 20% were observed in adults (>18 years of age), demonstrating the significant lifetime risk of adrenal crisis in patients with 21-hydroxylase deficiency CAH. Data on the occurrence of adrenal crisis support epidemiological data, showing an increased risk of death from infections; in addition, some of the deaths attributed to vascular causes may be due to adrenal crisis.
Bone Metabolism
All corticosteroids increase calcium excretion and reduce the rate of bone remodelling. Reduction in the rate of bone remodelling during therapy with supraphysiological doses of glucocorticoids is associated with a decrease in markers of both bone formation and resorption.
No reductions in bone mineral density (BMD) were observed in patients receiving long-term replacement therapy for AI in two small studies (n = 15, PAI or SAI and n = 25, PAI). In another two studies conducted in individuals with PAI (n = 30 and n = 33), BMD was reduced only in postmenopausal women. Another small study reported reduced BMD in women with PAI but not in men with PAI (n = 14). However, a further study conducted in 29 patients with PAI found that BMD was inversely correlated with the cumulative glucocorticoid dose per kilogram body weight and with the duration of treatment, although reduced BMD values were observed mainly in men with low levels of testosterone. In a study by Koetz et al., patients with PAI who received lower daily doses of hydrocortisone than normally used as replacement therapy (mean, 12·0 ± 2·7 mg/m) had BMD values within the normal reference range, whereas those receiving prednisone had a lower BMD. Zelissen et al. found that BMD decreased linearly with increasing doses of hydrocortisone per kilogram body weight in men. In contrast, reduced spinal BMD (found in 71·4% of patients) did not correlate with the weight-adjusted hydrocortisone dose in a retrospective cohort study, which involved 28 participants with PAI for whom there were available dual-energy X-ray absorptiometry (DXA) data. In a two-cohort study (n = 292), both male and female patients with PAI on long-term glucocorticoid replacement therapy were shown to have reduced BMD compared with a control population, although the patients in this study were receiving, on average, more than 30 mg hydrocortisone daily. This study also showed that BMD decreased with increasing daily glucocorticoid dose (calculated as hydrocortisone equivalent/kg body weight) in one of the two cohorts (from Norway).
The risk of hip fracture was analysed in a population-based cohort study of 3219 patients who had received a diagnosis of PAI when they were at least 30 years of age and who had not had a previous hip fracture. Compared with age- and sex-matched controls, the overall hazard ratio for hip fracture in these individuals was 1·8 (95% CI, 1·6–2·1; P < 0·001), with the highest risk in women who were aged 50 years or younger when PAI was diagnosed (odds ratio, 2·7; 95% CI, 1·6–4·5). In patients with PAI, the absolute risk of hip fracture was 784/100 000 person-years, corresponding to an excess risk of 350/100 000 person-years. The same study also found that the relative risk of hip fracture was highest in the first year after diagnosis of PAI, indicating that the increased fracture risk may not be attributable only to long-term high-dose glucocorticoid replacement therapy.
In patients with SAI, reduced BMD has been reported in those with adult-onset GH deficiency in conjunction with deficiency of other pituitary hormones. A cross-sectional study of 365 patients with hypopituitarism showed that glucocorticoid replacement is independently associated with reduced BMD and an increased frequency of osteopenia in women with ACTH insufficiency. An earlier study, however, of 23 patients with PAI and 23 with SAI concluded that BMD in patients with AI is generally normal and does not require routine follow-up. Fracture frequency in patients with hypopituitarism has been shown to be increased in some studies, particularly in women with childhood-onset disease, but this has not been associated with the presence or absence of AI.
In summary, it appears from the available data that BMD may be reduced in patients with AI, particularly in patients with SAI and those receiving supraphysiological doses of glucocorticoids over an extended period of time. The limited data available suggest that fracture frequency is increased in patients with PAI and those with hypopituitarism, but its association with AI and glucocorticoid replacement has not been established. A recent study suggests that BMD is normal in patients with more physiological glucocorticoid replacement therapy. It will be interesting to undertake long-term follow up of changes in BMD to determine whether lower and more physiological glucocorticoid replacement has any clinically deleterious effect on bone physiology.
Blood Pressure and Vascular Function
Glucocorticoids are known to exert cardiovascular effects, probably through a permissive enhancement of other vasoactive agents, such as angiotensin II and catecholamines. High doses of hydrocortisone can cause hypertension, salt and water retention, and increased excretion of potassium owing to saturation of the 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD-2) isoenzyme in the kidney, allowing access of hydrocortisone to the mineralocorticoid receptor. Changes in body weight and blood pressure due to high doses of hydrocortisone may contribute to the increased number of premature cardiovascular deaths in patients with AI.
Glucocorticoid excess has been suggested to induce cardiovascular risk factors and to promote progression of atheromatous vascular disease. In a study of 68 781 individuals taking high-dose glucocorticoids for a variety of conditions and 82 202 not receiving glucocorticoids, the relative risk for a cardiovascular event after adjustment for known covariates was 2·56 (95% CI, 2·18–2·99) in those receiving glucocorticoids. It should be emphasized, however, that the majority of glucocorticoid use is for inflammatory diseases that may themselves be cardiovascular risk factors. The prevalence of atherosclerosis was also significantly increased in patients with hypopituitarism (n = 34) compared with that of matched controls in a cross-sectional study, although there is no evidence that this was related to glucocorticoid use as opposed to untreated GH deficiency.
Glucose and Lipid Metabolism
Glucocorticoids are known to affect glucose metabolism. They increase hepatic gluconeogenesis, inhibit peripheral glucose utilization and promote hepatic glycogen synthesis by making substrates available for an acute stress response. Elevated levels of glucocorticoids lead to protein catabolism and lipolysis, providing amino acids and free fatty acids as substrates for hepatic gluconeogenesis. Animal data have shown that high afternoon and evening levels of cortisol lead to a worsening of lipid and glucose homeostasis.
Patients with hypopituitarism are known to be at increased risk of developing signs and symptoms of the metabolic syndrome (a combination of cardiovascular risk factors, including hypertension, dyslipidaemia, central obesity and insulin insensitivity). Filipsson et al. found that daily hydrocortisone-equivalent doses of 20 mg or more were associated with an unfavourable metabolic profile in a large cohort of adults with hypopituitarism; increased serum levels of total cholesterol, triglycerides and low-density lipoprotein cholesterol were associated with higher daily doses of glucocorticoid.
Abnormal glucose tolerance is more common in patients receiving long-term glucocorticoid replacement therapy than in healthy individuals. In addition, type 1 diabetes mellitus is prevalent in patients with PAI owing to its common APS type 2 aetiology. Given that cortisol strongly influences glucose metabolism, insulin requirements are often higher in patients with PAI and type 1 diabetes mellitus, particularly in the afternoon, than in those with type 1 diabetes mellitus alone.
Standard therapies for type 1 diabetes mellitus and AI do not precisely mimic normal physiology. There is therefore a risk that either condition may decompensate, with a vicious cycle of clinical deterioration. Conventional oral hydrocortisone replacement is associated with supraphysiological glucocorticoid exposure in the early evening, which may have asymptomatic but adverse metabolic effects. Insufficient glucocorticoid therapy, however, may result in acute decompensation, and it is useful to consider the impact of primary derangements in each condition on clinical status.
Body Composition
In a study of European patients with GH deficiency receiving hydrocortisone (n = 1168), cortisone acetate (n = 487) or prednisone/dexamethasone (n = 52), a positive relationship was found between the glucocorticoid replacement dose and body mass index after adjustment for sex and age. Patients treated with glucocorticoids in the same study had a greater waist circumference and waist:hip ratio than those without ACTH insufficiency (n = 717).
Reductions in mean total body fat of 7·1 kg over 6 months were found in patients with SAI after reducing the dose of hydrocortisone by 50% (to 10–15 mg daily), although no changes in body weight were demonstrated in two previous studies after a dose reduction from 30 to 15 mg daily, or from 30 to 20 mg daily. In a recent open-label randomized study, patients received the same daily dose of once-daily oral dual-release hydrocortisone and thrice-daily immediate-release hydrocortisone. Although the daily doses were the same in the two groups, the pharmacokinetic properties of once-daily hydrocortisone meant that there was, on average, 20% less bioavailable hydrocortisone in patients receiving once-daily treatment. After 12 weeks of treatment, the mean reduction in body weight was significantly greater in the patients on once-daily treatment than in those on thrice-daily treatment. It is not clear whether this reduction in weight was due to the 20% reduced bioavailability of dual-release hydrocortisone compared with immediate-release hydrocortisone or whether it was due to the markedly different exposure pattern produced by the dual-release formulation.
Quality of Life
Current hydrocortisone replacement regimens fail to normalize health-related quality of life (QoL) in patients with AI. Perceived health status and vitality were reduced in individuals with PAI (n = 79) in Norway who were receiving conventional replacement therapy. Analysis of 989 individuals with PAI or SAI from Denmark showed a significantly higher rate of affective disorders (2·68 times greater) and depressive disorders (2·12 times greater) than a control group of patients with osteoarthritis. Subjective health status was significantly impaired (P < 0·001 compared with sex- and age-matched controls) in patients with PAI (n = 132) and SAI (n = 78) from Germany, irrespective of concomitant disease. In this study, occupational changes due to AI were reported by 40% of participants; 18·3% reported that they were out of work and receiving a disability pension compared with 4·1% of the general population. Similarly, Bleicken et al. found that a large number of patients were out of work (25%) and receiving a disability pension. These results are supported by a Norwegian study in which 30% of 425 individuals with PAI were receiving disability benefits. Additionally, patients with AI had an impaired subjective health status compared with controls, but those who received a correct diagnosis within 3 months of the onset of symptoms had a significantly better subjective health status than those for whom diagnosis was delayed.
A worldwide survey of 1245 patients with AI (84% PAI) reported that 64% of individuals had a compromised subjective health status, necessitating changes to physical activity or social, work or family life, 40% reported absence from work or school in the past 3 months, 76% were concerned about the long-term side effects of therapy and 38% had been hospitalized in the past year. The findings from this large study are supported by a recent study of 54 patients with PAI on stable treatment compared with 54 healthy matched controls (P < 0·01). The patients with PAI reported psychological morbidity and impaired QoL compared with controls P < 0·01). There was also a strong relationship between the mean daily hydrocortisone dose and both reduced QoL and the prevalence of maladaptive personality traits (P < 0·05). The impaired QoL and maladaptive personality traits were associated with higher doses of hydrocortisone. The association between dose of hydrocortisone and QoL has also been demonstrated recently in a large cohort of patients with SAI. In 1750 hypopituitary patients with ACTH deficiency, QoL decreased with increasing doses of glucocorticoid; the highest QoL was achieved with hydrocortisone equivalent doses of <10 mg and the worst QoL with doses of more than 25 mg (P = 0·006 for trend). The apparent negative relationship between QoL and hydrocortisone dose may, however, have been due to an empirical increment in hydrocortisone dose in patients with non-specific symptoms.
It thus appears that conventional glucocorticoid replacement therapy does not restore QoL in patients with AI, and that higher replacement doses are associated with a negative effect on QoL.
Source...