Management of dyslipidemia in endocrine diseases

Synopsis:

Most endocrine disorders are chronic in nature, and thus even a minor effect to increase risk for cardiovascular disease can lead to a significant impact over prolonged duration. While robust therapies exist for many endocrine disorders (eg suppression of excess hormone amounts, or replacement of hormone deficiencies), the therapies do not perfectly restore normal physiology. Thus, individuals with endocrine disorders are at potential increased cardiovascular disease risk, and maximizing strategies to reduce that risk are needed. This article will review various endocrine conditions that can impact lipid levels and/or CVD risk.

Introduction/ History/ Definitions/ Background

The majority of lipid management guidelines recommend assessment of lipid levels and cardiovascular risk factors within an individual, then based on estimated cardiovascular disease (CVD) risk, suggest lipid lowering therapy and/or lipid goals¹. Overall, this approach works well. However, one of the limitations is that the cardiovascular risk factors typically included in risk calculators and guidelines are somewhat limited: smoking, hypertension, diabetes, age, sex, and race, are the typical factors. In some, but not all, guidelines additional factors such as family history of premature CVD events, chronic inflammatory disorders, chronic kidney disease, and premature menopause (age < 40 years) are often also noted as risk enhancing factors. However, these factors are fairly difficult to turn into a quantitative risk, and thus often not included in risk calculators, and may be overlooked by providers.

Most endocrine disorders are chronic in nature, and thus even a relatively minor effect to increase risk for CVD can lead to a significant impact when duration of exposure is considered. While robust therapies exist for many endocrine disorders (whether it be suppression of excess hormone amounts, or replacement of hormone deficiencies), the therapies do not perfectly restore normal physiology. Thus, individuals with endocrine disorders are at potential increased CVD risk, and maximizing strategies to reduce that risk are needed. The Endocrine Society recently published a guideline explicitly assessing the lipid profile and CVD risk and thus indications for lipid lowering therapy in individuals with endocrine diseases². This new guideline suggests that some endocrine diseases, including hyperthyroidism, hypothyroidism, Cushing disease, chronic glucocorticoid therapy with doses above physiological needs, obesity, postmenopausal hormone replacement therapy use, and premature menopause should be included in the list of risk enhancing factors. This article will review various endocrine conditions that can impact lipid levels and/or CVD risk.

Case study:

A 47 year old woman presents to clinic to establish care. She has recently moved to your region. She was diagnosed with autoimmune adrenal insufficiency at age 12, and takes glucocorticoid replacement therapy. She currently takes 20 mg every morning and 5 to 10 mg in the afternoon. She tells you that her previous endocrinologist kept attempting to reduce her hydrocortisone dose, but every time she decreased the dose she suffered severe fatigue and would go back up to this dose. She states she was trained in sick day management, and doubles the dose for 2–4 days each time she has nausea, vomiting, fever or severe fatigue. She estimates that she doubles her dose at least 5 times each year.

At age 22 she presented with Graves’ disease, was treated with radioactive iodine ablation, and subsequently became hypothyroid treated with levothyroxine. Her current dose is 150 mcg daily. She was diagnosed with endometriosis in her teens. She had one pregnancy complicated by pre-eclampsia at age 32, and had hysterectomy and oophorectomy at age 37. She took estrogen for 2–3 years afterwards, but then discontinued it as she didn’t see the need. She has struggled with her weight for years, but has recently put on about 10 lbs (4.5 kg) with the stress of the move, and her current weight is 216 lbs (98.2 kg). She reports her height as 5’6” (1.67 m); and her BMI is 34.9 kg/m2. She has long standing depression controlled with citalopram 40 mg daily. She has never smoked.

• Her current medications are:

• Hydrocortisone, 25–30 mg daily

• Levothyroxine 150 mcg daily

• Multivitamin, one daily

• Citalopram, 40 mg daily.

Her blood pressure is 132/74 mmHg. Her pulse is 82 beats per minute. Her examination is notable for generalized obesity with numerous pale striae on her abdomen. Her thyroid is not palpable. The remainder of the exam is normal.

• Fasting labs:

• Metabolic panel normal, with fasting glucose 97 mg/dl and eGFR > 60 ml/min/m2

• Hemogram: normal

• TSH 6.9 mIU/ml (normal range 0.5–5.0 mIU/ml)

• Free T4 0.6 ng/dl (normal range 0.8–1.8 ng/dl)

• Total cholesterol 234 mg/dl

• LDL cholesterol 148 mg/dl

• HDL cholesterol 32 mg/dl

• Triglycerides 268 mg/dl

• The ASCVD risk calculator estimates her 10 year risk as 2.8%.

Clinical Questions:

Does the ASCVD risk calculator 10 year risk estimate reassure you?

Are there any other factors you need to consider?

Discussion

On the surface, this patient appears to have several endocrine issues that need to be addressed, but cardiovascular risk does not appear to be one. She is obese, has premature menopause, inadequately treated hypothyroidism, and over-treated adrenal insufficiency. Managing these chronic conditions is important, and would likely be addressed by most endocrinologists. However, the 10 year CVD risk estimate may seem reassuring, and it would not be surprising if most providers did not consider CVD risk reduction and lipid lowering therapy as high priorities. The recent guidelines published by the endocrine society suggest that her endocrine co-morbidities may impact her cardiovascular risk, and should be considered². In the following sections we will address these (and others) one by one.

Changes in lipids with hyper- and hypo-thyroidism

Altered thyroid hormone levels can have profound effects on lipoprotein metabolism and thus lipid levels. Thyroid hormone decreases intestinal absorption of cholesterol, increases biliary secretion leading to decreased hepatic cholesterol content and compensatory increase in LDL-receptors, increases HMGCoA reductase activity, and increases enzymes involved in LDL metabolism including lipoprotein lipase, hepatic lipase, cholesteryl ester transfer protein and lecithin cholesterol acyltransferase^3–7. Thus, hyperthyroidism tends to lead to accelerated metabolism of lipoprotein particles such that total cholesterol and LDL-c are often low, while TG and HDL-c are not usually affected. Conversely, hypothyroidism tends to lead to elevated total cholesterol, LDL-c, HDL-c and TG. The Endocrine Society meta-analysis⁸ found that treatment of hyperthyroidism (with surgery, radioactive iodine or anti-thyroid medication) led to significant increases in total cholesterol and LDL-c and HDL-c with restoration of euthyroid state; however, this was only true in overt hyperthyroidism and not in subclinical hyperthyroidism. Treatment of overt hypothyroidism correspondingly led to significant decreases of TC, LDL-c, HDL-c and TG⁸. Indeed, levothyroxine has been studied as a therapy for elevated lipid levels⁹ (Table 1).

Table 1.

Change in Lipid Parameters after Treatment of Thyroid Disease

Mean % change with treatment	TC	LDL-c	HDL-c	TG
Overt hypothyroidism	↓22%	↓24%	↓7%	↓18%
Subclinical hypothyroidism	↓5%	↓8%	0%	↓4%
Overt hyperthyroidism	↑28%	↑35%	↑12%	↑7%
Subclinical hyperthyroidism	↑5%	↑6%	0%	↓30%

Data adapted from a meta-analysis of n=3–72 studies per parameter, with total patients from 104–4588 per parameter⁸.

Whether hyper- or hypothyroidism directly influence CVD (rather than indirectly, via dyslipidemia) is unknown. Epidemiologic studies suggest that coronary disease prevalence is increased in hypothyroidism compared with euthyroid controls^{10, 11}. Hypothyroidism can affect cardiac contractility and exacerbate angina; however, it is not clear if there are direct pathophysiological changes related to thyroid hormone abnormalities, or if the association is mediated through dyslipidemia.

Considerations for clinical management

The Endocrine Society guidelines recommend both screening for hypothyroidism as a cause of dyslipidemia and deferring treatment decisions until after restoration of euthyroid status when a patient has either hyper- or hypothyroidism. These recommendations are prudent and will help avoid unnecessary lipid-lowering therapy in the setting of hypothyroidism (if the repeated lipid panel shows resolution of hyperlipidemia), and avoid a missed therapeutic opportunity in the setting of hyperthyroidism if the lipid panel is falsely reassuring when the patient is hyperthyroid.

Thus, for the case vignette presented, it would be prudent to adjust her levothyroxine dose, and repeat the lipid panel 6–8 weeks later when she would be expected to be euthyroid. Although her freeT4 is only marginally low, restoration of euthyroid state could lead to some improvement in her dyslipidemia.

Changes in lipids with glucocorticoid excess or therapy

Elevated glucocorticoid levels (whether endogenous such as in Cushing syndrome, or exogenous, as in the case vignette) can lead to elevations in total cholesterol, LDL-c and TG. Glucocorticoids stimulate preadipocyte differentiation and increased adipose tissue, especially visceral adipose, and also promote fatty acid and cholesterol synthesis in the liver, leading to hepatic steatosis¹². Chronic elevations in glucocorticoids increase metabolic syndrome prevalence, with associated hypertension, insulin resistance and prothrombotic state, all of which contribute to increased CVD risk. Patients cured of Cushing syndrome typically experience improvements in dyslipidemia, as well reduced obesity, hypertension and insulin resistance¹³. For patients using exogenous steroids the literature is conflicted, but at least one study indicates a glucocorticoid dose dependent increase in total cholesterol, LDL-c and TG levels^{14, 15}. Furthermore, several studies suggest increased CVD in patients using exogenous glucocorticoids, particularly for those with iatrogenic hypercortisolism and Cushing syndrome^16–19.

Considerations for clinical management

The Endocrine Society guidelines address lipid screening and management in settings of excess endogenous or exogenous glucocorticoids. The guidelines recommend screening lipid levels in adults with Cushing Syndrome and those on chronic glucocorticoid therapy above standard physiologic replacement doses. The guidelines go on to suggest statin therapy in addition to lifestyle modification in adults with persistent Cushing syndrome to reduce CVD risk, regardless of CV risk score. There is not any evidence to guide recommendation of lipid lowering therapies (statins) in individuals with exogenous glucocorticoids above physiological doses, but certainly, there is accumulating evidence that supra-physiological doses of glucocorticoids convey health risks. Thus, at a minimum it is prudent to recommend decreasing glucocorticoid doses which may help decrease lipid levels and CVD risk.

Thus, for the case vignette presented, it would be prudent to taper her hydrocortisone dose down, review sick day rules to minimize excessive dosing, and repeat the lipid panel when stable on a lower dose. Collectively, her use of glucocorticoid therapy should be considered along with other risk factors when making a decision about lipid lowering therapy.

Changes in lipids with obesity

Obesity prevalence is high and rising, thus a common concern for health care providers. In particular, when obesity is mainly central it often exists as part of the metabolic syndrome (elevated TG, reduced HDL, increased waist circumference, hyperglycemia, increased BP) where dyslipidemia is highly prevalent, and robust evidence indicates an increased risk for CVD. Even without metabolic syndrome, obesity is associated with elevations in TG and decreases in HDL-c, and while LDL-c may not be elevated, the particles are often small and dense, which are thought to be more atherogenic^20–22. In addition, delayed lipoprotein metabolism leads to prolonged and exacerbated postprandial hyperglycemia^23–25. Furthermore, elevated BMI, or increased waist circumference or waist/hip ratio are predictors of CVD mortality^{24, 26–28}.

Considerations for clinical management

Weight loss, whether induced by caloric restriction, medications, or surgery leads to improvements in the lipid profile. As little as 5% body weight reduction can lead to improvements in a number of comorbidities of obesity²⁹. There is a corresponding improvement in lipids with increased weight loss, so that 3 kg weight loss is associated with a TG decrease of 15 mg/dl (0.17 mmol/L), but weight loss of 5–8 kg is associated with decreases in LDL-c of −5 mg/dl (−0.13 mmol/L) and increases in HDL-c of 2–3 mg/dl (0.5–0.8 mmol/L)³⁰. A meta-analysis showed that the most consistent and favorable effect of weight loss in a lowering of TG³¹. When patients are actively undergoing weight loss a number of changes in lipid levels can occur, including paradoxical drops in HDL-c during active weight loss^{32, 33}; thus, the Endocrine Society recommends reassessment of lipids after weight loss once weight has stabilized.

Lipid lowering therapy with statins have been clearly shown to decrease CVD events in patients with and without obesity. Despite the fact that LDL-c is not always high in obesity, statins (which target LDL-c) are highly efficacious in lowering CVD risk^{34, 35}. Conversely, fibrates (which target TG) have not been consistently shown to decrease CVD risk. Thus, if there are indications for lipid lowering therapy then most guidelines will recommend statins as the first line therapy. Of note, when TG are elevated to the extent that pancreatitis is a risk there is uniform agreement of using fibrates to lower TG.

Thus, for the case vignette presented, it would be appropriate to both screen her for metabolic syndrome (based on information provided she does have metabolic syndrome as she has high TG, low HDL-c, and high systolic BP; even without knowing her fasting glucose or waist circumference), and counsel her on weight loss strategies. The Endocrine Society recommends use of a risk calculator to assess 10-year CVD risk, and initiation of lipid lowering therapy if indicated by the calculator. In this vignette the 10-year CVD risk was 2.8%, and thus she does not clearly meet recommendations for lipid lowering therapy. However, as discussed throughout this manuscript, the risk calculator does not adjust for potential CVD impact of her combined metabolic disorders, and may underestimate her risk.

Changes in lipids with menopause

Although the changes in lipid levels from pre-menopause to post-menopause are fairly small, epidemiological evidence consistently indicates an increase in CVD risk. Studies have yielded variable results, but in general the lipid panel shows decreases in HDL-c, increases in total cholesterol, and a shift in LDL particle size towards a small, dense phenotype; collectively these are pro-atherogenic changes. The mechanisms behind shifts in lipid levels likely relate to the decrease in estrogen. Estrogen affects VLDL synthesis, insulin sensitivity, LDL-receptors and PCSK9^{36, 37}. Estrogen replacement therapy leads to increases in HDL-c and decreases in LDL-c; however, progestins tend to decrease HDL-c^{38, 39}. In patients with an underlying predisposition to hypertriglyceridemia (due to genetics, or other risk factors such as diabetes, obesity, insulin resistance and others) estrogen therapy can cause significant increases in TG levels and increase risk for pancreatitis. However, while this is not uncommonly seen in specific individuals, population based studies have not found a significant increase of pancreatitis with estrogen therapy⁴⁰.

Considerations for clinical management

Several decades ago the standard of care was to recommend hormone replacement therapy (estrogen, with or without progestins for uterine protection) with the expectation this would decrease CVD risk in postmenopausal women. However, several large randomized controlled trials of estrogen (+/− progestins) found increased CVD events, thought to be due to increased thrombosis, especially when introduced late after onset of menopause (> 60 yrs, or > 10 yrs since last menstrual period)^{41, 42}. Thus, most guidelines recommend caution for initiation of estrogen, especially in older women.

When menopause occurs early, regardless of whether it is spontaneous or surgically induced, CVD risk is increased: a younger age at menopause appears to be an independent risk factor for CVD^43–46. The estimate is that the risk of ASCVD is 1.5-fold higher in women with menopause < 40 years, and 1.3-fold higher in women with menopause occurring at age 40–44 years, compared to women who entered menopause at age 50–51 years⁴⁶. Several recent guidelines now recognize early menopause as a CV risk factor. Statin therapy in postmenopausal women has been shown to lower CVD risk, for both women using or not using hormone replacement therapy^{47, 48}. As further evidence, a meta-analysis of estrogen therapy in younger women reported a significant reduction in CVD⁴⁹. Thus, the Endocrine Society recommends use of statin therapy in postmenopausal women with dyslipidemia, or those on hormone therapy with other risk factors for CVD, and encourages consideration of CVD risk in patients who enter menopause early.

Thus, for the case vignette presented, there are several factors to consider. Her hypothyroidism is undertreated, her glucocorticoid dose is supra-physiological, she has obesity, and she has early menopause. Although the 10-year risk calculator estimates her risk as fairly low (2.8%) none of these factors are included in that calculator. Collectively, this patient has a number of endocrine comorbidities that each confer increased CVD risk, and initiation of statin therapy would be expected to help reduce her risk. However, if the provider does not consider each of these co-morbidities, and therapeutic opportunity may be missed.

Other lipid-influencing endocrine diagnoses.

Beyond the topics raised by the case vignette, the Endocrine Society addressed several other endocrine diagnoses that affect lipid levels and CVD risk. These are summarized below.

Changes in lipids with polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is characterized by insulin resistance and often a similar lipid profile to that seen in metabolic syndrome: increased TG, low HDL-c and normal or increased LDL-c^{50, 51}. In addition, increased levels of Lp(a) may be seen, particularly in nonobese women with PCOS^52–54. The dyslipidemia exists throughout the reproductive years, tends to be worse in anovulatory women^{53, 55}, and may be further exacerbated after the onset of menopause. Thus, although the dyslipidemia may be relatively mild, it may be of long duration. Despite the dyslipidemia, it is not clear if there is increased ASCVD risk in PCOS per se, or if the risk is explained by the obesity and metabolic syndrome components⁵⁶.

Considerations for clinical management

Unlike the impact on dyslipidemia in obesity and metabolic syndrome, weight loss achieved via lifestyle therapy (diet and exercise) in PCOS appears to have minimal effects on the dyslipidemia, although improvements in body composition, insulin resistance and ovulation are seen^{57, 58}. Other common therapies used in PCOS include metformin and oral contraceptives. Metformin monotherapy appears to have minimal effects on the dyslipidemia in PCOS, though metformin in combination with other medications such as statins, thiazolidinediones, oral contraceptives, or inositol, can induce improvements^59–61. However, treatment with oral contraceptives confers the risk of further elevations in TG in susceptible women, although beneficial effects of estrogen on LDL-c and HDL-c can be seen⁶².

There have been several trials determining the effect of lipid lowering medications (mainly statins) in PCOS; collectively, statins are efficacious at lipid-lowering in women with PCOS^63–67, but cardiovascular outcomes, and reproductive outcomes remain unclear.

The Endocrine Society recommends obtaining a lipid panel in all women with PCOS, but using lipid lowering therapies only as indicated for lipid lowering, and not for the treatment of hyperandrogenism or infertility.

Changes in lipids with male hypogonadism and testosterone therapy

Men with hypogonadism tend to have elevations in LDL-c and TG with lower levels of HDL-c⁶⁸; however, repletion of testosterone tends to have minimal effects on lipid levels^{69, 70}. Thus, use of testosterone therapy is not recommended as a treatment for dyslipidemia. Furthermore, illicit use of testosterone or other androgens is not uncommon, and elevated (supraphysiologic) androgen levels can dramatically suppress HDL-c, as well as increase apoB and decrease Lp(a)^{71, 72}.

Considerations for clinical management

Although hypogonadism is associated with increased CVD risk⁷³, there are multiple factors involved, including insulin resistance, obesity, increased prevalence of metabolic syndrome, and increased free fatty acids, in addition to the dyslipidemia. However, it remains controversial if testosterone therapy alters CVD risk, with some benefit perhaps seen in appropriate dosing of certain sub-populations, but no global benefit⁷⁴.

The Endocrine Society recommends using testosterone therapy for hypogonadism symptoms, but not as a treatment of dyslipidemia or CVD risk. Moreover, in patients with very low HDL-c but without high TG, androgen abuse should be considered as an etiology of the dyslipidemia.

Changes in lipids with gender affirming hormone therapy

The use of gender affirming hormone therapy is increasing, but there is still a paucity of long term outcome data to guide CVD recommendations. Numerous small studies and a meta-analysis reported an increase in TG and LDL-c and a drop in HDL-c with use of testosterone therapy for transmen^75–77. In transwomen treated with estrogen therapy most studies have reported an increase in TG with use of oral estrogens, but not with use of transdermal estrogens⁷⁷. Although an increase in HDL-c may be expected, this has not always been confirmed.

Considerations for clinical management

There is minimal data to guide use of lipid lowering therapy in the transgender population. The Endocrine Society recommends evaluation of CVD risk using the same guidelines as in cisgender adults.

Changes in lipids with growth hormone deficiency or growth hormone excess

In the setting of growth hormone deficiency, elevations in total cholesterol and LDL-c are common seen, while TG and HDL-c changes are variable^78–80. Conversely, in growth hormone excess, such as seen in acromegaly, increased TG is commonly observed with variable effects on cholesterol and LDL-c levels^{81, 82}. Growth hormone inhibits hepatic lipase and lipoprotein lipase activity^{81, 83} and can increase hepatic LDL-R expression and decrease PCSK9 expression^{84, 85}.

Considerations for clinical management

Hypopituitarism, of which growth hormone deficiency is the most common hormone abnormality, is associated with increased premature mortality, including increased risk of CVD^86–88. Growth hormone deficiency itself may affect the myocardial and endothelial tissue, cardiac performance and coronary calcification, though the direct mechanisms linking growth hormone deficiency and CVD are not fully understood. Although long term growth hormone therapy improves the dyslipidemia in growth hormone deficiency^89–91 it is not clear that growth hormone therapy can decrease mortality⁹².

Summary

Numerous endocrine disorders affect lipid levels, and thus may confer risk for CVD. As the current CVD risk calculators that are widely used to guide lipid lowering therapy decisions do not consider endocrine co-morbidities, CVD risk may be under-estimated, and thus lipid lowering therapy may be under-utilized in these populations. Consideration of the additional impact of endocrine diseases when assessing individuals for dyslipidemia, CVD risk, and lipid lowering therapy is urged.

Key points:

• treatment of dyslipidemia decreases cardiovascular risk

• many endocrine diseases are associated with dyslipidemia

• many endocrine diseases can affect cardiovascular risk, but are not considered when assessing risk

Clinics Care Points.

• Many endocrine diseases affect lipid levels and thus may confer CVD risk

• CVD risk calculators do not currently assess the impact of endocrine co-morbidities

• Clinicians should consider impact of endocrine co-morbidities on CVD risk when making treatment decisions regarding lipid lowering medications