Therapeutic Controversies

LIPOPROTEIN DISORDERS IN WOMEN: WHICH WOMEN ARE THE BEST CANDIDATES FOR HORMONE REPLACEMENT THERAPY? Susyn L Plushner

To review the data examining hormone replacement therapy (HRT) in the treatment of lipoprotein disorders in women.

OBJECTIVE:

DATA SOURCE: A MEDLINE search (1975-1995) of the Englishlanguage literature wa~ performed to identify pertinent primary literature and review articles. Articles were also identified through bibliographies of selected articles.

Controlled and uncontrolled studies evaluating the effects of lipoprotein concentrations on coronary risk and the effects of estrogen and HRT on coronary heart disease and lipoprotein concentrations in women were evaluated. Trials pertaining to adverse effects of therapy were also examined. Emphasis was placed on recent clinical trials.

DATA EXTRACOON:

The National Cholesterol Education Program's (NCEP's) 1993 report recommends estrogen replacement therapy as a treatment option in postmenopausal women with hyperlipidemia. Recent trials suggesting that triglycerides and high-density lipoproteins are more closely related to coronary risks in women necessitate an improved understanding of estrogen and progestin's effects on lipoprotein concentrations. A recent trial has clarified the lipoprotein effects of HRT in women with normal lipid concentrations and suggests that beneficial effects on low-density lipoproteins are maintained, although progestins attenuate beneficial changes in high-density lipoproteins and triglyceride elevations persist. The few trials evaluating estrogen use in women with hyperlipidemia suggest a beneficial effect as well.

DATA SYNTHESIS:

In the absence of contraindications, postmenopausal women with hyperlipidemia should be offered estrogen replacement therapy as conjugated equine estrogen 0.625 mg/d. Pending further information, NCEP's recommendations should be followed regarding goals of therapy.

CONCLUSIONS:

Ann PhaT71Ulcother 1996;30:98-107.

in postmenopausal women is cardiovascular disease, specifically coronary heart disease (CHD)Y The incidence of CHD increases sharply with age and is rare in women before the age of 55 years. 3 Epidemiologic evidence indicates that estrogen replacement therapy (ERT) in postmenopausal women reduces

THE LEADING CAUSE OF DEATH

Susyn L Plushner PharmD RCPS. Clinical Pharmacy Specialist. Internal Medicine. Franklin Medical Center. Kaiser Permanente, 2045 Franklin St, Denver, CO 80205, FAX 303/861-3664 Reprints: Susyn L Plushner PharmD RCPS

98 • The Annals ofPhaf71U1cotherapy •

the risk of CHD by approximately 50%,4 with greater reductions apparent in women with established coronary disease (80-90%).5 The importance ofCHD and ERT in this population is reflected in the National Cholesterol Education Program's (NCEP's) 1993 report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel 11).6 The new guidelines add female sex and age older than 55 years, or premature menopause in women not receiving ERT, as coronary risk factors. Additionally, NCEP added estrogen replacement as a treatment option in postmenopausal women with hyperlipidemia. In individualizing therapy, however, many questions remain unanswered regarding the optimal dosage, type, and duration of estrogen therapy. Little information is available regarding concurrent progestin use and subsequent effects on lipoproteins and coronary risks. Finally, the specific lipoprotein targets in treating women are unclear, as are the patient-specific factors that must be considered prior to treatment initiation. To optimally resolve these issues would require large, controlled clinical trials in women with hyperlipidemia where hormone replacement therapy (HRT) is compared to either placebo or conventional treatment. While such trials have yet to be completed, several large, controlled clinical studies examining the effects of hormone therapy in healthy women (Women's Health Initiative) and women with established heart disease (Hormone Estrogen/Progestin Replacement Study) are currently under way. The results, however, will not be available for at least 2-9 years. Treatment decisions may now be made by examining how lipoprotein concentrations relate to CHD risk in women by identifying which women obtain cardiovascular benefits from ERT and by evaluating the specific lipoprotein changes involved with these benefits. Additionally, the relative risks and benefits of ERT as well as HRT, which is defined as combination estrogen and progestin regimens, must be evaluated. The intent of this article is to outline the current controversies involved in treating hyperlipidemia in women, including effects of treatment on CHD and ultimate treatment goals, and to review recent trials pertaining to ERT and HRT to assist practitioners in formulating an

1997 January, Volume 31

optimal approach to treatment. Only oral routes of administration of HRT are included in this article as there is no information regarding other routes' association with CHD. Nonlipid mechanisms contributing to the cardioprotective effects of estrogens and lipoprotein changes not routinely measured in clinical practice are not discussed. The reader is directed to a number of reviews and clinical trials for a detailed discussion of these mechanisms. 7•10

lipid Concentrations and Increased Coronary Heart Disease Risk in Women Several epidemiologic studies demonstrate an increased risk of CHD in women as lipoprotein concentrations increase. 1l ,12 Concurrent risks such as smoking, hypertension, and diabetes have been irregularly reported; whether results would have remained significant after controlling for these factors cannot be determined. Additionally, not reporting menopausal status (natural vs. surgical) may have affected study results as the incidence of CHD in surgically menopausal women may be greater than that in naturally menopausal women. 13 In the Donolo-Tel Aviv study,l1 the 20-year incidence of definite coronary events in 1481 women aged 35-64 years increased from 3% in women with total cholesterol concentrations less than 200 mgldL (n =306) to 10% in those with total cholesterol concentrations greater than 264 mgldL (n =362). At baseline, mean ± SD total cholesterol and high-density lipoprotein (HDL) concentrations ranged from 216 ± 38 to 252 ± 44 mgldL and 56 ± 15 to 57 ± 16 mgldL, respectively. HDL percentage, defined as the percent of serum total cholesterol bound to HDL, was a stronger predictor of CHD mortality regardless of total cholesterol concentrations. Women with a low HDL percentage «23%) had an increased risk for coronary events compared with those with a high HDL percentage (2::23%). Risk factors were not reported, nor was information provided regarding menopausal status or ERT. Similarly, the Framingham study12 reported an increased incidence of CHD in women aged 49 years and older with increased total cholesterol and reduced HDL concentrations. In this study, HDL was not an independent predictor of CHD, but jointly predicted CHD risk with total cholesterol concentrations in multivariate analysis controlling for systolic blood pressure, smoking, weight, and age. Results were not adjusted for estrogen use or smoking. Menopausal status was not addressed, but a later analysis of women aged 45-54 years in this cohort found no significant difference in risk for naturally versus surgically menopausal women; both groups showed an increased risk of coronary events.14 The relationship between mortality and lipoprotein concentrations was further explored in a subset of 1405 white women aged 50-69 years from the Lipid Research Clinics Follow-up Study. IS One of the major objectives of this study was to assess whether HDL was a more important determinant of cardiovascular disease risk in women than was total cholesterol. Baseline total cholesterol, HDL, lowdensity lipoprotein (LDL), and triglyceride concentrations, expressed as mean ± SD, were 246 ± 51, 61 ± 18, 162 ±

47, and 159 ± 229 mgldL, respectively. Over 70% of subjects had HDL concentrations exceeding 50 mg/dL and 80% of triglyceride concentrations were less than 200 mgldL. After an average of 14 years, 89 deaths attributable to CHD occurred. The specific incidences of mortality secondary to CHD alone could not be determined as cardiovascular disease was defined as deaths caused by heart disease and stroke. While an elevated risk of cardiovascular disease mortality was again noted in women with elevated total cholesterol concentrations, neither total cholesterol nor LDL was predictive of cardiovascular disease mortality. HDL concentrations less than 50 mgldL and triglyceride concentrations exceeding 400 mgldL were predictive of cardiovascular death. Cardiovascular disease death rates were also examined by categorical concentrations of total cholesterol, LDL, and triglycerides, after stratifying HDL into high (>50 mg/dL) and low «50 mg/dL) concentrations. The cardiovascular disease death rates (per 100 personyears) were significantly higher in women with low HDL and borderline or high total cholesterol concentrations than in women with these same total cholesterol concentrations and high HDL concentrations. At total cholesterol concentrations less than 200 mgldL, however, the cardiovascular disease mortality was higher in women with high HDL concentrations, a result unexplained by the authors. Cardiovascular disease mortality was also three- to fourfold higher at all concentrations of LDL in women with low versus high HDL concentrations. Finally, the combination of low HDL and high triglycerides carried approximately an eightfold increased risk of cardiovascular disease death over low HDL and normal triglycerides. With high HDL concentrations, there was a threefold increased risk of cardiovascular disease mortality with normal and high triglycerides. ls After adjusting for age, hypertension, diabetes mellitus, smoking, history of heart disease, and estrogen use, HDL and triglycerides persisted as independent predictors of cardiovascular disease mortality (Table I). LDL was only a significant predictor at concentrations of 130-159 mgldL. The authors concluded that HDL is a stronger predictor of cardiovascular disease risk in women aged 50-69 years than are other commonly measured lipid concentrations and that elevated triglycerides are also an independent risk factor, particularly in women with low HDL concentrations. As only 3% (n =41) of subjects had triglyceride concentrations of 400 mg/dL or greater, further studies are needed to confirm these findings. IS

Estrogens and Hypolipidemic Agents in Hyperlipidemic Women Few studies have examined the effects of any type of antilipemic therapy in women with hyperlipidemia. Therefore, the effects of HRT or conventional lipid-lowering treatment on lipoproteins or mortality in this population are virtually unknown. In the Scandinavian Simvastatin Survival Study l6 (n = 4444), a subset of 827 women with CHD, mean age 60.5 years, was analyzed for the effect of simvastatin 20-40 mg/d on total mortality and coronary

The Annals of Pharmacotherapy •

1997 January, Volume 31 •

99

Table 1. Adjusted Relative Risk for Cardiovascular Mortality by Lipid Category" LIPID (mg/dL)

Total cholesterol 160 and HDL >40 mg/dL, TC .j, IS, LDL .j, 28, HDL i 23, TG i 19d; for any LDL < 160 and HDL 232 mg/dL

IS'

CEE =conjugated equine estrogen; DUR =duration; EE =ethinyl estradiol; EV =estradiol valerate; HDL =high-density lipoprotein; LDL = low-density lipoprotein; MPA = medroxyprogesterone acetate; TC =total cholesterol; TG =trigylcerides. 8p < 0.0001. b p < 0.001. ·p400 mg/dL excluded

-2 -3 -6

-3 -10 -13

-2 9 3

-3 14 14

-6

-12

2

II

--4

-II

7

14

TC changes only significantly different from placebo (p < 0.00 I) in CEE-MPA groups; all treatment groups significantly different from placebo (p < 0.001) for HDL. LDL. and TG

CEE =conjugated equine estrogen; DUR =duration; EE =ethinyl estradiol; EV =estradiol valerate; HDL =high-density lipoprotein; LDL = low-density lipoprotein; MP =micronized progesterone; MPA = medroxyprogesterone acetate; N/A = not available; NC =no change; NR =not reported; PEPI =Postmenopausal EstrogenlProgestin Interventions; TC = total cholesterol; TG = trigylcerides. ·p