Levels of Evidence for Clinical Decisions: Menopausal Hormone Therapy Revisited
Therapeutics Letter Issue 30 / Jun - Jul 1999
Confusion and misconceptions exist about evidence and levels of evidence. Consensus guidelines proliferate and purport to follow rules of evidence. Yet, there are no clear rules as to what to recommend when no Level I evidence (see Table) is available, or when evidence is conflicting.
|Large double blind RCTs, or meta-analyses of smaller RCTs, clinically relevant outcomes|
|Small RCTs, non-blinded RCTs, RCTs using valid surrogate markers|
|Non-randomised controlled studies, observational (cohort) studies, case-control studies, or cross-sectional studies|
|Opinion of expert committees or respected authorities|
To illustrate the impact that Level I evidence can have on a therapeutic decision we present the following case.
Mrs. AG is new to your practice. You are seeing her for the second time and she wants to know whether she should be taking hormone therapy as she read an article that stated that she should. She is a 67-year-old married Caucasian retired teacher, who was well until 2 years ago, when she suffered an acute inferior myocardial infarction. She recovered completely from that episode and is asymptomatic. She has been taking 80 mg of ASA and 40 mg of nadolol daily since her infarct. She has had no surgery. Her last menstrual period occurred at age 52 and she experienced only mild hot flashes for a year. She has a 30 pack-year history of smoking, but quit 7 years ago. She is a non-drinker. She has a negative family history for cardiovascular disease and osteoporosis. When you first saw her you did a complete physical exam and found no significant abnormal findings. She is 5ft 2in. and weighs 60 kg., BP 136/74 mm Hg., Pulse 66/min. Total cholesterol 4.8 mmol/L, LDL 2.9 mmol/L, HDL 1.5 mmol/L, Triglycerides 2.0 mmol/L.
Is there evidence available to assist us in this case?
Up until August 1998, there was no Level I evidence to answer this patient’s question. In August 1998 the first randomised controlled trial (RCT) relating to this question was published: the Heart and Estrogen/progestin Replacement Study, or (HERS) trial1, 2.
Objective: To determine if estrogen plus progestin therapy alters the risk for coronary heart disease events in postmenopausal women with established coronary disease.
Design: Randomised, double-blinded, placebo controlled secondary prevention trial in 20 US outpatient centers.
Intervention: Conjugated equine estrogens, 0.625 mg, plus medroxyprogesterone acetate, 2.5 mg, Hormone Therapy (HT) in 1 tablet daily (n =1380) or an identical appearing placebo (n = 1383).
Participants: Menopausal women (mean age 67 yrs), intact uterus, average follow-up 4.1 years.
Outcome measures: Primary: non-fatal myocardial infarction or CHD death. Secondary: total mortality, coronary revascularization, unstable angina, congestive heart failure, stroke, etc.
Results: No statistically significant difference between groups in primary outcome: 12.5% HT, 12.7% placebo, or any predefined secondary outcome (total mortality 9.5% HT, 8.9% placebo), was seen. The only statistically significant finding was more frequent adverse outcomes in the HT group: venous thromboembolic events HT 2.5%, placebo 0.9%; and gall bladder disease HT 6.1%, placebo 4.5%. There were no statistically significant differences in any cancer or any fracture categories. Interestingly, there was a statistically significant 11% net reduction in LDL and 10% net increase in HDL associated with the HT, yet there was no cardiovascular benefit. The sum of all adverse events in both groups (includes individuals with more than one event) was: HT=1298 events and placebo=1274 events.
Authors’ Conclusion: "We do not recommend starting this treatment for the purpose of secondary prevention of CHD."
Is there other RCT evidence available?
An extensive literature search found 6 additional RCTs, which reported some outcome data and lasted at least 1 year. These trials, 1 in men3, and 5 in women4 - 8, are smaller and except for the one in men were not designed to measure clinically relevant outcomes. However, the limited data from these trials corroborates the evidence from the HERS trial.
What should we advise her?
Based on this well designed large RCT with clinically relevant outcomes (Level I evidence), and the fact that she meets the inclusion criteria of this trial, we can advise her that HT is unlikely to prevent further CHD or other cardiovascular events. In addition HT is associated with a small increased risk of thromboembolic events and gall bladder disease. There is also no Level I evidence that attempting to lower her LDL cholesterol by other means would be beneficial.
What can be learned from this example?
Before the HERS trial the available evidence was from observational studies and RCTs looking at surrogate markers. Controversy was present, but many clinicians and guidelines were recommending HT for primary and secondary prevention of CHD and osteoporosis. The observational evidence (see Therapeutics Letter #14 & Therapeutics Letter #16) suggested a consistent reduction in risk of coronary heart disease with estrogens and combined HT (pooled relative risk 0.65 [0.59-0.71]). This magnitude of reduced risk (Level III evidence) is inconsistent with the HERS trial evidence and is likely due to 2 types of bias that can occur with observational studies: 1. Selection bias—women who choose HT are healthier at baseline than those who do not or 2. Compliance bias—people who comply with placebo have better outcomes than those who do not2. There was also RCT evidence that various regimens of HT had beneficial effects on lipid levels (possible surrogate markers for cardiovascular events, Level II evidence). The HERS trial has demonstrated that reduction in lipid levels with HT is not a valid surrogate for CHD outcomes.
What if our patient was merely at risk for CHD (primary prevention)?
In this case the risk of CHD events would be less and therefore the opportunity for benefit would be less. On the other hand the chance of adverse outcomes would most likely remain the same. It is unlikely that a drug would be beneficial in primary prevention if there is no demonstrable benefit for secondary prevention.
What if our patient had a previous osteoporotic fracture instead of an MI?
At the present time the evidence for the use of HT (or estrogen or progesterone alone) for osteoporosis is based on the same levels of evidence for CHD before the HERS trial, i.e. observational data and RCT evidence using bone mineral density as the surrogate. The same obervational data also suggests a small increased risk of breast cancer for women taking HT for 8 or more years (Letter # 14). Fortunately, further large RCTs studying HT are in progress.
Dosing and cost data on the drugs available for Menopausal Hormone Therapy can be found in the updated Table 1, Therapeutics Letter #14.
Is the controversy over prescribing HT resolved by the HERS trial? Probably not. Recent publications, while including the HERS trial in their list of references, are still advising doctors to encourage use of HT for prevention of CHD9 and osteoporosis10. It will likely take some time for the full implications of the Level I evidence from the HERS trial to be reflected in the literature and practice.
|This Therapeutics Letter contains an assessment and synthesis of published (and whenever possible peer-reviewed) publications up to July, 1999. We attempt to maintain the accuracy of the information contained in the Therapeutics Letter by extensive literature searches and verification by both the authors and the editorial board. In addition this Therapeutics Letter was submitted for review to 75 experts and primary care physicians in order to correct any identified shortcomings or inaccuracies and to ensure that the information is concise and relevant to clinicians.|
- Hulley S, Grady D, Bush T et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605-613.
- Editorial: Petitti DB, Hormone therapy and heart disease prevention: Experimentation trumps observation. JAMA 1998;280:650-52.
- The Coronary Drug Project Research Group. The Coronary Drug Project: findings leading to discontinuation of the 2.5 mg / day estrogen group. JAMA.1973;226:652-657.
- The Writing Group for the PEPI trial. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. JAMA 1995;273:199-208.
- Komulainen M, Kröger H, Tuppurainen MT, Heikkinen A-M, et al. Prevention of femoral and lumbar bone loss with hormone replacement therapy and vitamin D3 in early postmenopausal women: A population based 5-year randomized trial. J Clin Endocrin Metab 1999;84;546-552.
- Lufkin EG, Wahner HW, O'Fallon WM, et al. Treatment of postmenopausal osteoporosis with transdermal estrogen. Ann Int Med 1992;117:1-9.
- Hall G, Pripp U, Schenck-Gustafsson K and Landgren B-M. Longterm effects of hormone replacement therapy on symptoms of angina pectoris, quality of life and compliance in women with coronary artery disease. Maturitas1998;28:235-242.
- Speroff L, Rowan J, Symons J, et al. The comparative effect on bone density, endometrium, and lipids of continuous hormones as replacement therapy. (CHART Study). JAMA 1996;276:1397-1403.
- Wise MR, Stewart DE, Liu P Abramson BL. Use of hormone replacement therapy among cardiac patients at a Canadian academic centre. CMAJ 1999; 161:33-36.(Editorial Grover SA, Estrogen replacement for women with cardiovascular disease: Why don't physicians and patients follow the guidelines? CMAJ 1999;161:42-43.).
- Recker RR, Davies M, Dowd RM, Heaney RP. The effect of low-dose continuous estrogen and progesterone therapy with calcium and vitamin D on bone in elderly women. A randomized controlled trial. Ann Intern Med 1999;130:897-904.