Postmenopausal estrogen replacement therapy is associated with adverse breast cancer prognostic indices.

Previous studies have reported that breast cancer patients who used estrogen replacement therapy (ERT) have more favorable tumor characteristics and decreased mortality compared with nonusers. However, these findings may be due partly to increased medical surveillance in ERT users and detection of early stage tumors. Postmenopausal women with biopsy-proven breast cancer (n = 108) were identified based on their participation in screening mammography. Based on self-administered questionnaires completed at the time of mammography, 29 of these were users of ERT. Tumor characteristics (histology size, nodal status, and estrogen receptor content) of ERT users were compared with those of nonusers. After adjusting for potentially confounding variables, the odds ratios (OR) describing the relationship between ERT use and the risk of invasive histopathology (OR = 1.35, 95% CI = 0.48, 3.75), positive nodes (OR = 2.43, 95% CI = 0.59, 10.10), size > or = 2.0 cm (OR = 2.34, CI = 0.66, 8.27), or negative estrogen receptor status (OR = 1.08, 95% CI = 0.18, 9.38) were > 1, although none reached statistical significance. When the subjects were separated into two prognostic groups based on the presence or absence of adverse prognostic indices, ERT users had a statistically significantly increased risk of being in the poor prognostic group (tumor size > or = 2.0 cm or positive nodes or negative estrogen receptor content) (OR = 4.48, 95% CI = 1.10, 18.30). The risk was highest in current users (OR = 6.28, 95% CI = 1.16, 34.00), users for 5 or more years (OR = 7.77, 95% CI = 1.09, 55.60), and users of nonconjugated estrogen (OR = 9.63, 95% CI = 1.18, 78.60). Although our sample size is small and we do not currently have information on long-term outcomes, the findings from this screening population suggest that ERT may have an adverse effect on important breast cancer prognostic indices.

Postmenopausal estrogen use and invasive versus in situ breast cancer risk.

To examine the effect of cancer histopathology on the relationship between estrogen-replacement therapy (ERT) use and breast cancer risk, we performed a case-control study of 109 postmenopausal women 45 years or older with in situ or invasive breast cancer matched to 545 controls. When in situ and invasive tumors were combined, the overall odds ratio (OR) describing the association between ERT use and breast cancer risk was not statistically significantly elevated (adjusted OR = 1.48, 95% confidence interval [CI] = 0.89-2.47). When the analyses were confined to women with invasive disease, risk estimates were uniformly higher (adjusted OR = 1.85, 95% CI = 1.00-3.45). In contrast, the overall estimate for the relationship between ERT use and in situ breast cancer was close to 1 (adjusted OR = 1.08, 95% CI = 0.42-2.77). The positive association between ERT use and invasive breast cancer we observed, and the lack of association in women with in situ disease, may represent a distinct biological difference or may be related to the small sample size of our study.

The postmenopausal estrogen/breast cancer controversy.

OBJECTIVE: To provide an overview of the postmenopausal estrogen/breast cancer controversy emphasizing the sources of disagreement in the literature and their clinical and research implications. DATA SOURCE AND SELECTION: A MEDLINE search of the English-language literature and a review of bibliographies of meta-analyses describing the association between postmenopausal estrogen use and breast cancer risk. DATA EXTRACTION: Twenty-four original articles and three meta-analyses were reviewed. In addition, five studies that attempted to minimize detection bias were reviewed to assess the potential role of this bias on risk estimates. DATA SYNTHESIS: Among the original articles, risk estimates ranged from a protective to an adverse effect in women who ever used estrogens; no consistent quantitative effects of estrogens on breast cancer risk were found. In the meta-analyses, summary risk estimates were not significantly elevated in women who ever used estrogen. Findings from European-based studies may account for the increased risk associated with increasing duration of use reported in one meta-analysis. In studies that controlled for detection bias, risk estimates were 1 or less in the ever-used category; there was no consistent effect across other categories of use. CONCLUSION: These findings do not support an overall increased risk of breast cancer in women who ever used postmenopausal estrogens or a conclusive or consistent effect across other measures of use. Cross-national differences in estrogen use and inequalities in breast cancer detection between estrogen users and nonusers may account for the increased risk estimates reported in some studies. Newer estrogen and progestin-opposed regimens need to be evaluated further.

Integrating general medicine and rheumatology training in the outpatient setting: a practice model.

The authors describe a primary care-based educational and practice model that integrates general medicine resident education in outpatient rheumatology with specialty fellowship training. Compared with the use of traditional specialty clinics, the model provides better access and service to patients and more appropriate training for residents. Revenues from clinical service delivered by faculty-supervised residents and fellows support 80% of the operating costs and educational activities of the model. The conceptual framework for the model reconciles the educational goals and practice philosophies of general medicine and specialty training and is applicable to training in other predominantly outpatient specialty areas.

Evidence against the association between alcohol use and ischemic stroke risk.

To examine the relationship between alcohol consumption and ischemic stroke risk, we used data from our case-control study of stroke risk. Eighty-nine patients admitted to the hospital with ischemic stroke documented by computed tomography of the head were matched to 178 controls. Alcohol use was defined by an estimate of customary use (heavy, moderate, light, or none). We found no consistent or significant association between any level of alcohol use and ischemic stroke risk (odds ratios: any, 1.3; heavy, 0.5; moderate, 1.5; and light, 1.5). We repeated the analysis of our study using a control group assembled according to the study criteria of another case-control study that reported a significant association in men with heavy alcohol use (odds ratio, 4.2). We demonstrated that the association in the prior study may be spurious due to methodological problems.

The contributions of individual factors to thromboembolic stroke.

To determine the importance of individual factors to thromboembolic stroke (TES) risk, the authors performed a hospital-based case-control study. Ninety cases (56 men and 34 women, ages 15 to 65) discharged from the hospital between January 1981 and December 1984 with a diagnosis of TES supported by computed tomography were matched to 174 control patients (106 men and 68 women). Data on potential risk factors were obtained from the medical record and telephone interview. Using multivariate analysis, three variables were significantly associated with TES risk: hypertension (odds ratio [OR] = 3.4; 95% confidence interval [CI] 1.9-6.0), diabetes (OR = 4.0; 95% CI 2.0-8.3), and smoking (OR = 2.0; 95% CI 1.2-3.6). The data were also analyzed using a direct risk assessment method. This analysis describes the risk in patients with any one factor compared with patients without any of the factors. The direct estimates of risk increased by 71% for hypertension (OR = 5.8), 28% for diabetes (OR = 5.1), and 90% for smoking (OR = 3.8). The authors conclude that hypertension, diabetes, and smoking are the major risk factors for TES in patients 65 years old or younger.

A controlled study of ischemic stroke risk in migraine patients.

To help resolve the uncertain relationship between migraine headache and ischemic stoke, we performed a hospital-based, case-control study. Eighty-nine cases ages 15-65 with a head computed tomography (CT) scan supported diagnosis of ischemic stroke were matched to 178 control subjects. Using information obtained by telephone interview, the patients were placed into three categories according to explicit criteria: classic migraine; common migraine; and no migraine headache. Overall, the association between migraine and ischemic stroke is significantly increased only in patients with classic migraine [odds ratio (OR) = 2.6, 95% confidence interval (CI) 1.1-6.6]. In addition, classic migraine does not appear to increase ischemic stroke when hypertension, diabetes or smoking are present; however, when these other risk factors are absent, classic migraine is strongly and significantly associated with the risk of ischemic stroke [no hypertension, OR = 5.7 (95% CI 1.6-20.2); no diabetes, OR = 3.4 (95% CI 1.2-9.3); non-smoker OR = 4.3 (95% CI 1.2-15.0)]. Since none of the migraine patients in our study had a migrainous stroke, an underlying disorder other than prolonged vasospasm may be responsible for the observed increased risk. Our data suggest that classic migraine may be a marker for patients at increased risk for ischemic stroke unrelated to a migraine attack.

The association between migraine and cerebral vascular events: an analytical review.

This paper explores the uncertain relationship between migraine headache and thromboembolic stroke. In reviewing the literature that links migraine with thromboembolic cerebral vascular events, a distinction is made between two stroke events that occur in migraine patients: stroke associated with a migraine attack (a migrainous stroke) and stroke unrelated to a migraine attack (a non-migrainous stroke). In a recent community-based stroke register, migrainous strokes occurred at a rate of 3.4 per 100,000 per annum. Prevalence rates for migraine in young stroke populations (11-28%) are similar to those in the general population and do not support an additional long-term risk of non-migrainous stroke in migraine patients. The only study providing a controlled estimate of long-term thromboembolic stroke risk (odds ratio = 1.7; 95% CI 1.3, 2.2) included only women and has not been independently confirmed. The contribution of migraine to other known risk factors for thromboembolic stroke needs to be examined further by controlled studies.

Stroke and migraine in the Oxfordshire Community Stroke Project.

A total of 323 cases of first-ever stroke were registered in the first 2 years of the Oxfordshire Community Stroke Project. Of these patients, 244 (76%) had a stroke due to cerebral infarction. There was a past history of migraine headaches in 56 (17%) of the 323 cases of stroke and in 44 (18%) of the 244 cases of cerebral infarction. A past history of migraine headaches was no commoner in patients with stroke due to cerebral infarction than in those with stroke due to intracranial haemorrhage. One hundred and seventy-three (71%) patients with cerebral infarction had at least one risk factor for ischaemic stroke; the frequency of such risk factors was similar in patients with and without a history of migraine. In 7 (3%) of the 244 patients the cerebral infarction was presumed to be "migrainous"; however, only 3 of these 7 (1.2% of the 24) were free of risk factors for ischaemic stroke. If all 7 cases were considered migrainous, the incidence rate of first migrainous cerebral infarction was 3.36 per 100,000 per year (95% confidence limits 0.87-5.86). If only the 3 patients who were free of risk factors were included, the incidence was 1.44 per 100,000 per year (95% confidence limits 0-3.07).