Osteoporosis and Calcium

A bisphosphonate drug was shown to help prevent fractures in some persons taking high-dose prednisone for at least 1 year. A selective estrogen receptor modulator, raloxifene, was shown to favorably alter several markers of complications of the postmenopausal state without increasing the risk for breast cancer. Finally, the incidence of primary hyperparathyroidism has decreased dramatically since 1975, although the reasons for this decline are unclear.

Etidronate Reduced Fractures in Patients Taking Prednisone

Adachi JD, Bensen WG, Brown J, et al. Intermittent etidronate therapy to prevent corticosteroid-induced osteoporosis. N Engl J Med. 1997; 337:382-7.

Much attention is now being given to the detection and treatment of postmenopausal osteoporosis, but we continue to have difficulty treating glucocorticoid-induced osteoporosis. Glucocorticoids inhibit calcium absorption, but their primary contribution to the treatment of osteoporosis is their ability to decrease bone formation.

Adachi and colleagues used cyclic etidronate therapy to determine whether corticosteroid-induced osteoporosis could be prevented. They conducted a 12-month, randomized, placebo-controlled study in 141 men and women who had recently begun high-dose glucocorticoid therapy and were expected to continue receiving this therapy for at least 1 year. The patients were ambulatory and ranged in age from 19 to 87 years. They had received glucocorticoids for various reasons, primarily rheumatoid arthritis or polymyalgia rheumatica. They were randomly assigned to receive etidronate, 400 mg/d, or placebo for 14 days. This regimen was followed by 76 days of calcium carbonate, 500 mg/d. This cycle was repeated three times. The primary outcome measure was lumbar bone density. Secondary outcome measures included bone density in the femur, trochanter, and radius and the rate of new vertebral fractures.

Lumbar bone density increased 0.6% in the treated group and decreased 3.2% in the placebo group. Similar differences were seen at other bone sites. Vertebral compression fractures occurred in 8.8% of patients receiving etidronate and in 15.4% of those receiving placebo, but this difference was not statistically significant. Many of the fractures in the placebo group were multiple. When only postmenopausal women were considered, the proportion of etidronate recipients with vertebral fractures decreased by 85%. Etidronate was well tolerated.

This study is important because previous studies of the effect of calcium, vitamin D, and calcitonin on glucocorticoid-induced osteoporosis have shown small increases in bone density but have not shown a reduction in the incidence of fractures. This study also has several important limitations. First, it remains unknown whether this protective effect will be sustained beyond 1 year. Second, the data cannot be generalized to all patients taking glucocorticoids because most of these patients had polymyalgia rheumatica or rheumatoid arthritis. Third, the authors measured markers of bone formation and bone resorption but did not perform histomorphometric analyses of bone.

Raloxifene Increased Bone Density and Improved Lipid Levels

Delmas PD, Bjarnason NH, Mitlak BH, et al. Effects of raloxifene on bone mineral density, serum cholesterol concentrations, and uterine endometrium in postmenopausal women. N Engl J Med. 1997; 337:1641-7.

Raloxifene is a selective estrogen receptor modulator that inhibits the growth of estrogen receptor-dependent mammary tumors and reduces the occurrence of mammary tumors in animals. Raloxifene is marketed for postmenopausal women who are at high risk for developing breast cancer.

This 24-month, multicenter, placebo-controlled, double-blind study examined the effects of raloxifene on bone density, lipid levels, and endometrial thickness. Postmenopausal women were eligible if their spinal bone density was within 2.5 SDs of the mean spinal bone density of premenopausal women. The 601 women were randomly assigned to receive placebo or one of three dosages of raloxifene (30, 60, or 150 mg/d). All women were given 400 to 600 mg of calcium per day and were evaluated every 3 months. Ninety-nine percent of patients were white, and 55% had some degree of osteopenia.

With each raloxifene dosage, statistically significant increases occurred in lumbar spine, hip, and total-body bone density. In all raloxifene groups, serum concentrations of total and low-density lipoprotein cholesterol decreased but high-density lipoprotein cholesterol and triglyceride levels did not change. No significant difference in endometrial thickness was seen between the raloxifene and placebo groups. The proportion of women who reported hot flashes or vaginal bleeding was the same in all groups. The investigators also found an overall reduction in markers of bone formation and bone resorption in patients receiving raloxifene.

Raloxifene is approved for the prevention of osteoporosis in postmenopausal women, but its effect on actual fracture risk is unknown. The drug should be given at a dosage of 60 mg/d with or without food, and supplemental calcium should be added if daily calcium intake is inadequate. Raloxifene is contraindicated in women who are or may become pregnant and in women with a history of venous thromboembolic disease. This study did not show an increase in venous thromboembolic events with raloxifene, but this association has been shown previously; the greatest risk seems to occur during the first 4 months of therapy [1]. Other studies have shown that raloxifene is associated with hot flashes and possibly with leg cramps (Unpublished data on file, Lilly Research Laboratories, Indianapolis, Indiana).

The effect of raloxifene on total-body bone density is similar to that seen in women treated with conjugated estrogens and progesterone or 5 mg of alendronate daily. In contrast, the effect of raloxifene on lumbar spine bone density is smaller than that seen in women after 24 months of treatment with estrogen or alendronate. The changes in the lipid profile that occurred after raloxifene treatment suggest that raloxifene may favorably affect the incidence of cardiovascular disease. However, actual cardiovascular outcomes have not been measured.

Prevalence of Primary Hyperparathyroidism Has Waned

Wermers RA, Khosla S, Atkinson EJ, et al. The rise and fall of primary hyperparathyroidism: a population-based study in Rochester, Minnesota, 1965-1992. Ann Intern Med. 1997; 126:433-40.

The introduction of routine measurement of serum calcium levels in the early 1970s resulted in a dramatic increase in the number of cases of primary hyperparathyroidism diagnosed. Most of this increase was found in asymptomatic patients.

Wermers and colleagues evaluated trends in the incidence of primary hyperparathyroidism since the mid-1970s. This population-based descriptive study involved residents of Rochester, Minnesota, who received a diagnosis of primary hyperparathyroidism between 1965 and 1992. The patients were included if they had a pathologic diagnosis of primary hyperparathyroidism, had hypercalcemia with inappropriately elevated parathyroid hormone levels, or had hypercalcemia that lasted for more than 1 year and no apparent cause for this condition other than primary hyperparathyroidism.

Between 1965 and 1975, the incidence of primary hyperparathyroidism increased from about 5 per 100 000 patient-years to 112 per 100 000 patient-years. The incidence then steadily declined until 1992, when it was 4 per 100 000 patient-years. The proportion of patients who presented with symptoms of hypercalcemia declined from 22% in 1965 to 2% between 1983 and 1992. During this 22-year study, only 39 patients (8%) had complications attributable to primary hyperparathyroidism: kidney stones in 25 patients, osteoporosis or fractures in 5, hypercalcemic crisis in 5, peptic ulcer disease in 2, pseudogout in 1, and band keratopathy in 1. Treatment also changed. From 1965 to 1974, 29% of the patients underwent surgery within 6 months of diagnosis; between 1983 and 1992, only 13% had surgery. Similarly, 56% of patients from 1965 to 1970 were observed without therapy compared with 85% from 1983 to 1992.

The authors suggest that the declining incidence of primary hyperparathyroidism could be due to the increasing use of estrogen replacement therapy. However, U.S. trends in the use of estrogen replacement therapy are not consistent with trends for primary hyperparathyroidism, and only 10% to 20% of postmenopausal women in the United States receive long-term estrogen replacement therapy. In addition, the incidence of primary hyperparathyroidism declined in both women and men. The authors also suggest that dietary changes may have modified the risk for the disease. Calcium requirements increase with age. If elderly persons do not maintain adequate calcium intake, parathyroid stimulation could occur, and chronic stimulation could lead to adenomatous transformation of the glands. The authors posit that the decrease might be due to the effects of ionizing radiation. Other studies have reported a fourfold increase in the risk for primary hyperparathyroidism among atomic bomb survivors and an 11-fold increase in risk among patients who receive head and neck irradiation [2, 3]. Finally, these results could have occurred if the pool of patients with previously occult hyperparathyroidism had been given a diagnosis through screening programs and was now exhausted.

The decline in the incidence of primary hyperparathyroidism has important implications for clinicians with respect to the prior probability of disease and the clinical usefulness of measuring serum calcium as part of a screening panel. In an era of cost control, evidence such as this could argue against measuring serum calcium as part of routine screening. If calcium is measured less frequently, most patients will come to a clinician's attention when symptoms occur. This may not be bad practice because only 2% of the patients recognized in the past decade actually had complications, and those complications were rarely life-threatening.

Thyroid Disease

The treatment of Graves disease is often complicated by worsening of ophthalmopathy, and one study showed that the addition of prednisone to radioactive iodine therapy could alleviate some of the complications. The diagnosis of thyroid nodules continues to be controversial after investigators found a high prevalence of nonpalpable nodules in patients at high risk for developing nodular thyroid disease.

Prednisone Blocked Ophthalmopathy from Radioactive Iodine Treatment

Bartalena L, Marcocci C, Bogazzi F, et al. Relation between therapy for hyperthyroidism and the course of Graves' ophthalmopathy. N Engl J Med. 1998; 338:73-8.

The relation between the treatment of Graves disease and the development of Graves ophthalmopathy is controversial. In general, thyroidectomy and antithyroid drug therapy are not associated with worsening of preexisting eye disease. In contrast, various effects of radioactive iodine on eye disease have been reported.

Bartalena and associates sought to resolve this controversy in a randomized study of the effects of treatment of Graves hyperthyroidism with methimazole, radioactive iodine, or radioactive iodine combined with glucocorticoids. They studied 443 patients with Graves hyperthyroidism who had slight or no ophthalmopathy. All received methimazole therapy lasting 3 to 4 months and were then randomly assigned to one of three groups: radioactive iodine alone, radioactive iodine followed by a 3-month course of prednisone (0.4 to 0.5 mg/kg of body weight), or methimazole. A single examiner did ocular examinations every 2 months for 18 months.

Of the patients who received radioactive iodine alone, ophthalmopathy developed or worsened in 15%. Smoking was more common in this group, and more patients who had worsening of eye disease smoked. Of the patients who received radioactive iodine followed by prednisone, 67% had regression of ophthalmopathy and none showed worsening of their eye disease. Of the patients treated with methimazole, 4% had improvement of their eye disease and 3% had new or worsening ophthalmopathy.

The authors suggested several reasons for the worsening of ophthalmopathy with radioactive iodine. First, the confounding factor of smoking was raised. Second, ophthalmopathy may also be a result of thyroid antigens released after radiation injury and subsequent enhancement of autoimmune responses directed toward antigens shared by the thyroid and the orbit. This mechanism has also been postulated as an explanation for the occurrence of ophthalmopathy after irradiation of the neck for nonthyroidal disorders or after other thyroid-destructive processes.

One might conclude from this article that patients with ophthalmopathy should be treated with methimazole while waiting for the hyperthyroidism and ophthalmopathy to subside. However, permanent control is difficult to achieve with pharmacologic agents. Radioiodine gives permanent control. In addition, the negative effects of radioiodine on ophthalmopathy are transient and can be prevented by the short-term use of corticosteroids.

Ultrasonography Was Extremely Sensitive in Detecting Nodules

Schneider AB, Bekerman C, Leland J, et al. Thyroid nodules in the follow-up of irradiated individuals: comparison of thyroid ultrasound with scanning and palpation. J Clin Endocrinol Metab. 1997; 82:4020-7.

Thyroid ultrasonography is a very sensitive technique that can detect nodules as small as 2 mm. No radiation is involved; thus, ultrasonography may be a good tool for screening for nodular thyroid disease. Because it is so sensitive, however, it has the potential to be misused in the evaluation, treatment, and follow-up of patients.

Schneider and coworkers studied the role of ultrasonography in patients with higher-than-average risk for developing thyroid nodular disease. They had previously assembled a cohort of 4296 persons who had undergone therapeutic head and neck irradiation during childhood. A subgroup of 54 persons seen between 1974 and 1976 had normal thyroid scans and no palpable nodules. These persons underwent thyroid ultrasonography, (^99m) Tc-pertechnetate scanning, and thyroid palpation.

Eighty-seven percent of the patients had one or more discrete nodules on ultrasonography, and 75% of the nodules were less than 1 cm in diameter. Of the ultrasonography-detected nodules that were greater than 1 cm, 67% were present on (^99m) Tc-pertechnetate scanning. Serum thyroglobulin levels correlated with the number of thyroid nodules but not with the volume of thyroid nodules.

Although it is generally believed that nodules can be palpated when they grow to about 1.5 cm, less than half of the nodules that size were palpable. Nodules were more likely to be palpable if they were anterior and protruded from the gland.

Thyroid ultrasonography is more sensitive than physical examination and scanning. Ultrasonography reveals nodules in 13% to 67% of healthy persons who have no radiation exposure. In this high-risk group, ultrasonography revealed nodules in 87% of patients. The manner in which ultrasonography should be used when a thyroid nodule is not palpable is still controversial. Patients with no history of radiation exposure should not have ultrasonographic screening when findings on thyroid examination are normal. In patients at higher risk, thyroid ultrasonography performed at intervals may be beneficial. Because the test is so sensitive, however, great caution must be used in interpretation of the results.

Prolactinomas

The treatment of choice for prolactin-secreting pituitary tumors is a dopamine agonist. Bromocriptine and pergolide rapidly normalize serum prolactin and induce tumor shrinkage, but adverse effects, such as nausea and the development of drug resistance, often limit the usefulness of these drugs.

Cabergoline Was Effective When Bromocriptine Failed

Caloa A, Di Sarno A, Sarnacchiaro F, et al. Prolactinomas resistant to standard dopamine agonists respond to chronic cabergoline treatment. J Clin Endocrinol Metab. 1997; 82:876-83.

Cabergoline is a new dopamine agonist that has high specificity for the dopamine D2 receptor and a very long half-life. These Italian researchers studied cabergoline in 27 patients 15 to 64 years of age who had prolactinomas. All patients had gonadal dysfunction, and 19 had macroadenomas. Several had previously undergone surgery, and all had received unsuccessful treatment with bromocriptine for at least 3 months before study entry. Cabergoline was given to all patients at a dosage that began at 0.25 mg weekly and was titrated to 0.5 mg twice weekly. Outcome measures were serum prolactin levels and change in tumor size.

In all patients with microadenomas and in about half of patients with macroadenomas, prolactin levels returned to normal. Gonadal function improved in 67% of patients with the largest tumors and in 88% of those with microadenomas. Tumors shrank in 44% of patients as early as 3 months after therapy began. Cabergoline was well tolerated, even in patients who had noted adverse effects with bromocriptine. Almost all side effects resolved after the second week of treatment.

The mechanism by which resistance to dopamine agonists occurs is poorly understood. In some patients, the density of the high-affinity D2 dopamine receptor sites decreases during therapy. Cabergoline has a much higher affinity for these sites than does bromocriptine, and this higher affinity may explain why cabergoline seems to provide a superior response. The longer half-life and slower elimination may also be responsible for the efficacy of cabergoline in bromocriptine-resistant patients.

Lipid Disorders

Careful reanalysis of data reported in a large study of secondary prevention of coronary artery disease found that this therapy was particularly beneficial for women, elderly persons, and patients with diabetes mellitus. In addition, a review of clinical trials showed that cholesterol control may help prevent strokes, although probably not to the degree to which this intervention helps prevent coronary artery events.

Lipid Control Was the Most Effective Prevention Measure for Diabetic Patients

Pyorala K, Pedersen TR, Kjekshus J, et al. Cholesterol lowering with simvastatin improves prognosis of diabetic patients with coronary heart disease. A subgroup analysis of the Scandinavian Simvastatin Survival Study (4S). Diabetes Care. 1997; 20:614-20.

The risk for developing and dying of coronary artery disease is dramatically increased in patients with diabetes. Patients with type 2 diabetes tend to have more diffuse and more extensive nonocclusive coronary artery disease, and this predisposes patients to acute coronary syndromes. Because revascularization is less effective in these patients, inhibiting progression of coronary artery disease with cholesterol-lowering agents could be of great value.

Pyorala and colleagues' study was a subgroup analysis of the original Scandinavian Simvastatin Survival Study (4S), which had shown that simvastatin was effective in the secondary prevention of coronary artery disease [4]. The purpose of this new analysis was to determine the effect of simvastatin on clinically apparent coronary heart disease in patients with diabetes. Study participants were 202 diabetic patients who had previously had acute coronary events and who had serum cholesterol levels of 5.5 to 8.0 mmol/L (213 to 312 mg/dL) and serum triglyceride levels less than 2.5 mmol/L (221 mg/dL). These patients were randomly assigned to receive placebo or simvastatin, 20 to 40 mg/d as necessary. End points were total mortality, major coronary artery events, other acute atherosclerotic events, and myocardial revascularization procedures.

Over the 5.4-year median follow-up period, simvastatin produced changes in serum lipid levels in diabetic patients that were similar to the changes noted in nondiabetic patients. The results, shown in Table 1, strongly suggest that simvastatin therapy improves the prognosis of diabetic patients with coronary artery disease. The absolute clinical benefit achieved through use of this drug may be greater in diabetic patients with coronary artery disease than in nondiabetic patients with coronary artery disease because diabetic patients have a higher absolute risk for recurrent coronary and other atherosclerotic events. The absolute risk reduction among diabetic patients was impressive: It was approximately 25%, compared with the absolute risk reduction of approximately 9% in persons without diabetes (number needed to treat, 4).

These data indicate that use of the lipid-lowering drug simvastatin is the only therapy proven to reduce the risk for recurrent coronary events in diabetic patients. Control of hyperglycemia may also improve outcome, but this has not been shown in a clinical trial. Aggressive management of lipid disorders is the best way to prevent recurrent coronary events in diabetic patients with established coronary disease.

Combined hypercholesterolemia and hypertriglyceridemia is a common problem in patients with type 2 diabetes. The proper way to treat these patients is unclear. Of particular concern is the relative importance of reducing cholesterol levels compared with reducing triglyceride levels. The relation of hypertriglyceridemia to coronary artery disease remains uncertain. For patients with moderately high triglyceride levels (200 to 400 mg/dL), the 3-hydroxy-methylglutaryl coenzyme A reductase inhibitors ("statins") are excellent agents. Atorvastatin at 40 to 80 mg/d or simvastatin at 80 mg/d will predictably reduce triglyceride levels by 30% to 40%. In patients with even higher triglyceride levels, gemfibrozil or fenofibrate should be selected. The role of fish oil in the treatment of diabetic dyslipidemia is unclear: Its effect is dose-related, and patients may have to take up to 10 capsules a day. This may decrease overall compliance.

Elderly Persons and Women Benefited from Simvastatin

Miettinen TA, Pyorala K, Olsson AG, et al. Cholesterol-lowering therapy in women and elderly patients with myocardial infarction or angina pectoris: findings from the Scandinavian Simvastatin Survival Study (4S). Circulation. 1997; 96:4211-8.

The role of cholesterol lowering in women and elderly persons has been unclear. Women, in general, have a lower prevalence of coronary artery disease, but the outcome in women with disease is often worse than the outcome in men. The role of cholesterol lowering in elderly persons is important, but the effectiveness of this therapy has been uncertain.

In this study, women of any age and patients at least 65 years of age in 4S were subjected to subgroup analysis. A total of 821 women and 1021 patients at least 65 years of age were evaluated. The study design and outcomes were the same as those described in Pyorala and colleagues' study on cholesterol lowering in diabetic patients.

Results of this study are summarized in Table 2. It is clear that cholesterol lowering with simvastatin is of major benefit to women with coronary heart disease or hypercholesterolemia and that the high prevalence of coronary artery disease in elderly persons makes them excellent candidates for risk reduction. Although the relative risk reduction in elderly persons is of the same order of magnitude as that in younger patients, the absolute risk reduction is greater. Therefore, fewer patients need to be treated to prevent coronary events.

Statins Decreased Risk for Stroke

Herbert PR, Gaziano JM, Chan KS, et al. Cholesterol lowering with statin drugs, risk of stroke, and total mortality. An overview of randomized trials. JAMA. 1997; 278:313-21.

At least four meta-analyses published in the past 2 years have addressed the relation of cholesterol lowering to the development of stroke. Until these articles were published, a reduction in risk for stroke had not been shown.

Herbert and associates systematically reviewed randomized trials that studied cholesterol lowering with statin drugs and included stroke as an end point. The selected articles were from the English-language literature and were published from 1985 through 1995. The authors also included data from the Cholesterol and Recurrent Events (CARE) study, published in 1996. Overall, 16 trials with 29 000 patients who were followed for an average of 3.5 years were identified.

The use of statins for primary and secondary prevention of coronary heart disease was found to reduce the risk for stroke by approximately 25%. An important finding was that no evidence suggested that noncardiovascular death or death from cancer was increased as a result of cholesterol lowering. However, because the absolute risk for stroke was low compared with the risk for coronary events, the absolute risk reduction was only 7.4 per 1000 patients. One would have to treat 135 patients for 3.5 years to prevent one stroke.

In summary, the total effect of statin drugs in preventing stroke is not great. It is, however, an important additional benefit for patients who require cholesterol lowering for prevention of coronary events.

Diabetes

The previous definition of diabetes was published in 1979, but subsequent data have suggested that the earlier threshold of diagnosis was probably too high. As a result, new definitions were published last year. In addition, more evidence indicated that intense control of hyperglycemia after myocardial infarction improves outcomes.

New Diabetes Threshold Was Set

Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 1998; 21:S5-19.

The definition of diabetes in the United States was guided by a 17-member expert panel of the American Diabetes Association. By 1995, data on the complications of hyperglycemia were so compelling that the expert panel had begun to redefine the diagnostic criteria for diabetes.

This report recommended several changes in the approach to the diagnosis of diabetes. It recommended avoiding the terms insulin-dependent diabetes and non-insulin-dependent diabetes, which were defined by the National Diabetes Data Group in 1979, and it encouraged use of the terms type 1 diabetes and type 2 diabetes. Perhaps most important, the panel established new criteria for the diagnosis of diabetes (Table 3).

The panel also identified a new category, known as impaired plasma glucose, which is defined as a fasting plasma glucose level between 6.11 and 6.94 mmol/L (110 and 125 mg/dL). As the fasting plasma glucose level increases within the normal range, the area under the curve during an oral glucose tolerance test becomes progressively greater. Although patients may technically have fasting plasma glucose levels within the normal range, those whose levels are at the upper limit of the normal range have impaired glucose tolerance or an increased risk for developing impaired glucose tolerance. An impaired plasma glucose level is not an indication of performance of a glucose tolerance test because the result will probably be abnormal.

This group also addressed screening for diabetes on the basis of estimates that 12% to 15% of adults have diabetes. These recommendations are listed in Table 4.

These recommendations simplify the diagnosis of diabetes and reflect a growing appreciation of the relation of glycemia to microvascular disease. The better the control of hyperglycemia, the lower the risk for developing microvascular complications in type 1 and type 2 diabetes. Correction of insulin resistance and its associated metabolic abnormalities will undoubtedly play an important role in preventing macrovascular complications as well.

Intensive Glucose Control Produced Better Long-Term Outcomes after Myocardial Infarction

Malmberg K. Prospective randomised study of intensive insulin treatment on long term survival after acute myocardial infarction in patients with diabetes mellitus. DIGAMI (Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction) Study Group. BMJ. 1997; 314:1512-5.

Patients with diabetes who sustain a myocardial infarction have worse short- and long-term prognoses than patients without diabetes. Several factors, including more diffuse and extensive coronary artery disease, diabetic cardiomyopathy, altered autonomic tone, and abnormal fibrinolytic and platelet function, may contribute to this difference. These investigators previously showed that short-term mortality after acute myocardial infarction can be reduced by 30% with use of intensive insulin treatment in the coronary care setting.

This study evaluated the effect of intensive insulin therapy at presentation and during the period immediately after acute myocardial infarction on long-term outcome. The investigators recruited 620 diabetic patients who had had myocardial infarction. The patients were randomly assigned to receive either standard care or standard care plus intensive glycemic control for the first 24 hours, followed by subcutaneous insulin four times daily for at least 3 months.

After a mean of 3.4 years, mortality rates were 33% in the intensively treated group and 44% in the control group (relative risk, 0.72 [95% CI, 0.55 to 0.92]; number needed to treat to prevent one death, 9). The effects were most pronounced in patients who had not received insulin before the coronary event, suggesting that the benefit may have been due more to stopping therapy with sulfonylurea drugs than to starting insulin treatment.

The adverse effects of hyperglycemia are also well known in patients who have had stroke. As a result, patients who are hospitalized for stroke or myocardial infarction should receive insulin for at least 3 months to aggressively control their plasma glucose levels.

Dr. Kreisberg: Baptist Health Center, 840 Montclair Road, Birmingham, AL 35213.

Dr. Roberts (Series Editor): Madrona Medical Group, 3199 Steller Court, Bellingham, WA 98226.