The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 331:1188-1193 November 3, 1994 Number 18
NextNext

Improvement in Glucose Tolerance and Insulin Resistance in Obese Subjects Treated with Troglitazone
John J. Nolan, Bernhard Ludvik, Patricia Beerdsen, Mary Joyce, and Jerrold Olefsky

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background Troglitazone decreases insulin resistance and hyperglycemia in patients with non-insulin-dependent diabetes mellitus (NIDDM), but its effects on subjects without diabetes are not known.

Methods We performed oral and intravenous glucose-tolerance tests, studies with the euglycemic-hyperinsulinemic clamp, meal-tolerance tests, and 24-hour blood-pressure measurements at base line and after the administration of troglitazone, 200 mg orally twice daily, or placebo for 12 weeks in 18 nondiabetic obese subjects, 9 of whom had impaired glucose tolerance.

Results The mean (±SD) rates of glucose disposal increased from 4.7 ±1.7 to 6.0 ±1.7 mg per kilogram of body weight per minute (P = 0.004) and from 9.0 ±1.8 to 9.9 ±1.3 mg per kilogram per minute (P = 0.02) during insulin infusions of 40 and 300 mU per square meter of body-surface area per minute, respectively, in the troglitazone group. The insulin-sensitivity index, calculated from the results of intravenous glucose-tolerance tests, increased from 0.7 ±0.6 x 10-4 to 1.6 ±0.9 x 10-4 in subjects given troglitazone, and their glycemic response to oral glucose and to mixed meals decreased. The mean fasting plasma insulin concentration decreased by 48 percent (P = 0.002), and the plasma insulin response to oral glucose and mixed meals decreased by 40 and 41 percent, respectively. The changes were similar in the subjects with normal glucose tolerance and those with impaired glucose tolerance. Systolic and diastolic blood pressure decreased by 5 ±2 mm Hg (P = 0.05) and 4 ±2 mm Hg (P = 0.04), respectively, after treatment with troglitazone. There were virtually no changes in the placebo group.

Conclusions Troglitazone decreases insulin resistance and improves glucose tolerance in obese subjects with either impaired or normal glucose tolerance. The ability of troglitazone to reduce insulin resistance could be useful in preventing NIDDM.


Non-insulin-dependent diabetes mellitus (NIDDM) is likely to develop in persons with impaired glucose tolerance. The proportion of study subjects in whom the condition progresses to diabetes depends on their characteristics, the length of follow-up, and the means of assessment1,2,3,4,5. Insulin resistance is characteristic of impaired glucose tolerance and is a metabolic abnormality that precedes the development of glucose intolerance and NIDDM6,7,8,9. In subjects with impaired glucose tolerance, insulin resistance and hyperinsulinemia are often associated with dyslipidemia and hypertension,10 and these people also have an increased risk of premature atherosclerosis. Since insulin resistance appears to be central to the etiologic process of this entire syndrome, treatments that improve the action of insulin could be beneficial. Although weight reduction and exercise have such a beneficial effect, compliance and widespread application are problematic. Until recently, no drugs for the treatment of insulin resistance were available.

This situation changed with the introduction of a class of drugs termed thiazolidinediones. Troglitazone is a member of this class of antidiabetic drugs that improve insulin resistance. Although its exact mechanism of action is unknown, troglitazone appears to improve the action of insulin in liver, skeletal muscle, and adipose tissue directly,11,12,13 and it reduces elevated plasma glucose and insulin concentrations in diabetic animals with hyperinsulinemia12.

The administration of troglitazone to patients with NIDDM improves both fasting and postprandial hyperglycemia and insulinemia14. This reduction in hyperglycemia is associated with a near normalization of the rates of hepatic glucose production and a 40-to-60-percent increase in insulin-mediated glucose disposal as measured by the glucose-clamp technique14. In the aggregate, these results are consistent with an effect of troglitazone on the insulin resistance of the liver and skeletal muscle. In this study we measured a spectrum of metabolic variables in a group of obese subjects with insulin resistance who did not have diabetes, half of whom had impaired glucose tolerance, before and after 12 weeks of treatment with troglitazone or placebo, in a randomized double-blind fashion.

Methods

Study Subjects and Protocol

We studied 18 obese subjects (15 men and 3 women), 9 of whom had impaired glucose tolerance according to criteria outlined by the World Health Organization15 -- that is, a fasting plasma glucose concentration below 140 mg per deciliter (7.8 mmol per liter) and a plasma glucose concentration between 140 and 199 mg per deciliter (7.8 and 11.0 mmol per liter) two hours after an oral-glucose challenge. The characteristics of the subjects at base line are shown in Table 1. Obesity was defined as a body-mass index (the weight in kilograms divided by the square of the height in meters) of more than 27; subjects with values above 39 were excluded. Persons with a history or laboratory evidence of chronic illness were also excluded, as were women who were capable of childbearing and persons taking medications that could influence glucose metabolism. The protocol was approved by the human subjects committee, and all subjects gave written informed consent.

View this table:
[in this window]
[in a new window]
 
Table 1. Base-Line Clinical Characteristics of the Study Subjects.

 
Starting at least five weeks before the study and then for the duration of the study, the subjects followed a weight-maintaining diet (28 to 32 kcal per kilogram of body weight) consisting of 50 percent carbohydrate, 35 percent fat, and 15 percent protein. Inpatient meals were prepared under the supervision of the metabolic nutritionist; the outpatient diet was prepared according to specific instructions given to each subject, and the subjects were asked to record their food intake at home. Fat-free body mass was measured by weighing the subjects under water16. The subjects received either troglitazone, 200 mg twice daily orally (12 subjects; 7 with impaired glucose tolerance), or placebo (6 subjects; 2 with impaired glucose tolerance) for 12 weeks in accordance with preassigned randomization codes. The pharmacists, investigators, and study subjects were not aware of the treatment code, which was broken only after the completion of all analyses. The subjects were seen and examined weekly to monitor their general well-being and potential adverse reactions. Compliance was monitored by weekly pill counts. Each subject's dietary records were reviewed weekly by the metabolic nutritionist to ensure compliance. Physical activity was maintained at a constant level throughout the study, and caloric intake was adjusted as necessary to ensure weight maintenance. Body weight, blood pressure, fasting plasma glucose concentrations, and serum chemical and hematologic profiles were determined at each visit.

Base-Line and Final Metabolic Assessments

The subjects were hospitalized for 7 to 10 days for the base-line and final studies. Blood pressure was recorded continuously for 24 hours with a Spacelabs (Redmond, Wash.) portable recording system. On separate days, in random order, the subjects underwent a standard three-hour oral glucose-tolerance test (75 g), a seven-hour meal-tolerance test (with two identical liquid-formula meals of the same composition as the study diet, each containing 33 percent of the daily calories, prepared in the metabolic kitchen, which were consumed at 8:30 a.m. and 12:30 p.m.), and an intravenous glucose-tolerance test (300 mg of glucose per kilogram)17. On two other days, in random order, studies with the euglycemic-hyperinsulinemic clamp in which insulin was infused at rates of 40 and 300 mU per square meter per minute for at least 240 minutes were performed as previously described14. Glucose turnover was measured by a concomitant infusion of [3-3H]glucose, with the variable tracer method,18 and calculated with the Steele equations19. Plasma glucose, insulin, and glucagon concentrations were measured as described previously14.

Insulin sensitivity was estimated by two methods. The insulin-sensitivity index represents the ability of endogenous insulin to increase the disappearance of glucose from the extracellular fluid, both by inhibiting hepatic glucose production and by stimulating peripheral glucose use. It is calculated from the plasma insulin and glucose values measured after the intravenous administration of glucose17. The clamp-derived insulin-sensitivity index is an index of the effect of a change in the plasma insulin concentration on glucose clearance (the rate of glucose uptake divided by the plasma glucose concentration) per unit of body-surface area. The index for each subject was calculated with the use of the basal hepatic glucose output, basal plasma insulin concentrations, the glucose-disposal rate, and mean plasma insulin concentrations from the studies with the euglycemic-hyperinsulinemic clamp in which the insulin-infusion rate was 40 mU per square meter per minute. Thus, the clamp-derived insulin-sensitivity index = ({Delta}GDR/{Delta}I x G), where GDR (glucose-disposal rate) is expressed in milligrams per square meter per minute, I (insulin) is expressed in microunits per milliliter, and G (glucose) is expressed in milligrams per deciliter, as previously described20.

Statistical Analysis

Single-variable statistical comparisons were made with paired t-tests for pretreatment and post-treatment results. The results of glucose-tolerance and meal-tolerance tests were subjected to analysis of variance for repeated measures. The incremental area under the curve (for the glucose- and meal-tolerance tests) was calculated by summing the areas of successive triangles and rectangles under the graph for mean increases above the basal values. All statistical tests were two-tailed.

Results

The subjects' weights and levels of physical activity did not change during the study. Troglitazone was well tolerated by all subjects, and none had any subjective side effects or persistent abnormalities in laboratory variables.

The mean fasting plasma glucose concentration decreased slightly in the troglitazone group and did not change in the placebo group (Table 2). There was a 48 percent decrease in the mean fasting plasma insulin concentration in the former group (P = 0.002), and no substantial change in the latter. The fasting plasma glucagon concentrations and rates of basal hepatic glucose output did not change substantially in either group.

View this table:
[in this window]
[in a new window]
 
Table 2. Summary of Metabolic Measurements before and after the Administration of Troglitazone or Placebo for 12 Weeks in Nondiabetic Obese Subjects.

 
Oral Glucose-Tolerance Tests

For the troglitazone group as a whole, the average base-line plasma glucose concentrations met the criteria for impaired glucose tolerance, whereas the values were normal after treatment (Figure 1). For example, mean (±SD) values at two hours were 146 ±25 and 126 ±17 mg per deciliter (8.1 ±1.4 and 7.0 ±0.9 mmol per liter) before and after troglitazone treatment, respectively, and the incremental area under the glucose curve decreased by 25 percent. All but one of the seven subjects in the troglitazone group who had impaired glucose tolerance before treatment had normal glucose tolerance after treatment. The mean two-hour values in these seven subjects were 164 ±14 and 131 ±19 mg per deciliter (9.1 ±0.8 and 7.3 ±1.1 mmol per liter) before and after troglitazone treatment, respectively, and the incremental area under the curve decreased by 36 percent (P = 0.03). Troglitazone also reduced plasma insulin concentrations, with a 40 percent decrease in the incremental area under the curve (P = 0.002) (there was a 48 percent decrease in this variable in the seven subjects with impaired glucose tolerance; P = 0.008). There was no change in either plasma glucose or insulin concentrations in the placebo group.


View larger version (25K):
[in this window]
[in a new window]
 
Figure 1. Results of Oral Glucose-Tolerance Tests before and after the Administration of Troglitazone or Placebo for 12 Weeks in Obese Subjects without Diabetes.

Values are means ±SE. To convert values for glucose to millimoles per liter, multiply by 0.056. To convert values for insulin to picomoles per liter, multiply by 6.

 
Meal-Tolerance Tests

The plasma glucose and insulin responses to the two mixed meals are shown in Figure 2. Treatment with troglitazone resulted in a 24 percent reduction in postprandial plasma glucose concentrations (incremental area under the glucose curve) and an even greater fall (41 percent, P = 0.006) in plasma insulin concentrations. The plasma glucose and insulin concentrations were virtually unchanged in the placebo group. The effect of troglitazone was greater among the subjects with impaired glucose tolerance, in whom the area under the glucose curve decreased by 40 percent after troglitazone therapy. Plasma free fatty acid and glucagon concentrations did not change in either group.


View larger version (47K):
[in this window]
[in a new window]
 
Figure 2. Results of Meal-Tolerance Tests before and after the Administration of Troglitazone or Placebo for 12 Weeks in Obese Subjects without Diabetes.

Values are means ±SE. To convert values for glucose to millimoles per liter, multiply by 0.056. To convert values for insulin to picomoles per liter, multiply by 6.

 
Measurements of Insulin Resistance

The results of the glucose-clamp studies are shown in Figure 3. At infusion rates of both 40 and 300 mU of insulin per square meter per minute, there was a significant increase in the rate of glucose disposal in the group given troglitazone. At the lower (and more physiologic) rate of insulin infusion, the mean rate of glucose disposal increased from 4.7 ±1.7 to 6.0 ±1.7 mg per kilogram per minute (P = 0.004); 10 of 12 subjects had increased values. At the higher rate of insulin infusion, the rate of glucose disposal increased from 9.0 ±1.8 to 9.9 ±1.3 mg per kilogram per minute (P = 0.02). There were virtually no changes in the placebo group. The values for the clamp-derived insulin-sensitivity index, which approximate the slope of the effect of this range of insulinemia (from basal concentrations to concentrations of approximately 90 micro U per milliliter [approximately 540 pmol per liter]) in stimulating glucose disposal, increased from 1.6 to 2.8 (P = 0.007) after troglitazone treatment (the increase was more than twofold in the subgroup with impaired glucose tolerance but did not change in the placebo group). The values for the insulin-sensitivity index as calculated from the results of the intravenous glucose-tolerance test17 more than doubled, from 0.7 ±0.6 x 10-4 to 1.6 ±0.9 x 10-4 (P = 0.002), in the troglitazone group (Figure 3); 10 of the 12 subjects had improved values. There was a small but not statistically significant decrease in the placebo group.


View larger version (24K):
[in this window]
[in a new window]
 
Figure 3. Measurements of Insulin Resistance before and after the Administration of Troglitazone or Placebo for 12 Weeks in Obese Subjects without Diabetes.

Values are means ±SE. Panel A shows the results of studies with the euglycemic-hyperinsulinemic clamp in which insulin was infused at a rate of 40 mU per square meter per minute. Panel B shows the results of studies with the euglycemic-hyperinsulinemic clamp in which insulin was infused at a rate of 300 mU per square meter per minute. Panel C shows the values for the insulin-sensitivity index as calculated from basal values and values measured during the studies with the euglycemic-hyperinsulinemic clamp in which insulin was infused at a rate of 40 mU per square meter per minute. Panel D shows the values for the insulin-sensitivity index as calculated from the results of intravenous glucose-tolerance tests.

 
Blood Pressure and Plasma Lipid Concentrations

Troglitazone treatment was associated with a statistically significant reduction in systolic blood pressure of 5 ±2 mm Hg (P = 0.05) and in diastolic blood pressure of 4 ±2 mm Hg (P = 0.04) (Table 2). There was an increase in systolic blood pressure in the placebo group. The mean plasma low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglyceride concentrations at base line were similar in both groups, and the values did not change substantially in either group.

Discussion

A 12-week course of treatment with troglitazone in a group of nondiabetic subjects with insulin resistance decreased insulin insensitivity, glucose intolerance, hyperinsulinemia, and hypertension. These abnormalities have been reported as potential risk factors for the development of cardiovascular disease10. Our results suggest that primary treatment of insulin resistance may improve the overall risk profile for cardiovascular disease.

These results are consistent with the reported effects of troglitazone in vitro and in animals. Thus, although its exact mechanism of action is unknown, this drug works directly on skeletal muscle, liver, and adipose tissue to potentiate the action of insulin13. It also has potent glucose- and insulin-reducing effects and reduces insulin resistance in diabetic animals with hyperinsulinemia12. Treating patients with NIDDM with troglitazone reduces hyperglycemia, hyperinsulinemia, and hypertriglyceridemia and improves insulin sensitivity11,14.

The measures of insulin resistance we used did not all show the same degree of improvement. For example, after troglitazone treatment the increase in glucose disposal during the glucose-clamp study in which a high dose of insulin was infused was not as great as the increase during the study in which a lower dose of insulin was infused. This finding is fully consistent with our previous observations that impaired glucose tolerance is characterized by decreased insulin sensitivity with little change in insulin responsiveness21. In other words, insulin resistance is manifested at physiologic plasma insulin concentrations, with a normal response to pharmacologic concentrations. Treatment that improves the insulin resistance in subjects with impaired glucose tolerance would therefore have its greatest effect at physiologic insulin concentrations. The results of the calculations of insulin sensitivity support this conclusion. In both tests, which measure insulin sensitivity over the physiologic range of insulin concentrations, troglitazone treatment was associated with an even greater improvement in this index than in glucose disposal. Although the exact cellular mechanisms causing insulin resistance in the type of subjects we studied are not known, our results suggest that troglitazone exerts its effects at a cellular site relevant to the basic defect in these subjects.

Oral glucose tolerance was improved after troglitazone treatment; the mean postprandial plasma glucose concentrations decreased, as did the concentrations measured during glucose-tolerance tests. Predictably, this improvement was greatest in the subjects whose glucose values were highest initially; six of seven subjects with impaired glucose tolerance had normal glucose tolerance after troglitazone treatment.

It has been proposed that insulin resistance, hyperinsulinemia, or both can cause hypertension in some subjects, but the mechanisms are poorly understood10. Our results, showing that troglitazone, a drug that improves the action of insulin, can also lower blood pressure, strengthen the association between the syndrome of insulin resistance and hypertension.

The fact that troglitazone can improve insulin resistance, lower plasma insulin concentrations, and normalize glucose tolerance in subjects with impaired glucose tolerance could have implications for the prevention of NIDDM. Insulin resistance is a primary metabolic defect in most patients with NIDDM6,7,8,9. Thus, insulin resistance and hyperinsulinemia are present many years before glucose tolerance deteriorates6,7. Acquired factors, such as obesity, a sedentary lifestyle, and aging, may be contributory, but insulin resistance is probably a primary inherited feature in most patients with NIDDM. In the presence of normal beta-cell function, this will lead to hyperinsulinemia but relatively normal glucose tolerance9. Thus, in the compensated insulin-resistant, hyperinsulinemic state one has either normal glucose tolerance or impaired glucose tolerance, but not diabetes. In patients in whom diabetes develops, the process of compensatory hyperinsulinemia eventually fails and beta-cell function declines, leading to hyperglycemia. Therefore, interventions that might ameliorate insulin resistance in the prediabetic state could be of potential benefit in preventing the development of NIDDM, assuming that the agents' effects on insulin resistance would be sustained during prolonged periods of treatment.

In summary, treatment of nondiabetic obese subjects who have insulin resistance with the new antidiabetic agent troglitazone decreases insulin resistance, reduces hyperinsulinemia, improves glucose tolerance, and reduces both systolic and diastolic blood pressure.

Supported in part by a grant from the National Institutes of Health (DK 33651), a grant from the General Clinical Research Centers of the National Institutes of Health (MO1 RR00827), and a grant from the Sankyo Pharmaceutical Company. Dr. Ludvik is a recipient of a Max-Kade Foundation Postdoctoral Fellowship Award.

Dr. Olefsky is a consultant to Parke Davis Company, which holds the license for troglitazone in the United States.

We are indebted to the staff of the Special Diagnostic and Treatment Unit at the San Diego Veterans Affairs Medical Center, to Ms. Elizabeth Hansen for assistance in the preparation of the manuscript, to Ms. Reena Deutsch for statistical advice, and to the study participants, who gave generously of their time and energy.


Source Information

From the Department of Medicine, University of California, San Diego, Division of Endocrinology and Metabolism, La Jolla (J.J.N., B.L., P.B., J.O.), and the Veterans Affairs Medical Center, San Diego, Calif. (M.J.).

Address reprint requests to Dr. Olefsky at the Division of Endocrinology and Metabolism, 9111-G, University of California, San Diego, La Jolla, CA 92093.

References

  1. Kosaka K, Hagura R, Kuzuya T. Insulin responses in equivocal and definite diabetes, with special reference to subjects who had mild glucose intolerance but later developed definite diabetes. Diabetes 1977;26:944-952. [Medline]
  2. Keen H, Jarrett RJ, McCartney P. The ten-year follow-up of the Bedford survey (1962-1972): glucose tolerance and diabetes. Diabetologia 1982;22:73-78. [Medline]
  3. Kadowaki T, Miyake Y, Hagura R, et al. Risk factors for worsening to diabetes in subjects with impaired glucose tolerance. Diabetologia 1984;26:44-49. [Medline]
  4. Knowler WC, Pettitt DJ, Saad MF, Bennett PH. Diabetes mellitus in the Pima Indians: incidence, risk factors and pathogenesis. Diabetes Metab Rev 1990;6:1-27. [Medline]
  5. Haffner SM, Stern MP, Mitchell BD, Hazuda HP, Patterson JK. Incidence of type II diabetes in Mexican Americans predicted by fasting insulin and glucose levels, obesity, and body-fat distribution. Diabetes 1990;39:283-288. [Abstract]
  6. Warram JH, Martin BC, Krolewski AS, Soeldner JS, Kahn CR. Slow glucose removal rate and hyperinsulinemia precede the development of type II diabetes in the offspring of diabetic parents. Ann Intern Med 1990;113:909-915.
  7. Eriksson J, Franssila-Kallunki A, Ekstrand A, et al. Early metabolic defects in persons at increased risk for non-insulin-dependent diabetes mellitus. N Engl J Med 1989;321:337-343. [Abstract]
  8. Ho LT, Chang ZY, Wang JT, et al. Insulin insensitivity in offspring of parents with type 2 diabetes mellitus. Diabetic Med 1990;7:31-4.
  9. Lillioja S, Mott DM, Spraul M, et al. Insulin resistance and insulin secretory dysfunction as precursors of non-insulin-dependent diabetes mellitus: prospective studies of Pima Indians. N Engl J Med 1993;329:1988-1992. [Free Full Text]
  10. Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595-1607. [Abstract]
  11. Iwamoto Y, Kuzuya T, Matsuda A, et al. Effects of new oral antidiabetic agent CS-045 on glucose tolerance and insulin secretion in patients with NIDDM. Diabetes Care 1991;14:1083-1086. [Abstract]
  12. Fujiwara T, Yoshioka S, Yoshioka T, Ushiyama I, Horikoshi H. Characterization of new oral antidiabetic agent CS-045: studies in KK and ob/ob mice and Zucker fatty rats. Diabetes 1988;37:1549-1558. [Abstract]
  13. Ciaraldi TP, Gilmore A, Olefsky JM, Goldberg M, Heidenreich KA. In vitro studies on the action of CS-045, a new antidiabetic agent. Metabolism 1990;39:1056-1062. [CrossRef][Medline]
  14. Suter SL, Nolan JJ, Wallace P, Gumbiner B, Olefsky JM. Metabolic effects of new oral hypoglycemic agent CS-045 in NIDDM subjects. Diabetes Care 1992;15:193-203. [Abstract]
  15. Diabetes mellitus: report of a WHO study group. World Health Organ Tech Rep Ser 1985;727:9-17. 
  16. Goldman RF, Burskirk ER. Body volume measurement by underwater weighing: description of a method. In: Brozek J, Henschel A, eds. Techniques for measuring body composition. Washington, D.C.: National Research Council, 1961:78-89.
  17. Welch S, Gebhart SSP, Bergman RN, Phillips LS. Minimal model analysis of intravenous glucose tolerance test-derived insulin sensitivity in diabetic subjects. J Clin Endocrinol Metab 1990;71:1508-1518. [Abstract]
  18. Finegood DT, Bergman RN, Vranic M. Estimation of endogenous glucose production during hyperinsulinemic-euglycemic glucose clamps: comparison of unlabeled and labeled exogenous glucose infusates. Diabetes 1987;36:914-924. [Abstract]
  19. Steele R. Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y Acad Sci 1959;82:420-430.
  20. Bergman RN, Prager R, Volund A, Olefsky JM. Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp. J Clin Invest 1987;79:790-800.
  21. Kolterman OG, Gray RS, Griffin J, et al. Receptor and postreceptor defects contribute to the insulin resistance in noninsulin-dependent diabetes mellitus. J Clin Invest 1981;68:957-969.

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved.