Correspondence
Glycemic Durability of Monotherapy for Diabetes
N Engl J Med 2007; 356:1378-1380March 29, 2007
- Article
To the Editor:
Kahn et al. (Dec. 7 issue)1 report on A Diabetes Outcome Progression Trial (ADOPT), which assessed the glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. It is difficult to maintain target levels of glycated hemoglobin with the use of traditional oral antidiabetic drugs, owing to declining β-cell function.2 In ADOPT, the annual increase in glycated hemoglobin levels over a 4-year period in patients with newly diagnosed diabetes was greatest with glyburide, intermediate with metformin, and least with rosiglitazone. Our longitudinal data (for 2005–2006) from a nationwide general-practitioner database in Germany (Disease Analyzer)3,4 indicate similar findings. Among 12,304 patients with diabetes who were using oral antidiabetic drugs, the mean of the individual relative differences in glycated hemoglobin levels (those in 2006 divided by those in 2005) was 1.018 (95% confidence interval, 1.016 to 1.020). The mean relative difference in glycated hemoglobin levels was highest for glinides, followed by sulfonylureas, acarbose, metformin, and glitazones (Table 1Table 1
Baseline Glycated Hemoglobin Levels and Annual Change among Patients with Diabetes Who Were Treated with Oral Agents in Primary Care Practices.). In multivariate logistic-regression analyses, the use of glitazones was associated with a significantly smaller increase in glycated hemoglobin levels than was the use of sulfonylureas, after adjustment for age, sex, the other oral antidiabetic drugs, health care use, and practice characteristics (P=0.007); the same trend was observed for metformin (P=0.048). Though all the oral antidiabetes drugs we studied are associated with progression of glycemia, our primary care data indicate the potential value of and need for new agents in the treatment of diabetes.2 Wolfgang Rathmann, M.D., M.S.P.H.
German Diabetes Center, 40225 Düsseldorf, Germany
rathmann@ddz.uni-duesseldorf. de Karel Kostev, M.A.
IMS Health, 60528 Frankfurt, GermanyBurkhard Haastert, Ph.D.
German Diabetes Center, 40225 Düsseldorf, GermanyMr. Kostev reports being an employee of IMS Health, a consulting company that received a grant from Eli Lilly, Germany, to carry out a database study of longitudinal changes in glycated hemoglobin values in patients with diabetes in primary care. Drs. Rathmann and Haastert report receiving consulting fees from IMS Health for this study.
4 References1
Kahn SE ,Haffner SM ,Heise MA , et al. Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med 2006;355:2427-2443
Full Text | Web of Science | Medline2
UK Prospective Diabetes Study (UKPDS) Group
CrossRef | Web of Science | Medline3
Rathmann W ,Haastert B ,Roseman JM ,Gries FA ,Giani G . Prescription drug use and costs among diabetic patients in primary health care practices in Germany. Diabetes Care 1998;21:389-397
CrossRef | Web of Science | Medline4
Icks A ,Haastert B ,Giani G ,Rathmann W . Incremental prescriptions and drug costs during the years preceding diabetes diagnosis in primary care practices in Germany. Exp Clin Endocrinol Diabetes 2006;114:348-355
CrossRef | Web of Science | Medline
To the Editor:
Table 1 in the article by Kahn et al., which details the baseline characteristics of the study population, does not mention the type of antihypertensive therapy subjects in the three groups were taking. As shown in previous studies, some types of antihypertensive treatment (i.e., angiotensin-converting–enzyme inhibitors,1 angiotensin-receptor blockers, and calcium-channel blockers2,3) can improve glycemic control and can even result in normoglycemia in patients with impaired fasting glucose levels. On the other hand, thiazide diuretics4 and beta-blockers5 are known to worsen glycemic control.
Thus, the type of antihypertensive therapy subjects received in the trial conducted by Kahn et al. might have been a factor affecting the primary outcome and should have been documented as a baseline characteristic. Randomization according to this factor at the beginning of the study could have added sophistication to the study design and credibility to the results. Similarly, documentation of the use of aspirin therapy as prophylaxis against coronary artery disease in the three groups might have shed more light on the differences seen in the rates of cardiovascular events.
5 ReferencesAmitabh Parashar, M.D.
Anjali Varma, M.D.
Carilion Clinic, Roanoke, VA 24018
aparashar@carilion.com 1
The DREAM Trial Investigators
Full Text | Web of Science | Medline2
Abuissa H ,Jones PG ,Marson SP ,O'Keefe JH Jr . Angiotensin-converting enzyme inhibitors or angiotensin receptor blockers for prevention of type 2 diabetes: a meta-analysis of randomized clinical trials. J Am Coll Cardiol 2005;46:821-826
CrossRef | Web of Science | Medline3
Manica G ,Grassi G ,Zanchetti A . New-onset diabetes and antihypertensive drugs. J Hypertens 2006;24:3-10
CrossRef | Web of Science | Medline4
Lindholm H ,Persson M ,Alaupovic P ,Carlberg B ,Svensson A ,Samuelsson O . Metabolic outcome during 1 year in newly detected hypertensives: results of the Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation (ALPINE study). J Hypertens 2003;21:1563-1574
CrossRef | Web of Science | Medline5
Thornley-Brown D ,Wang X ,Wright JT Jr , et al. Differing effects of antihypertensive drugs on the incidence of diabetes mellitus among patients with hypertensive kidney disease. Arch Intern Med 2006;166:797-805
CrossRef | Web of Science | Medline
To the Editor:
The conclusion by Kahn et al. that rosiglitazone slows the loss of β-cell function seems unwarranted. The three treatment groups started with similar baseline values for β-cell function, determined with the use of homeostasis model assessment (HOMA 2), and at the end of the study there was no significant difference in the values between the rosiglitazone group and the glyburide group and only a small difference between the rosiglitazone group and the metformin group. The rate of decline in β-cell function was greatest in the glyburide group because of an initial increase in β-cell function according to HOMA 2; it does not represent the true state of β-cell function. The stopping of all drugs or switching of all patients to a similar regimen for 3 months might have made clearer assessment possible.
Rajesh Garg, M.D.
Brigham and Women's Hospital, Boston, MA 02115
rgarg@partners.org Dr. Garg reports receiving speaking fees from Merck and Novartis.
To the Editor:
ADOPT focused on the effect of diabetes medications on an end point of questionable importance to patients (the need for a second medication). In fact, ADOPT results leave patients in the dark as to whether they are better off in important ways (the quality of life and risks of illness and death) with any of these agents, particularly given the burden of their adverse effects as reported for this trial.
ADOPT is not a special case: only 20% of randomized studies of diabetes have reported important patient outcomes.1 We believe the time has come for a broad consensus on a standard set of important outcomes for patients in diabetes trials, like the Outcome Measures in Rheumatology (OMERACT) initiative,2 in order to improve the relevance of such evidence for clinical decision making.
2 ReferencesGunjan Y. Gandhi, M.D.
Victor M. Montori, M.D.
Mayo Clinic College of Medicine, Rochester, MN 55905
gandhi.gunjan@mayo. edu 1
Montori VM ,Wang YG ,Alonso-Coello P ,Bhagra S . Systematic evaluation of the quality of randomized controlled trials in diabetes. Diabetes Care 2006;29:1833-1838
CrossRef | Web of Science | Medline2
Boers M ,Brooks P ,Simon LS ,Strand V ,Tugwell P . OMERACT: an international initiative to improve outcome measurement in rheumatology. Clin Exp Rheumatol 2005;23:Suppl 39:S10-S13
Web of Science | Medline
Author/Editor Response
Gandhi and Montori discuss the need for clinical trials to examine outcomes such as morbidity and mortality. We agree, but this was not the aim of our study. Our study was designed to determine whether the loss of glucose control in patients with type 2 diabetes could be slowed, and we were able to answer this important question definitively. Since improved glucose control clearly reduces microvascular complications,1 slowing the progression of hyperglycemia would be expected to reduce morbidity and increase the quality of patients' lives.
Parashar and Varma correctly mention that antihypertensive agents have dissimilar effects on glucose metabolism. Table 2 of our article shows the percentages of subjects with concomitant use of various antihypertensive agents at some time during the study. There was no significant difference in the use of these medications among the three groups; thus, differential use cannot explain the differences in glycemia that we observed. Similarly, concomitant use of aspirin ranged from 31.6% of patients in the glyburide group to 33.0% in the metformin group and therefore also cannot explain the differences in outcomes.
The importance of the rate of loss of β-cell function in determining the progression of type 2 diabetes is well documented.2 Garg is correct that β-cell function according to HOMA 2 was similar in the rosiglitazone group and the glyburide group at baseline and after 4 years of treatment. However, since the two medications have different mechanisms of action and establishment of their maximal biologic effectiveness takes months, we prespecified that our analysis would commence at 6 months. From then on, the rate of change differed markedly between the two groups, which partly explains the greater durability of the lowering of glucose levels with rosiglitazone. Since insulin sensitivity is a critical determinant of β-cell secretory demand,3 the similar β-cell function according to HOMA 2 among patients receiving rosiglitazone and those receiving glyburide, with better insulin sensitivity among patients receiving rosiglitazone, indicates that at the end of the study, the β cells of the patients receiving rosiglitazone were performing qualitatively better. With regard to the use of glucose-lowering medications before the study commenced, since no patients were taking these medications at the time of randomization, there was no need to discontinue their use or to use the same agent initially before switching to one of the three study treatments.
The primary care data reported by Rathmann et al. confirm our finding that glucose-lowering agents have differential effects in slowing the rate of progression of glycemia, providing support for our conclusion that the choice of initial monotherapy has to be guided by clinical efficacy along with a consideration of adverse events and costs. The challenge now is to develop new agents and approaches that can slow progression even more effectively.
3 ReferencesSteven E. Kahn, M.B., Ch.B.
Veterans Affairs Puget Sound Health Care System, Seattle, WA 98108
skahn@u.washington. edu Giancarlo Viberti, M.D., F.R.C.P.
King's College London School of Medicine, London SE1 9RT, United Kingdomfor the ADOPT Steering Committee
1
UK Prospective Diabetes Study (UKPDS) Group
CrossRef | Web of Science | Medline2
U.K. Prospective Diabetes Study Group. U.K. Prospective Diabetes Study 16: overview of 6 years' therapy of type II diabetes: a progressive disease. Diabetes 1995;44:1249-1258[Erratum, Diabetes 1996;45:1655.]
CrossRef | Web of Science | Medline3
Kahn SE . The relative contributions of insulin resistance and beta-cell dysfunction to the pathophysiology of type 2 diabetes. Diabetologia 2003;46:3-19
CrossRef | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
Bernd Richter, Elizabeth Bandeira-Echtler, Karla Bergerhoff, Christine Clar, Susanne H Ebrahim, Bernd Richter. 2007. Rosiglitazone for type 2 diabetes mellitus. .
CrossRef
