Elsevier

Journal of Diabetes and its Complications

Volume 25, Issue 6, November–December 2011, Pages 355-361
Journal of Diabetes and its Complications

The duration of diabetes affects the response to intensive glucose control in type 2 subjects: the VA Diabetes Trial,☆☆

https://doi.org/10.1016/j.jdiacomp.2011.10.003Get rights and content

Abstract

Background

The goal of the VA Diabetes Trial (VADT) was to determine the effect of intensive glucose control on macrovascular events in subjects with difficult-to-control diabetes. No significant benefit was found. This report examines predictors of the effect of intensive therapy on the primary outcome in this population.

Methods

This trial included 1791 subjects. Baseline cardiovascular risk factors were collected by interview and the VA record. The analyses were done by intention to treat.

Findings

Univariate analysis at baseline of predictors of a primary cardiovascular (CV) event included a prior CV event, age, insulin use at baseline, and duration of diagnosed diabetes (all P<.0001). Multivariable modeling revealed a U-shaped relationship between duration of diabetes and treatment. Modeled estimates for the hazard ratios (HRs) for treatment show that subjects with a short duration (3 years or less) of diagnosed diabetes have a nonsignificant increase in risk (HR >1.0) after which the HR is below 1.0. From 7 to 15 years' duration at entry, subjects have HRs favoring intensive treatment. Thereafter the HR approaches 1.0 and over-21-years' duration approaches 2.0. Duration over 21 years resulted in a HR of 1.977 (CI 1.77–3.320, P<.01). Baseline c-peptide levels progressively declined up to 15 years and were stable subsequently.

Interpretation

In difficult-to-control older subjects with type 2 DM, duration of diabetes altered the response to intensive glucose control. Intensive therapy may reduce CV events in subjects with a duration of 15 years or less and may increase risks in those with longer duration.

Introduction

People with type 2 diabetes mellitus (DM) (Stamler, Vaccaro, Neaton, & Wentworth, 1993) have a two- to fourfold greater chance of developing cardiovascular (CV) disease than those without diabetes. Nearly two thirds of people with type 2 DM die from macrovascular disease (Del Prato, 2009). Lipid and blood pressure abnormalities are part of the cardiometabolic syndrome, and other risks such as increased LDL are common in diabetes. Reduction of BP and improvements in lipid levels decrease, but do not eliminate the risk of CV events (Kirkman, McCarren, Shah, Duckworth, & Abraira, 2006). Improving glucose levels has long been considered potentially beneficial for preventing macrovascular disease without convincing prospective controlled data.

The UK Prospective Diabetes Study (UKPDS) examined intensive glucose control in newly diagnosed subjects and found a decrease in myocardial infarctions reaching near significant improvements (P=.052) (UK Prospective Diabetes Study Group (UKPDS), 1998). Prospective trials in other populations have had variable results (Abraira et al., 1997, The Action to Control Cardiovascular Risk in Diabetes Study Group, 2008, The ADVANCE Collaborative Group, 2008). A major difficulty has been that the various studies have examined different populations.

Three long-term trials, Action in Diabetes and Vascular Disease Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) (The ADVANCE Collaborative Group, 2008), Action to Control Cardiovascular Risk in Diabetes (ACCORD) (The Action to Control Cardiovascular Risk in Diabetes Study Group, 2008), and the Veteran's Affairs Diabetes Trial (VADT) (Duckworth et al., 2009), all with the stated aim of determining the role of intensive glucose control on CV events in type 2 DM, have recently been reported. The VADT was a 7.5-year trial in veterans with poor glucose control and inadequate response to initial therapy. Failure of therapy was defined as elevated glycosylated hemoglobin levels after maximal doses of at least one oral hypoglycemic agent and/or on insulin. Participants were randomized to standard vs. intensive treatment arms. Subjects at entry had mean HbA1c levels of 9.4±2.0% and had diagnosed diabetes for 11.4 years (mean). Nonglucose risk factors were treated intensively and equally in both groups (Duckworth et al., 2009).

Intensive glucose control failed to reduce major CV events significantly. The hazard ratio (HR) for the primary event was 0.88 with 95% confidence intervals (CI) of 0.73–1.06 (P=.14). The failure to achieve a significant benefit has led to a post hoc examination of possible contributors to this. In this population, duration of diagnosed diabetes was found to be a potential factor. The findings are presented in this report, which is a post hoc analysis and should be considered hypothesis generating.

Section snippets

Methods

The VADT was a trial of the effects of intensive glucose control on CV events in type 2 diabetes. The end point was first occurrence of one of a composite of CV events (myocardial infarction, CV death, CVA, amputation due to ischemia, surgical intervention for vascular disease, new or worsening congestive failure, and inoperable coronary artery disease). All end points were adjudicated by an endpoints committee that was blinded to participant identity and treatment assignment. The mean duration

Results

Baseline predictors of the first primary endpoint using univariate Cox regression are shown in Table 1A. Non-Hispanic whites had an increased risk (P<.0001) of having a primary event during the trial. Older age, insulin use at baseline, and prior CV event increased the risk with prior CV event having the greatest effect (HR 3.565, CI 2.961–4.292, P<.0001). Among the other predictors was duration of diagnosed diabetes (HR 1.034, CI 1.023–1.045, P<.001). Other significant baseline predictors

Discussion

Although long-term prospective controlled trials have not shown significant decreases in CV events with intensive glucose control, all four of the major trials (UKPDS, ADVANCE, ACCORD, and VADT) have shown nonsignificant improvements. A meta-analysis of the results of the studies involving over 27,000 subjects showed a significant 9% improvement in CV risks (HR 0.91, CI 0.84–0.99) (Turner, 1998).

The meta-analysis supports the benefit of intensive glucose control, but also suggests that the

References (20)

There are more references available in the full text version of this article.

Cited by (0)

A portion of these data was presented at the June 2009 Annual Meeting of the ADA in New Orleans, LA, USA.

☆☆

This study was supported by the VA Cooperative Studies Program, Department of Veterans Affairs Office of Research and Development. All rights are reserved by the United States Department of Veteran Affairs. All authors have filed conflict-of-interest statements. Additional support was received from the American Diabetes Association and the National Eye Institute of the NIH. Pharmaceutical, other supplies, and financial assistance were provided by GlaxoSmithKline, Novo Nordisk, Roche Diagnostics, Sanofi-Aventis, Amylin, and Kos Pharmaceuticals. The assistance of the co-chairs' coordinators, Susan Collier (Phoenix) and Christina Paul (Miami), in the preparation of this manuscript is gratefully acknowledged. Lizy Thottapurathu (Hines) assisted in the data analysis. Special recognition is given to Rebecca Miller (Phoenix), Georgia Roode (Phoenix), and Janet Fawcett, Ph.D. (Phoenix), for preparing the final draft and for consistent support throughout the study, and to our veterans who participated in and made possible this study.

View full text