European Association for the Study of Diabetes
European Association for the Study of Diabetes
September 28, 2013
3 min read

UKPDS: Risk for diabetes-related death, all-cause mortality no longer significant

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The risk for diabetes-related death and all-cause mortality, which was initially an area of concern in the United Kingdom Prospective Diabetes Study 15 years ago, no longer appears to be statistically significant, according to data presented at  the European Association for the Study of Diabetes annual meeting.

According to Rury R. Holman, FRCP, director of the University of Oxford Diabetes Trials Unit and honorary consultant physician to the Oxford Radcliffe Hospitals NHS Trust, the effect of the early addition of metformin to sulfonylurea does not appear to have continuing adverse effects.

“We’re seeing a diminishing risk ratio, and at this point, after this degree of follow-up, close to 15 years in total for these patients as opposed to the original 6.6 (years), there doesn’t appear to be evidence of harm,” Holman said during a presentation.

Unexpected data

To better understand the effects of combination therapy, Holman and colleagues conducted a sulfonylurea plus metformin substudy of the United Kingdom Prospective Diabetes Study (UKPDS). Normal and overweight patients (n=537) were randomly assigned to either continue sulfonylurea alone (n=269) with an aim for a fasting plasma glucose of <15 mmol/L or to add metformin to the existing sulfonylurea (n=268) with the goal of FPG <6 mmol/L.

Post-trial monitoring substudy results indicate that 57 deaths occurred among the sulfonylurea alone group vs. 60 deaths in the sulfonylurea plus metformin group (RR=1.18; 95% CI, 0.82-1.69), according to data presented by Holman.

“Ten years after cessation of the randomized trial, the apparent increased risks of diabetes-related deaths and all-cause mortality are greatly diminished and certainly no longer statistically significant,” he said.

According to Holman, the substudy findings may be paradoxical for the following reasons:

  • Major effects appeared to be fewer deaths in the sulfonylurea alone group, not more with sulfonylurea plus metformin.
  • Small numbers of deaths were subject to “an evil play of chance” (ie, 26 deaths on combined therapy vs. 14 deaths on sulfonylurea alone).
  • The incidence of fatal and nonfatal events were not different (coronary heart disease: RR=1.09; 95% CI, 0.67-1.78; stroke: RR=1.21; 95% CI, 0.58-2.55).
  • Epidemiological analysis of all patients showed no increased risk in those taking combination therapies.
  • Metformin and sulfonylurea when given separately both decreased risk for diabetes complications.

Global impact

According to David R. Matthews, MA (Oxon), DPhil, BM, BCh, FRCP, medical tutor at Harris Manchester College and professor of diabetes medicine at the University of Oxford, all guidelines quote the UKPDS as the justification for using metformin as the first-line of therapy.

“The UKPDS dramatically changed the role of metformin to be the background medication on which all other pharmacotherapy was added, and the principle of intensive therapy is widely accepted,” Matthews said during a presentation.

Further, Matthews said the UKPDS was the first definitive evidence that intensive glycemic control reduced macrovascular events by a large and highly significant margin.

He said because of the UKPDS, millions of patients will have better outcomes and better lives.

Amanda Adler, MD, PhD, FRCP, of the Institute of Metabolic Science at the Wolfson Diabetes and Endocrine Clinic with Addenbrooke’s Hospitals in Cambridge, said although some have said the UKPDS is an outdated, old study, bias has been well described. Adler chaired the session at EASD in which these presentations were delivered.

“We learned from [Holman] and [Matthews] that in the 15 years after the presentation of the UKPDS, it set the standard for treatment of type 2 diabetes in a relatively uncomplicated policy to achieve blood glucose of less than 6%. It provided epidemiologic data that determined and documented the frequency of complications in type 2 diabetes and it set the standard for subsequent trials,” Adler said.

Moving forward

Holman and colleagues said they are looking forward to the Glucose Lowering in Nondiabetic Hyperglycemia Trial (GLINT), in conjunction with the University of Cambridge.

They have begun the UK-based, multicenter cardiovascular primary prevention trial with 500 patients in a feasibility phase and nearly 12,000 patients in the 5-year main trial.

The double blind study will examine men and women aged at least 40 years, with HbA1c levels between 5.5% and 6.5%, assessing their 10-year CV disease risk. Patients will be assigned to metformin XR 1,500 mg daily or placebo. The primary endpoint will be time to composite CV outcome or CV death, nonfatal myocardial infarction or nonfatal stroke. Secondary endpoints will include the incidence of cancer.

“I look forward to this daughter or son of UKPDS finally helping us answer one of the unresolved conundrums,” Holman said.

For more information:

Holman RR. UKPDS: 15 years on from Barcelona. Post trial monitoring results of the UKPDS sulfonylurea plus metformin substudy.

Matthews D. UKPDS: 15 years on from Barcelona. Global impact of UKPDS findings. Both presented at: 49th Annual Meeting of the European Association for the Study of Diabetes; Sept. 24-27; Barcelona, Spain.

Disclosure: Holman reports honoraria from Amylin, Bayer, Lilly, Merck, Merck Serono, Novartis and Novo Nordisk. Matthews reports multiple disclosures with Eli Lilly, GlaxoSmithKline, Janssen, Johnson & Johnson, NIHR, Novartis, Novo Nordisk, Sanofi-Aventis and Servier.