Hypoglycemia, 25 Years After DCCT

Richard M. Plotzker, MD


June 27, 2018

Later this year, one of the most monumental research projects evaluating the treatment of diabetes will turn 25 years old. The Diabetes Control and Complications Trial[1] (DCCT) dominated endocrinology news, and master clinicians spent 9 years speculating about what the results would tell us and how they would transform clinical practice.

As the endocrine fellow at one of the 29 renowned centers, I had no involvement in the study, but across from the examination areas was a small room containing an unobtrusive, faux wood, plastic plaque with white DCCT lettering that, in effect, subsidized the other activities of the diabetes section. The safety precautions of the study and the project itself were quite intricate, so no expense was spared, making it the single most expensive federally funded medical research project of its time.

The Diabetes Control and Complications Trial

In this landmark study,[1] 1440 young adults with type 1 diabetes were recruited with the researchers' intent to follow them for 10 years. The study was terminated and reported 1 year earlier than planned.

Participants had no end organ involvement at study initiation because they had been diagnosed with diabetes less than 5 years before enrollment. They were meticulously evaluated during the study for end organ involvement using state-of-the-art surveillance.

Participants were randomly assigned to conventional therapy or intensive therapy. Conventional therapy at the time was some combination of neutral protamine Hagedorn (NPH) and regular insulin on a once- or twice-daily schedule, similar to human 70/30 insulin now. The intensive therapy group was given insulin by injection or an insulin pump three or more times daily, according to the master prescriber's best judgment, with the intent of lowering hemoglobin A1c as low as safely possible.

By the end of the study, those who were treated more intensely had less microvascular involvement but a lot more hypoglycemia.

Although the study design remains virtually unassailable, one lasting criticism is that hypoglycemia was defined as requiring the assistance of another person for recovery. This definition separates the perspectives of researchers—who see patients monthly and have resources to address this—from those of office clinicians, who, like myself, are left holding the bag when the paramedics arrive or who receive patient calls about more manageable symptoms with accelerated frequency.

After publication of this study, grand rounds were often funded by insulin manufacturers who brought the researchers to the podium with prepared data slides. Many clinicians accepted the findings, restructured insulin programs, and then backed off when the hypoglycemia resulted in considerable dissatisfaction among insulin users.

I inherited three of the participants after completion of the study, all of whom were assigned to the control group. All were advised to take multidose insulin on the basis of the results. One patient fractured his humerus during a hypoglycemic seizure shortly thereafter, adding me to the ranks of skittish clinicians functioning with ordinary resources. Not until the availability of analogue insulins was the DCCT method of administering intensive multidose insulin to lower the A1c to about 7% (another target that owes itself to that study) regularly implemented and acceptable to both prescriber and patient.

Contending With Hypoglycemia

Yes, hypoglycemia was and is the barrier to normalizing glucose. And it's not only with insulin. Chlorpropamide may have been among the most effective glucose-lowering pills, but it is now in the medical history books. Those who used it remember admitting patients with symptomatic low glucose to the hospital and running D5 for 3 days until the stuff was metabolized. We still contend with glyburide, which is used less frequently than it once was because of excess prolonged hypoglycemia.

When insulin is combined with other drugs that do not cause hypoglycemia when taken alone, insulin starts to do what insulin does. We get phone calls from patients. We get phone calls from emergency departments. We give consults for those whose misadventure got them admitted. We sometimes fly by the seat of our pants, doing gastric emptying scans or prescribing metoclopramide empirically to minimize the occurrence of this when we do not have the option of discontinuing insulin.

Hypoglycemia, nearly always a result of well-intentioned therapy—though on occasion a result of a suicide attempt—absorbs a lot of resources.

And we haven't yet included the hospitalized patients who have a much smaller therapeutic window for glucose control, particularly during critical illness. Inadequate insulin impairs recovery. Unfortunately, hypoglycemia during critical illness correlates with mortality—not just in the ICU but up to 90 days after discharge, when recovery seems to be almost a sure thing.

An Effort to Elucidate the Challenges of Hypoglycemia

It came as welcome news that a consortium was assembled in Europe to investigate the broad challenges that hypoglycemia imposes on the optimal care of patients with diabetes in a formal, scientific way.

The project is called Hypo-RESOLVE, with representation from 23 institutions.

Focusing on medical care in Europe, the task seems vast. Even some very basic aspects, including mechanisms by which hypoglycemia occurs in treated patients or identifying the most reliable predictors of hypoglycemia, still need to be elucidated. Outcomes of those affected are often in conflict, ranging from an annoying but necessary evil without lasting effects, as in the original DCCT data, to more ominous prognostic outcomes of the various studies on critically ill patients—something the Hypo-RESOLVE project will attempt to clarify.

Hypoglycemia is associated with the financial costs of visits to the emergency department, automotive mishaps, hospitalization, and disability, as well as the indirect cost of reticence to treat hyperglycemia in a suitably aggressive manner. The current financial and opportunity costs of hypoglycemia remain unknown, but a part of this project's mission is to better define and improve these costs.

This knowledge is vital to those of us who need to establish credibility and earn the trust of our patients with diabetes, and for health systems to plan services to optimize diabetes care. Acquiring the needed data requires a major investment of 26.8 million euros, which may seem like a high cost. However, the current damage from hypoglycemia around the world far exceeds that sum in both economic and human costs.

Those of us who contend with hypoglycemia as a daily barrier to the medical care we aspire to provide enthusiastically look forward to data that will help us do better for our patients.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: