Deprescribing Antihyperglycemics

Researchers at the deprescribing project based at the Bruyère Research Institute in Ottawa, Canada, have just published another deprescribing algorithm, this one focused on antihyperglycemic agents.^[1] The aims is to guide healthcare professionals in stopping, switching, or lowering the dose of these drugs in patients at risk for hypoglycemia or other antihyperglycemic adverse effects or in whom the drug’s benefit is uncertain due to frailty, dementia, or limited life expectancy. Lead author Barbara Farrell, PharmD, spoke with Medscape about the guideline and its recommendations.

“The intensity of glycemic control needed in older people is quite controversial, with different diabetes guidelines recommending different targets,” explained Dr Farrell. Around 5 years ago, treatment guidelines, which had previously favored a glycated hemoglobin (A1c) target of <7% (53 mmol/mol) for most people, set less stringent targets for the elderly (age ≥65 years). Many current international and national guidelines now recommend approximate treatment goals of <7.5% (58 mmol/mol) in healthy older adults and <8.5% (69 mmol/mol) in the very frail elderly.^[2,3,4,5,6] These changes in targets result from clinical trials showing that compared with conventional glycemic control, intensive control did not significantly reduce all-cause or cardiovascular mortality.^[7] It did, however, increase the risk for hypoglycemia and serious adverse events,^[8,9] especially in elderly patients.^[10,11]Hypoglycemia is associated with cardiovascular events, cognitive impairment, fractures, death, and reduced quality of life.^[12,13] It is a leading cause of emergency department visits in older adults in the United States.^[14]Hospitalization for hypoglycemia is associated with a poor prognosis.^[15]

Need for New Guidelines

Despite the new guidelines and “Choosing Wisely” campaigns specifically cautioning against intensive glycemic control in the elderly,^[16,17] recent studies in the United States and Canada have shown that many older patients with diabetes are still being treated to A1c <7%.^{[18,19,20,21,22]}

“In the United States and Canada, the concept of treating to specific number targets has been very entrenched in the medical communities and among the public, so it is a difficult thing to change,” Dr Farrell stressed. “The diabetes guidelines all talk about how to start these drugs, and some of them discuss how to adjust doses for kidney function or for age, but they don’t specifically address how to reduce a dose or how often to monitor while you are reducing it. That is why we developed the deprescribing guideline, to fill that gap,” Dr Farrell explained. “For patients with low blood sugar, in particular the frail elderly, we wanted to be able to provide guidance to prescribers about when it might be appropriate to start reducing doses or just stopping some of those medications that can contribute to low blood sugar. What we are most concerned about is ensuring that people are switched off those medications and either continue on their remaining medications or with safer ones substituted. We need to be looking at safety as an increasing priority over the potential long-term benefits of keeping blood sugar very low. It is challenging because every person is different in terms of identifying when the balance of risk of medication versus benefit changes.”

The guideline does not set specific glycemic targets, referring to Canadian Diabetes Association and other, similar guidelines that address the need for less stringent targets in the elderly.

Which Drugs to Stop?

“There was a particular worry about glyburide, the long-acting sulfonylurea that frequently causes hypoglycemia,” Dr Farrell said. This usually happens because glyburide was started when patients were younger, and they have been taking it for many years, she explained. Switching glyburide to short- or long-acting gliclazide may reduce but will not eliminate the risk for hypoglycemia. Options other than a sulfonylurea should be considered.

“The other medication that can cause hypoglycemia is insulin, so cut back on the doses if the blood sugar is too low,” Dr Farrell added. The highest risk is with neutral protamine Hagedorn (NPH) insulin, and the guideline advises switching NPH insulin or mixed insulin to insulin detemir or glargine to reduce nocturnal hypoglycemia.

The guideline also lists the antihyperglycemic medications with no or low risk for hypoglycemia (Table).

Table. Medications With Low or No Risk for Hypoglycemia

Tapering and Monitoring

“There is no evidence that one tapering approach is better than another,” Dr Farrell noted. “With antihyperglycemic drugs, the only adverse effect from stopping is a rise in blood sugar that needs to be monitored. So if we have a patient who has very low blood sugar and no real problems with high blood sugar, we would just stop the sulfonylurea. If it’s a low dose of insulin, we would stop that. In other cases, we might gradually reduce the dose, either because we think the patient would still benefit from having the drug even at a lower dose, or sometimes the patient is reluctant to just stop suddenly.”

Frequency of monitoring the effects of deprescribing these drugs is highly individualized, depending on the patient’s blood sugar levels and on the drugs that have been changed, Dr Farrell added. Monitoring need not be done daily unless insulin is being deprescribed. “If we are stopping glyburide or switching from glyburide to gliclazide, we might check once a week for a couple of weeks. Typically after most antihyperglycemic drugs are stopped, changes in blood glucose will be seen within 1 or 2 weeks,” she said.

Other Causes of Hypoglycemia

Another unique aspect of the new guideline is the inclusion of different situations that can contribute to hypoglycemia, such as taking insulin and then not eating, taking other drugs that cause hyper- or hypoglycemia, or drug interactions with antihyperglycemic medication. “Sometimes we see that a patient has recently stopped a drug that causes hyperglycemia, but the dose of the diabetes drug has not been reduced,” she explained. “For example, someone who is taking high-dose prednisone for a short period of time might experience a steep rise in blood sugar, and the dose of insulin or other antihyperglycemic drug is increased to cope with that. Then the prednisone is stopped, and the patient’s blood sugar plummets. We might simply avoid the particular drug that causes that drug interaction, or we would probably consider lowering the dose of the antihyperglycemic drug.”

Discussion With Patients and Families

One of the goals of the guideline is to encourage healthcare providers to discuss the rationale and benefits of deprescribing with patients and their families. “One of the biggest challenges is to explain the targets and the need to focus on safety versus not always focusing on preventing the long-term problems of diabetes,” Dr Farrell said. “This can be especially difficult with older people who may have had diabetes for 30 years, and it may seem as though suddenly, out of the blue, we’re not concerned about their blood sugar any longer. It is not an easy conversation to have, and it can be time-consuming,” she acknowledged.

“When these medications are first prescribed to a newly diagnosed person, and targets are discussed, it would be helpful to also explain that when they are older or if they start having hypoglycemic episodes, these targets may change,” she suggested. “We would establish that very intensive control is important in a younger person to prevent long-term complications but that as they get older we will revisit these targets, so they would know that modifying their targets as they get older is a normal part of the program.”

Updating the Guidelines

Dr Farrell pointed out that the systematic review of evidence done for preparation of the new guideline^[23] identified only two relevant studies of deprescribing antihyperglycemic agents.^[24,25] Both studies concluded that the approach is safe and feasible, but the quality of the evidence was low, the guideline authors commented. “We would benefit from having more studies of deprescribing so that we can get a better understanding about whether there is an optimal approach to tapering, monitoring, and follow-up,” Dr Farrell said. “Our eventual goal is that all prescribing guidelines will include deprescribing sections, so that specialists, family physicians, and all other healthcare professionals are on board with the approach.”