Premixed Human Insulin Ups Inpatient Hypoglycemia Rate in Type 2 Diabetes

October 22, 2015

Inpatient treatment with premixed human insulin produces glycemic control similar to that of basal-bolus analog regimens in those with type 2 diabetes, but the former led to far higher rates of hypoglycemia, according to a new study.

The difference in hypoglycemia rates was so dramatic that the prospective, open-label trial had to be stopped early, Virginia Bellido, MD, PhD, of the department of endocrinology at the Hospital Universitario de Cruces, Vizcaya, Spain, and colleagues report online October 12 in Diabetes Care.

"The use of premixed insulin has been associated with a higher rate of hypoglycemia in ambulatory patients. We hypothesized that a similar finding would be observed in the hospital setting," Dr Bellido told Medscape Medical News. She was shocked by the threefold higher rate of hypoglycemia with premixed insulin, "which is a much higher value than the reported prevalence in ambulatory patients," she noted.

The high rate of hypoglycemia is likely the result of poor oral intake and high illness severity, she noted.

Clinical guidelines from the Endocrine Society and the American Association of Clinical Endocrinologists/American Diabetes Association recommend the use of basal-bolus regimens in non–intensive-care inpatient settings.

However, many health professionals consider this approach to be difficult and inconvenient because of the high number of injections, among other factors, and premixed insulins are still commonly used in hospitals worldwide, the authors write.

Dr Bellido told Medscape Medical News, "Despite the simplicity and [ease of use], premixed insulin formulations should be avoided in the hospital setting. People in the hospital are sick and have poor nutritional intake, and many times it is impossible to predict when they will skip meals."

She added, "Our study indicates that a premixed insulin regimen should be used with caution, especially in patients with altered oral intake or with changing insulin requirements. This formulation may be acceptable in selected patients with normal renal function and good oral intake; but in general it should be avoided."

Threefold Difference in Hypoglycemia Rates

The study, conducted at two hospitals in Spain, had planned to enroll a total of 109 non–critically ill general medical and surgical inpatients with previously diagnosed type 2 diabetes. Half were to be randomized to receive a basal-bolus regimen using glargine once daily and glulisine before meals, while the other half were to be given premixed 30% regular and 70% NPH human insulin twice daily.

A total of 33 in the basal-bolus and 39 in the premixed groups completed the study before it was discontinued, with mean hospital stays of 17 and 24 days, respectively.

Both regimens resulted in "prompt and sustained improvement" in mean daily blood glucose levels, with no differences between the groups in the percentage of values between 80 and 180 mg/dL during their stays.

However, an interim analysis revealed that hypoglycemia (defined as capillary blood glucose level < 70 mg/dL) had occurred in 24% (eight) of the basal-bolus patients, compared with 64% (25) of the premixed insulin recipients (P = .001).

Because the rate of hypoglycemia in the premixed group exceeded the prespecified threshold of 50%, the study had to be stopped with only half of the planned subjects enrolled.

Basal-bolus insulin was also associated with less glycemic variability throughout the day (P = .037), the authors note.

Differences in hypoglycemia between the two groups were most pronounced before lunch and overnight. Overall, hypoglycemia was more common among patients who had been treated with insulin prior to hospitalization (62% vs 29% among those not receiving insulin previously, P = .04).

None of the patients lost consciousness or had seizures due to hypoglycemia. No associations were found between hypoglycemia rates and age, gender, diabetes duration, HbA1c, hospital department, or length of stay.

Daily insulin doses were similar at the start of treatment, but by the end the premixed group was receiving an average of 0.17-units/kg more insulin per day (P = .014).

The large difference in hypoglycemia occurred despite the fact that the premixed group was given snacks between breakfast and lunch, whereas the basal-bolus group was not, while total calories were kept the same. (Individuals in both groups were given bedtime snacks if their blood glucose was less than 140 mg/dL.)

"In our institutions, patients treated with premixed insulin regimens receive three main meals and snacks at midmorning and afternoon. One can assume that in the absence of snacks, such as in the US, the frequency of hypoglycemia may be even higher," Dr Bellido commented.

What About Premixed Analogs?

One study limitation, the authors acknowledge, is that they did not examine the effect of premixed insulin analogs, which might be associated with lower rates of hypoglycemia compared with the NPH/regular formulation.

"We do not know if they are safer than premixed human insulin. Unfortunately, no previous studies have investigated the use of these analogs in the hospital setting….Future studies should determine the safety and efficacy of premixed insulin analogs in the hospital setting and evaluate the difference between premixed analogs and basal-bolus regimens," Dr Bellido told Medscape Medical News.

This investigator-initiated study was supported by an unrestricted grant from Sanofi (Asturias, Spain). Dr Bellido has no relevant financial relationships; disclosures for the coauthors are listed in the paper.

Diabetes Care. Published online October 12, 2015. Abstract


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.