Human insulin

Additional information

Three meta-analyses have looked at pooled data from randomized clinical trials on sex-differences and treatment outcome in patients with type 2 diabetes. The majority of patients included in the three meta-analyses are from the same studies.

Kautzky-Willer et al performed a meta-analysis published in 2014/2015 based on pooled data from six randomized clinical trial data (1349 men, 1251 women). The patients had inadequately controlled type 2 diabetes treated with oral antidiabetics. Insulin glargine or NPH insulin was added and evaluated after 24-36 weeks [1].

McGill et al performed a meta-analysis published in 2013 with pooled data from nine similarly designed randomized controlled studies. They evaluated 1651 adult men and 1287 women who were treated with insulin glargine or comparators (NPH insulin, insulin lispro, premixed insulin, oral antidiabetic drugs, or diet), respectively, for 24 weeks [2].

Owens et al performed a meta-analysis published in 2017 with pooled data from 16 randomized treat-to-target clinical trials including a total of 3188 patients (1680 men, 1508 women). Insulin glargine 100 U/mL was added to oral antidiabetic agents and evaluated after 24 weeks [3].

Pharmacokinetics and dosing

Kautzky-Willer et al found a higher insulin dose/kg after 24 weeks of treatment, (0.47 U/kg for women vs 0.42 U/kg for men) [1]. McGill et al found that at study end, men were receiving lower weight-adjusted insulin doses of insulin glargine vs comparators than (least squares mean difference for men  compared to women  -0.03 IU/kg) [2]. It is unlikely that these subtle differences have any clinical relevance.

Effects

Kautzky-Willer et al studied the effect of insulins on HbA1c, when insulin glargine and NPH insulin were added to oral antidiabetic drugs. Comparison of baseline demographics showed no effect of patient sex for glargine/NPH insulin use, HbA1c levels or fasting blood glucose levels (FBG). In the overall comparison, men had a lower HbA1c at study end (HbA1c 7.6% vs 7.8% in women), with a greater change in HbA1c for men (-1.36 vs -1.22 for women). Also in the overall comparison, a sex difference was seen in patients achieving a glycemic target of HbA1c 7% (53 mmol/mol) (32.99% of men vs 26.54% of women). A difference in change in FBG from baseline was seen (-4.33 mmol/L for women vs -3.93 mmol/L for men) [1].

McGill et al found no difference between men and women in the likelihood of achieving a glycemic target of HbA1c 7% for patients receiving comparator treatment, nor in the overall comparison (comparator treatment or insulin glargine) between men and women. For patients receiving insulin glargine, women were less likely than men to achieve a glycemic target of HbA1c 7%, (54.5% vs 60.8%), with a mean HbA1c of 6.9% (0.9) for men vs 7.1% (0.9) for women (unknowen if significant). Men receiving NPH insulin were similarly more likely to reach HbA1c 7% than women (OR 1.59 (1.06-2.40)). No difference in FBG was seen at baseline. At study end, women receiving insulin glargine or comparators had a 3.1 mg/dl (0,02 mmol/L) greater reduction from baseline in FBG compared with men [2].

Owens et al found that women had an overall slightly smaller reduction in HbA1c from baseline to week 24, reduced by 1.4% reaching 7.3% compared to a reduction by 1.6% for men, reaching HbA1c 7.1% (unknown if significant). No sex differences were found in reductions in HbA1c for patients using SU only prior to initiation of insulin treatment [3].

It is unclear if the subtle differences described above have any clinical implication.

Adverse effects

The three above mentioned meta-analyses all found a higher frequency of hypoglycemia, defined as blood glucose levels 70 mg/dl (4 mmol/l), in women than in men.

Kautzky-Willer et al found that a higher proportion of women overall (oral antidiabetic drugs (OAD) + insulin glargine (Lantus) or NPH insulin) experienced a severe hypoglycaemic episode (3.28% of women compared to 1.85% of men). Severe nocturnal hypoglycaemia was seen overall in 2.24% of women vs 0.59% of men  [1].

McGill et al found that women overall (treated with insulin glargine or comparator treatment NPH insulin, insulin lispro, premixed insulin, oral antidiabetes drugs, or diet) were more likely to experience at least one severe hypoglycemic event than men (OR 1.85, 95% CI 1.03-3.34 ) [2].

Owens et al found an overall (OAD + insulin glargine) incidence of hypoglycemia of 44.9% for women vs 41.3% for men, and an event rate (events per patient-year) of 4.8 vs 3.7 [3].

One study retrospectively analysed four large trials including 713 patients (47% women) with type 2 diabetes. Insulin glulisine was added to a basal treatment with OAD and insulin glargine. Female gender was identified as a predictor of nocturnal and symptomatic hypoglycemia (OR 1.82; 95% CI 1.07-3.11 and OR 1.89; 95% CI 1.31-2.78) [4].

Reproductive health issues

Concurrent administration of human insulin and oral contraceptives may decrease the effect of human insulin [5]. Regarding drug-drug interactions aspects, please consult Janusmed Interactions (in Swedish, Janusmed interaktioner).

Human insulin can be used in pregnant women. Insulin requirements may change during pregnancy, and quickly return to normal after delivery [5]. Regarding teratogenic aspects, please consult Janusmed Drugs and Birth Defects (in Swedish, Janusmed fosterpåverkan).

Other information

Two observational studies from the 1990’s with a total of 418 patients (43 men, 374 women) reported intentional insulin omission among 1/3 of women to control their weight [6,7].

In a retrospective study in 124 women, perimenstrual changes in selfreported glucose concentrations were found in 61%. Use of oral contraceptives did not diminish variability in blood glucosis [8]. In another study based on questionnaires (406 women) 67% of the participants reported changes in blood glucose levels or glycosuria premenstrually and 70% during the menstrual phase. Those with more cravings had larger elevations in blood glucosis levels suggesting that giving in to cravings might cause the changes [9].