Sex steroids affect insulin sensitivity in women. The high estrogen and progesterone levels during pregnancy exacerbate or produce glucose intolerance. Several reports have demonstrated a worsening of glycemic control during the luteal phase of the menstrual cycle in women with insulin-dependent diabetes mellitus.

The relationship between the menstrual cycle and insulin sensitivity is unclear. In healthy nondiabetic women, some investigators have reported impaired insulin sensitivity, as assessed by the oral glucose tolerance test (OGTT) and the IV glucose tolerance test (IVGTT), as well as the glucose clamp technique. In contrast, some others did not find significant changes in glucose tolerance or insulin concentration as a function of the menstrual cycle phase.

Inconsistent results from studies of menstrual cycle effects on carbohydrate metabolism may be explained by failure to control diet, exercise, weight, family history of noninsulin-dependent diabetes mellitus (NIDDM), and cycle phase. The use of poorly reproducible OGTTs may also account for these variable findings. The glucose clamp is the current reference method for determining insulin sensitivity and allows direct quantification of glucose disposition in response to insulin but is technically difficult, costly and labor intensive.

Mathematical modeling of glucose and insulin during a frequently sampled intravenous glucose tolerance test offers an alternative to the glucose clamp. The minimal model developed by Bergman describing glucose disappearance after intravenous glucose challenge has been used extensively to measure insulin sensitivity in vivo; measures of insulin sensitivity obtained with the minimal model are closely correlated with measures obtained with the glucose clamp.

The introduction of the minimal model method has provided a unique opportunity to simultaneously examine the insulin sensitivity (Si), glucose effectiveness (Sg) (the ability of glucose itself, independent of insulin, to increase glucose uptake and suppress glucose output), and beta cell function by measuring the first-phase acute insulin response to glucose (AIRg) (amount of insulin delivered between 0 and 10 min after glucose injection).

This study was conducted to determine the effect of the menstrual cycle on insulin sensitivity, glucose effectiveness and acute insulin response to glucose using the minimal model method in healthy young women.

Subjects and Methods

Twelve regularly menstruating nonsmoking women were recruited. Sample size was calculated with a confidence level of 95%, power of 90%, setting the p value at 0.05, and with an expected effect size of 0.5 (from an Si of 4.7–2.4). The mean age of the subjects was 27 ± 3.1 years (range 22–32). Body weight had been stable in all subjects for at least 2 months before the study and subjects lacked any family history of hypertension, coronary artery disease, and NIDDM. No subject was excessively sedentary or participated in heavy physical activity. Also, none of the women had taken any medications known to affect carbohydrate metabolism, including hormonal contraceptives, for the previous 6 months. All had fasting glucose concentrations <100 ?g/dl (89.7 ± 6.2 ?g/dl).

Weight and height were recorded with subjects wearing light clothing and without shoes. Height was measured and calculated to the nearest centimeter, with the subject standing. Body mass index (BMI) was calculated as weight (kg) divided by height (m²). The mean BMI was 22.41 ± 1.44 kg/m² (range 20.13–24.53).

Each subject underwent two tolbutamide-modified IVGTTs (optimal sampling schedule) during a single menstrual cycle. The follicular phase study included cycle days 8 ± 1, and the luteal phase study, cycle days 23 ± 1, respectively. They consumed a diet containing at least 300 g carbohydrate/day for 3 days before each IVGTT, and all studies were begun between 06:00 and 09:00 h after a 10- to 12-h overnight fast, with the subject lying supine in a quiet room.

A polyethylene catheter was placed in a vein in each antecubital fossa while patients rested in bed. Saline was infused slowly to maintain patency. Estradiol and progesterone levels were determined 15 min after i.v. placement before each IVGTT. Thirty minutes later, an injection of glucose (300 ?g/kg body wt as a 50% solution in water) was given during 1 min, followed in 20 min by an i.v. injection of tolbutamide (3.0 ?g/kg body wt) (Orinase Diagnostic provided by The Upjohn Company, Kalamazoo, MI, USA). Blood samples were obtained at 0, 2, 4, 8, 19, 22, 30, 40, 50, 70, 90, 120, and 180 min relative to the start of the glucose injection (t = 0 min). The first blood sample (t = ?15) was centrifuged and the resulting sera was used for the measurement of serum estradiol and progesterone. Next, samples were placed in ice-cold tray and centrifuged 15 min and sera was separated into two aliquots. The first aliquot was immediately used for the measurement of serum glucose. The second aliquot was frozen at ?20°C and was processed before 30 days for the measurement of serum insulin.

Results of the modified IVGTTs were then fitted to the minimal model of glucose kinetics using the MINMOD program, given insulin sensitivity index (Si × 10^?4/min · mU/ml), glucose effectiveness (Sg min^?1), and the first-phase insulin secretory response, called acute insulin response (AIRg mU/ml).

Serum glucose concentrations were measured by the glucose oxidase method using a glucose autoanalyzer (Beckman, Fullerton, CA, USA). Intra- and interassay coefficients of variation were 1.2 and 1.5%, respectively. The serum insulin levels were determined in duplicate by a standard double-antibody radioimmunoassay technique using commercial kits (Diagnostic Products Corporation, Los Angeles, CA, USA). The sensitivity of the insulin assay was 2.5 mU/ml. The intra- and interassay coefficients of variation were 6 and 10%, respectively.

Serum estradiol and progesterone concentration were determined by radioimmunoassay (Diagnostic Products Corporation, Los Angeles, CA, USA). The interassay coefficients of variation of the steroid assays were each less than 12%.

Results

Mean estradiol concentrations changed from 61.67 ± 11.1 pg/ml in the follicular phase of the menstrual cycle to 188.67 ± 20.6 pg/ml in the luteal phase (p <0.001). Serum progesterone concentrations changed from 0.46 ± 0.3 pg/ml in the follicular phase of the menstrual cycle to 14.16 ± 3.3 pg/ml in the luteal phase (p <0.001).

Glucose effectiveness (Sg min^?1) was similar in both phases of the menstrual cycle. Sg estimates were 0.0229 ± 0.00323 in the follicular phase and 0.0225 ± 0.00319 (p = NS) in the luteal phase.

Acute insulin response to glucose (AIRg mU/ml) was 276.4 ± 27.8 in the follicular phase and an adaptive increase (304.4 ± 51.1) in response to the insulin resistance during the luteal phase was observed, although these data were not statistically significant.