Low-frequency electroacupuncture attenuates insulin resistance in a rat model of diet-induced maternal obesity

Maternal obesity is highly prevalent worldwide, conferring elevated risks of maternal/fetal complications and adult-onset disease in offspring. Obesity and pregnancy are both states of insulin resistance. Percutaneous electrical stimulation of skeletal muscle, better known as electroacupuncture (EA), mitigates insulin resistance in polycystic ovarian syndrome and diabetes, but its efficacy in pregnancy is unknown. We aimed to determine if EA can attenuate insulin resistance in pregnant Wistar rats with high-fat, high-sucrose (HFS) diet-induced obesity.

Nineteen HFS-exposed dams receiving 12-14 sessions of EA of rectus abdominis, tibialis anterior, soleus and gastrocnemius (3/15 Hz frequency, 10 mA intensity, 30 min) under isoflurane anesthesia (HFS+EA group) were compared with 19 untreated HFS-exposed dams receiving control anesthesia (HFS group) and 10 nonobese, nonpregnant controls (Ctrl group). On embryonic day (E)19 ± 1, insulin sensitivity was measured as the steady-state glucose infusion rate (GIR) during a gold-standard hyperinsulinemic-euglycemic clamp. Body composition, fasting blood glucose, fasting plasma insulin, serum corticosterone, blood pressure, pancreatic ß-cell mass and hepatic malonaldehyde/triglyceride content were determined.

Compared to the Ctrl group, the HFS group exhibited a decreased GIR (mean ± SEM = 13.3 ± 1.33 vs. 18.2 ± 1.37 mg/kg/min, p = 0.014)-reflecting insulin resistance-as well as increased pancreatic weight (1.31 ± 0.077 vs. 1.06 ± 0.111 g, p = 0.04) and increased hepatic malonaldehyde content (0.24 ± 0.041 vs. 0.12 ± 0.003 µM, p = 0.016). These changes were reversed by EA; compared to the HFS group, the HFS+EA group had a higher GIR (16.8 ± 1.01 mg/kg/min, p = 0.029), lower pancreatic weight (1.03 ± 0.051 g, p = 0.006) and lower hepatic malonaldehyde content (0.14 ± 0.016 µM, p = 0.019). Visceral adiposity, insulin, corticosterone and hepatic triglyceride content were increased in the model (p < 0.05) but not significantly impacted by EA. There were no measurable effects of pregnancy or EA on blood pressure or ß-cell mass, although there were significantly fewer small islets of Langerhans and significantly more large islets in HFS versus Ctrl groups. Raw fetal/placental weights did not differ; however, placental efficiency (fetal weight per unit placenta) was increased (4.09 ± 0.007 vs. 3.79 ± 0.006, p = 0.027) and litter size marginally decreased (median [interquartile range] = 16 [14-18] vs. 14 [11-16], p = 0.012) in HFS+EA versus HFS groups.

EA attenuated insulin resistance in this rat model of maternal obesity. The effects of EA on fetoplacental growth and litter size require further investigation, especially given the latter observation could signal a potential safety issue.