Effective Postprandial Glucose Management in Pregnancy with Type 1 Diabetes

Maintaining good glucose control before conception and throughout pregnancy is critical for obtaining ideal maternal and fetal outcomes. The risks of hyperglycemia to a pregnant mother include worsening of diabetic complications, euglycemic diabetic ketoacidosis, preeclampsia, preterm labor, and poor obstetrical outcomes (e.g., cesarean section). ^{1 –5} The risks of hyperglycemia to a fetus or infant include miscarriage or stillbirth, congenital anomalies, large-for-gestational age (LGA, birth weight >90th percentile for gestational age), macrosomia (birth weight >4 kg or >4.5 kg), small-for-gestational age (birth weight <10th percentile for gestational age), birth injuries, cardiac dysfunction, neonatal hypoglycemia, respiratory distress syndrome, and hyperbilirubinemia to name a few. ^{1 –6}

The American Diabetes Association (ADA) lowered its recommended target glycated hemoglobin A1C (A1C) level for women with preexisting diabetes to 6%–6.5% during pregnancy, but <6% “may be optimal as pregnancy progresses,” if targets can be met without hypoglycemia. ⁷ Furthermore, optimal glycemic targets, if achievable without significant hypoglycemia, are fasting glucose levels of ≤90 mg/dL and postprandial glucose levels of ≤130–140 mg/dL at 1 h and ≤120 mg/dL at 2 h. ^7,8

Achieving near-normal glucose levels in pregnancy and maintaining those levels for 40 weeks of gestation is, understandably, a large undertaking. A number of improvements to insulins and diabetes technologies have helped countless women approach or reach these glycemic targets. Rapid-acting insulin analogues have shown favorable results of reducing hypoglycemia and hyperglycemia with similar maternal and fetal outcomes compared with regular human insulin and neutral protamine Hagedorn (NPH). ^{9 –11} Continuous subcutaneous insulin infusion (CSII) therapy, ^12,13 continuous glucose monitoring, ^14,15 and sensor-augmented pump therapy ^16,17 have all shown benefits in diabetes associated with pregnancy. For a more detailed overview please see a recent review article. ¹⁸

Evidence suggests that it is not just overall glycemic control (reflected in the A1C) but also glycemic variability that plays a role in pregnancy outcomes. ^8,19 De Veciana et al. found that in a trial where women with gestational diabetes were randomized to pre- or postprandial glucose monitoring, the risk for macrosomia and neonatal hypoglycemia may be better predicted by postprandial glucose levels than by preprandial levels. ¹⁹

Maresh et al. prospectively assessed pregnancy outcomes in 725 women with type 1 diabetes mellitus (T1DM) at 26 and 34 weeks gestation. Women were categorized into quintiles of A1C ranging from <6% to >7.5%, with pregnancy outcomes being compared with the <6% group. The odds of having preeclampsia, LGA, neonatal hypoglycemia, severe hyperbilirubinemia, and preterm delivery all increased starting at an A1C of 6%–6.4% in a dose-dependent manner based on A1C quintile. ²⁰ Moreover, increases in preprandial average glucose more than 90 mg/dL at 26 and 34 weeks gestation were associated with increased risks for LGA, severe neonatal hypoglycemia, severe hyperbilirubinemia (34 weeks only), and preterm delivery, whereas increases in average 1-h postprandial glucose more than 108 mg/dL at 26 and 34 weeks gestation increased the risk for LGA. ²⁰ These data elucidate the importance of optimizing glucose levels throughout the day, and not just fasting levels.

In this issue of Diabetes Technology & Therapeutics, Bongiovanni et al. report data from a retrospective study in four Italian diabetes pregnancy centers of 101 women with T1DM who were followed throughout pregnancy with weekly or biweekly assessments of insulin use (basal and bolus) and carbohydrate-to-insulin (CHO/I) ratios. ²¹ Most women were using CSII therapy before conception (68/101, 67.3%), whereas the remaining subjects initiated CSII therapy in the first trimester (on average). Dietary recommendations included 45%–50% of total calories from carbohydrates daily, 20% from protein, and 30%–35% from fat (divided across meals and snacks).

In this study, the ADA guidelines were used to set glycemic targets as mentioned. Median CHO/I ratios for 2-week intervals were reported (6 values each interval) as measured ratios, as well as calculations for insulin resistance (CHO/IRI) based on the ratios of 300/24-h insulin need (300 formula) and 500/24-h insulin need (500 formula) using self-monitored blood glucose values. Measured ratios were only considered if fasting glucose values were 70–100 mg/dL and 1-h postprandial values were 100–145 mg/dL.

In this study, total daily insulin requirements increased nearly 1.6-fold from 5–6 weeks to 37–38 weeks gestation, with the majority of the increase coming from bolus insulin (twofold change). Average weight gain was 35.2 pounds among participants, which is consistent with the Institute of Medicine guidelines for weight gain in pregnancy ²² for women with a body mass index (BMI) in the normal range (average preconception BMI 23.5 kg/m² in this study). Linear regression analysis suggested that the 300 and 500 formulas' optimal uses varied by meal and gestational age (Table 1). The CHO/I ratios became more aggressive over time until the 36th week, after which point they became more conservative. The median CHO/I ratios were the most aggressive at breakfast (peak at 33–34 weeks), with relatively equal ratios at lunch and dinner from 5 to 38 weeks gestation (peaks at 31–32 weeks for lunch and 33–34 weeks for dinner). ²¹

The relative increased insulin resistance in the morning and the disproportionate increase in bolus insulin compared with basal insulin throughout the pregnancy have been observed by others, ²³ as mentioned by the authors. There are also studies that have helped uncover the pathophysiology surrounding postprandial hyperglycemia in pregnancy. Murphy et al. enrolled 10 women with T1DM for 24 h at a time in early (12–16 weeks) and late pregnancy (28–32 weeks). ²⁴ Hybrid closed-loop CSII therapy was initiated for each time period specified, such that premeal insulin boluses were calculated according to glucose level, anticipated carbohydrate consumption, and insulin:carbohydrate ratio for breakfast and dinner meals. A glucose tracer was infused ([6,6-²H₂]glucose). ²⁴

Among these 10 women, there was a significant increase in prandial insulin boluses with advanced gestation, along with increased hepatic insulin resistance and decreased peripheral insulin sensitivity. The quickest appearance of postprandial glucose was after breakfast, though unaffected by stage of gestation. Although fasting and postprandial glucose appearance rates did not change, there were delays in postprandial glucose disposal with advanced gestation (103 ± 17 vs. 125 ± 21 min for breakfast early vs. late gestation; 112 ± 22 vs. 142 ± 34 min for dinner early vs. late gestation, p = 0.003), which was, in part, because of delayed insulin absorption (mean delays of 26 and 32 min for dinner and breakfast, respectively, from early to late gestation, p = 0.0002). ²⁴

Data from Murphy's study and from a second study in 22 women who were encouraged to walk postprandially at 19–23 weeks gestation were compiled to examine pharmacokinetic properties of Aspart insulin in pregnancy. ²⁵ Moderate-intensity physical activity reduced average time-to-peak insulin concentration from 55 to 40 min after breakfast. In addition, Aspart absorption was delayed ∼50% at 38 versus 8 weeks gestation. ²⁵ Taken together, prandial glucose excursions and duration of excursions both increase as pregnancy progresses, thus it is advisable to premeal bolus earlier and earlier.

Bongiovanni et al. have contributed evidence to the growing body of literature that describe the necessary increases in insulin over time from mid to late gestation. They suggest that formulas may help guide therapy for prandial bolusing. ²¹ It is also important to take into account the duration of time between premeal boluses and carbohydrate consumption with advanced gestation. ^24,25 As bolus insulin increases more than basal insulin with progression of pregnancy, innovative approaches to improving postprandial hyperglycemia should be examined carefully. The high stakes of maternal and fetal outcomes warrant more research into effective strategies for both achieving and maintaining tight glucose control throughout pregnancies complicated by diabetes.