Continuous subcutaneous insulin infusion does not correspond with pregnancy outcomes despite better glycemic control as compared to multiple daily injections in type 1 diabetes - Significance of pregnancy planning and prepregnancy HbA1c.

Objective: The aim of the study was to compare pregnancy outcomes with glycemic control, total increase in insulin requirement, and body weight gain in the women with Type 1 Diabetes Mellitus (T1DM) using continuous subcutaneous insulin infusion (CSII) or multiple daily injections (MDI).

Material And Methods: This was a single center retrospective observational study involving 209 pregnant Caucasian women. Among the study participants, 95 subjects were treated with MDI and 114 patients were using CSII therapy. The primary outcomes were pregnancy results, while secondary ones were HbA1c, increase in daily dose of insulin (DDI), and body weight gain.

Results: At baseline, the CSII users were older (P = 0.0373), they were diagnosed with T1DM at a younger age (P = 0.047), and more often planned pregnancy (P = 0.032). A majority of the women were classified as class D, according to the White classification. Among the CSII users, a significantly higher proportion of the subjects in class B was noted than in the MDI users, with no differences in the proportion of the remaining White classes. Prepregnancy HbA1c was insignificantly lower in the CSII group, however, a significantly higher proportion of the CSII users reached the target value of HbA1c (P = 0.008). A prepregnancy daily dose of insulin (both total and per kg of body weight), body weight, and body mass index (BMI) did not differ between the groups. The 1st and 2nd trimester HbA1c was lower among the CSII users (6.83 ± 1.38 vs 7.52 ± 2.11%, P = 0.01 and 6.17 ± 0.9 vs 6.57 ± 1.12%, P = 0.009, respectively), while the 3rd trimester HbA1c as well as the total change in HbA1c were comparable. Neither DDI and body weight in concecutive trimesters, nor their total gestational increase, differed between the groups. The rate of pregnancy loss, such as abortions, fetal and neonatal death did not differ between the groups. As regards composite pregnancy loss, prepregnancy HbA1c was 8.41%±2.81% among the MDI cohort vs 7.22%±1.31% in the CSII users (P = 0.517). No differences were found in the gestational age at delivery, the mode of delivery, neonatal birth weight, the rate of macrosomy, LGA or SGA. A higher Apgar score was noted among the CSII users (8.63 ± 1.63 vs 8.03 ± 2.49%, P = 0.047), however, the proportion of neonates with an Apgar score lower than 7 points was similar. In the women planning pregnancy, as compared to the subjects who did not, HbA1c was significantly lower in the 1st trimester, together with a significantly higher rate of the women achieving the target HbA1c value during planning as well as in the 1st trimester. In the group of women planning pregnancy, significantly lower 1st trimester HbA1c and composite outcome of pregnancy loss were observed in the CSII users vs the MDI treated women. Lack of pregnancy planning and a high HbA1c level in the 1st trimester were independent predictors of both LGA (OR = 4.99 [95%CI 1.12-21.0], P = 0.033 and OR = 3.02 [95%CI 1.19-7.65], P = 0.019, respectively) and macrosomia (OR = 8.43 [95%CI 1.36-51.93], P = 0.021 and OR = 5.47 [95%CI 1.77-16.87], P = 0.003, respectively).

Conclusions: The course of pregnancy and obstetric outcomes were not dependent on the mode of insulin delivery, but only on pregnancy planning and HbA1c in early pregnancy. Further studies are needed to explore more precise parameters describing both glycemic control in pregnant women as well as perinatal infant well-being.