Parity Increases Insulin Requirements in Pregnant Women With Type 1 Diabetes

Parity Increases Insulin Requirements in Pregnant Women With Type 1 Diabetes Abstract Context Tight glycemic control throughout pregnancy in women with type 1 diabetes is crucial, and knowledge about which factors that affect insulin sensitivity could improve the outcome for both mother and offspring. Objective To evaluate insulin requirements in women with type 1 diabetes during pregnancy and test whether parity affects insulin requirements. Design Observational cohort study consisting of women with type 1 diabetes who gave birth at Aarhus University Hospital, Denmark, from 2004 to 2014. Main Outcome Measure Daily insulin requirement (the hypothesis that parity could affect insulin resistance was formulated before data collection). Results A total of 380 women with a total of 536 pregnancies were included in the study. Mean age was 31.1 years, and prepregnancy hemoglobin A1c was 60 mmol/mol. Parity was as follows: P0, 43%; P1, 40%; P2, 14%; and P3+4, 3%. Insulin requirements from weeks 11 to 16 decreased significantly by 4% (P = 0.0004) and rose from week 19 to delivery with a peak of 70% (P < 0.0005) at weeks 33 to 36. Overall, insulin requirements increased significantly with parity. The unadjusted differences between P0 and P1, P2, and P3+4 were 9% (P < 0.0005), 12% (P < 0.0005), and 23% (P < 0.0011), respectively. After adjustment for confounders, differences were 13% (P < 0.0005), 20% (P < 0.0005), and 36% (P < 0.0005). We also observed an adjusted difference between P1 and P3+4 of 20% (P < 0.0012). Conclusions The data show changes in insulin requirements from week to week in pregnancy and indicate that insulin requirements increase with parity. This suggests that the patient's parity probably should be considered in choosing insulin dosages for pregnant women with type 1 diabetes. Alterations in insulin requirements and insulin resistance during pregnancies complicated by diabetes constitute a challenge for both patients and clinicians. Pregnancies in women with type 1 diabetes are associated with an increased risk for congenital malformations, obstetric complications, and neonatal morbidity—a risk that is directly correlated to glycemic control immediately before and during pregnancy (1, 2). Thus, tight glycemic control throughout pregnancy is crucial and can be achieved only if a successful collaboration between patients and clinicians is established. To allow patients and clinicians to act preventively, knowledge about alterations in insulin requirements over time during pregnancy is essential to enable implementation of appropriate insulin dosage preemptively. During normal pregnancy, high levels of diabetogenic placental hormones lead to a physiological decrease in peripheral insulin sensitivity as pregnancy proceeds (3). Another factor that could be affecting insulin requirement is parity. Some studies have reported an increase in the incidence of type 2 diabetes with increasing parity (4–6), whereas others have found no such association (7–10). Several of these studies have included a limited number of patients, and the influence of parity on insulin requirement during pregnancy in women with type 1 diabetes remains uncertain. In type 2 diabetes and gestational diabetes, increasing parity may increase insulin requirement because of exhaustion of the pancreatic β cells due to the high insulin need during pregnancy. In type 1 diabetes, however, there is essentially no insulin production in the β cells. Any increase in insulin requirements with increasing parity would be determined by insulin resistance in target tissues probably related to such factors as age, increasing body mass index (BMI), or altered hormone secretion from the placenta. The progressions in insulin requirements during pregnancy vary considerably between individuals. Some women experience subtle changes, whereas others have to triple or quadruple insulin doses late in pregnancy. We and others (11–13) have studied insulin resistance during pregnancy. Insulin resistance can be estimated in different ways; analyzing insulin requirements during pregnancy is one approach. This has previously been done only in relatively small cohorts of patients with type 1 diabetes (11–14). In the current study, we evaluated insulin requirements during 536 pregnancies in 380 patients with type 1 diabetes. Using the same dataset, we examined whether parity affects insulin requirements. Patients and Methods A retrospective observational cohort study was conducted. The cohort consisted of women with type 1 diabetes who gave birth at Aarhus University Hospital, Denmark, between January 2004 and December 2014. All pregnant women with type 1 diabetes residing in Central Denmark Region are treated at Aarhus University Hospital, which is a tertiary center with highly specialized service. The Central Denmark Region has 1.3 million residents and is very diverse in terms of socioeconomic status covering both rural and urban areas. The ethnicity of the cohort was predominantly white (15), but the ethnicity of each participant was not recorded. Exclusion criteria were multiple pregnancy, stillbirth, termination of pregnancy, and spontaneous abortion. The women were seen at the outpatient clinic for pregnant women with diabetes at Aarhus University Hospital or contacted by telephone every 2 to 4 weeks from gestational week 4 to week 28 and every 1 to 2 weeks from week 28 to delivery. A diabetologist adjusted insulin dose according to current hemoglobin 1Ac (HbA1c) and the patient's own daily recordings of fasting and postprandial glucose measurements. The target HbA1c values for pregnant women with type 1 diabetes were according to the Danish National Guidelines: 48 mmol/mol (6.5%) before gestational week 20 and 37 mmol/mol (5.6%) after gestational week 20 (16). Diabetes regulation was managed by self-monitoring of glucose and insulin adjustments, aiming at capillary plasma glucose levels of 4 to 6 mmol/L preprandial and 4 to 8 mmol/L 1.5 hours postprandial. An obstetrician or a sonographer obtained serial ultrasound scans to monitor fetal growth, and a dietitian gave nutritional advice. From the patients’ medical records, we obtained the following data at every visit: daily insulin requirements, current HbA1c, and weight. We also extracted the following prepregnancy data: weight, height, parity, last known daily insulin dosage before pregnancy, last known HbA1c (not older than 6 months), and number of years since diabetes onset (duration of diabetes). Relevant data on the offspring, such as gestational age at delivery, sex, and birth weight, were also obtained. Daily insulin requirement was calculated by adding long-acting and short-acting insulin. The mean daily insulin requirement was determined at the following time intervals: prepregnancy, weeks 5 to 10, weeks 11 to 14, weeks 15 to 18, weeks 19 to 22, weeks 23 to 28, weeks 29 to 32, weeks 33 to 36, and weeks 37 to 40. Our primary outcome was mean daily insulin requirement (IU) at the above-mentioned time intervals. Secondary outcomes were (1) percentage change from prepregnancy insulin requirement at the above-mentioned time intervals, (2) the effect of parity on mean daily insulin requirement for the whole pregnancy, and (3) the effect of fetal sex on insulin requirement. The Danish Health Authority (no. 3-3013-625/1) and the Danish Data Protection Agency (no. 1-16-02-347-14) approved our study. Statistics The Biostatistical Advisory Service at the Faculty of Health, Aarhus University, helped perform all statistical analyses by using the statistical software Stata 13 (Stata Corp., College Station, TX). Continuous data are expressed as mean with standard deviation and categorical variables as percentages. Repeated-measurement analysis of variance was used to compare the paired data between the groups. Both the repeated measurements in every pregnancy and the repeated measurements for each of the woman's pregnancies were taken into account by using nested random effects in a mixed model. Measurements were analyzed on log scale, and results were back-transformed to obtain relative comparisons. Model validation was performed by visual inspection of residuals, fitted values, and random effect estimates, which did not give cause to reject the model. For the subgroup analyses, paired t test was used. Statistical significance was defined as P < 0.05. Five different models were made to adjust for factors that could affect insulin requirement: (1) unadjusted; (2) adjusted for prepregnancy BMI; (3) adjusted for prepregnancy BMI and age; (4) adjusted for prepregnancy BMI, age, and prepregnancy HbA1c; and (5) adjusted for prepregnancy BMI, age, prepregnancy HbA1c, and duration of diabetes. Results We reviewed 583 pregnancies. Exclusion criteria were multiple pregnancy (n = 34), stillbirth (n = 3), termination of pregnancy (n = 5), spontaneous abortion (n = 2), and missing data (n = 3). Thus, 380 women with a total 536 pregnancies were included in the study. Several of the women gave birth more than once within the time period. Two hundred thirty-six women (62%) gave birth once, 132 women (35%) gave birth twice, and 12 women (3%) gave birth three or four times. Table 1 presents the clinical features of each pregnancy in the cohort. When we divided the cohort into parity groups (P0, P1, P2, and P3+4), we found no statistical difference in prepregnancy BMI or prepregnancy HbA1c between the groups. Table 1. Demographic Characteristics of Patients With Type 1 Diabetes Included in Study Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  A total of 380 women gave birth 536 times. Two hundred thirty-six women (62%) gave birth one time, 132 women (35%) gave birth two times, and 12 women (3%) gave birth three times. View Large Table 1. Demographic Characteristics of Patients With Type 1 Diabetes Included in Study Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  A total of 380 women gave birth 536 times. Two hundred thirty-six women (62%) gave birth one time, 132 women (35%) gave birth two times, and 12 women (3%) gave birth three times. View Large We compared duration of diabetes among the groups and, as expected, found that the P0 group had statistically significantly shorter diabetes duration, at 14.4 years, than the P1 group, at 15.6 years, and P2 group, at 16.7 years. The P3+4 group, however (n = 16), had an even shorter duration, at 12.8 years, probably due to the small number of included individuals and the influence of a few extremes affecting the average. Mean age overall was 31.1 years, and age increased, also as expected, with parity: P0, 29.1 years; P1, 32.0 years; P2, 33.6 years; and P3+4, 36.7 years. Mean HbA1c was 55.0 mmol/L [95% confidence interval (CI), 54.7 to 55.4 mmol/L], 48.3 mmol/L (95% CI, 48.1 to 48.6 mmol/L), and 48.3 mmol/L (95% CI, 48.0 to 48.5 mmol/L) for first, second, and third trimesters, respectively. Mean daily insulin requirements at the different time intervals for all included pregnancies (n = 536) are shown in Fig.1. Figure 1 also shows the mean daily insulin requirements in percentages relative to prepregnancy levels. Insulin requirements at weeks 11 to 16 decreased significantly by 4% (95% CI, 2% to 6%; P = 0.004) and rose significantly from week 19 to delivery, with a peak at weeks 33 to 36 at 70% (95% CI, 65% to 75%; P < 0.0005), relative to prepregnancy levels. After week 37, the insulin requirement decreased and terminated at 66% (95% CI, 59% to 72%) relative to prepregnancy levels. Figure 1. View largeDownload slide Mean daily insulin requirements at the different time intervals for all included pregnancies (with 95% CI error bars). The dashed line represents the prepregnancy insulin requirement, and numbers in percentages indicate the increase or decrease in percentage relative to prepregnancy levels. *Statistical significance (P < 0.05). Figure 1. View largeDownload slide Mean daily insulin requirements at the different time intervals for all included pregnancies (with 95% CI error bars). The dashed line represents the prepregnancy insulin requirement, and numbers in percentages indicate the increase or decrease in percentage relative to prepregnancy levels. *Statistical significance (P < 0.05). When comparing mean daily insulin requirements for the entire pregnancy, we found that insulin requirements increased with parity. Figure 2 shows how mean daily insulin requirement developed during pregnancy for P0, P1, and P2. Figure 2. View largeDownload slide Difference in mean daily insulin requirement at the different time intervals between parity groups. Figure 2. View largeDownload slide Difference in mean daily insulin requirement at the different time intervals between parity groups. The unadjusted differences between P0 and P1, P2, and P3+4 were 9% (95% CI, 4% to 14%; P < 0.0005,), 12% (95% CI, 4% to 20%; P < 0.0005), and 23% (95% CI, 5% to 44%; P = 0.011), respectively. To avoid any confounding, we adjusted for age, prepregnancy BMI, prepregnancy HbA1c, and duration of diabetes. After adjustment for all four variables, the differences between P0 and P1, P2, and P3+4 were 13% (95% CI, 8% to 18%; P < 0.0005), 20% (95% CI, 12% to 30%; P < 0.0005), and 36% (95% CI, 17% to 38%; P < 0.0005), respectively. An additional significant difference between P1 and P3+4 in mean daily insulin requirements of 20% (95% CI, 4% to 39%; P = 0.012) was found. Table 2 presents all estimates, both unadjusted and adjusted. Table 2. Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  a Statistically significant (P < 0.05). View Large Table 2. Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  a Statistically significant (P < 0.05). View Large To confirm our results, we performed a subgroup analysis that included only women who were in the cohort more than once (n = 147). We compared mean daily insulin requirements in the woman's first pregnancy in the cohort with her subsequent pregnancies. We found a significant difference in insulin requirement between the pregnancies of 11% (95% CI, 9% to 13%; P < 0.0005). This was across all parities. Table 3 shows all estimates and 95% CIs for comparisons of the parities. Table 3. Subgroup Analysis: Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  a Statistically significant (P < 0.05). View Large Table 3. Subgroup Analysis: Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  a Statistically significant (P < 0.05). View Large The sex ratio at birth (live male newborns/total live newborns) in our study was 0.496, and we found no difference in mean daily insulin requirements when comparing women with male offspring and female offspring (data not shown). Discussion The fall and rise in insulin requirements during pregnancy are well known. The S-shaped curve has been reported before (11, 14), but here we have clearly confirmed it with a large number of participants. The decrease in insulin requirements around weeks 10 to 14 is important to acknowledge in order to avoid hypoglycemia in early pregnancy. With our results, we have attained a more detailed estimate of this decrease, and we hope this will help lower the incidence of potentially dangerous hypoglycemia in early pregnancy (17, 18). We have also identified the time span within which the rapid increase in insulin requirements occurs, thereby allowing clinicians to reduce hyperglycemia and maintain steady and tight glycemic control during the last part of the pregnancy. In addition, we have established that after the maximal insulin requirement levels are reached around week 37, a minor decline of 4% is observed. It is well known that a decline in insulin requirement of ≥15% in late pregnancy can be a sign of placental insufficiency (19). The observed 4% decline is probably a part of the normal type 1 diabetes pregnancy and should not by itself lead to induction of birth. Finally, we established that parity is a factor that should be considered when treating pregnant women with type 1 diabetes. We have found a statistically and clinically significant difference in mean daily insulin requirement during pregnancy between the P0 group and P1, P2, and P3+4 groups at 9%, 12%, and 23% respectively. Because this difference could be due to a difference between the groups in BMI, age, or prepregnancy HbA1c, we also presented an adjusted model, which increased the differences to 13%, 20%, and 36%, respectively. In addition, after adjustment we found a significant difference between the P1 group and P3+4 group of 20%. In our subgroup analysis, where we compared mean daily insulin requirement in the woman's first pregnancy in the cohort with the following pregnancy, we were able to confirm these findings, thereby diminishing the effect of variation between individuals. This further strengthens the concept that parity predicts increased insulin requirements as this clearly shows how women who had given birth several times in our cohort increased their individual total insulin requirement during pregnancy with each pregnancy. Furthermore, when we looked at the insulin requirements before pregnancy, we found no difference between parity groups, indicating that the difference observed only concerns pregnancy. The mechanisms behind the parity-associated increase in insulin requirements are unknown. Studies have shown that anthropometric measures increase with parity and that this could influence insulin resistance (20, 21). Although we did not find any difference when we adjusted for BMI, other anthropometric measures (e.g., subcutaneous fat or intra-abdominal fat) could increase with parity and thereby influence our findings (22). Sex of the offspring could also potentially affect the insulin resistance. The sex ratio at birth in the general population is 0.515 (23), and in a cohort like ours, Garcia-Patterson et al. (24) found a sex ratio at birth at 0.509. Our sex ratio at birth at 0.496 was, as expected, not significantly different from any of the cohorts mentioned. Male fetuses are known to grow faster in the womb, be larger at birth, and have more efficient placentas (25–27). Moreover, women carrying a male fetus have a higher risk of developing gestational diabetes (28). We thus hypothesized that male fetuses for this reason could make the mother more insulin resistant during pregnancy, but our analysis did not show any difference in insulin requirement between mothers with male and female offspring. A possible mechanism for the observed increase in insulin requirements with parity could be pregnancy-induced endogenous insulin secretion (29, 30). It could be speculated that during the first pregnancy, there is limited residual endogenous insulin secretion that is attenuated during the next pregnancies. Unfortunately, we have no measurements of C-peptide to support such a hypothesis. Exercise and diet could also change with parity. It could be that mothers of more than one child have less time for exercise and healthy eating during pregnancy. This would negatively affect insulin sensitivity and thus increase insulin requirement. Because of the lack of available data on these possible confounders, we were not able to adjust for them. We did not record the patients' ethnicity, and this could be a weakness in our study. We know that ethnicity affects insulin sensitivity, and this too could be a confounder. Our cohort consisting of Danish women is, however, almost exclusively white and thus highly ethnically homogeneous. Hence, the influence of ethnicity in this study must be inconsiderable. Our study is to our knowledge the largest study on insulin requirements during pregnancy to date, which is an obvious strength and produces convincingly low P values and narrow CIs. In conclusion, our data show that parity per se increases insulin requirements during pregnancy between 9% and 36% in type 1 diabetes and confirm that insulin dosages exhibit a characteristic pattern with a modest early decrease and a pronounced late increase during pregnancy. Furthermore, the large number of participants in our study has yielded more accurate information on how parity affects insulin requirements in pregnant women with type 1 diabetes and shows that the more times a woman with type 1 diabetes gives birth, the more insulin she needs. The finding of increasing insulin requirements with increasing parity is not only a highly interesting physiological observation. Our findings also provide valuable information the clinician can use to achieve tight glycemic control throughout pregnancy. They indicate that adjustment of insulin doses should consider not only HbA1c values, blood glucose values, fetal size, and awareness but also the patient's parity. Abbreviations: Abbreviations: BMI body mass index CI confidence interval HbA1c hemoglobin 1Ac Acknowledgments The authors thank the Biostatistical Advisory Service at the Faculty of Health, Aarhus University, for helping with all statistical analysis. Financial Support: G.Ø.S and U.K. are supported by the Danish Diabetes Academy, funded by the Novo Nordisk Foundation. P.G.O. received financial support from the Novo Nordisk Foundation. J.F. has no financial disclosures. The funding sources had no role in the study design, inclusion process, data analysis, or statistical considerations or in the writing of this report. Disclosure Summary: The authors have nothing to disclose. References 1. Nielsen GL, Sørensen HT, Nielsen PH, Sabroe S, Olsen J. Glycosylated hemoglobin as predictor of adverse fetal outcome in type 1 diabetic pregnancies. Acta Diabetol . 1997; 34( 3): 217– 222. Google Scholar CrossRef Search ADS PubMed  2. Suhonen L, Hiilesmaa V, Teramo K. Glycaemic control during early pregnancy and fetal malformations in women with type I diabetes mellitus. Diabetologia . 2000; 43( 1): 79– 82. Google Scholar CrossRef Search ADS PubMed  3. Barbour LA, McCurdy CE, Hernandez TL, Kirwan JP, Catalano PM, Friedman JE. Cellular mechanisms for insulin resistance in normal pregnancy and gestational diabetes. Diabetes Care . 2007; 30( Suppl 2): S112– S119. Google Scholar CrossRef Search ADS PubMed  4. Naver KV, Lundbye-Christensen S, Gorst-Rasmussen A, Nilas L, Secher NJ, Rasmussen S, Ovesen P. Parity and risk of diabetes in a Danish nationwide birth cohort. Diabet Med . 2011; 28( 1): 43– 47. Google Scholar CrossRef Search ADS PubMed  5. Simmons D, Shaw J, McKenzie A, Eaton S, Cameron AJ, Zimmet P. Is grand multiparity associated with an increased risk of dysglycaemia? Diabetologia . 2006; 49( 7): 1522– 1527. Google Scholar CrossRef Search ADS PubMed  6. Cheung NW. Is parity associated with earlier diagnosis of type 2 diabetes? Diabetes Res Clin Pract . 2004; 66( 3): 287– 291. Google Scholar CrossRef Search ADS PubMed  7. Iversen DS, Støy J, Kampmann U, Voss TS, Madsen LR, Møller N, Ovesen PG. Parity and type 2 diabetes mellitus: a study of insulin resistance and β-cell function in women with multiple pregnancies. BMJ Open Diabetes Res Care . 2016; 4( 1): e000237. Google Scholar CrossRef Search ADS PubMed  8. Gunderson EP, Lewis CE, Tsai AL, Chiang V, Carnethon M, Quesenberry CP Jr, Sidney S. A 20-year prospective study of childbearing and incidence of diabetes in young women, controlling for glycemia before conception: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Diabetes . 2007; 56( 12): 2990– 2996. Google Scholar CrossRef Search ADS PubMed  9. Boyko EJ, Alderman BW, Keane EM, Baron AE. Effects of childbearing on glucose tolerance and NIDDM prevalence [published correction appears in Diabetes Care. 1990;13(11):1138]. Diabetes Care . 1990; 13( 8): 848– 854. Google Scholar CrossRef Search ADS PubMed  10. Collins VR, Dowse GK, Zimmet PZ. Evidence against association between parity and NIDDM from five population groups. Diabetes Care . 1991; 14( 11): 975– 981. Google Scholar CrossRef Search ADS PubMed  11. García-Patterson A, Gich I, Amini SB, Catalano PM, de Leiva A, Corcoy R. Insulin requirements throughout pregnancy in women with type 1 diabetes mellitus: three changes of direction. Diabetologia . 2010; 53( 3): 446– 451. Google Scholar CrossRef Search ADS PubMed  12. Callesen NF, Ringholm L, Stage E, Damm P, Mathiesen ER. Insulin requirements in type 1 diabetic pregnancy: do twin pregnant women require twice as much insulin as singleton pregnant women? Diabetes Care . 2012; 35( 6): 1246– 1248. Google Scholar CrossRef Search ADS PubMed  13. Padmanabhan S, Jiang S, Mclean M, Cheung NW. Effect of pregnancy on insulin requirements differs between type 1 and type 2 diabetes: a cohort study of 222 pregnancies. Aust N Z J Obstet Gynaecol . 2016; 56( 4): 352– 357. Google Scholar CrossRef Search ADS PubMed  14. Fuglsang J, Lauszus F, Flyvbjerg A, Ovesen P. Human placental growth hormone, insulin-like growth factor I and -II, and insulin requirements during pregnancy in type 1 diabetes. J Clin Endocrinol Metab . 2003; 88( 9): 4355– 4361. Google Scholar CrossRef Search ADS   15. Citizenship and origin of the municipality of Aarhus [machine translation from Danish]. Available at: http://www.aarhus.dk/statistik. https://goo.gl/9CitkU. Accessed 10 January 2018. 16. Damm P, Ovesen P, Andersen LL, Møller M, Jensen DM. Kliniske retningslinier for diabetesbehandling ved graviditet hos kvinder med kendt diabetes diabetes (type 1 og type 2) før graviditeten. [Danish national clinical guidelines for diabetes treatment in pregnancy in women with known diabetes (type 1 and type 2) before pregnancy]. Available at: http://www.endocrinology.dk/kliniske retningslinier - diabetes og graviditet.pdf. 2010. Accessed 10 January 2018. 17. Ringholm L, Pedersen-Bjergaard U, Thorsteinsson B, Damm P, Mathiesen ER. Hypoglycaemia during pregnancy in women with Type 1 diabetes. Diabet Med . 2012; 29( 5): 558– 566. Google Scholar CrossRef Search ADS PubMed  18. Secher AL, Stage E, Ringholm L, Barfred C, Damm P, Mathiesen ER. Real-time continuous glucose monitoring as a tool to prevent severe hypoglycaemia in selected pregnant women with type 1 diabetes - an observational study. Diabet Med . 2014; 31( 3): 352– 356. Google Scholar CrossRef Search ADS PubMed  19. Padmanabhan S, McLean M, Cheung NW. Falling insulin requirements are associated with adverse obstetric outcomes in women with preexisting diabetes. Diabetes Care . 2014; 37( 10): 2685– 2692. Google Scholar CrossRef Search ADS PubMed  20. Ingram KH, Hunter GR, James JF, Gower BA. Central fat accretion and insulin sensitivity: differential relationships in parous and nulliparous women. Int J Obes . 2017; 41( 8): 1214– 1217. Google Scholar CrossRef Search ADS   21. Sohlström A, Forsum E. Changes in adipose tissue volume and distribution during reproduction in Swedish women as assessed by magnetic resonance imaging. Am J Clin Nutr . 1995; 61( 2): 287– 295. Google Scholar CrossRef Search ADS PubMed  22. Blaudeau TE, Hunter GR, Sirikul B. Intra-abdominal adipose tissue deposition and parity. Int J Obes . 2006; 30( 7): 1119– 1124. Google Scholar CrossRef Search ADS   23. Davis DL, Gottlieb MB. Reduced ratio of male to female births in several industrial countries a sentinel health indicator? JAMA . 1998; 279: 1018– 1023. Google Scholar CrossRef Search ADS PubMed  24. García-Patterson A, Miñambres I, Adelantado JM, Gich I, Puig T, de Leiva A, Corcoy R. Sex ratio at birth is associated with type 1 diabetes characteristics. Acta Diabetol . 2016; 53( 6): 1025– 1035. Google Scholar CrossRef Search ADS PubMed  25. Forsén T, Eriksson JG, Tuomilehto J, Osmond C, Barker DJ. Growth in utero and during childhood among women who develop coronary heart disease: longitudinal study. BMJ . 1999; 319( 7222): 1403– 1407. Google Scholar CrossRef Search ADS PubMed  26. Pedersen JF. Ultrasound evidence of sexual difference in fetal size in first trimester. BMJ . 1980; 281( 6250): 1253. Google Scholar CrossRef Search ADS PubMed  27. Eriksson JG, Kajantie E, Osmond C, Thornburg K, Barker DJ. Boys live dangerously in the womb. Am J Hum Biol . 2010; 22( 3): 330– 335. Google Scholar CrossRef Search ADS PubMed  28. Retnakaran R, Kramer CK, Ye C, Kew S, Hanley AJ, Connelly PW, Sermer M, Zinman B. Fetal sex and maternal risk of gestational diabetes mellitus: the impact of having a boy. Diabetes Care . 2015; 38( 5): 844– 851. Google Scholar CrossRef Search ADS PubMed  29. Ilic S, Jovanovic L, Wollitzer AO. Is the paradoxical first trimester drop in insulin requirement due to an increase in C-peptide concentration in pregnant Type I diabetic women? Diabetologia . 2000; 43( 10): 1329– 1330. Google Scholar CrossRef Search ADS PubMed  30. Nielsen LR, Rehfeld JF, Pedersen-Bjergaard U, Damm P, Mathiesen ER. Pregnancy-induced rise in serum C-peptide concentrations in women with type 1 diabetes. Diabetes Care . 2009; 32( 6): 1052– 1057. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Clinical Endocrinology and Metabolism Oxford University Press

Parity Increases Insulin Requirements in Pregnant Women With Type 1 Diabetes

Loading next page...
 
/lp/ou_press/parity-increases-insulin-requirements-in-pregnant-women-with-type-1-qgFoiqTbE0
Publisher
Endocrine Society
Copyright
Copyright © 2018 Endocrine Society
ISSN
0021-972X
eISSN
1945-7197
D.O.I.
10.1210/jc.2018-00094
Publisher site
See Article on Publisher Site

Abstract

Abstract Context Tight glycemic control throughout pregnancy in women with type 1 diabetes is crucial, and knowledge about which factors that affect insulin sensitivity could improve the outcome for both mother and offspring. Objective To evaluate insulin requirements in women with type 1 diabetes during pregnancy and test whether parity affects insulin requirements. Design Observational cohort study consisting of women with type 1 diabetes who gave birth at Aarhus University Hospital, Denmark, from 2004 to 2014. Main Outcome Measure Daily insulin requirement (the hypothesis that parity could affect insulin resistance was formulated before data collection). Results A total of 380 women with a total of 536 pregnancies were included in the study. Mean age was 31.1 years, and prepregnancy hemoglobin A1c was 60 mmol/mol. Parity was as follows: P0, 43%; P1, 40%; P2, 14%; and P3+4, 3%. Insulin requirements from weeks 11 to 16 decreased significantly by 4% (P = 0.0004) and rose from week 19 to delivery with a peak of 70% (P < 0.0005) at weeks 33 to 36. Overall, insulin requirements increased significantly with parity. The unadjusted differences between P0 and P1, P2, and P3+4 were 9% (P < 0.0005), 12% (P < 0.0005), and 23% (P < 0.0011), respectively. After adjustment for confounders, differences were 13% (P < 0.0005), 20% (P < 0.0005), and 36% (P < 0.0005). We also observed an adjusted difference between P1 and P3+4 of 20% (P < 0.0012). Conclusions The data show changes in insulin requirements from week to week in pregnancy and indicate that insulin requirements increase with parity. This suggests that the patient's parity probably should be considered in choosing insulin dosages for pregnant women with type 1 diabetes. Alterations in insulin requirements and insulin resistance during pregnancies complicated by diabetes constitute a challenge for both patients and clinicians. Pregnancies in women with type 1 diabetes are associated with an increased risk for congenital malformations, obstetric complications, and neonatal morbidity—a risk that is directly correlated to glycemic control immediately before and during pregnancy (1, 2). Thus, tight glycemic control throughout pregnancy is crucial and can be achieved only if a successful collaboration between patients and clinicians is established. To allow patients and clinicians to act preventively, knowledge about alterations in insulin requirements over time during pregnancy is essential to enable implementation of appropriate insulin dosage preemptively. During normal pregnancy, high levels of diabetogenic placental hormones lead to a physiological decrease in peripheral insulin sensitivity as pregnancy proceeds (3). Another factor that could be affecting insulin requirement is parity. Some studies have reported an increase in the incidence of type 2 diabetes with increasing parity (4–6), whereas others have found no such association (7–10). Several of these studies have included a limited number of patients, and the influence of parity on insulin requirement during pregnancy in women with type 1 diabetes remains uncertain. In type 2 diabetes and gestational diabetes, increasing parity may increase insulin requirement because of exhaustion of the pancreatic β cells due to the high insulin need during pregnancy. In type 1 diabetes, however, there is essentially no insulin production in the β cells. Any increase in insulin requirements with increasing parity would be determined by insulin resistance in target tissues probably related to such factors as age, increasing body mass index (BMI), or altered hormone secretion from the placenta. The progressions in insulin requirements during pregnancy vary considerably between individuals. Some women experience subtle changes, whereas others have to triple or quadruple insulin doses late in pregnancy. We and others (11–13) have studied insulin resistance during pregnancy. Insulin resistance can be estimated in different ways; analyzing insulin requirements during pregnancy is one approach. This has previously been done only in relatively small cohorts of patients with type 1 diabetes (11–14). In the current study, we evaluated insulin requirements during 536 pregnancies in 380 patients with type 1 diabetes. Using the same dataset, we examined whether parity affects insulin requirements. Patients and Methods A retrospective observational cohort study was conducted. The cohort consisted of women with type 1 diabetes who gave birth at Aarhus University Hospital, Denmark, between January 2004 and December 2014. All pregnant women with type 1 diabetes residing in Central Denmark Region are treated at Aarhus University Hospital, which is a tertiary center with highly specialized service. The Central Denmark Region has 1.3 million residents and is very diverse in terms of socioeconomic status covering both rural and urban areas. The ethnicity of the cohort was predominantly white (15), but the ethnicity of each participant was not recorded. Exclusion criteria were multiple pregnancy, stillbirth, termination of pregnancy, and spontaneous abortion. The women were seen at the outpatient clinic for pregnant women with diabetes at Aarhus University Hospital or contacted by telephone every 2 to 4 weeks from gestational week 4 to week 28 and every 1 to 2 weeks from week 28 to delivery. A diabetologist adjusted insulin dose according to current hemoglobin 1Ac (HbA1c) and the patient's own daily recordings of fasting and postprandial glucose measurements. The target HbA1c values for pregnant women with type 1 diabetes were according to the Danish National Guidelines: 48 mmol/mol (6.5%) before gestational week 20 and 37 mmol/mol (5.6%) after gestational week 20 (16). Diabetes regulation was managed by self-monitoring of glucose and insulin adjustments, aiming at capillary plasma glucose levels of 4 to 6 mmol/L preprandial and 4 to 8 mmol/L 1.5 hours postprandial. An obstetrician or a sonographer obtained serial ultrasound scans to monitor fetal growth, and a dietitian gave nutritional advice. From the patients’ medical records, we obtained the following data at every visit: daily insulin requirements, current HbA1c, and weight. We also extracted the following prepregnancy data: weight, height, parity, last known daily insulin dosage before pregnancy, last known HbA1c (not older than 6 months), and number of years since diabetes onset (duration of diabetes). Relevant data on the offspring, such as gestational age at delivery, sex, and birth weight, were also obtained. Daily insulin requirement was calculated by adding long-acting and short-acting insulin. The mean daily insulin requirement was determined at the following time intervals: prepregnancy, weeks 5 to 10, weeks 11 to 14, weeks 15 to 18, weeks 19 to 22, weeks 23 to 28, weeks 29 to 32, weeks 33 to 36, and weeks 37 to 40. Our primary outcome was mean daily insulin requirement (IU) at the above-mentioned time intervals. Secondary outcomes were (1) percentage change from prepregnancy insulin requirement at the above-mentioned time intervals, (2) the effect of parity on mean daily insulin requirement for the whole pregnancy, and (3) the effect of fetal sex on insulin requirement. The Danish Health Authority (no. 3-3013-625/1) and the Danish Data Protection Agency (no. 1-16-02-347-14) approved our study. Statistics The Biostatistical Advisory Service at the Faculty of Health, Aarhus University, helped perform all statistical analyses by using the statistical software Stata 13 (Stata Corp., College Station, TX). Continuous data are expressed as mean with standard deviation and categorical variables as percentages. Repeated-measurement analysis of variance was used to compare the paired data between the groups. Both the repeated measurements in every pregnancy and the repeated measurements for each of the woman's pregnancies were taken into account by using nested random effects in a mixed model. Measurements were analyzed on log scale, and results were back-transformed to obtain relative comparisons. Model validation was performed by visual inspection of residuals, fitted values, and random effect estimates, which did not give cause to reject the model. For the subgroup analyses, paired t test was used. Statistical significance was defined as P < 0.05. Five different models were made to adjust for factors that could affect insulin requirement: (1) unadjusted; (2) adjusted for prepregnancy BMI; (3) adjusted for prepregnancy BMI and age; (4) adjusted for prepregnancy BMI, age, and prepregnancy HbA1c; and (5) adjusted for prepregnancy BMI, age, prepregnancy HbA1c, and duration of diabetes. Results We reviewed 583 pregnancies. Exclusion criteria were multiple pregnancy (n = 34), stillbirth (n = 3), termination of pregnancy (n = 5), spontaneous abortion (n = 2), and missing data (n = 3). Thus, 380 women with a total 536 pregnancies were included in the study. Several of the women gave birth more than once within the time period. Two hundred thirty-six women (62%) gave birth once, 132 women (35%) gave birth twice, and 12 women (3%) gave birth three or four times. Table 1 presents the clinical features of each pregnancy in the cohort. When we divided the cohort into parity groups (P0, P1, P2, and P3+4), we found no statistical difference in prepregnancy BMI or prepregnancy HbA1c between the groups. Table 1. Demographic Characteristics of Patients With Type 1 Diabetes Included in Study Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  A total of 380 women gave birth 536 times. Two hundred thirty-six women (62%) gave birth one time, 132 women (35%) gave birth two times, and 12 women (3%) gave birth three times. View Large Table 1. Demographic Characteristics of Patients With Type 1 Diabetes Included in Study Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  Variable  Parity  All  P0  P1  P2  P3+4  Patients, n (%)  229 (43)  216 (40)  75 (14)  16 (3)  536 (100)  Age, y  29.1  32.0  33.6  36.7  31.1  Duration of diabetes, y  14.4  15.6  16.7  12.8  15.2  Prepregnancy HbA1c              In mmol  60  59  62  65  60   In %  7.6  7.5  7.8  8.1  7.6  A total of 380 women gave birth 536 times. Two hundred thirty-six women (62%) gave birth one time, 132 women (35%) gave birth two times, and 12 women (3%) gave birth three times. View Large We compared duration of diabetes among the groups and, as expected, found that the P0 group had statistically significantly shorter diabetes duration, at 14.4 years, than the P1 group, at 15.6 years, and P2 group, at 16.7 years. The P3+4 group, however (n = 16), had an even shorter duration, at 12.8 years, probably due to the small number of included individuals and the influence of a few extremes affecting the average. Mean age overall was 31.1 years, and age increased, also as expected, with parity: P0, 29.1 years; P1, 32.0 years; P2, 33.6 years; and P3+4, 36.7 years. Mean HbA1c was 55.0 mmol/L [95% confidence interval (CI), 54.7 to 55.4 mmol/L], 48.3 mmol/L (95% CI, 48.1 to 48.6 mmol/L), and 48.3 mmol/L (95% CI, 48.0 to 48.5 mmol/L) for first, second, and third trimesters, respectively. Mean daily insulin requirements at the different time intervals for all included pregnancies (n = 536) are shown in Fig.1. Figure 1 also shows the mean daily insulin requirements in percentages relative to prepregnancy levels. Insulin requirements at weeks 11 to 16 decreased significantly by 4% (95% CI, 2% to 6%; P = 0.004) and rose significantly from week 19 to delivery, with a peak at weeks 33 to 36 at 70% (95% CI, 65% to 75%; P < 0.0005), relative to prepregnancy levels. After week 37, the insulin requirement decreased and terminated at 66% (95% CI, 59% to 72%) relative to prepregnancy levels. Figure 1. View largeDownload slide Mean daily insulin requirements at the different time intervals for all included pregnancies (with 95% CI error bars). The dashed line represents the prepregnancy insulin requirement, and numbers in percentages indicate the increase or decrease in percentage relative to prepregnancy levels. *Statistical significance (P < 0.05). Figure 1. View largeDownload slide Mean daily insulin requirements at the different time intervals for all included pregnancies (with 95% CI error bars). The dashed line represents the prepregnancy insulin requirement, and numbers in percentages indicate the increase or decrease in percentage relative to prepregnancy levels. *Statistical significance (P < 0.05). When comparing mean daily insulin requirements for the entire pregnancy, we found that insulin requirements increased with parity. Figure 2 shows how mean daily insulin requirement developed during pregnancy for P0, P1, and P2. Figure 2. View largeDownload slide Difference in mean daily insulin requirement at the different time intervals between parity groups. Figure 2. View largeDownload slide Difference in mean daily insulin requirement at the different time intervals between parity groups. The unadjusted differences between P0 and P1, P2, and P3+4 were 9% (95% CI, 4% to 14%; P < 0.0005,), 12% (95% CI, 4% to 20%; P < 0.0005), and 23% (95% CI, 5% to 44%; P = 0.011), respectively. To avoid any confounding, we adjusted for age, prepregnancy BMI, prepregnancy HbA1c, and duration of diabetes. After adjustment for all four variables, the differences between P0 and P1, P2, and P3+4 were 13% (95% CI, 8% to 18%; P < 0.0005), 20% (95% CI, 12% to 30%; P < 0.0005), and 36% (95% CI, 17% to 38%; P < 0.0005), respectively. An additional significant difference between P1 and P3+4 in mean daily insulin requirements of 20% (95% CI, 4% to 39%; P = 0.012) was found. Table 2 presents all estimates, both unadjusted and adjusted. Table 2. Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  a Statistically significant (P < 0.05). View Large Table 2. Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0  9a (4.3–13.6)  12a (3.9 to 20.4)  23a (4.9 to 43.8)   P1    3 (−4.2 to 10.1)  13 (−3.6 to 32)   P2      10 (−5.8 to 28.1)  Adjusted for BMI         P0  8a (3.4–12)  10a (3.2 to 18.3)  17a (1.5 to 35.6)   P1    3 (−3.7 to 9.5)  9 (−5.6 to 26)   P2      6 (−9.6 to 22.4)  Adjusted for BMI and Age         P0  15a (9.9–20.3)  25a (15.2 to 34.6)  40a (20.0 to 64.6)   P1    8a (1.4 to 15.7)  22a (5.1 to 41.2)   P2      12 (−2.4 to 29.6)  Adjusted for BMI, age, prepregnancy HbA1c         P0  13a (8.2–18.3)  21a (11.5 to 30.2)  34a (15.5 to 56.3)   P1    6 (−0.3 to 13.7)  19a (2.8 to 37.1)   P2      11 (−2.9 to 28.0)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  13a (8.1–18.2)  20a (11.5 to 30.1)  36a (16.8 to 58.1)   P1    7 (−0.2 to 13.8)  20a (4.1 to 38.9)   P2      13 (−1.7 to 29.6)  a Statistically significant (P < 0.05). View Large To confirm our results, we performed a subgroup analysis that included only women who were in the cohort more than once (n = 147). We compared mean daily insulin requirements in the woman's first pregnancy in the cohort with her subsequent pregnancies. We found a significant difference in insulin requirement between the pregnancies of 11% (95% CI, 9% to 13%; P < 0.0005). This was across all parities. Table 3 shows all estimates and 95% CIs for comparisons of the parities. Table 3. Subgroup Analysis: Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  a Statistically significant (P < 0.05). View Large Table 3. Subgroup Analysis: Comparison of Total Insulin Requirement for Whole Pregnancy Between Parity Groups: Unadjusted and Adjusted Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  Model  Insulin Requirement per Parity Group (%)  P1  P2  P3+4  Unadjusted         P0
   11a (5.7–16.0)  15a (5.4 to 25.3)  44a (20.2–73.3)   P1
     4 (−4.2 to 12.5)  30a (8.7–56.3)   P2      26a (5.5–49.4)  Adjusted for BMI         P0  8a (3.5–13.2)  12a (2.7 to 20.9)  31a (10.5–56.1)   P1    3 (−4.5 to 11.0)  21a (2.3–43.9)   P2      18a (0–38.8)  Adjusted for BMI and age         P0  17a (10.9–24.2)  31a (18.0 to 46.0)  65a (35.8–101)   P1    12a (2.9 to 21.5)  41a (17.6–68.4)   P2      26a (6.7–48.4)  Adjusted for BMI, age, and prepregnancy HbA1ca         P0  15a (8.5–21.3)  27a (13.9 to 40.6)  59a (30.9–91.8)   P1    10a (1.6 to 19.7)  38a (16.0–64.5)   P2      25a (6.7–47.1)  Adjusted for BMI, age, prepregnancy HbA1c, and duration of diabetes         P0  15a (8.2–21.1)  26a (13.4 to 40.2)  59a (30.6–91.5)   P1    7a (1.5 to 19.6)  39a (16.0–64.6)   P2      25a (6.8–47.3)  a Statistically significant (P < 0.05). View Large The sex ratio at birth (live male newborns/total live newborns) in our study was 0.496, and we found no difference in mean daily insulin requirements when comparing women with male offspring and female offspring (data not shown). Discussion The fall and rise in insulin requirements during pregnancy are well known. The S-shaped curve has been reported before (11, 14), but here we have clearly confirmed it with a large number of participants. The decrease in insulin requirements around weeks 10 to 14 is important to acknowledge in order to avoid hypoglycemia in early pregnancy. With our results, we have attained a more detailed estimate of this decrease, and we hope this will help lower the incidence of potentially dangerous hypoglycemia in early pregnancy (17, 18). We have also identified the time span within which the rapid increase in insulin requirements occurs, thereby allowing clinicians to reduce hyperglycemia and maintain steady and tight glycemic control during the last part of the pregnancy. In addition, we have established that after the maximal insulin requirement levels are reached around week 37, a minor decline of 4% is observed. It is well known that a decline in insulin requirement of ≥15% in late pregnancy can be a sign of placental insufficiency (19). The observed 4% decline is probably a part of the normal type 1 diabetes pregnancy and should not by itself lead to induction of birth. Finally, we established that parity is a factor that should be considered when treating pregnant women with type 1 diabetes. We have found a statistically and clinically significant difference in mean daily insulin requirement during pregnancy between the P0 group and P1, P2, and P3+4 groups at 9%, 12%, and 23% respectively. Because this difference could be due to a difference between the groups in BMI, age, or prepregnancy HbA1c, we also presented an adjusted model, which increased the differences to 13%, 20%, and 36%, respectively. In addition, after adjustment we found a significant difference between the P1 group and P3+4 group of 20%. In our subgroup analysis, where we compared mean daily insulin requirement in the woman's first pregnancy in the cohort with the following pregnancy, we were able to confirm these findings, thereby diminishing the effect of variation between individuals. This further strengthens the concept that parity predicts increased insulin requirements as this clearly shows how women who had given birth several times in our cohort increased their individual total insulin requirement during pregnancy with each pregnancy. Furthermore, when we looked at the insulin requirements before pregnancy, we found no difference between parity groups, indicating that the difference observed only concerns pregnancy. The mechanisms behind the parity-associated increase in insulin requirements are unknown. Studies have shown that anthropometric measures increase with parity and that this could influence insulin resistance (20, 21). Although we did not find any difference when we adjusted for BMI, other anthropometric measures (e.g., subcutaneous fat or intra-abdominal fat) could increase with parity and thereby influence our findings (22). Sex of the offspring could also potentially affect the insulin resistance. The sex ratio at birth in the general population is 0.515 (23), and in a cohort like ours, Garcia-Patterson et al. (24) found a sex ratio at birth at 0.509. Our sex ratio at birth at 0.496 was, as expected, not significantly different from any of the cohorts mentioned. Male fetuses are known to grow faster in the womb, be larger at birth, and have more efficient placentas (25–27). Moreover, women carrying a male fetus have a higher risk of developing gestational diabetes (28). We thus hypothesized that male fetuses for this reason could make the mother more insulin resistant during pregnancy, but our analysis did not show any difference in insulin requirement between mothers with male and female offspring. A possible mechanism for the observed increase in insulin requirements with parity could be pregnancy-induced endogenous insulin secretion (29, 30). It could be speculated that during the first pregnancy, there is limited residual endogenous insulin secretion that is attenuated during the next pregnancies. Unfortunately, we have no measurements of C-peptide to support such a hypothesis. Exercise and diet could also change with parity. It could be that mothers of more than one child have less time for exercise and healthy eating during pregnancy. This would negatively affect insulin sensitivity and thus increase insulin requirement. Because of the lack of available data on these possible confounders, we were not able to adjust for them. We did not record the patients' ethnicity, and this could be a weakness in our study. We know that ethnicity affects insulin sensitivity, and this too could be a confounder. Our cohort consisting of Danish women is, however, almost exclusively white and thus highly ethnically homogeneous. Hence, the influence of ethnicity in this study must be inconsiderable. Our study is to our knowledge the largest study on insulin requirements during pregnancy to date, which is an obvious strength and produces convincingly low P values and narrow CIs. In conclusion, our data show that parity per se increases insulin requirements during pregnancy between 9% and 36% in type 1 diabetes and confirm that insulin dosages exhibit a characteristic pattern with a modest early decrease and a pronounced late increase during pregnancy. Furthermore, the large number of participants in our study has yielded more accurate information on how parity affects insulin requirements in pregnant women with type 1 diabetes and shows that the more times a woman with type 1 diabetes gives birth, the more insulin she needs. The finding of increasing insulin requirements with increasing parity is not only a highly interesting physiological observation. Our findings also provide valuable information the clinician can use to achieve tight glycemic control throughout pregnancy. They indicate that adjustment of insulin doses should consider not only HbA1c values, blood glucose values, fetal size, and awareness but also the patient's parity. Abbreviations: Abbreviations: BMI body mass index CI confidence interval HbA1c hemoglobin 1Ac Acknowledgments The authors thank the Biostatistical Advisory Service at the Faculty of Health, Aarhus University, for helping with all statistical analysis. Financial Support: G.Ø.S and U.K. are supported by the Danish Diabetes Academy, funded by the Novo Nordisk Foundation. P.G.O. received financial support from the Novo Nordisk Foundation. J.F. has no financial disclosures. The funding sources had no role in the study design, inclusion process, data analysis, or statistical considerations or in the writing of this report. Disclosure Summary: The authors have nothing to disclose. References 1. Nielsen GL, Sørensen HT, Nielsen PH, Sabroe S, Olsen J. Glycosylated hemoglobin as predictor of adverse fetal outcome in type 1 diabetic pregnancies. Acta Diabetol . 1997; 34( 3): 217– 222. Google Scholar CrossRef Search ADS PubMed  2. Suhonen L, Hiilesmaa V, Teramo K. Glycaemic control during early pregnancy and fetal malformations in women with type I diabetes mellitus. Diabetologia . 2000; 43( 1): 79– 82. Google Scholar CrossRef Search ADS PubMed  3. Barbour LA, McCurdy CE, Hernandez TL, Kirwan JP, Catalano PM, Friedman JE. Cellular mechanisms for insulin resistance in normal pregnancy and gestational diabetes. Diabetes Care . 2007; 30( Suppl 2): S112– S119. Google Scholar CrossRef Search ADS PubMed  4. Naver KV, Lundbye-Christensen S, Gorst-Rasmussen A, Nilas L, Secher NJ, Rasmussen S, Ovesen P. Parity and risk of diabetes in a Danish nationwide birth cohort. Diabet Med . 2011; 28( 1): 43– 47. Google Scholar CrossRef Search ADS PubMed  5. Simmons D, Shaw J, McKenzie A, Eaton S, Cameron AJ, Zimmet P. Is grand multiparity associated with an increased risk of dysglycaemia? Diabetologia . 2006; 49( 7): 1522– 1527. Google Scholar CrossRef Search ADS PubMed  6. Cheung NW. Is parity associated with earlier diagnosis of type 2 diabetes? Diabetes Res Clin Pract . 2004; 66( 3): 287– 291. Google Scholar CrossRef Search ADS PubMed  7. Iversen DS, Støy J, Kampmann U, Voss TS, Madsen LR, Møller N, Ovesen PG. Parity and type 2 diabetes mellitus: a study of insulin resistance and β-cell function in women with multiple pregnancies. BMJ Open Diabetes Res Care . 2016; 4( 1): e000237. Google Scholar CrossRef Search ADS PubMed  8. Gunderson EP, Lewis CE, Tsai AL, Chiang V, Carnethon M, Quesenberry CP Jr, Sidney S. A 20-year prospective study of childbearing and incidence of diabetes in young women, controlling for glycemia before conception: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Diabetes . 2007; 56( 12): 2990– 2996. Google Scholar CrossRef Search ADS PubMed  9. Boyko EJ, Alderman BW, Keane EM, Baron AE. Effects of childbearing on glucose tolerance and NIDDM prevalence [published correction appears in Diabetes Care. 1990;13(11):1138]. Diabetes Care . 1990; 13( 8): 848– 854. Google Scholar CrossRef Search ADS PubMed  10. Collins VR, Dowse GK, Zimmet PZ. Evidence against association between parity and NIDDM from five population groups. Diabetes Care . 1991; 14( 11): 975– 981. Google Scholar CrossRef Search ADS PubMed  11. García-Patterson A, Gich I, Amini SB, Catalano PM, de Leiva A, Corcoy R. Insulin requirements throughout pregnancy in women with type 1 diabetes mellitus: three changes of direction. Diabetologia . 2010; 53( 3): 446– 451. Google Scholar CrossRef Search ADS PubMed  12. Callesen NF, Ringholm L, Stage E, Damm P, Mathiesen ER. Insulin requirements in type 1 diabetic pregnancy: do twin pregnant women require twice as much insulin as singleton pregnant women? Diabetes Care . 2012; 35( 6): 1246– 1248. Google Scholar CrossRef Search ADS PubMed  13. Padmanabhan S, Jiang S, Mclean M, Cheung NW. Effect of pregnancy on insulin requirements differs between type 1 and type 2 diabetes: a cohort study of 222 pregnancies. Aust N Z J Obstet Gynaecol . 2016; 56( 4): 352– 357. Google Scholar CrossRef Search ADS PubMed  14. Fuglsang J, Lauszus F, Flyvbjerg A, Ovesen P. Human placental growth hormone, insulin-like growth factor I and -II, and insulin requirements during pregnancy in type 1 diabetes. J Clin Endocrinol Metab . 2003; 88( 9): 4355– 4361. Google Scholar CrossRef Search ADS   15. Citizenship and origin of the municipality of Aarhus [machine translation from Danish]. Available at: http://www.aarhus.dk/statistik. https://goo.gl/9CitkU. Accessed 10 January 2018. 16. Damm P, Ovesen P, Andersen LL, Møller M, Jensen DM. Kliniske retningslinier for diabetesbehandling ved graviditet hos kvinder med kendt diabetes diabetes (type 1 og type 2) før graviditeten. [Danish national clinical guidelines for diabetes treatment in pregnancy in women with known diabetes (type 1 and type 2) before pregnancy]. Available at: http://www.endocrinology.dk/kliniske retningslinier - diabetes og graviditet.pdf. 2010. Accessed 10 January 2018. 17. Ringholm L, Pedersen-Bjergaard U, Thorsteinsson B, Damm P, Mathiesen ER. Hypoglycaemia during pregnancy in women with Type 1 diabetes. Diabet Med . 2012; 29( 5): 558– 566. Google Scholar CrossRef Search ADS PubMed  18. Secher AL, Stage E, Ringholm L, Barfred C, Damm P, Mathiesen ER. Real-time continuous glucose monitoring as a tool to prevent severe hypoglycaemia in selected pregnant women with type 1 diabetes - an observational study. Diabet Med . 2014; 31( 3): 352– 356. Google Scholar CrossRef Search ADS PubMed  19. Padmanabhan S, McLean M, Cheung NW. Falling insulin requirements are associated with adverse obstetric outcomes in women with preexisting diabetes. Diabetes Care . 2014; 37( 10): 2685– 2692. Google Scholar CrossRef Search ADS PubMed  20. Ingram KH, Hunter GR, James JF, Gower BA. Central fat accretion and insulin sensitivity: differential relationships in parous and nulliparous women. Int J Obes . 2017; 41( 8): 1214– 1217. Google Scholar CrossRef Search ADS   21. Sohlström A, Forsum E. Changes in adipose tissue volume and distribution during reproduction in Swedish women as assessed by magnetic resonance imaging. Am J Clin Nutr . 1995; 61( 2): 287– 295. Google Scholar CrossRef Search ADS PubMed  22. Blaudeau TE, Hunter GR, Sirikul B. Intra-abdominal adipose tissue deposition and parity. Int J Obes . 2006; 30( 7): 1119– 1124. Google Scholar CrossRef Search ADS   23. Davis DL, Gottlieb MB. Reduced ratio of male to female births in several industrial countries a sentinel health indicator? JAMA . 1998; 279: 1018– 1023. Google Scholar CrossRef Search ADS PubMed  24. García-Patterson A, Miñambres I, Adelantado JM, Gich I, Puig T, de Leiva A, Corcoy R. Sex ratio at birth is associated with type 1 diabetes characteristics. Acta Diabetol . 2016; 53( 6): 1025– 1035. Google Scholar CrossRef Search ADS PubMed  25. Forsén T, Eriksson JG, Tuomilehto J, Osmond C, Barker DJ. Growth in utero and during childhood among women who develop coronary heart disease: longitudinal study. BMJ . 1999; 319( 7222): 1403– 1407. Google Scholar CrossRef Search ADS PubMed  26. Pedersen JF. Ultrasound evidence of sexual difference in fetal size in first trimester. BMJ . 1980; 281( 6250): 1253. Google Scholar CrossRef Search ADS PubMed  27. Eriksson JG, Kajantie E, Osmond C, Thornburg K, Barker DJ. Boys live dangerously in the womb. Am J Hum Biol . 2010; 22( 3): 330– 335. Google Scholar CrossRef Search ADS PubMed  28. Retnakaran R, Kramer CK, Ye C, Kew S, Hanley AJ, Connelly PW, Sermer M, Zinman B. Fetal sex and maternal risk of gestational diabetes mellitus: the impact of having a boy. Diabetes Care . 2015; 38( 5): 844– 851. Google Scholar CrossRef Search ADS PubMed  29. Ilic S, Jovanovic L, Wollitzer AO. Is the paradoxical first trimester drop in insulin requirement due to an increase in C-peptide concentration in pregnant Type I diabetic women? Diabetologia . 2000; 43( 10): 1329– 1330. Google Scholar CrossRef Search ADS PubMed  30. Nielsen LR, Rehfeld JF, Pedersen-Bjergaard U, Damm P, Mathiesen ER. Pregnancy-induced rise in serum C-peptide concentrations in women with type 1 diabetes. Diabetes Care . 2009; 32( 6): 1052– 1057. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society

Journal

Journal of Clinical Endocrinology and MetabolismOxford University Press

Published: Mar 22, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off