Abstract
Objective
To compare perinatal outcomes for patients with diet‐controlled gestational diabetes mellitus (GDM), based on advancing gestational age at delivery.
Methods
Retrospective cohort study including patients delivering singleton, live born infants at ≥39 weeks’ gestation between November 2010 and February 2024. Among patients with diet‐controlled GDM at the time of delivery, perinatal outcomes were compared based on delivery timing, then compared to patients without diabetes. Statistical analyses included Mantel–Haenszel and Breslow–Day tests.
Results
Of 156,214 total deliveries, 4467 were to patients with diet‐controlled GDM at ≥39 weeks. When comparing deliveries at ≥41 weeks to those at 39 weeks, odds of primary cesarean delivery (CD) (adjusted odds ratio [aOR], 1.55; 95% confidence interval [CI], 1.22, 1.97), forceps delivery (aOR, 1.41; 95% CI, 1.27, 1.58), third‐ and fourth‐degree lacerations (aOR, 1.51; 95% CI, 1.32, 1.72), shoulder dystocia (aOR, 1.87; 95% CI, 1.49, 2.36), large for gestational age (LGA) infants (aOR, 1.48; 95% CI, 1.19, 1.83), and brachial plexus injury (aOR, 2.10; 95% CI, 1.53, 2.87) were higher after adjusting for age and parity. When compared to patients without diabetes, patients with diet‐controlled GDM experienced higher rates of most examined outcomes, but the rate of increase with advancing gestational age was not higher, except for primary CD and LGA infants (Breslow–Day p = 0.03 and 0.09, respectively).
Conclusion
Patients with diet‐controlled GDM have increased perinatal risks compared to those without GDM, which rise with advancing gestational age after 39 weeks. The risk of primary CD and LGA infants with advancing gestational age is exaggerated for patients with diet‐controlled GDM.
Keywords: delivery timing, diet‐controlled gestational diabetes, expectant management, gestational diabetes
1. INTRODUCTION
Gestational diabetes mellitus (GDM), the development of intolerance to carbohydrates during pregnancy, has been estimated to occur at a rate of 2%–11% in the United States and up to 14% globally [1, 2, 3, 4, 5]. GDM is associated with increased rates of maternal morbidity to include cesarean delivery (CD), pre‐eclampsia, and development of overt diabetes later in life as well as neonatal morbidity to include macrosomia, shoulder dystocia, and birth trauma [6, 7, 8, 9, 10, 11, 12]. Per the American College of Obstetricians and Gynecologists (ACOG), for patients with GDM who require insulin for management and who are well controlled on medication, delivery at 39 weeks is recommended [1, 13, 14]. For patients who are diet controlled, delivery prior to 39 weeks of gestation is not encouraged and expectant management up to 41 weeks is permissible [1].
There is mixed data on maternal and neonatal outcomes for patients with GDM who are scheduled for induction at 39 weeks compared to those who are managed expectantly up to 41 weeks. Some studies show no difference in neonatal outcomes and others demonstrate increased rates of large for gestational age (LGA) infants in patients expectantly managed or increased rates of hyperbilirubinemia in cases of induced patients [15, 16]. Further, data on the effect of expectantly managed GDM on rates of CD vary. Rates of CD in some retrospective studies are shown to be increased in cases of expectantly managed GDM and reduced in patients who are induced at 39 weeks [17]. While in some randomized controlled trials, there were no differences found when GDM patients were randomized to induction at 38 weeks versus expectant management up to 41 weeks of gestation [15, 16]. Additionally, there are few large, randomized trials that compare induction to expectant management for GDM and many studies include both patients with diet‐controlled GDM and patients with GDM requiring insulin together which can confound the findings [18]. There are no randomized controlled trials that compare induction to expectant management for patients with diet‐controlled GDM specifically; thus, recommendations for timing of delivery of these patients has remained the same. The objective of this study is to examine the effect of increasing gestational age at delivery on maternal and neonatal outcomes in patients with diet‐controlled GDM.
2. MATERIALS AND METHODS
This is a retrospective cohort study of patients with a diagnosis of diet‐controlled GDM at the time of delivery who delivered singleton, live born term infants at a single hospital between November 2010 and February 2024. Diagnosis and treatment of diet‐controlled GDM, including delivery at 420 weeks, was uniform based on institutional protocols and did not vary between providers. This study was approved by the institutional review board at our institution.
All patients at our institution receive a 50 g 1 h glucose tolerance test (GTT) during pregnancy at 24–28 weeks or earlier if they have a history of GDM in a prior pregnancy, glucosuria on urine dip, random glucose greater than or equal to 130 mg/dL, or symptoms of diabetes. If the 1 h GTT is greater than 140 mg/dL, the patient undergoes a 100 g 3 h GTT. At our institution we use the National Diabetes Data Group criteria for the 3 h GTT [19]. If on the 3 h GTT, patients have two or more abnormal values, they are diagnosed with GDM and sent to a centralized maternal‐fetal medicine (MFM) clinic for further evaluation. Patients with normal early testing for GDM will undergo repeat testing at 24–28 weeks as previously described.
In our MFM clinic, patients with a diagnosis of GDM attend a nutrition class and undergo education on how to use a glucometer. Our nutrition class that is provided consists of a standardized diet counseling class performed by registered dieticians in conjunction with diabetic nurse educators with additional one‐to‐one counseling as needed. Patients are instructed to check their glucose levels fasting and 2 h postprandial after every meal for a week. The week after this visit, they are seen by a provider and are instructed to come to the clinic fasting. Their capillary blood glucose (CBG) levels from the week before are reviewed and a fasting CBG is checked. Per ACOG recommendations, we encourage our patients to have glucose goals of fasting less than 95 and 2 h postprandial values less than 120 [1]. With the information obtained from their log and their fasting CBG, the decision is made to either continue management of GDM with diet or to start insulin if greater than 50% of glucose values are above goal. Patients with GDM are seen in our GDM clinic every 2 weeks with continued assessment of their CBGs at each visit. Patients who were initially diagnosed with diet‐controlled GDM but ultimately were started on medications were reclassified as insulin‐dependent GDM and were excluded from this analysis. We do not perform routine antenatal testing such as non‐stress tests or biophysical profiles on our diet‐controlled GDM population. Additionally, we do not routinely use ultrasound in this population for delivery planning nor do we routinely perform rate of growth ultrasounds in our diet‐controlled GDM population. Patients with diet‐controlled GDM are managed in our MFM clinic until delivery at 42 weeks.
Delivery of patients with diet‐controlled GDM prior to 42 weeks is only recommended for indications other than GDM. Indications for delivery other than GDM, although non‐encompassing here, include oligohydramnios, hypertension, diabetes requiring insulin, fetal growth restriction, and systemic lupus nephritis. Patients with the diagnoses noted previously were excluded from both cohorts in this analysis as they would likely have indications for earlier delivery prior to 39 weeks. Additionally, of note, although the ARRIVE trial was published during this time period, elective induction of labor at 39 weeks was not introduced as a practice change in our population. We expectantly managed our low‐risk patients until 42 weeks because of the volume of deliveries at our institution and thus the limited resources available on labor and delivery to accommodate a potential further increase in delivery volume. These practices are uniform at our institution and were in place for the duration of the study. Specifically for our patients with GDM, there were no practice changes made during the study period.
Maternal demographics and delivery and neonatal outcomes for patients with diet‐controlled GDM were compared based on advancing gestational age and then were secondarily compared to the non‐diabetic population. Patients excluded from the analysis included multiple gestations, those who delivered at 38 weeks or less and those who were diagnosed with GDM requiring insulin or pregestational diabetes. We do not routinely collect hemoglobin A1c levels on patients with GDM. Patients who were identified to have a history of diabetes outside of pregnancy and who were previously on medications for this purpose were diagnosed with overt, pregestational diabetes and thus were excluded from this analysis. Neonates with major fetal malformations were excluded from the analysis of neonatal outcomes. The data for this study were obtained from an established obstetrical quality database that contains data extracted from the electronic medical record (EMR) by dedicated research nurses according to standard protocols and definitions contained within a manual of operations and routinely validated with cross‐checks.
Maternal demographics examined included maternal age, race, body mass index (BMI), and parity. Delivery outcomes of interest include mode of delivery, rate of labor induction, rate of third‐ or fourth‐degree vaginal lacerations, rate of forceps assisted vaginal delivery, rate of postpartum hemorrhage (PPH), and rate of shoulder dystocia. Vacuum‐assisted vaginal deliveries are not performed at our institution, so were not included in the delivery outcomes. For the calculations of rates of vaginal and primary CD, patients who underwent a repeat CD were excluded. For the calculations of rates of forceps deliveries, third‐ or fourth‐degree lacerations, and shoulder dystocia, only patients who underwent a vaginal delivery were included. Neonatal outcomes of interest included rates of small for gestational age (SGA), LGA, neonatal birthweight (in grams), brachial plexus palsy, clavicular fracture, major malformations, hyperbilirubinemia, hypoglycemia, and development of transient tachypnea of the newborn (TTN).
For patients with diet‐controlled GDM, perinatal outcomes were compared across gestational age categories using the Pearson chi‐square for categorical measures and analysis of variance for continuous measures. Trend tests were incorporated (Cochrane–Armitage for categorical measures and linear regression for continuous measures) to examine the alternate hypothesis of monotonic increase or decrease in rates or means, respectively. When examining the similarity of the association between outcome and patients with diet‐controlled GDM versus patients without GDM across gestational age, the Breslow–Day test was used to discern if the odds ratios (OR; the effect size for the strength of the relationship) were similar. As a conservative nature, a significance level of 0.10 was used for significance in this test. If the Breslow–Day test was significant then the association between diabetes classification and outcome was judged different across gestational age, meaning that the association depends on gestational age. In this case, ORs (95% confidence intervals) were estimated at each gestational age independently. If the Breslow–Day test was not significant then the association between diabetes classification (patients with diet‐controlled GDM vs. others) and outcome is assumed independent of gestational age. The overall gestational age adjusted ORs (95% confidence intervals [CI]) was estimated by the Cochran–Mantel–Haenszel method. CIs for rates were determined using the Clopper–Pearson method. Statistical analyses were conducted with the use of SAS software, version 9.4.
3. RESULTS
Of 156,214 total deliveries, there were 8208 deliveries of patients with diet‐controlled GDM. A total of 4467 (54%) of these patients were delivered at ≥39 weeks. A total of 2456 (55%) patients with diet‐controlled GDM delivered at 390–396 weeks, 1376 (30.8%) at 400–406 weeks, and 635 (14%) at ≥410 weeks (Figure 1). Patients who delivered at ≥410 weeks were older (31.3 ± 6.4 years), more commonly African American (10.2% vs. 7.4% vs. 8.2%, p < 0.001), and more likely to be nulliparous (28% vs. 15.6% vs. 22.8%, p < 0.001), when compared to patients who delivered at 390–396 and 400–406 weeks (Table 1).
FIGURE 1.
Flow diagram of study population. *Reasons for exclusion overlap such that the totals of the populations reported do not account for potential overlap with the other criteria for exclusion. GDM, gestational diabetes mellitus.
TABLE 1.
Maternal demographics based on advancing gestational age in patients with diet‐controlled GDM.
| Diet‐controlled GDM—gestational age at delivery in weeks | Non‐GDM—gestational age at delivery in weeks | |||||||
|---|---|---|---|---|---|---|---|---|
| Demographic |
39 weeks N = 2456 |
40 weeks N = 1376 |
≥41 weeks N = 635 |
p value * |
39 weeks N = 43,335 |
40 weeks N = 31,205 |
≥41 weeks N = 17,890 |
p value ** |
| Maternal age | 32.0 ± 6.1 | 31.6 ± 6.4 | 31.3 ± 6.4 | <0.001 | 27.6 ± 6.2 | 26.8 ± 6.3 | 26.7 ± 6.1 | <0.001 |
| Race/ethnicity | <0.001 | <0.001 | ||||||
| Black | 182 (7.4) | 113 (8.2) | 65 (10.2) | 5963 (13.8) | 4008 (12.8) | 2569 (14.4) | ||
| White (Non‐ Hispanic) | 56 (2.3) | 23 (1.7) | 22 (3.5) | 1373 (3.2) | 974 (3.1) | 552 (3.1) | ||
| White (Hispanic) | 2070 (84.3) | 1169 (85) | 518 (81.6) | 34,542 (79.7) | 25,201 (80.8) | 14,203 (79.4) | ||
| Other | 148 (6.0) | 71 (5.2) | 30 (4.7) | 1457 (3.4) | 1022 (3.3) | 566 (3.2) | ||
| BMI at delivery | 34.1 ± 5.7 | 34.0 ± 5.9 | 34.2 ± 5.6 | 0.88 | 32.3 ± 5.9 | 32.3 ± 5.8 | 33.0 ± 6.0 | <0.001 |
| Nulliparous | 382 (15.6) | 314 (22.8) | 178 (28.0) | <0.001 | 10,768 (24.8) | 10,860 (34.8) | 7068 (39.5) | <0.001 |
Note: Data reported as mean ± SD or n (%).
Abbreviations: BMI, body mass index; GDM, gestational diabetes mellitus.
p value compares the GDM population based on advancing gestational age.
p value compares the non‐GDM population based on advancing gestational age.
When looking at delivery outcomes, the rate of primary CD increased with increasing gestational age (17.1% vs. 16.8% vs. 24.8%, p < 0.001). This included an increased rate of primary CD for both fetal distress (6.7% vs. 8.1% vs. 11.4%, p < 0.001) and failure to progress in labor (4.5% vs. 5.8% vs. 8.1%, p < 0.001). Of patients who had a vaginal delivery, rates of forceps delivery (3% vs. 4.5% vs. 5.2%, p = 0.05), third‐ or fourth‐degree lacerations (2.5% vs. 2.6% vs. 3.9%, p = 0.27), and shoulder dystocia (1.5% vs. 2.2% vs. 4.0%, p = 0.11) were not statistically significant (Table 2). After excluding neonates with malformations, birthweight (3526 ± 459 g vs. 3651 ± 471 g vs. 3756 ± 480 g, p < 0.001) and rate of LGA (17.9% vs. 23.1% vs. 23.3%, p < 0.001) increased with increasing gestational age. There were no differences found in terms of rates of brachial plexus palsy, clavicular fracture, hyperbilirubinemia, or TTN (Table 2).
TABLE 2.
Delivery and neonatal outcomes based on advancing gestational age in patients with diet‐controlled GDM.
| Gestational age at delivery in weeks | ||||
|---|---|---|---|---|
| Outcome | 39 weeks N = 2456 | 40 weeks N = 1376 | ≥41 weeks N = 635 | p value * |
| Repeat CD | 911 (37.1) | 88 (6.4) | 21 (3.3) | <0.001 |
| Primary CD a | 264 (17.1) | 217 (16.8) | 152 (24.8) | <0.001 |
| Fetal distress a | 103 (6.7) | 104 (8.1) | 70 (11.4) | <0.001 |
| Presentation a | 42 (2.7) | 15 (1.2) | 16 (2.6) | 0.01 |
| FTP a | 69 (4.5) | 75 (5.8) | 50 (8.1) | <0.001 |
| Other a | 50 (3.2) | 23 (1.8) | 16 (2.6) | 0.05 |
| Vaginal delivery a | 1281 (82.9) | 1071 (83.2) | 462 (75.2) | <0.001 |
| Forceps b | 38 (3.0) | 48 (4.5) | 24 (5.2) | 0.05 |
| Third‐ or fourth‐degree laceration b | 32 (2.5) | 28 (2.6) | 18 (3.9) | 0.27 |
| PPH | 221 (9) | 77 (5.6) | 52 (8.2) | <0.001 |
| Shoulder dystocia b | 19 (1.5) | 24 (2.2) | 14 (4.0) | 0.11 |
| SGA (<10th percentile) | 162 (6.6) | 52 (3.8) | 25 (3.9) | <0.001 |
| LGA (>90th percentile) | 440 (17.9) | 318 (23.1) | 148 (23.3) | <0.001 |
| Birthweight (g) | 3526 ± 459 | 3651 ± 471 | 3756 ± 480 | <0.001 |
| Brachial plexus injury | 8 (0.3) | 11 (0.8) | 2 (0.3) | 0.10 |
| Clavicular fracture | 36 (1.5) | 35 (2.5) | 13 (2.0) | 0.06 |
| Hyperbilirubinemia | 56 (2.3) | 25 (1.8) | 8 (1.3) | 0.22 |
| Hypoglycemia | 98 (6.7) | 63 (7.4) | 31 (9.0) | 0.33 |
| TTN | 11 (0.5) | 5 (0.4) | 7 (1.1) | 0.08 |
Note: Data reported as n (%).
Abbreviations: CD, cesarean delivery; FTP, failure to progress (encompasses both failure to progress in active labor, cephalopelvic disproportion, and failed induction); LGA, large for gestational age; PPH, postpartum hemorrhage; SGA, small for gestational age; TTN, transient tachypnea of the newborn.
Denominator representing patients without repeat CD (n = 3447).
Denominator representing patients only with vaginal deliveries (n = 2814).
p value comparing outcomes of patients with diet‐controlled GDM based on gestational age.
A logistic regression adjusted for age and parity was performed comparing the same outcomes based on advancing gestational age. The rate of primary CD again was shown to increase when comparing patients delivered at gestational ages ≥ 41weeks to both patients delivered at 39 weeks and 40 weeks, respectively (OR 1.55 (95% CI 1.22, 1.97) and OR 1.58 (95% CI 1.23, 2.03)). The adjusted ORs for primary CD for fetal distress or failure to progress were only significant when comparing patients ≥41 weeks to those delivered at 39 weeks (OR, 1.72; 95% CI, 1.24, 2.40 and OR, 1.77; 95% CI, 1.19, 2.63). For patients who underwent a vaginal delivery, adjusted odds of forceps delivery and third‐ or fourth‐degree vaginal lacerations were elevated for advancing gestational age but were highest when comparing those who delivered at ≥41 weeks compared to those who delivered at 39 weeks (OR, 1.41; 95% CI, 1.27, 1.58 and OR, 1.51; 95% CI, 1.32, 1.72, respectively). The adjusted odds of shoulder dystocia and clavicular fracture also increased when comparing those who were delivered at ≥41 weeks to those delivered at 39 and 40 weeks and were highest for those who delivered at ≥41 weeks compared to those who delivered at 39 weeks (OR, 1.87; 95% CI, 1.49, 2.36 and OR, 1.58; 95% CI, 1.35, 1.84, respectively). The adjusted odds of LGA increased with increasing gestational age but again were highest for those who delivered ≥41 weeks when compared to those who delivered at 39 weeks (OR, 1.48; 95% CI, 1.19, 1.83). The adjusted odds of brachial plexus injury increased with increasing gestational age but were highest for those who delivered at ≥41 weeks when compared to those who delivered at 39 weeks (OR, 2.10; 95% CI, 1.53, 2.87) (Table 3).
TABLE 3.
Logistic regression and adjusted logistic regression models for delivery outcomes in patients with diet‐controlled GDM based on advancing gestational age.
| Gestational age at delivery in weeks | ||||||
|---|---|---|---|---|---|---|
| Outcome | 40 versus 39 weeks c | 41 versus 39 weeks c | 41 weeks versus 40 weeks c | 40 versus 39 weeks d | 41 versus 39 weeks d | 41 versus 40 weeks d |
| Repeat CD | 0.12 (0.09, 0.15) | 0.06 (0.04, 0.09) | 0.50 (0.31, 0.81) | 0.11 0.09, 0.15) | 0.06 (0.04, 0.09) | 0.52 (0.32, 0.86) |
| Primary CD a | 0.98 (0.81, 1.20) | 1.60 (1.27, 2.0) | 1.62 (1.28, 2.05) | 0.98 (0.80, 1.21) | 1.55 (1.22, 1.97) | 1.58 (1.23, |
| Forceps b | 1.19 (1.08, 1.31) | 1.47 (1.32, 1.64) | 1.24 (1.11, 1.38) | 1.17 (1.06, 1.29) | 1.41 (1.27, 1.58) | 1.21 (1.08, 1.35) |
| Third‐ or fourth‐degree laceration b | 1.21 (1.08, 1.36) | 1.59 (1.39, 1.81) | 1.31 (1.15, 1.49) | 1.18 (1.05, 1.33) | 1.51 (1.32, 1.72) | 1.27 (1.12, 1.45) |
| PPH | 0.60 (0.46, 0.78) | 0.90 (0.66, 1.24) | 1.50 (1.04, 2.17) | 0.57 (0.43, 0.74) | 0.83 (0.60, 1.15) | 1.47 (1.02, 2.12) |
| Shoulder dystocia b | 1.54 (1.25, 1.89) | 1.85 1.47, 2.33) | 1.20 (0.97, 1.49) | 1.54 (1.25, 1.89) | 1.87 (1.49, 2.36) | 1.22 (0.98, 1.51) |
| SGA (<10th percentile) | 0.51 (0.37, 0.72) | 0.60 (0.39, 0.93) | 1.18 (0.72, 1.94) | 0.48 (0.34, 0.67) | 0.53 (0.34, 0.82) | 1.11 (0.68, 1.83) |
| LGA (>90thpercentile) | 1.37 (1.16, 1.62) | 1.39 (1.13, 1.72) | 1.02 (0.81, 1.27) | 1.41 (1.19, 1.66) | 1.48 (1.19, 1.83) | 1.05 (0.84, 1.32) |
| Brachial plexus injury | 1.55 (1.16, 2.09) | 2.20 (1.61, 3.01) | 1.42 (1.05, 1.92) | 1.50 (1.12, 2.02) | 2.10 (1.53, 2.87) | 1.40 (1.03, 1.90) |
| Clavicular fracture | 1.41 (1.24, 1.61) | 1.47 (1.27, 1.71) | 1.04 (0.90, 1.21) | 1.48 (1.30, 1.69) | 1.58 (1.35, 1.84) | 1.06 (0.91, 1.24) |
| Hyperbilirubinemia | 1.08 (0.98, 1.20) | 1.07 (0.95, 1.21) | 0.99 (0.87, 1.12) | 1.06 (0.96, 1.17) | 1.03 (0.92, 1.16) | 0.98 (0.86, 1.11) |
| Hypoglycemia | 1.12 (1.00, 1.26) | 1.09 (0.95, 1.26) | 0.97 (0.84, 1.13) | 1.13 (1, 1.27) | 1.10 (0.95, 1.26) | 0.97 (0.84, 1.12) |
| TTN | 0.88 (0.70, 1.11) | 1.08 (0.83, 1.39) | 1.22 (0.92, 1.61) | 0.85 (0.67, 1.07) | 1.02 (0.78, 1.32) | 1.20 (0.91, 1.58) |
Abbreviations: CI, confidence interval; CD, cesarean delivery; FTP, Failure to progress; LGA, large for gestational age; PPH, postpartum hemorrhage; SGA, small for gestational age; TTN, transient tachypnea of the newborn.
Denominator representing patients without repeat CD (n = 3447).
Denominator representing patients only with vaginal deliveries (n = 2814).
Unadjusted odds ratio with 95% CI.
Odds ratio with 95% CI adjusted for age and parity.
Comparing outcomes between patients with diet‐controlled GDM and those without diabetes, all outcomes were increased in the cohort of patients with diet‐controlled GDM except for SGA, hyperbilirubinemia, and TTN (Table 4). Shoulder dystocia specifically was 2.51 times more likely to occur in patients with diet‐controlled GDM when compared to the cohort without diabetes. The rate of increase of each outcome was not substantially higher in patients with diet‐controlled GDM with Breslow–Day p values > 0.10 except in the case of primary CD and rate of LGA infants (Table 4). Primary CD and LGA yielded Breslow–Day p values of 0.03 and 0.09, respectively, which represented that the rate of primary CD and LGA infants increased at a higher rate in patients with diet‐controlled GDM with increasing gestational age when compared to patients without diabetes (Figure 2).
TABLE 4.
Adjusted logistic regression and hypothesis of equal strengths of association between patients with diet‐controlled GDM and non‐diabetics and outcomes of interest.
| Outcome | Odds ratio (95% CI) a | Breslow–Day b |
|---|---|---|
| Primary CD d | 1.26 (1.14, 1.38) | 0.03 |
| Cesarean for fetal distress d | 1.14 (1.00, 1.30) | 0.24 |
| Cesarean for failure to progress d | 1.33 (1.14, 1.56) | 0.57 |
| Forceps c | 1.31 (1.07, 1.60) | 0.49 |
| Third‐ or fourth‐degree laceration c | 1.45 (1.14, 1.85) | 0.83 |
| PPH | 1.10 (0.98, 1.23) | 0.28 |
| Shoulder dystocia c | 2.51 (1.89, 3.34) | 0.94 |
| SGA | 0.84 (0.73, 0.97) | 0.50 |
| LGA | 1.54 (1.42, 1.66) | 0.09 |
| Brachial plexus palsy | 1.81 (1.13, 2.89) | 0.23 |
| Clavicular fracture | 1.37 (1.09, 1.72) | 0.51 |
| Hyperbilirubinemia | 0.93 (0.75, 1.16) | 0.12 |
| TTN | 1.25 (0.82, 1.92) | 0.14 |
Abbreviations: CI, confidence interval; CD, cesarean delivery; LGA, large for gestational age; PPH, postpartum hemorrhage; SGA, small for gestational age; TTN, transient tachypnea of the newborn.
Adjusted odds ratio (OR) comparing all term patients with diet‐controlled GDM to term non‐diabetic patients, OR adjusted for age, nulliparity, race, and body mass index.
Breslow–Day test examining the hypothesis of equal strengths of association between patients with diet‐controlled GDM or non‐diabetics and outcome of interest.
Denominator representing patients only with vaginal deliveries (n = 2814).
Denominator representing patients without repeat CD (n = 3447).
FIGURE 2.
Rate of primary CD with increasing gestational age. Breslow–Day test p value = 0.03, indicating the frequency of primary CD increases at a faster rate for women with diet‐controlled GDM when compared to non‐diabetics. CI, confidence interval; CD, cesarean delivery; GDM, gestational diabetes mellitus.
4. DISCUSSION
Morbidity increased with advancing gestational age in patients with GDM. Primary CD and LGA rates specifically are at baseline higher in patients with diet‐controlled GDM and both worsen with increasing gestational age, especially for patients who are delivered at ≥41 weeks. Importantly, the rate at which primary CD and LGA infants occurred with advancing gestational age was more rapid than that seen in patients without diabetes, emphasizing an additional benefit of 39 week delivery in a population with GDM.
Studies on timing of delivery in patients with GDM are limited and often report mixed results with some studies showing maternal benefit with scheduled induction and others showing no difference with worsening of neonatal outcomes. A retrospective study by Lurie and colleagues in the 1990s demonstrated no difference in the rate of CD in cases of expectantly managed diet‐controlled GDM and GDM requiring insulin that delivered after 40 weeks of gestation [20]. Randomized controlled trials have been subsequently completed to further examine delivery timing in the setting of GDM. Kjos and colleagues as well as Alberico and colleagues attempted to compare outcomes in cases of expectantly managed GDM versus planned delivery at 38 weeks gestation [15, 16]. Both studies showed that there was no difference in rate of CD, although they both noted higher rates of shoulder dystocia in the patients expectantly managed. Kjos and colleagues specifically reported on a greater incidence of LGA infants specifically in the expectantly managed group [15]. This study included patients with GDM requiring insulin and even pregestational diabetes while the study by Alberico and colleagues included all types of GDM. Rates of outcomes for those with diet‐controlled GDM versus GDM requiring insulin were not available in the study by Kjos et al. and rates for patients with diet‐controlled GDM in general were not reported by Alberico et al [16]. Both of these studies were not specifically powered to detect a difference in rate of CD for patients with diet‐controlled GDM.
Contrary to the above‐mentioned studies, a more contemporary, retrospective study by Melamed et al. demonstrated that routine induction at 38 or 39 weeks was associated with lower risk of CD, but this study did note increased potential risk of admission of the neonate to the neonatal intensive care unit (NICU) [17]. Again, this study included both patients with diet‐controlled GDM and patients with GDM requiring insulin. A decision‐analytic model was built to compare outcomes of patients with diet‐controlled GDM specifically in a theoretical cohort of 100,000 women by Niu and colleagues in 2014. Their model demonstrated that delivery of patients with diet‐controlled GDM at 38 weeks reduced the rate of stillbirths, maternal death, and CD [21]. Although these studies provide important information, decisions for delivery timing cannot be based solely on a theoretical model or the currently available small studies including all types of GDM.
Our study agrees with prior data that show increased risk of maternal and neonatal outcomes in patients with diet‐controlled GDM when compared to non‐diabetics [6, 7, 8, 9, 10, 11, 12]. Our study also aligns with previous findings by Melamed et al. and Niu et al. of increased rates of CD in patients with diet‐controlled GDM expectantly managed after 39 weeks. It also shows increased risk of common associated adverse outcomes associated with GDM such as shoulder dystocia, third‐ and fourth‐degree lacerations, brachial plexus injuries, and clavicular fracture, especially when comparing patients delivered greater than 40 weeks. The data did not show that neonatal outcomes or maternal outcomes in the exception of the primary CD rate and the rate of LGA infants increased at a higher rate with increasing gestational age when compared to non‐diabetics. Of the current clinical data that are available, our study is the only study that has focused on patients with diet‐controlled GDM alone and shows statistically trends of worsening morbidity with increasing gestational age, especially after 40 weeks, as it relates within and outside of the population with diet‐controlled GDM.
This study implies that patients with diet‐controlled GDM do have increased risk for maternal and neonatal morbidity and thus should not be considered similar to low‐risk pregnancies. In general, patients with diet‐controlled GDM can be managed expectantly safely but there should be consideration of the interplay of gestational age, the presence of GDM, and the rate of primary CD and LGA. Although overall maternal and neonatal morbidity does not increase at a faster rate in the patient population with diet‐controlled GDM when compared to the non‐diabetic population, the rate of primary CD and LGA infants is significant enough that consideration should be given to earlier delivery to reduce associated risks of primary CD and LGA infants.
Although the ARRIVE trial was introduced during the time frame of our current study, its findings, although important, were not used to implement any changes to our labor induction practices. The patients in this study were cared for at a single large, county hospital where elective induction is not offered and routine induction if no other indications exist is not planned until 42 weeks. Since we have not implemented changes based on the ARRIVE trial, we believe this study aligns with the ARRIVE trial showing that possibly labor induction could result in a lower rate of CD, even in a population that is not low risk such as diet‐controlled GDM [22].
Further research is needed at multiple centers that explore the same findings presented here so that the results can be more generalizable. Prospective data in the way of randomized controlled trials are sorely needed to further support the findings in this study and to optimize care of patients with diet‐controlled GDM specifically. The optimal timing of delivery of patients with diet‐controlled GDM continues to be uncertain but further prospective research can shed more light on how to optimize care for this population.
One of the major strengths of this study is it is one of the few to show retrospectively, the effect of increasing gestational age on maternal and neonatal outcomes specifically in the population with diet‐controlled GDM. It also not only combines outcomes between patients with diet‐controlled GDM alone but also compares this to non‐diabetics and shows trends of outcomes and how these compare between those with and without diet‐controlled GDM. Another major strength of these data is that these are collected from a single institution, where the diagnosis and management of diet‐controlled GDM is uniform and all patients with GDM managed with diet only are not delivered until 42 weeks unless other indications for delivery arise prior to this set gestational age.
One limitation of this study is that it is retrospective in nature. This population is not representative of the entire obstetric population worldwide so the findings of this study may not be completely generalizable. The threshold we use for our 1‐h GTT is a less conservative screening approach and the cutoffs we use for the 3‐h GTT are less inclusive such that the data here may not be applicable to practices that use more conservative screening thresholds such as 130 or 135 for the 1‐h GTT or the Carpenter Coustan cutoffs for the 3‐h GTT. There was a potential for provider bias as providers may not manage labor the same in patients with GDM compared to those without. Notably, at our institution, labor is protocolized with use of the same induction agents and labor curves, thus we believe this provider bias if present was likely low. An additional limitation is that we did not have data on overall glucose control of the patients included in this study so we cannot make a statement about their control and how this affected their delivery outcomes. Finally, the non‐diabetic population that was chosen for comparison of outcomes to the patient population with diet‐controlled GDM is not a representation of only low‐risk pregnancies. Patients delivered at 39 weeks were higher risk patients as they had indications for earlier delivery. This may provide bias in our conclusions but we assume that the findings here may actually underestimate the differences described. We hypothesize prospective studies may show that outcomes, including the rate of CD and LGA infants, are even worse in patients delivered after 39 weeks if other comorbidities are controlled for.
5. CONCLUSION
Patients with diet‐controlled GDM are at increased risk of both maternal and neonatal morbidity when compared to a non‐diabetic population. These risks increase with increasing gestational age but for the most part not at a rate that is higher than the non‐diabetic population. The only exceptions are the rate of primary CD and LGA infants which increase with increasing gestational age at a faster rate when compared to patients without diabetes. Delivery at 39–40 weeks for these patients may be associated with lower rates of primary CD and LGA infants but further prospective research is needed to truly define the optimal delivery timing for this population.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
FUNDING INFORMATION
The authors received no specific funding for this work.
This research was presented as a poster presentation at the Society of Maternal‐Fetal Medicine Pregnancy Meeting on January 30, 2025, Denver, Colorado.
REFERENCES
- 1. 2018. “ACOG Practice Bulletin No. 190: Gestational Diabetes Mellitus.” Obstetrics and Gynecology 131(2): e49–e64. 10.1097/aog.0000000000002501. [DOI] [PubMed] [Google Scholar]
- 2. Osterman, M. , Hamilton B., Martin J. A., Driscoll A. K., and Valenzuela C. P.. 2021. “Births: Final Data for 2020.” National Vital Statistics Reports 70(17): 1–50. [PubMed] [Google Scholar]
- 3. Sweeting, A. , Hannah W., Backman H., Catalano P., Feghali M., Herman W. H., Hivert M.‐F., et al. 2024. “Epidemiology and Management of Gestational Diabetes.” Lancet 404(10448): 175–92. 10.1016/s0140-6736(24)00825-0. [DOI] [PubMed] [Google Scholar]
- 4. 2023. “QuickStats: Percentage of Mothers With Gestational Diabetes,* by Maternal Age—National Vital Statistics System, United States, 2016 and 2021.” MMWR Morbidity and Mortality Weekly Report 72(1): 16. 10.15585/mmwr.mm7201a4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gregory, E. C. and Ely D. M.. 2022. “Trends and Characteristics in Gestational Diabetes: United States, 2016‐2020.” National Vital Statistics Reports 71(3): 1–15. [PubMed] [Google Scholar]
- 6. Yogev, Y. , Xenakis E. M., and Langer O.. 2004. “The Association Between Preeclampsia and the Severity of Gestational Diabetes: The Impact of Glycemic Control.” American Journal of Obstetrics and Gynecology 191(5): 1655–60. 10.1016/j.ajog.2004.03.074. [DOI] [PubMed] [Google Scholar]
- 7. Ehrenberg, H. M. , Durnwald C. P., Catalano P., and Mercer B. M.. 2004. “The Influence of Obesity and Diabetes on the Risk of Cesarean Delivery.” American Journal of Obstetrics and Gynecology 191(3): 969–74. 10.1016/j.ajog.2004.06.057. [DOI] [PubMed] [Google Scholar]
- 8. Kim, C. , Newton K. M., and Knopp R. H.. 2002. “Gestational Diabetes and the Incidence of Type 2 Diabetes: A Systematic Review.” Diabetes Care 25(10): 1862–8. 10.2337/diacare.25.10.1862. [DOI] [PubMed] [Google Scholar]
- 9. Ehrenberg, H. M. , Mercer B. M., and Catalano P. M.. 2004. “The Influence of Obesity and Diabetes on the Prevalence of Macrosomia.” American Journal of Obstetrics and Gynecology 191(3): 964–8. 10.1016/j.ajog.2004.05.052. [DOI] [PubMed] [Google Scholar]
- 10. 2020. “Macrosomia: ACOG Practice Bulletin, Number 216.” Obstetrics and Gynecology 135(1): e18–e35. 10.1097/aog.0000000000003606. [DOI] [PubMed] [Google Scholar]
- 11. HAPO Study Cooperative Research Group ; Metzger, B. E. , Lowe L. P., Dyer A. R., Trimble E. R., Chaovarindr U., Coustan D. R., et al. 2008. “Hyperglycemia and Adverse Pregnancy Outcomes.” New England Journal of Medicine 358(19): 1991–2002. 10.1056/NEJMoa0707943. [DOI] [PubMed] [Google Scholar]
- 12. Casey, B. 1997. “Pregnancy Outcomes in Women With Gestational Diabetes Compared With the General Obstetric Population.” Obstetrics and Gynecology 90(6): 869–73. 10.1016/s0029-7844(97)00542-5. [DOI] [PubMed] [Google Scholar]
- 13. 2013. “ACOG Committee Opinion no. 560: Medically Indicated Late‐preterm and Early‐term Deliveries.” Obstetrics and Gynecology 121(4): 908–10. 10.1097/01.Aog.0000428648.75548.00. [DOI] [PubMed] [Google Scholar]
- 14. Spong, C. Y. , Mercer B. M., D'Alton M., Kilpatrick S., Blackwell S., and Saade G.. 2011. “Timing of Indicated Late‐Preterm and Early‐Term Birth.” Obstetrics and Gynecology 118(2 Pt 1): 323–33. 10.1097/AOG.0b013e3182255999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kjos, S. L. , Henry O. A., Montoro M., Buchanan T. A., and Mestman J. H.. 1993. “Insulin‐Requiring Diabetes in Pregnancy: A Randomized Trial of Active Induction of Labor and Expectant Management.” American Journal of Obstetrics and Gynecology 169(3): 611–5. 10.1016/0002-9378(93)90631-r. [DOI] [PubMed] [Google Scholar]
- 16. Alberico, S. , Erenbourg A., Hod M., Yogev Y., Hadar E., Neri F., Ronfani L., and Maso G., 2017. “Immediate Delivery or Expectant Management in Gestational Diabetes at Term: the GINEXMAL Randomised Controlled Trial.” BJOG 124(4): 669–77. 10.1111/1471-0528.14389. [DOI] [PubMed] [Google Scholar]
- 17. Melamed, N. , Ray J. G., Geary M., Bedard D., Yang C., Sprague A., Murray‐Davis B., Barrett J., and Berger H., 2016. “Induction of Labor Before 40 Weeks Is Associated With Lower Rate of Cesarean Delivery in Women With Gestational Diabetes Mellitus.” American Journal of Obstetrics and Gynecology 214(3): 364.e1–e8. 10.1016/j.ajog.2015.12.021. [DOI] [PubMed] [Google Scholar]
- 18. Witkop, C. T. , Neale D., Wilson L. M., Bass E. B., and Nicholson W. K.. 2009. “Active Compared With Expectant Delivery Management in Women With Gestational Diabetes: A Systematic Review.” Obstetrics and Gynecology 113(1): 206–17. 10.1097/AOG.0b013e31818db36f. [DOI] [PubMed] [Google Scholar]
- 19. 1979. “Classification and Diagnosis of Diabetes Mellitus and Other Categories of Glucose Intolerance. National Diabetes Data Group.” Diabetes 28(12): 1039–57. 10.2337/diab.28.12.1039. [DOI] [PubMed] [Google Scholar]
- 20. Lurie, S. , Matzkel A., Weissman A., Gotlibe Z., and Friedman A.. 1992. “Outcome of Pregnancy in Class A1 and A2 Gestational Diabetic Patients Delivered Beyond 40 Weeks' Gestation.” American Journal of Perinatology 9(5‐6): 484–8. 10.1055/s-2007-999294. [DOI] [PubMed] [Google Scholar]
- 21. Niu, B. , Lee V. R., Cheng Y. W., Frias A. E., Nicholson J. M., and Caughey A. B.. 2014. “What Is the Optimal Gestational Age for Women With Gestational Diabetes Type A1 to Deliver?.” American Journal of Obstetrics and Gynecology 211(4): 418.e1–e6. 10.1016/j.ajog.2014.06.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Grobman, W. A. , Rice M. M., Reddy U. M., Tita A. T. N., Silver R. M., Mallett G., Hill K., et al. 2018. “Labor Induction Versus Expectant Management in Low‐Risk Nulliparous Women.” New England Journal of Medicine 379(6): 513–23. 10.1056/NEJMoa1800566. [DOI] [PMC free article] [PubMed] [Google Scholar]