If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Department of Clinical Science and Education, Karolinska Institutet, Women’s clinic, Sodersjukhuset, 118 83 Stockholm, Sweden.
Sweden had 29 different guidelines for type 1 diabetes during pregnancy and labor with no consensus.
•
Medical staff’s adherence to the three local guidelines in Stockholm was low.
•
A national guideline might improve the health of pregnant women with type 1 diabetes and their fetuses.
Abstract
Aims
Type 1 diabetes (DM1) during pregnancy and labor is associated with an increased risk of maternal and fetal complications. Evidence-based care is therefore provided in accordance with guidelines.
In this study, we aimed to compare all the Swedish guidelines for DM1 during pregnancy and labor in terms of the variables emphasized in the national guidelines from the US and from England and Wales. The second aim was to measure adherence to local guidelines at the four hospitals in Stockholm that cared for pregnant women with DM1 during 2016 and to describe the pregnancy and labor outcomes.
Methods
All the Swedish guidelines for DM1 during pregnancy and labor were reviewed on 31 variables. The medical records of 114 women were reviewed according to whether ≥70% of 22 variables in the guidelines were followed.
Results
No consensus was found in the Swedish guidelines for any of the 31 variables. Some guidelines were contradictory. The pregnancy guidelines were followed in 17.5% of the medical records, 18.4% followed the labor guidelines, and 5.3% followed both guidelines. The onset of labor, mode of delivery and HbA1c in the third trimester varied significantly, depending on the adherence to guidelines.
Conclusions
The Swedish guidelines for DM1 during pregnancy and labor lack both consensus and adherence. A national guideline on DM1 during pregnancy and childbirth with high adherence could improve care for pregnant Swedish women with DM1 and their fetuses.
]. DM1 during pregnancy and labor is associated with an increased risk of maternal and fetal complications, and pregnant women with DM1 have an increased risk of developing preeclampsia. The fetus also has a higher risk of neonatal morbidity and mortality due to congenital malformations, being large for gestational age (LGA), shoulder dystocia, cesarean sectio, preterm birth, hypoglycemia, hyper-bilirubin, and stillbirth [
Previous research has shown that pregnancy complications among women with DM1 are associated with increased hemoglobin A1c (HbA1c) up to three months before and throughout pregnancy. The current consensus is that evidence-based pre-pregnancy counseling and care during pregnancy and labor are paramount for the optimal management of pregnancies complicated with DM1 [
In Sweden, 42 obstetric clinics offer care for pregnant women with DM1. In some clinics, care is provided in special maternity wards, while in other settings, it is provided in collaboration with endocrinology/diabetology departments. Swedish antenatal care is free of charge, and each patient has an equal right to optimal, evidence-based care. To support clinicians in providing evidence-based care, local guidelines are used in everyday practice. By contrast, England and Wales [
] have national guidelines for this group of pregnant women.
In this study, we aimed to compare all the Swedish guidelines for DM1 during pregnancy and labor in terms of the variables emphasized in the US and the England and Wales national guidelines. The second aim was to measure adherence to local guidelines at the four hospitals in Stockholm that cared for pregnant women with DM1 during 2016 and to describe the pregnancy and labor outcomes.
2. Methods
We asked each obstetrical clinic in Sweden to send us their local guidelines on DM1 during pregnancy and labor that were in use between 2015 and 2017. The guidelines were reviewed in terms of 31 variables (Table 1) recommended in the US national guideline on pregestational diabetes [
]. All the Swedish guidelines were reviewed according to which of the 31 variables were recommended. If the variables were recommended, the recommendations were categorized into groups according to the level of compliance (<70% and ≥70%) and compared.
Table 1Swedish guidelines for type 1 diabetes during pregnancy and labor.
In the second part of the study, pregnancy and labor data were collected after obtaining written consent from 114 women with DM1 who gave birth at one of the four Stockholm hospitals that cared for this patient group during 2016. Only women with singleton pregnancies and who received prenatal and intrapartum care at the same hospital were invited to participate. Their medical records were reviewed according to 22 predetermined variables to see whether the medical staff had followed their local guidelines (Table 2). We chose the cutoff of >70% for adherence to the guidelines, based on a study that concluded that adherence to a guideline for diabetes was typically between 60% and 80% [
]. The 22 variables were the same as the previous 31 but were modified to fit the review of the women’s medical records. If the guideline did not recommend a variable, none of the medical records that followed that guideline were reviewed for adherence to that specific variable.
Table 2Adherence to the guidelines in the Stockholm cohort.
Recommendation
N
%
First antenatal visit
Yes
51
44.7
No
6
5.3
Not in the guideline
57
50.0
Insulin regime
Yes
57
50.0
No
57
50.0
Not in the guideline
0
0.0
Folic acid
Yes
0
0.0
No
0
0.0
Not in the guideline
114
100.0
Duration of folic acid intake
Yes
0
0.0
No
0
0.0
Not in the guideline
114
100.0
First meeting with the profession(s) suggested in the guideline
Yes
51
44.7
No
6
5.3
Not in the guideline
57
50.0
Ocular assessment
Yes
30
26.3
No
41
36.0
Not in the guideline
43
37.7
Renal function
Yes
65
57.0
No
49
43.0
Not in the guideline
0
0.0
Hemoglobin A1c
Yes
1
0.9
No
56
49.1
Not in the guideline
57
50.0
Preprandial p-glucose
Yes
9
7.9
No
105
92.1
Not in the guideline
0
0.0
Frequency and number of daily p-glucose tests
Yes
15
13.2
No
56
49.1
Not in the guideline
43
37.7
Postprandial time specified
Yes
5
4.4
No
52
45.6
Not in the guideline
57
50.0
Continuous glucose measurement
Yes
0
0.0
No
0
0.0
Not in the guideline
114
100.0
Fetal ultrasound conducted according to the guideline
Yes
82
71.9
No
32
28.1
Not in the guideline
0
0.0
Cardiotocography conducted according to the guideline
Yes
83
72.8
No
30
26.3
Not in the guideline
0
0.0
Not applicable
1
0.9
When to induce labor
Yes
36
31.6
No
3
2.6
Not in the guideline
0
0.0
Not applicable
75
65.8
P-glucose measured at the guideline timepoints during labor
Yes
30
26.3
No
67
58.8
Not in the guideline
0
0.0
Not applicable
17
14.9
Interventions in p-glucose during labor according to the guideline
Yes
40
35.1
No
74
64.9
Not in the guideline
0
0.0
Insulin ordered during labor according to the guideline
Yes
38
33.3
No
60
52.6
Not in the guideline
0
0.0
Not applicable
16
14.0
Insulin administered during labor according to the guideline
Yes
28
24.6
No
69
60.5
Not in the guideline
0
0.0
Not applicable
17
14.9
Intravenous infusion of glucose ordered according to the guideline
HbA1c values were compared on three occasions during pregnancy (first, second, and third trimester) and categorized into three groups: <42 mmol/mol, 42–48 mmol/mol, and >48 mmol/mol.
2.1 Statistical analyses
Data were collected from the obstetric medical record system Obstetrix (Cerner, North Kansas, US) and the medical record system Take Care (CompuGroup Medical Sweden AB, Solna, Sweden). The statistical analyses were performed using SPSS software version 25.0 (IBM Corp., Armonk, NY, USA).
The statistical method for comparing the medians of the continuous variables was the Mann-Whitney U-test. When comparing the binary variables, the calculations were performed either by Fisher’s exact test in groups <5 or by the Chi-square test for the remaining variables. To consider statistical significance, the p-value was set to <0.05. All data are presented as medians ± range (min–max) or as numbers and percentages.
2.2 Ethical approval
The study complied with the World Medical Association Declaration of Helsinki regarding the ethical conduct of research involving human subjects. The research was approved by the independent regional Research Ethics Committee (Karolinska Institute, no. 2017/1310-31). Oral and written informed consent was obtained from all the women before inclusion in the study.
3. Results
Of the 42 Swedish delivery wards caring for women with DM1, 30 different local guidelines for pregnancy and labor were used, including 29 different local guidelines for pregnancy and 29 different local guidelines for labor. Twenty-one hospitals used 21 different guidelines, 10 hospitals had 5 different guidelines, and 9 hospitals had 3 different guidelines. One clinic had a guideline only for DM1 during pregnancy but not for labor, and one clinic had just a guideline for labor (Fig. 1). No consensus was found among the 29 guidelines for any of the 31 variables studied (Table 1).
Fig. 1Guidelines of clinics in Sweden that care for DM1 pregnancies.
Pre-conception care was recommended in 14% of the guidelines. While folic acid supplementation was recommended by some, the doses and durations differed. Less than half (45%) of the guidelines recommended multidisciplinary care. Assessment of renal function was recommended by 69% of the guidelines, but the frequency and gestational week varied. Overall, 48% of the guidelines had no recommendations for HbA1c during pregnancy. The remainder recommended 9 different values (38 to <57 mmol/mol). Only 59% of the guidelines had recommendations regarding post-prandial plasma glucose (p-glucose) during pregnancy, providing 8 various recommendations.
Near consensus was found for fetal growth ultrasound. Most (79%) of the guidelines recommended examination at 28 weeks; the remainder recommended it between 24 and 32 weeks. Overall, 72% of the guidelines recommended cardiotocography (CTG) during pregnancy but at six different gestational weeks (33–38 weeks of gestation). Five guidelines had no recommendations at all. Three did not recommend CTG during pregnancy if no abnormalities were observed. The remainder (n=21) recommended it for everyone.
The recommended time for induction of labor varied between 38+0 and 41+0 weeks. The recommendations for monitoring intrapartum p-glucose varied included three different intervals. One guideline lacked recommendations for target p-glucose levels during labor, and the rest had 10 various suggestions. A p-glucose level of 4–7 mmol/l was the most common recommendation (34.5%). The insulin recommendations during labor varied and some were contradictory. The least consistent recommendation in these guidelines was on glucose infusion during labor. Eighteen different recommendations were found among the 29 guidelines (Table 1).
In the second part of the study, women with DM1 before pregnancy who delivered in Stockholm during 2016 were studied. In total, 126 women with DM1 gave birth in 1 of the 4 hospitals in Stockholm during that year. Twelve women were excluded as they did not meet the study’s inclusion criteria, so 114 women constituted our final study population (Fig. 2).
When analyzing the 114 medical records, we found that the medical staff followed their local guideline for DM1 and pregnancy with ≥70% adherence in 17.5% of the medical records. The medical staff followed their own local guidelines for labor with ≥70% adherence in 18.4% of the medical records. In six (5.3%) medical records, the local guidelines for both labor and pregnancy were followed at a rate of ≥70% by the medical staff.
No significant differences in the maternal or fetal background data were found when comparing the different groups (Table 3). The onset of labor varied. Induction of labor was more common when labor guidelines were followed with ≥70% adherence (66.7 vs. 41.9%, p = 0.02). Further, planned cesarean sectio were more common before the start of labor when labor guidelines were followed < 70% (32.3 vs. 4.8 %, p = 0.007).
Table 3Background and adherence to pregnancy and labor guidelines.
Guideline
Pregnancy
Pregnancy
Labor
Labor
All labors in Sweden in 2016 (n = 120,110) Newborns in Sweden in 2016 (n = 121,843)
Abbreviations: BE, base excess; BMI, body mass index; LGA, large for gestational age; NICU, neonatal intensive care unit; SGA, small for gestational age.
Values are numbers (%) or median (range).
a Considered statistically significant if p < 0.05.
Differences in mode of delivery were also identified. Non-instrumental vaginal labor occurred in 80.0% vs. 23.4% and 66.7% vs. 25.8% (p = <0.001) when the guidelines for pregnancy or the guidelines for labor were followed by >70% in the medical records (Table 4).
Table 4Labor outcomes and adherence to pregnancy and labor guidelines.
The most significant cause of operative or instrumental intervention when the guidelines for pregnancy had been followed >70% was a clinical sign of LGA, 75% vs. 18.1% (p = 0.03; Table 4). However, the frequency of an LGA at birth did not differ between the groups (30% vs. 37%, p = 0.54; Table 3).
In the third trimester, HbA1c was significantly lower when adherence to pregnancy guidelines was > 70% compared to those medical records that showed compliance with the pregnancy guidelines of <70% (63.2% vs. 27.5%, p = 0.005; Table 5). At the same time, significantly more medical records that adhered to the pregnancy guidelines <70% had HbA1c levels of 42–48 mmol/mol compared to the medical records that complied with the pregnancy guidelines >70% (47.5% vs. 21.2%, p = 0.047; Table 5).
Table 5HbA1c adherence in pregnancy, labor, and both guidelines.
This study’s main finding was the lack of consensus among the various Swedish guidelines for DM1 during pregnancy and labor. Swedish healthcare system serves around 500 pregnant women with DM1 every year, and no standard guideline for how these pregnancies and labor should be handled exists. All the Swedish guidelines were collected and compared on 31 predetermined factors that are considered necessary according to international guidelines. No consensus was identified for any of the 31 variables. Adherence with local guidelines was also poor. If the level of adherence to guidelines was set at ≥70%, 17.5% of the medical staff in Stockholm followed their guidelines on DM1 during pregnancy, 18.5% followed their local guidelines during labor, and only 5.3% followed the combination of the two guidelines. DM1 represents one of the most challenging medical complications of pregnancy because of the need for frequent monitoring and adjustment of insulin, considering the potential for maternal and fetal complications. These findings suggest an urgent need to overcome the lack of consensus when caring for pregnant women living with DM1.
Cutchie et al. compared the international guidelines of women with diabetes and pregnancy with reported practice in 2005 among New Zealand maternity centers [
]. They found that international guidelines for the care of women with diabetes in pregnancy remained fragmented. They concluded that the development of one set of guidelines based on the consensus of global best practices could overcome many of the misconceptions associated with diabetes in pregnancy. Subsequent work in 2007 by Zeck et al. compared the Austrian and Australian national guidelines for gestational and pre-gestational diabetes [
]. They estimated the level to which physicians complied with their country’s guidelines and concluded that standardizing the screening criteria and diagnostic methods for gestational and pre-gestational diabetes was difficult. They recommended that national and international consensus be reached to manage diabetes in pregnancy. These studies were published more than 15 years ago, and no further progress has been made. Many factors have been found to challenge adherence to diabetes guidelines, including how the guidelines are written and the responsible organization itself [
]. The low compliance seen in our study is comparable to that reported in other studies, where, for example, adherence to a diabetic guideline was 38% and adherence to measuring HbA1c was 20% [
Two specialized care areas, obstetrics and diabetology, are combined when dealing with pregnancy in women with DM1. Therefore, interdisciplinary collaboration is recommended in international guidelines [
] demonstrated a reduction in the malformation rate of fetuses of women with DM1 who received pre-conception care. In our review of the 29 different Swedish guidelines for women with DM1, we found that only 4 of the 29 recommend a pre-conception visit. The resources invested in the first visit to the antenatal care center should be spent early, before pregnancy, as this has been proven to be more cost-efficient and to generate better maternal and fetal outcomes [
A study published by Jensen et al. concluded that monitoring the level of HbA1c during pregnancy is essential, as high HbA1c levels (> 98 mmol/mol) during pregnancy increase the risk of malformations in the fetus by 16% [
]. Previous studies have also concluded that even HbA1c levels < 57 mmol/mol three months before conception increase the risk of fetal primary cardiac defects more than twofold [
]. According to our study, the Swedish guidelines have no consensus regarding the appropriate level of HbA1c during pregnancy. We found a significant difference in the level of HbA1c in the third trimester if the local guideline during pregnancy was followed >70% or not. This is likely to have a significant impact on the well-being of women and the fetus during pregnancy.
Continuous glucose monitoring (CGM) with specific time in range (TIR) recommendations is a more specific tool than HbA1c. CGM measures interstitial glucose in real time, while HbA1c estimates average p-glucose for the last two to three months and physiological falls during pregnancy. In 2017 Feig et al. [
] recommended CGM for pregnant women with DM1. Their study CONCEPTT demonstrated improved neonatal outcomes for the newborns when using CGM during pregnancy with DM1. Therefore, the international consensus is now to use CGM instead of or together with HbA1c. After the study by Feig et al. [
] now also recommend CGM for all pregnant women with DM1. Some local Swedish guidelines have been updated with this recommendation. In 2020, Stockholm Region published one local guideline for all the hospitals in Stockholm caring for DM1 pregnancies [
]; however, this guideline still only has recommendations for the level of HbA1c and no recommendations for TIR on its CGM recommendation.
In the second part of our study, we found that in the medical records where guidelines for pregnancy were followed ≥70%, significantly more women (80%) had a non-instrumental labor compared to just over 23% in records where guidelines for pregnancy were followed <70%; fear of an LGA fetus was the only significant different cause of an operative/instrumental intervention. When the guidelines for labor were followed ≥70%, more women were induced to labor, risk of fetal asphyxia was a significantly higher reason for intervention, and a non-instrumental vaginal delivery occurred in more than 66% of the women, compared to only 25% among those where the guidelines were followed <70%. During labor, maternal p-glucose could increase the risk of labor complications. High maternal p-glucose levels during labor may affect the fetus’s oxygenation and the CTG trace. This, in turn, is associated with a higher rate of cesarean sectio’s observed among pregnancies complicated by DM1 [
]. However, in our results, we found no difference in umbilical cord Ph or Apgar scores after delivery between the groups. This suggests that following the guidelines during labor is protective when there is a risk of fetal asphyxia.
A high maternal p-glucose level during labor also triggers insulin production in the fetus. This generates neonatal hypoglycemia within 24 h after delivery, including transfer to the neonatal intensive care unit (NICU) [
]. The Swedish guidelines have no consensus regarding when and how to monitor maternal p-glucose during labor in pregnant women with DM1. They also have no agreement regarding how insulin should be administered during labor or how glucose infusion should occur. We conclude that some of the Swedish recommendations are contradictory, which may pose a risk to the well-being of the fetus, the mother, and the outcomes of the labor. In our study, there was no difference when the guidelines were followed by >70 or not in terms of neonatal hypoglycemia or transfer to the NICU. About 20% of the newborns were moved from ordinary postnatal care to the NICU.
Women living with DM1 might have high degrees of anxiety, depression, and negative thoughts during pregnancy. This is linked to their impotence in regulating their blood sugar during the pregnancy and the lack of support and knowledge they receive from caregivers [
]. The Swedish guidelines lack consensus on how to manage a pregnancy complicated with DM1. If the caregivers relied on the same, evidence-based, knowledge on how to care for pregnant women with DM1, this might increase the feeling of support and wellbeing for pregnant women living with DM1.
A strength of this study is that a similar review of Swedish guidelines for pregnant women living with DM1 has never been performed. Identifying this significant problem and improving outcomes for this patient group is important. The study’s limitations include that medical care for these pregnancies might be more extensive than the guidelines indicate. We suspect that the guidelines might not be the primary tool used when these women are cared for in daily clinical work.
This study represents the first time that all 29 Swedish guidelines for DM1 during pregnancy and labor have been collected, studied, and compared in terms of the variables that reach a consensus with international guidelines. This work is also, to our knowledge, the first time that the adherence of Swedish guidelines for DM1 during pregnancy and labor has been reviewed.
In 1989, a meeting between the World Health Organization and the International Diabetes Federation resulted in the St. Vincent Declaration. This declaration stated that women living with DM1 have the right to achieve similar pregnancy, maternal, and fetal outcomes to those enjoyed by non-diabetic women [
]. Care must be evidence-based and delivered equally to all women, even when the pregnancies are complicated by DM1.
5. Conclusion
A lack of consensus and adherence persists among the Swedish guidelines regarding the care provided for women with DM1 during pregnancy and labor. A national guideline for DM1 during pregnancy and childbirth, followed by assurance of high compliance, could improve care for pregnant Swedish women with DM1 and their fetuses.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors Alexandra Goldberg, Carina Ursing, Cecilia Ekéus och Eva Wiberg-Itzel, have no conflict of interest to declare.
Acknowledgments
We thank all the women that participated in the study and our colleagues for sending us their guidelines.
00137-6&id=gr1.jpg){kind=link}
00137-6&id=gr2.jpg){kind=link}