Original Article
Pedro Marques, MD1; Maria Raquel Carvalho, MD2; Luísa Pinto, MD3; Sílvia Guerra, MD2 ABSTRACT Objective: The use of metformin in pregnant women is still controversial, despite the increasing reports on metformin’s safety and effectiveness. We aimed to evaluate the maternal and neonatal safety of metformin in subjects with gestational diabetes mellitus (GDM). Methods: We retrospectively reviewed the clinical records of 186 pregnancies complicated with GDM surveilled at Hospital de Santa Maria, Lisboa, between 2011 and 2012. The maternal and neonatal outcomes of 32 females who took metformin during pregnancy were compared with 121 females controlled with diet and 33 insulintreated females. Results: Of the 186 GDM subjects, 32 (17.2%) received metformin during pregnancy. No statistical differences between the diet and metformin groups were found with regard to the rates of abortion, prematurity, preeclampsia, macrosomy, small-for-gestational-age (SGA) or largefor-gestational-age (LGA) newborns, cesarean deliveries, neonatal intensive care unit (NICU) admissions, and birth malformations or neonatal injuries. Similarly, there were no differences between the metformin and insulin groups with regard to the referred outcomes. No abortions or perinatal deaths were recorded in the metformin group. Ten out of 32 metformin patients required additional insulin.
Submitted for publication January 14, 2013 Accepted for publication April 1, 2014 From the 1Department of Endocrinology, Instituto Português de Oncologia de Lisboa; Departments of 2Endocrinology and 3Obstetrics and Gynecology, Hospital de Santa Maria, CHLN, Lisboa, Portugal. Address correspondence to Dr. Pedro Marques, Endocrinology Department, Instituto Português de Oncologia de Lisboa, Rua Professor Lima Basto, 1099023, Lisboa, Portugal. E-mail: [email protected] Published as a Rapid Electronic Article in Press at http://www.endocrine practice.org on May 1, 2014. DOI:10.4158/EP14018.OR To purchase reprints of this article, please visit: www.aace.com/reprints. Copyright © 2014 AACE.
1022
Conclusion: This retrospective study suggests that metformin is a safe alternative or additional treatment to insulin in females with GDM. Metformin was not associated with a higher risk of maternal or neonatal complications when compared to the insulin or diet groups. (Endocr Pract. 2014;20:1022-1031) Abbreviations: BMI = body mass index; DM = diabetes mellitus; FPG = fasting plasma glucose; GDM = gestational diabetes mellitus; HbA1C = glycated hemoglobin; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; LGA = large-for-gestational-age; NICU = neonatal intensive care unit; OGTT = oral glucose tolerance test; SGA = small-for-gestational-age; T2DM = type 2 diabetes mellitus INTRODUCTION Gestational diabetes mellitus (GDM) is defined as any degree of glucose intolerance with onset or initially identified during pregnancy (1). The overall incidence of GDM has been increasing as pregnant women become older and more obese (2,3), and it varies widely depending on the diagnostic criteria and characteristics of the studied population, such as ethnicity (4,5). The prevalence rates of GDM are higher in Asian/Pacific Islander, Non-Hispanic Black and Hispanic populations than in Non-Hispanic White populations (6). The prevalence of GDM ranges from 1.1 to 25.5% of all pregnancies in the U.S. (7), and it is estimated to affect 2.2% and 15% of South American and Indian pregnancies, respectively (8,9). GDM is associated with an increased risk of maternal and perinatal complications including preeclampsia, cesarean delivery, macrosomia, birth injuries and trauma, prematurity, hypoglycemia, and neonatal respiratory distress (10). Several studies have demonstrated that the risk of adverse pregnancy outcomes increases continuously with elevated glucose levels and can be reduced with effective
1023
treatment (10,11). Interventions to modify lifestyle have been shown to improve perinatal outcomes, and if hyperglycemia persists, additional treatments, traditionally insulin, are often administered (12). While effective, insulin therapy has several disadvantages for pregnant women, including multiple daily injections, risk of hypoglycemia, increased risk of maternal weight gain, and higher cost. Moreover, it requires adjustments based on maternal weight, glucose levels, and diet and physical activity, necessitating frequent assessments. Therefore, a safe and effective oral agent would offer advantages over insulin (13,14). Metformin, as the firstline medication for type 2 diabetes mellitus (T2DM) in nonpregnant patients is an oral agent candidate for GDM. It decreases hepatic gluconeogenesis and improves peripheral glucose uptake and is not associated with weight gain or hypoglycemia (15). Metformin crosses the placenta with a maternal-to-fetal transfer rate estimated at 10 to 16%, so it may directly affect fetal physiology (16). The long-term effects of metformin on future metabolic disorders in the offspring are currently unknown and constitute an important barrier for its use (13). However, it is speculated that metformin induces a more favorable intrauterine environment and may reduce the offspring’s long-term metabolic complications (e.g., obesity or diabetes mellitus [DM]) (13,17). Although several studies have reported that metformin is safe and suitable for GDM (5,11,13,14,18-24), its use in pregnant women remains controversial. Furthermore, inconsistencies in clinical outcomes across studies and limited data on the safety, benefits, and risks of metformin in GDM have made it difficult to establish definitive conclusions (14). In this retrospective study, we aimed to evaluate the maternal and neonatal safety of metformin in GDM by comparing several maternal and neonatal outcomes in patients with GDM treated with metformin, insulin, or dietary regimens. METHODS We conducted a retrospective study of pregnancies complicated with GDM followed in the endocrine department of the Hospital de Santa Maria, Lisboa, between 2011 and 2012. Patients were identified from the database of the obstetrics and gynecology department. Women were eligible if they had a GDM diagnosis established according to the International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria (25) adopted in Portugal in January 2011 (26), if the pregnancy and delivery were entirely surveilled at our hospital, and if the clinical records were complete. GDM was diagnosed when the fasting plasma glucose (FPG) value prior
to 13 weeks of gestation was between 92 and 125 mg/dL (5.1-6.9 mmol/L) or when the 2-hour 75-g oral glucose tolerance test (OGTT) at 24 to 28 weeks of gestation yielded 1-hour glycemia ≥180 mg/dL (10.0 mmol/L) and/or 2-hour glycemia of 153 to 199 mg/dL (8.5-11.0 mmol/L) (25). Overt DM was identified before 13 weeks of gestation when FPG was ≥126 mg/dL (7.0 mmol/L), random plasma glucose was ≥200 mg/dL (11.1 mmol/L), and/or glycated hemoglobin (HbA1C) was ≥6.5%. It was also diagnosed at 24 to 28 weeks of gestation through 2-hour OGTT for FPG ≥126 mg/dL (7.0 mmol/L) and/or 2-hour glycemia ≥200 mg/dL (11.1 mmol/L) (25). We excluded these cases, which were considered as prepregnancy DM, as well as cases of multiple pregnancy, subjects with severe underlying diseases (cardiovascular, renal or liver), and those on corticosteroids. We identified 186 GDM women. Demographic, clinical, and management parameters and maternal and neonatal outcomes were collected and analyzed. All women with GDM received prenatal care and counseling on diet and physical exercise from a multidisciplinary team (obstetrician, endocrinologist, and nutritionist). Carbohydrate intake was restricted, and the diet was divided into 3 meals and 3 snacks. An exercise program of 30 minutes of walking per day was recommended. The women performed glucose self assessments 4 times per day (overnight fast, 1 hour after breakfast, 1 hour after lunch, and 1 hour after dinner). The targets were FPG levels ≤90 mg/dL and 1-hour postprandial glucose levels ≤120 mg/dL. The frequency of glycemic log revision was determined by the metabolic control and the type of treatment of each subject, varying between 1 and 4 weeks. Those with GDM inadequately controlled by diet were assigned to pharmacological treatment, receiving either insulin or metformin. There were no strict criteria defining which treatment should be started, but subjects FPG levels ≥140 mg/dL and/or 2-hour postprandial glucose levels ≥180 mg/dL were assigned to receive insulin. Pharmacological treatment was started when 3 readings were above the target values, except when women identified a factor explaining the abnormal readings. If glycemic targets were achieved, diet intervention was considered to be adequate. HbA1C levels were monitored at the first visit, every 1 to 2 months during pregnancy, and near partum. Metformin was initiated in 32 women (all after 12 weeks of gestation). An initial dose of 500 mg twice daily was normally used, with an increase of 500 to 1,000 mg to achieve glycemic targets, up to a maximum daily dose of 2,500 mg. Metformin was continued until delivery and no patient stopped treatment. Insulin was added if glycemic targets were not achieved despite maximum dose. Ten out of 32 patients who received metformin required
1024
supplementary neutral protamine Hagedorn (NPH) insulin. There was no need to initiate prandial insulins in the metformin group. Insulin therapy (without metformin) was used in 33 patients. An initial daily dose of 0.1 to 0.2 U/kg was normally used, and the patients received titration instructions. The basal insulin was usually NPH insulin, although 4 subjects were treated with long-acting insulin analogs, and 4 required prandial insulin. Data was collected according to baseline features of the patients and maternal/neonatal outcomes. Maternal baseline variables were age, parity, ethnicity, education, family history of DM, body mass index (BMI), and initial HbA1C. Maternal variables were weight gain during pregnancy, difference of end-pregnancy and initial HbA1C, preeclampsia (blood pressure >140/90 mm Hg with proteinuria >0.3 g/24 hours), abortion, cesarean delivery, and prematurity. Neonatal variables were birth weight, macrosomia (≥4,000 g), birth weight percentile for gestational age (small-for-gestational-age [SGA] if percentile 90th), 5-minute Apgar score, malformations, perinatal injuries or death, and neonatal intensive care unit (NICU) admission. In order to assess the maternal and neonatal safety of metformin, our primary study purpose, the group treated with metformin was individually compared to the diet and insulin groups. The diet group was considered a good comparator to reinforce the safety profile of metformin as it includes GDM cases that are theoretically easier to manage. An OGTT was performed in 154 women at 6- to 8-weeks postpartum independently of the metabolic control and pregnancy outcomes. The OGTT diagnostic cutoff values for impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or DM were adopted from the American Diabetes Association (ADA) (1). All the data and statistical tests were carried out using SPSS (version 20.0, IBM Inc, Armonk, NY). Descriptive statistics were used to summarize the data. Categorical variables were compared with c2 and Fisher’s exact tests. Student’s t-tests and analysis of variance (ANOVA) were used to compare continuous variables. Univariate logistic regression analysis was carried out to evaluate treatment modality safety/risk. A P value
Pedro Marques, MD1; Maria Raquel Carvalho, MD2; Luísa Pinto, MD3; Sílvia Guerra, MD2 ABSTRACT Objective: The use of metformin in pregnant women is still controversial, despite the increasing reports on metformin’s safety and effectiveness. We aimed to evaluate the maternal and neonatal safety of metformin in subjects with gestational diabetes mellitus (GDM). Methods: We retrospectively reviewed the clinical records of 186 pregnancies complicated with GDM surveilled at Hospital de Santa Maria, Lisboa, between 2011 and 2012. The maternal and neonatal outcomes of 32 females who took metformin during pregnancy were compared with 121 females controlled with diet and 33 insulintreated females. Results: Of the 186 GDM subjects, 32 (17.2%) received metformin during pregnancy. No statistical differences between the diet and metformin groups were found with regard to the rates of abortion, prematurity, preeclampsia, macrosomy, small-for-gestational-age (SGA) or largefor-gestational-age (LGA) newborns, cesarean deliveries, neonatal intensive care unit (NICU) admissions, and birth malformations or neonatal injuries. Similarly, there were no differences between the metformin and insulin groups with regard to the referred outcomes. No abortions or perinatal deaths were recorded in the metformin group. Ten out of 32 metformin patients required additional insulin.
Submitted for publication January 14, 2013 Accepted for publication April 1, 2014 From the 1Department of Endocrinology, Instituto Português de Oncologia de Lisboa; Departments of 2Endocrinology and 3Obstetrics and Gynecology, Hospital de Santa Maria, CHLN, Lisboa, Portugal. Address correspondence to Dr. Pedro Marques, Endocrinology Department, Instituto Português de Oncologia de Lisboa, Rua Professor Lima Basto, 1099023, Lisboa, Portugal. E-mail: [email protected] Published as a Rapid Electronic Article in Press at http://www.endocrine practice.org on May 1, 2014. DOI:10.4158/EP14018.OR To purchase reprints of this article, please visit: www.aace.com/reprints. Copyright © 2014 AACE.
1022
Conclusion: This retrospective study suggests that metformin is a safe alternative or additional treatment to insulin in females with GDM. Metformin was not associated with a higher risk of maternal or neonatal complications when compared to the insulin or diet groups. (Endocr Pract. 2014;20:1022-1031) Abbreviations: BMI = body mass index; DM = diabetes mellitus; FPG = fasting plasma glucose; GDM = gestational diabetes mellitus; HbA1C = glycated hemoglobin; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; LGA = large-for-gestational-age; NICU = neonatal intensive care unit; OGTT = oral glucose tolerance test; SGA = small-for-gestational-age; T2DM = type 2 diabetes mellitus INTRODUCTION Gestational diabetes mellitus (GDM) is defined as any degree of glucose intolerance with onset or initially identified during pregnancy (1). The overall incidence of GDM has been increasing as pregnant women become older and more obese (2,3), and it varies widely depending on the diagnostic criteria and characteristics of the studied population, such as ethnicity (4,5). The prevalence rates of GDM are higher in Asian/Pacific Islander, Non-Hispanic Black and Hispanic populations than in Non-Hispanic White populations (6). The prevalence of GDM ranges from 1.1 to 25.5% of all pregnancies in the U.S. (7), and it is estimated to affect 2.2% and 15% of South American and Indian pregnancies, respectively (8,9). GDM is associated with an increased risk of maternal and perinatal complications including preeclampsia, cesarean delivery, macrosomia, birth injuries and trauma, prematurity, hypoglycemia, and neonatal respiratory distress (10). Several studies have demonstrated that the risk of adverse pregnancy outcomes increases continuously with elevated glucose levels and can be reduced with effective
1023
treatment (10,11). Interventions to modify lifestyle have been shown to improve perinatal outcomes, and if hyperglycemia persists, additional treatments, traditionally insulin, are often administered (12). While effective, insulin therapy has several disadvantages for pregnant women, including multiple daily injections, risk of hypoglycemia, increased risk of maternal weight gain, and higher cost. Moreover, it requires adjustments based on maternal weight, glucose levels, and diet and physical activity, necessitating frequent assessments. Therefore, a safe and effective oral agent would offer advantages over insulin (13,14). Metformin, as the firstline medication for type 2 diabetes mellitus (T2DM) in nonpregnant patients is an oral agent candidate for GDM. It decreases hepatic gluconeogenesis and improves peripheral glucose uptake and is not associated with weight gain or hypoglycemia (15). Metformin crosses the placenta with a maternal-to-fetal transfer rate estimated at 10 to 16%, so it may directly affect fetal physiology (16). The long-term effects of metformin on future metabolic disorders in the offspring are currently unknown and constitute an important barrier for its use (13). However, it is speculated that metformin induces a more favorable intrauterine environment and may reduce the offspring’s long-term metabolic complications (e.g., obesity or diabetes mellitus [DM]) (13,17). Although several studies have reported that metformin is safe and suitable for GDM (5,11,13,14,18-24), its use in pregnant women remains controversial. Furthermore, inconsistencies in clinical outcomes across studies and limited data on the safety, benefits, and risks of metformin in GDM have made it difficult to establish definitive conclusions (14). In this retrospective study, we aimed to evaluate the maternal and neonatal safety of metformin in GDM by comparing several maternal and neonatal outcomes in patients with GDM treated with metformin, insulin, or dietary regimens. METHODS We conducted a retrospective study of pregnancies complicated with GDM followed in the endocrine department of the Hospital de Santa Maria, Lisboa, between 2011 and 2012. Patients were identified from the database of the obstetrics and gynecology department. Women were eligible if they had a GDM diagnosis established according to the International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria (25) adopted in Portugal in January 2011 (26), if the pregnancy and delivery were entirely surveilled at our hospital, and if the clinical records were complete. GDM was diagnosed when the fasting plasma glucose (FPG) value prior
to 13 weeks of gestation was between 92 and 125 mg/dL (5.1-6.9 mmol/L) or when the 2-hour 75-g oral glucose tolerance test (OGTT) at 24 to 28 weeks of gestation yielded 1-hour glycemia ≥180 mg/dL (10.0 mmol/L) and/or 2-hour glycemia of 153 to 199 mg/dL (8.5-11.0 mmol/L) (25). Overt DM was identified before 13 weeks of gestation when FPG was ≥126 mg/dL (7.0 mmol/L), random plasma glucose was ≥200 mg/dL (11.1 mmol/L), and/or glycated hemoglobin (HbA1C) was ≥6.5%. It was also diagnosed at 24 to 28 weeks of gestation through 2-hour OGTT for FPG ≥126 mg/dL (7.0 mmol/L) and/or 2-hour glycemia ≥200 mg/dL (11.1 mmol/L) (25). We excluded these cases, which were considered as prepregnancy DM, as well as cases of multiple pregnancy, subjects with severe underlying diseases (cardiovascular, renal or liver), and those on corticosteroids. We identified 186 GDM women. Demographic, clinical, and management parameters and maternal and neonatal outcomes were collected and analyzed. All women with GDM received prenatal care and counseling on diet and physical exercise from a multidisciplinary team (obstetrician, endocrinologist, and nutritionist). Carbohydrate intake was restricted, and the diet was divided into 3 meals and 3 snacks. An exercise program of 30 minutes of walking per day was recommended. The women performed glucose self assessments 4 times per day (overnight fast, 1 hour after breakfast, 1 hour after lunch, and 1 hour after dinner). The targets were FPG levels ≤90 mg/dL and 1-hour postprandial glucose levels ≤120 mg/dL. The frequency of glycemic log revision was determined by the metabolic control and the type of treatment of each subject, varying between 1 and 4 weeks. Those with GDM inadequately controlled by diet were assigned to pharmacological treatment, receiving either insulin or metformin. There were no strict criteria defining which treatment should be started, but subjects FPG levels ≥140 mg/dL and/or 2-hour postprandial glucose levels ≥180 mg/dL were assigned to receive insulin. Pharmacological treatment was started when 3 readings were above the target values, except when women identified a factor explaining the abnormal readings. If glycemic targets were achieved, diet intervention was considered to be adequate. HbA1C levels were monitored at the first visit, every 1 to 2 months during pregnancy, and near partum. Metformin was initiated in 32 women (all after 12 weeks of gestation). An initial dose of 500 mg twice daily was normally used, with an increase of 500 to 1,000 mg to achieve glycemic targets, up to a maximum daily dose of 2,500 mg. Metformin was continued until delivery and no patient stopped treatment. Insulin was added if glycemic targets were not achieved despite maximum dose. Ten out of 32 patients who received metformin required
1024
supplementary neutral protamine Hagedorn (NPH) insulin. There was no need to initiate prandial insulins in the metformin group. Insulin therapy (without metformin) was used in 33 patients. An initial daily dose of 0.1 to 0.2 U/kg was normally used, and the patients received titration instructions. The basal insulin was usually NPH insulin, although 4 subjects were treated with long-acting insulin analogs, and 4 required prandial insulin. Data was collected according to baseline features of the patients and maternal/neonatal outcomes. Maternal baseline variables were age, parity, ethnicity, education, family history of DM, body mass index (BMI), and initial HbA1C. Maternal variables were weight gain during pregnancy, difference of end-pregnancy and initial HbA1C, preeclampsia (blood pressure >140/90 mm Hg with proteinuria >0.3 g/24 hours), abortion, cesarean delivery, and prematurity. Neonatal variables were birth weight, macrosomia (≥4,000 g), birth weight percentile for gestational age (small-for-gestational-age [SGA] if percentile 90th), 5-minute Apgar score, malformations, perinatal injuries or death, and neonatal intensive care unit (NICU) admission. In order to assess the maternal and neonatal safety of metformin, our primary study purpose, the group treated with metformin was individually compared to the diet and insulin groups. The diet group was considered a good comparator to reinforce the safety profile of metformin as it includes GDM cases that are theoretically easier to manage. An OGTT was performed in 154 women at 6- to 8-weeks postpartum independently of the metabolic control and pregnancy outcomes. The OGTT diagnostic cutoff values for impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or DM were adopted from the American Diabetes Association (ADA) (1). All the data and statistical tests were carried out using SPSS (version 20.0, IBM Inc, Armonk, NY). Descriptive statistics were used to summarize the data. Categorical variables were compared with c2 and Fisher’s exact tests. Student’s t-tests and analysis of variance (ANOVA) were used to compare continuous variables. Univariate logistic regression analysis was carried out to evaluate treatment modality safety/risk. A P value
