Period of gestational diabetes mellitus diagnosis and maternal and fetal morbidity

Transcription

1 Acta Obstet Gynecol Scand 2005: 84: Copyright # Acta Obstet Gynecol Scand 2005 Printed in UK. All rights reserved Acta Obstetricia et Gynecologica Scandinavica ORIGINAL ARTICLE Period of gestational diabetes mellitus diagnosis and maternal and fetal morbidity MARÍA JOSÉ BARAHONA 1,NURIA SUCUNZA 1,APOLONIA GARCÍA-PATTERSON 1,MARTA HERNÁNDEZ 1,JUAN M. ADELANTADO 2, GEMMA GINOVART 3,ALBERTO DE LEIVA 1 AND ROSA CORCOY 1 From the Departments of 1 Endocrinology, 2 Gynecology and Obstetrics, and 3 Pediatrics, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Acta Obstet Gynecol Scand 2005; 84: # Acta Obstet Gynecol Scand Background. The aim of the study was to analyze the association between the period of diagnosis of gestational diabetes mellitus (GDM) and maternal and neonatal outcome. Methods. In this retrospective study, 1708 offspring (1571 singleton, 119 twins, and 18 triplets) born to women with GDM who attended the Diabetic and Pregnancy Clinic were included. Pregnancies were divided into three groups according to the gestational age at GDM diagnosis. The association of the period of diagnosis with maternal and fetal outcome was assessed adjusting for potentially confounding variables (logistic regression analysis). Results. The period of diagnosis was a predictor in two out of three maternal outcomes (pregnancy-induced hypertension and insulin treatment) and in four out of 12 fetal outcomes (preterm birth, 5-min Apgar <7, perinatal mortality, and hyperbilirubinemia). Whereas pregnancy-induced hypertension was higher in women diagnosed with GDM in the second period, the other outcomes displayed higher occurrences with earlier diagnosis. Conclusions. Diagnosis of GDM earlier in pregnancy is a predictor of adverse maternal and neonatal outcome. Key words: gestational diabetes mellitus; maternal outcome; perinatal outcome; period of diagnosis Submitted 27 November, 2003 Accepted 3 May, 2004 Introduction Gestational diabetes mellitus (GDM) is associated with adverse maternal and fetal outcome (1), but its diagnosis and treatment remain controversial (2). According to the Fourth Workshop Conference on Gestational Diabetes Mellitus, early screening is only recommended for women with high-risk characteristics, whereas those at low risk could even skip screening (3). However, Bartha et al. have reported a worse Abbreviations: GDM: gestational diabetes mellitus; HbA1c: glycosylated hemoglobin; OGTT: oral glucose tolerance test; BMI: body mass index; OR: odds ratio; CI: confidence interval; NS: not significant; BG: blood glucose; LGA: large for gestational age. outcome for women with GDM diagnosed in the first 23 weeks of pregnancy, although results were not adjusted for potentially confounding variables (4). We aimed to analyze the usefulness of early diagnosis of GDM to predict adverse maternal and neonatal outcomes adjusting for potentially confounding variables. Materials and methods Patients This is a retrospective study of 2049 offspring consecutively born between January 1986 and December 2001 to women with GDM who attended the Diabetic and Pregnancy Clinic and delivered at a gestational age of 23 weeks. Seven singleton pregnancies were excluded, because the date of

2 Perinatal outcome in GDM 623 GDM diagnosis was missing. Analysis was performed on 1708 offspring (1571 singleton, 119 twins, and 18 triplets) without any missing data. For the analysis, the information included in an existing database was used. Chart review and data entry are performed regularly by the endocrinologic team shortly after delivery as part of a continuous quality improvement strategy. GDM screening and diagnosis The diagnosis of GDM was established with the help of an oral glucose tolerance test (OGTT) in accordance with the Second and Third Workshop Conference Criteria on GDM (5,6). The GDM screening protocol in our center includes a 50-g glucose challenge test at the first visit, between 24 and 28 weeks, and between 31 and 34 weeks of pregnancy, whenever a GDM diagnosis has not been performed. This GDM screening protocol has been running in the center since 1986, when a pilot study indicated a significant contribution of each screening period to the prevalence of GDM (7). The population was divided into three groups according to the gestational age at diagnosis: the first period: <24 weeks (302 newborns, 17.7%), the second period: weeks (653 newborns, 38.2%), and the third period: >31 weeks (753 newborns, 44.1%). GDM management Therapy was initiated with a normocaloric diet and selfmonitoring of blood glucose. Glycosylated hemoglobin (HbA1c) was measured monthly. Insulin was administered when blood glucose values rose above critical target values adapted to the reflectance meter: 5.8 mmol/l fasting and 7.8 mmol/l 1-h postprandial before 1990 when capillary blood glucose was measured with Reflolux I 1 (Boehringer- Mannheim, Germany) and 5.0 mmol/l fasting and 6.7 mmol/l 1-h postprandial when Reflolux II 1 (Boehringer-Mannheim) was introduced (8). As relevant outcome variables, all maternal and perinatal outcome variables potentially related to GDM diagnosis were considered: pregnancy-induced hypertension, insulin treatment, cesarean section, 5-min Apgar <7, macrosomia, large for gestational age (LGA), perinatal mortality, obstetric trauma, minor and major malformations, neonatal hypoglycemia, hypocalcemia, jaundice, respiratory distress, and preterm birth. Outcome definitions Pregnancy-induced hypertension was defined as a blood pressure reading of 140/90 mmhg presenting at a gestational age of >20 weeks or the first week after delivery, confirmed as of 6 h later in a woman without a former diagnosis of hypertension, or previous hypertension worsening during pregnancy and requiring additional treatment. Insulin therapy was defined as a treatment with insulin according to the aforementioned criteria for any length of time. Perinatal mortality was defined as any fetal or neonatal death occurring from gestation of 22 weeks through the first 4 weeks after birth (9). Neonatal hypoglycemia was defined in accordance with Cornblath and Reisner s criteria (10) for capillary blood glucose, which take into consideration both birth weight and prematurity. Blood glucose level was measured with a reflectance meter to allow immediate initiation of treatment when necessary. Hypocalcemia was defined as a serum calcium level of <1.75 mmol/l (11). Neonatal jaundice was defined as hyperbilirubinemia requiring treatment according to Maisels (12). All types of respiratory distress, including transient tachypnea, were considered. Preterm birth was defined as a birth before 37 complete gestational weeks. Macrosomia was defined as a birthweight of >4000 g. LGA newborns are defined as those with a birthweight of >P90 for the corresponding sex and gestational age (13). Obstetric management Obstetric management followed current obstetric principles with an additional evaluation of fetal growth in the third trimester and regular assessment of fetal well-being (cardiotocography once/twice in a week) from the 36th pregnancy week onward. In the 16-year span of the present study, several modifications have been introduced in the obstetric protocol (high-resolution sonography at weeks for malformation detection, biochemical screening for chromosomal abnormalities, systematic screening for toxoplasma infection, and streptococcus carriage), but none specifically affecting the management of women with GDM. Statistical methods HbA1c was expressed in standard deviations around the mean. The distribution of quantitative variables was tested with a Shapiro-Wilks test. The distribution was found to be non-normal, so that variables are expressed as median (range). Differences in maternal and fetal outcomes between the periods were tested using a 2 -test (bivariate analysis). Kruskal Wallis analysis of variance was then used to compare age, body mass index, and variables indicating GDM severity (blood glucose values in the diagnostic OGTT, HbA1c after diagnosis, and mean HbA1c during pregnancy). Next, forward logistic regression analyses were performed using each of the outcomes as the dependent variable and the period of diagnosis, maternal age, body mass index, and variables indicating GDM severity as potentially predictive variables. In the case of macrosomia, weight increment during pregnancy, maternal smoking, and macrosomia in previous pregnancies were also included as potential predictors. A first series of analyses was performed for singleton, a second for newborns of multiple pregnancies, and a third for the whole group, including multiple pregnancy as a potential predictor in the latter. A P-value of <0.05 was considered significant. Results Maternal characteristics were as follows: age 32 years (17 45), body mass index 23.2 kg/m 2 ( ), gestational age at diagnosis 34 weeks (8 42), blood glucose values in the diagnostic OGTT (mmol/l) 4.7 ( ) fasting, 11.7 ( ) at 1 hr, 10.2 ( ) at 2 hr, 7.7 ( ) at 3 h, HbA1c after diagnosis 0.45 standard deviations around the mean ( ), and mean HbA1c during pregnancy 0.40 standard deviations around the mean ( ). In the bivariate analysis evaluating the association between the period of diagnosis and outcomes (Table I), the period was found to be associated with two out of three maternal outcomes: pregnancy-induced hypertension, which was higher in women diagnosed in the second period, insulin treatment, and cesarean section,

3 624 R. Corcoy et al. Table I. Maternal and fetal outcomes according to the period of diagnosis of gestational diabetes mellitus (GDM) Global First period Second period Third period p * Maternal Pregnancy-induced hypertension Insulin treatment <0.001 Cesarean section <0.001 Fetal 5-min Apgar < <0.05 Macrosomia Not significant Large for gestational age Not significant Perinatal mortality <0.001 Obstetric trauma Not significant Minor malformations Not significant Major malformations Not significant Hypoglycemia Not significant Hypocalcemia Not significant Hyperbilirubinemia <0.001 Respiratory distress <0.001 Preterm birth <0.001 Figures have been given in percentage. * 2 -test between periods. whose rate increased with earlier GDM diagnosis. As to fetal outcomes, the period was found to be associated with five out of 12 (5-min Apgar <7, perinatal mortalitity, hyperbilirubinemia, respiratory distress, and preterm birth), and their occurrence was also shifted toward the first period. No differences were found in macrosomia, LGA, obstetric trauma, minor and major malformations, hypoglycemia, or hypocalcemia. Regarding potentially confounding variables, maternal age was higher in women diagnosed in the second period; body mass index, fasting, and 1-h blood glucose at diagnosis were shifted toward higher values in women with an early diagnosis of GDM and the reverse was true for 3-h blood glucose at diagnosis and HbA1c after diagnosis. Two-hour blood glucose at diagnosis and mean HbA1c during pregnancy were not significantly different across the periods (see Table II). We performed separate analyses in women with single and multiple pregnancies with similar patterns being observed (data not shown). Additional analyses were also performed excluding subsequent pregnancies in women with more than one pregnancy in the study period. Moreover, the observed predictors were similar (data not shown). The results reported in this study (Table III) are those of the global analyses (single and multiple pregnancies, all the pregnancies of every woman in the study period, the only limit being the absence of any missing data). In addition to other variables, the period was a predictor in two out of three maternal outcomes (pregnancy-induced hypertension and insulin Table II. Maternal characteristics according to the period of diagnosis of gestational diabetes mellitus (GDM) First period Second period Third period p * Maternal age (years) 32 (21 43) 33 (20 44) 32 (17 45) <0.01 Body mass index (kg/m 2 ) 23.8 ( ) 23.2 ( ) 22.9 ( ) <0.01 Diagnostic OGTT (mmol/l) Fasting blood glucose 4.9 ( ) 4.7 ( ) 4.6 ( ) <0.001 One-hour blood glucose 11.9 ( ) 11.7 ( ) 11.5 ( ) <0.001 Two-hour blood glucose 10.3 ( ) 10.2 ( ) 10.2 ( ) Not significant Three-hour blood glucose 7.5 ( ) 7.8 ( ) 7.7 ( ) <0.05 HbA1sc After diagnosis 0.62 ( ) 0.56 ( ) 0.40 ( ) <0.05 Mean 0.40 ( ) 0.45 ( ) 0.40 ( ) Not significant *Kruskal Wallis ANOVA between periods. HbA1c, glycosylated hemoglobin, expressed in standard deviations around the mean. OGTT, oral glucose tolerance test.

4 Perinatal outcome in GDM 625 Table III. Predictors of maternal and fetal outcomes in women with gestational diabetes mellitus (GDM) Outcomes Predictive variable Odds ratio 95% CI p-value of the model* ROC curve area Maternal Pregnancy-induced hypertension < Prepregnancy BMI Second Third Multiple pregnancy Insulin treatment < Prepregnancy BMI Fasting BG at diagnosis Three-hour BG at diagnosis Second Third Cesarean section < Maternal age Prepregnancy BMI Mean HbA1c One-hour BG at diagnosis Multiple pregnancy Neonatal Preterm birth < Age Second Third Multiple pregnancy Five-min Apgar < Second Third Multiple pregnancy Macrosomia < Prepregnancy BMI Gestational weight gain Smoking No 1 Stopped during pregnancy Active during pregnancy Previous macrosomia Multiple pregnancy LGA < Prepregnancy BMI Gestational weight gain Smoking No 1 Stopped during pregnancy Active during pregnancy Previous macrosomia Multiple pregnancy Perinatal mortality Second Third Multiple pregnancy Obstetric trauma Not significant Minor malformations Not significant Major malformations Fasting BG at diagnosis Hypoglycemia Prepregnancy BMI HbA1c after diagnosis

5 626 R. Corcoy et al. Table III. Conted Predictors of maternal and fetal outcomes in women with gestational diabetes mellitus (GDM) Outcomes Predictive variable Odds ratio 95% CI p-value of the model* ROC curve area Hypocalcemia Not significant Hyperbilirubinemia < Second Third Multiple pregnancy Respiratory distress < Multiple pregnancy BMI, body mass index; HbA1c, glycosylated hemoglobin; BG, blood glucose; ROC, receiver operator curve. *Multiple logistic regression analysis. treatment). The rate of pregnancy-induced hypertension was highest in women diagnosed in the second period, whereas the rate of insulin treatment decreased from the first to the third period of diagnosis. Besides, the period of diagnosis was a predictor in four out of 12 fetal outcomes (preterm birth, 5-min Apgar <7, perinatal mortality, and hyperbilirubinemia). Preterm birth and 5-min Apgar <7 decreased gradually from the first to the third period, although for 5-min Apgar <7 differences between contiguous periods were non-significant. Perinatal mortality was highest in women diagnosed in the first period; it decreased significantly between the first and second periods, and non-significantly between the first and the third. Finally, neonatal jaundice decreased in offspring of women diagnosed in the third, compared with the first period with a nonsignificant increment in offspring of mothers diagnosed in the second period. The period of diagnosis was not included among the predictors of macrosomia, LGA, major malformations, hypoglycemia, or respiratory distress. Finally, no predictor was found for obstetric trauma, minor malformations, or hypocalcemia. Discussion Studies addressing the relationship between gestational age at GDM diagnosis and perinatal outcome are scarce. In addition to the paper of Bartha et al. (4), only Berkowitz et al. (14) and Svare et al. (15) have performed similar analyses. There is an agreement in maternal outcomes; early diagnosis is associated with higher rates of insulin treatment (4,14,15) and higher (4) or borderline increased hypertension rates (14). Early GDM diagnosis has also been consistently associated with diabetes at follow-up (15 among others). Other authors have addressed maternal characteristics in women with a GDM diagnosis early in pregnancy, but perinatal outcome was not explored (16,17). Results reported in this study confirm a positive association between earlier diagnosis and requirement of insulin therapy, but unexpectedly, we found that the rate of pregnancy-induced hypertension was increased in women diagnosed with GDM in the second period. In this paper, we have not addressed postpartum glucose tolerance, but we have described elsewhere that early diagnosis is a predictor of diabetes at midterm follow-up (18). In relation to fetal outcomes, Berkowitz et al. and Svare et al. (14,15) do not report differences in women diagnosed with GDM early in pregnancy, probably due to a lack of statistical power. However, Bartha et al. (4) report an increased rate of perinatal mortality and hypoglycemia in these infants. In the present paper, and after adjusting for potential confounders, we confirm the association between perinatal mortality and diagnosis of GDM early in pregnancy, and extend the findings to preterm birth, 5-min Apgar <7, and hyperbilirubinemia, but we do not corroborate an association with neonatal hypoglycemia. In this paper, fasting blood glucose is a predictor of major congenital malformations, but we have recently reported that in women with GDM, prepregnancy body mass index has a greater role than diabetic severity in this prediction (19). In the aforementioned paper, specific models for congenital malformations were constructed using a backward method among other differences and, in fact, body mass index would be included in the predictive model of major congenital malformations in the present study, if a backward method was used (data not shown). However, in the present study, we preferred to use the same strategy in the prediction of all outcomes. We conclude that despite intensive metabolic management, increased obstetric surveillance, and good overall outcome, earlier diagnosis of

6 Perinatal outcome in GDM 627 GDM is associated with a less favorable outcome, especially for the newborn. Acknowledgments The Investigation Centre belongs to the Centre Network of the ISCIIII, C03/08, Molecular determinants of Metabolism, Nutrition, and Hormonal Biocommunication ( ). References 1. Persson B, Hanson U. Neonatal morbidities in gestational diabetes mellitus. Diabetes Care 1998; 21: B79 B HAPO Study Cooperative Research Group. The hyperglycemia adverse pregnancy outcome (HAPO) study. Int J Gynecol Obstet 2002; 78: Metzger BE, Coustan DR. Summary and Recommendations of the Fourth International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes Care 1998; 21: Bartha J, Martinez P, Comino R. Gestational diabetes mellitus diagnosed during early pregnancy. Am J Obstet Gynecol 2000; 182: Summary and Recommendations of the Second International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes 1985; 34: Metzger BE. Summary and Recommendations of the Third International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes 1991; 40: Corcoy R, Cerqueira MJ, Codina M, Ordóñez J, Leiva A, Cabero L. Diagnóstico de la diabetes gestacional: importancia del screening rutinario y utilidad relativa de los factores de riesgo. (Diagnosis of gestational diabetes: importance of routine screening and relative usefulness of risk factors). Av Diabetol 1988; 1: Jovanovic L, Ilic S, Pettitt D, Hugo K, Gutierrez M, Bowsher R et al. Metabolic and immunologic effects of insulin lispro in gestational diabetes. Diabetes Care 1999; 22: Working group on the very low birthweight infant. European community collaborative study of outcome of pregnancy between 22 and 28 weeks gestation. Lancet 1990; 336: Cornblath M, Reisner SH. Blood glucose in the neonate and its clinical significance. N Engl J Med 1965; 273: Robson AM. Consideraciones generales sobre la asistencia del niño enfermo: tetania hipocalcémica. In: Behrman RE, ed. Tratado de Pediatría. Madrid, Spain: Interamericana-Mcgrawhill, 1992: Maisels MJ. Bilirubin: on understanding and influencing is metabolism in the newborn infant. Pediatr Clin North Am 1972; 19: Santamaria Lozano R, Verdú Martín I, Martín Caballero C, García López G. Tablas españolas de pesos neonatales según edad gestacional. Barcelona, Spain: Master-Graf SL, 1998: Berkowitz G, Roman S, Lapinski R. Maternal characteristics, neonatal outcome, and the time of diagnosis of gestational diabetes. Am J Obstet Gynecol 1992; 167: Svare J, Hansen B, Molsted-Pedersen L. Perinatal complications in women with gestational diabetes mellitus. Acta Obstet Gynecol Scand 2001; 80: Meyer WJ, Carbone J, Gauthier DW, Gottmann DA. Early gestational glucose screening and gestational diabetes. J Reprod Med 1996; 41: Super D, Edelberg S, Philipson E, Hertz R, Kalhan S. Diagnosis of gestational diabetes in early pregnancy. Diabetes Care 1991; 14: Albareda M, Caballero A, Badell G, Piquer S, Ortiz A, de Leiva A et al. Diabetes and abnormal glucose tolerance in women with previous gestational diabetes. Diabetes Care 2003; 26: Garcia-Patterson A, Erdozain L, Ginovart G, Adelantado JM, Cubero JM, Gallo G et al. In human gestational diabetes mellitus congenital malformations are related to pre-pregnancy body mass index and to severity of diabetes. Diabetologia 2004; 47: Address for correspondence: Rosa Corcoy Department of Endocrinology Hospital de la Santa Creu i Sant Pau C/Sant Antoni M. Claret Barcelona Spain rcorcoy@santpau.es