A comparison of glyburide and insulin in women with gestational diabetes mellitus.

BACKGROUND: Women with gestational diabetes mellitus are rarely treated with a sulfonylurea drug, because of concern about teratogenicity and neonatal hypoglycemia. There is little information about the efficacy of these drugs in this group of women. METHODS: We studied 404 women with singleton pregnancies and gestational diabetes that required treatment. The women were randomly assigned between 11 and 33 weeks of gestation to receive glyburide or insulin according to an intensified treatment protocol. The primary end point was achievement of the desired level of glycemic control. Secondary end points included maternal and neonatal complications. RESULTS: The mean (+/-SD) pretreatment blood glucose concentration as measured at home for one week was 114+/-19 mg per deciliter (6.4+/-1.1 mmol per liter) in the glyburide group and 116+/-22 mg per deciliter (6.5+/-1.2 mmol per liter) in the insulin group (P=0.33). The mean concentrations during treatment were 105+/-16 mg per deciliter (5.9+/-0.9 mmol per liter) in the glyburide group and 105+/-18 mg per deciliter (5.9+/-1.0 mmol per liter) in the insulin group (P=0.99). Eight women in the glyburide group (4 percent) required insulin therapy. There were no significant differences between the glyburide and insulin groups in the percentage of infants who were large for gestational age (12 percent and 13 percent, respectively); who had macrosomia, defined as a birth weight of 4000 g or more (7 percent and 4 percent); who had lung complications (8 percent and 6 percent); who had hypoglycemia (9 percent and 6 percent); who were admitted to a neonatal intensive care unit (6 percent and 7 percent); or who had fetal anomalies (2 percent and 2 percent). The cord-serum insulin concentrations were similar in the two groups, and glyburide was not detected in the cord serum of any infant in the glyburide group. CONCLUSIONS: In women with gestational diabetes, glyburide is a clinically effective alternative to insulin therapy.

Dietary therapy for gestational diabetes: how long is long enough?

OBJECTIVE: To determine the length of time required for dietary therapy alone to effect good glycemic control and whether the need for insulin treatment can be predicted at diagnosis of gestational diabetes mellitus (GDM). METHODS: Women with GDM were treated with dietary therapy for 4 weeks. Each measured her blood glucose using a memory-based reflectance glucometer, and those in poor glycemic control (mean glucose exceeding 105 mg/dL) after 4 weeks of dietary therapy were prescribed insulin. Women were stratified by fasting plasma glucose value of 3-hour glucose tolerance tests (GTTs). RESULTS: Women with fasting glucose at or below 95 mg/dL were significantly more likely to achieve good glycemic control after 2 weeks of dietary therapy than were those with values above 95 mg/dL whose control did not improve during the study. Receiver operating characteristic (ROC) analysis determined that fasting values of GTT between 91 and 95 mg/dL best predicted that insulin would be needed for good glycemic control. CONCLUSION: Women with GDM should be prescribed dietary therapy alone for at least 2 weeks before they are prescribed insulin. In those with fasting glucose above 95 mg/dL, insulin may be prescribed after 1 week of dietary therapy, or at diagnosis.

Electronic fetal monitoring: what's reassuring?

BACKGROUND: To determine which combinations of fetal heart rate pattern abnormalities are associated with normal outcome in term pregnancies. METHODS: A cohort of 2200 consecutive deliveries was examined and the fetal heart rate tracings analyzed. Singleton, term patients without chorioamnionitis or serious malformations were used to perform logistic regression analysis to select those FHR patterns associated with increased risk for Apgar<7 and cord pH<7.15, or immediate adverse outcome. RESULTS: Patients having no fetal heart rate abnormalities, mild variable decelerations, decreased variability, mild bradycardia, or accelerations present, constituted 84% of all fetal heart rate tracings. These tracings alone, or in combination, predicted 5 minute Apgar score> or =7 in 99.7%, cord pH> or =7.15 in 96.9% and no adverse neonatal sequelae in 96.2% of cases. Accelerations were reassuring regardless of FHR pattern. When these patterns were not present, non-reassuring tracings, the risk for immediate adverse outcome increased 50%. The non-reassuring tracings were both without accelerations and had tachycardia, prolonged bradycardia, severe variable or late decelerations, or a combination of these patterns. This increased risk was independent of the risk of confounders: i.e. thick meconium (1.8-fold), prolonged second stage of labor (>50 min., 1.5-fold), maternal disease (e.g. kidney, respiratory, 3-fold), or hypertensive disease (1.9-fold). CONCLUSIONS: The great majority of fetal heart rate pattern abnormalities can be considered reassuring as they are within normal variations of a healthy fetus. The non-reassuring ones identify infants that truly require further evaluation by fetal scalp, vibroacoustic stimulation, or fetal scalp blood sampling.

Meconium: a marker for peripartum infection.

OBJECTIVE: To test the hypothesis that the presence of meconium-stained amniotic fluid (AF) is associated with maternal and neonatal infection, both before and after delivery. METHODS: Nine hundred thirty-six laboring women were analyzed for the presence of meconium in amniotic fluid and occurrence of peripartum infection. Meconium was assessed clinically as thin, moderate, or thick. Intra-amniotic infection and endometritis were diagnosed by standard definitions. All patients were tested for vaginal group B streptococcus, bacterial vaginosis, and other aerobic organisms. RESULTS: Meconium-stained AF was present in 28% of the study participants (9% thin, 12% moderate, 7% thick). The presence of meconium was associated with increased intra-amniotic fluid (17% versus 9%, relative risk [RRI 1.98, 95% confidence interval [CI] 1.3, 3.1), endometritis (10% versus 5%, RR 2.38, 95% CI 1.3, 4.4), and total infection (25% versus 13%, RR 2.19, 95% CI 1.5, 3.2). Thick meconium had higher infection rates than clear AF (44% versus 13%, RR 5.18, 95% CI 2.9, 9.3). Meconium was associated independently with peripartum infection by multiple logistic regression (RR 1.28, 95% CI 1.1, 1.6). CONCLUSION: Meconium-stained AF is associated with increased peripartum infection, independent of other risk factors for infection. Thick meconium, in particular, is associated with a marked increase in peripartum infectious morbidity.

Perinatal outcome and the type and number of maneuvers in shoulder dystocia.

OBJECTIVES: To ascertain the type and order of maneuvers that should be used for the treatment of shoulder dystocia, and to determine its correlation with perinatal outcome. METHODS: We reviewed all consecutive cases of shoulder dystocia from January 1986 to August 1994 in our institution to obtain the type, order and number of maneuvers used for delivery. Patients were stratified by the number of maneuvers required for delivery. Outcome parameters included cord pH, Apgar score, neonatal trauma (Erb's palsy and fracture), and maternal trauma. RESULTS: The incidence of shoulder dystocia was 0.7% (39,280 total vaginal deliveries). Use of only two maneuvers, McRoberts and suprapubic pressure, resulted in resolution in 58% of cases. The addition of the Woods screw maneuver and/or delivery of the posterior arm was sufficient in all remaining cases. The rates of neonatal palsy and fracture, and maternal four-degree laceration, increased with the number of maneuvers. CONCLUSIONS: The McRoberts maneuver and suprapubic pressure should be first-line treatment for shoulder dystocia. More difficult and damaging maneuvers such as Woods screw and delivery of the posterior arm may be reserved for refractory cases. Additional maneuvers are rarely necessary for delivery. The number of maneuvers may serve as a measure of the severity of the shoulder dystocia.

Perinatal outcome in growth-restricted fetuses: do hypertensive and normotensive pregnancies differ?

OBJECTIVE: To test the hypothesis that fetal growth restriction (FGR) associated with a maternal hypertensive disorder results in worse perinatal outcome than FGR in pregnancies without maternal hypertension. METHODS: All consecutive, singleton, nondiabetic, small for gestational age (SGA) deliveries (birth weight at or below the tenth percentile for gestational age) in a 15-year computerized data base were analyzed for pregnancy outcome. Perinatal outcome was compared after stratification by presence or absence of hypertensive disorders and by gestational age at delivery. RESULTS: Eleven thousand two hundred twenty-seven SGA pregnancies were analyzed. The morbidity and mortality profiles differed between hypertensive and normotensive pregnancies delivered preterm and those delivered at term. Perinatal mortality was significantly higher in the normotensive than in the hypertensive group in preterm deliveries (30.3 versus 18.7%, odds ratio [OR] 1.9 [confidence interval (CI) 1.3-2.9]). At term, hypertensive pregnancies demonstrated significantly higher mortality than normotensive pregnancies (4.6 versus 1.9%, OR 2.42 [95% CI 1.7-3.4]). In both preterm and term gestations, cesarean rates were significantly higher in hypertensive pregnancies than in normotensive pregnancies. Using logistic regression analysis, hypertension was independently associated with a 39% reduction in risk of perinatal mortality preterm, compared with a twofold increased risk of perinatal mortality at term. CONCLUSION: Before term, FGR in normotensive women resulted in significantly higher perinatal mortality than FGR in hypertensive women. In contrast, at term, FGR in pregnancies complicated by hypertension had poorer perinatal outcomes than FGR in normotensive women.

Cellular alterations in collembolan midgut cells as a marker of heavy metal exposure: ultrastructure and intracellular metal distribution.

The ultrastructure of the midgut epithelial cells of Tetrodontophora bielanensis (Collembola) fed either with lead-, cadmium- or zinc-enriched food or kept under control conditions was compared by transmission electron microscopy (TEM). Atomic absorption spectrophotometry (AAS) showed accumulation of these three metals in the body tissues of the collembolans. Intracellular localization of zinc in the midgut epithelial cells was investigated by electron spectroscopic imaging (ESI) and electron energy loss spectroscopy (EELS). The presence of lead in the midgut tissue was shown by laser microprobe mass spectrometry (LAMMS). Under heavy metal conditions, the midgut cells showed different ultrastructural alterations, the degree of which was found to be dose-dependent. Independent of the type of the metal in the food, the endoplasmic reticulum appeared fenestrated or vesiculated, and often large vacuoles, which were shown to be ER-derived, occurred throughout the cells. Also the mitochondrial membranes were affected by heavy metal stress. In epithelial cells of individuals exposed to either lead or cadmium, an increase of myelin-like structures could be observed. In the case of exposure to zinc in the highest applied concentration, the cytoplasm showed condensation and portions of the microvillous border appeared destroyed. Additionally, an increase of mineral congregations (type-A spherites) could be observed under heavy metal influence. Intracellularly stored zinc could be localised at highly affected mitochondrial membranes, in the microvillous border, and in the heterochromatin.

Do growth-retarded premature infants have different rates of perinatal morbidity and mortality than appropriately grown premature infants?

OBJECTIVE: To determine if perinatal morbidity and mortality differ in growth-retarded, small for gestational age (SGA), premature infants and appropriate for gestational age (AGA) infants. METHODS: All consecutive, singleton, nondiabetic, preterm pregnancies delivered over a 15-year period were analyzed. Infants were categorized as SGA (at or below the tenth percentile) or AGA (11th to the 89th percentiles), then stratified by birth weight and gestational age categories. Perinatal morbidity and mortality were examined. RESULTS: We studied 4183 preterm deliveries, 1012 of them SGA and 3171 of them AGA. Overall, we found significantly higher rates of fetal and neonatal death in the SGA group. Stratification by gestational age revealed significantly higher rates of neonatal death for the SGA group compared with the AGA group in each gestational age category. Overall, comparison also revealed significantly higher rates of fetal heart rate abnormality in the SGA group but no difference in neonatal sepsis, birth trauma, cesarean delivery, hyaline membrane disease, or congenital anomalies. CONCLUSION: Growth-retarded premature infants have a significantly higher risk of morbidity and mortality, both before and after delivery, than do appropriately grown infants.

Low-dose versus high-dose oxytocin augmentation of labor--a randomized trial.

OBJECTIVE: Our purpose was to compare the efficacy and safety of low-dose versus high-dose oxytocin regimens in the augmentation of labor. STUDY DESIGN: Three hundred ten term pregnancies requiring augmentation of labor underwent randomization to receive either a low-dose or high-dose oxytocin augmentation regimen. Maternal demographics, labor-delivery data, and neonatal outcome were compared. RESULTS: The high-dose oxytocin group had a significantly lower cesarean section rate, regardless of parity (10.4% vs 25.7%, p < 0.001), with no differences in maternal complications and neonatal outcomes. The time needed to correct the labor abnormality was also significantly decreased (1.24 +/- 1.4 hours vs 3.12 +/- 1.6 hours, p < 0.001) in the high-dose group. CONCLUSIONS: The use of high-dose oxytocin regimen benefits both nulliparous and multiparous women requiring labor augmentation by significantly lowering both the time necessary to correct the labor abnormality and the need for cesarean section.

Efficiency of lower threshold criteria for the diagnosis of gestational diabetes.

OBJECTIVES: To determine the incidence of adverse outcome in normal untreated gravidas with minimal hyperglycemia, classified as having gestational diabetes mellitus (GDM) by threshold criteria lower than current standards; to determine how efficient the different criteria are in identifying infants at risk for morbidity; and to explore the pathophysiology of minimal hyperglycemia using the glucose tolerance test (GTT) periodicity concept. METHODS: Seven hundred eight subjects considered nondiabetic by current ACOG criteria were reclassified by the criteria of Coustan (fasting 95, 1 hour 180, 2 hours 155, and 5 hours 140 mg/dL), Sacks (96, 172, 152, and 131 mg/dL), or Langer (at least one abnormal ACOG value). Glucose tolerance test periodicity, the incidence of large for gestational age (LGA) neonates, and macrosomia were then determined for each gravida diagnosed as having GDM by these criteria. RESULTS: Both Coustan and Langer criteria identified a significantly greater incidence of LGA infants compared with non-GDM (23.6 and 25.3%, respectively, versus 14%, P < .05), and identified them as efficiently as current criteria, approximately one LGA infant for every four GDM subjects treated. The incidence of LGA did not differ between the Sacks GDM and non-GDM groups. Glucose tolerance test periodicity for newly diagnosed GDM gravidas was significantly longer than non-GDM for Coustan and Langer criteria (3.9 and 4.06 versus 3.3 hours, P < .01) but not for the Sacks criteria. CONCLUSION: Using lower threshold criteria to diagnose GDM identified morbidity at an incidence and efficiency comparable to current standards. These newly diagnosed GDM gravidas had abnormal GTT characteristics, with each group exceeding the 3.5-hour GTT periodicity limit previously found for nondiabetic gravidas. Sack's conversion of existing standards may be too low to efficiently identify pregnant subjects at risk for increased morbidity.

Glucose tolerance test periodicity: the effect of glucose loading.

OBJECTIVE: To test the hypothesis that glucose abnormality, as shown by glucose tolerance test (GTT) periodicity, is not affected by different glucose loads, allowing for the identification of gestational diabetes mellitus (GDM) under varying glucose challenges. METHODS: Eighty subjects were tested by multiple GTTs 1 week apart. Each woman served as her own control, undergoing a standard 3-hour, 100-g GTT; then, half of the subject group randomly underwent a 50-g and the other half a 75-g, 2-hour GTT. Subjects were classified using National Diabetes Data Group thresholds for the 100-g GTT. Those with two or more abnormal values were classified as gestational diabetic (GDM group); the rest of the women were considered to be nondiabetic. The projected time for the GTT to revert to fasting value, GTT periodicity, was then determined for each glucose load in the GDM and nondiabetic groups. RESULTS: All glucose values for the nondiabetic group were significantly lower at 1 and 2 hours than those for the GDM group, regardless of the glucose load (P < .04). There was a statistically significant difference within the GDM and nondiabetic groups between glucose values of the 100- and 50-g GTTs at 1 hour (P < .02) and between all loads at 2 hours (P < .04). The GTT periodicity for the 3-hour, 100-g test was significantly longer for patients with GDM, as shown previously (5.6 +/- 1.9 versus 3.2 +/- 1.7 hours, P < .0001). In addition, similar values were found for nondiabetic and GDM subjects for the 75-g (5.1 +/- 2 versus 3.6 +/- 1.8 hours, P < .04), but not the 50-g load (2.2 +/- .6 versus 1.34 +/- .8 hours, P < .01). CONCLUSION: Glucose tolerance test periodicity will identify subjects with GDM regardless of GTT load because the physiologic disturbance of glucose level measured by this time period remains comparably longer than in normal subjects. We speculate that the relatively shorter cycle of the 50-g load may reflect an insufficient challenge to pancreatic function.

Meconium-stained amniotic fluid: increased risk for adverse neonatal outcome.

OBJECTIVE: To determine the risk of adverse neonatal outcome associated with meconium-stained amniotic fluid independent of that related to antepartum or intrapartum abnormalities. METHODS: A cohort of 2200 consecutive deliveries was examined and the fetal heart rate (FHR) tracings analyzed independently. Singleton term pregnancies without fatal malformations were stratified by the consistency of meconium and compared. RESULTS: Moderate or thick meconium increased the risk for adverse outcome more than threefold (relative risk 3.2, 95% confidence interval 2.0-5.2). This risk was independent of fetal heart tracing abnormalities or maternal hypertensive, kidney, or heart disease. CONCLUSION: Thick meconium alone should alert the physician to a high-risk fetal condition. This phenomenon requires continuous FHR monitoring and reassurance of fetal well-being by acid-base assessment or the equivalent, regardless of maternal disease status or the presence of abnormal FHR tracings.

The effect of clinical chorioamnionitis on cord blood gas at term.

OBJECTIVE: We sought to determine the effect of clinical chorioamnionitis on cord blood gas values in term pregnancies not complicated by any other disease. STUDY DESIGN: 2200 consecutive deliveries were studied. Following exclusion of twins, non-viable malformations and stillbirths, as well as mothers with high-risk pregnancy diseases--i.e. hypertension, diabetes, preterm labor, third-trimester bleeding, IUGR, postdates, oligohydramnios, i.v. drug abused, decreased fetal movement, maternal viral infection, UTI or pneumonia (n = 897)--two groups of patients remained: term pregnancies complicated only by clinical chorioamnionitis (n = 81) and uncomplicated term pregnancies (n = 1246). RESULTS: Evaluation of mean cord blood gas values revealed a significant difference in pH, PO2, PCO2 and BE values, with the infants of mothers with clinical chorioamnionitis having lower pH values (7.23 +/- 0.07 vs. 7.28 +/- 0.07). However, evaluating the independent effect of chorioamnionitis on arterial cord blood pH (using a logistic regression model) showed that clinical chorioamnionitis, by itself, did not contribute to this change in arterial cord blood pH. CONCLUSION: Chorioamnionitis was neither the explanation nor the cause for differences in arterial cord blood pH found between the two groups in our study. In cases of chorioamnionitis, these differences were attributed to other factors, such as length of labor, mode of delivery, method of delivery and presence of meconium.

Intensified versus conventional management of gestational diabetes.

OBJECTIVE: We tested the hypothesis that intensified management of gestational diabetes mellitus on the basis of stringent glycemic control, verified glucose data, and adherence to an established criterion for insulin initiation results in near normoglycemia control and reduction of adverse outcomes. STUDY DESIGN: A prospective, population-based study compared the effect on perinatal outcome of conventional (n = 1316) and intensified (n = 1145) management. Group assignment was based on availability of memory-based reflectance meters at entry to the program. A contemporaneous randomized control group (nondiabetic, n = 4922) was selected. RESULTS: The diabetic groups were comparable in demographic characteristics and in factors associated with higher risk for adverse pregnancy outcome, such as previous macrosomia, previous gestational diabetes mellitus, and family history of diabetes. The control group was younger, less obese, and had a lower rate of previous macrosomia. The intensified management group had rates of macrosomia, cesarean section, metabolic complications, shoulder dystocia, stillbirth, neonatal intensive care unit days, and respiratory complications lower than those in the conventional management group and comparable to those of the nondiabetic controls. Other maternal complication rates, such as for preeclampsia, chronic hypertension, and infection, were similar for the three groups. Mean blood glucose levels were a good predictor of perinatal outcome. Gestational age at delivery, previous history of macrosomia, and overall mean blood glucose levels were the only significant predictors of birth weight percentile in both diabetic groups (logistic regression). CONCLUSION: The intensified management approach is significantly associated with enhanced perinatal outcome. This management strategy clarifies the relationship between glycemic control and neonatal outcome.

Is fetal heart rate variability a good predictor of fetal outcome?

OBJECTIVE: The purpose of this study was to investigate, in a prospective manner, whether fetal heart rate (FHR) variability serves as a reliable single predictor of fetal outcome. METHODS: We undertook a prospective study of 2,200 consecutive deliveries. Preterms < 37 weeks, twins, stillbirths and fetal malformations were excluded from the study. FHR tracings were analyzed for variability in every delivery. Three windows were evaluated: 1) early in labor (30 min); 2) in the active phase of labor - 1 h prior to complete cervical dilatation (30 min); and 3) throughout the second stage of delivery, in segments of 30 min. Using the tracings taken at admission and prior to delivery, FHR variability was scored according to five previously used evaluative methods, and was correlated to the following three fetal outcome parameters: a) arterial cord blood pH < 7.2, > 7.2; b) 5-min Apgar >6, < 7; and c) immediate outcome ('good' defined as newborn going home after 2-5 days, with no neonatal intensive care unit (NICU) admission). RESULTS: 1. The majority of cases with adverse fetal outcome exhibited good FHR variability (mean, 80%). 2. Low FHR variability is not common in newborns with adverse general outcome (mean, 11.5%). 3. The predictability of FHR variability (using the five most common scoring methods) for evaluating adverse fetal outcome revealed low sensitivity (mean, 20.3%) and low predictive value (mean, 11.6%). CONCLUSION: FHR variability by itself cannot serve as the only indicator of fetal wellbeing. The presence of low variability should alert the physician; however, good FHR variability should not be interpreted as reassuring.

Recurrent stillbirth. Significance and characteristics.

We sought to identify the significance of recurrent stillbirth and to determine the contributory etiologic factors for this phenomenon. Data were analyzed and retrospective chart review conducted for all stillbirths occurring during a 13-year period. Subjects were divided into two groups: those for whom the current stillbirth was the first and those who had had a previous stillbirth. The study included 48,479 consecutive multiparous women, of whom 403 had delivered stillborn infants (8.31/1,000 live births). For 34 of these subjects, this represented a recurrent stillbirth (84.36/1,000 live births). The recurrent-stillbirth group had a 10.15-fold higher risk for stillbirth. Additionally, this group had a twofold higher incidence of diabetes and hypertensive disease than did those women experiencing their first stillbirths; furthermore, the gestational age and birth weight of the stillborn infants were significantly lower in the recurrent-stillbirth group (P < .0004 and < .007, respectively). Such factors as socioeconomic class, chorioamnionitis and erythroblastosis fetalis, traditionally cited as contributing to repeated fetal loss, were not significant. Although recurrent stillbirth remains an unsolved problem, improving health care to specific groups within high-risk populations may reduce fetal loss.

Glucose tolerance test: degree of glucose abnormality correlates with neonatal outcome.

OBJECTIVE: To determine how well the extent of glucose abnormality, as reflected by the number of abnormal values on the 3-hour oral glucose tolerance test (GTT), correlates with the level of carbohydrate intolerance during pregnancy and with the severity of adverse outcome. METHODS: We followed 764 gestational diabetic women under a once-per-week fasting and 2-hour postprandial serum glucose monitoring system. The subjects were stratified by the number of abnormal values on their GTTs. The level of glucose control and incidence of large for gestational age (LGA) infants were then determined and compared with the findings in 636 gravidas with abnormal screening but all normal GTT values. RESULTS: Patients with one or more abnormal GTT values had comparable incidences of LGA infants, which were all significantly greater than that in the 0-abnormal group (23-27% versus 13%; P < .01). This difference was due to subjects with poor glucose control. The means of the GTT values for each sampling time were greater and the GTT periodicity (the time for the GTT curve to return to the fasting level) was longer with an increasing number of GTT abnormal values (zero versus one versus two versus three versus four abnormal values, P < .02). The mean fasting, 2-hour postprandial, and overall mean glucose values during the study were positively associated with the number of abnormal GTT values. CONCLUSIONS: One or more abnormal GTT values were associated with comparably elevated incidences of LGA infants in patients with poor glycemic control. Achievement of recommended glucose control decreased adverse outcomes to near normal levels.