Search

Eleven pregnant women were treated for hyperthyroidism with carbimazole (CZ) and one with propylthiouracil (PTU). Based upon a previous study it was decided that lactation should be permitted if the dose required after delivery did not exceed 15 mg of CZ or 150 mg of PTU. In the patients studied here the daily dose of CZ varied from 5 to 15 mg and that of PTU was 125 mg. TSH was measured in cord blood and in the blood of the newborn infants usually after 2 and 3 weeks of lactation. Serum T4 was measured serially in the infants' blood from day 4 up to 21 day of age, at least. In all instances the TSH concentration in cord blood remained below 45 mU/l the level used in screening for neonatal hypothyroidism. Serum TSH and T4 were all within the appropriate reference limits during the 3 weeks of study with only one exception. In the infant whose mother was treated with PTU the serum T4 measured 5 d after birth was slightly below the lower limit but later returned to normal. Since serum TSH and T4 did not deviate from the reference range in newborn infants during lactation, we conclude that breast-feeding can be permitted if the daily dose of CZ does not exceed 15 mg (or 150 mg of PTU) and if facilities are available for measuring neonatal serum TSH and T4.

The effect of a rapid rise in blood glucose on glycosylated hemoglobin (HbA1) was measured by two different cation exchange chromatographic methods. Blood samples were taken after fasting and one hour after breakfast from 32 insulin treated diabetics for the measurement of 1) HbA1 with a microcolumn procedure with and without prior incubation of the samples in saline overnight and 2) HbA1 and HbA1c with the macrocolumn chromatographic method of Trivelli with and without prior dialysis of the hemolysates. Mean (+/- SEM) blood glucose concentration rose significantly one hour after breakfast from 9.8 +/- 0.8 mmol/l to 14.9 +/- 0.8 mmol/l (p less than 0.001). The corresponding HbA1-concentrations measured by microcolumn chromatography without incubation of the samples were 15.2 +/- 0.5% (fasting) of total hemoglobin and 15.8 +/- 0.6% (postprandial) (p less than 0.001) and after incubation 13.3 +/- 0.4% and 13.2 +/- 0.4%, respectively (NS). The HbA1-concentrations measured by the method of Trivelli without dialysis were 14.7 +/- 0.5% (fasting) and 14.9 +/- 0.5% (postprandial) (p less than 0.01) and after dialysis of the hemolysates 14.3 +/- 0.5% and 14.3 +/- 0.5%, respectively (NS). The HbA1c-concentrations measured by the macrocolumn method without dialysis were 11.1 +/- 0.4% (fasting) and 11.2 +/- 0.4% (postprandial) (p less than 0.05) and after dialysis 10.9 +/- 0.4% and 10.8 +/- 0.4%, respectively (NS). Acute hyperglycemia may cause an increase in HbA1-values determined by microcolumn chromatography, the magnitude of which can exceed 20% in an individual case. The acute effect can be abolished by incubation of the samples in saline overnight.

Three hundred eighty amniocenteses in 200 parturients were reviewed for success or failure in obtaining amniotic fluid, number of attempts for each amniocentesis, premature rupture of the fetal membranes, presence of blood, and the perinatal outcome. The highest success rate was in cases in which amniocentesis was performed between the symphysis and the presenting part of the fetus. The total incidence of spontaneous rupture of fetal membranes within 5 days after the last amniocentesis was 13.5%. When the amniocentesis was performed behind the fetal neck, the membranes ruptured significantly more frequently (P less than 0.05) than after amniocentesis at other sites. It is concluded that the area behind the fetal neck should be avoided if possible at amniocentesis. There were no severe be avoided if possible at amniocentesis. There were no severe fetal or neonatal complications from amniocentesis in these patients.

Cardiac tumours are rarities in newborn infants. The majority of them are rhabdomyomas as a manifestation of the basic disease-tuberous sclerosis. The clinical picture may vary from serious arrhythmias to A-V valve obstruction and gross myocardial dysfunction. Therefore these patients are extremely poor candidates for invasive diagnostic procedures and surgical treatment. Our patient had intrauterine cardiac arrhythmias, and a prenatal echocardiogram showed features of A-V septal defect malformation. Therefore the echocardiogram was repeated soon after birth and it revealed intracardiac tumours. The patient was operated on the first day of life and four intracavitary tumours were successfully removed.

Thirty-four hypothyroid women on thyroid hormone substitution were followed through 37 pregnancies, and 16 women having previous surgery for thyroid carcinoma and thereafter placed on suppressive thyroxine treatment were followed through 19 pregnancies. The thyroxine treatment needed readjustment in 13 pregnancies (23%) to maintain euthyroidism. At delivery, the maternal free thyroxine index was 126 nmol/L in the group of patients treated for hypothyroidism and 146 nmol/L in the patients with treated thyroid carcinoma. The amniotic fluid thyroxine level in normal pregnancies was 6.7 nmol/L, in hypothyroid patients 6.7 nmol/L, and in patients with thyroid carcinoma 5.6 nmol/L. The amniotic fluid reverse triiodothyronine level in normal pregnancies was 0.51 nmol/L, in hypothyroid patients 0.66 nmol/L, and in patients with thyroid carcinoma 0.70 nmol/L. All infants were euthyroid.

Telemetry and conventional cardiotocography were compared by monitoring the labor of 60 patients with an uneventful pregnancy and delivery in the 38th-42nd week of pregnancy. 31 patients were monitored by telemetry and 29 by cardiotocography. The patients were matched for age (+/- 5 years), duration of pregnancy (+/- 7 days) and parity (I or II). The husband attended labor and delivery in 42% of the cases in the telemetry group and in 59% of the cases in the control group. Induction of labor by amniotomy was performed in 32% of the cases in the telemetry group and in 24% of the cases in the cardiotocography group. The patients monitored subjective pain every half hour during the opening phase. The telemetric patients were encouraged to sit or walk during the first stage. No maternal or fetal complications occurred. All infants were born in good condition with APGAR scores greater than or equal to 7 recorded at one and five minutes. There were 4 operative deliveries in the telemetry group and 5 in the control group. Indications for these were maternal or uterine exhaustion with the exception of two control patients where fetal asphyxia was suspected. The duration of the first stage of labor did not differ significantly between the telemetry and the cardiotocography groups. The telemetric patients received less analgesics than the controls but this difference was not significant. In spite of less analgesia in the telemetry group, the secondparas of the telemetry group experienced significantly less (p less than 0.01) labor pain than the controls. In addition, the secondparas of the telemetry group considered the present labor less painful than the previous one significantly more often than the controls. Among the primiparous patients there was no difference in the amount of pain experienced by the patients.

Serum folate concentrations, blood counts, and antiepileptic drug concentrations were measured during 133 pregnancies of 125 women with epilepsy. There was an inverse correlation between serum folate concentrations and concentrations of phenytoin and phenobarbitone. The number of epileptic seizures during pregnancy showed no association with serum folate concentrations. No cases of maternal tissue folate deficiency or fetal damage attributable to low maternal serum folate were observed. Maternal serum folate concentrations for infants with structural birth defects, "fetal hydantoin syndrome," or perinatal death were similar to those for healthy babies. A low dose (100 to 1000 micrograms daily) of folate supplement appeared sufficient for pregnant women with epilepsy despite the antifolic action of antiepileptic medication. Monitoring folate concentrations in pregnant women with high serum concentrations of phenytoin or phenobarbitone is recommended.

The electroencephalograms (EEGs) of epileptic women in late pregnancy were compared with their EEGs outside pregnancy and the puerperium. The number of epileptic interictal discharges was not modified by pregnancy. Visual scoring and frequency analysis demonstrated a slight increase in the alpha band during pregnancy. No correlations were found between the EEG findings and changes in seizure frequency.

To assess the value of various methods for long-term follow-up of diabetic patients, we compared the concentrations of fructosamine in serum with those of various glycated proteins: hemoglobin (HbA1C), total serum proteins (G-prot), and albumin (G-alb), assayed in 30 pregnant insulin-dependent diabetics every two weeks after initial determination of a 24-h blood glucose profile. HbA1C correlated best with the 24-h glucose profile during the succeeding 10-35 days (r = 0.65-0.68, P less than 0.001). G-prot and G-alb correlated nearly as well as HbA1C 10-20 days after the glucose profile (r = 0.54-0.64, P less than 0.01-0.001), but only weakly after 25-35 days. Values for fructosamine did not correlate significantly with the glucose profile 10-35 days after it (r = 0.23-0.36). Evidently the fructosamine assay is not an adequate alternative to HbA1C, G-alb, or G-prot as an index to long-term control of blood glucose in such patients.

Eleven pregnant women with concomitant hyperthyroidism were treated with antithyroid drugs. At monthly intervals serum thyroxine (T4) and triiodothyronine (T3) were measured with radioimmunoassay, the Sephadex uptake of radioactive triiodothyronine (T3U) determined the free T4 and T3 indices calculated (FT4I, FT3I). TSH-binding inhibiting immunoglobulins (TBII) were determined by the radiomembrane assay. Serum TSH and T4 were measured at delivery from cord blood and/or from the newborn infants some days after birth. Serum TSH was significantly elevated in one infant. There was an inadequate post-partal rise in serum T4 concentration in this child and in another who showed only a marginal elevation of TSH. The mothers of these infants were given carbimazole in doses of 30 and 25 mg/day, respectively, at the time of delivery. No significant changes were seen in other infants, the daily doses being 20 mg of carbimazole or less. There was no clinical indication of hypo- or hyperthyroidism in any of the newborn. The TBII were positive in most patients and there was a trend of normalization during treatment. No relationship between the dose of antithyroid drug and the level of TBII could be seen. During treatment the dose was adjusted according to the FT3I values. This seems to be an adequate laboratory test for this purpose.

Haemoglobin AIc (Hb AIc) was measured in 112 insulin-dependent diabetic pregnancies on an average 4.7 times. Hb AIc was high in early pregnancy in all three patients with severe fetal malformations. There were six pregnancies with a perinatal death. The maximum Hb AIc values were significantly higher in the second trimester (p less than 0.001) in these pregnancies than in the other diabetic pregnancies. When neonatal hypoglycaemia was present, the mean maternal Hb AIc was significantly higher in the second (p less than 0.005) and third (p less than 0.02) trimesters of pregnancy than in the group without neonatal hypoglycaemia. In the group with neonatal hyperbilirubinaemia the mean maternal Hb AIc was significantly higher in the third trimester (p less than 0.02) than in the group with no neonatal hyperbilirubinaemia. The results suggest that poor metabolic control of maternal diabetes during the second trimester is associated with a clearly increased risk of perinatal death and during the second and third trimesters with metabolic derangements in the neonatal period. The results also indicate that Hb AIc could be used to detect the pregnant diabetics at special risk.

Maternal haemoglobin A1c (HbA1c) values were measured before the end of the 15th week of gestation in 142 pregnancies in women with insulin dependent diabetes. In pregnancies complicated by fetal malformations (n = 17) the mean initial HbA1c value was 9.5 (SD 1.8)% of the total haemoglobin concentration, which was significantly (p less than 0.001) higher than in pregnancies without malformations (8.0 (SD 1.4)%; n = 125). HbA1c values did not differ between pregnancies complicated by minor and major fetal malformations, but the rate of malformations showed a positive relation to the HbA1c value in early pregnancy (chi 2 = 11.9; p = 0.001). Fetal malformations occurred in six out of 17 pregnancies (35.3%) in mothers whose initial HbA1c value was 10% or more, in eight out of 62 pregnancies (12.9%) in mothers with initial values between 8.0% and 9.9%, and in only three out of 63 pregnancies (4.8%) in mothers with an initial value below 8.0%. These data support the hypothesis that the increased incidence of fetal malformations in mothers with insulin dependent diabetes is associated with maternal hyperglycaemia during organogenesis. Hence diabetic women who are planning to have a child--especially those with a high HbA1c value--should receive intensified metabolic control.

Using continuous cardiotocography, the effect of maternal position on fetal heart rate (FHR) was studied during extradural analgesia (EA) with either 50 or 100 mg of etidocaine without adrenaline for labour and vaginal delivery. Of 70 healthy parturients, 39 were supine and 40 were in the lateral position. FHR was normal in all patients during the control period before EA, and remained normal during EA in 53. Transient abnormal patterns occurred in 26 patients, 22 (56%) in the supine group and four (10%) in the lateral group, a significant difference (P less than 0.001). The decrease in arterial pressure (AP) did not differ between the groups, but the frequency of abnormal FHR associated with a small or moderate decrease in AP (30% or less) was greater in the supine (51%) than in the lateral group (3%).

Nonstress fetal heart rate (FHR) recording was used as a primary test to detect fetal distress in 145 pregnant women with insulin-dependent diabetes. Testing was performed every second day beginning with the 32nd week of pregnancy and daily after the 34th week until delivery. One hundred eighteen (81.4%) had normal, nine (6.2%) suspicious, and 18 (12.4%) pathologic FHR recordings. Poor metabolic control of diabetes was observed in 25 (17.2%) of the 145 pregnancies during the last trimester of pregnancy. Nine of these 25 women (35%) with poor metabolic control had a suspicious or pathologic FHR recording, which was significantly more frequent (P less than .02) than in women with good metabolic control (18 of 120, 15%). The mean value (+/- SD) of hemoglobin AIc during the last trimester in diabetic women with pathologic FHR records was 7.63 +/- 0.87%, which was significantly higher (P less than .02) than in diabetic women with normal FHR records (6.91 +/- 0.83%). None of the 145 fetuses monitored died in utero. It was concluded that no obvious iatrogenic morbidity was caused by early intervention in cases with pathologic FHR recordings.

Although maternal ingestion of antiepileptic drugs is strongly suspected of causing congenital defects, particularly oral clefts, the effect of epilepsy itself or a combined effect of drug intake and epilepsy have not been excluded as etiological factors. Very little is known about fetal oxygenation during a maternal grand mal epileptic seizure. We describe two cases in which fetal heart rate was recorded during a maternal epileptic seizure during labor. The first fetus became clearly asphyctic as judged from the fetal heart rate recording: immediately after the epileptic seizure there was a 13-minute continuous bradycardia wave with decreased short-term variability. After the bradycardia a phase of tachycardia with decreased short-term and long-term variability occurred. In the other fetus there was only a short period of bradycardia, which was followed by a phase of tachycardia and decreased short-term and long-term variability. Both fetuses were vigorous at birth 43 and 87 minutes, respectively, after the epileptic seizures of their mothers. We conclude that a maternal grand mal epileptic seizure can be ominous to the fetus. It is therefore important that epileptic seizures are controlled by optimal medication throughout pregnancy.

The time of ovulation and conception of 24 insulin-dependent (Type I) diabetic women was estimated by monitoring the basal body temperature (BBT). The mean (+/- SD) duration of the menstrual cycle was 28 (+/- 2) days. Conception occurred, on the average, on the 18th day of the cycle. The fetal crown-rump length was measured by ultrasound scan every 2 weeks between 6 and 14 weeks of pregnancy. When menstrual history was used for dating, there was an apparent early fetal growth delay by a mean of 4 days. When the dates were corrected by BBT, the fetal growth was identical with the standard mean crown-rump length diagram by Robinson & Fleming. The findings of our study suggest that early fetal growth delay reported by Pedersen & Mølsted-Pedersen does not exist in diabetic pregnancy. The small fetuses observed could be explained by an error in the gestational age.

We have developed a rapid chromatographic method for determination of glycosylated haemoglobins by high-performance liquid chromatography with a new cation-exchange column. The haemoglobins are eluted with a three-step gradient in 7 min, and total assay time including re-equilibration of the column is 15 min. The method permits separation and quantitation of HbAlc, even in the presence of elevated levels of HbF. HbA1a, A1b and A2 can also be determined. The results correlate well (r = 0.94) with those obtained by the macro-column method of Trivelli et al. [New Engl. J. Med., 284 (1971) 353] for determination of HbA1c. The method has been fully automated by the use of an automatic injector. The within-assay and between-assay coefficient of variation of the method is 2-3%.

The major surfactant protein with a molecular weight of 35 kd and also saturated phosphatidylcholine and phosphatidylglycerol were analyzed in specimens of amniotic fluid; 68 were from cases of maternal diabetes, 41 from preeclampsia or maternal hypertension, 26 from premature rupture of the fetal membranes, and 45 from normal pregnancies. The relationship between the individual surfactant components was studied after covariance adjustment for the length of gestation. In severe early-onset preeclampsia, the 35 kd surfactant protein/saturated phosphatidylcholine ratio was significantly higher than in the other pregnancies. In diabetic pregnancies (classes B to D without preeclampsia), the phosphatidylglycerol/saturated phosphatidylcholine ratio was lower than in the other pregnancies. Isolated surfactant complex showed similar abnormalities. In severe early-onset preeclampsia and insulin-dependent diabetes without vascular disease, the phosphatidylglycerol/saturated phosphatidylcholine ratio correlated negatively with fetal growth. In four samples of amniotic fluids from cases of severe early-onset preeclampsia, the 35 kd protein falsely predicted lung maturity. All had abnormally high 35 kd protein/saturated phosphatidylcholine ratios (greater than 2 SD of controls). According to the present results, the 35 kd protein may give a false mature test result in severe preeclampsia.

In the human fetus, elevated plasma erythropoietin levels have been found in high-risk pregnancies at delivery. We examined the relationship of amniotic fluid erythropoietin and umbilical plasma erythropoietin at delivery in 17 normal pregnancies, 41 hypertensive pregnancies, and 37 insulin-treated diabetic pregnancies terminated by elective cesarean section without labor. An additional 27 insulin-treated diabetic patients were studied after undergoing variable durations (86-1184 minutes) of labor. Erythropoietin was analyzed using a highly sensitive and specific radioimmunoassay technique. Fetal plasma erythropoietin concentrations were elevated above the control upper range (50.3 mU/mL) in 59% of the hypertensives and in 38% of the diabetics. The amniotic fluid erythropoietin values were significantly lower than the umbilical plasma erythropoietin values in each study group. Although the umbilical plasma erythropoietin values in the abnormal pregnancy groups differed considerably from the corresponding levels in the controls, the ratio of amniotic fluid erythropoietin to umbilical plasma erythropoietin was approximately the same in controls, hypertensives, and diabetics. Furthermore, the plasma and amniotic fluid levels (In transformed) correlated highly significantly in all three individual groups in absence of labor. In the diabetic labor group, this relationship was nonsignificant. We conclude that in the absence of labor, amniotic fluid erythropoietin reflects fetal plasma erythropoietin. We speculate that amniotic fluid erythropoietin may be an antepartum indicator of fetal hypoxemia.

Rapidly adhering cells (RA cells) from the amniotic fluid of a pregnancy with fetal anencephaly were investigated by immunofluorescence assay with an antiserum against glial cells. After 24 hours' cultivation a high proportion of the cells showed positive glial-specific fluorescence, whereas no staining was seen in cells from samples of normal amniotic fluid. At the 24th week the mother was delivered of a stillborn infant with anencephaly. Immunofluorescence staining of RA cells with glial-specific antiserum may be used for the differential diagnosis of fetal abnormalities associated with a high alpha-fetoprotein concentration in amniotic fluid.

A comparison of 150 pregnancies in women with epilepsy and 150 pregnancies in matched nonepileptic control women showed similar rates of pregnancy-induced hypertension, albuminuria, premature contractions, premature labor, and bleeding in pregnancy. Duration of labor, blood loss at delivery, cesarean section rates, and vacuum extraction rates were also similar among epileptic and control groups. There were five perinatal deaths in the epileptic group and two in the control group. A fetal heart rate tracing during a maternal grand mal seizure showed bradycardia, reduced short-term and long-term variability, and late decelerations suggesting asphyxia. It is concluded that grand mal seizures during pregnancy should be avoided by the use of antiepileptic drugs. Women with epilepsy require antenatal neurological and obstetric follow-up during pregnancy.

Intrauterine hydrops was diagnosed by two dimensional echocardiography. The fetus had a pericardial effusion and a thick interatrial septum without a foramen ovale flap. The condition was treated by giving the mother digitalis. Postnatally, the effusion had disappeared and a parachute mitral valve was found.

Natural surfactants consist of unique proteins and lipids. Their effectiveness in improving subnormal lung function in surfactant deficiency should be established prior to any clinical trials. Rigorous tests are required to document batch to batch variability in surface activity and to exclude toxic contaminants. Up to this date randomized clinical trials in small preterm infants have demonstrated a striking improvement in lung function, and a decrease in incidence of acute complications (pneumothorax, interstitial emphysema). Administration of human surfactant at birth or in severe RDS decreased deaths and incidence of bronchopulmonary dysplasia. Although homologous surfactant may not be more advantageous than the heterologous one in terms of its acute beneficial effects on lung function, the safety and efficacy of animal surfactant in improving the outcome remains to be established. Human surfactant may serve as a model for unlimited natural surfactant produced by gene technology. The pharmacodynamics aspects of surfactant substitution, the indications of exogenous surfactant, and the management of the patients undergoing surfactant substitution remain to be studied. Exogenous surfactant offers a potential to treat or prevent severe respiratory failure in infants, children and adults.

111 pregnancies of epileptic women on phenytoin therapy were observed in a prospective study. Maternal serum phenytoin concentrations were measured monthly or bi-weekly during pregnancy, labor and puerperium. The concentration decreased towards the end of pregnancy and was lowest at delivery. In 48% of the patients the drug dosage had to be increased to combat the increased seizure frequency.

The obstetric records of 125 infants weighing 500-1250 g who were born in 1978 and 1979 at the Helsinki University Central Hospital were studied retrospectively to find out any associations between fetal heart rate (FHR) patterns and perinatal outcome. FHR recordings of 79 fetuses were available for study. Normal reactive patterns were observed in nine, nonreactive in 54, decelerations in 58, silent patterns in 45 and combined distress patterns in 36 fetuses. The interval from the first pathological sign to delivery was up to 27 days. Five fetuses in the monitored group died in utero, 26 infants died neonatally and two later. The risk of neonatal death after pathological FHR patterns was lower than after an FHR pattern without abnormality. Idiopathic respiratory distress syndrome (RDS) was diagnosed in 32 infants, of whom 16 died. Forty-seven fetuses (59%) of the monitored group were delivered abdominally and 32 vaginally. The risk of RDS was not significantly associated with FHR patterns and mode of delivery. We conclude that unlike term fetuses, low-birthweight fetuses seem to benefit from a period of intrauterine stress reflected in abnormal FHR patterns, if caesarean section is used liberally.

Repeated amniotic fluid erythropoietin measurements in 23 Rh-immunized pregnancies were done to evaluate erythropoietin levels of amniotic fluid as an indicator of fetal distress (umbilical artery, pH 7.14 or less, or 1-minute Apgar score of 4 or less). Amniotic fluid erythropoietin levels did not vary significantly between 168 and 273 gestational days in the pregnancies without fetal distress. Increasing levels of amniotic fluid erythropoietin predicted highly reliably severe fetal distress at birth. Whether erythropoietin levels of amniotic fluid can also predict fetal distress in other pathologic pregnancies needs further study.

Forty consecutive pregnant patients with insulin-dependent (Type I) diabetes mellitus were randomized at the end of the first trimester for treatment with conventional insulin therapy (CIT) or continuous subcutaneous insulin infusion therapy (CSII). Nine patients randomized into the CSII group declined the pump treatment. The mean glycosylated haemoglobin (Hb AIc) decreased (p less than 0.001) both in the CIT and the CSII groups with no difference between the groups. Some deterioration in retinopathy was found in 2/18 patients in the CIT group, in 5/13 in the CSII group, and in 3/9 of those who declined the pump treatment. The proportion of patients whose retinopathy progressed did not differ significantly between the groups, and in the majority the deterioration was mild. However, two patients in the CSII group developed acute ischaemic retinopathy, which progressed to proliferative stage in spite of laser treatment. In these two cases the decrease in the Hb AIc level was among the greatest and fastest in the study. These data suggest that a rapid near normalization of glycaemic control by CSII during pregnancy can accelerate the progress of retinopathy in poorly controlled diabetic patients.

Determination of glycohemoglobin in blood (HbA1) represents an established measure of glycemic control in diabetic patients. In patients with uremia, however, the determination can be subject to pitfalls which may limit its reliability. In order to evaluate the clinical usefulness of HbA1 determinations in diabetic patients with nephropathy, concentrations of HbA1 and its subfractions HbA1c and HbA1a+b were measured by micro- and macrocolumn chromatography in 58 diabetic and 80 non-diabetic patients with impaired renal function. Fifteen diabetic patients without nephropathy and 15 healthy subjects served as controls. The concentrations of HbA1 and its subfractions were significantly higher in non-diabetic patients with nephropathy than in healthy controls. A positive correlation was seen between HbA1 and plasma glucose concentrations in all subjects, and between HbA1 and serum urea and creatinine concentrations in the non-diabetic subjects. When measured repeatedly in the same patient there was a positive correlation between HbA1 and plasma glucose concentrations in diabetic patients with azotemia. There was no change in HbA1 concentrations measured immediately before and after hemo- or peritoneal dialysis. The increase of chromatographically determined HbA1 concentrations in azotemic patients is most likely due to the joint action of carbamylation of hemoglobin with urea derived cyanate and deterioration of glycemic control induced by azotemia. Despite these problems, chromatographically determined HbA1 is still a clinically useful measure of glycemic control in diabetic patients with nephropathy. This presumes repeated measurements in the same patient and the use of appropriate reference levels which consider the degree of renal impairment.

A study was designed to investigate whether asthma, when carefully managed, is associated with an increased risk of complications in connection with pregnancy. One hundred and eighty one asthmatic women were monitored during 198 pregnancies. Antiasthmatic treatment consisted of inhaled beta 2 adrenergic drugs, beclomethasone, sodium cromoglycate, oral theophylline, and systemic corticosteroids as needed. Postpartum information on asthmatic symptoms and infant feeding was collected by means of a questionnaire. A control group of 198 non-asthmatic pregnant women was matched for age and parity. Atopic women had less severe asthma than non-atopic women. During pregnancy 40% of the patients were managed with the same antiasthmatic medication as before pregnancy; 18% needed less and 42% more medication. Pre-eclampsia occurred more often in asthmatic than control subjects, especially in patients with severe asthma. Hypoglycaemia occurred more often in infants of mothers with severe asthma than in infants of mothers with less severe disease. Theophylline medication at term did not influence labour or delivery. Asthma caused no emergencies during labour. Among the asthmatic subjects 28% of babies were delivered by caesarean section compared with 17% in the control group. There was no difference between asthmatic and control subjects with regard to length of gestation, birth weight, incidence of perinatal deaths, low Apgar scores, neonatal respiratory difficulties, hyperbilirubinaemia, or malformations. It is concluded that severe asthma or systemic corticosteroid treatment (or both) during pregnancy seems to increase the incidence of mild pre-eclampsia in the mother and hypoglycaemia in the infant. The findings suggest that careful supervision of asthma during pregnancy and labour by obstetricians and chest physicians working in close collaboration should prevent most of the serious obstetric and neonatal complications of asthma in pregnancy reported by previous authors.