Type 1 diabetes and pregnancy: trends in birth weight over 40 years at a single clinic

OBJECTIVE

To examine changes in perinatal mortality and birth weight of babies born to mothers with pregestational type 1 diabetes over 40 years in a single teaching hospital clinic.

METHODS

This was a retrospective survey of cases from the combined diabetes and obstetrics antenatal clinic at the Royal Infirmary of Edinburgh and Simpson Memorial Maternity Pavilion, Edinburgh, Scotland. Birth weight, standardized birth weight, and perinatal mortality were obtained from 643 singleton babies born after 28 weeks of gestation to mothers with pregestational type 1 diabetes between 1960 and 1999.

RESULTS

There was a dramatic improvement in perinatal mortality rate, falling from 225 (per 1,000 total births after 28 weeks of gestation) in the 1960s to 102 in the 1970s, 21 in the 1980s, and 10 in the 1990s (P < .001 for effect of birth year). In contrast, standardized birth weight (adjusted for sex, gestational age, and parity), which was significantly higher than the background population (+1.41 standard deviations above the population norm, P < .001) showed no significant change over time.

CONCLUSION

Changes in diabetic management and obstetric practice over the 40 years of our survey have resulted in enormous improvements in the outlook for offspring of mothers with diabetes. Somewhat surprisingly this has not been associated with a reduction in overgrowth of the fetus.

LEVEL OF EVIDENCE

II-2.

The relation of insulin, leptin and IGF-1 to birthweight in offspring of women with type 1 diabetes

OBJECTIVE

Maternal diabetes is associated with excess foetal growth. We have assessed the influence of maternal diabetes on hormones associated with foetal growth and the relationship of these hormones to birthweight.

DESIGN

Case-control study.

PATIENTS

Singleton offspring of mothers with type 1 diabetes (ODM, n = 140) and control mothers (Control, n = 49).

MEASUREMENTS

Birthweight, cord blood insulin, proinsulin, 32-33 split proinsulin, leptin, IGF-1, IGFBP-3, cortisol.

RESULTS

Maternal diabetes was associated with higher birthweight (ODM 3.80 +/- 0.69 kg; Control; 3.56 +/- 0.52 kg, P = 0.02) and marked increases in insulin (median [interquartile range]: ODM 110 [60-217] pmol/l; Control 22 [15-37] pmol/l; P < 0.0001) and leptin (ODM 32 [15-60] ng/ml; Control 9 [4-17] ng/ml; P < 0.0001) but no absolute difference in IGF-1 (ODM 7.9 [6.2-9.8] nmol/l, Control 7.5 [6.2-9.8] nmol/l, P = 0.24) or its principle binding protein IGFBP-3 (ODM 1.63 +/- 0.38 micro g/ml, Control 1.63 +/- 0.28 micro g/ml; P = 0.12). Individually, insulin, insulin propeptides, leptin, IGF-1 and IGFBP-3 were significantly (P < 0.05) correlated with birthweight (in ODM and Control). IGF-1 and leptin were positively related to birthweight independently of each other and insulin in both ODM and Control. By contrast, insulin showed independent relationships to birthweight in ODM (P < 0.0001) but not in Control (P = 0.4).

CONCLUSIONS

Maternal diabetes is associated with marked elevation of insulin and leptin in cord blood of their offspring. Hormonal correlates of birthweight differ between ODM and Control with an independent relationship of insulin to birthweight observed only in ODM.

Type 1 diabetes-related antibodies in the fetal circulation: prevalence and influence on cord insulin and birth weight in offspring of mothers with type 1 diabetes

During pregnancy, maternal type 1 diabetes-associated autoantibodies may cross the placenta. It is proposed that insulin antibodies (IA) allow transfer of insulin across the placenta, contributing to fetal hyperinsulinemia and macrosomia. We assessed the prevalence of IA, the tyrosine phosphatase IA-2, and glutamic acid decarboxylase (GADA) in cord blood from offspring of mothers with type 1 diabetes (ODM, n = 138) and control mothers (control, n = 47) and further assessed cross-sectional relationships of antibody titers to birth weight and fetal insulin. In ODM, antibodies were frequently present in cord blood; 124 ODM (95%) were positive for IA, 82 (59%) were positive for GADA antibodies, and 61 (44%) were positive for IA-2 antibodies. In controls, GADA and IA-2 antibodies were absent, whereas seven controls (15%) were positive for IA at low titers (P < 0.0001 ODM vs. controls for all).ODM with IA (IA positive) or without IA (IA negative) had similar birth weights (mean +/- sd: IA positive, 3.8 +/- 0.7 kg; IA negative, 4.0 +/- 0.6 kg; P = 0.31) and cord insulin concentrations (IA positive: median, 112 pmol/liter; interquartile range, 62-219 pmol/liter; IA negative: median, 114 pmol/liter; interquartile range, 59-194 pmol/liter; P = 0.96). Similarly, the presence of GADA and/or IA-2 autoantibodies (n = 103) was not associated with differences in birth weight or insulin concentrations. Antibody titers were not associated with birth weight or insulin as continuous variables in either controls or ODM. Islet autoantibodies and IA are a common finding in cord blood of ODM, but we found no evidence that they influence offspring insulin concentrations or weight at birth.

Insulin and insulin propeptides at birth in offspring of diabetic mothers

Maternal diabetes during pregnancy is associated with excess fetal growth and increased fetal insulin production. We hypothesized that insulin propeptides (proinsulin and 32-33 split proinsulin) might be more robust indicators of chronic fetal overproduction of insulin. We examined insulin-like molecules in cord blood (ILM) (insulin, proinsulin, and 32-33 split proinsulin) in relation to birth weight, maternal glycemia, and cord glucose in 140 offspring of mothers with type 1 diabetes (ODM) and 49 offspring of mothers who did not have diabetes (CONTROL) as well as degradation of ILM in response to sampling conditions at birth. Insulin propeptides were abundant in cord blood, comprising 50% of ILM in CONTROL and 36% in ODM (P < 0.0001) and more resistant to degradation than insulin (P < 0.05). Concentrations of all three ILM were highly intercorrelated with median values 2- to 5-fold higher in ODM than CONTROL [e.g. median (range): insulin ODM 110 (60-217) pmol/liter; CONTROL 22 (15-37) pmol/liter; P < 0.0001]. In ODM, 32-33 split proinsulin and proinsulin were more closely related to birth weight (Spearman r for ILM: r(32-33 split)= 0.54; r(PROINSULIN): r = 0.54; r(INSULIN) = 0.40: r(32-33 split) and r(PROINSULIN) > r(INSULIN)P < 0.05) and fetal leptin (r(32-33 split)= 0.55; r(PROINSULIN); r = 0.54; r(INSULIN) = 0.22: r(32-33 split) and r(PROINSULIN) > r(INSULIN)P < 0.05) than insulin). By contrast, insulin was more closely related to cord glucose (r(32-33 split) = 0.15; r(PROINSULIN): r = 0.10; r(INSULIN) = 0.42: r(INSULIN) > r(32-33 split) and r(PROINSULIN)P < 0.05). In CONTROL, 32-33 split proinsulin was also more closely related to fetal leptin r(32-33 split)= 0.61; r(PROINSULIN): r = 0.29; r(INSULIN) = 0.33: r(32-33 split) > r(INSULIN)P < 0.05). In ODM, 32-33 split proinsulin and proinsulin have closer relationships to fetal growth and leptin concentrations at birth than insulin. Measurement of insulin propeptides may be advantageous in assessment of the influence of maternal hyperglycemia on the newborn.

Factors affecting fetal weight distribution in women with type I diabetes

OBJECTIVE

To identify factors independently affecting fetal weight in women with type I diabetes.

DESIGN

Prospectively recorded data in consecutive women with type I diabetes, between 1975-1992.

SETTING

Simpson Memorial Maternity Hospital, Edinburgh. Population Three hundred and two pregnancies with type I diabetes identified before pregnancy, with antenatal care and delivery in the Simpson Memorial Maternity Hospital, a singleton pregnancy, and the same diabetic physician.

METHODS

Normal ranges for birthweight were established for the total hospital population. All cases and the total population had pregnancy dating by ultrasound. The relation between standardised birthweight and explanatory variables was investigated using correlation analysis, t tests and chi2 tests as appropriate, and subsequently using multiple linear regression.

RESULTS

Standardised birthweight in cases, compared with the reference population, showed a unimodal, approximately normal distribution, markedly shifted to the right (mean + 1.26 SD). The most predictive variable was glycated haemoglobin concentration at 27-33 weeks, which explained 6.3% of the birthweight variance, while smoking explained 2.7% and maternal weight 2.0%. There was a trend towards a negative relationship with glycated haemoglobin concentration at 6-12 weeks. Smoking and glycated haemoglobin concentration were strongly intercorrelated.

CONCLUSIONS

Most of the variance in standardised birthweight remains unexplained, but glycated haemoglobin concentration at 27-33 weeks is the most powerful explanatory variable. Possible reasons why there is not a stronger relationship between markers of maternal glycaemia and birthweight are discussed.

What do pregnant women think about the HIV test? A qualitative study

Before any new antenatal screening test is introduced, the opinions of pregnant women should be considered. This is particularly relevant with HIV testing. This qualitative study reports the views of 29 women attending an antenatal clinic in a large maternity hospital in Scotland where a trial of different ways of offering HIV testing on a universal, voluntary basis occurred. Women were in favour of a test offer, although they did not necessarily wish to accept testing for themselves. Generally they were more worried about having an unhealthy baby. There was a commonly held view that routine testing would cause less anxiety because it would eliminate the stigma of saying yes to testing. A move towards the HIV test being recommended to pregnant women as opposed to merely offered is likely to be acceptable, would probably increase uptake rates and should therefore be assessed.

Uptake and acceptability of antenatal HIV testing: randomised controlled trial of different methods of offering the test

OBJECTIVE

To determine the uptake and acceptability of different methods of a universal offer of voluntary HIV testing to pregnant women.

DESIGN

Randomised controlled trial involving four combinations of written and verbal communication, followed by the direct offer of a test. The control group received no information and no direct offer of a test, although testing was available on request.

SETTING

Hospital antenatal clinic covering most of the population of the city of Edinburgh.

SUBJECTS

3024 pregnant women booking at the clinic over a 10 month period.

MAIN OUTCOME MEASURES

Uptake of HIV testing and women's knowledge, satisfaction, and anxiety.

RESULTS

Uptake rates were 6% for those in the control group and 35% for those directly offered the test. Neither the style of leaflet nor the length of discussion had an effect on uptake. Significant independent predictors of uptake were a direct test offer; the midwife seen; and being unmarried, previously tested, and younger age. Knowledge of the specific benefits of testing increased with the amount of information given, but neither satisfaction nor anxiety was affected by the type of offer.

CONCLUSIONS

The universal offer of HIV testing is not intrusive and is acceptable to pregnant women. A policy of offering the HIV test to all women resulted in higher uptake and did not increase anxiety or dissatisfaction. Uptake depends more on the midwife than the method of offering the test. Low uptake rates and inadequate detection of HIV infection point to the need to assess a more routine approach to testing.

Effect of absent end diastolic flow velocity in the fetal umbilical artery on subsequent outcome

Sixty babies, delivered over a six and a half year period, who had absent end diastolic frequency (AEDF) in the umbilical artery, were studied. Individually matched control pregnancies for gestational age, birth-weight, maternal clinical condition and date of delivery, in whom umbilical artery recordings showed end diastolic frequency, were also studied. Matching was achieved in 36 cases. Neonates from case pregnancies showed no increase in necrotising enterocolitis, intraventricular haemorrhage, pneumothorax, neonatal death or bronchopulmonary dysplasia. However, they were significantly less likely to require ventilation for respiratory distress syndrome (P = 0.02). Although AEDF indicates a fetus under vascular stress, this finding alone will include a spectrum of response in the baby, from the well compensated to the irreversibly damaged. Delivery at different points in the deteriorating fetal environment may explain discrepant study results. This intrauterine stress, by increasing fetal corticosteroid and thyroid hormones, may account for enhanced lung maturity. Predictions of neonatal course need to be based on more comprehensive awareness of fetal status.

The effect of human immunodeficiency virus infection and drug use on birth characteristics

OBJECTIVE

To explore the effect of human immunodeficiency virus (HIV) infection and drug use on birth weight, length, and gestational duration at delivery.

METHODS

Subjects had a history of injection drug use or a sexual partner who was an injection drug user, were Scottish, and their HIV serostatus during pregnancy was known. Control pregnancies were matched for age, parity, ethnic group, year of delivery, and postal code sector of home address. In addition, some were matched for smoking and housing deprivation score. Birth weights were standardized for gestational age by expressing them as z scores with a mean of zero and a standard deviation of unity. Statistical analysis was by univariate and multiple regression with multilevel modeling.

RESULTS

Regression analysis for birth weight, gestational age, and gestation-adjusted birth weights (z score) included 789 pregnancies in 693 women. Human immunodeficiency virus seropositivity was associated with a z score that was 0.27 lower (P = .03), but there was no significant difference in gestational duration at delivery. Current oral or injection drug use were associated with a reduction in standardized birth weight (z score -0.27, P = .06, and z score -0.28, P = .04, respectively), and injection drug use with a reduction in occipitofrontal circumference only (1.8 cm reduction, P = .05). Injection drug use, but not the other factors, had an effect on gestational age at delivery (1.54 weeks earlier, P < .001).

CONCLUSION

Although HIV seropositivity is associated with a small reduction in standardized birth weight, this effect is less than that attributable to smoking and may not be of clinical significance. The effect seems to be associated with placental size. Opiate use, regardless of route, had a small association with reduced birth weight, suggesting a specific drug effect. However, only injection drug use had a strong association with early delivery, and this effect was likely to be clinically significant at the population level.

Smoking during pregnancy: the dose dependence of birthweight deficits

OBJECTIVE

To assess whether a simple urine based estimate of relative daily nicotine intake could predict smoking related birthweight deficits more accurately than self-reported cigarette consumption.

DESIGN

Active smokers were identified by a simple qualitative colorimetric urine test procedure and their relative nicotine intakes assessed by determining the ratios of the urinary concentrations of nicotine plus its metabolites to creatinine using automated colorimetric methods.

SETTING

A large teaching hospital.

PARTICIPANTS

Three thousand and thirty-eight mothers from whom smoking histories had been elicited and who gave birth to live singleton babies after 28 weeks of gestation.

MAIN OUTCOME MEASURES

Birthweights (adjusted for maternal weight, maternal age, baby's sex, parity and length of gestation), maternal weight gains during pregnancy and placental weights.

RESULTS

The adjusted birthweight deficits of babies born to proven active smokers averaged 226 g (95% confidence interval 194 g to 258 g), but dose dependent effects were only apparent when nicotine intake was based on urinary nicotine metabolites/creatinine ratios. Among the smokers, adjusted birthweights fell linearly with increasing nicotine intakes but gave a predicted mean birthweight for nonsmokers that was 102 g (95% CI 50 g to 154 g) lighter than that actually found (P < 0.0001). Maternal weight gains during pregnancy were substantially reduced in smokers and correlated more closely with urinary nicotine metabolite excretions than with reported daily cigarette consumptions. Placental weights were unaffected by smoking.

CONCLUSION

There was a closer dose-effect relationship between birthweight deficits and urinary nicotine metabolites/creatinine ratios than with self-reported daily cigarette consumptions. The influence of nicotine exposure on birthweight appears to be biphasic, with one mechanism operating at very low levels of tobacco smoke intake and the other causing seemingly linearly related effects over the whole range of nicotine intakes of active smokers. These findings support recent evidence that passive smoking can cause substantial birthweight deficits.

Clinical and ultrasound prediction of macrosomia in diabetic pregnancy

OBJECTIVE

To study prospectively the prediction power, at different gestations, of clinical and ultrasound measurements for fetal size in diabetic pregnancy.

SETTING

A large combined obstetric diabetic clinic in a teaching hospital.

PARTICIPANTS

One hundred and eighty-one pregnancies in which women had scans at least two of three specific time points and who were delivered of singletons after 34 weeks: 73% were pre-gestational insulin-dependent diabetics, the others were pre-gestational White class A or gestational diabetics.

INTERVENTIONS

Clinical estimates of fundal height and fetal size and ultrasound estimates of abdominal circumference and head circumference were routinely carried out at gestational ages of 28, 34 and 38 weeks or before delivery.

MAIN OUTCOME MEASURES

Standardised birthweight, corrected for gestation and parity. The relation with clinical and ultrasound measurements was investigated using multiple linear regression and the capability of the measurements to predict macrosomic births (> 95th centile of normals) using receiver-operator characteristic curves.

RESULTS

All measurements are poor predictors of eventual standardised birthweight. Prediction improves with closeness to delivery. Prediction is significantly improved by adding ultrasound to clinical information, but at 34 weeks or later this only contributes 8% of the variance. There is no difference in the prediction power for macrosomia between clinical and ultrasound measurements.

CONCLUSIONS

Even regular serial scanning and clinical examination will not always diagnose the macrosomic fetus in diabetic pregnancy. In our hands, clinical examination is as predictive as ultrasound measurements. Ultrasound does add to clinical prediction power but only to a small extent. Ultrasound should be used in a selected way, as defined by clinical findings, and with recognition and understanding of the errors and biases involved.

Guidelines for the management of insulin-dependent diabetes mellitus in pregnancy

The infant of an insulin-dependent diabetic mother is at increased risk of perinatal death, neonatal problems and major congenital malformations: Many of these problems are preventable. All young women with diabetes should receive contraceptive advice and information about pregnancy. The objects of pre-pregnancy care are to assess suit-ability for pregnancy, to optimise control in early pregnancy and to improve pregnancy outcome through the provision of individualised education and information. Pre-pregnancy care can reduce the congenital malformation rate to approximately that of the nondiabetic. In each area there should be one designated diabetologist and one designated obstetrician who, together with their team, should see all pregnant women in a combined clinic in a hospital with an intensive care baby unit. All pregnant women with diabetes should have 24-hour access to the specialist team. Tight glycaemic control during pregnancy can reduce complications of pregnancy greatly, improving infant mortality and morbidity. Insulin requirements usually change during pregnancy. Education about hypoglycaemia and avoidance of ketoacidosis is essential. Women should have regular examination of the fundi and renal function. They should have ultrasound scanning to assess gestation, to look for abnormalities and to assess fetal growth. Fetal monitoring should be used, particularly for those at high risk. Women with good diabetic control and no complications of diabetes or pregnancy may be delivered at 39 to 40 weeks but those at high risk earlier. During labour or caesarean section blood glucose should be normalised using intravenous glucose and insulin supervised by a specialist team. An experienced paediatrician should be available. Breast feeding should be encouraged.

Maternal HIV and drug use: effect on health and social morbidity

A retrospective analysis of routine child health surveillance information was performed on health visitor records of 459 children, to examine the independent effects of maternal HIV infection and drug use during pregnancy on morbidity in the first 3 years of life. No significant differences were observed in the developmental progress of children born to HIV infected or drug using women when compared to community controls. The pattern of medical consultations in the first 18 months of life was significantly different, maternal drug use exerting a negative influence on outpatient visits (odds ratio 0.6, 95% confidence interval 0.4 to 1.0). At 6 weeks, the majority of children lived with their birth parent(s), and no differences were observed between the groups. By 10 months of age, only 81% of children born to HIV infected drug using women lived with their parent(s). While maternal drug use and HIV did not have adverse effects on child health and development, these findings highlight the social implications for children affected by the heterosexual spread of HIV.

Maternal HIV infection, drug use, and growth of uninfected children in their first 3 years

OBJECTIVE

To determine the separate effects of maternal HIV infection and drug use during pregnancy on growth of uninfected children in their first 3 years.

DESIGN

Retrospective analysis of measurements from health visitor records made during routine child health surveillance at 6 weeks, 10 months, and 3 years of age. Multilevel analysis allowed for between-infant variation in fitted growth lines, and adjustment for other factors. Growth was described in terms of an intercept (z score at term) and growth slopes (change in z score per year) up to, and from, 4 months.

SUBJECTS

290 case babies delivered in Edinburgh hospitals to women who reported injection drug use by either themselves or their HIV infected partner, and 186 community controls. A total of 131 (45%) of the case babies were born to women who used drugs, predominantly opiates, during pregnancy and 93 (32%) to HIV infected women. The eight infected children were excluded from analysis.

MAIN OUTCOME MEASURES

Age and sex standardised z scores for height, weight, and body mass index.

RESULTS

459 (96%) of the 476 records for cases and controls were traced, yielding 1432 weight and 939 height measurements. Maternal HIV infection was not found to affect growth; at 3 years the estimated effect on weight z score was 0.16 with 95% confidence interval (-0.25 to 0.57) and for height 0.18 (-0.19 to 0.55). Drug use during pregnancy was associated with lighter babies at 40 weeks followed by depressed growth in the first four months, these infants remaining just slightly smaller at 3 years with an estimated effect on z scores of -0.5 for weight with 95% confidence interval (-0.89 to -0.11) and -0.37 (-0.72 to -0.02) for height.

CONCLUSIONS

Maternal HIV infection does not adversely affect growth in uninfected infants, and the effect of drug use during pregnancy is limited to small decrease in size at 3 years.

Does early growth delay occur in diabetic pregnancy?

OBJECTIVE

To identify the date of ovulation in pregnant women with Type 1 diabetes in order to assess the validity of the concept of early growth delay.

DESIGN

Identification of ovulation by measurement of urinary luteinising hormone and assessment of fetal growth using ultrasound scan.

SETTING

Diabetic pre-pregnancy and antenatal clinic in a teaching hospital.

SUBJECTS

Twenty women with Type 1 diabetes who had attended a pre-pregnancy clinic.

MEASURES

Urinary LH, by laboratory and kit methods, during conception cycles. Human chorionic gonadotrophin measured in early pregnancy. Early ultrasound scans by a single observer blind to menstrual and ovulation dates.

OUTCOME

Gestation calculated from ovulation date and gestation estimated from menstrual dates, compared with gestation at age indicated by early ultrasound scan.

RESULTS

When the date of ovulation was identified in 20 women with Type 1 diabetes there was no evidence of growth delay in any pregnancy. When gestation was estimated from menstrual dates there was apparent early growth delay in six pregnancies.

CONCLUSION

This study, together with others discussed, indicates that early growth delay is probably an artefact of incorrectly estimated ovulation date.

Spread of maternal HIV infection in Scotland from 1990 to 1992

In Scotland since January 1990, unlinked anonymous testing of Guthrie cards has documented maternal HIV-1 antibody in neonatal blood. District postcode and quarter year of birth determined prevalence and spread of infection. The Fujirebio particle agglutination assay screened for HIV-1 antibody, with confirmation by ELISA and full western blotting. Births to known HIV infected women were reported to the Royal College of Obstetricians and Gynaecologists. 0.3/1000 childbearing women were infected with HIV-1 with no significant increase from 1990 to 1992. Spread of infection from 11 to 26 districts has occurred. In 1990, 74%(14/19) of HIV positive deliveries were known to obstetricians falling to 33%(7/21) in 1992. Spread of HIV-1 infection has occurred to mothers who live outside closely defined areas and who do not belong to recognised high risk groups. In Scotland, two thirds of mothers and their infants will not receive early prophylactic care for their HIV disease.

A population-based, controlled study of the relation between HIV infection and cervical neoplasia

OBJECTIVE

To study a geographically defined population of women exposed to a drug-related risk for human immunodeficiency virus (HIV), together with carefully matched neighbourhood controls, in order to examine (1) the proportion of different groups having cervical cytopathology screening; (2) the association between HIV infection and cervical intraepithelial neoplasia; (3) the independent effect of CD4+ lymphocyte count and duration of HIV infection; and (4) the correspondence between cervical cytopathology and colposcopically directed biopsy.

DESIGN

A population-based study.

SUBJECTS

All women domiciled in Lothians with the following characteristics: between 1983 and 1987 they had a history of injection drug use or a seropositive partner with a history of injection drug use; they had a pregnancy after that exposure where their serostatus was known. In addition, neighbourhood controls were identified by computer matching for the following criteria--post-code sector, housing deprivation score, age, parity, pregnancy outcome, ethnic group, year of pregnancy, smoking.

MAIN OUTCOME MEASURES

Search was carried out for the cervical smear nearest in time to the index pregnancy end date, providing serostatus was known at that time. All identified smears were assessed by a cytopathologist without knowledge of clinical information, study group or serostatus.

RESULTS

Of 376 women, appropriate cervical smears were identified for 336. The proportions screened in the different groups were similar. There were more abnormal smears in the seropositive group than in the drug-related seronegative (P < 0.01) or the neighbourhood control groups (P < 0.001). HIV-infected women with abnormal smears had lower CD4+ lymphocyte counts (P < 0.0005). There was a reasonable correspondence between cytopathological classification and histological grading, and this was of similar strength in cases and controls (weighted kappa 0.72, 0.74).

CONCLUSIONS

There is a definite relation between HIV infection and cervical intraepithelial neoplasia. This is related to immune depletion but whether this is the sole mechanism for the association is unclear.

Lymphocyte subpopulations in early human pregnancy

OBJECTIVE

To examine changes in lymphocyte subpopulations in early pregnancy using a methodologically appropriate study design that addresses previous sources of bias.

METHODS

Thirty-seven healthy women without risk factors for human immunodeficiency virus (HIV) were reviewed when less than 9 weeks pregnant (median 51 days, range 44-61) and again at least 4 weeks following termination of pregnancy. No woman took the oral contraceptive pill. Blood was taken on each occasion at the same time of day under the same conditions of rest and food intake, transported immediately to the laboratory, and directly prepared for analysis. Lymphocyte surface markers were determined by staining with dual-colored, isotype-matched monoclonal antibody fluorescent conjugates, followed by whole blood lysis and subsequent flow cytometric analysis.

RESULTS

Pregnancy was associated with a significant reduction in total lymphocytes (P < .0001) and also in CD4+ cells, whether expressed as a percentage of lymphocytes (P = .004), an absolute count (P = .0006), or a ratio (P = .01). Change was independent of the basal level except for lymphocytes, and almost all indices had significant correlations between pregnant and nonpregnant values.

CONCLUSIONS

In this study design, each woman served as her own control and all factors remained constant except the pregnancy state. Early pregnancy causes a reduction in total lymphocytes of about 6% expressed as a percentage of total white cell count, and in CD4+ cells by 3% as a percentage of lymphocytes, or 100/mm3. We believe this fall can be accepted as definitive.

Immunohistochemical characterization of endometrial lymphoid cell populations in women infected with human immunodeficiency virus

OBJECTIVE

To determine whether lymphocytic infiltration of the endometrium accompanies human immunodeficiency virus (HIV) infection.

METHODS

Endometrial samples from 12 HIV-infected women and from rigorously matched controls were examined. The following markers were used: common leukocyte antigen (CD45), T lymphocytes (CD3), monocytes-macrophages (CD68), and CD4 and CD8 lymphocytes. Cell counts were performed without knowledge of HIV status. Factors considered in relation to these markers were menstrual symptoms, pelvic pain, peripheral blood CD4+ count, and time since seroconversion.

RESULTS

Histology showed conventional features of chronic endometritis in only one case. In the remainder, the endometrium of HIV-infected women, compared with controls, showed an increase in CD45 cells (P < .02) and an increase in CD3 staining cells (P < .05). This appeared to be restricted to those with menstrual symptoms, and this group also had lower peripheral blood CD4 counts. There was no difference in cells of the monocyte-macrophage series (CD68). In contrast to control samples, CD4 lymphocytes were infrequent or absent in the endometrium of HIV-infected women, regardless of peripheral blood CD4 count or presence of menstrual symptoms; however, this was not universal, as one sample showed an area of dense CD4 cell infiltration. The ratio of CD4 to CD8 was reduced in HIV-seropositive samples compared with controls (P < .02).

CONCLUSION

We hypothesize that chronic endometritis of a nonclassical form may be common in advancing HIV disease, possibly directed against HIV-infected cells or self-determined antigens. This could be associated with morbidity and may represent a reservoir of infection. Endometrial depletion of CD4 cells is a common, but not universal, feature and may be independent of immune compromise.

Lack of association between maternal phosphoglucomutase-1 phenotype and fetal macrosomia in diabetic pregnancy

OBJECTIVE

To assess the reports that maternal phosphoglucomutase-1 (PGM1) phenotype is highly related to macrosomia in diabetic pregnancy. This could be either a direct metabolic phenomenon, or the PGM1 locus could be a marker for a tightly linked gene involved in the maternal control of fetal growth.

DESIGN

A comparative biochemical genetic study.

SETTING

A large diabetic pregnancy clinic.

SUBJECTS

One hundred and fifty-two women who had diabetes during pregnancy, 136 being insulin dependent before pregnancy. Two hundred and thirty-six women without pre-existing medical or pregnancy complications who functioned as a control group.

MEASURES

PGM1 phenotype was assessed by conventional electrophoresis and subgroups were examined using iso-electric focusing.

OUTCOME

Standardised birthweight was corrected for sex, maternal parity and gestation confirmed in every case by early pregnancy ultrasound. Maternal diabetes control was assessed by glycosylated haemoglobin.

RESULTS

No differences were found in the observed phenotype frequencies for diabetics and control pregnant women. No association between PGM1 phenotype and macrosomia in diabetic pregnancy was found. PGM1 did not make a significant contribution to birthweight, standardised birthweight, length or ponderal index of the baby as assessed by multiple regression.

CONCLUSIONS

Our study of a larger number of insulin dependent diabetics in Scotland makes the claim that macrosomia in diabetic pregnancy is associated with PGM1 phenotype unlikely to be of general significance.

HIV testing and prevalence in pregnancy in Edinburgh

The objective was to study the changes in pregnancy HIV prevalence with time. Data were collected from multiple sources to provide a comprehensive record of all HIV seropositive pregnant women identified in the Edinburgh area (Scotland) until December 1992. There were 177 pregnancies in 108 HIV seropositive identified women. Risk factors were injection drug use (79% of pregnancies) and a known HIV seropositive injection drug-using partner (16%). Prevalence has decreased for Edinburgh City women from 0.5% of all pregnancies in 1986 to 0.1% in 1992; It was higher for induced abortion (0.6%) than for delivery (0.2%). HIV testing in pregnancy has declined. Comparison with unlinked anonymized testing showed that in 1990-1991, 20/22 seropositive women were known. In 1992, only 3 of 10 seropositive pregnancies were identified. The cohort initially infected by exposure to a 'drug related' risk factor between 1983 and 1985 may have increasingly finished childbearing, deliberately decided against pregnancy because of HIV status, and declined because of death, illness and emigration from the area, There may not have been major early tertiary heterosexual spread; however, data from 1992 suggest that this could now be impacting on pregnancy prevalence. Local testing policies have not adapted to this possible change.

Insulin requirements during pregnancy in women with type I diabetes

OBJECTIVES

To document individual variations in the rise in insulin requirements during type I diabetic pregnancies, to relate the degree of increase to maternal characteristics and fetal outcome, and to examine these factors in a subgroup of patients experiencing a large fall in insulin requirement in the third trimester.

METHODS

Insulin dose was documented in 237 pregnancies in women with type I diabetes. Multiple regression analysis was performed to identify significant associations with maternal and fetal characteristics. Eighteen pregnancies with a fall in insulin requirement of 30% or more in the third trimester were considered in detail.

RESULTS

The mean absolute increase in insulin requirement was 52 units. The degree of rise was significantly related to maternal weight gain between 20-29 weeks and maternal weight at booking, and was inversely related to duration of diabetes. It was not related to the degree of diabetes control, complications of pregnancy, White class, or outcome of pregnancy. In the 18 women experiencing a large fall in insulin requirement, there was no relation with maternal characteristics or fetal outcome.

CONCLUSION

There is a wide individual variation in the change in insulin requirements in type I diabetic pregnancy. The degree of increase is related only to maternal weight gain during weeks 20-29 and maternal weight at booking, and is inversely related to duration of diabetes. Large falls in insulin requirement remain unexplained and may not be associated with placental failure.

Fetal growth, gestation length and phosphoglucomutase-1 phenotype

This study investigates reports that phosphoglucomutase-1 (PGM1) phenotype is associated with fetal growth and gestation length. A total of 350 women were studied, 234 having uncomplicated pregnancies and 114 with a baby weighing greater than 90th centile, corrected for parity, gestation and fetal sex. All women had gestation confirmed by early ultrasound. Conventional cellulose acetate electrophoresis was used to distinguish the three common PGM1 phenotypes and polyacrylamide gel isoelectric focusing to distinguish the ten PGM1 subtypes. Neither PGM1 phenotype nor subtype were found to be associated with gestation length or standardised birth weight. Logistic regression, where maternal age, parity, fetal sex, maternal weight, gestation and smoking were introduced as explanatory variables in addition to PGM1 phenotype testing against the dependent variables birth weight, standardised birth weight and gestation length, did not show differences related to PGM1 phenotype. Two possible reasons for the discrepancy with previously published data are discussed. We conclude that the study provides no support for the belief that PGM1 phenotype is related to fetal growth or gestation length and that the original observations could have arisen as a result of statistical artefact due to multiple testing.

The effect of introduction of umbilical Doppler recordings to obstetric practice

OBJECTIVE

To assess the effect on obstetric practice of clinician access to umbilical artery Doppler ultrasound results.

DESIGN

Randomised controlled trial.

SETTING

A large teaching hospital.

SUBJECTS

Two thousand two hundred and eighty-nine pregnancies defined as being at risk by referral for Doppler or fetal monitoring.

INTERVENTIONS

Continuous wave Doppler studies of umbilical artery. Results immediately available to clinicians.

MAIN OUTCOME MEASURES

Fetal outcome: perinatal mortality, Apgar score and admission to the neonatal unit. Obstetric intervention: admission to hospital, induction of labour and caesarean section. Use of of fetal well being: cardiotocography, biophysical profile and ultrasound biometry.

RESULTS

The treatment and control groups were comparable in age, parity, gestation at point of entry and risk features. There were no overall differences in perinatal outcome, obstetric intervention or use of fetal monitoring. Examination of a subset recruited only because of hypertension or suspected intrauterine growth retardation (n = 754) similarly showed no difference attributable to group randomisation. Comparison of only those pregnancies retrospectively defined as low risk and high risk showed more use of cardiotocography in the high risk group with access to Doppler (P = 0.007) but no difference in the low risk group.

CONCLUSION

Doppler umbilical artery recording has been shown to perform well in prediction power of antenatal fetal compromise. What has been examined in this study is the response of clinicians to the test. The results suggest that obstetricians do not use the test to modify their risk assessment, and, therefore, the need for fetal monitoring in particular pregnancies. There is a real need for accumulation of information from very large data sets, particularly in the prediction power of Doppler for antenatal fetal compromise from apparently chronic utero-placental cause to guide use of monitoring resources. If simply added to existing fetal monitoring techniques in a hospital where these are widely used, then umbilical artery Doppler recordings may at present simply involve extra resources of staff and expenses, without benefit.

Maternal factors in HIV transmission

In order to identify features associated with an increased risk of transmission of HIV from seropositive women to their offspring, 70 children of 58 HIV seropositive mothers were studied. Fifty-six children were followed prospectively from pregnancy; in 14 identified after the puerperium, obstetric notes were reviewed and stored serum was tested. Twelve infants of 10 mothers were HIV infected. Risk of transmission was increased in the first year after seroconversion; 5/9 infants born at this time were infected compared with 7/61 born subsequently (P < 0.001). Progression to stage IV in transmitters was more likely, occurring in the mothers of 9 infected children at a median of 3 years (range 0.5-6.5) and in mothers of 19 non-infected children at a median of 5 years (range 1-7) (P = 0.032). Maternal CD4+ counts < 400 x 10(6)/l were found in 7/12 transmitting and 7/49 non-transmitting pregnancies (P < 0.01). Differences in HIV antigenaemia did not reach significance. These factors may influence the counselling of mothers regarding their child's and their own prognosis.

Management of pregnancy in women with HIV infection

HIV infection is a common medical problem in pregnancy in many parts of the world. Management involves supportive multiagency care, easy and effective communication, an awareness and understanding of other issues (which may be more important to the mother than her HIV infection), and specialists with both experience and an up-do-date knowledge base.

Survival time after AIDS in pregnancy

OBJECTIVE

To examine the suggestion, based on theoretical considerations and case reports, that pregnancy decreases survival time after AIDS (acquired immunodeficiency syndrome).

DESIGN

A total population study in Edinburgh.

SETTING

A city with a moderately high prevalence of human immunodeficiency virus (HIV) infection in women.

SUBJECTS

AIDS has been diagnosed in 22 women, five of whom had a pregnancy.

MAIN OUTCOME MEASURES

Clinical characteristics, disease presentation, lymphocyte markers, pregnancy outcome, subsequent progress and survival time.

RESULTS

Pregnancy was not obviously associated with a difference in clinical findings. The mean survival time for the three women with a pregnancy who died was 24 months and for the 11 women without a pregnancy it was 15 months. (P = 0.63 log rank test).

CONCLUSIONS

The clinical presentation, severity of the illness and laboratory findings were not obviously different in pregnancy. All three women who had Pneumocystis carinii pneumonia for the first time in pregnancy survived this initial episode. Survival time was not obviously reduced by the conjunction of pregnancy with AIDS.

Anticardiolipin antibodies and pregnancy outcome in women with human immunodeficiency virus infection

Abnormal anticardiolipin antibodies are associated with adverse pregnancy outcome in the general obstetric population. We studied 55 pregnancies in women infected with human immunodeficiency virus type 1 (HIV-1) to examine two hypotheses: that a substantial proportion of these women would have raised anticardiolipin antibodies, and that the affected pregnancies would be more likely to be complicated by preterm delivery and intrauterine growth retardation. Blood was taken at the initial antenatal visit, and was also taken from 15 HIV-seronegative intravenous (IV) drug users and 20 controls with no drug history. The reference for immunoglobulin G class antibodies to cardiolipin was calculated from 50 healthy young adults, and the upper limit of normal was defined as 4 standard deviations above the mean for this reference series. Twenty-four percent of the infected women, but none of the HIV-seronegative IV drug users and none of the control group, had raised anticardiolipin antibodies. There was no statistically significant relationship with birth weight, standardized birth weight, length, gestation, occipitofrontal circumference, or adverse pregnancy outcome. No women gave a history of thromboembolism or recurrent abortion, and in no pregnancy was there clinically apparent thromboembolism. We conclude that abnormal anticardiolipin antibodies provoked by HIV infection are common, but probably have a narrow specificity for cardiolipin, are unrelated to phospholipid antibody syndrome, and cannot explain adverse pregnancy outcome in HIV-infected women.

Doppler umbilical artery flow velocity waveforms in diabetic pregnancy

OBJECTIVE

To assess the effect of uncomplicated diabetes on umbilical artery flow velocity waveforms (FVWs); to investigate the relation between glycaemic control and FVWs and the predictive value of umbilical artery FVWs for antenatal fetal compromise.

DESIGN

Prospective descriptive study.

SETTING

A large diabetic pregnancy clinic in a teaching hospital.

SUBJECTS

128 pregnancies complicated by diabetes mellitus. 170 non diabetic women with no pre-existing or pregnancy complications.

INTERVENTIONS

In diabetic pregnancies, umbilical artery resistance index (RI) Doppler recordings and glycosylated haemoglobin were measured every 2 weeks from 28 weeks.

MAIN OUTCOME MEASURES

Umbilical artery RI and antenatal fetal compromise defined as a non reactive, decelerative cardiotocograph and/or a biophysical profile score persistently less than 6 and leading to immediate caesarean section.

RESULTS

Uncomplicated diabetic pregnancies had FVW values similar to those in the non-diabetic range. Glycaemic control was unrelated to umbilical artery FVW values. Abnormal umbilical artery RI was found in nine pregnancies, these were more likely to show evidence of fetal compromise and to be associated with birthweights of less than 50th centile. In seven pregnancies there was evidence of fetal compromise, but only three of these pregnancies had abnormal FVW values.

CONCLUSIONS

The non-diabetic range of umbilical artery RI values is appropriate for diabetic pregnancies. Long-term glycaemic control, within the range in this study, does not seem to affect umbilical artery RI. Abnormal umbilical artery RI is a significant predictor of fetal compromise in diabetic pregnancy, but fetal compromise can occur in association with normal RI values. Undue reliance should not be placed on normal FVW values in diabetic pregnancies.