الفهرس 
ClassificationsOnly 14 pages are availabe for public view
 |  | from | 369 |  |  |
| | | from | 369 | | |
Abstract
The aims of this work were:
1-To determine the incidence of gestational diabetes in Jahra Governorate of Kuwait.
2-To determine if diabetes occurred more in pregnant women with risk factors for developing diabetes than in those
with no similar factors.
3-To know which is the most important risk factor(s) of diabetes with pregnancy.
4-To test our protocol for the management of pregnant mothers with diabetes mellitus and to evaluate its effect on maternal and foetal outcome.
5-To test our protocol in the immediate care of the infants of diabetic mothers (IDMs) and to evaluate its effect on perinatal morbidity and mortality. Material and Methods:
The study involved 500 successive Kuwaiti pregnant women regularly attending the Jahra Antenatal Clinic in the period from November 1983 to October 1986. They were divided into three groups:
Group I - 100 established diabetics seen at different stages and distributed according to White’s Classi-fication (1965).
Group II- 200 women, pregnant 32 weeks or more, having
one or more historic risk factor for developing di-abetes mellitus.
-305-
Group III- 200 women, pregnant 32 weeks or more, with no risk factors for diabetes mellitus.
A full history includig personal, social, cultural,
menstrual, obstetrical, past, and family history was taken
from every patient. A complete general, abdominal, and
pelvic examination was done. A random urine sample was tested for protein and sugar. If the urine was positive
for sugar, a second fasting sample ws tested in each sub-
sequent visit.
For groups II and III, a 1-h, 50-gm glucose load test (Oral Glucose Challenge Test ”OGCT”), as proposed by O’Sulli-van et al.(1973), was performed in the non-fasting state. Plasma glucose values of 150 mg/dl (8.3 mmo1/1) were consid-ered the cutoff level for the test to be positive. Women with a +ve OGCT were submitted to a 3-h, 100-gm Oral Glucose Tolerance Test (OGTT); the results of which were considered according to the criteria of the National Diabetes Data Group (1979), to diagnose diabetes with pregnancy. Women were redistributed so that those with abnormal OGTT were shifted to the diabetic group under their proper White’s Classes and those with normal OGTT were detained under
their own groups.
Women with normal OGTT were followed up on outpatient basis till delivery. Those with abnormal OGTT were managed in the hospital along with established diabetics for control of their carbohydrate intolerance. A 3-h, 75-gm OGTT was performed on abnormal 100-gm OGTT patients 6 to 8 weeks postpartum, to confirm diagnosis of gestational diabetes.
-306-
The goal of glycwmic control in diabetic patients was to attain a mean daily plasma glucose of 80-100 mg/d1 (4.4-5.8 mmo1/1), which was the average of a fasting and three 6-hourly preprandial values. The daily food intake was planned to supply a caloric intake of 30-35 cal/kg of ideal body weight, fractioned to contain 20 to 25% as protein, 45 to 50% as carbohydrate, and 30 to 35% as fat. Most of the patients were requested to consume 25, 30, and 30% of the calories for breakfast, lunch and supper. The remaining 15% were provided by 2 to 3 snacks per 24 hours. Insulin was administered as a mixture of regular and intermediate-acting insulin, in 2 subcutaneous doses before breakfast and supper. Two thirds of the total dose were given in the morninig and one third in the evening. Adjustments were made on daily basis for inpatients and weekly basis for outpatients, aiming at a fasting plasma glucose level of <100 mg/dl (5.6 mmo1/1) and of <120 mg/d1 (6.7 mmo1/1) at any of the other samples. Patients brought under control were discharged from the hospital and followed up in the Outpatient Clinic till the 28th week to be readmit-ted for 5 to 7 days for review of their control. Further admission was carried out at the 32nd week till delivery. Non-controlled, difficult-to control, or complicated cases were retained in the hospital for longer periods.
Admitted patients were submitted to daily clinical examination and 4-hourly urine testing; twice weekly diabet-ic profile including plasma glucose, cholesterol and trigly-cerides, serum creatinine and BUN levels; weekly ultra-
-307-
sound scans, plasma oestriol, Hb A1c and serum alpha-foeto protein levels. At the 36th week, ultrasound for biophysical profile and placental localization, NST and foetal kick-count charting were requested.
Delivery was attempted always after the completed 37th week depending on the clinical, biochemical and bio-
physical profiles. The vaginal route was preferred unless there was an obstetric indication for caesarean section.
Upon delivery, all IDMs were examined for general appearance, Apgar score, birth weight, length, gestational age, head and chest circumferences, congenital anomalies and other complications. Blood glucose was estimated at one hour, 2 hours, 4 hours and later,at 6 hours. Serum calcium, electrolytes, bilirubin, hwmatocrit were estimated daily along with routine analysis. X-raying was requested, if needed. Results:
Screening for gestational diabetes:
OGCT was positive in 21 subjects (10.5%) with historic risk factors and in 5 subjects (2.5%) with no such factors (P<0.05), with an overall prevalence of 6.5% in the pooled sample of 400 subjects. 13 patients (6.5%) in the risk group and 3 patients (1.5%) in the control group had abnor-mal 100-gm OGTT (P<0.05), with an overall figure of 4%. 75-gm OGTT at the end of puerperium was abnormal in only 2 women with historic risk factors and abnormal 100-gm
-308-
OGTT during pregnancy. The incidence of gestational diabetes was 5.5% with risk factors and 1.5% without risk factors (P<0.0%), with an overall incidence of 3.5% in the pooled sample.
The predictive value of positive OGCT to detect diabetes was 61.3%. Combination of risk factors was the most frequent (81.3%) finding in patients with abnormal OGTTs, while grand-multiparity was the most frequent (56.3%) single risk factor in these patients.
Patient redistribution:
Depending on the OGTT results, the diabetic group included 116 patients distributed into 49 (42.2%) Class A (non-insulin -dependent) and insulin-dependent Classes B 32(27.6%), C 19 (16.4%), D 12(10.3%), and R/F 4(3.5%). The risk group includ-ed 189 women, and the control group included 195 women.
Course of pregnancy:
Minority of our diabetic patients fell outside the glycmmic range specified. Diurnal plasma glucose ranges were signifi-cantly higher in diabetic than in non-diabetic patients (P<0.05). Plasma glucose profiles were significantly higher in IDDs than in NIDDs at most checktimes (P<0.05). Hb A1c levels were significantly higher in the diabetic group compared to the non-diabetic groups (P<0.05) in the third trimester in spite of metabolic control. They were signifi-cantly higher in IDDs than in NIDDs at all checktimes throu-ghout pregnancy (P<0.05). Hb A1c levels correlated with
-309
maternal mean plasma glucose in the prior 6-8 weeks reflect-ing the degree of metabolic control (r=0.42, P<0.05). Levels did not correlate with maternal age, weight, parity, family history of diabetes or gestational age in all groups. No significant correlation was found between maternal Hb A1c levels and neonatal birthweight in the study groups. In the diabetic group, however, a weak correlation was found only in the latter part of pregnancy. We failed to demonst-rate a correlative relationship between maternal Hb A1c levels and congenital malformations in the 3 study groups.
There was no significant intergroup differences in fasting plasma cholesterol levels. Triglyceride and high-density lipoprotein cholesterol levels, however, occupied a midposition in diabetic patients. Plasma lipids were generally higher in NIDDs than in IDDs (P<0.05). No signifi-cant intergroup differences were found in serum creatinine and BUN levels, which were nearly 10% lower than those of non-pregnant ranges.
Mean levels of plasma oestriol were generally rising with the progress of pregnancy. However, there were no significant intergroup differences. Mean last reading before delivery was higher, although non significantly, in diabetic than in non-diabetic, and in risk than in control women (P<0.05). Levels did not correlate with neonatal Apgar score. In diabetics alone, plasma oestriol’s last reading
before delivery significantly correlated with neonatal
birthweight and with placental weight (P<0.05). Low levels,
-310-
however, did not strongly correlate with non-reactive NSTs. Significant DROP in plasma oestriol was observed in 50% of diabetics, being accompanied by abnormal NST and/or biophysical profile scoring in only 30% of them.
Maternal serum alpha-foetoprotein significantly decreas-ed throughout pregnancy. In diabetics, they were significant-
ly higher from one trimester to another (13(0.05) till they
finally dropped from 32 weeks till delivery. Around their time of delivery, levels were persistently higher in 13% of diabetics, 4.3% of risk and 3.1% of control subjects. High levels associated 27.3%, 25% and 33.3% of congenital malformations in diabetic, risk and control groups respectiv-ely. Ultrasonography was more sensitive for the same indicat-
ion.
Significantly highest frequency of non-reactive and lowest frequency of reactive NSTs were found amongst diabet-ic patients (P<0.05). Opposite findings were found in cont-rol subjects. Predictive value of reactive NST was more than 90% regarding Apgar score, and more than 80% for intra-partum foetal heart tracing and neonatal metabolic disturb-ances, but the predictive value of non-reactive NST was
generally poor
Ultrasound gestational age prediction was sensitive in 90%, 83% and 76% of diabetic, risk and control groups respectively. For less than 2. week-error predicting the date of delivery, diabetic scans were 18% better than both dating and clinical estimation, compared to 17% and 14% difference in risk and control scans. For less than one
-311-
week-error, differences were minimized to 14% and 15% in diabetics, 14% and 22% in risk and 14% and 17% in control groups respectively. Repeatedly reactive NSTs were more specific to predict a normal neonatal course than correspond-ingly normal Ultrasonic biophysical profile. Also, non-reactive NSTs were more sensitive, compared to ultrasound, to predict an abnormal neonatal course.
Of our diabetics, 42.2% were NIDD and 57.8% were IDD. Of the diet-controlled cases, one third were gestational diabetics and two thirds were established cases. 3 of the gestational diabetics (7%) required insulin during pregnancy and the rest were managed with diet alone. Dietary treatment was considered inappropriate if fasting and postprandial plasma glucose values exceeded 125 and 165 mg/dl (6.9 and 9.2 mmo1/1), respectively in which case insulin was instituted. The preprandial concentrations of glucose exceeded 150 mg/dl (8.3 mmol/1) in 20% of diabetics in the third trimester. Maternal insulin requirement increased as term approached reaching their maximum around the 32nd week. 90% of IDDs required 2 daily doses of insulin mixtures. For the other 10%, th evening dose was split to give the intermediate-acting insulin at bedtime. Ketoacidosis was encountered in 7 (3.4%) of our NIDDs but this was managed easily with no harmful effects on the foetus. Following delivery, insulin demands dropped in 95.5% to their prepreg-nancy doses.
-312-
Labour and delivery:
18% of our diabetics were primigravidae and 82% were multigravidae. Of the latter, 21 had had one or more abort-ions or premature deliveries. 2 diabetics and one in the risk group were delivered to twins. Most of our women did not require amniocentesis for estimation of the L/S ratio of lung maturity. Induction of labour was resorted to in 17 (15.2%) and succeded in 15 (12.9%) of our diabetics. 3.2% and 2% were induced in the risk and control groups. Only one case was difficult to induce in the risk group. Failed inductions ended up by caesarean sections. Abnormal intrapartum foetal heart tracing was recorded in 14.5%, 14.3% and 8.2% of diabetic, risk and control parturients respectively. Type I dips were the commonest abnormalities in the tracings. 82.8% of diabetics were delivered vaginally. Of them 8.6% went to post-term due to non-compliance. In 12.1% of the vaginal births, instrumental assistance was needed, through vacuum (8.6%) or low forceps (3.5%) extract-ion. 4.3% had difficult labour and 2.6% had shoulder dysto-cia. The rate of operative vaginal delivery was nearly double that in the risk and control rates (P<0.05). The caesarean section frequency was 17.2% in diabetics, compared to 9.2% and 6.6% in risk and control groups. Differences were significant (P<0.05).
Foetal/Neonatal outcome (Infant of Diabetic Mother ”IDM”)” Of 118 infants born to diabetic mothers, 54.2% were
males and 45.8% were females. Mean gestational ages were
-313-
not statistically different between infants born to carbo-hydrate-intolerant and carbohydrate-tolerant women lingering
around 38 weeks. Significantly higher numbers of infants of diabetic mothers were depressed compared to those of
non-diabetics (P<0.05). After a minute, 18.8% had an Apgar score of <7 and after 5 minutes , 4.3% had a score of <7. 29% of IDMs, 2/3 of whom belonged to IDD mothers, were macro
-somic with a mean gestational age of 38.4 weeks at delivery
IDMs were significantly heavier than those of risk mothers (P<0.05). The latter were similarly heavier than those of control mothers (P<0.05). Infant birthweight correlated with maternal mean diurnal plasma glucose levels in the third trimester only (r=0.4, P=0.03). A relationship between infant birthweight and maternal body weight (before pregn-ancy and gestational weight gain) as well as with grand multiparity was found but the correlation was not signifi-cant. Small for gestational age represented 3.5%, 9% and 1.5% of infants in the diabetic, risk and control groups respectively (P<0.05). Birth trauma t pneumothorax complicat-ed 3.4% of IDMs. Respiratory distress syndrome prevailed in 4.3% of IDMs, which was 4 times that in the risk and 9 times in the control infants (P<0.05). Hypoglycaemia was frequent in 11.1 of IDMs; 85% of which was in in those of IDD mothers. Significantly lower figures of 3.7% and 1% were in risk and control groups (P<0.05). Incidence of hypocalcwmia was 2.6% in IDMs compared to 0.5% in other infants(P<0.05). Byperbilirubinaemia complicated 19.7% of IDMs which was 2.5 times that in the risk and nearly 10 times that in the control groups (P<0.05).
-314-
The congenital malformation rate was 9.4% in IDMs; an incid-ence that was 5 times that in potential diabetics and 6 times that in the normal population (P<0.05). The malform-ations were equally divided between major and minor with a predominance of multiple anomalies. Neural tube defects were met with in IDMs alone, where ultrasound and alpha-foetoprotein were equally sensitive to diagnose them. Perinatal morbidity as a whole was correlated to maternal diurnal plasma glucose in the third trimester (r=0.48, P=0.01).
Two antenatal deaths occurred, one in the diabetic and the other in the risk groups. Intrauterine death in the diabetic group was due to multiple congenital anomalies, while that in the risk group was explained by abruptio placentae. 3 infants in the diabetic group died in the early neonatal period of RDS, traumatized meningocele and VSD respectively. One infant in the risk group also died neonatally of RDS. Perinatal mortality rate (PNMR) was signi-ficantly higher for diabetics (33.9 per 1000 live births) versus in risk patients (10.6 / 1000). Corrected PNMR was 8.5 versus 5.3 per 1000. Control group had no mortalities.