British Journal of Obstetrics and Gynaecology March 1977. Vol84. pp 197-204

PLASMA HUMAN PLACENTAL LACTOGEN PROFILES OVER 24 HOURS IN NORMAL AND DIABETIC PREGNANCY BY

M. D. G. GILLMER, Lecturer R. W. BEARD, Professor Department of Obstetrics and Gynaecology, St Mary’s Hospital Medical School, London W2

N. W. OAKLEY, Senior Lecturer FIONAM. BROOKE,Research Assistant Department of Human Metabolism and Alexander Simpson Laboratory St Mary’s Hospital Medical School, London W2 M. BRUDENELL, Consultant Department o,f Obstetrics and Gynaecology, King’s College Hospital, London SE5 AND

T. CHARD, Professor Department of Obstetrics and Gynaecology, St Bartholomew’s Hospital, London ECI

Summary Plasma human placental lactogen (HPL) concentrations have been measured at frequent intervals over 24 hours in 24 normal women, 13 chemical and 13 insulin dependent diabetic patients during the third trimester of pregnancy. Most of the women studied displayed variation in HPL levels during the day, but there was no evidence of a diurnal rhythm, nor significant changes following meals or during the nocturnal fast. The mean plasma HPL concentrations over 24 hours in the chemical and insulin dependent diabetic patients ( 5 . 9 I S D 1.3 and 5.9*1.7 pg/ml respectively) were greater than those in normal women ( 5 . l i l - 2 pg/rnl) but this difference was not significant. No significant change in plasma HFL concentrations was observed during a five-hour oral glucose tolerance test in either the normal or chemical diabetic group. It is suggested that in normal pregnant women HPL does not itself contribute to glucose homeostasis but acts as a ‘regulator’ which alters the actions of certain maternal hormones to achieve a favourable environment for fetal growth.

are poorly understood. A rise in plasma concentration has been observed during insulin induced hypoglycaemia (Gaspard et al, 1974) and after prolonged fasting in mid-pregnancy (Kim and Felig, 1971). Intravenous glucose injection has been shown to reduce plasma HPL in some studies (Burt et a1, 1970; Gaspard et al, 1974) but not in others (Saaman et al, 1966; Ajabor and Yen, 1972) while no change

HUMANplacental lactogen (HPL) displays a close structural and biological similarity to human growth hormone (HGH) (Niall et al, 1971 ; Li, 1972) and there is general agreement that it probably contributes to the increased free fatty acid mobilization, peripheral insulin resistance and enhanced insulin response to glucose of late pregnancy (Yen, 1973). The mechanisms for control of HPL secretion 197

198

GILLMER, BEARD, OAKLEY, BROOKE, BRUDENELL AND CHARD

has been observed after oral glucose administration (Grumbach et al, 1968; Pavlou et al, 1973; Kiihl et al, 1975). The present study had two aims: (i) to ascertain the relation between plasma glucose and plasma HPL concentrations during the course of a day in normal and diabetic women in an attempt to clarify the mechanisms which control placental HPL secretion, and (ii) to determine whether the abnormal plasma glucose levels in diabetic patients are associated with alterations in plasma HPL concentrations. PATIENTS AND METHODS Fifty patients were recruited to the study. Thirteen were insulin dependent diabetic women who had been on insulin treatment for 6 to 26 years. The remaining patients comprised 31 with one or more of the features of potential diabetes (W.H.O., 1965) and 6 control patients. Each patient was admitted to hospital and the study started at 1000 hours. Blood samples were collected hourly from an indwelling catheter in an antecubital vein until 2000 hours. During this time the patient was encouraged to be up and about. A total daily carbohydrate intake of 180 g was provided of which 40 g was consumed at home as part of the patient’s breakfast. Main meals containing 40g of carbohydrate were taken at 1200 hours and 1800 hours and snacks containing 20g of carbohydrate at 1100, 1500 and 2100 hours. From 2200 hours until 0800 hours the following morning, blood samples were taken every two

hours. At 0900 hours following the overnight fast, a five hour 50 g oral glucose tolerance test (GTT) was started on those patients not requiring insulin. During the test the patient remained awake, semi-recumbent in bed and was not allowed to drink or smoke. Patients were assigned either to a ‘normal’ or to a ‘chemical diabetic’ group depending on the result of the GTT. The response to this test was assessed by measuring the area under the three hour glucose curve (Harding et al, 1973). In this study the upper limit of normal for the third trimester was taken as 750 area units (Gillmer et al, 1975). Twenty-four of the patients studied were classified on the result of the GTT as ‘normal’ and 13 as ‘chemical diabetic’. (No difference was found between the results of the 18 potential diabetics with a normal GTT and the six controls and these two groups have therefore been combined.) I n calculating the mean data and performing between-group comparisons, three of the patients studied have been omitted. These were one normal woman (Patient No. 17) with a twin pregnancy and two insulin dependent diabetic patients (Nos. 51 and 53) who were delivered of congenitally malformed infants. The details of the remaining 47 patients are presented in Table I . The insulin dependent patients were studied after at least two weeks in hospital during which optimal diabetic control had been achieved with careful dietary carbohydrate management and twice daily injections of soluble and isophane insulin. From Table I it may be seen that the three groups of patients

TABLE1 Details of the patients studied

Normal Chemical diabetic Insulin dependent diabetic

Maternal weight at

Height

delivery (kg)

(cm)

Gestation at study (weeks)

Number

Age (years)

23

26.3 &4.7

72.71 $15.64

159.26 d15.35

33.9 k2.2

27.5 $6.1

73.59 19.77

161.5 1t5.70

33.2 4-2.9

25.5 53.9

70.98 ik7.41

162.91 34.7 16.46 1-1.2 _____________~

13 11

Gestation Birth at delivery Birth weight weight (weeks) (kg) (centile)

38.8 k1.28

3.215 Ik0.485

44.91 123.77

38.5 LI1.13

J

3.387 0.407

59.38 -i-22.78

3.327 10.466

60.91 126.67

~.~

- ~ _ _

37.6 +0.51

_ _ _ ~ -

24 HOUR PLASMA HPL PROFILES IN DIABETIC PREGNANCY were well matched for age, weight, height and gestation at study. Blood was collected into lithium heparin tubes and immediately centrifuged for plasma separation. Plasma glucose was measured by an automated glucose oxidase-peroxidase method (Cramp, 1967) within 24 hours of collection, on plasma kept at 4 "C. The plasma for the HPL determinations was stored at -20 "C, and HPL concentration was measured by a semi-automated radio-immunoassay (Letchworth et al, 1971). The within-assay precision for this method, expressed as the coefficient of variation, is 5 per cent. The coefficient of variation for the between-assay precision using the sanie batch of 125 I-HPL label is also 5 per cent but rises to 11.5 per cent when different batches of label are used. The mean of all the plasma glucose and plasma HPL values obtained from the start of the study at 1000 hours until its conclusion at 0800 hours the following morning has been calculated and these have been called the 'mean 24-hour plasma glucose' and 'mean 24-hour HPL' respectively. Statistical analyses were performed using the SPSS package available at the University of London Computer Centre. Student's 't' test was used to assess the significance of the difference between the means of the study groups. Pearson's correlation coefficient was used to assess the correlation between mean 24-hour glucose and HPL. Pearson's coefficient of variation was used to assess the variation in HPL concentration over

24 hours relative to the variation inherent in the radio-immunoassay method.

RESULTS The individual plasma HPL concentrations in the controls, chemical diabetics, and insulin dependent diabetics are shown in Figures 1, 2 and 3. Despite the apparent lack of diurnal variation in many of the women studied, the coefficient of variation calculated from all the CHEMICAL DIABETIC PATIENTS 7-

Case Coeffic~ent Number of Variation

a

5_ \_ ~ _ 0 %

=3

I

fb4b

2

I VB

t

931

t

59.

r

16

f 10.19

26

f 17.21

27

i 56.37

4

f

4

n

. 12md

Gpm 12mn TlMC (hours)

f 11.75 &

IS

? a 6 7

1

f 9.41

9

Coefficient of Variation

t

199

%/----J-\-----

41

i3.81

e 15

r

11 1 3

1

6am

FIG.1 Plasma HPL concentrations in the six control subjects.

12md

6pm l2mn TIME (hours)

6am

FIG.2 Plasma HPL concentrations in the 13 chemical diabetic patients.

200

GILLMER, BEARD, OAKLEY, BROOKE, BRUDENELL AND CHARD

INSULIN DEPENDENT DIABETIC PATIENTS Case Coefficient Number of Variation

II 7 42

f w.57

43

f 24.36

44

f 7-

45

f 13.W

46

? 4.76

47

i irm

48

+ 11.11

49

f l1.W

w

? 11.87

51

f

52

f 4.15

53

2

15.49

ti4

f

11.15

6-

4-

F h

8

5 *. -

11.91

I

12md

6pm l2mn TIME (hours)

6am

FIG.3 Plasma HPL concentrations in the 13 insulin dependent diabetic patients.

HPL values of each patient exceeded the within assay variation of 5 per cent in all but four of the patients (one normal, one chemical diabetic, and two insulin dependent diabetics). The mean 24-hour plasma glucose and HPL values calculated for each group are shown in Table 11. The mean 24-hour plasma glucose of the chemical diabetics (5.61 *SD 1.03 mmol/l), and the insulin dependent patients (6.02*SD 1.26 mmol/l) were both significantly greater than that of the normal women ( 4. 70kSD 0.38 mmol/l); P