Calorie Restriction for Treatment of Gestational Diabetes ANNE DORNHORST, JONATHAN S.D. NICHOLLS, FAY PROBST, CATHERINE M. PATERSON, KAYE L HOLLER, ROBERT S. ELKELES, AND RICHARD W. BEARD

Birth weights of infants of 35 gestational diabetic mothers treated with calorie restriction alone (1200-1800 kcal) were compared with those of infants of 2337 nondiabetic women, including two control groups (A and B) matched for race, body mass index, age, and parity. All women were screened for gestational diabetes with the O'Sullivan screening method, and a 3-h oral glucose tolerance test was performed on all abnormal results. Control group A mothers had a normal screen, and control group B mothers had an abnormal screen with a normal glucose tolerance test. Pregnancy weight gain was significantly less for the gestational diabetic mothers (mean ± SD 4.6 ± 4.9 kg) than for the general prenatal population (9.3 ± 5.3 kg), group A control subjects (9.7 ± 5.3 kg), and group B control subjects (9.7 ± 5.4 kg; P < 0.0005). No infant of a gestational diabetic mother was below the 10th percentile for weight, and birth weights were similar to those of the control groups even though weight gain after the 28th wk of gestation was only 1.7 ± 1.6 kg. The frequency of macrosomia (birth weight > 4000 g) was similar among the gestational diabetic mothers (9.3%), the general prenatal population (7.4%), and group A mothers (11.6%) but significantly higher for the group B control subjects (20.9%; ** = 8.57, P < 0.005). This study demonstrated that gestational diabetic mothers who are calorie restricted have infants with normal birth weights and a frequency of macrosomia less than that of screen-positive nondiabetic women with similar macrosomic risk factors. Diabetes 40 (Suppl. 2): 161-64, 1991

F

etal macrosomia is a major factor contributing to the increased maternal and perinatal morbidity associated with gestational diabetes mellitus (GDM) (1,2). Pedersen postulated that diabetes-related macrosomia was a consequence of fetal hyperinsulinemia caused by maternal hyperglycemia (3,4). O'Sullivan et al. (5) reported that the incidence of macrosomia fell from 13.1 to 4.3% when GDM was treated with diet and insulin. In this study, weight gain during late pregnancy was curtailed by 2.2 kg, which would have contributed to the reduction in macrosomia independently of any effect of insulin. In addition to diabetes, high weight gain in pregnancy, maternal obesity, and increasing age and parity are all recognized macrosomic risk factors (6,7), with obesity contributing the greatest risk (8). Although patients with GDM are frequently obese and excessive weight gain is discouraged (9), frank calorie restriction remains controversial and is seldom advocated (10). This study examined birth weights of infants of GDM women treated with calorie restriction alone. Comparisons were made with the birth weights of nondiabetic mothers, including two control groups, one with a normal and one with an abnormal O'Sullivan screening test, who were otherwise matched with the GDM mothers for the known macrosomic risk factors.

RESEARCH DESIGN AND METHODS

From the Unit of Metabolic Medicine, Department of Obstetrics and Gynaecology, and Department of Nutrition and Dietetics, St. Mary's Hospital Medical School, Imperial College of Science, Technology and Medicine, London, United Kingdom. Address correspondence and reprint requests to Dr. Anne Dornhorst, Unit of Metabolic Medicine, St. Mary's Hospital Medical School, Imperial College of Science, Technology and Medicine, London W2 1 NY, UK.

DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991

GDM mothers. All 35 women with GDM who were treated with diet alone throughout pregnancy while attending the prenatal clinic at St. Mary's Hospital in London between February 1988 and March 1989 are included in this study. The 8 GDM women seen during the study who required insulin treatment during their pregnancy are not included and are excluded from the analysis. The GDM mothers attended the Joint Diabetic Antenatal Clinic fortnightly until delivery.

161

CALORIE RESTRICTION IN GESTATIONAL DIABETES

TABLE? Maternal booking details of the gestational diabetic women and control subjects

n Age (yr) Booking BMI (kg/m2) BMI (kg/m2) >27 4 White Nonwhite

Gestational diabetic women

General prenatal population

Screen negative (group A)

Screen positive (group B)

35 33.1 ± 5.4* 28.7 ± 5.0*

2337 28.9 ± 5.8 24.1 ± 4.1

35 31.7 ± 5.1 28.1 ± 4.9

35 30.5 ± 6.5 26.6 ± 4.5

24* 11*

472 1865

21 14

14 21

20%* 66% 14%* 3 (9%)* 35(91%)*

48.8% 46.5% 4.7% 1352(57.8%) 995 (42.2%)

29% 57% 14% 3 (9%) 35(91%)

34% 57% 9% 3 (9%) 35(91%)

Data are means ± SD. BMI, body mass index. Maternal booking details of 35 gestational diabetic women, 2337 nondiabetic women, and two matched control groups: group A, with normal O'Sullivan screening value; and group B, with abnormal screening value but normal glucose-tolerance test. *P < 0.001 vs. general prenatal population.

All 35 GDM subjects were prescribed a calorie-restricted diet, with complex carbohydrates replacing refined carbohydrates and fats. Diets varied between 1200 and 1800 kcal (20-30 kcal/kg) and were calculated individually with the known or estimated prepregnancy weight, ideal body weight, and present calorie intake. The total daily calorie intake prescribed was calculated to be 30% less than that consumed before pregnancy. The composition of the prescribed diet was 50% carbohydrate, 15% protein, and 35% fat, with most carbohydrate taken as unrefined carbohydrate; protein as fish, white meat, beans, and dairy products; and fat as polyunsaturated fat. The diets were approved by the local health authority ethical committee. Weight gain after 28 wk gestation was discouraged. Blood glucose and HbA1c were randomly measured at each clinic visit, and a diurnal glucose profile was obtained if blood glucose exceeded 9 mM or HbA1c exceeded 9% (normal range < 8%). The criteria for remaining on dietary treatment alone was a diurnal glucose profile averaged over six measurements of 4, 14% vs. 4.7%); fewer were primagravidas (20% vs. 48.8%); and a greater percentage were nonwhite (91 vs. 42.2%) (P < 0.001 for all comparisons; Table 1). The GDM mothers and their control groups were completely matched for racial origin, and there was no significant difference in age or parity; however, BMI remained higher in the GDM group (28.7 ± 5.0) than control groups A (28.1 ± 4.9) and B (26.6 ± 4.5) (Wilcoxon's signed-rank test, P < 0.001; Table 1). Obesity was common among the women with GDM, with 24 subjects having a BMI >27 kg/m2. Effect of calorie restriction. The individual mean blood glucose concentration for the GDM mother ranged from 4.7 to 8.5 mM (mean 6.6 ± 0.9 mM). The individual mean HbA1c ranged from 4.8 to 8.7% (mean 6.5 ± 0.4%).

DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991

A. DORNHORST AND ASSOCIATES

TABLE 2 Maternal weight gain from booking and 28 wk gestation to delivery

n Booking gestation (wk) Weight gain (kg) Booking to delivery 28 wk to delivery

Gestational diabetic women

General prenatal population

Screen negative (group A)

Screen positive (group B)

35 16.0 ± 4 . 2

2337 16.7 ± 7.4

35 16.0 ± 6.8

35 17.5 ± 6.5

9.3 ± 5.3

9.7 ± 5.3 4.1 ± 3.1

9.7 ± 5.4 4.6 ± 2.9

4.6 ± 4.9* 1.7 ± 1.6*

Data are means ± SD. Maternal weight gain from booking and 28 wk gestation to delivery of 35 gestational diabetic women treated with calorie restriction and control subjects, 2337 nondiabetic women and two matched control groups: group A, with normal O'Sullivan screening value, and group B, with abnormal screening value but normal glucose-tolerance test. *P < 0.005 vs. general prenatal population and control subjects in groups A and B.

Weight gain from booking to delivery was significantly less for the 35 GDM mothers (4.6 ± 4.9 kg) than for the general prenatal population (9.3 ± 5.3 kg) or subjects in control groups A (9.7 ± 5.3 kg) and B (9.7 ± 5.4 kg) (P < 0.0005), despite no difference in booking gestation (Table 2). Weight gain from 28 wk to delivery was significantly less for the GDM mothers (1.7 ± 1.6 kg) than for control groups A (4.1 ± 3.1 kg) and B (4.6 ± 2.9 kg) (P < 0.0001; Table 2). Birth weight. Birth weight, birth weight percentile, and gestational age for infants of the GDM mothers, the general prenatal population, and control groups A and B were similar. No baby of a GDM mother had a birth weight below the 10th percentile when gestational age was considered. The GDM group had a similar incidence of macrosomia (birth weight > 4000 g) (6%, 2 of 35) as the general prenatal population (7.5%, 178 of 2337) and the group A control subjects (9%, 3 of 35), whereas the group B control mothers had a significantly higher frequency (26%, 9 of 35) (x2 = 3.88, P < 0.05; Table 3). The increased rate of macrosomia in the group B mothers could not be explained by obesity alone, because five of the nine mothers with macrosomic infants were not obese (BMI < 27 kg/m2), and the group B control subjects had a significantly lower BMI than the women with gestational diabetes (26.6 ± 4.5 vs. 28.7 ± 5.0 kg/m2) (Wilcoxon's signed-rank test, P < 0.001). Admission to Special Care Baby Unit. Two babies of GDM mothers were admitted to the Special Care Baby Unit. One baby was premature (34 of 40 wk) and required admission for >48 h, and the other required admission for 12 h for routine observation after a cesarean section. Two babies from group A and two from group B control subjects were admitted to the Unit: one baby from each group for routine

observation after a cesarean section, one with respiratory distress (group A), and one who was premature (group B).

DISCUSSION

Our study confirms those of O'Sullivan et al. (11) and Maresh et al. (8), which found that the woman with GDM is older, heavier, and of higher parity than the nondiabetic woman attending the general prenatal clinic. It also confirms that GDM is more frequent in nonwhites than whites (13), probably because of a combination of genetic susceptibility and diet. The diagnostic criteria used for the diagnosis of GDM predict both neonatal hypoglycemia and development of impaired glucose tolerance or non-insulin-dependent diabetes mellitus in later life (14). There is general agreement that diet is the first-line treatment for GDM, although precise recommendations for energy intake and composition have not yet been established (9). The calorie intake in this study was considerably less than the 2000 kcal • day" 1 minimum recommended by the American and British authorities on diabetes for pregnancy (9). Dietary compliance with calorie restriction in our patients is shown by the limited weight gain in the GDM mothers compared with their control subjects. This finding emphasizes that women during pregnancy are highly motivated and that changes in dietary eating patterns are obtainable. There is a good case for using a calorierestricted diet for the treatment of GDM women, because it will minimize obesity after the pregnancy and thereby lessen the risk of future diabetes (14,15). Concerns about calorie restriction in pregnancy arise from in vitro animal studies that showed ketosis in the rat to be

TABLE 3 Birth weight and percentile weight of infants of gestational diabetic women and control subjects Gestational diabetic women

General prenatal population

Screen negative (group A)

Screen positive (group B)

35 38.5 ± 1.4 3285 ± 520 2 (6%) 0

2337 38.9 ± 2.4 3228 ± 604 178(7.5%) 258(11%)

35 38.9 ± 1.7 3347 ± 492 3 (9%) 2 (6%)

35 39.1 ± 1.8 3448 ± 630 8 (23%)* 2 (6%)

n Gestational age (at delivery) Birth weight (g) Birth weight >4000 g Percentile weight

Calorie restriction for treatment of gestational diabetes.

Birth weights of infants of 35 gestational diabetic mothers treated with calorie restriction alone (1200-1800 kcal) were compared with those of infant...
529KB Sizes 0 Downloads 0 Views