Metabolic Effects of Hypocaloric Diets in Management of Gestational Diabetes ROBERT H. KNOPP, M. SCOTT MAGEE, VIDMANTAS RAISYS, AND THOMAS BENEDETTI
Although hypocatoric diets have been advocated for the management of the obese gravida and the obese mother with gestational diabetes, there is no general agreement on how severely calories should be restricted or on how this therapeutic approach compares with insulin therapy. The lack of consensus is in part because of the lack of studies comparing insulin management with the effects of different degrees of hypocaloric feeding and its effects on metabolism and glycemic status. We review the effects of 50 and 33% calorie restriction on glycemic status and intermediary fuel status in obese gestational diabetic subjects and compare the results with the administration of 20 U NPH and 10 U regular insulin every morning, a therapy of proven value in reducing macrosomia in gestational diabetes. When the two calorie-restriction regimens were compared after a 9-h overnight fast, glycemic status improved 10-20% on both. Ketonuria increased about twofold with 50% calorie restriction, but on average no increase in ketonuria was seen on the 33% calorie-restriction regimen. Both calorie-restriction programs led to a reduction in levels of plasma triglyceride, a correlate of infant birth weight. In contrast, the insulin regimen diminished ketonuria, but glycemic status improved little, and plasma triglyceride concentrations did not decline. Although more studies are needed to confirm these trends, the beneficial effect of 33% calorie restriction, which occurred without marked ketonuria, is consistent with previous studies in gestational diabetes. In addition, the simultaneous improvements observed in plasma glucose and triglyceride concentrations suggest that moderate calorie restriction may be valuable in preventing macrosomia in the offspring of the obese subject with gestational diabetes. A clinical trial to test this hypothesis is warranted. Diabetes 40 (Suppl. 2):165-71, 1991
From the Northwest Lipid Research Clinic and Departments of Medicine, Obstetrics and Gynecology, and Laboratory Medicine, University of Washington School of Medicine, Seattle, Washington. Address correspondence and reprint requests to Dr. Robert H. Knopp, Northwest Lipid Research Clinic, 326 Ninth Avenue, Seattle, WA 98104.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
S
everal decades ago, calorie restriction during pregnancy was a standard treatment to hedge against the development of preeclampsia and eclampsia (1). This treatment was discarded when the National Academy of Sciences recommended ~30 Ib (13.6 kg) of weight gain during normal pregnancy instead of the 15-20 Ib (6.8-9.1 kg) previously allowed (1). This mandate left unresolved whether calorie restrictions are appropriate in the dietary management of the obese pregnant woman and particularly the obese pregnant woman with gestational diabetes. The experience of some research groups indicated the potential benefit of hypocaloric feeding without adverse effect on infant outcome (2-4). In fact, one of the earliest experts on diabetes in pregnancy, Jorgen Pedersen, advocated calorie restriction for obese pregnant women from the earliest days of his experience, beginning in the 1950s (5). Finally, obese women who become pregnant often gain little weight, indicating a spontaneous net loss in maternal tissues when the weight of the conceptus (uterus, fetus, and attendant membranes and fluid) is subtracted (6). The argument that starvation in pregnancy might injure the intellectual function of the offspring arose from the studies of Berendes and Churchill, who associated ketonuria at delivery with impaired intellectual function in the progeny (7). However, after a more careful analysis, this association was not consistent and may have been caused by chorioamnionitis in individuals with ketonuria at delivery (8). Stehbens et al. (9) also associated ketonuria in diabetic pregnancy with lower IQ scores in the progeny, but this finding also was not confirmed by later research (10,11). Finally, Dutch army recruits who were born during the Dutch famine had no impairment in intellectual function compared with recruits not born then (12). Thus, the physiological ketonemia and ketonuria associated with calorie restriction in human pregnancy has never been consistently associated with impaired fetal outcome in epidemiological surveys
165
HYPOCALORIC DIETS
8
Day of Study
9
10
11
12
13
14
FIG. 1. Design of study to investigate effect of 50% caloric restriction in pregnancy. Subjects were given equivalent full-calorie diet of -2400 kcal daily for 1st wk on metabolic ward. Twenty-four-hour glucose profile was performed on day 6 and glucose-tolerance test on day 7 in both groups. In 2nd wk, control group was continued on ~ 2400-kcal diet, and calorie-restricted group received -1200 kcal daily. Daily glucose and insulin monitoring was performed, and 24-h glucose profile and glucose-tolerance tests were repeated on day 6 and 7, respectively. © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
(for review, see ref. 13). Indeed, ketonuria is not uncommon in normal pregnancy (14-16). Despite these considerations, a general recommendation favoring the use of hypocaloric diets in the management of the obese woman with gestational diabetes has been impeded by a lack of information on the effects of specific amounts of calorie restriction on plasma fuel metabolism and glycemic control under carefully controlled study conditions. To address this deficiency, we performed two short-term metabolic-ward studies of restricted-calorie feeding in obese subjects with gestational diabetes to determine the effects on plasma glucose control, plasma fatty acid mobilization, and plasma ketone body levels and excretion. In addition, we performed limited studies on hepatic glucose production and insulin sensitivity in these subjects. Results of these studies were recently published (13,17). Finally, we looked at whether calorie restriction benefits plasma fuel abnormalities in obese subjects with gestational diabetes more than does insulin therapy. All studies were approved by the University of Washington Human Subjects Review Committee, and written consent of each subject was obtained. EXPERIMENT 1: RESEARCH DESIGN AND METHODS AND RESULTS Effect of 50% calorie restriction. Subjects with gestational diabetes diagnosed by the criteria of Carpenter and Cous-
tan (18) who had weighed > 120% of their ideal body weight prepregnancy were admitted to a metabolic ward and fed a calorie-sufficient (-2400 kcal) diet daily for 1 wk. Daily fasting glucose and insulin concentrations were obtained; on day 6, 24 measurements of plasma glucose were made over 24 h; and on day 7, a3-h, 100-g oral glucose tolerance test was performed. Subjects were then randomly assigned either to continue their 2400-kcal diet or to restrict dietary intake to -1200 kcal/day. The observations made in wk 1 were then repeated in wk 2 (Fig. 1). The diet consisted of 50% carbohydrate, 30% fat, and 20% protein and contained 11 g fiber. Subjects received 25% of calories at each of the major meals and 12.5% of calories at both 1500 and 2200. Blood samples were obtained the next morning, 9 h after the last snack. Characteristics of the subjects are shown in Table 1. Control subjects receiving a full-calorie diet during both weeks tended to be older and somewhat lighter, but the two groups were similar in the numbers of previous pregnancies and the weeks of gestation in which they were studied. The actual mean calorie intake in the control subjects was 2307 kcal during both wk 1 and 2, whereas the restricted subjects received a mean of 2476 kcal during the 1 st wk and a mean of 1238 kcal in the second, with the calorie intake calculated with the Harris-Benedict formula (17). Effects on plasma fuels and insulin are shown in Table 2. Plasma glucose concentrations changed very little in the control group, declining 4% between wk 1 and 2. In addition, the 24-h mean of 24 observations of plasma glucose for wk 1 and 2 (Fig. 2) could be superimposed. Similarly, in the control group, plasma immunoreactive insulin (IRI) concentrations, free fatty-acid (FFA) concentrations, and (3-hydroxybutyric acid concentrations were not different in the 2nd wk, and no ketonuria was observed in this group of women after the 9-h overnight fast in either week. In contrast, 50% calorie-restricted pregnant subjects experienced a 17% reduction in mean daily fasting glucose concentrations and a 21 % fall in the 24-h mean glucose level. The 24-h glucose profile showed that both fasting and postprandial glucose concentrations were lower during the calorie-restricted week, with the greatest reductions in the postprandial time points (Fig. 2). Plasma IRI concentrations were even more strikingly reduced, decreasing 45% from base line. Increased fatty acid mobilization was apparent in the 2nd wk, with an increase in plasma FFA concentrations of 64% compared with wk 1, and (3-hydroxybutyric acid concentrations increased from 0.29 to 0.78 mM, or 2.7-fold. Urinary ketone excretion semiquantitated with Ketostix (Ames, Miles, Elkhart, IN) increased 2.1-fold. Because the restricted subjects had a higher base-line glucose and were also
TABLE 1 Characteristics of subjects studied
Control Restricted diet
n
Age (yr)
Weight (kg)
Prepregnancy ideal body weight (%)
Prior pregnancies (n)
Gestation (wk)
5 7
36 30
88 96
147 171
7 5
30 31
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. Based on Magee et al. (17).
166
DIABETES, VOL 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES
TABLE 2 Effects of 50% calorie restriction on plasma fuels and insulin in obese women with gestational diabetes Control
Glucose (mM) Fasting 24-h mean Immunoreactive insulin (pM) Free fatty acids (mg/l) 3-Hydroxybutyric acid (mM) Urine ketones (+) Triglyceride (mM)
Restricted
Week
Week
Week
Change
2
Change (%)
Week
1
1
2
(%)
5.4 6.7 165 330 220 0 3.2
5.2 6.7 165 390 210 0 3.7
-4 0 0 18 -5 0 16
5.9 6.8 265 220 290 1.0 3.1
4.9 5.4 145 360 780 2.1 2.9
-17 -21 -45 64 169 210 -7
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. Based on Magee et al. (17).
heavier than the control subjects, they had a greater tendency to be ketonuric even in the 1st wk of full-calorie feeding (Table 2). Despite the heightened fatty-acid mobilization and ketogenesis, plasma triglyceride concentrations tended to decline in the calorie-restricted group (Table 2). Glucose tolerance did not change in the calorie-restricted group or the control group (Fig. 3). This surprising result occurred even though plasma insulin concentrations decreased markedly in the calorie-restricted group (Table 2). This observation suggests that the improvement in glycemic status is not related to an improvement in the rate of glucose
disposal. In other words, although insulin sensitivity, as reflected in the reduction in fasting plasma insulin concentrations, might be improved, the reduction in insulin concentration appears to nullify any gain in glucose-disposal efficiency in the new steady state. Because of the effect of this degree of calorie restriction on ketonemia and uncertainties about the possible effect on fetal growth and development, a specific dietary recommendation based on these
o u
D
o 03
15 OL
8
10 12 14 Time (hrs)
18 20 22 24
FIG. 2. Twenty-four-hour glucose profile of control subjects (fop) and calorie-restricted subjects (bottom). 0,1st wk of calorie-sufficient feeding for both groups; • , calorie restriction of 2nd wk in calorie-restricted group (bottom). © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
o
60 Time
120
180
(mm)
FIG. 3. Glucose-tolerance tests done at end of 1st (O) and 2nd (•) wk on control (fop) and calorie-restricted (bottom) subjects. Legend is the same as in Fig. 2. No change in glucose tolerance is seen in either group. © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
167
HYPOCALORIC DIETS
TABLE 3 Subject characteristic of group with 33% calorie restriction in pregnancy
Normal Obese Obese with gestational diabetes Insulin Decreased calories
n
Age
Gravida
Prepregnancy ideal body weight (%)
9 7 6 3 3
24.1 25.7 28.8* 26.3 31.3
2 2 3.2 3 3.3
100 128 132t 124 139
Glucose 1 h after 50 g (mM)
Diet (kcal)
Week studied
5.8 6.6 9.1t 8.9 9.4
2352 2374 2415 2504 1655
31.4 32 31.8 33.3 30.3
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. *P < 0.025, tP < 0.005, for trend vs. normal and obese.
data was not possible. Nonetheless, this study demonstrated the marked extent to which calorie restriction can improve the hyperglycemia of gestational diabetes.
subjects are shown in Table 4. The highest fasting plasma glucose concentrations were seen in the obese gestational diabetic group on a full-calorie diet studied in the 1st wk of the metabolic-ward observation. Fasting plasma glucose was unchanged in three subjects with gestational diabetes treated with insulin therapy in the 2nd wk, and 24-h mean glucose was slightly improved, as was the 24-h glucose profile (Fig. 4). However, obese subjects with gestational diabetes receiving 33% calorie-restricted diets had fasting plasma glucose concentrations below those of the normal subjects and a 24-h mean glucose that was also slightly less than that of the normal subjects and lower than those of the obese or obese gestational diabetic groups. The 24-h glucose profile was also lower than in the women treated with insulin (Fig. 4). Plasma insulin concentrations were higher in the obese gestational diabetic group than in the obese and normal nondiabetic pregnant women. Insulin therapy resulted in a slight increase in plasma IRI, as expected, but a lower insulin concentration was seen in the calorie-restricted group. Mean plasma FFA concentrations were elevated, as were plasma triglyceride concentrations in the obese women with gestational diabetes before therapy and remained so in the gestational diabetic subjects treated with insulin in the 2nd wk. However, FFA and triglyceride concentrations were lower when gestational diabetic women were treated with calorie restriction. These trends were confirmed when individual pre- and posttreatment values were compared (13). Urinary ketone excretion was highest in the obese gestational diabetic group on a full-calorie diet and lower with insulin therapy as well as with calorie restriction. Again, individual pretreatment-posttreatment comparisons confirmed group differences (data not
EXPERIMENT 2: RESEARCH DESIGN AND METHODS AND RESULTS Effect of 3 3 % calorie restriction. The effect of a 3 3 %
calorie-restricted diet was next investigated as a closer approximation to a diet that might be not only tolerable but also beneficial and less likely to promote excessive fat mobilization and ketogenesis (17,19). Six obese subjects with gestational diabetes were studied as shown in Table 3. These subjects were fed a full-calorie diet in the 1st wk from the metabolic ward and were compared with nine nonobese nondiabetic pregnant women and seven obese nondiabetic pregnant women also fed a full-calorie diet. In the 2nd wk, the women with gestational diabetes were randomly assigned either to an insulin-therapy regimen consisting of 20 U of NPH insulin and 10 U of regular insulin, as proposed by Coustan and Imarah (20) or to a 33% calorie-restricted diet. Characteristics of these subjects are shown in Table 3. The mean ages of the different groups ranged from 24 to 31 yr. Prior gravidity ranged between 2 and 3.3, and the obese subjects with gestational diabetes weighed between 124 and 139% of ideal body weight (somewhat less overweight than in the previous study). Plasma glucose concentrations 1 h after a 50-g oral glucose load were clearly elevated in the obese gestationally diabetic group compared with normal subjects, but not significantly so in the obese nondiabetic group compared with the normal group. All subjects were studied between 30 and 33 wk gestation. Mean basal metabolic fuels and hormones of the study
TABLE 4 Basal metabolic fuels and hormones Plasma
Normal Obese Obese with gestational diabetes Insulin Reduced calories
Fasting glucose (mM)
Mean 24-h glucose (mM)
Immunoreactive insulin (pM)
Free fatty acids (mg/l)
3.9 4.0 4.2 4.2 3.7
4.7 4.8 5.1 4.9 4.6
127 164 199* 225 138
100 116 140t 157 83
All values obtained after a 9 h overnight fast. From Knopp et al. (13). *P < 0.01, \P < 0.025, tP < 0.1, for trend vs. normal and obese.
168
Triglyceride (mM)
Urine ketone
(+) 1.7 1.9 2.3+ 2.6 1.5
0.7 0.9 3.Of 1.3 1.7
1991 John Wiley & Sons, Inc.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES mg/dL
mM
TABLE 6 Comparative effects of insulin and 2 levels of calorie restriction on metabolism in gestational diabetes
• — • GDM full calorie (6) 0---0 GDM 2/3 calorie (3) A
A GDM insulin
(3)
Calorierestricted diet
Fasting glucose 24-h glucose Fasting immunoreactive insulin Free fatty acids Ketonuria Triglyceride
50%
33%
Insulin
-17 -21 -45 64 210 -6
-22 -10 -31 -47 -43 -35
0 -4 28 12 -57 13
Results are % change. From Knopp et al. (13). © 1991 John Wiley & Sons, Inc.
7
8
9
10
11
12
differences in the 1st observational wk. Neither insulin therapy nor calorie restriction altered this sensitivity, consistent with the absence of improvement in glucose tolerance in the 50% calorie-restricted study. The amounts of glucose needed to maintain plasma glucose concentrations during insulin infusion were not significantly different, but tended to be lower in the obese women compared with pregnant women of normal weight. This measurement was only slightly lower in gestational diabetic subjects compared with obese nondiabetic subjects. Finally, calculations of insulin sensitivity also showed only slightly lower values in the gestational diabetic subjects, again not statistically different. Thus, our results do not agree with those of Ryan et al., who found increased insulin resistance in gestational diabetic pregnancy (22). Our results are more in keeping with the results of Buchanan et al. (23), who found no increase in insulin resistance in gestational diabetic pregnancy with the intravenous glucose-tolerance test and the Bergman minimal-model technique. Some of the differences among the studies may reflect differences in technique. With insulin or calorie restriction as a form of therapy, the insulin-clamp technique used in our gestational diabetic subjects did not disclose major changes in insulin action or hepatic glucose production, reinforcing the idea that the predominant mechanism of glycemic improvement with 33% calorie restriction is the reduction in intake of calories. Comparison of effects of 50 and 33% calorie restrictions and insulin. A comparison of the effects of the two degrees of calorie restriction is shown in Table 6. Fasting plasma glucose concentrations were reduced by 17% in the 50% calorie-restriction program and 22% with 33% calorie-
1
FIG. 4. Twenty-four-hour glucose profiles in gestational diabetic subjects treated with insulin or with 33% calorie restriction.
shown; see ref. 13). Although the numbers of gestational diabetic subjects treated with insulin or diet are small and do not allow for secure conclusions about the differences between insulin and calorie restriction, 33% calorie restriction did not aggravate fat mobilization and ketogenesis in these subjects. To determine the mechanism of the effect of calorie restriction on glycemic status in gestational diabetic subjects, [6,6-2H2]glucose (dideuterated) was used to measure hepatic glucose output under basal conditions and during an infusion of insulin designed to raise the plasma insulin concentration ~16 |xll/ml (for methodological details, see ref. 13). Observations are shown in Table 5. Basal hepatic glucose output (HGO) was indistinguishable among normal, obese, and obese gestational diabetic pregnant subjects before initiation of insulin or calorie restriction, confirming the observations of Kalhan et al. (21). HGO tended to be higher in the two insulin-treated subjects during the 2nd wk of observation. In contrast, the obese subjects with gestational diabetes treated with calorie restriction had a slightly lower basal hepatic glucose output. However, these differences are small, are not statistically significant, and are unlikely to be so even if more subjects were studied. The calculated amount insulin must be increased above basal to reduce HGO by 50% (ED50) also showed only small, nonsignificant TABLE 5 Measurements of insulin sensitivity and hepatic glucose production
n Normal Obese Obese with gestational diabetes Insulin Decreased calories
5-8 6-7 4-5 2 3
Basal hepatic glucose output (mg • k g " 1 • min" 1 ) 2.99 2.88 2.96 3.21 2.800
± ± ± ± ±
0.45 0.45 0.50 0.10 0.063
ED 50 (%) 136 115 147 140 157
± ± ± ± ±
33 19 41 6 22
Ml 12.7 10.6 9.4 9.3 9.3
ED50, % of basal insulin level at which hepatic glucose production is calculated to be reduced 50%. (mg • k g " 1 • min" 1 )/mean insulin level (|xll/ml) x 100. S,, insulin sensitivity index.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
± ± ± ± ±
s, 3.8 3.4 1.5 2.9 1.1
2.5 3.0 2.1 2.1 1.5
± ± ± ± ±
1.3 1.7 1.1 0.9 0.2
glucose infusion rate
169
HYPOCALORIC DIETS
these studies suggests that calorie restriction may be a more effective means than insulin to limit the global excess Normal: in plasma fuels seen in the women with gestational diabetes (26). A model of the increased diversion of circulating Glucose maternal fuels to the fetal circulation in the obese subject Amino acids with gestational diabetes is illustrated in Fig. 5. Because of FFA lipoprotein lipase in the placenta (27,28), plasma triglycerTriglyceride ide fatty acids potentially have access to the fetal circulation. Increases in transport of this substrate along with the increases in plasma amino acids and glucose concentrations Gestational (26) should, in concert, enhance fetal growth and macrosoDiabetic: mia. This model is in keeping with the expanded Pedersen Obesity hypothesis proposed by Freinkel and Metzger in 1979 (29). Weight gain The importance of the calorie-restriction program described Insulin here is that it might decrease the concentration of all resistance maternal fuels accessing the fetal circulation. Because O'Sullivan found that even modest doses of insulin given to gestational diabetic mothers (as little as 10 U daily) can reduce infant macrosomia (24), insulin therapy FIG. 5. Model of excess glucose and triglyceride transport to fetus in remains a valuable tool in the management of gestational obese gestational diabetic. Dark lines in gestational diabetic subject (bottom) denote increased transfer of nutrients to fetal circulation. diabetic women, particularly those with more marked gluNote important role of placental lipoprotein lipase in mediating cose intolerance or elevations in fasting glucose concentratriglyceride fatty-acid transport to fetus. tions. On the other hand, one may not wish to use insulin if restriction program, whereas insulin produced no improve- avoidable because it can provoke insulin allergy or serious ment. The 24-h mean glucose concentrations were reduced hypoglycemia or aggravate infant macrosomia by the transby 2 1 % in the 50% calorie-restricted subjects and 10% with placental transport of immunoglobulin- bound maternal or the 33% calorie-restricted subjects and only 4% with insulin exogenously administered insulin (30). For these important therapy. Thus, even with the 33% calorie restriction, glyce- reasons and to differentiate more clearly the metabolic mic status is more improved than with a standard insulin- effects of insulin therapy from those of calorie-restriction therapy regimen (18), a regimen in excess of that previously therapy, it is necessary to perform more such carefully reported to reduce macrosomia in gestational diabetes (24). monitored metabolic-ward studies in more subjects and to In fact, the 33% calorie restriction was sufficient to restore initiate randomized clinical trials in outpatient gestational the gestational diabetic subjects not only to the normal diabetic subjects to judge the effect of the two treatment range but to a level equal to or better than that in normal approaches on perinatal morbidity and future well-being of subjects (Table 4). The extent of plasma IRI reduction with mother and offspring. Only in this deliberate and collabora33% calorie restriction was - 6 6 % of that in the 50% re- tive way can we resolve the persisting questions about how stricted group, consistent with the relative amounts of calorie best to treat gestational diabetes. restriction imposed. The amount of FFA mobilization was Note added in proof.The recent report of the association of elevated in the 50% calorie restriction and reduced in the third trimester ketonemia with reduced intelligence sores at 33%. Similar effects were seen for ketonuria. Finally, reduc- 3, 4, and 5 yr in the offspring (31) underscores the need for tion in plasma triglycerides was seen in both groups of further studies of dietary approaches to the management of subjects, but a greater triglyceride reduction was suggested the gestational diabetic subject. in the 33% group. Reducing plasma triglyceride concentrations is important because plasma triglyceride concentrations are elevated in gestational diabetic subjects and the ACKNOWLEDGMENTS triglyceride elevation is more strongly associated with infant This work was supported by Grant DK-35816 (Clinical Numacrosomia than is fasting or postprandial plasma glucose trition Research Unit), Clinical Research Center Grant MO1RR-0037, and Diabetes and Endocrinology Research concentrations (25). Center Grant DK-17047, all from the National Institutes of Overall, the 33% calorie-restricted regimen, which is an Health. This study was also supported by Training Grant -1600- to 1800-kcal • day" 1 diet, is sufficient to normalize DK-07247 in endocrinology and metabolism (M.S.M.), a glycemic status without apparent adverse metabolic effect. research fellowship from the American Diabetes Association This diet is similar to those used by previous workers (2-5), Washington Affiliate, and a generous grant from the WashDomhorst et al. (this issue, p. 161) in the management of ington State Eagles. obese subjects with gestational diabetes or obese pregnant We express our gratitude to Dr. Zane Brown for assissubjects without adverse effect on fetal outcome. More tance in identifying patients and to Al Greeves for typing the research is needed to extend and confirm these limited manuscript. observations.
Mother
Placenta
Fetus
DISCUSSION
The greater glycemic improvement associated with two levels of calorie restriction compared with insulin therapy in
170
REFERENCES 1. Committee on Maternal Nutrition/Food and Nutrition Board, National Research Council: Maternal Nutrition and the Course of Pregnancy. Washington, DC, Natl. Acad. Sci., 1970, p. 1-124
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES 2. Borberg C, Gillmer MDG, Brunner EJ, Gunn PJ, Oakley NW, Beard RW: Obesity in pregnancy: the effect of dietary advice. Diabetes Care 3:476-81,1980 3. Algert S, Shragg P, Hollingworth DR: Moderate caloric restriction in obese women with gestational diabetes. Obstet Gynecol 65:487-91, 1985 4. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:103957,1979 5. Pedersen J: The Pregnant Diabetic and Her Newborn. 2nd ed. Baltimore, MD, Williams & Wilkins, 1977, p. 223 6. Naeye RL: Weight gain and the outcome of pregnancy. Am J Obstet Gynecol 135:3-9, 1979 7. Churchill JA, Berendes HW: Intelligence of children whose mothers have acetonuria in pregnancy. In Perinatal Factors Affecting Human Development. Washington, DC, Pan Am. Health Org., 1969 (Sci. publ. no. 185) 8. Naeye RL, Chez RA: Effects of maternal acetonuria and low pregnancy weight gain on children's psychomotor development. Am J Obstet Gynecol 139:189-93, 1981 9. Stehbens JA, Baker Gl, Kitchell M: Outcome at ages 1, 3, and 5 years of children born to diabetic mothers. Am J Obstet Gynecol 127:408-13, 1977 10. Naeye RL: The outcome of diabetic pregnancies: a prospective study. In Pregnancy Metabolism, Diabetes and the Fetus. Amsterdam, Excerpta Med., 1979, p. 227-41 (Ciba Found. Symp. no. 63) 11. Persson B, Gentz J: Follow-up of children of insulin-dependent and gestational diabetic mothers: neuropsychological outcome. Acta Paediatr Scand 73:349-58, 1984 12. Stein Z, Susser M, Saenger G, Marolla F: Famine and Human Development. New York, Oxford Univ. Press, 1975, p. 1-284 13. Knopp RH, Magee MS, Raisys V, Benedetti T, Bonet B: Hypocaloric diets and ketogenesis in the management of obese gestational diabetic women. J Am Coll Nutr 10:649-67, 1991 14. Paterson P, Sheath J, Taft P, Wood C: Maternal and fetal ketone concentrations in plasma and urine. LanceM:862-65, 1967 15. Coetzee EJ, Jackson WPU, Berman PA: Ketonuria in pregnancy—with special reference to calorie-restricted food intake in obese diabetics. Diabetes 29:177-81, 1980 16. Chez RA, Curcio FD III: Ketonuria in normal pregnancy. Obstet Gynecol 69:272-74, 1987 17. Magee MS, Knopp RH, Benedetti TJ: Metabolic effects of a 1200-kcal diet in obese pregnant women with gestational diabetes. Diabetes 39:234-40, 1990
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
18. Carpenter MW, Coustan DR: Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol 144:768-73, 1982 19. Magee MS, Knopp R, Raisys V, Prince E, Brown Z, Benedetti T: Pathophysiology and response to insulin or dietary treatment in gestational diabetes (Abstract). Diabetes 37 (Suppl. 1):111A, 1988 20. Coustan DR, Imarah J: Prophylactic insulin treatment of gestational diabetes reduces the incidence of macrosomia, operative delivery and birth trauma. Am J Obstet Gynecol 150:836-43, 1984 21. Kalhan SC, D'Angelo LJ, Savin SM, Adam PAJ: Glucose production in pregnant women at term gestation: sources of glucose for human fetus. J Clin Invest 63:388-94, 1979 22. Ryan EA, O'Sullivan MJ, Skyler JS: Insulin action during pregnancy: studies with the euglycemic clamp technique. Diabetes 34:380-89,1985 23. Buchanan TA, Metzger BE, Freinkel N, Bergman RN: Insulin sensitivity and B-cell responsiveness to glucose during late pregnancy in lean and moderately obese women with normal glucose tolerance or mild gestational diabetes. Am J Obstet Gynecol 162:1008-14, 1990 24. O'Sullivan JB: Insulin treatment for gestational diabetes. In Early Diabetes in Early Life. Camerini-Davalos RA, Cole HS, Eds. New York, Academic, 1975, p. 447-54. 25. Knopp RH, Magee MS, Larson MP, Benedetti T: Alternative screening tests and birth weight associations in pregnant women with abnormal glucose screening (Abstract). Diabetes 37 (Suppl. 1):110A, 1988 26. Metzger BE, Phelps RL, Freinkel N, Navickas IA: Effects of gestational diabetes on diurnal profiles of plasma glucose, lipids, and individual amino acids. Diabetes Care 3:402-409, 1980 27. Knopp RH, Warth MR, Charles D, Childs M, Li JR, Mabuchi H, Van Allen M: Lipoprotein metabolism in pregnancy, fat transport to the fetus, and the effects of diabetes. Biol Neonate (Paris) 50:297-317, 1986 28. Bonet B, Gown AM, Brunzell JD, Knopp RH: Role of lipoprotein lipase in triglyceride metabolism by placental trophoblast and macrophages (Abstract). Arteriosclerosis 7:642A, 1987 29. Freinkel N, Metzger BE: Pregnancy as a tissue culture experience: the critical implications of maternal metabolism for fetal development. In Pregnancy Metabolism, Diabetes and the Fetus. Amsterdam, Excerpta Med., 1979, p. 3-23 (Ciba Found. Symp. no. 63) 30. Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F, Khoury JC: Transplacental passage of insulin in pregnant women with insulindependent diabetes mellitus. N EnglJ Med 323:309-14, 1990 31. Rizzo T, Metzger BE, Burns WJ, Burns K: Correlations between antepartum maternal metabolism and intelligence of offspring. N Engl J Med 325:911-16, 1991
171
Although hypocatoric diets have been advocated for the management of the obese gravida and the obese mother with gestational diabetes, there is no general agreement on how severely calories should be restricted or on how this therapeutic approach compares with insulin therapy. The lack of consensus is in part because of the lack of studies comparing insulin management with the effects of different degrees of hypocaloric feeding and its effects on metabolism and glycemic status. We review the effects of 50 and 33% calorie restriction on glycemic status and intermediary fuel status in obese gestational diabetic subjects and compare the results with the administration of 20 U NPH and 10 U regular insulin every morning, a therapy of proven value in reducing macrosomia in gestational diabetes. When the two calorie-restriction regimens were compared after a 9-h overnight fast, glycemic status improved 10-20% on both. Ketonuria increased about twofold with 50% calorie restriction, but on average no increase in ketonuria was seen on the 33% calorie-restriction regimen. Both calorie-restriction programs led to a reduction in levels of plasma triglyceride, a correlate of infant birth weight. In contrast, the insulin regimen diminished ketonuria, but glycemic status improved little, and plasma triglyceride concentrations did not decline. Although more studies are needed to confirm these trends, the beneficial effect of 33% calorie restriction, which occurred without marked ketonuria, is consistent with previous studies in gestational diabetes. In addition, the simultaneous improvements observed in plasma glucose and triglyceride concentrations suggest that moderate calorie restriction may be valuable in preventing macrosomia in the offspring of the obese subject with gestational diabetes. A clinical trial to test this hypothesis is warranted. Diabetes 40 (Suppl. 2):165-71, 1991
From the Northwest Lipid Research Clinic and Departments of Medicine, Obstetrics and Gynecology, and Laboratory Medicine, University of Washington School of Medicine, Seattle, Washington. Address correspondence and reprint requests to Dr. Robert H. Knopp, Northwest Lipid Research Clinic, 326 Ninth Avenue, Seattle, WA 98104.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
S
everal decades ago, calorie restriction during pregnancy was a standard treatment to hedge against the development of preeclampsia and eclampsia (1). This treatment was discarded when the National Academy of Sciences recommended ~30 Ib (13.6 kg) of weight gain during normal pregnancy instead of the 15-20 Ib (6.8-9.1 kg) previously allowed (1). This mandate left unresolved whether calorie restrictions are appropriate in the dietary management of the obese pregnant woman and particularly the obese pregnant woman with gestational diabetes. The experience of some research groups indicated the potential benefit of hypocaloric feeding without adverse effect on infant outcome (2-4). In fact, one of the earliest experts on diabetes in pregnancy, Jorgen Pedersen, advocated calorie restriction for obese pregnant women from the earliest days of his experience, beginning in the 1950s (5). Finally, obese women who become pregnant often gain little weight, indicating a spontaneous net loss in maternal tissues when the weight of the conceptus (uterus, fetus, and attendant membranes and fluid) is subtracted (6). The argument that starvation in pregnancy might injure the intellectual function of the offspring arose from the studies of Berendes and Churchill, who associated ketonuria at delivery with impaired intellectual function in the progeny (7). However, after a more careful analysis, this association was not consistent and may have been caused by chorioamnionitis in individuals with ketonuria at delivery (8). Stehbens et al. (9) also associated ketonuria in diabetic pregnancy with lower IQ scores in the progeny, but this finding also was not confirmed by later research (10,11). Finally, Dutch army recruits who were born during the Dutch famine had no impairment in intellectual function compared with recruits not born then (12). Thus, the physiological ketonemia and ketonuria associated with calorie restriction in human pregnancy has never been consistently associated with impaired fetal outcome in epidemiological surveys
165
HYPOCALORIC DIETS
8
Day of Study
9
10
11
12
13
14
FIG. 1. Design of study to investigate effect of 50% caloric restriction in pregnancy. Subjects were given equivalent full-calorie diet of -2400 kcal daily for 1st wk on metabolic ward. Twenty-four-hour glucose profile was performed on day 6 and glucose-tolerance test on day 7 in both groups. In 2nd wk, control group was continued on ~ 2400-kcal diet, and calorie-restricted group received -1200 kcal daily. Daily glucose and insulin monitoring was performed, and 24-h glucose profile and glucose-tolerance tests were repeated on day 6 and 7, respectively. © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
(for review, see ref. 13). Indeed, ketonuria is not uncommon in normal pregnancy (14-16). Despite these considerations, a general recommendation favoring the use of hypocaloric diets in the management of the obese woman with gestational diabetes has been impeded by a lack of information on the effects of specific amounts of calorie restriction on plasma fuel metabolism and glycemic control under carefully controlled study conditions. To address this deficiency, we performed two short-term metabolic-ward studies of restricted-calorie feeding in obese subjects with gestational diabetes to determine the effects on plasma glucose control, plasma fatty acid mobilization, and plasma ketone body levels and excretion. In addition, we performed limited studies on hepatic glucose production and insulin sensitivity in these subjects. Results of these studies were recently published (13,17). Finally, we looked at whether calorie restriction benefits plasma fuel abnormalities in obese subjects with gestational diabetes more than does insulin therapy. All studies were approved by the University of Washington Human Subjects Review Committee, and written consent of each subject was obtained. EXPERIMENT 1: RESEARCH DESIGN AND METHODS AND RESULTS Effect of 50% calorie restriction. Subjects with gestational diabetes diagnosed by the criteria of Carpenter and Cous-
tan (18) who had weighed > 120% of their ideal body weight prepregnancy were admitted to a metabolic ward and fed a calorie-sufficient (-2400 kcal) diet daily for 1 wk. Daily fasting glucose and insulin concentrations were obtained; on day 6, 24 measurements of plasma glucose were made over 24 h; and on day 7, a3-h, 100-g oral glucose tolerance test was performed. Subjects were then randomly assigned either to continue their 2400-kcal diet or to restrict dietary intake to -1200 kcal/day. The observations made in wk 1 were then repeated in wk 2 (Fig. 1). The diet consisted of 50% carbohydrate, 30% fat, and 20% protein and contained 11 g fiber. Subjects received 25% of calories at each of the major meals and 12.5% of calories at both 1500 and 2200. Blood samples were obtained the next morning, 9 h after the last snack. Characteristics of the subjects are shown in Table 1. Control subjects receiving a full-calorie diet during both weeks tended to be older and somewhat lighter, but the two groups were similar in the numbers of previous pregnancies and the weeks of gestation in which they were studied. The actual mean calorie intake in the control subjects was 2307 kcal during both wk 1 and 2, whereas the restricted subjects received a mean of 2476 kcal during the 1 st wk and a mean of 1238 kcal in the second, with the calorie intake calculated with the Harris-Benedict formula (17). Effects on plasma fuels and insulin are shown in Table 2. Plasma glucose concentrations changed very little in the control group, declining 4% between wk 1 and 2. In addition, the 24-h mean of 24 observations of plasma glucose for wk 1 and 2 (Fig. 2) could be superimposed. Similarly, in the control group, plasma immunoreactive insulin (IRI) concentrations, free fatty-acid (FFA) concentrations, and (3-hydroxybutyric acid concentrations were not different in the 2nd wk, and no ketonuria was observed in this group of women after the 9-h overnight fast in either week. In contrast, 50% calorie-restricted pregnant subjects experienced a 17% reduction in mean daily fasting glucose concentrations and a 21 % fall in the 24-h mean glucose level. The 24-h glucose profile showed that both fasting and postprandial glucose concentrations were lower during the calorie-restricted week, with the greatest reductions in the postprandial time points (Fig. 2). Plasma IRI concentrations were even more strikingly reduced, decreasing 45% from base line. Increased fatty acid mobilization was apparent in the 2nd wk, with an increase in plasma FFA concentrations of 64% compared with wk 1, and (3-hydroxybutyric acid concentrations increased from 0.29 to 0.78 mM, or 2.7-fold. Urinary ketone excretion semiquantitated with Ketostix (Ames, Miles, Elkhart, IN) increased 2.1-fold. Because the restricted subjects had a higher base-line glucose and were also
TABLE 1 Characteristics of subjects studied
Control Restricted diet
n
Age (yr)
Weight (kg)
Prepregnancy ideal body weight (%)
Prior pregnancies (n)
Gestation (wk)
5 7
36 30
88 96
147 171
7 5
30 31
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. Based on Magee et al. (17).
166
DIABETES, VOL 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES
TABLE 2 Effects of 50% calorie restriction on plasma fuels and insulin in obese women with gestational diabetes Control
Glucose (mM) Fasting 24-h mean Immunoreactive insulin (pM) Free fatty acids (mg/l) 3-Hydroxybutyric acid (mM) Urine ketones (+) Triglyceride (mM)
Restricted
Week
Week
Week
Change
2
Change (%)
Week
1
1
2
(%)
5.4 6.7 165 330 220 0 3.2
5.2 6.7 165 390 210 0 3.7
-4 0 0 18 -5 0 16
5.9 6.8 265 220 290 1.0 3.1
4.9 5.4 145 360 780 2.1 2.9
-17 -21 -45 64 169 210 -7
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. Based on Magee et al. (17).
heavier than the control subjects, they had a greater tendency to be ketonuric even in the 1st wk of full-calorie feeding (Table 2). Despite the heightened fatty-acid mobilization and ketogenesis, plasma triglyceride concentrations tended to decline in the calorie-restricted group (Table 2). Glucose tolerance did not change in the calorie-restricted group or the control group (Fig. 3). This surprising result occurred even though plasma insulin concentrations decreased markedly in the calorie-restricted group (Table 2). This observation suggests that the improvement in glycemic status is not related to an improvement in the rate of glucose
disposal. In other words, although insulin sensitivity, as reflected in the reduction in fasting plasma insulin concentrations, might be improved, the reduction in insulin concentration appears to nullify any gain in glucose-disposal efficiency in the new steady state. Because of the effect of this degree of calorie restriction on ketonemia and uncertainties about the possible effect on fetal growth and development, a specific dietary recommendation based on these
o u
D
o 03
15 OL
8
10 12 14 Time (hrs)
18 20 22 24
FIG. 2. Twenty-four-hour glucose profile of control subjects (fop) and calorie-restricted subjects (bottom). 0,1st wk of calorie-sufficient feeding for both groups; • , calorie restriction of 2nd wk in calorie-restricted group (bottom). © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
o
60 Time
120
180
(mm)
FIG. 3. Glucose-tolerance tests done at end of 1st (O) and 2nd (•) wk on control (fop) and calorie-restricted (bottom) subjects. Legend is the same as in Fig. 2. No change in glucose tolerance is seen in either group. © 1990 by the American Diabetes Association, Inc. Reprinted by permission from Magee et al. (17).
167
HYPOCALORIC DIETS
TABLE 3 Subject characteristic of group with 33% calorie restriction in pregnancy
Normal Obese Obese with gestational diabetes Insulin Decreased calories
n
Age
Gravida
Prepregnancy ideal body weight (%)
9 7 6 3 3
24.1 25.7 28.8* 26.3 31.3
2 2 3.2 3 3.3
100 128 132t 124 139
Glucose 1 h after 50 g (mM)
Diet (kcal)
Week studied
5.8 6.6 9.1t 8.9 9.4
2352 2374 2415 2504 1655
31.4 32 31.8 33.3 30.3
From Knopp et al. (13). © 1991 John Wiley & Sons, Inc. *P < 0.025, tP < 0.005, for trend vs. normal and obese.
data was not possible. Nonetheless, this study demonstrated the marked extent to which calorie restriction can improve the hyperglycemia of gestational diabetes.
subjects are shown in Table 4. The highest fasting plasma glucose concentrations were seen in the obese gestational diabetic group on a full-calorie diet studied in the 1st wk of the metabolic-ward observation. Fasting plasma glucose was unchanged in three subjects with gestational diabetes treated with insulin therapy in the 2nd wk, and 24-h mean glucose was slightly improved, as was the 24-h glucose profile (Fig. 4). However, obese subjects with gestational diabetes receiving 33% calorie-restricted diets had fasting plasma glucose concentrations below those of the normal subjects and a 24-h mean glucose that was also slightly less than that of the normal subjects and lower than those of the obese or obese gestational diabetic groups. The 24-h glucose profile was also lower than in the women treated with insulin (Fig. 4). Plasma insulin concentrations were higher in the obese gestational diabetic group than in the obese and normal nondiabetic pregnant women. Insulin therapy resulted in a slight increase in plasma IRI, as expected, but a lower insulin concentration was seen in the calorie-restricted group. Mean plasma FFA concentrations were elevated, as were plasma triglyceride concentrations in the obese women with gestational diabetes before therapy and remained so in the gestational diabetic subjects treated with insulin in the 2nd wk. However, FFA and triglyceride concentrations were lower when gestational diabetic women were treated with calorie restriction. These trends were confirmed when individual pre- and posttreatment values were compared (13). Urinary ketone excretion was highest in the obese gestational diabetic group on a full-calorie diet and lower with insulin therapy as well as with calorie restriction. Again, individual pretreatment-posttreatment comparisons confirmed group differences (data not
EXPERIMENT 2: RESEARCH DESIGN AND METHODS AND RESULTS Effect of 3 3 % calorie restriction. The effect of a 3 3 %
calorie-restricted diet was next investigated as a closer approximation to a diet that might be not only tolerable but also beneficial and less likely to promote excessive fat mobilization and ketogenesis (17,19). Six obese subjects with gestational diabetes were studied as shown in Table 3. These subjects were fed a full-calorie diet in the 1st wk from the metabolic ward and were compared with nine nonobese nondiabetic pregnant women and seven obese nondiabetic pregnant women also fed a full-calorie diet. In the 2nd wk, the women with gestational diabetes were randomly assigned either to an insulin-therapy regimen consisting of 20 U of NPH insulin and 10 U of regular insulin, as proposed by Coustan and Imarah (20) or to a 33% calorie-restricted diet. Characteristics of these subjects are shown in Table 3. The mean ages of the different groups ranged from 24 to 31 yr. Prior gravidity ranged between 2 and 3.3, and the obese subjects with gestational diabetes weighed between 124 and 139% of ideal body weight (somewhat less overweight than in the previous study). Plasma glucose concentrations 1 h after a 50-g oral glucose load were clearly elevated in the obese gestationally diabetic group compared with normal subjects, but not significantly so in the obese nondiabetic group compared with the normal group. All subjects were studied between 30 and 33 wk gestation. Mean basal metabolic fuels and hormones of the study
TABLE 4 Basal metabolic fuels and hormones Plasma
Normal Obese Obese with gestational diabetes Insulin Reduced calories
Fasting glucose (mM)
Mean 24-h glucose (mM)
Immunoreactive insulin (pM)
Free fatty acids (mg/l)
3.9 4.0 4.2 4.2 3.7
4.7 4.8 5.1 4.9 4.6
127 164 199* 225 138
100 116 140t 157 83
All values obtained after a 9 h overnight fast. From Knopp et al. (13). *P < 0.01, \P < 0.025, tP < 0.1, for trend vs. normal and obese.
168
Triglyceride (mM)
Urine ketone
(+) 1.7 1.9 2.3+ 2.6 1.5
0.7 0.9 3.Of 1.3 1.7
1991 John Wiley & Sons, Inc.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES mg/dL
mM
TABLE 6 Comparative effects of insulin and 2 levels of calorie restriction on metabolism in gestational diabetes
• — • GDM full calorie (6) 0---0 GDM 2/3 calorie (3) A
A GDM insulin
(3)
Calorierestricted diet
Fasting glucose 24-h glucose Fasting immunoreactive insulin Free fatty acids Ketonuria Triglyceride
50%
33%
Insulin
-17 -21 -45 64 210 -6
-22 -10 -31 -47 -43 -35
0 -4 28 12 -57 13
Results are % change. From Knopp et al. (13). © 1991 John Wiley & Sons, Inc.
7
8
9
10
11
12
differences in the 1st observational wk. Neither insulin therapy nor calorie restriction altered this sensitivity, consistent with the absence of improvement in glucose tolerance in the 50% calorie-restricted study. The amounts of glucose needed to maintain plasma glucose concentrations during insulin infusion were not significantly different, but tended to be lower in the obese women compared with pregnant women of normal weight. This measurement was only slightly lower in gestational diabetic subjects compared with obese nondiabetic subjects. Finally, calculations of insulin sensitivity also showed only slightly lower values in the gestational diabetic subjects, again not statistically different. Thus, our results do not agree with those of Ryan et al., who found increased insulin resistance in gestational diabetic pregnancy (22). Our results are more in keeping with the results of Buchanan et al. (23), who found no increase in insulin resistance in gestational diabetic pregnancy with the intravenous glucose-tolerance test and the Bergman minimal-model technique. Some of the differences among the studies may reflect differences in technique. With insulin or calorie restriction as a form of therapy, the insulin-clamp technique used in our gestational diabetic subjects did not disclose major changes in insulin action or hepatic glucose production, reinforcing the idea that the predominant mechanism of glycemic improvement with 33% calorie restriction is the reduction in intake of calories. Comparison of effects of 50 and 33% calorie restrictions and insulin. A comparison of the effects of the two degrees of calorie restriction is shown in Table 6. Fasting plasma glucose concentrations were reduced by 17% in the 50% calorie-restriction program and 22% with 33% calorie-
1
FIG. 4. Twenty-four-hour glucose profiles in gestational diabetic subjects treated with insulin or with 33% calorie restriction.
shown; see ref. 13). Although the numbers of gestational diabetic subjects treated with insulin or diet are small and do not allow for secure conclusions about the differences between insulin and calorie restriction, 33% calorie restriction did not aggravate fat mobilization and ketogenesis in these subjects. To determine the mechanism of the effect of calorie restriction on glycemic status in gestational diabetic subjects, [6,6-2H2]glucose (dideuterated) was used to measure hepatic glucose output under basal conditions and during an infusion of insulin designed to raise the plasma insulin concentration ~16 |xll/ml (for methodological details, see ref. 13). Observations are shown in Table 5. Basal hepatic glucose output (HGO) was indistinguishable among normal, obese, and obese gestational diabetic pregnant subjects before initiation of insulin or calorie restriction, confirming the observations of Kalhan et al. (21). HGO tended to be higher in the two insulin-treated subjects during the 2nd wk of observation. In contrast, the obese subjects with gestational diabetes treated with calorie restriction had a slightly lower basal hepatic glucose output. However, these differences are small, are not statistically significant, and are unlikely to be so even if more subjects were studied. The calculated amount insulin must be increased above basal to reduce HGO by 50% (ED50) also showed only small, nonsignificant TABLE 5 Measurements of insulin sensitivity and hepatic glucose production
n Normal Obese Obese with gestational diabetes Insulin Decreased calories
5-8 6-7 4-5 2 3
Basal hepatic glucose output (mg • k g " 1 • min" 1 ) 2.99 2.88 2.96 3.21 2.800
± ± ± ± ±
0.45 0.45 0.50 0.10 0.063
ED 50 (%) 136 115 147 140 157
± ± ± ± ±
33 19 41 6 22
Ml 12.7 10.6 9.4 9.3 9.3
ED50, % of basal insulin level at which hepatic glucose production is calculated to be reduced 50%. (mg • k g " 1 • min" 1 )/mean insulin level (|xll/ml) x 100. S,, insulin sensitivity index.
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
± ± ± ± ±
s, 3.8 3.4 1.5 2.9 1.1
2.5 3.0 2.1 2.1 1.5
± ± ± ± ±
1.3 1.7 1.1 0.9 0.2
glucose infusion rate
169
HYPOCALORIC DIETS
these studies suggests that calorie restriction may be a more effective means than insulin to limit the global excess Normal: in plasma fuels seen in the women with gestational diabetes (26). A model of the increased diversion of circulating Glucose maternal fuels to the fetal circulation in the obese subject Amino acids with gestational diabetes is illustrated in Fig. 5. Because of FFA lipoprotein lipase in the placenta (27,28), plasma triglycerTriglyceride ide fatty acids potentially have access to the fetal circulation. Increases in transport of this substrate along with the increases in plasma amino acids and glucose concentrations Gestational (26) should, in concert, enhance fetal growth and macrosoDiabetic: mia. This model is in keeping with the expanded Pedersen Obesity hypothesis proposed by Freinkel and Metzger in 1979 (29). Weight gain The importance of the calorie-restriction program described Insulin here is that it might decrease the concentration of all resistance maternal fuels accessing the fetal circulation. Because O'Sullivan found that even modest doses of insulin given to gestational diabetic mothers (as little as 10 U daily) can reduce infant macrosomia (24), insulin therapy FIG. 5. Model of excess glucose and triglyceride transport to fetus in remains a valuable tool in the management of gestational obese gestational diabetic. Dark lines in gestational diabetic subject (bottom) denote increased transfer of nutrients to fetal circulation. diabetic women, particularly those with more marked gluNote important role of placental lipoprotein lipase in mediating cose intolerance or elevations in fasting glucose concentratriglyceride fatty-acid transport to fetus. tions. On the other hand, one may not wish to use insulin if restriction program, whereas insulin produced no improve- avoidable because it can provoke insulin allergy or serious ment. The 24-h mean glucose concentrations were reduced hypoglycemia or aggravate infant macrosomia by the transby 2 1 % in the 50% calorie-restricted subjects and 10% with placental transport of immunoglobulin- bound maternal or the 33% calorie-restricted subjects and only 4% with insulin exogenously administered insulin (30). For these important therapy. Thus, even with the 33% calorie restriction, glyce- reasons and to differentiate more clearly the metabolic mic status is more improved than with a standard insulin- effects of insulin therapy from those of calorie-restriction therapy regimen (18), a regimen in excess of that previously therapy, it is necessary to perform more such carefully reported to reduce macrosomia in gestational diabetes (24). monitored metabolic-ward studies in more subjects and to In fact, the 33% calorie restriction was sufficient to restore initiate randomized clinical trials in outpatient gestational the gestational diabetic subjects not only to the normal diabetic subjects to judge the effect of the two treatment range but to a level equal to or better than that in normal approaches on perinatal morbidity and future well-being of subjects (Table 4). The extent of plasma IRI reduction with mother and offspring. Only in this deliberate and collabora33% calorie restriction was - 6 6 % of that in the 50% re- tive way can we resolve the persisting questions about how stricted group, consistent with the relative amounts of calorie best to treat gestational diabetes. restriction imposed. The amount of FFA mobilization was Note added in proof.The recent report of the association of elevated in the 50% calorie restriction and reduced in the third trimester ketonemia with reduced intelligence sores at 33%. Similar effects were seen for ketonuria. Finally, reduc- 3, 4, and 5 yr in the offspring (31) underscores the need for tion in plasma triglycerides was seen in both groups of further studies of dietary approaches to the management of subjects, but a greater triglyceride reduction was suggested the gestational diabetic subject. in the 33% group. Reducing plasma triglyceride concentrations is important because plasma triglyceride concentrations are elevated in gestational diabetic subjects and the ACKNOWLEDGMENTS triglyceride elevation is more strongly associated with infant This work was supported by Grant DK-35816 (Clinical Numacrosomia than is fasting or postprandial plasma glucose trition Research Unit), Clinical Research Center Grant MO1RR-0037, and Diabetes and Endocrinology Research concentrations (25). Center Grant DK-17047, all from the National Institutes of Overall, the 33% calorie-restricted regimen, which is an Health. This study was also supported by Training Grant -1600- to 1800-kcal • day" 1 diet, is sufficient to normalize DK-07247 in endocrinology and metabolism (M.S.M.), a glycemic status without apparent adverse metabolic effect. research fellowship from the American Diabetes Association This diet is similar to those used by previous workers (2-5), Washington Affiliate, and a generous grant from the WashDomhorst et al. (this issue, p. 161) in the management of ington State Eagles. obese subjects with gestational diabetes or obese pregnant We express our gratitude to Dr. Zane Brown for assissubjects without adverse effect on fetal outcome. More tance in identifying patients and to Al Greeves for typing the research is needed to extend and confirm these limited manuscript. observations.
Mother
Placenta
Fetus
DISCUSSION
The greater glycemic improvement associated with two levels of calorie restriction compared with insulin therapy in
170
REFERENCES 1. Committee on Maternal Nutrition/Food and Nutrition Board, National Research Council: Maternal Nutrition and the Course of Pregnancy. Washington, DC, Natl. Acad. Sci., 1970, p. 1-124
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
R.H. KNOPP AND ASSOCIATES 2. Borberg C, Gillmer MDG, Brunner EJ, Gunn PJ, Oakley NW, Beard RW: Obesity in pregnancy: the effect of dietary advice. Diabetes Care 3:476-81,1980 3. Algert S, Shragg P, Hollingworth DR: Moderate caloric restriction in obese women with gestational diabetes. Obstet Gynecol 65:487-91, 1985 4. National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:103957,1979 5. Pedersen J: The Pregnant Diabetic and Her Newborn. 2nd ed. Baltimore, MD, Williams & Wilkins, 1977, p. 223 6. Naeye RL: Weight gain and the outcome of pregnancy. Am J Obstet Gynecol 135:3-9, 1979 7. Churchill JA, Berendes HW: Intelligence of children whose mothers have acetonuria in pregnancy. In Perinatal Factors Affecting Human Development. Washington, DC, Pan Am. Health Org., 1969 (Sci. publ. no. 185) 8. Naeye RL, Chez RA: Effects of maternal acetonuria and low pregnancy weight gain on children's psychomotor development. Am J Obstet Gynecol 139:189-93, 1981 9. Stehbens JA, Baker Gl, Kitchell M: Outcome at ages 1, 3, and 5 years of children born to diabetic mothers. Am J Obstet Gynecol 127:408-13, 1977 10. Naeye RL: The outcome of diabetic pregnancies: a prospective study. In Pregnancy Metabolism, Diabetes and the Fetus. Amsterdam, Excerpta Med., 1979, p. 227-41 (Ciba Found. Symp. no. 63) 11. Persson B, Gentz J: Follow-up of children of insulin-dependent and gestational diabetic mothers: neuropsychological outcome. Acta Paediatr Scand 73:349-58, 1984 12. Stein Z, Susser M, Saenger G, Marolla F: Famine and Human Development. New York, Oxford Univ. Press, 1975, p. 1-284 13. Knopp RH, Magee MS, Raisys V, Benedetti T, Bonet B: Hypocaloric diets and ketogenesis in the management of obese gestational diabetic women. J Am Coll Nutr 10:649-67, 1991 14. Paterson P, Sheath J, Taft P, Wood C: Maternal and fetal ketone concentrations in plasma and urine. LanceM:862-65, 1967 15. Coetzee EJ, Jackson WPU, Berman PA: Ketonuria in pregnancy—with special reference to calorie-restricted food intake in obese diabetics. Diabetes 29:177-81, 1980 16. Chez RA, Curcio FD III: Ketonuria in normal pregnancy. Obstet Gynecol 69:272-74, 1987 17. Magee MS, Knopp RH, Benedetti TJ: Metabolic effects of a 1200-kcal diet in obese pregnant women with gestational diabetes. Diabetes 39:234-40, 1990
DIABETES, VOL. 40, SUPPL. 2, DECEMBER 1991
18. Carpenter MW, Coustan DR: Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol 144:768-73, 1982 19. Magee MS, Knopp R, Raisys V, Prince E, Brown Z, Benedetti T: Pathophysiology and response to insulin or dietary treatment in gestational diabetes (Abstract). Diabetes 37 (Suppl. 1):111A, 1988 20. Coustan DR, Imarah J: Prophylactic insulin treatment of gestational diabetes reduces the incidence of macrosomia, operative delivery and birth trauma. Am J Obstet Gynecol 150:836-43, 1984 21. Kalhan SC, D'Angelo LJ, Savin SM, Adam PAJ: Glucose production in pregnant women at term gestation: sources of glucose for human fetus. J Clin Invest 63:388-94, 1979 22. Ryan EA, O'Sullivan MJ, Skyler JS: Insulin action during pregnancy: studies with the euglycemic clamp technique. Diabetes 34:380-89,1985 23. Buchanan TA, Metzger BE, Freinkel N, Bergman RN: Insulin sensitivity and B-cell responsiveness to glucose during late pregnancy in lean and moderately obese women with normal glucose tolerance or mild gestational diabetes. Am J Obstet Gynecol 162:1008-14, 1990 24. O'Sullivan JB: Insulin treatment for gestational diabetes. In Early Diabetes in Early Life. Camerini-Davalos RA, Cole HS, Eds. New York, Academic, 1975, p. 447-54. 25. Knopp RH, Magee MS, Larson MP, Benedetti T: Alternative screening tests and birth weight associations in pregnant women with abnormal glucose screening (Abstract). Diabetes 37 (Suppl. 1):110A, 1988 26. Metzger BE, Phelps RL, Freinkel N, Navickas IA: Effects of gestational diabetes on diurnal profiles of plasma glucose, lipids, and individual amino acids. Diabetes Care 3:402-409, 1980 27. Knopp RH, Warth MR, Charles D, Childs M, Li JR, Mabuchi H, Van Allen M: Lipoprotein metabolism in pregnancy, fat transport to the fetus, and the effects of diabetes. Biol Neonate (Paris) 50:297-317, 1986 28. Bonet B, Gown AM, Brunzell JD, Knopp RH: Role of lipoprotein lipase in triglyceride metabolism by placental trophoblast and macrophages (Abstract). Arteriosclerosis 7:642A, 1987 29. Freinkel N, Metzger BE: Pregnancy as a tissue culture experience: the critical implications of maternal metabolism for fetal development. In Pregnancy Metabolism, Diabetes and the Fetus. Amsterdam, Excerpta Med., 1979, p. 3-23 (Ciba Found. Symp. no. 63) 30. Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F, Khoury JC: Transplacental passage of insulin in pregnant women with insulindependent diabetes mellitus. N EnglJ Med 323:309-14, 1990 31. Rizzo T, Metzger BE, Burns WJ, Burns K: Correlations between antepartum maternal metabolism and intelligence of offspring. N Engl J Med 325:911-16, 1991
171
