- Email: [email protected]
In pregnancies with gestational diabetes mellitus and intensive therapy, perinatal outcome is worse in small-for-gestational-age newborns
In pregnancies with gestational diabetes mellitus and intensive therapy, perinatal outcome is worse in small-for-gestational-age newborns Apolonia García-Patterson, MD, Rosa Corcoy, PhD, Montserrat Balsells, MD, Orenci Altirriba, PhD, Juan M. Adelantado, PhD, Lluis Cabero, PhD, and Alberto de Leiva, PhD Barcelona, Spain OBJECTIVE: This study analyzed the relationship between birth weight and perinatal outcome among women with gestational diabetes mellitus. STUDY DESIGN: The relationship between perinatal outcome and birth weight was analyzed for 821 pregnancies of women with gestational diabetes mellitus attended in a tertiary hospital and receiving intensive metabolic therapy (normocaloric diet, self-monitoring of blood glucose level and individually tailored insulin regimen when needed). The Mantel-Haenszel test was used to adjust for preterm delivery. RESULTS: Seven percent of neonates were small for gestational age, 85% were appropriate for gestational age, and 8% were large for gestational age. After adjustment for preterm delivery the rates of adverse fetal outcome, low 1-minute Apgar score, and hypoglycemia were greater among small for gestational age neonates than among appropriate and large for gestational age infants (odds ratios 3.08, 2.51, and 3.17, respectively). CONCLUSION: Among women with gestational diabetes mellitus who are receiving intensive therapy, perinatal outcome is worse for small for gestational age neonates than for appropriate and large for gestational age neonates. (Am J Obstet Gynecol 1998;179:481-5.)
Key words: Gestational diabetes mellitus, birth weight, perinatal outcome
Neonates born to women with gestational diabetes mellitus have a perinatal mortality rate similar to that of the control population1 but increased rates of macrosomia and morbidity in terms of neonatal asphyxia, birth trauma, hypoglycemia, hypocalcemia, hyperbilirubinemia, and respiratory distress, among others.1,2 The excess of morbidity is due to a hyperinsulinemic state of the infant.3 Because of a common physiopathologic pathway, we could expect macrosomia and neonatal morbidity and mortality to be associated. Most authors agree that this holds true in cases of insulin-dependent diabetes mellitus, especially for hypoglycemia.3,4 For gestational diabetes mellitus, Stenninger et al2 did not find any relationship between macrosomia and hypoglycemia, whereas other authors describe a higher rate of obstetric trauma but similar rates of 6 additional morbidity items (hypoglycemia, hypocalcemia, respiratory distress, polycythemia, jaundice, and malformations).4 In the series of
Langer et al,5 in which women had gestational diabetes mellitus that was kept under strict metabolic control, morbidity followed either an L-shaped pattern for hyperbilirubinemia (higher incidence among small for gestational age [SGA] than among appropriate for gestational age [AGA] and large for gestational age [LGA] infants), a U-shaped pattern for neonatal intensive care unit stay (higher incidence among SGA and LGA infants than among AGA infants), a J-shaped pattern for polycythemia (higher incidence among LGA infants than among SGA and AGA infants), or no pattern at all for hypoglycemia. A higher risk of perinatal mortality for the macrosomic fetus is nearly an axiom in diabetes and pregnancy management, although most reports document a higher mortality rate among SGA infants.6 The aim of this study was to analyze the relationship between birth weight and perinatal outcome among women with gestational diabetes mellitus. Patients and methods
From the Endocrinology, Pediatric, and Obstetric Departments, Hospital de Sant Pau, Universitat Autònoma de Barcelona. Received for publication September 3, 1997; revised December 13, 1997; accepted January 15, 1998. Reprint requests: Rosa Corcoy, PhD, Servei d’Endocrinologia, Hospital de Sant Pau, Avgda Sant Antoni Ma Claret 167, Barcelona 08025, Spain. Copyright © 1998 by Mosby, Inc. 0002-9378/98 $5.00 + 0 6/1/89116
Between 1986 and 1991, 821 pregnancies in women with gestational diabetes mellitus were attended in the Diabetic and Pregnancy Clinic. Three pregnancies ended in an embryonic loss, whereas the others progressed beyond 22 weeks: 804 singleton pregnancies, 13 twin pregnancies, and 1 triplet pregnancy. Because we intended to analyze the relationship between birth weight and mor481
482 García-Patterson et al
August 1998 Am J Obstet Gynecol
Table I. Maternal characteristics of women with gestational diabetes mellitus Variable Age (y) Body mass index (kg/m2) Gestational age (wk) At diagnosis When care was initiated Weight gain during pregnancy (kg) Women receiving insulin therapy (%) Mean dose in women treated with insulin (IU/kg/d) Third-trimester hemoglobin A1c*
Mean ± SD 31.43 ± 4.63 24.05 ± 3.97 30.00 ± 6.45 32.50 ± 5.76 9.48 ± 4.17 66.5 0.34 ± 0.25 –0.62 ± 1.4
*Hemoglobin A1c is expressed in SDs around the mean for the nondiabetic population.
bidity in an unselected population of women with gestational diabetes mellitus, all neonates (singleton and from multiple pregnancies) were included. The diagnosis of gestational diabetes mellitus was established by an oral glucose tolerance test in accordance with the National Diabetes Data Group Criteria.7 The gestational diabetes mellitus screening protocol in this hospital includes an O’Sullivan test after the first visit and at between 24 and 28 and between 32 and 35 weeks of pregnancy. Diabetes control. Therapy was initiated with normocaloric diet and self-monitoring of blood glucose, with insulin therapy added when blood glucose values rose above the target range. Normocaloric diet energy requirements were derived according to basal metabolic rate (Harris-Benedict equation), estimated energy expenditure of activity, and pregnancy requirement. Before 1990 the caloric supplementation for pregnancy was 150 kcal before 20 weeks of gestation and 350 kcal from then on.8 After 1990 this supplement consisted of 150 kcal from 20 weeks onward.9 Self-monitoring of blood glucose with BM-Test Glycemie 20-800 R (Boehringer-Mannheim, Mannheim, Germany) was performed by all these women to promote optimal metabolic control. Initially strips were checked weekly in the hospital (Reflolux I [BoehringerMannheim]). Afterward, Reflolux II (BoehringerMannheim) meters were available to be loaned to women with gestational diabetes mellitus during pregnancy. They were requested to test their blood glucose 4 times a day (fasting, before lunch, before dinner, and 1 variable 1hour postprandial blood glucose level) and the procedure accuracy was assessed during the first visits after training. Insulin was administered when blood glucose values rose above critical targets adapted to the reflectance meter: 105 mg/dL fasting/preprandial and 140 mg/dL 1-hour postprandial with Reflolux I and 90 mg/dL and 120 mg/dL, respectively, with Reflolux II, both roughly equivalent to venous plasma values of 95 mg/dL fasting and 125 mg/dL postprandially.10,11 Insulin therapy was usually initiated as a bedtime intermediate insulin injection, with regular insulin being added before
meals when necessary. Eventually a program of 4 injections/day could be reached. Maternal characteristics are summarized in Table I. Metabolic control is defined in terms of third-trimester hemoglobin A1c level, expressed in standard deviations around the mean for the nondiabetic population. Outcome definitions. Perinatal death was defined as any fetal or neonatal death occurring from 22 weeks’ gestation through the first 4 weeks after birth.12 Preterm birth was defined as birth before 37 complete gestational weeks. Neonatal weight was classified as SGA, AGA, or LGA (birth weight <10th percentile, between 10th and 90th percentiles, and >90th percentile, respectively) according to gestational age and sex.13 Neonatal hypoglycemia was defined in accordance with Cornblath and Reisner’s criteria14 for capillary blood glucose, which take in consideration both birth weight and prematurity. Blood glucose level was measured with a reflectance meter to allow immediate initiation of treatment when necessary. Hypocalcemia was defined as a serum calcium level <1.75 mmol/L.15 Neonatal jaundice was defined as hyperbilirubinemia requiring treatment according to Maisels.16 Polycythemia was defined as a hematocrit >60% in cord blood.17 All types of respiratory distress, including transient tachypnea, were considered. Low Apgar score was defined as a 1- or 5-minute Apgar score <7. Statistical methods. The χ2 test was used to analyze morbidity and mortality distributions among groups. When the percentage of expected frequencies <5 was >25% of cells, infants belonging to a given birth weight group were grouped into whichever other group held the more similar rate for the specific morbidity. The Mantel-Haenszel test was used to adjust for preterm birth. Results Birth weight distribution of neonates born to women with gestational diabetes mellitus was as follows: 61 SGA (7.32%), 708 AGA (85.0%), and 64 LGA (7.68%). In the SGA group low 1-minute Apgar score was recorded in 20.3% of cases, a rate markedly higher than seen in the other groups (7.3% and 6.3% in the AGA and LGA groups, respectively, P < .01). The Apgar score at 5 minutes was low in 1.67% of SGA, 0.86% of AGA, and 0% of LGA neonates (P not significant). We observed a high frequency of hypoglycemia (25.4%) among the SGA infants, compared with 8.2% in the AGA group and 9.4% in the LGA group (P < .0001). The rates of hyperbilirubinemia were 11.9% among SGA, 2.8% among AGA, and 0% among LGA infants, with the frequency in the SGA group significantly higher than in the AGA and LGA groups (P < .01). With respect to respiratory distress, its rate was 8.5% for SGA, 3.3% for AGA, and 3.1% for LGA infants, with the significance of SGA versus AGA and LGA being bor-
García-Patterson et al 483
Volume 179, Number 2 Am J Obstet Gynecol
Table II. Morbidity in infants of women with gestational diabetes mellitus according to birth weight
Adverse fetal outcome 1* Adverse fetal outcome 2† Low 1-min Apgar score Hypoglycemia Hyperbilirubinemia Respiratory distress Major malformation Hypocalcemia Polycythemia Low 5-min Apgar score Obstetric trauma Minor malformation Mortality
All groups (N = 833)
SGA (n = 61)
AGA (n = 708)
LGA (n = 64)
Significance
24.9 19.7 8.2 9.5 3.2 3.7 2.3 2.7 1.0 0.86 2.3 3.8 0.2
53.2 43.5 20.3 25.4 11.9 8.5 0.0 4.3 1.7 1.67 1.7 1.7 0.0
22.4 17.5 7.3 8.2 2.8 3.3 2.6 2.2 0.9 0.86 2.2 3.8 0.12
20.8 17.4 6.3 9.4 0.0 3.1 1.6 6.3 1.6 0 4.7 7.8 1.6
P < .0001 P < .001 P < .01 P < .0001 P < .01‡ P = .056‡ P = NS‡ P = NS§ P = NS§ P = NSll P = NSll P = NSll P = NSll
NS, Not significant. *Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and death. †Adverse fetal outcome 2 was defined as adverse fetal outcome 1 but with Apgar score at 5 minutes instead of 1 minute. ‡SGA versus AGA and LGA. §SGA and LGA versus AGA. llSGA and AGA versus LGA.
derline (P = .056). We did not find any differences between groups in rates of obstetric trauma, hypocalcemia, polycythemia, minor and major malformations, and perinatal mortality, even after grouping according to the aforementioned criteria. Adverse fetal outcome 1 (defined as ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality) was present in 53.2% of SGA, 22.4% of AGA and 20.8% of LGA infants (P < .0001). Adverse fetal outcome 2 used 5-minute Apgar score and was recorded for 43.5% of SGA, 17.5% of AGA, and 17.4% of LGA infants (P < .001; Table II). The relationship between birth weight category and the rate of any morbidity is displayed in Fig. 1. The rates of preterm birth before 37 weeks of gestation were 25.4% in the SGA group, 4.4% in the AGA group, and 3.1% in the LGA group (P < .001), whereas birth before 34 weeks of gestation did not differ among groups (4.91% in SGA, 1.13% in AGA, and 0% in LGA, P not significant). Because low Apgar score,18 hypoglycemia,19 respiratory distress,20 and hyperbilirubinemia20,21 are more frequent among preterm infants, the observed morbidity in the SGA group could be attributable to the prematurity itself, indicating the need to adjust for this variable. Because morbidity did not differ between the AGA and LGA groups, these groups were analyzed together when performing the adjustment for prematurity (SGA versus non-SGA). Elective delivery because of growth retardation did not contribute significantly to preterm delivery in SGA infants (6.67% in SGA, 0% in AGA, and 0% in LGA, P not significant). After adjustment for preterm birth (Table III), the rate of adverse fetal outcome was higher in the SGA group both when including 1-minute Apgar score (odds ratio 3.08, 95% confidence interval 1.63 to 5.81) and when in-
cluding 5-minute Apgar score (odds ratio 2.48, 95% confidence interval 1.27 to 4.82). The same pattern was observed for 1-minute Apgar scores <7 (odds ratio 2.51, 95% confidence 1.23 to 5.15), with a similar trend for scores <4 (overall rate 1.93%, odds ratio for SGA 3.11, 95% CI 0.924 to 10.5). Hypoglycemia remained associated with low birth weight after adjustment for prematurity (odds ratio 3.17, 95% confidence interval 1.60 to 6.28). When the effect of preterm birth was controlled for, however, hyperbilirubinemia (odds ratio 1.71, 95% confidence interval 0.69 to 4.24) and respiratory distress (odds ratio 1.25, 95% confidence interval 0.42 to 3.71) were no longer associated with low birth weight. In the subgroups of term and preterm infants, birth weight influenced morbidity in the same direction for all outcomes (data not shown), even though statistical significance was reached only for neonatal hypoglycemia in the term subgroup (odds ratio 3.01, 95% confidence interval 1.38 to 6.58). The exception was hyperbilirubinemia, for which the patterns in term and preterm subgroups were divergent. Whereas low birth weight seemed to protect against hyperbilirubinemia in the preterm group (odds ratio 0.34, 95% confidence interval 0.06 to 1.79), it was clearly associated with hyperbilirubinemia in term infants (odds ratio 9.29, 95% confidence interval 3.03 to 28.5). If we limit the analysis to singleton pregnancies, the rates of hypoglycemia (odds ratio 3.06, 95% confidence interval 1.43 to 6.55) and adverse fetal outcome with 1-minute Apgar score (odds ratio 2.20, 95% confidence interval 1.06 to 4.55) were also higher in the SGA group after controlling for prematurity. Comment In this group of infants of women with gestational diabetes mellitus with tight metabolic control according to
484 García-Patterson et al
August 1998 Am J Obstet Gynecol
Fig 1. Rates of adverse fetal outcome 1 among SGA and non-SGA neonates adjusted for gestational age (P < .05). Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality.
Table III. Morbidity outcomes among infants of women with gestational diabetes mellitus under tight metabolic control Morbidity outcomes Adverse fetal outcome 1 Adverse fetal outcome 2† Low 1-min Apgar score Hypoglycemia Hyperbilirubinemia Respiratory distress
Odds ratio
95% Confidence interval
3.08 2.48 2.51 3.17 1.71 1.25
1.63-5.81 1.27-4.82 1.23-5.15 1.60-6.28 0.69-4.24 0.42-3.71
*Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality. †Adverse fetal outcome 2 was defined as adverse fetal outcome 1 but with Apgar score at 5 minutes instead of 1 minute.
third-trimester hemoglobin A1c a good overall perinatal outcome was achieved. This compares favorably with other reports of women with intensively treated gestational diabetes.4,5 The somewhat high rate of hypoglycemia (9.4%) may be partly attributed to the analytic method (capillary blood, reflectance meter) chosen to allow an early therapy in case of a low blood glucose measurement. Regarding the distribution of morbidity and mortality among the 3 birth weight subgroups, the LGA subgroup did not show a higher rate for any of the analyzed entities. In contrast, SGA neonates after adjustment for preterm delivery demonstrated a higher rate of adverse fetal outcome, low Apgar score in the first minute after
delivery, and hypoglycemia than did either AGA or LGA infants (L-shaped pattern of morbidity). No significant differences could be demonstrated across birth weight categories for malformations, hypocalcemia, polycythemia, obstetric trauma, and mortality. Our results agree with those of Stenninger et al2 in that there is no excess hypoglycemia seen among macrosomic infants and with those of Megía et al4 in that the rates of hypoglycemia, hyperbilirubinemia, respiratory distress, malformations, hypocalcemia, and polycythemia are similar in the macrosomic and nonmacrosomic subgroups. However, Megía et al4 did find an excess of obstetric trauma in the macrosomic infants, which we did not observe. In the series of Langer et al,5 hyperbilirubinemia had an L-shaped pattern, polycythemia had a J-shaped pattern, and neonatal intensive care unit admission had a Ushaped pattern. Taken together, these patterns would indicate that morbidity in the SGA subgroup is at least as important as in the LGA subgroup. It is noteworthy that some of the aforementioned authors did not consider the morbidity in the SGA subgroup.2,4 In these articles, as well as in several others,22 the rate of SGA among infants is not reported. In our opinion, however, the percentage of SGA infants should be included in the outcome analysis of diabetic pregnancies, not only because of its relationship with neonatal morbidity but also because of its association with adult health outcomes such as diabetes mellitus, hypertension, and obesity.23 We have shown that, after adjustment for preterm
Volume 179, Number 2 Am J Obstet Gynecol
birth, SGA neonates of mothers with gestational diabetes mellitus receiving intensive therapy have a greater risk than do those who are AGA or LGA for low 1-minute Apgar score, hypoglycemia, and adverse fetal outcome. Additionally, subgroups of term and preterm infants behave similarly to the pattern that has been described for neonates of nondiabetic mothers. For preterm neonates low birth weight is a risk factor for hypoglycemia19 and a protective factor against hyperbilirubinemia21 but has either no effect or a deleterious effect24 on respiratory distress, with the same trend being observed among the preterm babies in our study. For term neonates, on the other hand, low birth weight is a risk factor for hypoglycemia,19 hyperbilirubinemia,21 low Apgar score,25 and respiratory distress.25 The same pattern was observed in this study, although statistical significance was attained only for hypoglycemia and hyperbilirubinemia. We conclude that in this group of 833 infants born to women with gestational diabetes mellitus managed with an intensified treatment, perinatal outcomes were similar for LGA and AGA neonates but worse, and not attributable to prematurity, for SGA babies.
REFERENCES
1. Jacobson JD, Cousins L. A population-based study of maternal and perinatal outcome in patients with gestational diabetes. Am J Obstet Gynecol 1989;161:981-6. 2. Stenninger E, Schollin J, Aman J. Neonatal macrosomia and hypoglycemia in children of mothers with insulin-treated gestational diabetes mellitus. Acta Paediatr Scand 1991;80:1014-8. 3. Sosenko IR, Kitzmiller JL, Loo SW, Blix P, Rubenstein A, Gabbay KH. The infant of diabetic mother: correlation of increased cord C-peptide levels with macrosomia and hypoglycemia. N Engl J Med 1979;301:859-62. 4. Megía A, Luna R, López A, Grande C, Jáñez M, Pallardo LF. Macrosomía fetal en la gestante diabética. Av Diabetol 1991;4:43-9. 5. Langer O, Levy S, Brustman L, Anyaegbunam A, Merkatz R, Divon M. Glycemic control in gestational diabetes mellitus— how tight is tight enough: small for gestational age versus large for gestational age? Am J Obstet Gynecol 1989;161:646-53. 6. Driscoll SG, Benirschke K, Curtis GW. Neonatal deaths among infants of diabetic mothers. Am J Dis Child 1960;100:818-35. 7. National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979;28:1039-57.
García-Patterson et al 485
8. Food and Nutrition Board. Calorías. In: Raciones dietéticas recomendadas. Barcelona (Spain): Ediciones Consulta SA; 1991. p. 23-37. 9. Durnin JG. Energy requirements of pregnancy. Diabetes 1991;40(Suppl 2):152-6. 10. Weiner CP, Faustich MW, Burns J, Fraser M, Whitaker L, Klugman M. Diagnosis of gestational diabetes by capillary blood samples and a portable reflectance meter: derivation of threshold values and prospective validation. Am J Obstet Gynecol 1987;156:1085-9. 11. Carr S, Coustan DR, Martelly P, Brosco F, Rotondo L. Precision of reflectance meters in screening for gestational diabetes. Obstet Gynecol 1989;73:727-31. 12. Working group on the very low birthweight infant. European Community collaborative study of outcome of pregnancy between 22 and 28 weeks gestation. Lancet 1990;336:782-4. 13. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963;32:793-800. 14. Cornblath M, Reisner SH. Blood glucose in the neonate and its clinical significance. N Engl J Med 1965;273:378-81. 15. Robson AM. Consideraciones generales sobre la asistencia del niño enfermo: tetania hipocalcémica. In: Behrman RE, editor. Tratado de pediatría. Madrid (Spain): Interamericana–McGrawHill; 1992. p. 254-6. 16. Maisels MJ. Bilirubin: on understanding and influencing its metabolism in the newborn infant. Pediatr Clin North Am 1972;19:447-503. 17. Ramamurthy RS, Brans YW. Neonatal polycythemia I. Criteria for diagnosis and treatment. Pediatrics 1981;68:168-73. 18. Catlin EA, Carpenter MW, Brann BS, Mayfield SR, Shaul PW, Goldstein M, et al. The Apgar score revisited: influence of gestational age. J Pediatr 1986;109:865-8. 19. Lubchenco LO, Bard H. Incidence of hypoglycemia in newborn infants classified by birthweight and gestational age. Pediatrics 1971;47:831-8. 20. Wigton TR, Tamura RK, Wickstrom E, Atkins V, Deddish R, Socol ML. Neonatal morbidity after preterm delivery in the presence of documented lung maturity. Am J Obstet Gynecol 1993;169:951-5. 21. Friedman L, Lewis PJ, Clifton P, Bulpitt CJ. Factors influencing the incidence of neonatal jaundice [letter]. BMJ 1978;2:1235-7. 22. Hod M, Merlob P, Friedman S, Schoenfeld A, Ovadia J. Gestational diabetes mellitus: a survey of perinatal complications in the 1980s. Diabetes 1991;40(Suppl 2):74-8. 23. Robinson S, Walton RJ, Clark PM, Barker DJ, Hales CN, Osmond C. The relation of fetal growth to plasma glucose in young men. Diabetologia 1992;35:444-6. 24. Piper JM, Xenakis EM, McFarland M, Elliott BD, Berkus MD, Langer O. Do growth-retarded premature infants have different rates of perinatal morbidity and mortality than appropriately grown premature infants? Obstet Gynecol 1996;87:169-74. 25. Lin YS, Chang FM, Liu CH. Comparison of umbilical blood gas and acid-base status of small-for-dates and normal Chinese newborns. J Formos Med Assoc 1992;91:396-9.
Key words: Gestational diabetes mellitus, birth weight, perinatal outcome
Neonates born to women with gestational diabetes mellitus have a perinatal mortality rate similar to that of the control population1 but increased rates of macrosomia and morbidity in terms of neonatal asphyxia, birth trauma, hypoglycemia, hypocalcemia, hyperbilirubinemia, and respiratory distress, among others.1,2 The excess of morbidity is due to a hyperinsulinemic state of the infant.3 Because of a common physiopathologic pathway, we could expect macrosomia and neonatal morbidity and mortality to be associated. Most authors agree that this holds true in cases of insulin-dependent diabetes mellitus, especially for hypoglycemia.3,4 For gestational diabetes mellitus, Stenninger et al2 did not find any relationship between macrosomia and hypoglycemia, whereas other authors describe a higher rate of obstetric trauma but similar rates of 6 additional morbidity items (hypoglycemia, hypocalcemia, respiratory distress, polycythemia, jaundice, and malformations).4 In the series of
Langer et al,5 in which women had gestational diabetes mellitus that was kept under strict metabolic control, morbidity followed either an L-shaped pattern for hyperbilirubinemia (higher incidence among small for gestational age [SGA] than among appropriate for gestational age [AGA] and large for gestational age [LGA] infants), a U-shaped pattern for neonatal intensive care unit stay (higher incidence among SGA and LGA infants than among AGA infants), a J-shaped pattern for polycythemia (higher incidence among LGA infants than among SGA and AGA infants), or no pattern at all for hypoglycemia. A higher risk of perinatal mortality for the macrosomic fetus is nearly an axiom in diabetes and pregnancy management, although most reports document a higher mortality rate among SGA infants.6 The aim of this study was to analyze the relationship between birth weight and perinatal outcome among women with gestational diabetes mellitus. Patients and methods
From the Endocrinology, Pediatric, and Obstetric Departments, Hospital de Sant Pau, Universitat Autònoma de Barcelona. Received for publication September 3, 1997; revised December 13, 1997; accepted January 15, 1998. Reprint requests: Rosa Corcoy, PhD, Servei d’Endocrinologia, Hospital de Sant Pau, Avgda Sant Antoni Ma Claret 167, Barcelona 08025, Spain. Copyright © 1998 by Mosby, Inc. 0002-9378/98 $5.00 + 0 6/1/89116
Between 1986 and 1991, 821 pregnancies in women with gestational diabetes mellitus were attended in the Diabetic and Pregnancy Clinic. Three pregnancies ended in an embryonic loss, whereas the others progressed beyond 22 weeks: 804 singleton pregnancies, 13 twin pregnancies, and 1 triplet pregnancy. Because we intended to analyze the relationship between birth weight and mor481
482 García-Patterson et al
August 1998 Am J Obstet Gynecol
Table I. Maternal characteristics of women with gestational diabetes mellitus Variable Age (y) Body mass index (kg/m2) Gestational age (wk) At diagnosis When care was initiated Weight gain during pregnancy (kg) Women receiving insulin therapy (%) Mean dose in women treated with insulin (IU/kg/d) Third-trimester hemoglobin A1c*
Mean ± SD 31.43 ± 4.63 24.05 ± 3.97 30.00 ± 6.45 32.50 ± 5.76 9.48 ± 4.17 66.5 0.34 ± 0.25 –0.62 ± 1.4
*Hemoglobin A1c is expressed in SDs around the mean for the nondiabetic population.
bidity in an unselected population of women with gestational diabetes mellitus, all neonates (singleton and from multiple pregnancies) were included. The diagnosis of gestational diabetes mellitus was established by an oral glucose tolerance test in accordance with the National Diabetes Data Group Criteria.7 The gestational diabetes mellitus screening protocol in this hospital includes an O’Sullivan test after the first visit and at between 24 and 28 and between 32 and 35 weeks of pregnancy. Diabetes control. Therapy was initiated with normocaloric diet and self-monitoring of blood glucose, with insulin therapy added when blood glucose values rose above the target range. Normocaloric diet energy requirements were derived according to basal metabolic rate (Harris-Benedict equation), estimated energy expenditure of activity, and pregnancy requirement. Before 1990 the caloric supplementation for pregnancy was 150 kcal before 20 weeks of gestation and 350 kcal from then on.8 After 1990 this supplement consisted of 150 kcal from 20 weeks onward.9 Self-monitoring of blood glucose with BM-Test Glycemie 20-800 R (Boehringer-Mannheim, Mannheim, Germany) was performed by all these women to promote optimal metabolic control. Initially strips were checked weekly in the hospital (Reflolux I [BoehringerMannheim]). Afterward, Reflolux II (BoehringerMannheim) meters were available to be loaned to women with gestational diabetes mellitus during pregnancy. They were requested to test their blood glucose 4 times a day (fasting, before lunch, before dinner, and 1 variable 1hour postprandial blood glucose level) and the procedure accuracy was assessed during the first visits after training. Insulin was administered when blood glucose values rose above critical targets adapted to the reflectance meter: 105 mg/dL fasting/preprandial and 140 mg/dL 1-hour postprandial with Reflolux I and 90 mg/dL and 120 mg/dL, respectively, with Reflolux II, both roughly equivalent to venous plasma values of 95 mg/dL fasting and 125 mg/dL postprandially.10,11 Insulin therapy was usually initiated as a bedtime intermediate insulin injection, with regular insulin being added before
meals when necessary. Eventually a program of 4 injections/day could be reached. Maternal characteristics are summarized in Table I. Metabolic control is defined in terms of third-trimester hemoglobin A1c level, expressed in standard deviations around the mean for the nondiabetic population. Outcome definitions. Perinatal death was defined as any fetal or neonatal death occurring from 22 weeks’ gestation through the first 4 weeks after birth.12 Preterm birth was defined as birth before 37 complete gestational weeks. Neonatal weight was classified as SGA, AGA, or LGA (birth weight <10th percentile, between 10th and 90th percentiles, and >90th percentile, respectively) according to gestational age and sex.13 Neonatal hypoglycemia was defined in accordance with Cornblath and Reisner’s criteria14 for capillary blood glucose, which take in consideration both birth weight and prematurity. Blood glucose level was measured with a reflectance meter to allow immediate initiation of treatment when necessary. Hypocalcemia was defined as a serum calcium level <1.75 mmol/L.15 Neonatal jaundice was defined as hyperbilirubinemia requiring treatment according to Maisels.16 Polycythemia was defined as a hematocrit >60% in cord blood.17 All types of respiratory distress, including transient tachypnea, were considered. Low Apgar score was defined as a 1- or 5-minute Apgar score <7. Statistical methods. The χ2 test was used to analyze morbidity and mortality distributions among groups. When the percentage of expected frequencies <5 was >25% of cells, infants belonging to a given birth weight group were grouped into whichever other group held the more similar rate for the specific morbidity. The Mantel-Haenszel test was used to adjust for preterm birth. Results Birth weight distribution of neonates born to women with gestational diabetes mellitus was as follows: 61 SGA (7.32%), 708 AGA (85.0%), and 64 LGA (7.68%). In the SGA group low 1-minute Apgar score was recorded in 20.3% of cases, a rate markedly higher than seen in the other groups (7.3% and 6.3% in the AGA and LGA groups, respectively, P < .01). The Apgar score at 5 minutes was low in 1.67% of SGA, 0.86% of AGA, and 0% of LGA neonates (P not significant). We observed a high frequency of hypoglycemia (25.4%) among the SGA infants, compared with 8.2% in the AGA group and 9.4% in the LGA group (P < .0001). The rates of hyperbilirubinemia were 11.9% among SGA, 2.8% among AGA, and 0% among LGA infants, with the frequency in the SGA group significantly higher than in the AGA and LGA groups (P < .01). With respect to respiratory distress, its rate was 8.5% for SGA, 3.3% for AGA, and 3.1% for LGA infants, with the significance of SGA versus AGA and LGA being bor-
García-Patterson et al 483
Volume 179, Number 2 Am J Obstet Gynecol
Table II. Morbidity in infants of women with gestational diabetes mellitus according to birth weight
Adverse fetal outcome 1* Adverse fetal outcome 2† Low 1-min Apgar score Hypoglycemia Hyperbilirubinemia Respiratory distress Major malformation Hypocalcemia Polycythemia Low 5-min Apgar score Obstetric trauma Minor malformation Mortality
All groups (N = 833)
SGA (n = 61)
AGA (n = 708)
LGA (n = 64)
Significance
24.9 19.7 8.2 9.5 3.2 3.7 2.3 2.7 1.0 0.86 2.3 3.8 0.2
53.2 43.5 20.3 25.4 11.9 8.5 0.0 4.3 1.7 1.67 1.7 1.7 0.0
22.4 17.5 7.3 8.2 2.8 3.3 2.6 2.2 0.9 0.86 2.2 3.8 0.12
20.8 17.4 6.3 9.4 0.0 3.1 1.6 6.3 1.6 0 4.7 7.8 1.6
P < .0001 P < .001 P < .01 P < .0001 P < .01‡ P = .056‡ P = NS‡ P = NS§ P = NS§ P = NSll P = NSll P = NSll P = NSll
NS, Not significant. *Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and death. †Adverse fetal outcome 2 was defined as adverse fetal outcome 1 but with Apgar score at 5 minutes instead of 1 minute. ‡SGA versus AGA and LGA. §SGA and LGA versus AGA. llSGA and AGA versus LGA.
derline (P = .056). We did not find any differences between groups in rates of obstetric trauma, hypocalcemia, polycythemia, minor and major malformations, and perinatal mortality, even after grouping according to the aforementioned criteria. Adverse fetal outcome 1 (defined as ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality) was present in 53.2% of SGA, 22.4% of AGA and 20.8% of LGA infants (P < .0001). Adverse fetal outcome 2 used 5-minute Apgar score and was recorded for 43.5% of SGA, 17.5% of AGA, and 17.4% of LGA infants (P < .001; Table II). The relationship between birth weight category and the rate of any morbidity is displayed in Fig. 1. The rates of preterm birth before 37 weeks of gestation were 25.4% in the SGA group, 4.4% in the AGA group, and 3.1% in the LGA group (P < .001), whereas birth before 34 weeks of gestation did not differ among groups (4.91% in SGA, 1.13% in AGA, and 0% in LGA, P not significant). Because low Apgar score,18 hypoglycemia,19 respiratory distress,20 and hyperbilirubinemia20,21 are more frequent among preterm infants, the observed morbidity in the SGA group could be attributable to the prematurity itself, indicating the need to adjust for this variable. Because morbidity did not differ between the AGA and LGA groups, these groups were analyzed together when performing the adjustment for prematurity (SGA versus non-SGA). Elective delivery because of growth retardation did not contribute significantly to preterm delivery in SGA infants (6.67% in SGA, 0% in AGA, and 0% in LGA, P not significant). After adjustment for preterm birth (Table III), the rate of adverse fetal outcome was higher in the SGA group both when including 1-minute Apgar score (odds ratio 3.08, 95% confidence interval 1.63 to 5.81) and when in-
cluding 5-minute Apgar score (odds ratio 2.48, 95% confidence interval 1.27 to 4.82). The same pattern was observed for 1-minute Apgar scores <7 (odds ratio 2.51, 95% confidence 1.23 to 5.15), with a similar trend for scores <4 (overall rate 1.93%, odds ratio for SGA 3.11, 95% CI 0.924 to 10.5). Hypoglycemia remained associated with low birth weight after adjustment for prematurity (odds ratio 3.17, 95% confidence interval 1.60 to 6.28). When the effect of preterm birth was controlled for, however, hyperbilirubinemia (odds ratio 1.71, 95% confidence interval 0.69 to 4.24) and respiratory distress (odds ratio 1.25, 95% confidence interval 0.42 to 3.71) were no longer associated with low birth weight. In the subgroups of term and preterm infants, birth weight influenced morbidity in the same direction for all outcomes (data not shown), even though statistical significance was reached only for neonatal hypoglycemia in the term subgroup (odds ratio 3.01, 95% confidence interval 1.38 to 6.58). The exception was hyperbilirubinemia, for which the patterns in term and preterm subgroups were divergent. Whereas low birth weight seemed to protect against hyperbilirubinemia in the preterm group (odds ratio 0.34, 95% confidence interval 0.06 to 1.79), it was clearly associated with hyperbilirubinemia in term infants (odds ratio 9.29, 95% confidence interval 3.03 to 28.5). If we limit the analysis to singleton pregnancies, the rates of hypoglycemia (odds ratio 3.06, 95% confidence interval 1.43 to 6.55) and adverse fetal outcome with 1-minute Apgar score (odds ratio 2.20, 95% confidence interval 1.06 to 4.55) were also higher in the SGA group after controlling for prematurity. Comment In this group of infants of women with gestational diabetes mellitus with tight metabolic control according to
484 García-Patterson et al
August 1998 Am J Obstet Gynecol
Fig 1. Rates of adverse fetal outcome 1 among SGA and non-SGA neonates adjusted for gestational age (P < .05). Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality.
Table III. Morbidity outcomes among infants of women with gestational diabetes mellitus under tight metabolic control Morbidity outcomes Adverse fetal outcome 1 Adverse fetal outcome 2† Low 1-min Apgar score Hypoglycemia Hyperbilirubinemia Respiratory distress
Odds ratio
95% Confidence interval
3.08 2.48 2.51 3.17 1.71 1.25
1.63-5.81 1.27-4.82 1.23-5.15 1.60-6.28 0.69-4.24 0.42-3.71
*Adverse fetal outcome 1 was defined as the presence of ≥1 of the following: low 1-minute Apgar score, hypoglycemia, hypocalcemia, polycythemia, respiratory distress, hyperbilirubinemia, obstetric trauma, and mortality. †Adverse fetal outcome 2 was defined as adverse fetal outcome 1 but with Apgar score at 5 minutes instead of 1 minute.
third-trimester hemoglobin A1c a good overall perinatal outcome was achieved. This compares favorably with other reports of women with intensively treated gestational diabetes.4,5 The somewhat high rate of hypoglycemia (9.4%) may be partly attributed to the analytic method (capillary blood, reflectance meter) chosen to allow an early therapy in case of a low blood glucose measurement. Regarding the distribution of morbidity and mortality among the 3 birth weight subgroups, the LGA subgroup did not show a higher rate for any of the analyzed entities. In contrast, SGA neonates after adjustment for preterm delivery demonstrated a higher rate of adverse fetal outcome, low Apgar score in the first minute after
delivery, and hypoglycemia than did either AGA or LGA infants (L-shaped pattern of morbidity). No significant differences could be demonstrated across birth weight categories for malformations, hypocalcemia, polycythemia, obstetric trauma, and mortality. Our results agree with those of Stenninger et al2 in that there is no excess hypoglycemia seen among macrosomic infants and with those of Megía et al4 in that the rates of hypoglycemia, hyperbilirubinemia, respiratory distress, malformations, hypocalcemia, and polycythemia are similar in the macrosomic and nonmacrosomic subgroups. However, Megía et al4 did find an excess of obstetric trauma in the macrosomic infants, which we did not observe. In the series of Langer et al,5 hyperbilirubinemia had an L-shaped pattern, polycythemia had a J-shaped pattern, and neonatal intensive care unit admission had a Ushaped pattern. Taken together, these patterns would indicate that morbidity in the SGA subgroup is at least as important as in the LGA subgroup. It is noteworthy that some of the aforementioned authors did not consider the morbidity in the SGA subgroup.2,4 In these articles, as well as in several others,22 the rate of SGA among infants is not reported. In our opinion, however, the percentage of SGA infants should be included in the outcome analysis of diabetic pregnancies, not only because of its relationship with neonatal morbidity but also because of its association with adult health outcomes such as diabetes mellitus, hypertension, and obesity.23 We have shown that, after adjustment for preterm
Volume 179, Number 2 Am J Obstet Gynecol
birth, SGA neonates of mothers with gestational diabetes mellitus receiving intensive therapy have a greater risk than do those who are AGA or LGA for low 1-minute Apgar score, hypoglycemia, and adverse fetal outcome. Additionally, subgroups of term and preterm infants behave similarly to the pattern that has been described for neonates of nondiabetic mothers. For preterm neonates low birth weight is a risk factor for hypoglycemia19 and a protective factor against hyperbilirubinemia21 but has either no effect or a deleterious effect24 on respiratory distress, with the same trend being observed among the preterm babies in our study. For term neonates, on the other hand, low birth weight is a risk factor for hypoglycemia,19 hyperbilirubinemia,21 low Apgar score,25 and respiratory distress.25 The same pattern was observed in this study, although statistical significance was attained only for hypoglycemia and hyperbilirubinemia. We conclude that in this group of 833 infants born to women with gestational diabetes mellitus managed with an intensified treatment, perinatal outcomes were similar for LGA and AGA neonates but worse, and not attributable to prematurity, for SGA babies.
REFERENCES
1. Jacobson JD, Cousins L. A population-based study of maternal and perinatal outcome in patients with gestational diabetes. Am J Obstet Gynecol 1989;161:981-6. 2. Stenninger E, Schollin J, Aman J. Neonatal macrosomia and hypoglycemia in children of mothers with insulin-treated gestational diabetes mellitus. Acta Paediatr Scand 1991;80:1014-8. 3. Sosenko IR, Kitzmiller JL, Loo SW, Blix P, Rubenstein A, Gabbay KH. The infant of diabetic mother: correlation of increased cord C-peptide levels with macrosomia and hypoglycemia. N Engl J Med 1979;301:859-62. 4. Megía A, Luna R, López A, Grande C, Jáñez M, Pallardo LF. Macrosomía fetal en la gestante diabética. Av Diabetol 1991;4:43-9. 5. Langer O, Levy S, Brustman L, Anyaegbunam A, Merkatz R, Divon M. Glycemic control in gestational diabetes mellitus— how tight is tight enough: small for gestational age versus large for gestational age? Am J Obstet Gynecol 1989;161:646-53. 6. Driscoll SG, Benirschke K, Curtis GW. Neonatal deaths among infants of diabetic mothers. Am J Dis Child 1960;100:818-35. 7. National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979;28:1039-57.
García-Patterson et al 485
8. Food and Nutrition Board. Calorías. In: Raciones dietéticas recomendadas. Barcelona (Spain): Ediciones Consulta SA; 1991. p. 23-37. 9. Durnin JG. Energy requirements of pregnancy. Diabetes 1991;40(Suppl 2):152-6. 10. Weiner CP, Faustich MW, Burns J, Fraser M, Whitaker L, Klugman M. Diagnosis of gestational diabetes by capillary blood samples and a portable reflectance meter: derivation of threshold values and prospective validation. Am J Obstet Gynecol 1987;156:1085-9. 11. Carr S, Coustan DR, Martelly P, Brosco F, Rotondo L. Precision of reflectance meters in screening for gestational diabetes. Obstet Gynecol 1989;73:727-31. 12. Working group on the very low birthweight infant. European Community collaborative study of outcome of pregnancy between 22 and 28 weeks gestation. Lancet 1990;336:782-4. 13. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963;32:793-800. 14. Cornblath M, Reisner SH. Blood glucose in the neonate and its clinical significance. N Engl J Med 1965;273:378-81. 15. Robson AM. Consideraciones generales sobre la asistencia del niño enfermo: tetania hipocalcémica. In: Behrman RE, editor. Tratado de pediatría. Madrid (Spain): Interamericana–McGrawHill; 1992. p. 254-6. 16. Maisels MJ. Bilirubin: on understanding and influencing its metabolism in the newborn infant. Pediatr Clin North Am 1972;19:447-503. 17. Ramamurthy RS, Brans YW. Neonatal polycythemia I. Criteria for diagnosis and treatment. Pediatrics 1981;68:168-73. 18. Catlin EA, Carpenter MW, Brann BS, Mayfield SR, Shaul PW, Goldstein M, et al. The Apgar score revisited: influence of gestational age. J Pediatr 1986;109:865-8. 19. Lubchenco LO, Bard H. Incidence of hypoglycemia in newborn infants classified by birthweight and gestational age. Pediatrics 1971;47:831-8. 20. Wigton TR, Tamura RK, Wickstrom E, Atkins V, Deddish R, Socol ML. Neonatal morbidity after preterm delivery in the presence of documented lung maturity. Am J Obstet Gynecol 1993;169:951-5. 21. Friedman L, Lewis PJ, Clifton P, Bulpitt CJ. Factors influencing the incidence of neonatal jaundice [letter]. BMJ 1978;2:1235-7. 22. Hod M, Merlob P, Friedman S, Schoenfeld A, Ovadia J. Gestational diabetes mellitus: a survey of perinatal complications in the 1980s. Diabetes 1991;40(Suppl 2):74-8. 23. Robinson S, Walton RJ, Clark PM, Barker DJ, Hales CN, Osmond C. The relation of fetal growth to plasma glucose in young men. Diabetologia 1992;35:444-6. 24. Piper JM, Xenakis EM, McFarland M, Elliott BD, Berkus MD, Langer O. Do growth-retarded premature infants have different rates of perinatal morbidity and mortality than appropriately grown premature infants? Obstet Gynecol 1996;87:169-74. 25. Lin YS, Chang FM, Liu CH. Comparison of umbilical blood gas and acid-base status of small-for-dates and normal Chinese newborns. J Formos Med Assoc 1992;91:396-9.