Ferrara, Assiamira MD, PhD*; Kahn, Henry S. MD†; Quesenberry, Charles P. PhD*; Riley, Candice MS*; Hedderson, Monique M. MPH*
Gestational diabetes mellitus (GDM) is defined as carbohydrate intolerance with onset or first recognition during pregnancy1 and is associated with several perinatal complications.2 Women with GDM and their offspring are also at increased risk of developing diabetes later in life.2 Clinical recognition of GDM is important because it may lead to appropriate perinatal management and to the promotion of postpartum diabetes-prevention strategies.
GDM complicates between 4% and 14% of pregnancies in the United States, with prevalences being lower in non-Hispanic whites and higher in African Americans, Hispanics, Asians, and Native Americans.2 National surveys in the United States have shown that, among adults, the prevalence of diabetes (self-reported) increased from 4.9% in 1990 to 7.3% in 2000,3 and the prevalence of diabetes (nongestational, including undiagnosed cases) increased from 8.9% in the period of 1976 to 1980 to 12.3% by 1988–1994.4 However, it is unknown whether these observed increases in diabetes have been mirrored by an increase in GDM.
The aim of the present study was to estimate trends in the cumulative incidence of GDM among a multiethnic cohort of women who delivered live infants or had stillbirths between 1991 and 2000 and were aged 15–49 years at delivery. All the women were members of the Kaiser Permanente Medical Care Program in Northern California (KPMCP-NC).
MATERIALS AND METHODS
The Kaiser Permanente Medical Care Program in Northern California is a large group-practice prepaid health plan that provided comprehensive medical services to approximately 2.5 million members in 1991 and 3 million members in 2000 in a 14-county region. Approximately 30% of the population in the area served by the KPMCP-NC are members. According to the results of the 1990 and 2000 U.S. censuses, the KPMCP-NC membership is representative of the population living in the same geographic area with regard to demographics, ethnicity, and socioeconomic status, except that the KPMCP-NC membership under-represents the very poor and the very wealthy.5,6 Annually, from the years 1991 through 2000, approximately 800,000 KPMCP-NC members were in their childbearing years (women aged 15–49 years).
The cohort of pregnancies and of the GDM cases described below was identified through the KPMCP-NC Gestational Diabetes Registry. This registry was started in 1997 by retrospectively identifying pregnancies and GDM cases within the health plan since 1991. Since 1997, it has been an active surveillance registry that annually identifies pregnancies and pregnancies complicated by GDM. The registry identifies GDM cases from 3 KPMCP-NC automated data sources: hyperglycemia during pregnancy in the laboratory database, hospital discharge diagnoses of GDM in the hospitalization database, and outpatient diagnoses of GDM in the outpatient diagnosis database. For this study, the outpatient diagnosis database was not used for identification of GDM cases because the database was not available for case-finding before 1995.
The KPMCP-NC maintains a complete database of all hospitalizations at any Kaiser hospital, and data are complete within 3 months of patient discharge. Each member of the health plan receives a unique medical record number that is not reissued when an individual leaves the health plan, and when a member leaves and then returns to the health plan, he or she is reassigned the same original medical record number. In addition, when an infant is born, she or he is given a unique medical record in the hospitalization databases consisting of its number and the mother's medical record number, allowing identification of the mother–infant pair.
To identify pregnancies that occurred among women who delivered live infants or had stillbirths between January 1, 1991, and December 31, 2000, we searched the hospitalization database of the KPMCP-NC and the billing claim databases (for members who delivered in a non-Kaiser hospital) for admission and discharge diagnoses (of women aged 15–49 years) related to third-trimester pregnancy conditions, delivery, or stillbirth. We specifically sought pertinent codes for mothers and newborns in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9)7: V27.x (outcomes of delivery); V30–V39 (liveborn infants according to type of birth); 72.x–73.x, 74.0–74.2, and 74.4 (delivery procedures); 644.x (early or threatened labor after 22 weeks); 645.x (prolonged pregnancy); 650 (normal delivery); 658.1–658.3 (delayed delivery after spontaneous or artificial rupture of membranes); 760.x–763.x (maternal causes of perinatal morbidity and mortality), and 764.x–779.x (condition originating in the perinatal period). We excluded uncertain pregnancies (n = 4,065) identified by only one of the hospital discharge diagnoses that did not have a second pregnancy-related hospital discharge diagnosis (ICD-9 codes 640.x–677.x) or could not be matched to a newborn.
The Kaiser Permanente Diabetes Registry of Northern California8 was then searched to identify and exclude pregnancies that occurred in women with recognized chronic (ie, nongestational) diabetes before the index pregnancy. The Diabetes Registry systematically excludes women with GDM. In 1996, the sensitivity of the registry was estimated to be 96% with a 2% false-positive rate.8 Women were considered to have diabetes before a pregnancy if they were identified by the Diabetes Registry at least 9 months before their delivery date.
The women's characteristics, such as age and race–ethnicity, were obtained from the KPMCP-NC computerized hospitalization records that include 7 race–ethnic categories: white, African American, Hispanic, Asian, Native American, other, and unknown. Because of the small number of women in the categories of Native American, other, and unknown, the analyses by race–ethnicity were restricted to the 4 larger race–ethnic groups. The agreement between race–ethnicity recorded in the computerized discharge records and self-reported race–ethnicity recorded in the medical chart was 93%.9
A hierarchical approach was applied to define pregnancies as having been screened for GDM. We first defined pregnancies as having been screened for GDM if a 50-g, 1-hour oral glucose tolerance test was performed; then, if a 100-g, 3-hour or a 75-g, 2-hour oral glucose tolerance test was performed; then, if fasting plasma glucose was measured; and, finally, if 2-hour postprandial or random plasma glucose levels were measured.
Plasma glucose testing and results were obtained from the KPMCP-NC clinical laboratory database, which captures the results of all laboratory tests performed at all KPMCP-NC facilities. All plasma glucose measurements were performed by using the hexokinase method at the KPMCP-NC regional laboratory. This laboratory participates in the College of American Pathologists’ accreditation and monitoring program.
Among the pregnancies screened, GDM was defined according to laboratory-documented hyperglycemia identified during pregnancy, a hospital discharge diagnosis (ICD-9 code 648.8), or both. Hyperglycemia was defined as follows: 1) at least 2 plasma glucose measurements during the 100-g, 3-hour oral glucose tolerance test at or higher than the cutoffs recommended by the American Diabetes Association (ADA)10 and The American College of Obstetricians and Gynecologists (ACOG)11 (fasting, 5.3 mmol/L [95 mg/dL]; 1-hour, 10 mmol/L [180 mg/dL]; 2-hour, 8.6 mmol/L [155 mg/dL]; 3-hour, 7.8 mmol/L [140 mg/dL]); 2) a 2-hour postprandial plasma glucose or random plasma glucose concentration of 11.1 mmol/L or more (200 mg/dL) according to the ADA recommendation10; or 3) a fasting plasma glucose concentration of 7 mmol/L or more (126 mg/dL) or a 2-hour plasma glucose concentration after a 75-g oral glucose tolerance test of 7.8 mmol/L or more (140 mg/dL) according to the World Health Organization (WHO) criteria.12 A hierarchical approach was also applied to define pregnancies complicated by GDM. We first included those pregnancies with an abnormal 100-g, 3-hour oral glucose tolerance test result; then, those pregnancies with an abnormal fasting plasma glucose concentration or an abnormal plasma glucose concentration 2 hours after a 75-g oral glucose tolerance test; and, finally, those pregnancies with an abnormal random or 2-hour postprandial plasma glucose concentration. The remaining pregnancies that did not meet the definition of hyperglycemia were considered to be complicated by GDM if GDM was diagnosed at the time of hospital discharge.
To assess the accuracy in defining GDM based on our laboratory database, we conducted a manual review of the medical charts of 2,545 pregnancies that occurred during the study period and were screened for GDM. Of these pregnancies, according to the laboratory database 2,469 did not have plasma glucose values diagnostic of GDM, and 76 had GDM by the ADA plasma glucose criteria. The agreement between the laboratory database and the chart review data were 99.5% for normal glucose tolerance and 97.4% for the GDM. To assess accuracy of the inpatient database in diagnosis GDM, we manually reviewed the medical records of 145 women who delivered in the year 1991, 1995, or 1999; were screened for GDM; and had a hospital discharge diagnosis of GDM but without laboratory-confirmed hyperglycemia (according to the criteria used in this study). The review of the medical charts revealed that 93% of those women had a physician note of GDM in the inpatient or outpatient charts.
The yearly age-specific proportion of pregnancies screened and the yearly race–ethnic-specific and age-adjusted proportion of pregnancies screened were calculated. Women with more than 1 pregnancy during the study period were not eliminated from analyses after the first pregnancy unless chronic diabetes between pregnancies was confirmed by the Diabetes Registry. Similarly, women who had a pregnancy complicated by GDM contributed to analyses in the years when subsequent pregnancies (with or without GDM) occurred unless chronic diabetes made them ineligible.
Among the screened pregnancies, we calculated the yearly crude cumulative incidence of GDM (defined as the percentage of screened pregnancies in which GDM was recognized during each of the years of the study period) by age and ethnic groups. We estimated the cumulative incidence of GDM because we calculated the proportion of pregnancies with GDM among pregnancies that delivered or had a stillbirth for a given year and, therefore, women were at risk for GDM during the entire pregnancy rather a specific point in time during pregnancy.13 The yearly age-adjusted and age- and race–ethnicity-adjusted cumulative incidence of GDM and the 95% confidence intervals were calculated by the direct method, in which the age and race–ethnicity distribution of the entire study population was used as the standard. SAS 6.11 (SAS Institute, Inc, Cary, NC) was used for all analyses. This study was approved by the human subjects committee of the Kaiser Foundation Research Institute.
We identified 309,440 pregnancies occurring between the years 1991 and 2000 among members of the KPMCP-NC who were aged 15–49 years and delivered live infants or had stillbirths. We excluded 1,614 pregnancies that occurred in women who had recognized diabetes mellitus before pregnancy, leaving 307,826 pregnancies eligible for our study. Among these pregnancies, 267,051 (86.8%) were screened for GDM. Among the screened pregnancies, 98.2% were screened by use of a 50-g, 1-hour oral glucose tolerance test; 0.5% had a 3-hour oral glucose tolerance test that was not preceded by a 50-g, 1-hour oral glucose tolerance test; 0.05% received a 75-g, 2-hour oral glucose tolerance test; 0.2% had a fasting glucose test; 0.05% had a 2-hour postprandial plasma glucose test; and 1% had a random plasma glucose test.
The percentage of pregnancies screened for GDM increased from 1991 (64.8%) to 1996 (95.2%) and then leveled off from 1996 to 2000 (92.8%). Overall, pregnancies that occurred in women aged 25 years and over were more likely to be screened than pregnancies that occurred in women aged 25 years or younger (Figure 1). Also, pregnancies that occurred in white women were more likely to be screened than pregnancies that occurred in other race–ethnic groups (Figure 2). The relative differences in screening activity by age did not change over time, whereas the relative differences in screening activity by race–ethnicity decreased over time.
All subsequent analyses were based on the 267,051 screened pregnancies. These pregnancies occurred among 211,353 women because 48,682 had more than 1 pregnancy during the 10-year study period. The age of the screened women increased slightly from 1991 (mean [standard deviation] 28.2 [5.7]) to 2000 (28.8 ). The proportion of women from U.S. minority race–ethnic groups increased markedly from 1991 to 2000. The race–ethnicity distribution (1991 compared with 2000) was as follows: 60.7% and 43.4% white, 14.5% and 15.4% Asian, 14.1% and 24.9% Hispanic, 6.7% and 6.7% African American, 2% and 4.4% other, and 2% and 5.2% unknown.
A total of 14,175 screened pregnancies were defined as being complicated by GDM on the basis of hyperglycemia documented in the KPMCP-NC laboratory database. Of these, 95.2% were defined on the basis of an abnormal 100-g, 3-hour oral glucose tolerance test results; 3.5% on the basis of an abnormal fasting plasma glucose concentration or an abnormal plasma glucose concentration 2 hours after a 75-g oral glucose tolerance test; and 1.3% on the basis of an abnormal 2-hour postprandial or random plasma glucose concentration. An additional 2,743 screened pregnancies were defined as being complicated by GDM because of the presence of a hospital discharge diagnosis. Plasma glucose values in the laboratory database for these pregnancies were not compatible with GDM according to the diagnostic criteria used in this study.
The age- and race–ethnicity-adjusted yearly cumulative incidence of GDM among screened pregnancies is shown in Figure 3. The cumulative incidence of GDM defined by hyperglycemia, a hospital discharge diagnosis, or both increased steadily from 5.1% in 1991 to 7.4% in 1997 and then leveled off through 2000 (6.9%). The cumulative incidence of GDM defined by hyperglycemia (with or without a hospital discharge diagnosis) paralleled the increase observed for GDM defined by hyperglycemia, a hospital discharge diagnosis, or both: it increased from 3.7% in 1991 to 6.6% in 1997 and leveled off through 2000 (6.2%). The cumulative incidence of GDM defined by a hospital discharge diagnosis without hyperglycemia decreased from 1.4% in 1991 to 0.7% in 2000.
Because the observed increase in the cumulative incidence of GDM was not the result of an increase in the number of cases identified only by hospital discharge diagnosis and appeared to be the result of an increase in the proportion of pregnancies complicated by hyperglycemia, we report data on GDM defined by hyperglycemia, a hospital discharge diagnosis, or both in analyses stratified by age and ethnicity (Table 1). The cumulative incidence of GDM among screened pregnancies increased in all age and in all race–ethnicity groups, with the highest proportional increase in the youngest group and in white women. However, the absolute GDM cumulative incidence in all years was higher in older women and in Asians. The cumulative incidence of GDM leveled off in 1998–2000 in each race–ethnicity group except African Americans, in whom incidence continued to increase through 2000.
To test whether the observed increase in GDM was a consequence of including in the analyses repeated pregnancies or recurrent GDM cases, we excluded the 55,698 recurrent pregnancies and repeated the analyses among the 211,353 pregnancies that first occurred during the study period. A similar increase in the age- and race–ethnicity-adjusted cumulative incidence of GDM was observed: the age- and race–ethnicity- adjusted cumulative incidence of GDM among screened pregnancies increased from 5% in 1991 to 7.6% in 1997 and then leveled off through 2000 (7.1%).
We sought evidence for a trend in the severity of diagnosed GDM by examining the annual distributions of screening plasma glucose values obtained 1 hour after a 50-g oral glucose tolerance test among women with GDM. During our decade of observation, the 75th percentile of this screening glucose value increased slightly over time (182 mg/dL in 1991 and 185 mg/dL in 2000). We found less change in the 25th percentile of screening plasma glucose value (149 mg/dL in 1991 and 151 mg/dL in 2000).
In the cohort studied, the cumulative incidence of GDM among screened pregnancies increased from 1991 to 1997 and leveled off through 2000. The overall increase for the decade was approximately 35%. This increment is similar in magnitude to increases in the national prevalence of nongestational diabetes described for recent time intervals.4,14
The observed increase in cumulative incidence of GDM was not caused by the increasing proportion of screened pregnancies in race–ethnicity groups at higher risk in the later years because similar increases in GDM cumulative incidence were observed in all age and race–ethnic groups. However, the observed slight increase in the 75th percentile of screening plasma glucose value may indicate that there was a small increase in the proportion of GDM cases that were relatively severe. The observed increase was also not the result of the inclusion of milder cases of GDM over time (a theoretical consequence of increased screening activity) because the 25th percentile plasma glucose screening value obtained 1 hour after a 50-g oral glucose tolerance test did not decrease between 1991 and 2000. Furthermore, the increase was not caused by a change in physicians’ inclination over time to diagnose GDM at lower levels of hyperglycemia. When we assessed separately the yearly cumulative incidence of GDM defined by hyperglycemia and by hospital discharge diagnosis in pregnancies with normal glycemia, only the incidence of GDM by hyperglycemia increased over time.
In the results presented above, we defined GDM according to the criteria recommended by the ADA10 in 2000 and ACOG11 in 2001 or by the criteria issued by the WHO12 in 1999. In an alternative analysis, we used higher plasma glucose threshold for GDM diagnosis, such as those recommended by the National Diabetes Data Group15 and used by the ADA until 199916 or the higher WHO plasma glucose cutoff17 recommended until 1998. Similar trends in GDM incidence were observed by using the older and higher plasma glucose cutoff for GDM diagnosis (cumulative GDM incidence increased from 3.5% in 1991 to 5.1% in 2000). Therefore, the observed increase in the incidence of GDM was not caused by an increase in the proportion of pregnancies complicated by milder hyperglycemia.
Unfortunately, during the study period, our clinical databases did not include data on body mass index (BMI). Therefore, we were not able to assess whether the observed increase in GDM cumulative incidence would be explained by a concomitant increase in the proportion of women who were obese before pregnancy. Although we do not have data on obesity for our study cohort, it is possible that the observed increase in GDM cumulative incidence was partially the result of an increase in obesity over time because being overweight before pregnancy is one of the major risk factors for GDM.18 A national survey found that among U.S. women aged 18–49 years, the prevalence of self-reported obesity (defined as BMI of 30 kg/m2 or more) increased from about 11% in 1991 to about 17% in 1998.19 Similarly, preliminary data from a random sample of women aged 20–49 years who were members of the KPMCP-NC and responded to a mail survey in 1990 or 1999 showed an increase in obesity (defined as BMI of 30 kg/m2 or more) from 15% in 1990 to 19% in 1999.
Whatever the underlying reason of this observed increase in GDM cumulative incidence, the findings indicate that the health care system is faced with an increase in GDM. Therefore, this pregnancy complication will require increased resources. Women with a history of GDM are at increased risk of developing chronic diabetes,20 and recent clinical trials have shown that health behaviors, such as diet and physical activity, prevent or delay the onset of diabetes.21,22 Therefore, clinicians will increasingly have to promote plasma glucose testing and improve health behaviors at the postpartum visits of women who had GDM.
Gestational diabetes mellitus may play a crucial role in the increasing prevalence of diabetes and obesity. Approximately 50% of women with GDM are expected to develop non–insulin-dependent diabetes mellitus within 5 years of the index pregnancy.20 In 1980, Freinkel23 hypothesized that the human hyperglycemic intrauterine environment may have long-term effects. Offspring of rats with mild diabetes induced by streptozocin have reduced glucose tolerance during pregnancy, and their offspring (third generation) show inappropriate β-cell response during late pregnancy.24,25 Recent evidence in humans supports Freinkel's23 hypothesis. Infants of women with GDM or diabetes are at increased risk of developing obesity, impaired glucose tolerance, and diabetes as children or young adults,26–28 and the increased risk may be independent of genetic factors.29
Some strengths of the present study are the high proportion of pregnancies screened and the nearly uniform procedure used to screen and diagnose GDM as recommended by the ADA10 and ACOG.11 An additional strength is the availability of plasma glucose measurements from the KPMCP-NC laboratory database, which allowed us to rely on a standard definition of GDM. Our study also has some limitations. It is possible that the inclusion of the 2,743 pregnancies identified only on the basis of a hospital discharge diagnosis resulted in an overestimate of the yearly cumulative incidence of GDM. However, the inclusion of these cases did not affect the trend in GDM incidence. Moreover, 99.6% of those pregnancies had plasma glucose concentrations that, although not diagnostic of GDM by conventional criteria, have been shown to be associated with perinatal complications (such as one abnormal value on the diagnostic 100-g, 3-hour oral glucose tolerance test or plasma glucose screening concentration of 7.8 mmol/L or more [140 mg/dL]).30,31
A previous report from Australia32 compared the incidence of GDM in one large maternity hospital between 1979 and 1983 and 1984 and 1988. The authors demonstrated a doubling in the GDM incidence that appeared to apply similarly to mothers from many different ethnic backgrounds. However, they did not adjust for the changing age distribution among the pregnant women. By contrast, we adjusted for age, and within each race–ethnicity group, an increase in cumulative incidence of GDM was observed. A true increase in the incidence of GDM, aside from its adverse consequences for infants in the newborn period, might reflect or contribute to the ongoing pattern of increasing diabetes and obesity. The possible long-term effects on the immediate offspring and subsequent children of the female infants will not be known for decades. Therefore, coordinated efforts are required to alter these trends in GDM and to prevent chronic diabetes in GDM patients and their offspring.
1. Metzger BE. Summary and recommendations of the Third International Workshop: Conference on Gestational Diabetes Mellitus. Diabetes 1991;40(suppl 2):197–201.
2. Jovanovic L, Pettitt DJ. Gestational diabetes mellitus [review]. JAMA 2001;286:2516–8.
3. Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing epidemics of obesity and diabetes in the United States. JAMA. 2001;286:1195–200.
4. Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults: The Third National Health and Nutrition Examination Survey, 1988–1994. Diabetes Care 1998;21:518–24.
5. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation and application of a census-based methodology. Am J Public Health 1992;82:703–10.
6. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370–5.
7. International Classification of Diseases, 9th revision. Clinical Modification, 5th ed, color coded. 2000. Los Angeles (CA): PMIC; c1999.
8. Selby JV, Ray GT, Zhang D, Colby CJ. Excess costs of medical care for patients with diabetes in a managed care population. Diabetes Care 1997;20:1396–402.
9. Ferrara A, Hedderson MM, Quesenberry CP, Selby JV. Prevalence of gestational diabetes mellitus detected by the National Diabetes Data Group or the carpenter and coustan plasma glucose thresholds. Diabetes Care 2002;25:1625–30.
10. American Diabetes Association. Gestational diabetes mellitus. Diabetes Care 2000;23(suppl 1):S77–9.
11. Gestational Diabetes. ACOG Practice Bulletin No. 30 (replaces Technical Bulletin Number 200, December 1994). American College of Obstetricians and Gynecologists. Obstet Gynecol 2001;98:525–38.
12. World Health Organization. Definition, diagnosis and classification of diabetes mellitus and its complication. Report of a WHO consultation. Part 1: Diagnosis and classification of diabetes mellitus. WHO/NCD/NCS/99.2, 1–59. 1999. Geneva, Switzerland: World Health Organization; 1999.
13. Savitz DA, Hertz-Picciotto I, Poole C, Olshan AF. Epidemiologic measures of the course and outcome of pregnancy [review]. Epidemiol Rev 2002;24:91–101.
14. Mokdad AH, Ford ES, Bowman BA, Nelson DE, Engelgau MM, Vinicor F, et al. Diabetes trends in the U.S.: 1990–1998. Diabetes Care 2000;23:1278–83.
15. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance: National Diabetes Data Group. Diabetes 1979;28:1039–57.
16. American Diabetes Association. Gestational diabetes mellitus. Diabetes Care 1998;21:S60–1.
17. Diabetes mellitus. Report of a WHO Study Group [review]. World Health Organ Tech Rep Ser 1985;727:1–113.
18. Coustan DR. Gestational diabetes. In: Harris MI, editor. Diabetes in America. Bethesda (MD): National Institutes of Health; 1995. p. 703–16.
19. Mokdad AH, Serdula MK, Dietz WH, Bowman BA, Marks JS, Koplan JP. The spread of the obesity epidemic in the United States, 1991–1998. JAMA 1999;282:1519–22.
20. Kim C, Newton KM, Knopp RH. Gestational Diabetes and the Incidence of type 2 diabetes: a systematic review [review]. Diabetes Care 2002;25:1862–8.
21. Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–50.
22. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403.
23. Freinkel N. Banting Lecture 1980. Of pregnancy and progeny [review]. Diabetes 1980;29:1023–35.
24. Logotnetopoulos J. Islet cell regeneration and neogenesis. In: Steiner D, Freinkel N, editors. The endocrine pancreas handbook of physiology. Series. Baltimore, (MD): American Physiological Society; 1972. p. 67–76.
25. Hellerstrom C. Growth pattern of pancreatic islets in animals. In: Volk BW, Wellman KF, editors. The diabetic pancreas. New York (NY): Plenum Press; 1977. p. 61–97.
26. Pettitt DJ, Baird HR, Aleck KA, Bennett PH, Knowler WC. Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 1983;308:242–45.
27. Silverman BL, Rizzo TA, Cho NH, Metzger BE. Long-term effects of the intrauterine environment: The Northwestern University Diabetes in Pregnancy Center. Diabetes Care 1998;21(suppl 2):B142–9.
28. Pettitt DJ, Aleck KA, Baird HR, Carraher MJ, Bennett PH, Knowler WC. Congenital susceptibility to NIDDM: role of intrauterine environment. Diabetes 1988;37:622–8.
29. Dabelea D, Hanson RL, Lindsay RS, Pettitt DJ, Imperatore G, Gabir MM, et al. Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity: a study of discordant sibships. Diabetes 2000;49:2208–11.
30. Langer O, Brustman L, Anyaegbunam A, Mazze R. The significance of one abnormal glucose tolerance test value on adverse outcome in pregnancy. Am J Obstet Gynecol 1987;157:758–63.
31. Leikin EL, Jenkins JH, Pomerantz GA, Klein L. Abnormal glucose screening tests in pregnancy: a risk factor for fetal macrosomia. Obstet Gynecol 1987;69:570–3.
32. Beischer NA, Oats JN, Henry OA, Sheedy MT, Walstab JE. Incidence and severity of gestational diabetes mellitus according to country of birth in women living in Australia. Diabetes 1991;40(suppl 2):35–8.