ORIGINAL ARTICLE

Gestational After Delivery Short-term management and long-term risks

1 JOHN L. KITZMILLER, MD need to be addressed and mother-infant 2 LEONA DANG-KILDUFF, RN, CDE bonding encouraged. Cesarean delivery 3 M. MARK TASLIMI, MD and neonatal intensive care unit admis- sions are affected by parental and clini- cian input and institutional policies, as fter the intensified treatment often at least 6–12 weeks to determine their well as determined by glycemic control, required for treating gestational di- status. Many studies over 3 de- obesity, and hypertension during preg- A abetes mellitus (GDM), clinicians cades on all continents of the globe nancy. If medication has been used to may be tempted to relax after delivery of demonstrate the high risk of subsequent treat GDM, it is usually stopped at deliv- the baby. If it is assumed that no further diabetes in this female population. The ery. Encouragement and training for management is needed, an excellent op- degree of this risk is best assessed by healthy nutrition, planned physical activ- portunity to improve the future health glucose tolerance testing. Randomized ity, and weight reduction as needed, con- status of these high-risk women may be controlled trials have proven that sev- tinued cessation of smoking, facilitation lost. There are special concerns for the eral interventions (diet and planned ex- of breastfeeding, and effective planning early postpartum care of women with ercise 30–60 min daily at least 5 days for the next or no more pregnancies are of GDM. Encouragement and facilitation of per week and antidiabetic medications) high importance for all GDM mothers af- exclusive breastfeeding is very important can significantly delay or prevent the ter delivery. Considerations concerning because of the profound short-term as appearance of in the lactation (11,12) and contraception (13) well as long-term health benefits to the women with impaired glucose tolerance in women with GDM are presented else- infant and the reduced risks for subse- (IGT). The high-risk women can also be where in this supplement. quent obesity and glucose intolerance assessed for cardiovascular risk factors, Immediate postpartum persistence of demonstrated in many breastfeeding with appropriate management and fol- at the level of type 2 dia- women. A method of contraception low-up to reduce the risk of coronary betes is uncommon in women diagnosed should be chosen that does not increase heart disease, cardiomyopathy, and with GDM, and is even the risk of glucose intolerance in the stroke. These women should be edu- more unusual. Both can be ruled out by a mother. Some women with GDM will cated to seek specific preconception few fingerstick glucose tests in the first have persisting hyperglycemia in the days consultation before the next pregnancy days after delivery (to rule out diabetes: after delivery that will justify medical to avoid the teratogenic effect of unrec- fasting plasma glucose [FPG] Ͻ126 mg/ management for diabetes and perhaps for ognized diabetes. dl, Ͻ7 mmol/l; casual plasma glucose hypertension, microalbuminuria, and Ͻ200 mg/dl, Ͻ11.1 mmol/l). If diabetes EARLY POSTPARTUM CARE dyslipidemia. Treatment should be main- — is suspected and is confirmed by labora- tained according to the guidelines of the Provision of puerperal obstetrical and tory fasting or casual glucose tests (14), American Diabetes Association and other neonatal care is the first concern after vag- medical nutrition therapy, self-monitoring relevant organizations and adjusted for inal or cesarean delivery of women diag- of blood glucose, and planned physical ac- the needs of lactation. Treatment should nosed with GDM (1). GDM mothers who tivity are continued. The diabetic food be continued in adequate fashion to min- had imperfect glycemic control, obesity, plan should be designed for good glyce- imize risks to the early conceptus if there or hypertension may have an increased is a subsequent planned or unplanned frequency of preterm delivery and post- mic control, effective lactation, and infant pregnancy. partum complications (2–8). Hopefully, health. Consultation with a registered di- Most women with GDM will not a minority of babies will need to be man- etitian is desirable. If type 1 diabetes is have severe hyperglycemia after deliv- aged in the neonatal intensive care unit suspected and confirmed (14), ery. This group should be followed for (9,10), but if they do, parental anxieties therapy is reinstituted. If type 2 diabetes is suspected imme- ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● diately postpartum, the addition of oral From the 1Division of Maternal-Fetal Medicine, Santa Clara County Health System, San Jose, California; the agents can be considered. Glyburide 2California Program, Stanford, California; and the 3Department of Obstetrics and Gynecology, Stanford University Medical School, Stanford, California. (glibenclamide) or glipazide do not ap- Address correspondence and reprint requests to John L. Kitzmiller, MD, Santa Clara Valley Health System, pear in the breast milk of treated women PEP Services, Suite 340, 750 S. Bascom Ave., San Jose, CA 95128. E-mail: [email protected]. (15), and any type of insulin can be used Received for publication 28 March 2006 and accepted in revised form 9 May 2006. if needed during breastfeeding or bottle- This article is based on a presentation at a symposium. The symposium and the publication of this article were made possible by an unrestricted educational grant from LifeScan, Inc., a Johnson & Johnson company. feeding. Three small studies show that met- Abbreviations: CVD, cardiovascular disease; FPG, fasting plasma glucose; GDM, formin is excreted into breast milk with a mellitus; GTT, ; hsCRP, highly sensitive C-reactive protein; IFG, impaired fasting range of milk/plasma ratios of 0.35–0.71, glucose; IGT, impaired glucose tolerance. but with no indication of harmful effects on A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion factors for many substances. the infants (16–18). Larger studies are DOI: 10.2337/dc07-s221 needed to be able to determine that met- © 2007 by the American Diabetes Association. formin therapy is indicated for diabetic

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S225 GDM after delivery

Table 1—Reasons to perform glucose tolerance testing after pregnancies complicated by GDM BMI, weight gain after pregnancy, family 1. The substantial prevalence of glucose abnormalities detected by 3 months postpartum. , fasting and postchal- 2. Abnormal test results identify women at high risk of developing diabetes over the next lenge glucose levels during and after preg- nancy, postpartum and 5–10 years. ␤ 3. Ample clinical trial evidence in women with glucose intolerance that type 2 diabetes can inadequate -cell secretion, and the need be delayed or prevented by lifestyle interventions or modest and perhaps intermittent drug for pharmacological treatment during therapy. pregnancy (24,27–30,36,38,42,46,50– 4. Women with prior GDM and IGT or IFG have CVD risk factors. Interventions may also 52). However, the risk factors are unable reduce subsequent CVD, which is the leading cause of death in both types of diabetes. to predict all cases of subsequent type 2 5. Identification, treatment, and planning pregnancy in women developing diabetes after diabetes: the biggest risk factor is a GDM GDM should reduce subsequent early fetal loss and major congenital malformations. pregnancy. The prevalence of type 1 dia- betes identified 1–10 years after GDM in mostly European studies is 2.3–9.3% (Ta- women during lactation. The use of acar- first is the prevalence of abnormal results ble 2) (40–42,45) and can usually be pre- bose is attractive during lactation because by 3 months postpartum (cited references dicted by detection of ␤-cell–related absorption of the oral drug is quite limited, published since the last International autoantibodies during or after the GDM but there are no studies of the pharmaco- GDM Workshop in 1998: 22–33). Im- pregnancy (26,46,53–58). logical effect of delayed gastrointestinal ab- paired fasting glucose (IFG) is defined as The important third reason to iden- sorption of carbohydrate on the quality of FPG Ն100 or Ն110 mg/dl (Ն5.6 or Ն6.1 tify glucose abnormalities postpartum is lactation. Use of a thiazolidinedione could mmol/l), depending on study and guide- that six randomized clinical trials dem- also be an attractive choice, since the drugs line (14,34,35). IGT is defined as a 75-g onstrate the benefit of several interven- are highly protein-bound “and the large vol- glucose tolerance test (GTT) 2-h plasma tions (diet and exercise 30–60 min ume of distribution of the maternal com- glucose of 140–199 mg/dl (7.8–11.0 daily at least 5 days per week, acarbose, partment should ensure that relatively little mmol/l). Diabetes is defined as repeated metformin, or peroxisome proliferator– crosses into breast milk” (19), but there are FPG Ն126 mg/dl (Ն7 mmol/l) or 2-h glu- activated receptor-␥ agonists: thiazo- no studies as yet of the use of pioglitazone or cose Ն200 mg/dl (Ն11.1 mmol/l) lidinediones, “glitazones”) in delaying rosiglitazone in diabetic lactating women. (14,34,35). The prevalence of isolated or preventing type 2 diabetes in women Finally, the woman with diabetes should be IFG is 3–6%, that of IGT is 7–29%, and with glucose intolerance (59–72). educated about 1) the risk of early fetal loss that of diabetes is 5–14% 4–20 weeks af- These trials are reviewed elsewhere in and major congenital malformations if hy- ter pregnancy in women who were diag- this supplement (73). perglycemia is not controlled before the nosed with GDM and received treatment The fourth reason is that women with next pregnancy, 2) the possibility of preven- during gestation (Table 2). The variance prior GDM have a high frequency of CVD tion of diabetic complications with good in prevalence may depend on the fre- risk factors (30,43,46). Reduction of type control of glucose and blood pressure, and quency of obesity in the sample and dif- 2 diabetes in these women should be of 3) the long-term risks of cardiovascular dis- ferent diagnostic standards for GDM in great health benefit (74–77), since coro- ease (CVD) with type 2 diabetes and means the pregnancy, but does not seem to de- nary heart disease, heart failure, stroke, of its prevention (20,21). Women with pend much on geographic location of the and their associated mortality are com- GDM but without diabetes diagnosed im- study. The majority of recent studies of mon in this population (78). Clinical tri- mediately postpartum should be advised to early postpartum follow-up after GDM als are ongoing to test this hypothesis. have later glucose tolerance testing, to have yield an IGT prevalence of 17–23%. Intensified multifactorial treatment of a prepregnancy consultation before the next The second reason to identify glucose type 2 diabetes already present is of dem- pregnancy, and to request early glucose intolerance after pregnancy is that abnor- onstrated benefit in reducing diabetic and screening in the next pregnancy. mal test results identify women at in- cardiovascular complications (79). Clini- In the absence of obvious diabetes creased risk of developing type 1 diabetes cal trials are also needed to determine the soon after delivery, the timing of de- or especially type 2 diabetes over 1–15 best methods to reduce CVD risk factors layed postpartum glucose tolerance years follow-up (cited references pub- in women with prior GDM and to mea- testing (6–12 weeks or later) may de- lished since the last International GDM sure the impact on long-term health. pend on the length of continuation of Workshop in 1998: 36–46). Systematic The fifth reason to predict or iden- health insurance coverage. Obstetrical reviews of older studies conclude that 35– tify diabetes after GDM, but not least or medical care should continue until 60% of subjects develop type 2 diabetes important, is that women with undiag- delayed postpartum glucose status is by 10–20 years after a GDM pregnancy, at nosed hyperglycemia entering subse- determined and the patient is educated rates much greater than control groups quent pregnancy have high rates of about indicated preventive therapy or who did not have glucose intolerance major congenital malformations in their diabetes/CVD risk reduction. Patient re- during pregnancy (47–49). The higher infants (80–83), which can be reduced ferral may be necessary to continue this rates were in studies of particular ethnic by planning pregnancy and using inten- management. groups in the U.S. Recently, follow-up sified preconception care of diabetes programs elsewhere also have identified (84,85). Unfortunately, for many rea- POSTPARTUM GLUCOSE increasing rates of type 2 diabetes by 5–10 sons, women with type 2 diabetes have TESTING — The rationale for delayed years after GDM (Table 2): 9–43% type 2 been less likely to use preconception postpartum glucose testing (at 6–12 diabetes in Europe (37,40–42,45,46) care (83,86,87), so major public efforts weeks) of women with prior GDM is and 11–21% in Asia (38,43,44). The fre- are required to improve this dangerous based on five sets of facts (Table 1). The quency of type 2 diabetes is influenced by situation.

S226 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Kitzmiller, Dang-Kilduff, and Taslimi

Table 2—Observational studies published since 1998 identifying glucose abnormalities after pregnancies complicated by GDM (short- and longer-term follow-up)

Number Follow-up Percent Percent Percent type 2 Study Region GDM followed after delivery isolated IFG IGT diabetes Short-term follow-up Conway and Langer (22) Texas 179 4–13 weeks 3.3* 16.7 7.8 Ko et al. (23) China 801 6 weeks NA 22.7 13.1 Pallardo et al. (24) Spain 788 3–6 months 5.8* 14.1 5.4 Costa et al. (25) Spain 120 2–12 months 2.5* 10.8 2.5 Bartha et al. (26) Spain 102 3 months NA 6.9 8.8 Aberg et al. (27) Sweden 193 12 months NA 21.8 9.2 Schaefer-Graf et al. (28) California 1,636 1–4 months NA 21.8† 14.1 Jang et al. (29) Korea 311 6–8 weeks 2.3* 23.2 16.7 Pallardo et al. (30)‡ Spain 838 3–6 months 4.8* 10.4 3.5 Agarwal et al. (31) United Arab Emirates 549 4–8 weeks 5.5* 15.3 9.1 Winzer et al. (32) Austria 98 12 months 5.0§ 20.0 15.0 Lin et al. (33) China 127 Ͼ6 weeks NA 29.1 13.4 Current study California 527 6–21 weeks 6.3§ 23.3 4.7 Longer-term follow-up Buchanan et al. (36) California 91 1–2 years NA NA 15.4 Kousta et al. (37) U.K. 192 1–86 months 9.4* 27.1 24.0 Bian et al. (38) China 86 5–10 years NA 7.0 20.9 Linne et al. (39) Sweden 28 15 years NA NA 35 Albareda et al. (40) Spain 352 6 years 7.1* 17.3 11.1 (2.6% type 1 diabetes) Cypryk et al. (41) Poland 193 1–8 years 0.5* 13.5 18.7 (9.3% type 1 diabetes) Lauenborg et al. (42) Denmark 330 6–10 years NA 26.4† 37.0 (3.9% type 1 diabetes) Cho et al. (43) Korea 170 1–5 years NA 25.3 10.6 Cheung and Helmink (44) Australia 102 1–8 years NA 16 29 Hunger-Dathe et al. (45) Germany 173 2–10 years NA 19.1† 9.2 (2.3% type 1 diabetes) Lobner et al. (46) Germany 302 2–11 years NA NA 43.1‡ *1997 American Diabetes Association criteria (34). †Includes IFG by 1997 American Diabetes Association criteria. ‡Some overlap with prior study. §2003 American Diabetes Association criteria (14). ‡May include type 1 diabetes. NA, not available.

The purpose of delayed postpartum CURRENT STUDY: FAILURE meals. All patients received medical nu- glucose testing is to identify any type of OF FPG TO IDENTIFY CASES trition therapy from registered dietitians glucose abnormality present: IFG, IGT, OF IGT OR TYPE 2 and were trained to keep daily food type 1 diabetes, or type 2 diabetes (14). DIABETES 6–21 WEEKS records of carbohydrate intake in their Both isolated IFG and isolated IGT pre- AFTER PREGNANCY IN A own languages. They were taught to use dict (to different degrees) later risks of MULTIETHNIC POPULATION — daily planned physical activity whenever type 2 diabetes and of CVD, and com- In San Jose, CA, we evaluated the yield of possible. When the majority of glucose bined IFG-IGT generally has the greatest postpartum 2-h 75-g GTTs performed in values exceeded 99 mg/dl (5.5 mmol/l) predictive power (30,88–99). There has clinical laboratories in a multiethnic pop- fasting or 129 mg/dl (7.2 mmol/l) post- been debate about the applicability and ulation of women with GDM treated dur- prandial in any given week, medical ther- efficacy of different types of - ing 2000–2003. GDM was diagnosed by apy was instituted with glyburide ing in the postpartum state. It is claimed private clinicians based on a 50-g 1-h glu- (26.6%) or one of several insulin regi- that the greater stability and reproducibil- cose screening test value Ͼ199 mg/dl mens (36.8%). If hyperglycemia ex- ity of FPG compared with GTT suggests (Ͼ11.1 mmol/l) or a 100-g 3-h GTT with ceeded the stated limits using glyburide that FPG would be more easily and widely any two values Ն95 mg/dl fasting, 1-h therapy up to a maximum of 20 mg/day, applied for clinical screening and diagno- 180 mg/dl, 2-h 155 mg/dl, and 3-h 140 patients were changed to insulin therapy sis (14). In the case of women with GDM, mg/dl (5.3, 10.0, 8.6, 7.8 mmol/l, respec- (45.7% of those on glyburide), with dos- the pregnancy has already been the tively) (14). Patients were then referred to age adjusted sequentially as needed. All screening test for “glucose abnormality,” one of two diabetes and pregnancy edu- patients were given laboratory requisi- so what is needed after GDM is a diagnos- cation and treatment centers for coordi- tions before delivery and encouraged to tic test. It is generally agreed that random nated multidisciplinary management go for a postpartum 75-g 2-h GTT at glucose testing is not systematic and that under the supervision of one physician. 6–12 weeks after delivery, with the tim- assays for glycosylated hemoglobin or All patients were trained in daily finger- ing depending on the continuation of proteins are not sensitive to moderate hy- stick capillary self-monitoring of blood their health insurance coverage. perglycemia or glucose intolerance (14). glucose at fasting and 1 hour after main Of 527 women with GDM complet-

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S227 GDM after delivery

Table 3—Ethnicity and postpartum glucose abnormalities on 75-g 2-h GTTs 6–21 weeks after tolerance (109–111), neither breastfeed- delivery in women with prior GDM in San Jose, CA ing status nor use of low-dose combina- tion oral contraceptives influenced the Normal Isolated Isolated Combined Type 2 diagnosis of postpartum glucose abnor- Ethnic group GTT IFG IGT IFG-IGT diabetes Total malities in the present and other studies (105,112–114). Progestin-only prepara- Asian Indian 51 (66.2) 5 9 8 4 77 tions were not used in this population. Far East Asian 48 (51.1) 10 26 6 4 94 In our analysis of women with prior Southeast Asian 105 (68.2) 7 26 10 6 154 GDM, impaired fasting glucose (FPG Hispanic 64 (66.7) 3 15 9 5 96 Ն100 mg/dl, Ն5.6 mmol/l) (14) was not Non-Hispanic white* 78 (73.6) 8 11 3 6 106 sensitive (34%) in identifying IGT and Total 346 (65.7) 33 (6.3) 87 (16.5) 36 (6.8) 25 (4.7) 527 type 2 diabetes on the postpartum GTTs, Data are n or n (%). *Caucasian: European, Russian, or Middle Eastern origin. although of course it identifies its own category of glucose abnormality (Table 6). Regarding type 2 diabetes only, 44% ing the postpartum testing at 5–21 weeks 35.2% in the obese group. The distribu- had FPG Ͻ100 mg/dl (Ͻ5.6 mmol/l) on after delivery, the GTTs were diagnostic tion of postpartum glucose abnormalities the GTT. The lack of sensitivity of FPG of isolated IFG in 6.3%, isolated IGT in in the BMI categories of the represented persisted among the different ethnic 16.5%, combined IFG-IGT in 6.8%, and ethnic groups is presented in Table 4. groups and BMI categories (data not type 2 diabetes in 4.7% (14). Thus, 34.3% Even in Caucasian and Hispanic women, shown). The two-by-two tables show that of the group demonstrated postpartum postpartum glucose abnormalities were lower values of FPG set at Ն95 or Ն90 glucose abnormalities, a rate similar to common (19 and 26%, respectively) in also miss a substantial number of cases those in other recent U.S. reports (22,28). the groups with BMI Ͻ25 kg/m2,al- (51 and 38%, respectively) of IGT ϩ type Only 4 of the 25 women (16%) diagnosed though they were more common in 2 diabetes. All other studies but one (115) with type 2 diabetes had an FPG Ն126 groups with BMI Ն30 kg/m2. also show that postpartum IFG has low mg/dl (Ն7 mmol/l) on the GTT. The re- Postpartum glucose abnormalities sensitivity of predicting “postprandial” sults of the postpartum GTTs in the dif- were found in all GDM treatment groups, glucose intolerance and type 2 diabetes ferent ethnic groups are presented in although there was the expected variation after pregnancy (22,25,30,31,37,41, Table 3 (100). There were just a few in frequency with increasing intensity of 116,117) and at other times in women black/African-American, Native Ameri- treatment required to maintain normo- (93,96,118–121). Therefore, the 75-g can, and Pacific Islander women in this glycemia (Table 5). For women with prior GTT with fasting and 2-h glucose mea- population, and they were excluded from GDM not requiring pharmacotherapy surements is the best diagnostic test to the analysis. It is apparent that postpar- during pregnancy, 20.9% had postpar- identify important glucose abnormalities tum glucose abnormalities were common tum GTT abnormalities (mostly IGT). after pregnancy in women with prior (26.4–48.9%) in all represented ethnic Previous investigators demonstrated GDM. Use of the simpler FPG test alone is groups. many antepartum and postpartum pre- not recommended. Prepregnancy BMI (101,102) was dictors of a higher rate of type 2 diabetes atypical in this population of women with after pregnancy. These predictors include MANAGEMENT OF GDM in that 60.1% had BMI Ͻ25 kg/m2, advanced maternal age, elevated BMI in IGT AFTER PREGNANCY — How 25.6% were overweight (BMI 25–29.9 kg/ some studies, the degree of fasting and should the clinician manage the woman m2), and only 14.3% were obese (BMI postchallenge hyperglycemia during or with prior GDM and IGT identified after Ն30 kg/m2), probably because of the after pregnancy, earlier diagnosis of pregnancy? Certainly weight loss or high proportion of Far East and South- GDM, the need for pharmacotherapy, and weight maintenance medical nutrition eastern Asian women (only 0.9% obese by poor pancreatic ␤-cell compensation for therapy and 30–60 min exercise daily at National Institutes of Health standards). increased insulin resistance (26,28, least 5 days per week should be applied It has been suggested that specific BMI 29,36,40,43,105–108). These predictors (64,76). The 2-h 75-g GTT should be re- classifications should be validated for are not to be denied, but many so-called peated at some interval, since it can revert Asian women in the U.S. (103,104). Post- “low-risk” women with prior GDM also to normal or abnormal “spontaneously.” partum glucose abnormalities were of have postpartum glucose abnormalities, For women with persisting IGT after a similar frequency in all the prepregnancy so we recommend universal testing at good effort of medical nutrition therapy BMI categories: 33.4% with BMI Ͻ25 6–12 weeks or later after pregnancy. Al- and planned physical activity, clinical tri- kg/m2, 36.2% in those overweight, and though lactation may improve glucose als support the clinician adding pharma- cotherapy (73,74,122). Acarbose delays Table 4—Cross-tabulation of pre-pregnancy BMI, ethnic groups, and 6- to 21-week postpar- carbohydrate absorption and helps with tum glucose abnormalities in women with prior GDM in San Jose, CA postprandial glucose control, but side ef- fects limit usage (123). Metformin de- BMI (kg/m2) Asian Indian Far East Asian Southeast Asian Hispanic Caucasian creases hepatic glucose production and lipid oxidation, improves peripheral tis- Ͻ25 51.4 (35) 78.4 (46) 82.5 (31) 28.1 (26) 51.0 (19) sue insulin sensitivity and helps with 25–29.9 32.4 (28) 21.6 (60) 16.1 (40) 41.7 (35) 21.6 (35) weight loss (123,124). Thiazolidinedio- Ն30 16.2 (42) 0 1.5 (0) 30.2 (38) 27.5 (34) nes (glitazones) as peroxisome prolifera- Data are percent of cases in different BMI categories (percent with abnormal glucose tolerance). tor–activated receptor-␥ agonists increase

S228 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Kitzmiller, Dang-Kilduff, and Taslimi

Table 5—Frequency of postpartum glucose abnormalities according to GDM treatment group pared with GDM women with normal (San Jose data) glucose tolerance postpartum. The women with IGT destined to develop type Normal Isolated Type 2 2 diabetes by 36 months also initially Treatment group n GTT (%) IFG (%) IGT (%) diabetes (%) demonstrated reduced HDL cholesterol. LDL cholesterol levels were not different MNT only 192 79.1 1.8 17.2 1.9 postpartum in any of the GDM groups Glyburide 77 62.5 4.7 31.2 1.6 compared with control subjects, but LDL Glyburide Ͼ insulin 64 49.1 8.5 35.6 6.8 subfractions were not reported (136). The Insulin 194 59.3 9.5 23.8 7.4 authors concluded that factors other than MNT, medical nutrition therapy. lipids might contribute to the high prev- alence of cardiovascular morbidity in a insulin sensitivity and may improve lipid CVD events (68). Assays of inflammatory similar cohort of women with prior GDM balance and cardiovascular and renal markers for potential clinical use include followed for 12–18 years in Los Angeles function (72,125–127). Metformin and white blood cells, soluble adhesion mole- (137). A total of 56 former GDM mothers glitazones may help “take the load off” the cules, cytokines (the interleukins and tu- without IFG or IGT studied 5–6 years af- overworked pancreatic ␤-cells (75,128). mor necrosis factor-␣), and acute-phase ter pregnancy in Rhode Island had signif- As noted above, it is important to identify reactants (fibrinogen and highly sensitive icantly increased proportions of subjects and control type 2 diabetes before a sub- C-reactive protein [hsCRP]) (134). C-re- with elevated total cholesterol (39%), el- sequent pregnancy. This may justify con- active protein is a correlate of obesity in evated LDL cholesterol (13%), and sys- Ͼ tinued follow-up in the gynecological women with GDM (135). The American tolic blood pressure 140 mmHg (9%) setting of women with the potential to be- Heart Association/Centers for Disease compared with control subjects with sim- come pregnant, or close collaboration Control Scientific Statement concluded ilar BMI distribution (138). Elevated tri- with other physicians. that class IIA evidence supported use of glycerides and LDL cholesterol levels Published data are less helpful in de- hsCRP as the best inflammatory marker 6–11 years after pregnancy were also ciding management of women with iso- currently available (134). “Other inflam- noted in women with prior GDM com- lated IFG. At least follow-up with delayed matory markers (cytokines, other acute- pared with control subjects with similar repeat GTT testing is justified, since IFG phase reactants) should not be measured glucose and BMI parameters in Boston predicts risk of development of impaired for the determination of coronary risk in (139). Women with prior GDM were glucose tolerance or type 2 diabetes (71) addition to hsCRP.” Measurement of more likely to have elevated triglycerides and perhaps risk of CVD (88,89,96). hsCRP using standardized assays should and low HDL cholesterol than a control be done (in the absence of current infec- group in Denmark (140). Increased in- CARDIOVASCULAR RISKS tion or estrogen/progestogen hormone tramyocellular lipid concentration identi- IN WOMEN WITH PRIOR use) “twice (averaging results), optimally fied IGT in Austrian women with prior GDM — The question arises whether 2 weeks apart, fasting or nonfasting in GDM, compared with a glucose-tolerant and how to evaluate CVD risk markers in metabolically stable patients. If hsCRP control group (141). In Asia, total choles- women with glucose abnormalities per- level is Ͼ10 mg/l, the test should be re- terol, LDL cholesterol, and triglycerides sisting after pregnancies with GDM. In- peated and the patient examined for were significantly higher, and HDL cho- flammatory processes are now known to sources of infection or inflammation.” lesterol was significantly lower in 801 contribute to atherosclerosis (129,130). Otherwise, hsCRP levels are categorized women with prior GDM versus control Research continues on the role of lipopro- as low risk (Ͻ1 mg/l), average risk (1.0– subjects after adjustment for age, BMI, teins, cytokines, oxidative stress, loss of 3.0 mg/l), and high risk (Ͼ3.0 mg/l) and smoking (23), but only triglycerides nitric oxide bioactivity in the vessel wall (134). discriminated between IGT and normal (131), and effects of angiotensin and al- Several investigators have studied lip- glucose tolerance in the women who had dosterone (132,133). Addition of LDL ids in women with previous GDM. Latina GDM (23,43). subfractions (to detect small dense parti- women diagnosed with IGT 6–12 weeks On the other hand, standard lipopro- cles, LDL phenotype B) to the standard after pregnancies with GDM in Los Ange- tein concentrations were not different in lipid profile may help in predicting risk of les showed elevated triglycerides com- women with prior GDM compared with

Table 6—Predictive value of FPG measurements on postpartum oral GTT to identify 2-h glucose abnormalities (San Jose data)

FPG values IGT ϩ type 2 2-h glucose Predictive values of Predictive values of Predictive values of (mg/dl) diabetes normal FPG Ն100 mg/dl FPG Ն95 mg/dl FPG Ն90 mg/dl Ն100 50 33 Sensitivity 0.338 Sensitivity 0.486 Sensitivity 0.623 Ն95 72 66 Specificity 0.913 Specificity 0.826 Specificity 0.646 Ն90 92 134 Ͻ100 98 346 PPV* 0.602 PPV 0.522 PPV 0.417 Ͻ95 76 313 PNV† 0.779 PNV 0.805 PNV 0.814 Ͻ90 56 245 *Predictive value of a positive test; †predictive value of a negative test.

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S229 GDM after delivery control subjects 1–3 years after preg- HDL cholesterol. Inflammatory markers References nancy in Italy (142) and in Spain, despite were not measured (140). Total plasma ho- 1. Kjos SL: Postpartum care of the woman the increased BMI and waist circumfer- mocysteine level was a risk factor for the with diabetes. Clin Obstet Gynecol 43:75– ence (30). In the latter study, women with development of diabetes after GDM in Ko- 86, 2000 IFG had significantly increased odds ra- rea (43). As with lipids, continuing investi- 2. Adams KM, Li H, Nelson RL, Ogburn PL tios for obesity and hypertension than the gation of the CVD risk Jr, Danilenko-Dixon DR: Sequelae of un- women with postpartum IGT (30). Also factors is justified after GDM, to determine recognized gestational diabetes. Am J in Spain, the only lipid abnormality was the likelihood of CVD events and means of Obstet Gynecol 178:1321–1332, 1998 increased VLDL cholesterol levels at a their prevention. 3. Yang X, Hsu-Hage B, Zhang H, Zhang C, 5-year follow-up of 262 women with Zhang Y, Zhang C: Women with impaired prior GDM (143). Similar concentrations CONCLUSIONS — The initial post- glucose tolerance during pregnancy have of plasma lipids were found in white partum management of women with significantly poor pregnancy outcomes. Diabetes Care 25:1619–1624, 2002 nonobese women with prior GDM and in GDM should focus on maternal-infant 4. Hedderson MM, Ferrara A, Sacks DA: control subjects matched for age, BMI, well-being, encouragement and training Gestational diabetes mellitus and lesser and waist-to-hip ratio in a retrospective for healthy nutrition, planned physical degrees of pregnancy hyperglycemia: as- case-control study in Brazil (52) and in a activity, weight reduction as needed, con- sociation with increased risk of sponta- similar study in China (33). More re- tinued smoking cessation, breastfeeding, neous preterm birth. Obstet Gynecol 102: search is needed on lipid abnormalities in and provision of appropriate contracep- 850–856, 2003 women with prior GDM, with or without tion. We conclude that women with prior 5. Myles TD: An expanded description of IFG and IGT, and the relationship to sub- GDM have substantial rates of IFG, IGT, delivery-related maternal morbidity for sequent CVD. and type 2 diabetes after pregnancy best diabetic patients. Obstet Gynecol 101 Recent studies of insulin resistance identified by a 75-g 2-h oral GTT 6–12 (Suppl.):38S, 2003 and components of the “metabolic syn- weeks or later postpartum. Continued 6. Crowther CA, Hiller JE, Moss JR, McPhee AJ, Jeffries WS, Robinson JS, for drome” in women with prior GDM sug- prolonged follow-up is indicated to 1) of- the Australian Carbohydrate Intolerance gest that a chronic systemic inflammatory fer and apply treatment in women with Study in Pregnant Women: Effect of response may be present in glucose- IGT designed to delay or prevent devel- treatment of gestational diabetes melli- intolerant women and may be an early opment of type 2 diabetes, 2) follow tus on pregnancy outcomes. N Engl J Med feature of the cluster of CVD risk factors women with IFG or normal GTT to detect 352:2477–2486, 2005 known as the metabolic syndrome (144– later conversion to IGT or type 2 diabetes 7. Langer O, Yogev Y, Most O, Xenakis 147). A total of 23 women studied 1–10 (more research is needed to better define EMJ: Gestational diabetes: the conse- years after pregnancy with GDM in New the conversion rate in women receiving quences of not treating. Am J Obstet Gy- England had higher mean levels of BMI, appropriate advice on nutrition and phys- necol 192:989–997, 2005 waist circumference, triglycerides, ical activity 30–60 min daily at least 5 8. Saydah SH, Chandra A, Eberhardt MS: hsCRP, and interleukin-6 compared with days per week), and 3) identify diabetes Pregnancy experience among women with and without gestational diabetes in 23 control subjects (148). Insulin sensi- for intensified treatment before a subse- the U.S.: 1995 National Survey of Family tivity was similar to control subjects in quent pregnancy to lower the risk of ma- Growth. Diabetes Care 28:1035–1040, women with prior GDM in Barcelona, but jor congenital malformations in their 2005 insulin secretion was lower and waist cir- infants. 9. Kitzmiller JL, Elixhauser A, Carr S, Ma- cumference was higher, and the propor- During pregnancy women with GDM jor CA, De Veciana M, Dang-Kilduff L, tion with blood pressure Ͼ130/85 mmHg should be educated that glucose intoler- Weschler JM: Assessment of costs and was 42 versus 29% in the control subjects ance may not be temporary, that it can be benefits of management of gestational (143). Average hsCRP and interleukin-6, modified by behavior changes and that diabetes mellitus. Diabetes Care 21 but not tumor necrosis factor-␣, were postpartum testing will be important. (Suppl. 2):B123–B130, 1998 higher in a prior GDM group than in con- Presuming an increased risk for cardio- 10. Svarre JA, Hansen BB, Molsted-Pedersen trol subjects 3 months after delivery in vascular events in glucose-intolerant L: Perinatal complications in women with gestational diabetes mellitus: signif- Austria, with or without direct measures women with prior GDM (151), large icance of a diagnosis early in pregnancy. of insulin resistance (149). In Italy, hsCRP long-term follow-up studies are needed to Acta Obstet Gynecol Scand 80:899–904, and fibrinogen concentrations were sig- identify the frequency and value of CVD 2001 nificantly elevated in women 1–3 years risk markers and to determine if interven- 11. American Dietetic Association: Position after pregnancy with GDM, even exclud- tions (antioxidants, aspirin, behavior of the American Dietetic Association: ing women with IGT (142). Gestational modification, glucose and blood pressure promoting and supporting breastfeed- hyperglycemia predicted a high risk of control, and specific pharmacological ing. J Am Diet Assoc 105:810–818, 2005 later metabolic syndrome after adjust- agents) can reduce the frequency or mor- 12. Gunderson EP: Breastfeeding after gesta- ments for age and prepregnancy BMI in tality of coronary heart disease, heart fail- tional diabetes pregnancy: subsequent another Italian study (150). In the large 4- ure, or stroke in these women. Recent obesity and type 2 diabetes in women to 23-year follow-up program in Den- reports of low rates of postpartum glucose and their offspring. Diabetes Care 30 (Suppl. 2):S161–S168, 2007 mark, the prior GDM group of 481 tolerance testing (152,153) and of life- 13. Damm P, Mathiesen ER, Petersen KR, women had 68% impaired glucose regu- style modification (154–158) in women Kjos S: Contraception after gestational lation, 59% elevated fasting serum insu- with prior GDM show that a dramatic par- diabetes. Diabetes Care 30 (Suppl. 2): lin, 54% central obesity, 28% hypertension, adigm shift in clinical practice is neces- S236–S241, 2007 and 35% dyslipidemia, mainly character- sary to improve the lifelong health of 14. Expert Committee on the Diagnosis and ized by elevated triglycerides and reduced these women. Classification of Diabetes Mellitus: Fol-

S230 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Kitzmiller, Dang-Kilduff, and Taslimi

low-up report on the diagnosis of diabe- factors of developing diabetes mellitus in 38. Bian X, Gao P, Xiong X, Xu H, Qian M, tes mellitus. Diabetes Care 26:3160– women with gestational diabetes. Acta Liu S: Risk factors for development of 3167, 2003 Obstet Gynecol Scand 81:11–16, 2002 diabetes mellitus in women with a his- 15. Feig DS, Briggs GG, Kraemer JM, Am- 28. Schaefer-Graf UM, Buchanan TA, Xiang tory of gestational diabetes mellitus. Chi- brose PJ, Moskovitz DN, Nageotte M, AH, Peters RK, Kjos SL: Clinical predic- nese Med J 113:759–762, 2000 Donat DJ, Padilla G, Wan S, Klein J, Ko- tors for a high risk for the development 39. Linne Y, Barkeling B, Rossner S: Natural ren G: Transfer of glyburide and glipiz- of diabetes mellitus in the early puerpe- course of gestational diabetes mellitus: ide into breast milk. Diabetes Care 28: rium in women with recent gestational long-term follow-up of women in the 1851–1855, 2005 diabetes mellitus. Am J Obstet Gynecol SPAWN study. BJOG 109:1227–1231, 16. Hale TW, Kristensen JH, Hackett LP, 186:751–756, 2002 2002 Kohan R, Ilett KF: Transfer of metformin 29. Jang HC, Yim C-H, Han KO, Yoon H-K, 40. Albareda M, Caballero A, Badell G, into human milk. Diabetologia 45:1509– Han I-K, Kim M-Y, Yang J-H, Cho NH: Piquer S, Ortiz A, de Leiva A, Corcoy R: 1514, 2002 Gestational diabetes mellitus in Korea: Diabetes and abnormal glucose toler- 17. Gardiner SJ, Kirkpatrick CMJ, Begg EJ, prevalence and prediction of glucose in- ance in women with previous gestational Zhang M, Moore MP, Saville DJ: Transfer tolerance at early postpartum. Diabetes diabetes. Diabetes Care 26:1199–1205, of metformin into human milk. Clin Res Clin Pract 61:117–124, 2003 2003 Pharmacol Ther 73:71–77, 2003 30. Pallardo LF, Herranz L, Martin-Vaquero 41. Cypryk K, Czupryniak L, Wilczynski J, 18. Briggs GG, Ambrose PJ, Nageotte MP, P, Garcia-Ingelmo T, Grande C, Janez M: Lewinski A: Diabetes screening after ges- Padilla G, Wan S: Excretion of met- Impaired fasting glucose and impaired tational diabetes mellitus: poor perfor- formin into breast milk and the effect on glucose tolerance in women with prior mance of fasting plasma glucose. Acta nursing infants. Obstet Gynecol 105: gestational diabetes are associated with a Diabetol 41:5–8, 2004 1437–1441, 2005 different cardiovascular profile. Diabetes 42. Lauenborg J, Hansen T, Jensen DM, 19. Merlob P, Levitt O, Stahl B: Oral antihy- Care 26:2318–2322, 2003 Vestergaard H, Molsted-Pedersen L, perglycemic agents during pregnancy 31. Agarwal MM, Punnose J, Dhatt GS: Ges- Hornnes P, Locht H, Pedersen O, Damm and lactation. Pediatr Drugs 4:755–760, tational diabetes: implications of varia- P: Increasing incidence of diabetes after 2002 tion in post-partum follow-up criteria. gestational diabetes: a long-term fol- 20. Creager MA. Luescher TF, Cosentino F, Eur J Obstet Gynecol Reprod Sci 113:149– low-up in a Danish population. Diabetes Beckman JA: Diabetes and vascular dis- 153, 2004 Care 27:1194–1199, 2004 ease: pathophysiology, clinical conse- 32. Winzer C, Pacini G, Tura A, Wagner OF, 43. Cho NH, Lim S, Jang HC, Park HK, quences, and medical therapy: part I. Waldhausel W, Kautsky-Willer A: Metzger BE: Elevated homocysteine as a Circulation 108:1527–1532, 2003 Changes in glucose tolerance in women risk factor for the development of diabe- 21. Luescher TF, Creager MA, Beckman JA, with previous gestational diabetes tes in women with a previous history of Cosentino F: Diabetes and vascular dis- within one year after delivery: the Vien- gestational diabetes mellitus: a 4-year ease: pathophysiology, clinical conse- nese Post-Gestational Diabetes Project. prospective study. Diabetes Care 28: quences, and medical therapy: part II. Diabetologia 47 (Suppl. 1):A358, 2004 2750–2755, 2005 Circulation 108:1655–1661, 2003 33. Lin CH, Wen SF, Wu YH, Huang YY, 44. Cheung NW, Helmink D: Gestational 22. Conway DL, Langer O: Effects of new Huang MJ: The postpartum metabolic diabetes: the significance of persistent criteria for type 2 diabetes on the rate of outcome of women with previous gesta- fasting hyperglycemia for the subse- postpartum glucose intolerance in tional diabetes mellitus. Chang Gung Med quent development of diabetes mellitus. women with gestational diabetes. Am J J 28:794–800, 2005 J Diabetes Complications 20:21–25, 2006 Obstet Gynecol 181:610–614, 1999 34. Expert Committee on the Diagnosis and 45. Hunger-Dathe W, Mosebach N, Samann 23. Ko GTC, Chan JCN, Tsang LWW, Li Classification of Diabetes Mellitus: Re- A, Wolf G, Muller UA: Prevalence of im- C-Y, Cockram CS: Glucose tolerance port of the Expert Committee on the Di- paired glucose tolerance 6 years after and other cardiovascular risk factors in agnosis and Classification of Diabetes gestational diabetes. Exp Clin Endocrinol Chinese women with a history of gesta- Mellitus. Diabetes Care 20:1183–1197, Diabetes 114:11–17, 2006 tional diabetes. AustNZJObstet Gynecol 1997 46. Lobner K, Knopff A, Baumgarten A, Mol- 39:478–483, 1999 35. Alberti KGMM, Zimmet PZ, the WHO lenhauer U, Marienfeld S, Garrido- 24. Pallardo F, Herranz L, Garcia-Ingelmo Consultation: Definition, diagnosis, and Franco M, Bonifacio E, Ziegler A-G: T, Grande C, Martin-Vaquero P, Janez classification of diabetes mellitus and its Predictors of postpartum diabetes in M, Gonzalez A: Early postpartum meta- complications. Part 1. Diagnosis and women with gestational diabetes melli- bolic assessment with prior gestational classification of diabetes mellitus: provi- tus. Diabetes 55:792–797, 2006 diabetes. Diabetes Care 22:1053–1058, sional report of a WHO consultation. 47. Kim C, Newton KM, Knopp RH: Gesta- 1999 Diabet Med 15:539–553, 1998 tional diabetes and the incidence of type 25. Costa A, Carmona F, Martinez-Roman S, 36. Buchanan TA, Xiang AH, Kjos SL, Trigo 2 diabetes: a systematic review. Diabetes Quinto L, Levy I, Conget I: Post-partum E, Lee WP, Peters RK: Antepartum pre- Care 25:1862–1868, 2002 reclassification of glucose tolerance in dictors of the development of type 2 di- 48. Cheung NW, Byth K: Population health women previously diagnosed with ges- abetes in Latino women 11–26 months significance of gestational diabetes. Dia- tational diabetes. Diabet Med 17:595– after pregnancies complicated by gesta- betes Care 26:2005–2009, 2003 598, 2000 tional diabetes. Diabetes 48:2430–2436, 49. Ben-Haroush A, Yogev Y, Hod M: Epide- 26. Bartha JL, Martinez-del-Fresno P, 1999 miology of gestational diabetes mellitus Comino-Delgado R: Postpartum me- 37. Kousta E, Lawrence NJ, Penny A, Mill- and its association with type 2 diabetes. tabolism and autoantibody markers in auer BA, Robinson S, Dornhorst A, de Diabet Med 21:103–113, 2004 women with gestational diabetes mel- Swiet M, Steer PJ, Grenfell A, Mather 50. Kautzky-Willer A, Prager R, Waldhausl litus diagnosed in early pregnancy. HM, Johnston DG, McCarthy MI: Impli- W, Pacini G, Thomaseth K, Wagner OF, Am J Obstet Gynecol 184:965–970, cations of new diagnostic criteria for ab- Ulm M, Streli C, Ludvik B: Pronounced 2001 normal glucose homeostasis in women insulin resistance and inadequate beta- 27. Aberg AEB, Jonsson EK, Eskilsson I, with previous gestational diabetes. Dia- cell secretion characterize lean gesta- Landin-Olsson M, Frid AH: Predictive betes Care 22:933–937, 1999 tional diabetes during and after

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S231 GDM after delivery

pregnancy. Diabetes Care 20:1717– tes mellitus by changes in lifestyle glucose tolerance or impaired fasting 1723, 1997 among subjects with impaired glucose glucose: a randomized controlled trial. 51. Kousta E, Lawrence NJ, Godsland IF, tolerance. N Engl J Med 344:1343–1349, Lancet 368:1096–1105, 2006 Penny A, Anyaoku V, Millauer BA, Cela 2001 72. Xiang AH, Peters RK, Kjos SL, Marro- E, Johnston DG, Robinson S, McCarthy 61. Buchanan T, Xiang A, Peters R, Kjos S, quin A, Goico J, Ochoa C, Kawakubo M, MI: Insulin resistance and beta-cell dys- Marroquin A, Goico J, Ochoa C, Tan S, Buchanan TA: Effect of pioglitazone on function in normoglycemic European Berkowitz K, Hodis H, Azen S: Preserva- pancreatic ␤-cell function and diabetes women with a history of gestational dia- tion of pancreatic ␤-cell function and risk in Hispanic women with prior ges- betes. Clin Endocrinol (Oxf) 59:289–297, prevention of type 2 diabetes by phar- tational diabetes. Diabetes 55:517–522, 2003 macological treatment of insulin resis- 2006 52. Pimenta WP, Calderon IM, Cruz NS, tance in high-risk Hispanic women. 73. Ratner RE: Prevention of type 2 diabetes Santos ML, Aragon FF, Padovani CR: Diabetes 51:2796–2803, 2002 in women with previous gestation diabe- Subclinical abnormalities of glucose me- 62. Chiasson J, Josse R, Gomis R, Hanefield tes. Diabetes Care 30 (Suppl. 2):S242– tabolism in Brazilian women with a his- M, Karasik A, Laakso M: Acarbose for S245, 2007 tory of gestational diabetes mellitus. Acta prevention of type 2 diabetes mellitus: 74. American Diabetes Association and Na- Obstet Gynecol Scand 83:1152–1158, the STOP-NIDDM randomized trial. tional Institute of Diabetes and Digestive 2004 Lancet 359:2072–2077, 2002 and Kidney Diseases: The prevention or 53. Catalano PM, Tyzbir ED, Sims EA: Inci- 63. Diabetes Prevention Program Research delay of type 2 diabetes (Position State- dence and significance of islet cell anti- Group: Reduction in the incidence of ment). Diabetes Care 25:742–749, 2002 bodies in women with previous type 2 diabetes with lifestyle interven- 75. Buchanan TA: Prevention of type 2 dia- gestational diabetes. Diabetes Care 13: tion or metformin. N Engl J Med 346: betes: what is it really? Diabetes Care 26: 478–482, 1990 393–403, 2002 1306–1308, 2003 54. Damm P, Kuhl C, Buschard K, Jakobsen 64. Diabetes Prevention Program (DPP) Re- 76. Klein S, Sheard NF, Pi-Sunyer X, Daly A, BK, Svegaard A, Sodoyez-Goffaux F, search Group: The Diabetes Prevention Wylie-Rosett J, Kulkarni K, Clark NG: Shattock M, Bottazzo GF, Molsted-Ped- Program (DPP): description of lifestyle Weight management through lifestyle ersen L: Prevalence and predictive value intervention. Diabetes Care 25:2165– modification for the prevention and of islet cell antibodies and insulin auto- 2171, 2002 management of type 2 diabetes: ratio- antibodies in women with gestational di- 65. Diabetes Prevention Program Research nale and strategies: a statement of the abetes. Diabet Med 11:558–563, 1994 Group: Effects of withdrawal from met- American Diabetes Association, the 55. Beischer NA, Wein P, Sheedy MT, formin on the development of diabetes North American Association for the Mackay IR, Rowley MJ, Zimmet P: Prev- in the Diabetes Prevention Program. Di- Study of Obesity, and the American So- alence of antibodies to glutamic acid de- abetes Care 26:977–980, 2003 ciety of Clinical Nutrition. Diabetes Care carboxylase in women who have had 66. Xiang AH, Peters RK, Kjos SL, Goico J, 27:2067–2073, 2004 gestational diabetes. Am J Obstet Gynecol Ochoa C, Marroquin A, Tan S, Hodis 77. Tuomilehto J, Wareham N: Glucose 173:1563–1569, 1995 HN, Azen SP, Buchanan TA: Pharmaco- lowering and diabetes prevention: are 56. Mauricio D, Corcoy R, Codina M, Mo- logical treatment of insulin resistance at they the same? (Commentary) Lancet rales J, Balsells M, de Leiva A: Islet cell two different stages in the evolution of 368:1218–1219, 2006 antibodies and beta-cell function in ges- type 2 diabetes: impact on glucose toler- 78. Barrett-Connor E, Giardina EG, Gitt A, tational diabetic women: comparison to ance and beta-cell function. J Clin Endo- Gudat U, Steinberg HO, Tschoepe D: first-degree relatives of type 1 (insulin- crinol Med 89:2846–2851, 2004 Women and heart disease: the role of dependent) diabetic subjects. Diabet 67. Diabetes Prevention Program Research diabetes and hyperglycemia. Arch Intern Med 12:1009–1014, 1995 Group: Prevention of type 2 diabetes Med 164:934–942, 2004 57. Fuchtenbusch M, Ferber K, Standl E, with troglitazone in the Diabetes Preven- 79. Gaede P, Vedel P, Larsen N, Jensen GV, Ziegler AG: Prediction of type 1 diabetes tion Program. Diabetes 54:1150–1156, Parving H-H, Pedersen O: Multifactorial postpartum in patients with gestational 2005 intervention and cardiovascular disease diabetes mellitus by combined islet cell 68. Diabetes Prevention Program Research in patients with type 2 diabetes. N Engl autoantibody screening: a prospective Group: Impact of intensive lifestyle and J Med 348:383–393, 2003 multicenter study. Diabetes 46:1459– metformin therapy on cardiovascular 80. Towner D, Kjos S, Leung B: Congenital 1467, 1997 disease risk factors in the Diabetes Pre- malformations in pregnancies compli- 58. Jarvela IY, Juutinen J, Koskela P, Harti- vention Program. Diabetes Care 28:888– cated by NIDDM. Diabetes Care 11: kainen A-L, Kulmala P, Knip M, Tap- 894, 2005 1446–1451, 1995 anainen JS: Gestational diabetes identifies 69. Diabetes Prevention Program Research 81. Schaefer-Graf UM, Buchanan TA, Xiang women at risk for permanent type 1 and Group: Intensive lifestyle intervention or A: Patterns of congenital anomalies and type 2 diabetes in fertile age: predictive metformin on inflammation and coagu- relationship to initial maternal fasting role of autoantibodies. Diabetes Care 29: lation in participants with impaired glu- glucose levels in pregnancies compli- 607–612, 2006 cose tolerance. Diabetes 54:1566–1572, cated by type 2 and gestational diabetes. 59. Pan X, Li G, Hu Y, Wang J, Yang W, An 2005 Am J Obstet Gynecol 182:313–320, 2000 Z, Hu Z, Lin J, Xia J, Cao H, Liu P, Jiang 70. Diabetes Prevention Program Research 82. Farrell T, Neale L, Cindy T: Congenital X, Jiang Y, Wang J, Zheng H, Zhang H, Group: Role of insulin secretion and sen- anomalies in the offspring of women Bennet P, Howard B: Effects of diet and sitivity in the evolution of type 2 diabetes with type 1, type 2 and gestational dia- exercise in preventing NIDDM in people in the Diabetes Prevention Program: ef- betes. Diabet Med 19:322–326, 2002 with impaired glucose tolerance. Diabe- fects of lifestyle intervention and met- 83. Dunne F, Brydon P, Smith K, Gee H: tes Care 20:537–543, 1997 formin. Diabetes 54:2404–2414, 2005 Pregnancy in women with type 2 diabe- 60. Tuomilehto J, Lindstrom J, Eriksson J, 71. The DREAM (Diabetes Reduction As- tes: 12 years outcome data 1990–2002. Valle T, Hamalainen H, Ilanne-Parikka sessment with Ramipril and Rosiglita- Diabet Med 20:734–738, 2003 P, Keinanen-Kiukaanniemi S, Laakso M, zone Medication) Trial Investigators: 84. Kitzmiller JL, Buchanan TA, Kjos S, Louheranta A, Rastas M, Salminen V, Effect of rosiglitazone on the frequency Combs CA, Ratner RE: Pre-conception Uusitupa M: Prevention of type 2 diabe- of diabetes in patients with impaired care of diabetes, congenital malforma-

S232 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Kitzmiller, Dang-Kilduff, and Taslimi

tions, and spontaneous abortions (Tech- creased levels of coronary heart disease ables as predictors of postpartum glu- nical Review). Diabetes Care 19:514– risk factors: results from the Baltimore cose intolerance. Obstet Gynecol 86: 541, 1996 Longitudinal Study on Aging. Diabetes 97–101, 1995 85. MvElvy SS, Miodovnik M, Rosenn B, 53:2095–2100, 2004 108. Wein P, Beischer NA, Sheedy MT: Stud- Khoury JC, Siddiqi T, Dignan PS, Tsang 96. Heldgaard PE, Olivariust NF, Hinds- ies of postnatal diabetes mellitus in RC: A focused preconceptional and early berger C, Henriksen JE: Impaired fasting women who had gestational diabetes. pregnancy program in women with type glycemia resembles impaired glucose Part 2. Prevalence and predictors of dia- 1 diabetes reduces perinatal mortality tolerance with regard to cardiovascular betes mellitus after delivery. Aust N Z and malformation rates to general pop- risk factors: population-based, cross- Obstet Gynecol 37:420–423, 1997 ulation levels. J Matern Fetal Med 9:14– sectional study of risk factors for cardio- 109. Kjos SL, Henry O, Lee RM, Buchanan 20, 2000 vascular disease. Diabet Med 21:363– TA, Mishell DR Jr: The effect of lactation 86. Cundy T, Gamble G, Townend K, Hen- 370, 2004 on glucose and lipid metabolism in ley PG, MacPherson P, Roberts AB: Peri- 97. Tai ES, Goh ES, Lee JJM, Wong M-S, women with recent gestational diabetes. natal mortality in type 2 diabetes Heng D, Hughes K, Chew SK, Cutter J, Obstet Gynecol 82:451–455, 1993 mellitus. Diabet Med 17:33–39, 2000 Chew W, Gu K, Chia KS, Tan CE: Low- 110. McManus RM, Cunningham I, Watson 87. Feig D, Palda V: Type 2 diabetes in preg- ering the criterion for impaired fasting A, Harker L, Finegood DT: Beta-cell nancy: a growing concern. Lancet 359: glucose: impact on disease prevalence function and visceral fat in lactating 1690–1692, 2002 and associated risk of diabetes and isch- women with a history of gestational dia- 88. Tominaga M, Eguchi H, Manaka H, Iga- emic heart disease. Diabetes Care 27: betes. Metabolism 50:715–719, 2001 rashi K, Kato T, Seikkawa A: Impaired 1728–1734, 2004 111. Diniz JMM, Da Costa THM: Indepen- glucose tolerance is a risk factor for car- 98. Nichols GA, Brown JB: Higher medical dent of body adiposity, breast-feeding diovascular disease, but not impaired care costs accompany impaired fasting has a protective effect on glucose metab- fasting glucose: the Funagata Diabetes glucose. Diabetes Care 28:2223–2229, olism in young adult women. Br J Nutr Study. Diabetes Care 22:920–924, 1999 2005 92:905–912, 2004 89. Davies MJ, Raymond NT, Day JL, Hales 99. Vaccaro O, Riccardi G: Changing the 112. Molsted-Pedersen L, Skouby SO, Damm CN, Burden AC: Impaired glucose toler- definition of impaired fasting glucose: P: Preconception counseling and contra- ance and fasting hyperglycemia have dif- impact on the classification of individu- ception after gestational diabetes. Diabe- ferent characteristics. Diabet Med 17: als and risk definition. Diabetes Care 28: tes 40 (Suppl. 2):147–150, 1991 433–440, 2000 1786–1788, 2005 113. Petersen KR, Skouby SO, Jespersen J: 90. DECODE Study Group: Glucose toler- 100. National Center for Health Statistics and Contraception guidance in women with ance and cardiovascular mortality: com- March of Dimes: Peristats. 2002, http:// pre-existing disturbances in carbohy- parison of fasting and 2-hr diagnostic peristats.modimes.org drate metabolism. Eur J Contracept Re- criteria. Arch Intern Med 161:397–405, 101. National Institutes of Health, National prod Health Care 1:53–59, 1996 2001 Heart, Lung and Blood Institute: Clinical 114. Kjos SL, Peters RK, Xianh A, Thomas 91. Esposito K, Nappo F, Marfella R, guidelines on the identification, evalua- D, Schaefer U, Buchanan TA: Contra- Giugliano G, Giugliano F, Ciotola M, tion and treatment of overweight and ception and the risk of type 2 diabetes Quagliaro L, Ceriello A, Giugliano D: obesity in adults: the evidence report. mellitus in Latina women with prior Inflammatory cytokine concentrations Obes Res 6 (Suppl. 2):S51–S210, 1998 gestational diabetes mellitus. JAMA are acutely increased by hyperglyce- 102. Janssen I, Katzmarzyk PT, Ross R: Body 280:533–538, 1998 mia [IGT] in humans: role of oxidative mass index, waist circumference, and 115. Holt RIG, Goddard JR, Clarke P, stress. Circulation 106:2067–2072, health risk: evidence in support of cur- Coleman MAG: A postnatal fasting 2002 rent National Institutes of Health guide- plasma glucose is useful in determining 92. Hanefeld M, Koehler C, Fuecker K, lines. Arch Intern Med 162:2074–2079, which women with gestational diabetes Henkel E, Schaper F, Temelkova- 2002 should undergo a postnatal oral glucose Kurktschiev T: Insulin secretion and in- 103. Snehalatha C, Viswanathan V, Ram- tolerance test. Diabet Med 20:594–598, sulin sensitivity pattern is different in achandran A: Cutoff values for normal 2003 isolated impaired glucose tolerance and anthropometric variables in Asian In- 116. Cundy T, Ducker L, Wrathall K, Morri- impaired fasting glucose: the Risk Factor dian adults. Diabetes Care 26:1380– son J: Agreement between old and new in Impaired Glucose Tolerance for Ath- 1384, 2003 diagnostic criteria in postpartum testing erosclerosis and Diabetes Study. Diabe- 104. WHO Expert Consultation: Appropriate of women with gestational diabetes. Di- tes Care 26:868–874, 2003 body-mass index for Asian populations abetes Care 21:1579–1580, 1998 93. McNeely MJ, Boyko EJ, Leonetti DL, and its implications for policy and inter- 117. McElduff A, Hitchman R: Fasting plasma Kahn SE, Fujimoto WY: Comparison of vention strategies. Lancet 363:157–163, glucose values alone miss most abnor- a clinical model, the oral glucose toler- 2004 malities of glucose tolerance in the post- ance test, and fasting glucose for predic- 105. Catalano PM, Vargo KM, Bernstein IM, partum. Diabet Med 21:646–651, 2004 tion of type 2 diabetes risk in Japanese Amini SB: Incidence and risk factors as- 118. Pomerleau J, McKeigue PM, Chaturvedi Americans. Diabetes Care 26:758–763, sociated with abnormal postpartum glu- N: Relationship of fasting and post-load 2003 cose tolerance in women with ges- glucose levels to sex and alcohol con- 94. Schianca GPC, Rossi A, Sainaghi PP, tational diabetes. Am J Obstet Gynecol sumption: are American Diabetes Asso- Maduli E, Bartoli E: The significance of 165:914–919, 1991 ciation criteria biased against detection impaired fasting glucose versus im- 106. Metzger BE, Cho NH, Roston SM, Rod- of diabetes in women? Diabetes Care 22: paired glucose tolerance: the importance vany R: Prepregnancy weight and an- 430–433, 1999 of insulin secretion and resistance. Dia- tepartum insulin secretion predict 119. Shaw JE, Zimmet PZ, Hodge AM, de betes Care 26:1333–1337, 2003 glucose tolerance five years after gesta- Courten M, Dowse GK, Chitson P, 95. Blake DR, Meigs JB, Muller DC, Naijar tional diabetes. Diabetes Care 16: Toumilehto J, Alberti KG: Impaired fast- SS, Andres R, Nathan DM: Impaired 1598–1605, 1993 ing glucose: how low should it go? Dia- glucose tolerance, but not impaired 107. Greenberg LR, Moore TR, Murphy H: betes Care 23:34–39, 2000 fasting glucose, is associated with in- Gestational diabetes: antenatal vari- 120. Gomyo M, Sakane N, Kamae I, Sato S,

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S233 GDM after delivery

Suzuki K-I, Tominaga M, Kawazu S, Aldosterone promotes endothelial dys- drome at follow-up in women with and Yoshinaga H, Tsushita K, Sato J, Sato function via prostacyclin independent of without gestational diabetes mellitus in Y, Tsujii S, Yoshida T, Seino Y, Usui T, hypertension. Hypertension 46:29–30, index pregnancy. Metabolism Clin Exper Nanjo K, Hirata M, Kotani K, 2005 54:1115–1121, 2005 Hososako A, Kiyohara Y, Kuzuya H: 134. Pearson TA, Mensah GA, Alexander 144. Kautzky-Willer A, Fasching P, Jilma B, Effects of sex, age and BMI on screen- RW, Anderson JL, Cannon RO, Criqui Waldhausl W, Wagner OF: Persistent el- ing tests for impaired glucose toler- M, Fadl YY, Fortmann SP, Hong Y, My- evation and metabolic dependence of ance. Diabetes Res Clin Pract 64:129– ers GL, Rifai N, Smith SC, Taubert K, circulating E-selectin after delivery in 136, 2004 Tracy RP, Vinicor F: Markers of in- women with gestational diabetes melli- 121. Hanai K, Kiuchi Y, Wasada T: Prevalence flammation and cardiovascular dis- tus. J Clin Endocrinol Metab 82:4117– and progression of impaired fasting glu- ease: application to clinical and public 4121, 1997b cose homeostasis assessed by the differ- health practice: a statement for health- 145. Sriharan M, Reichelt AJ, Opperman ent criteria for IFG in Japanese adults. care professionals from the Centers for MLR, Duncan BB, Mengue SS, Crook Diabetologia 48:799–800, 2005 Disease Control and Prevention and MA, Schmidt MI: Total sialic acid and 122. Padwal R, Majumdar SR, Johnson JA, the American Heart Association. Circu- associated elements of the metabolic Varney J, McAlister FA: A systematic re- lation 107:499–511, 2003 syndrome in women with and without view of drug therapy to delay or prevent 135. Retnarkaran R, Hanley AJG, Raif N, previous gestational diabetes. Diabetes type 2 diabetes. Diabetes Care 28:736– Connelly PW, Sermer M, Zinman B: C- Care 25:1331–1335, 2002 744, 2005 reactive protein and gestational diabetes: 146. Di Cianni G, Lencioni C, Volpe L, Ghio 123. Krentz AJ, Bailey CJ: Oral antidiabetic the central role of maternal obesity. J Clin A, Cuccuru I, Pellegrini G, Benzi L, Mic- agents: current role in type 2 diabetes Endocrinol Metab 88:3507–3512, 2003 coli R, Del Prato S: C-reactive protein mellitus. Drugs 65:385–341, 2005 136. Kjos SL, Buchanan TA, Montoro M, and metabolic syndrome in women with 124. Kirpichnikov DM, McFarlane SI, Sowers Coulson A, Mestman JH: Serum lipids previous gestational diabetes. Diabetes JR: Metformin: an update. Ann Intern within 36 months of delivery in women Metab Res Rev 23:135–140, 2007 Med 137:25–33, 2002 with recent gestational diabetes. Diabetes 147. Farhan S, Winzer C, Tura A, Quehen- 125. St John Sutton M, Rendell M, Dandona 40 (Suppl. 2):142–146, 1991 berger P, Bieglmaier C, Wagner OF, P, Dole JF, Murphy K, Patwardhan R, 137. Mestman JH: Follow-up studies in Huber K, Waldhausl W, Pacini G, Patel J, Freed M: A comparison of the women with gestational diabetes mel- Kautzky-Willer A: Fibrinolytic dysfunc- effects of rosiglitazone and glyburide litus: the experience at Los Angeles tion in insulin-resistant women with on cardiovascular function and glyce- County/University of Southern Cali- previous gestational diabetes. Eur J Clin mic control in patients with type 2 di- fornia Medical Center. In Gestational Invest 36:345–352, 2006 abetes. Diabetes Care 25:2058–2064, Diabetes. Weiss PAM, Coustan DR, 148. Heitritter SM, Solomon CG, Mitchell 2002 Eds. New York, Springer-Verlag, 1987, GF, Skali-Ounis N, Seely EW: Subclini- 126. Lautamaki R, Airaksinen J, Seppanen M, p. 191–198 cal inflammation and vascular dysfunc- Toikka J, Luotolahti M, Ball E, Borra R, 138. Meyers-Seifer CH, Vohr BR: Lipid levels tion in women with previous gestational Harkonene R, Iozza P, Stewart M, Knu- in former gestational diabetic mothers. diabetes mellitus. J Clin Endocrinol Metab uti J, Nuutila P: Rosiglitazone improves Diabetes Care 19:1351–1356, 1996 90:3983–3988, 2005 myocardial glucose uptake in patients 139. Verma A Boney CM, Tucker R, Vohr BR: 149. Winzer C, Wagner O, Festa A, Schneider with type 2 diabetes and coronary artery Insulin resistance syndrome in women B, Roden M, Bancher-Todesca D, Pacini disease: a 16-week randomized, double- with prior history of gestational diabetes G, Funahashi T, Kautsky-Willer A: blind, placebo-controlled study. Diabe- mellitus. J Clin Endocrinol Metab 87: Plasma adiponectin, insulin sensitivity, tes 54:2787–2794, 2005 3227–3235, 2002 and subclinical inflammation in women 127. van Wijk JPH, de Koning EJP, Cabezas 140. Lauenborg J, Mathiesen E, Hansen T, with prior gestational diabetes mellitus. MC, Rabelink TJ: Rosiglitazone im- Glumer C, Jorgensen T, Borch-Johnsen Diabetes Care 27:1721–1727, 2004 proves postprandial triglyceride and free K, Hornnes P, Pedersen O, Damm P: The 150. Bo S, Monge L, Macchetta C, Menato G, fatty acid metabolism in type 2 diabetes. prevalence of the metabolic syndrome in Pinach S, Uberti B, Pagano G: Prior ges- Diabetes Care 28:844–849, 2005 a Danish population of women with pre- tational hyperglycemia: a long-term pre- 128. Stumvoll M, Tataranni PA, Stefan N, vious gestational diabetes mellitus is dictor of the metabolic syndrome. J Vozarova B, Bogardus C: Glucose al- three-fold higher than in the general Endocrinol Invest 27:629–635, 2004 lostasis. Diabetes 52:903–909, 2003 population. J Clin Endocrinol Metab 90: 151. Carpenter MW: Gestational diabetes, 129. Hansson GK: Inflammation, atheroscle- 4004–4010, 2005 pregnancy hypertension, and late vascu- rosis, and coronary artery disease. N Engl 141. Kautzky-Willer A, Krssak M, Winzer C, lar disease. Diabetes Care 30 (Suppl. 2): J Med 352:1685–1695, 2005 Pacini G, Tura A, Farhan S, Wagner O, S246–S250, 2007 130. Wellen KE, Hotamisligil GS: Inflamma- Brabant G, Horn R, Stingl H, Schneider 152. Clark HD, van Walraven C, Code C, tion, stress, and diabetes. J Clin Invest B, Waldhausl W, Roden M: Increased in- Karovitch A, Keely E: Did publication 115:1111–1119, 2005 tramyocellular lipid concentration iden- of a clinical practice guideline recom- 131. Bergholm R, Tiikkainen M, Vehkavaara tifies impaired glucose metabolism in mendation toscreen for type 2 diabetes in S, Tamminen M, Teramo K, Rissanen A, women with previous gestational diabe- women with gestational diabetes change Yki-Jarvinen H: Lowering of LDL choles- tes. Diabetes 52:244–251, 2003 practice? Diabetes Care 26:265–268, terol rather than moderate weight loss 142. DiBenedetto A, Russo GT, Corrado F, 2003 improves endothelium-dependent vaso- DiCesare E, Alessi E, Nicocia G, D’Anna 153. Smirnakis KV, Chasan-Taber L, Wolf M, dilation in obese women with previous R, Cucinotta D: Inflammatory markers Markenson G, Ecker JL, Thadhani R: gestational diabetes. Diabetes Care 26: in women with a recent history of gesta- Postpartum diabetes screening in 1667–1672, 2003 tional diabetes mellitus. J Endocrinol In- women with a history of gestational dia- 132. Thomas WG: Double trouble for angio- vest 28:34–38, 2005 betes. Obstet Gynecol 106:1297–1303, tensin receptors in atherosclerosis. 143. Albareda M, Caballero A, Badell G, 2005 N Engl J Med 352:506–508, 2005 Rodriguez-Espinosa J, Ordonez-Llanos 154. Dornhorst A, Frost G: The potential for 133. Johnson FK, Johnson RA, Durante W: J, de Leiva A, Corcoy R: Metabolic syn- dietary intervention postpartum in

S234 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Kitzmiller, Dang-Kilduff, and Taslimi

women with gestational diabetes. Diabe- 156. Wein P, Beischer N, Harris C, Permezel change after gestational diabetes. Diabe- tes Care 20:1635–1637, 1997 M: A trial of simple versus intensified tes Res Clin Pract 63:67–72, 2004 155. Feig DS, Chen E, Naylor CD: Self-per- dietary modification for prevention of 158. Smith BJ, Cheung NW, Bauman AE, ceived health status of women three to progression to diabetes mellitus in Zehle K, McLean M: Postpartum phys- five years after the diagnosis of gesta- women with impaired glucose tolerance. ical activity and related psychosocial tional diabetes: a survey of cases and AustNZJObstet Gynecol 39:161–165, factors among women with recent ges- matched controls. Am J Obstet Gynecol 1999 tational diabetes mellitus. Diabetes 178:386–393, 1998 157. Stage E, Ronneby H, Damm P: Lifestyle Care 28:2650–2654, 2005

DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 S235