Metabolic Culprits in ObesePregnancies and GestationalDiabetes Mellitus: Big Babies, BigTwists, Big PictureThe 2018 Norbert Freinkel Award LectureDiabetes Care 2019;42:718–726 | https://doi.org/10.2337/dci18-0048

Pregnancy has been equated to a “stress test” in which placental hormones andgrowth factors expose a mother’s predisposition toward metabolic disease,unleashing her previously occult insulin resistance (IR), mild b-cell dysfunction,and glucose and lipid surplus due to the formidable forces of pregnancy-induced IR.Althoughpregnancy-induced IR is intended toassureadequatenutrition to the fetusand placenta, in mothers with obesity, metabolic syndrome, or those who developgestational diabetesmellitus, thisovernutrition to the fetus carries a lifetime risk forincreased metabolic disease. Norbert Freinkel, nearly 40 years ago, coined thisexcess intrauterine nutrient exposure and subsequent offspring developmental risk“fuel-mediated teratogenesis,” not limited to only excess maternal glucose. Ourattempts to better elucidate the causes andmechanisms behind this double-edgedIR of pregnancy, to metabolically characterize the intrauterine environment thatresults in changes in newborn body composition and later childhood obesity risk,and to examine potential therapeutic approaches that might target maternalmetabolism are the focus of this article. Rapidly advancing technologies ingenomics, proteomics, and metabolomics offer us innovative approaches tointerrogate thesemetabolic processes in themother, hermicrobiome, theplacenta,andheroffspring that contribute toaphenotypeat risk for futuremetabolic disease.If we are successful in our efforts, the researcher, endocrinologist, obstetrician, andhealth care provider fortunate enough to care for pregnant women have theunique opportunity to positively impact health outcomes not only in the shortterm but in the long run, not just in one life but in twodand possibly, for the nextgeneration.

The power of pregnancy offers the clinician and scientist a unique opportunity tomeaningfully affect the outcome of two individuals. Applying observations from theimmense field of medicine to optimize the outcome of both mother and her infant,humbled by a placenta that will ultimately determine the exposure to the fetus, is anextraordinary challenge and opportunity. Trained as an internist and inspired by twoprofessors ofmedicine whowere both pioneers in obstetric medicine, I spentmy first12yearspursuingclinical research in thromboembolism inpregnancy, the leadingcaseofmaternalmortality at the time. Thefirstmajor twist alongmy career pathwasmadepossible by two professors of endocrinology who led me to appreciate the profoundendocrinologic changes remodelingmaternal physiology into an insulin-resistant (IR)

Divisions of Endocrinology, Metabolism and Di-abetes and Maternal Fetal Medicine, Universityof Colorado Anschutz Medical Campus, Aurora,CO

Corresponding author: Linda A. Barbour, lynn.barbour@ucdenver.edu

The 2018 Norbert Freinkel Award Lecture waspresented at the American Diabetes Associa-tion’s 78th Scientific Sessions, Orlando, FL,23 June 2018.

© 2019 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.

Linda A. Barbour

718 Diabetes Care Volume 42, May 2019

ADAAWARDLECT

URE

state. This transformation of maternalmetabolism and its effect on pregnanciescomplicated by obesity and gestationaldiabetes mellitus (GDM) became my re-search focus over the next 18 years andis the subject of this article.This IR state, not only characterized by

hyperglycemia, unearths the alreadypresent propensity for mothers to de-velop metabolic disease and may alsopromote the future development of met-abolic disease in her offspring (1,2). Infact, it was indeed an endocrinologist,Norbert Freinkel, who delivered his sem-inal 1980 Banting lecture (3) the year Istarted medical school, “Of Pregnancyand Progeny,” who is a founding fatherof the fetal overnutrition hypothesis.In essence, Dr. Freinkel transformedour thinking about the intrauterine en-vironment in mothers with diabetes,overnutrition, and obesity and the fu-ture health implications to the off-spring. His described that “fuel-mediatedteratogenesis”dnot limited to only mater-nal glucose but inclusive of maternal lipidsand amino acidsdcould lead to “alter-ations occurring subsequent to organo-genesis during the differentiation andproliferation of fetal cells,” which “couldcause long-range effects upon behav-ioral, anthropometric, and metabolicfunctions” (3). The work that followsis a direct result of many inspirationalcollaborators, especially Drs. Jacob (Jed)Friedman and Teresa (Teri) Hernandez(University of Colorado School of Med-icine), aiming to shed additional insightson the 1) causes and mechanisms thatunderlie the IR of normal pregnancy,obesity, and GDM; 2) metabolic char-acterization of the intrauterine meta-bolic environment that contributes toexcess infant adiposity; and 3) potentialtargets and timing of pregnancy inter-ventions to promote healthy short- andlong-term outcomes for both motherand child.

LESSONS LEARNED ON IR FROM AMOUSE WITH TOO MUCHPLACENTAL GROWTH HORMONE

Endocrinologists spend much of theirclinical time attempting to attenuateIR in their patients with obesity, meta-bolic syndrome, or diabetes. Yet, in thecontext of pregnancy, IR is a criticalnormal pathophysiologic mechanism toensure adequate nutrient substrateprovision from mother to a growing fetus

(4). Unfortunately, when IR in skeletalmuscle, adipose tissue, or liver and/orsome degree of compromised b-cell re-serve predates the pregnancy, often seenin women with obesity, metabolic syn-drome, and in women who developGDM, the increasing IR demands ofnormal pregnancy place further burdenon the maternal b-cell to step up se-cretion. The inability to meet this de-mand results in excess glucose, lipid,and amino acid exposure to the fetal-placental unit (5,6). This often leads tofetal hyperinsulinemia, which can occurby 16 weeks’ gestation, enhanced fe-tal growth, and an overabundance ofmaternal substrate for excess fetal fataccretion (7).

Textbooks have told us that humanplacental lactogen (hPL) was the primarycause of this physiologic IR of pregnancy.Yet the data supporting this causationwere limited as hPL has both insulin andanti-insulin effects and may actually bemore important to stimulate the growthof maternal pancreatic islets and increaselipolysis and free fatty acid (FFA) liber-ation from adipocytes as an energy sub-strate (8). The placenta also makes agrowth hormone (GH) variant, similarto pituitary GH, that has been knownto increase IR in pathologic states such asacromegaly (8). Human placental growthhormone (hPGH) is a product of thehuman GH variant gene that differsfrom pituitary GH by 13 amino acids.Its secretion is tonic, increases six- toeightfold throughout gestation, and vir-tually replaces pituitary GH in the ma-ternal circulation by;20 weeks (8,9). Weset out to prove whether hPGH, neverbefore characterized by its effect oninsulin action, also had the capacity tocause IR. Dr. Andrzej Bartke (SouthernIllinois University School of Medicine)was generous to share a transgenicmouse that expressed hPGH at levelssimilar to the third trimester of preg-nancy. Simple glucose tolerance testingrevealed that fasting insulin levels were;4 times higher than in control mice and;7 times higher 30 min after glucosestimulation, confirming pronounced IR.We further showed amarked decrease ininsulin sensitivity given that an insulininjection into these hPGH transgenic miceresulted in no significant decline in glu-cose disposal compared with wild-typemice, who demonstrated .65% declinein glucose levels (10).

How hPGH caused IR at a cellular levelwas our next puzzle. By focusing on theskeletalmuscle of these transgenic hPGHmice, we discovered an unforeseen twistthat hPGH selectively increased theexpression of the p85 regulatory unitof phosphatidlylinositol 3-kinase (PI3-kinase), a key effector enzyme neces-sary for stimulating glucose uptake ininsulin-sensitive tissues. In a series ofsubsequent experiments, the seeminglyparadoxical finding of an increase in asubunit of PI 3-kinase being associatedwith IR was ultimately solved. PI 3-kinaseis composedofboth ap85 regulatoryunitand a catalytic subunit (p110), whichmust form a p85-p110 heterodimerand bind to insulin receptor substrate1 (IRS-1) for PI 3-kinase activation tooccur. When the p85 monomer is selec-tively overexpressed by the action ofhPGH, it competes with the activep85-p110 heterodimer in a dominantnegative fashion for binding to IRS-1,resulting in a decrease in IRS-1–associatedPI 3-kinase activity. We confirmed thatthis decrease in PI 3-kinase activity re-sulted in the final step of the insulinsignaling pathway being attenuateddareduction in GLUT-4 translocation to theplasma membrane (11) (Fig. 1). Further,the excess GH-induced IR was com-pletely reversible using a GH-releasingantagonist (12).

INSULIN SIGNALING CHANGES INSKELETAL MUSCLE IN PREGNANTAND POSTPARTUM WOMEN WITHOBESITY OR GDM

Although hPGH mediated the IR in micethat expressed this placental hormone atlevels similar to the third trimester ofhuman pregnancy, ascribing this mech-anism to the IR in human pregnancy wasstill a leap. To translate these findingsto human pregnancy, we designed a lon-gitudinal study in which we collectedfour vastus lateralis muscle biopsiesfrom pregnant women who were over-weight/obese with BMI-matched diet-controlled GDM mothers before andafter a glucose load at ;30–32 weeksgestation and again at 9–10 weeks post-partum. Our intent was to also decipherthe dynamic insulin signaling changes inGDM mothers who continued to displayimpaired glucose tolerance postpartum.Not surprisingly, recruitment for thesestudies, which involved fasting andinsulin-stimulated muscle biopsies in

care.diabetesjournals.org Barbour 719

pregnant women both antepartum andpostpartum, was painfully slow. How-ever, after ;7 years, our prospectivestudy demonstrated that reduced abilityof insulin to phosphorylate the insulinreceptor and increased levels of the p85subunit ofPI 3-kinase, similar toourmice,were reversible after delivery, as mightbe expected given the disappearance ofhPGH after pregnancy. However, moth-ers with GDM also had a decrease in IRS-1and changes in tyrosine and serine phos-phorylation compared with BMI-matchedcontrol subjects (13), further dampeningsignaling. Importantly, somewomenwithGDMhad been shown to have a persistentinsulin signaling defect with increasedskeletal muscle tumor necrosis factor-a(TNF-a) up to 1 year postpartum (14). Wealso established that GDMwomen demon-strate impaired IRS-1 signaling–associatedincreased serine kinase activation, which

persisted in women who did not normal-ize their glucose tolerance postpartum(15) (Fig. 1).

Although we hoped that the physio-logical changes in hPGH in pregnancywould be a straightforward predictor ofthe degree of IR in pregnancy, McIntyreet al. (16) found hPGH not to be a simplebiomarker of the multiple biologic pro-cesses that contribute to the overall IRof pregnancy and excess fetal growth,which also include leptin, adiponectin,TNF-a, and IGFBP1. Catalano and col-leagues (17,18) have also shown a re-duction in adiponectin and PPARg inadipose tissue that is also associatedwith tissue-specific IR. While reducingexcessive skeletal muscle or adipose IR inpregnancies complicated by GDM mightbe a viable target for attenuating IR,the mechanisms driving these changeslikely have important roles in normal

pregnancy. We anticipate that the insulinsignaling defects we observed to persistin skeletal muscle of GDM women, inaddition to observations by Boyle et al.(19) that AMPK and oxidative phosphor-ylation are dramatically reduced in theskeletal muscle of GDM women, mightlead to specific targeted therapy or phys-ical activity interventions for preventionof GDM and the development of type 2diabetes postpartum.

WHY DO WOMEN WITH OBESITYALONEGIVE BIRTH TO BIG BABIES?

Obesity, which is often associated withIR, is now the leading cause of maternalmorbidity and affects nearly 40% ofwomen of child-bearing age (20). Infact, obesity alone accounts for a greaternumber of large-for-gestational age(LGA) infants than pregnancies compli-cated by preexisting diabetes or GDM

Figure 1—Simple schematic of the proximal insulin signaling pathway in skeletal muscle of normal pregnancy and GDM. Insulin stimulatestyrosine phosphorylation (P) of the insulin receptor (IR), defective in GDM, which activates the insulin receptor to dock IRS-1. After IRS-1 isphosphorylated on tyrosine domains, it triggers the recruitment of PI 3-kinase, a critical enzyme for glucose transport into the cell. GDMwomenalso have decreased tyrosine activation of IRS-1 and reduced IRS-1, in part due to increased serine phosphorylation that increases thedegradation of IRS-1 and dampens the insulin signal. The increase in basal IRS-1 serine phosphorylation does not resolve in GDM womenwho show persistent glucose intolerance postpartum and are at highest risk for type 2 diabetes. PI 3-kinase is composed of a regulatorysubunit (p85a) and a catalytic subunit (p110) that must form a heterodimer for PI 3-kinase activation to occur. When excess p85amonomersare stimulated by hPGH in normal pregnancy, it competes with the p85-p110 heterodimer for binding to IRS-1, thereby causing a decreasein the IRS-1–associated PI 3-kinase activity and reduction in GLUT-4 translocation to the plasma membrane. Elevated p85 reverses postpartumwith the disappearance of hPGH.

720 Norbert Freinkel Award Lecture Diabetes Care Volume 42, May 2019

(6,21,22). Many studies have reportedstrong associations between intrauterineexposure to maternal obesity, excessadiposity at birth, and offspring devel-opment of obesity in childhood (2). Fast-ing, 1-h, and 2-h glucose on a 75-g oralglucose tolerance test, lower than pre-viously considered as GDM, were corre-lated with LGA in the Hyperglycemia andAdverse Pregnancy Outcome (HAPO)Study (23) and were recently shown tocorrelate with glucose intolerance inthe 10- to 14-year-old offspring (24).However, Catalano et al. (25) demon-strated in a cohort of 89 offspring frommothers who were normoglycemic orwho had GDM that childhood adiposityat 9 years correlated much more stronglywith maternal obesity than did GDM,with an OR of ;5.5. In addition to ma-ternal prepregnancy BMI, the degreeof gestational weight gain (GWG) alsocontributes to excess fetal growth, al-though overall less so than prepregnancyBMI in our review of the literature (26).The association with excess GWG andadiposity at birth appeared strongest inoverweight mothers who gained exces-sive weight compared withmothers withobesity who gave birth to infants withthe highest mean fat mass, regardless ofGWG.We hypothesized that given the strong

contribution ofmaternal obesity alone toexcess fetal growth independent of GDM(27), women with obesity may havehigher fasting and postprandial glucosesthanwomenwhoarenormalweight (NW),although not high enough to meet cri-teria for GDM. In an extensive reviewdonewithDr. Teri Hernandez of glycemiapatterns in normal pregnancy (28), wedetermined that mean fasting glucoselevel in NW mothers was 71 mg/dL, 1-hpostprandial value was 109 mg/dL, andthe 2-h postprandial value was 99 mg/dLwith a 24-hmean glucose of 88mg/dL, allmuch lower than current therapeutictargets (gestational age 34 6 2 weeks).In a prospective study, spearheaded bytwo endocrine fellows, in which we con-trolled maternal diet while wearing con-tinuous glucose monitors (CGM), womenwith obesity were shown to have ;8%higher 24-h glycemic profiles comparedwith NW control subjects (29). We alsoidentified daytime, nocturnal, fasting,and postprandial glycemic profiles thatwere useful in characterizing fetalgrowth (30). In a second highly controlled

longitudinal study in which NW mothersand those with obesity received con-trolled eucaloric diets, we documentedmean fasting blood glucose values of;73 vs. ;77 mg/dL at 28 weeks andmean 2-h postprandial glucose values of87 vs. 96 mg/dL in our NW and obesegroups of women, respectively (31). Fur-thermore, the fasting and repeatedmeasures of postprandial glucose over4 h both early (;16 weeks) and later(;28 weeks) in pregnancy were also;10% higher in the mothers with obe-sity. Women with obesity had largerincreases in their 24-h glucose profilesanddaytime,nocturnal, andpostprandialglucose profiles from early to later preg-nancy compared with NW mothers (allP , 0.05) when studied on a eucaloriccontrolled diet (32). In the women withobesity, the postprandial glucose expo-sure over three meals early in pregnancywas predictive of newborn percentfat mass (NB %fat), and later in preg-nancy, the entire 24-h glucose profileswere predictive in women with obe-sity (all P, 0.05) but not in NW women(32). This subclinical pattern of mildhyperglycemia, which remains occultand untreated, may in part explainwhy the prevalence of LGA and macro-somia is higher in women with obesity(Fig. 2A). Interestingly, recent data by apostdoctoral fellow in our group dem-onstrated that women with obesity whohave mild sleep disordered breathing(apnea-hypopnea index.5) have higher24-h glucose profileswhenon afixeddietthan women without obesity. This sug-gests that mild sleep apnea, common inwomen with obesity, may be one of thecontributors to elevated 24-h glycemicprofiles that could be targeted for treat-ment (33).

Inaddition tohigherglucose levels, ourdata also showed that fasting and post-prandial insulin concentrations, aswell asIR estimated by HOMA-IR, are 50–60%higher in women with obesity both early(;16 weeks) and later in pregnancy(28 weeks) as compared with NW womenreceiving a fixed diet (31) (Fig. 2B). Thehigher degree of IR in women withobesity that predates the pregnancy,as shown by Catalano et al. (34) withhyperinsulinemic euglycemic clamp stud-ies, is clearly present early and worsenslater in pregnancy. This additional mater-nal IR, superimposed on the IR of normalpregnancy, may serve to increase the

availability of all nutrients (including lipidsand amino acids) to the fetal-placentalunit, resulting in excess fetal growth inwomen with obesity (Fig. 3).

THE UNDERAPPRECIATED ROLE OFMATERNAL LIPIDS AS ACONTRIBUTOR TO INFANTADIPOSITY

Although much of the focus on fetalgrowth has been through a glucocentriclens, there is increasing recognition thatit is not only maternal glucose that con-tributes to fetal fat accretion and thattriglycerides (TGs) and FFAs also play asignificant role (35–39). Another unex-pected twist in our earlier CGM data(29), during which women with obesityon a controlled diet demonstratedhigher 24-h glycemic profiles, wasthat a single fasting TG measured onceat;14–16 weeks in pregnancy correlatedmore strongly with NB %fat by skinfoldsthan any of the glycemic profiles by CGM.The placenta has lipase activity, and thetireless efforts of another postdoctoralfellow demonstrated that the activity ofplacental lipoprotein lipase, importantin hydrolyzing maternal TGs to FFAs thatcan be transported across the placenta, ishighly correlated with NB %fat by skin-folds at birth (r 5 0.59; P 5 0.006) andeven more so when corrected for ges-tational age at delivery (r 5 0.75; P 50.0001) (40).

Although a number of studies haveshown that fasting TGs and FFAs con-tribute to fetal fat accretion, prospectivestudies in which maternal diet was con-trolled and which also examined post-prandial FFAs and postprandial TGs werelacking (35). In our prospective study thatcharacterized the maternal metabolicenvironment in mothers who were NWor obese both early (;16 weeks) andlater in pregnancy (28 weeks), we dem-onstrated that maternal TGs both fastingand postprandially after a controlled dietwere a stronger predictor of NB %fat thanany of our glucose or insulin measures(31). Fasting TGs and postprandial TGsboth at 1 h and 2 h were already ;30–40% higher inmothers with obesity earlyinpregnancy comparedwithNWandalso30–40% higher at 28 weeks (Fig. 2C).Interestingly, in NW women, it was theincrease in TGs from early to later inpregnancy that correlated strongestwithNB %fat (r 5 0.57; P , 0.01) by dual-energy X-ray absorptiometry (DXA).

care.diabetesjournals.org Barbour 721

However, in mothers with obesity, it was16-week fasting TGs (r5 0.60;P50.006)and 1-h or 2-h postprandial TGs (r 50.69–0.71; P , 0.01) in early pregnancythat most strongly correlated with NB%fat, contributed by both VLDL-TG andchylomicron TG (CM-TG) (31). TGs aremodestly higher in mothers with GDMcompared with mothers with obesity(Fig. 2E), and these data, in addition toothers, suggest that maternal TGs area largely unrecognized substrate for fe-tal fat accretion, especially in motherswith obesity (31). A larger community-based trial in a free-living environmentto confirm the contribution and bettercharacterize the relationship of fastingand postprandial TGs to infant adipos-ity is the focus of a proposed futuretrial in women who are overweight orobese.

CAN WE INTERVENE WITH DIET?CHALLENGING THECONVENTIONAL LOWERCARBOHYDRATE DIET IN GDM

Efforts at deciphering theoptimal diet forpregnant women have been largely lim-ited to women with GDM given that diettherapy is the first-line therapy andmainstay of treatment. Conventionally,the “GDM diet” was one that limitedcarbohydrates in an attempt to attenu-ate the postprandial glycemic response(41). However, many mothers substitutefat for carbohydrate (42,43), which may

have unintentional consequences ofincreasing FFAs, which are a strong con-tributor to worsening maternal IR, pro-moting further shunting of all excessnutrients to the growing fetus (Fig. 3).Given the lack of rigorous data support-ing theuseofonemacronutrient over theother, the American Diabetes Asso-ciation abandoned specific diet recom-mendations for mothers with GDM in2005 (44) other than ensuring thatwomen receive, at the minimum, 175g of carbohydrate/day in the secondand third trimester. This amount is crit-ical for meeting the energy demands ofthe placenta and fetus, of which 80%are dependent on glucose.

My research colleague, Dr. TeriHernandez, as a former cardiac nurse,was extremely disconcerted about thepotential for a low-carbohydrate, higher-fat diet to have unintended consequencesand potentially worsen maternal IR, in-flammation and actually promote excessfetal fat accretion. We reconstructed thehighly quoted ;40% carbohydrate/45%fat/15% protein conventional diet de-scribed by Jovanovic-Peterson andPeterson (41) and developed a “challenge”diet precisely matched in calories andtypes of carbohydrates, fats, and pro-teins that only differed in the macronu-trient percentages (45). This diet, latertermed the CHOICE diet (ChoosingHealthy Options in Carbohydrate En-ergy), substituted 20% of the calories

from fat with higher-quality, more com-plex carbohydrates such that the mac-ronutrient composition was instead 60%carbohydrate/25% fat/15%protein. Bothdiets were matched for types of fatsand ,18% of total calories from bothdiets were from simple sugars. Impor-tantly, given that current literature lacksrandomized controlled trials (RCTs) com-paring diets in which all meals are pro-vided, critical in minimizing overlapbetween the diets, our design providedall meals to the mothers at the time ofGDMdiagnosis through the remainder ofpregnancy. Data from a randomized con-trolled crossover trial demonstratedthat the higher–complex carbohydrate/lower-fat CHOICE diet achieved similarglycemic profiles using CGM (6% higheron CHOICE) and that the mean 1-h(115 mg/dL) and 2-h (106 mg/dL) post-prandial glucoses were still well belowthe targets for GDM (45). The FFAs weresignificantly lower postprandially on theCHOICE diet. In a follow-up pilot studyin which glucose, insulin, FFAs, adiposetissue lipolysis, and NB %fat were mea-sured after the mothers had beenrandomized to one of the diets for 6–7 weeks, we found that the fasting bloodglucosewasactually loweron theCHOICEdiet (P 5 0.03). Furthermore, insulinsuppression of lipolysis in maternal ad-ipose tissue biopsies was better on theCHOICE diet, corresponding to lowerexpression of multiple adipose tissue

µµ

TG

sT

Gs

Figure2—Data fromcontrolledbreakfastmeal studies showing the4-hareaunder the curveof plasmaglucose (A), insulin (B), andTGs (C) inNWwomenvs. womenwith obesity (OB) early (12–14weeks) and later (26–28weeks) in pregnancy and the 4-h area under the curve of plasma glucose (D), insulin(E), and TGs (F) in NW, OB, and diet-controlled GDM at 28–30 weeks.

722 Norbert Freinkel Award Lecture Diabetes Care Volume 42, May 2019

proinflammatory genes (46). Impor-tantly, NB %fat trended lower on theCHOICE higher–complex carbohydrate/low-fat diet in a limited pilot sample of12 mother/offspring dyads. The findingsare consistent with other RCTs wheremeals were not provided but also sup-port that diets lower in glycemic index,low in saturated fat, and higher in fiberappear to attenuate excess fetal growth(43,47). The final data from the CHOICERCT powered for a diet difference in NB%fat as the primary outcome should beavailable within the next year.

THE SIGNIFICANCE ANDCHALLENGE IN MEASURING BODYCOMPOSITION IN BABIES ANDINFANTS

Birth weight is an imperfect proxy forincreased infant fat mass given that itincludes both lean mass and fat mass. In

the cohort of 89 offspring of motherswith normal glucose tolerance or GDM inthe study by Catalano et al. (25), in-creased adiposity at birth (but not birthweight) was correlated to adiposity at6 to 11 years of age by DXA. Adiposity atbirth and during the first 2 years of lifewhen infants triple their fat mass is animportant predictor for future obesityrisk andmetabolic syndrome,muchmorethan birth weight alone. DXA and air dis-placement plethysmography (PEAPOD) areconsidered methods with more precisionand reproducibility than skinfoldmeasures orbioelectrical impedance analysis and lessexpensive than MRI techniques (48,49).However, DXA requires infants to be stilland exposes them to a very low back-ground level of radiationda misinter-pretation (and unfortunate twist) thathalted our recruitment for nearly 1 year.Regrettably, most infants can no longer

fit into a PEAPOD by 6 months of life.Given the shortcomings of these differ-ent methods and the need to accuratelymeasure fat and fat-free mass through-out infancy, we prospectively measuredNB %fat in term infants using DXA, PEA-POD, and skinfolds at;10–14 days of life(all three measures taken on same day)and again at ;1 year of life by DXA andskinfolds. We evaluated two differentHologic software versions for DXA alongwith PEAPOD and skinfolds and deter-mined how well they correlated witheach other and, more importantly, theirdegree of agreement (48).

Consistent with our track record ofsurprising twists, we found that althoughthe Pearson correlations between theour original DXA software version(DXA1) with the newer software version(DXA2) in newborns were strong (r 50.89), aswithDXA1andPEAPOD(r50.74)

Figure 3—Diagrammatic portrayal of the effects of overnutrition on offspring in pregnancies complicated by obesity,metabolic syndrome, or GDMandthe potential life-cycle consequences. A woman becomes pregnant with variable degrees of preexisting metabolic dysfunction; her metabolism isfurthered altered by the IR of pregnancy mediated by placental factors. Maternal IR in muscle, adipose tissue, and liver along with dietary nutrient excessresults in excess glucose (G), amino acid (AA), FFA, and inflammatory cytokine exposure to the placenta. Placental angiogenesis and the transcriptomeare influencedbymaternal factors and the sexof the fetus,whichdeterminenutrient sensing,metabolism, transport, andultimatenutrient exposure tothe fetus. The fetus responds to excess glucose and AAs with fetal hyperinsulinemia, a potent growth factor, which also promotes organomegaly,adipogenesis, and lipid storage in subcutaneous (SubQ) and hepatic tissue. Intrauterine overnutrition affects stem cell differentiation, mitochondrialfunction, andappetite regulation in theoffspring, and thematernalmicrobiome is transmittedatdelivery. Postnatal factors suchasmodeof feedingandrapid weight (Wt) gain along with other early life exposures further promote the development of childhoodmetabolic disease. The daughter becomesa mother and the cycle perpetuates. EL, endothelial lipase; pLPL, placental lipoprotein lipase; uMSC, umbilical mesenchymal stem cell.

care.diabetesjournals.org Barbour 723

and with DXA1 and skinfold estimates(r 5 0.69), the actual agreement be-tween DXA1 and the other methodswas much less impressive (intraclasscorrelation coefficient [ICC] 0.53 withDXA2; 0.62 for PEAPOD and also 0.62for skinfolds). By 1 year of age, althoughthe correlation between the two DXAsoftware versions was identical to thatseen in the newborn (r 5 0.89), theagreement was poor (ICC 0.20). Thecorrelation between DXA1 and skinfoldsat 1 year was poor (r 5 0.50), and theagreement was even worse (ICC 0.11)(48). Unfortunately, these data demon-strate that while DXA, PEAPOD, andskinfolds are highly correlated at birth,the agreement is less impressive suchthat the methodologies cannot besubstituted for each other when follow-ing offspring longitudinally. Further, at1 year of life, both the correlation and theagreement with DXA and skinfolds arevery poor, with DXA estimating NB %fatat about ;3 times higher than by skin-folds in our study. These data supportthat the same methodology should beused in longitudinal studies. This is prob-lematic given that MRI is expensive and,like DXA, it is difficult to restrict infantand toddler movement, infants will out-grow the PEAPOD after 6 months, andtoddler PEAPOD (TODPOD)measures areless available and challenging.Another limitation of DXA, PEAPOD,

bioelectrical impedance analysis, or skin-folds is that they do not measure liveror visceral fat, likely a more importantestimate of metabolic risk (48). OnlyMRIspectroscopy is able to measure visceralor liver fat, which is expensive and re-quires that the infant or child lie still. Apediatric gastroenterologist colleaguehelped to design methodology, withthe assistance of radiology experts, toestimate the liver fat in the newbornoffspring from a cohort of GDMmotherswho were obese compared with new-borns from mothers who were NW. It isvery concerning that newborns fromGDM mothers with obesity already hada 68% increase in intrahepatocellularlipid at;2 weeks of age (50), and recentpilot data support a strong relationshipwith fasting TGs and postprandial TGsat GDM diagnosis and newborn liverfat (51). Given the steadily increasingprevalence of childhood nonalcoholicfatty liver disease (NAFLD) that pro-gresses more rapidly to nonalcoholic

steatohepatitis (NASH) in childhood,these findings and others in nonhumanprimates by McCurdy et al. (52) stronglysuggest that an intrauterine environ-ment characterized by excess maternalfuels from obesity, diabetes, or a high-fatdiet may result in TG accumulation inthe fetal liver. This may be especially im-portant if the exposure is early in preg-nancy, before subcutaneous fat storesare developed, and could potentiallyserve as a “first hit” toward the devel-opment of NAFLD (53), which may affectup to 40% of children with obesity.

TARGETING WHO, WHAT, ANDWHEN TO PREVENT ADVERSEPREGNANCY OUTCOMES ANDMETABOLIC DISEASE IN MOTHERSAND OFFSPRING

We and others have demonstrated thatthe metabolic heterogeneity in humanpregnancy, simply based on maternalBMI or the presence or absence of aGDM diagnosis (also influenced by var-ious criteria), is enormous (54). Not allwomen with normal BMI are insulinsensitive, not all women with a BMI inthe obese range are IR (6), and womenwith “GDM”have highly variable degreesof skeletal muscle, adipose, and hepaticIR predating the pregnancy in addition tothe normal IR of pregnancy (4). Further,GDM women have differing degrees ofb-cell defects that also predate the preg-nancy and some can mount markedlyhigh insulin responses and others cannot(5,6,54) (Fig. 1). Lipid metabolism inpregnancy is also highly variable, partlyrelated to the estrogen-induced influ-ence of VLDL production in pregnancyand CM-TG in the diet, superimposed onthe degree of maternal adipose tissue IR(31,39). The degree to which TGs andFFAsmay contribute to fetal fat accretionhas yet to be fully elucidated, but ob-servations that maternal glucose doesnot solely explain fetal overgrowth sup-port its future investigation as a thera-peutic target.

New technologies and advancementshave informed us how maternal diet,maternal body composition, and thematernal response to placental hor-mones and growth factors, in the contextof the mother’s own metagenomics andmetagenomics of her microbiome, caninfluence her IR and inflammatory state,b-cell responses, and the levels of glu-cose, amino acids, and lipids she exposes

to the fetal-placental unit (55) (Fig. 3).The placenta is the ultimate gatekeeperto what crosses into the fetal circulationand is affectedbyplacental angiogenesis,its own transcriptome that will influenceplacental hormone and growth factorproduction, metabolism, nutrient sens-ing and transporters, as well as the sex ofthe placenta (56–58) (Fig. 3). Clearly thetiming of these exposures to the fetal-placental unit, very early or later inpregnancy, has immense fetal program-ming implications (7,59,60). Indeed, mostlifestyle interventions to improve preg-nancy outcomes at midpregnancy havebeen largely disappointing (59). How thefetus responds to excess nutrient expo-sure and growth factors as a result oftiming in pregnancy and its own meta-genomics will affect its b-cell response,immune and metabolic responses, andfetal fat accretion in subcutaneous versusvisceral stores, as well as stem cell dif-ferentiation into adipocytes or myocytes(Fig. 3). Notably, Boyle et al. (61) haverecently shown that the adipogenic po-tential of offspring umbilical mesenchy-mal stem cells is higher in mothers withobesity. Furthermore, Baker et al. (62)demonstrated that metabolic changes inthese adipocytes were predictive of ad-iposity gain in the first 5 months of life.As described cogently by my colleagueDr. Jed Friedman (1), the peripartumand postnatal environment will continueto have a critical effect on childhooddevelopmental programming, partiallydriven by genetic and epigenetic factorsthat are influenced by mode of delivery,the transmission of the maternal micro-biome (63), feeding practices, antibioticuse, and other environmental exposuresand metabolic disruptors (64), the sub-ject of intense investigation in the Na-tional Institutes of Health (NIH)-fundedEnvironmental influences on Child Out-comes (ECHO) program (65). Dr. Freinkel(3) set the stage for us to consider farmore than simply maternal glucose as thedeterminant ofmaternal outcomes, fetalgrowth, and future maternal and offspringmetabolic risk. Technologic advancementsin metagenomics, proteomics, metabolo-mics, and the microbiome have broughtpreviously unimaginable new lenses to ourunderstanding and have humbled whatwe thought we understood.

Our current ability to identify preg-nancies “at risk” for maternal and off-spring short- and long-term adverse

724 Norbert Freinkel Award Lecture Diabetes Care Volume 42, May 2019

outcomes by using only maternal BMI,GWG, or differing diagnostic criteria todraw the line between GDM and no GDMis, unarguably, limited. “Risk engines”(60) that better identify high-risk mater-nal genetics, new maternal metabolicbiomarkers, and that utilize more ac-curate indicators of diet and lifestylepractices may dramatically alter ourdefinition of at-risk mothers. Personal-ized medicine may be critical in targetinginterventions that are much more likelyto be successful during prepregnancy orearly gestation that implement person-alized nutrition (likely influenced by themicrobiome), supplements or prebiotics,and activity prescriptionsdall based onnutrigenomics andmetabolomics. Whenit is necessary to prescribe medications,careful consideration of their pharmaco-kinetic properties (which may differ bythe mother’s pharmacogenomics), howbest to individualize their use to targetspecific metabolic abnormalities, andevaluation of their potential long-termprogramming effects for those that crossthe placenta argue against a one-size-fits-all interventional approach (66). Onething is clear: our current definition ofat-risk women simply based on maternalglucose requires modification, our tar-geted interventions need to be individ-ualized and occur early, and our effortscannot be halted at the time of delivery.The challenges are formidable, but thepotential formeaningfully impactful con-sequences on maternal and offspringmetabolic health call out loudly to futureinvestigators with new toolboxes andperspectives.

Acknowledgments.Mydeepest gratitude goesout to my closest colleagues, Drs. Jed Friedmanand Teri Hernandez; to all of my extraordinarycollaborators and friends who have made all ofthis research possible; to my inspirational men-tors, Drs. Richard Byyny, Richard Lee, BorisDraznin, Robert Eckel, and E. Chester (Chip)Ridgway; and to my husband (Pat), children(Sarah andGreg), and stepchildren (Katie, Shaun,Justin, Megan, and Mike), whose love and sup-port gives me purpose every day. This lecture isdedicated to my parents, Jeanne and EdmundBarbour, in special remembrance.Funding. This work was funded by NIH grantsR01DK078645, R01DK101659, and R21DK088324.Duality of Interest. No potential conflicts ofinterest relevant to this article were reported.

References1. Friedman JE. Developmental programming ofobesity and diabetes in mouse, monkey, and man

in 2018: where are we headed? Diabetes 2018;67:2137–21512. Catalano PM, Shankar K. Obesity and preg-nancy: mechanisms of short term and long termadverse consequences formother and child. BMJ2017;356:j13. Freinkel N. Banting lecture 1980. Of preg-nancy and progeny. Diabetes 1980;29:1023–10354. Hernandez TL, Friedman JE, Barbour LA. In-sulin resistance in pregnancy: implications formother and offspring. In Insulin Resistance: Child-hood Precursors toAdult Disease, 2nd ed. Zeitler P,Nadeau KJ, Eds. Totowa, NJ, Springer. In press5. Buchanan TA, Xiang A, Kjos SL, Watanabe R.What is gestational diabetes? Diabetes Care2007;30(Suppl. 2):S105–S1116. Barbour LA, Friedman JE, Hernandez TL. Met-abolic changes in normal and obese pregnanciesand in gestational diabetes. In Diagnosis andManagement of Diabetes During Pregnancy.ReeceEA, CoustanD, Eds. Philadelphia, PA, Wol-ters Kluwer Health. In press7. DesoyeG,Nolan CJ. The fetal glucose steal: anunderappreciatedphenomenon in diabetic preg-nancy. Diabetologia 2016;59:1089–10948. Handwerger S, Freemark M. The roles ofplacental growth hormone and placental lacto-gen in the regulation of human fetal growth anddevelopment. J Pediatr Endocrinol Metab 2000;13:343–3569. McIntyreHD,SerekR,CraneDI, etal. Placentalgrowth hormone (GH), GH-binding protein, andinsulin-like growth factor axis in normal, growth-retarded, and diabetic pregnancies: correlationswith fetal growth. J Clin Endocrinol Metab 2000;85:1143–115010. Barbour LA, Shao J, Qiao L, et al. Humanplacental growth hormone causes severe insulinresistance in transgenic mice. Am J Obstet Gyn-ecol 2002;186:512–51711. Barbour LA, Shao J, Qiao L, et al. Humanplacental growth hormone increases expressionof thep85 regulatoryunit of phosphatidylinositol3-kinase and triggers severe insulin resistance inskeletal muscle. Endocrinology 2004;145:1144–115012. Barbour LA,Mizanoor Rahman S, Gurevich I,et al. Increased P85a is a potent negative reg-ulator of skeletal muscle insulin signaling andinduces in vivo insulin resistance associated withgrowth hormone excess. J Biol Chem 2005;280:37489–3749413. Barbour LA, McCurdy CE, Hernandez TL,Kirwan JP, Catalano PM, Friedman JE. Cellularmechanisms for insulin resistance in normalpregnancy and gestational diabetes. DiabetesCare 2007;30(Suppl. 2):S112–S11914. Friedman JE, Kirwan JP, Jing M, Presley L,Catalano PM. Increased skeletal muscle tumornecrosis factor-a and impaired insulin signalingpersist in obese women with gestational diabetesmellitus 1 year postpartum. Diabetes 2008;57:606–61315. Barbour LA, McCurdy CE, Hernandez TL,Friedman JE. Chronically increased S6K1 is as-sociated with impaired IRS1 signaling in skeletalmuscle of GDM women with impaired glucosetolerance postpartum. J Clin Endocrinol Metab2011;96:1431–144116. McIntyre HD, Chang AM, Callaway LK, et al.;Hyperglycemia and Adverse Pregnancy Outcome

(HAPO) Study Cooperative Research Group. Hor-monal and metabolic factors associated withvariations in insulin sensitivity in human preg-nancy. Diabetes Care 2010;33:356–36017. Catalano PM, Nizielski SE, Shao J, Preston L,Qiao L, Friedman JE. Downregulated IRS-1 andPPARg in obese women with gestational diabe-tes: relationship to FFA during pregnancy. Am JPhysiol Endocrinol Metab 2002;282:E522–E53318. Catalano PM, Hoegh M, Minium J, et al.Adiponectin in human pregnancy: implicationsfor regulation of glucose and lipid metabolism.Diabetologia 2006;49:1677–168519. Boyle KE, Hwang H, Janssen RC, et al. Ges-tational diabetes is characterized by reducedmitochondrial protein expression and alteredcalcium signaling proteins in skeletal muscle.PLoS One 2014;9:e10687220. Flegal KM, Kruszon-Moran D, Carroll MD,Fryar CD, Ogden CL. Trends in obesity amongadults in the United States, 2005 to 2014. JAMA2016;315:2284–229121. American College of Obstetricians and Gyn-ecologists. ACOG practice bulletin No 156: obesityin pregnancy [published correction appears inObstet Gynecol 2016;128:1450]. Obstet Gynecol2015;126:e112–e12622. Dutton H, Borengasser SJ, Gaudet LM,Barbour LA, Keely EJ. Obesity in pregnancy:optimizing outcomes for mom and baby. MedClin North Am 2018;102:87–10623. Metzger BE, Lowe LP, Dyer AR, et al.; HAPOStudy Cooperative Research Group. Hyperglyce-mia and adverse pregnancy outcomes. N Engl JMed 2008;358:1991–200224. Scholtens DM, Kuang A, Lowe LP, et al.;HAPO Follow-up Study Cooperative ResearchGroup. Hyperglycemia and Adverse PregnancyOutcome Follow-up Study (HAPO FUS): maternalglycemia and childhood glucose metabolism.Diabetes Care 2019;42:381–39225. Catalano PM, Farrell K, Thomas A, et al.Perinatal risk factors for childhood obesity andmetabolic dysregulation. Am J Clin Nutr 2009;90:1303–131326. Nicklas JM, Barbour LA. Optimizing weightfor maternal and infant health - tenable, or toolate? Expert Rev EndocrinolMetab 2015;10:227–24227. Catalano PM, McIntyre HD, Cruickshank JK,et al.; HAPO Study Cooperative Research Group.The Hyperglycemia and Adverse Pregnancy Out-come Study: associations of GDM and obesitywith pregnancy outcomes. Diabetes Care 2012;35:780–78628. Hernandez TL, Friedman JE, Van Pelt RE,Barbour LA. Patterns of glycemia in normalpregnancy: should the current therapeutic tar-gets be challenged? Diabetes Care 2011;34:1660–166829. Harmon KA, Gerard L, Jensen DR, et al.Continuous glucose profiles in obese and nor-mal-weight pregnant women on a controlleddiet: metabolic determinants of fetal growth.Diabetes Care 2011;34:2198–220430. Hernandez TL, Barbour LA. A standard ap-proach to continuous glucose monitor data inpregnancy for the study of fetal growth andinfantoutcomes.DiabetesTechnol Ther2013;15:172–17931. Barbour LA, Farabi SS, Friedman JE, et al.Postprandial triglycerides predict newborn fat

care.diabetesjournals.org Barbour 725

more strongly than glucose in women withobesity inearlypregnancy.Obesity (Silver Spring)2018;26:1347–135632. Hernandez TL, Farabi SS, Van Pelt RE, et al.Early and late glycemic patterns in normal weight vsobese pregnancies: influence on neonatal adiposity(Poster 1485-P). Diabetes 2017;66(Suppl. 1):A39733. Farabi SS, Barbour LA, Heiss K, Hirsch NM,DunnE, Hernandez TL.Obstructive sleep apnea isassociated with altered glycemic patterns inpregnant women with obesity. J Clin EndocrinolMetab. 22 February 2019 [Epub ahead of print].DOI: 10.1210/jc.2019-0015934. Catalano PM, Huston L, Amini SB, Kalhan SC.Longitudinal changes in glucose metabolismduring pregnancy in obese women with normalglucose tolerance and gestational diabetes mel-litus. Am J Obstet Gynecol 1999;180:903–91635. Barrett HL, Dekker Nitert M, McIntyre HD,Callaway LK. Normalizing metabolism in diabeticpregnancy: is it time to target lipids? DiabetesCare 2014;37:1484–149336. Barbour LA, Hernandez TL. Maternal non-glycemic contributors to fetal growth in obesityand gestational diabetes: spotlight on lipids. CurrDiab Rep 2018;18:3737. Wang J, Moore D, Subramanian A, et al.Gestational dyslipidaemia and adverse birth-weight outcomes: a systematic review andmeta-analysis. Obes Rev 2018;19:1256–126838. Herrera E, Ortega-Senovilla H. Implicationsof lipids in neonatal body weight and fat mass ingestational diabetic mothers and non-diabeticcontrols. Curr Diab Rep 2018;18:739. Barbour LA, Hernandez TL. Maternal lipidsand fetal overgrowth: making fat from fat. ClinTher 2018;40:1638–164740. Heerwagen MJR, Gumina DL, Hernandez TL,et al. Placental lipoprotein lipase activity ispositively associated with newborn adiposity.Placenta 2018;64:53–6041. Jovanovic-Peterson L, Peterson CM. Dietarymanipulation as a primary treatment strategy forpregnancies complicated by diabetes. J Am CollNutr 1990;9:320–32542. Hernandez TL. Carbohydrate content in theGDMdiet: two views: view1: nutrition therapy ingestational diabetes: the case for complex car-bohydrates. Diabetes Spectr 2016;29:82–8843. Hernandez TL, Mande A, Barbour LA. Nutri-tion therapy within and beyond gestational di-abetes. Diabetes Res Clin Pract 2018;145:39–50

44. Metzger BE, Buchanan TA, Coustan DR, et al.Summary and recommendations of the FifthInternational Workshop-Conference on Gesta-tional Diabetes Mellitus. Diabetes Care 2007;30(Suppl. 2):S251–S26045. Hernandez TL, Van Pelt RE, Anderson MA,et al. A higher-complex carbohydrate diet ingestational diabetes mellitus achieves glucosetargets and lowers postprandial lipids: a random-ized crossover study. Diabetes Care 2014;37:1254–126246. Hernandez TL, Van Pelt RE, Anderson MA,et al. Women with gestational diabetes mellitusrandomized to a higher-complex carbohydrate/low-fat diet manifest lower adipose tissue in-sulin resistance, inflammation, glucose, andfree fatty acids: a pilot study. Diabetes Care2016;39:39–4247. Hernandez TL, Brand-Miller JC. Nutritiontherapy in gestational diabetes mellitus: timeto move forward. Diabetes Care 2018;41:1343–134548. Barbour LA, Hernandez TL, Reynolds RM,et al. Striking differences in estimates of infantadiposity by new and old DXA software, PEAPODand skin-folds at 2 weeks and 1 year of life.Pediatr Obes 2016;11:264–27149. Demerath EW, Fields DA. Body compositionassessment in the infant. Am J Hum Biol 2014;26:291–30450. Brumbaugh DE, Tearse P, Cree-Green M,et al. Intrahepatic fat is increased in theneonatal offspring of obese women withgestational diabetes. J Pediatr 2013;162:930–936.e151. Hernandez TL, Farabi SS, Hirsch NM, Dunn E,Haugen E, Brumbaugh D, et al. Maternal trigly-cerides in gestational diabetes are strongly as-sociated with increased newborn hepatic fatindependent of subcutaneous fat (Abstract).Diabetes. In press52. McCurdy CE, Bishop JM, Williams SM, et al.Maternal high-fat diet triggers lipotoxicity in thefetal livers of nonhuman primates. J Clin Invest2009;119:323–33553. BrumbaughDE, Friedman JE.Developmentalorigins of nonalcoholic fatty liver disease. PediatrRes 2014;75:140–14754. Powe CE, Allard C, Battista MC, et al. Het-erogeneous contribution of insulin sensitivityand secretion defects to gestational diabetesmellitus. Diabetes Care 2016;39:1052–1055

55. Lowe WL Jr, Bain JR, Nodzenski M, et al.;HAPO Study Cooperative Research Group. MaternalBMI and glycemia impact the fetal metabolome.Diabetes Care 2017;40:902–91056. Lewis RM,DesoyeG. Placental lipid and fattyacid transfer inmaternal overnutrition. Ann NutrMetab 2017;70:228–23157. Radaelli T, Lepercq J, VarastehpourA, Basu S,Catalano PM, Hauguel-DeMouzon S. Differentialregulation of genes for fetoplacental lipid path-ways in pregnancy with gestational and type 1diabetes mellitus. Am J Obstet Gynecol 2009;201:209.e1–209.e1058. Dimasuay KG, Boeuf P, Powell TL, Jansson T.Placental responses to changes in the maternalenvironment determine fetal growth. FrontPhysiol 2016;7:1259. Simmons D. Prevention of gestational di-abetes mellitus: where are we now? DiabetesObes Metab 2015;17:824–83460. McIntyre HD, Gibbons KS, Lowe J, Oats JJN.Development of a risk engine relating maternalglycemia and body mass index to pregnancyoutcomes. Diabetes Res Clin Pract 2018;139:331–33861. Boyle KE, Patinkin ZW, Shapiro AL, Baker PR2nd,DabeleaD, Friedman JE.Mesenchymal stemcells from infants born to obese mothers exhibitgreater potential for adipogenesis: the HealthyStart BabyBUMP Project. Diabetes 2016;65:647–65962. Baker PR 2nd, Patinkin ZW, Shapiro ALB,et al. Altered gene expression andmetabolism infetal umbilical cord mesenchymal stem cellscorrespond with differences in 5-month-old in-fant adiposity gain. Sci Rep 2017;7:1809563. Soderborg TK, Clark SE, Mulligan CE, et al.The gut microbiota in infants of obese mothersincreases inflammation and susceptibility toNAFLD. Nat Commun 2018;9:446264. Dabelea D. Diabetes in youthdlookingbackwards to inform the future: Kelly West AwardLecture 2017. Diabetes Care 2018;41:233–24065. Gillman MW, Blaisdell CJ. Environmentalinfluences on Child Health Outcomes, a researchprogramof theNational InstitutesofHealth. CurrOpin Pediatr 2018;30:260–26266. Barbour LA, Scifres C, Valent AM, et al. Acautionary response to SMFM statement:pharmacological treatment of gestational di-abetes. Am J Obstet Gynecol 2018;219:367.e1–367.e7

726 Norbert Freinkel Award Lecture Diabetes Care Volume 42, May 2019