Search

OBJECTIVE This study was conducted to determine the utility of tubular (urinary/plasma neutrophil gelatinase-associated lipocalin [NGAL] and urinary kidney injury molecule 1 [KIM-1]) and glomerular (estimated glomerular filtration rate [eGFR]) biomarkers in predicting preeclampsia (PE) in pregnant women with type 1 diabetes mellitus (T1DM) who were free of microalbuminuria and hypertension at the first trimester. RESEARCH DESIGN AND METHODS This was a prospective study of T1DM pregnancy. Maternal urinary and plasma NGAL, urinary KIM-1 (ELISA of frozen samples), and eGFR (Chronic Kidney Disease Epidemiology Collaboration equation) were determined at three study visits (V1: 12.4 ± 1.8; V2: 21.7 ± 1.4; V3: 31.4 ± 1.5 weeks’ gestation [mean ± SD]) in 23 women with T1DM with subsequent PE (DM+PE+), 24 who remained normotensive (DM+PE−), and, for reference, in 19 normotensive pregnant women without diabetes (DM−). The groups with diabetes were matched for age, diabetes duration, and parity. All subjects were normotensive and free of microalbuminuria or albuminuria at V1. All study visits preceded the onset of PE. RESULTS Urinary creatinine-corrected NGAL (uNGALcc, ng/mg) was significantly elevated at V1 in DM+PE+ vs. DM+PE− women (P = 0.01); this remained significant after exclusion of leukocyte-positive samples (5 DM+PE+ and 2 DM+PE−) (P = 0.02). Accounting for BMI, HbA1c, and total daily insulin dose, a doubling of uNGALcc at V1 conferred a sevenfold increase in risk for PE (P = 0.026). In contrast, neither plasma NGAL nor urinary KIM-1 predicted PE. Also at V1, eGFR was elevated in DM+PE+ vs. DM+PE− (P = 0.04). CONCLUSIONS Early tubular and glomerular dysfunction may predict PE in first trimester women with T1DM, even if free of microalbuminuria. These data suggest that subclinical renal tubular and glomerular injury, if present early in pregnancy, may predispose women with T1DM to PE.

The risk for preeclampsia (PE) is increased 4-fold in women with type 1 diabetes (T1D), in whom we previously demonstrated that elevated cholesterol-rich lipoproteins confer risk at the 1st trimester. Sphingolipids, including ceramides (Cer) and glycosphingolipids (hexosyl- (HexCer) and lactosyl-ceramides (LacCer)), are lipoprotein constituents that regulate cell signaling. We propose that these bio-active lipids could be modulators of PE risk. In a prospective T1D pregnancy cohort (studied at 12, 22, 32 weeks’ gestation), we used liquid chromatography/electrospray ionization-tandem mass spectrometry to determine plasma levels of 12 species of Cer, HexCer, and LacCer in 23 women with T1DM who did vs. 24 who did not develop PE, and in 19 normotensive nondiabetic pregnant women. All were free of microalbuminuria and hypertension at enrolment; all visits preceded clinical PE onset. Results: Plasma levels of C22-Cer (1st and 3rd study visits); C26:1-Cer (1st and 2nd visits); C26-Cer (1st visit); C18-Cer and C18:1-Cer (2nd visit) were significantly higher in women with T1DM who did vs. did not develop PE. C22-Cer and C26-Cer were significantly lower in T1DM vs. non-T1DM (1st visit). There were no overall PE- or diabetes-related differences in any HexCer- or LacCer species. Longitudinally, most Cer and HexCer species increased throughout pregnancy in all groups (except C14- and C18-Cer; and C14 and C26-HexCer). C16-LacCer increased, while C22:1-, C26-, and C26:1-LacCer decreased as pregnancy advanced, independent of PE status. Independent of LDL-C, total Cer increased throughout pregnancy in all groups. Conclusions: Maternal Cer species may serve as early biomarkers for PE in women with T1DM. T1DM was associated with lower levels of two Cer species. Like cholesterol-rich lipoproteins, Cer, HexCer and some LacCer species increase as pregnancy advances in all groups, but other LacCer species decrease: the significance of these findings requires further study. Disclosure C.B. Kelly: None. S.M. Hammad: None. R.L. Klein: None. M.F. Lopes-Virella: None. M.J. Leyva: None. J. Yu: None. A.J. Jenkins: Advisory Panel; Self; Abbott, Australian Diabetes Society, Medtronic. Research Support; Self; Abbott, GlySens Incorporated, Medtronic, Mylan. Speaker's Bureau; Self; Eli Lilly and Company, Novo Nordisk Inc. A.J. Nankervis: None. K.F. Hanssen: None. S.K. Garg: Advisory Panel; Self; Eli Lilly and Company, Roche Pharma, Sanofi-Aventis, Zealand Pharma A/S. Research Support; Self; Eli Lilly and Company, WebMD. J.A. Scardo: None. C.E. Aston: None. T. Lyons: None. Funding JDRF; Novo Nordisk; National Institutes of Health

Preeclampsia (PE) occurs four times more frequently in pregnant women with type 1 diabetes (T1D) than in nondiabetic women. We hypothesized that elevated plasma advanced glycoxidation products (AGEs) predict PE in T1D women. In a prospective T1D pregnancy cohort (study visits at 12, 22, 32 weeks' gestation), we used liquid chromatography/mass spectrometry, with internal stable heavy isotope substituted standards, to determine plasma levels of nine AGE and oxidation products (carboxymethyl-lysine [CML], carboxyethyl-lysine [CEL], methylglyoxal-hydroimidazolone [MGHI], 3-deoxyglucosone hydroimidazolones [3DGH], methionine sulfoxide [MetSO], glyoxal hydroimidazolone [GHI], 3-nitrotyrosine [3NT], dityrosine [DT], 2-aminoadipic acid [2AAA]) in 23 women with T1DM who developed PE vs. 24 who did not, and in 19 nondiabetic normotensive pregnant women. These products have been associated with CVD and renal failure in non-pregnant cohorts. All women were free of microalbuminuria and hypertension at enrolment, and all visits preceded clinical PE onset. GFR was estimated (Chronic Kidney Disease Epidemiology Collaboration equation). Results: Plasma CML, CEL, MGHI, 3DGH, MetSO, GHI, and 2AAA did not differ between the three groups at any study visit, and thus did not predict preeclampsia in T1D. In T1D women who later developed PE, eGFR was inversely associated at V3 with CML (p=0.008), CEL (p=0.03), MGH1 (p=0.03), 3DGH (p=0.03), GHI (p Conclusions: Plasma AGEs did not predict PE in well-controlled, previously healthy T1D women, and levels did not differ from healthy nondiabetic controls. Uniquely, in T1D women who developed PE, plasma AGEs were associated with renal function, consistent with the notion that subclinical alterations in renal function precede PE. Disclosure H. Karanchi: None. C.B. Kelly: None. S.M. Hammad: None. R.L. Klein: None. M.F. Lopes-Virella: None. M.J. Leyva: None. J. Yu: None. A.J. Jenkins: Advisory Panel; Self; Abbott, Australian Diabetes Society, Medtronic. Research Support; Self; Abbott, GlySens Incorporated, Medtronic, Mylan. Speaker’s Bureau; Self; Eli Lilly and Company, Novo Nordisk Inc. A.J. Nankervis: None. K.F. Hanssen: None. C.E. Aston: None. S. Howell: Employee; Self; PreventAGE Health Care, LLC. P.J. Beisswenger: Other Relationship; Self; PreventAGE Health Care, LLC. T. Lyons: None.

Statin therapy reduces cardiovascular risk in type 2 diabetes (T2DM), partly by improving dyslipidemia, and partly through ‘pleiotropic’ effects. Sphingolipids, including ceramides, are amphiphilic components of lipoprotein surface monolayers. Sphingolipids regulate lipoprotein surface fluidity, and include key signaling molecules: ceramides are specifically associated with apoptosis and risk for premature coronary heart disease. Effects of statins on the sphingolipid content of lipoproteins are poorly defined. Using liquid chromatography/electrospray ionization-tandem mass spectrometry, we determined fasting plasma concentrations of individual ceramides (Cer), sphingomyelins (SM), and glycosphingolipids (hexosyl- and lactosyl-Cer) in plasma from 19 men and 21 women with T2DM, before and after treatment with rosuvastatin (10 mg/day, 6 weeks). Exclusion criteria were use of ‘statins’ in the preceding 6 months, and any current use of lipid-modifying agents. Using existing lipoprotein compositional data, we calculated effects of rosuvastatin therapy on non-HDL sphingolipid content. Results: Rosuvastatin reduced total cholesterol (C) (by 33%), triglycerides (20%), LDL-C (49%), and non-HDL particle number (34%); HDL-C was unaffected. It decreased total SM (15%), Cer (29%), and hexosyl- (30%) and lactosyl- (34%) Cer, independent of gender. Assuming HDL composition was unaffected, rosuvastatin reduced non-HDL SM by 42% and Cer by 61%: these reductions remained significant after correction for non-HDL-C and/or non-HDL lipoprotein surface area. Conclusions: In T2DM, rosuvastatin lowers not only the number of non-HDL particles, but also the sphingolipid, and especially the Cer, content of each particle. This effect is independent of gender, potentially anti-atherogenic, and may represent a new pleiotropic effect of statins. Further studies are needed to confirm these findings in isolated lipoprotein classes, and to determine mechanism(s). Disclosure C.B. Kelly: None. S.M. Hammad: None. R.L. Klein: None. M.F. Lopes-Virella: None. A. Gosmanova: None. M.J. Leyva: None. C.E. Aston: None. T. Lyons: None.

Preeclampsia (PE), a major cause of maternal and fetal morbidity and mortality worldwide, has significantly higher incidence in women with type 1 diabetes mellitus (T1DM) than in nondiabetic populations (∼20% vs. 5%, respectively). Sphingolipids are key signaling molecules, and have many roles in the structure and regulation of cellular functions. Sphingomyelin (SM) is the most abundant sphingolipid in plasma lipoproteins. There is evidence that altered SM metabolism may contribute to cardiovascular and renal complications in diabetes. We aimed to determine whether plasma concentrations of 12 SM species were associated with PE in T1DM in a prospective study of pregnancy of 23 T1DM women who developed PE and 24 T1DM who remained normotensive. Additionally, 19 normotensive nondiabetic pregnant women were included for reference. All subjects were free of microalbuminuria and hypertension at enrolment, and all visits (12, 22, 32 weeks’ gestation) preceded clinical PE onset. Results: There were no cross-sectional differences in total SM, or in any individual SM species, between diabetic women with and without PE, or between normotensive women with and without diabetes at any stage of gestation. With advancing gestation in all groups, C18-SM, C18:1-SM, C20-SM, C20:1-SM, C22-SM, C22:1-SM, C24-SM, and C24:1-SM increased (all p Conclusions: With advancing gestation, plasma concentrations of most SM species increased, as did LDL and VLDL; however, two long-chain and two very long-chain SM species decreased. Although there were no differences in SM according to diabetes or PE status at any time point, further studies should address the significance of differential changes of SM species during pregnancy. Disclosure C.B. Kelly: None. S.M. Hammad: None. R.L. Klein: None. M.F. Lopes-Virella: None. M.J. Leyva: None. J. Yu: None. A.J. Jenkins: Advisory Panel; Self; Abbott, Australian Diabetes Society, Medtronic. Research Support; Self; Abbott, GlySens Incorporated, Medtronic, Mylan. Speaker's Bureau; Self; Eli Lilly and Company, Novo Nordisk Inc. A.J. Nankervis: None. K.F. Hanssen: None. S.K. Garg: Advisory Panel; Self; Eli Lilly and Company, Roche Pharma, Sanofi-Aventis, Zealand Pharma A/S. Research Support; Self; Eli Lilly and Company, WebMD. J.A. Scardo: None. C.E. Aston: None. T. Lyons: None. Funding JDRF; Novo Nordisk; National Institutes of Health

Preeclampsia (PE) occurs more frequently in pregnant diabetic than nondiabetic women (∼20% vs. ∼5%). Early identification of high risk women is needed. Dyslipidemia is implicated: we previously showed that early in pregnancy, increased cholesterol-rich lipoproteins are associated with subsequent PE in type 1 diabetes (T1D). Haptoglobin (Hp) is a plasma protein that binds free hemoglobin, and has two allelic forms, Hp-1, Hp-2, hence three phenotypes. Among people with diabetes, Hp 2-2 phenotype (present in ≈50%) has been associated with oxidative stress, cardiovascular risk and renal decline. We investigated whether maternal Hp phenotype is associated with PE in T1D, and/or modulates lipoprotein-related risks for PE. A prospective study of pregnancy included 23 T1DM women (cases) who developed PE, and 24 T1DM (controls) who remained normotensive. All were free of microalbuminuria and hypertension at enrolment. Hp phenotype was determined by ELISA (Savyon Diagnostics Ltd.). Lipid profiles were measured at three study visits (V1-V3), all preceding PE onset: (mean ± SD) 12.4 ± 1.8, 21.7 ± 1.4, and 31.3 ± 1.4 weeks gestation. Results: Hp phenotype did not differ between women with and without PE, and lipid profiles did not differ by Hp phenotype. In PE cases vs. controls, by univariate analysis, HDL-C was lower at V1 (1.9±0.4 vs. 2.2±0.5 mmol/l (mean ± SD)), while LDL-C was higher at V2 (3.1± 1.0 vs. 2.6± 0.8 mmol/l), as were triacylglycerols (1.6± 0.4 vs. 1.3± 0.4 mmol/l) (p Conclusion: In T1D women, lipoprotein-related risks for PE may be limited to those with Hp 2-2 phenotype. The data provide further evidence of a role for Hp phenotype as a modulator of vascular risk in diabetes. Disclosure C.B. Kelly: None. J. Yu: None. A. Jenkins: Research Support; Self; Medtronic, Mylan, Sanofi-Aventis. A.J. Nankervis: None. K.F. Hanssen: None. S.K. Garg: Research Support; Self; Dexcom, Inc., Eli Lilly and Company, Sanofi US. Advisory Panel; Self; Sanofi US. Research Support; Self; MannKind Corporation, Diasome Pharmaceuticals, Inc., Labstyle Innovations, Lexicon Pharmaceuticals, Inc., Medtronic. Advisory Panel; Self; Novo Nordisk A/S. C.E. Aston: None. T. Lyons: None.

OBJECTIVE Inflammation and endothelial dysfunction have been associated with the immunobiology of preeclampsia (PE), a significant cause of adverse pregnancy outcomes. The prevalence of PE is elevated several fold in the presence of maternal type 1 diabetes mellitus (T1DM). Although cross-sectional studies of pregnancies among women without diabetes have shown altered inflammatory markers in the presence of PE, longitudinal studies of diabetic women are lacking. In maternal serum samples, we examined the temporal associations of markers of inflammation with the subsequent development of PE in women with T1DM. RESEARCH DESIGN AND METHODS We conducted longitudinal analyses of serum C-reactive protein (CRP), adhesion molecules, and cytokines during the first (mean ± SD, 12.2 ± 1.9 weeks), second (21.6 ± 1.5 weeks), and third (31.5 ± 1.7 weeks) trimesters of pregnancy (visits 1–3, respectively). All study visits took place before the onset of PE. Covariates were BMI, HbA1c, age of onset, duration of diabetes, and mean arterial pressure. RESULTS In women with T1DM who developed PE versus those who remained normotensive, CRP tended to be higher at visits 1 (P = 0.07) and 2 (P = 0.06) and was significantly higher at visit 3 (P < 0.05); soluble E-selectin and interferon-γ–inducible protein-10 (IP-10) were significantly higher at visit 3; interleukin-1 receptor antagonist (IL-1ra) and eotaxin were higher and lower, respectively, at visit 2 (all P < 0.05). These conclusions persisted following adjustment for covariates. CONCLUSIONS In pregnant women with T1DM, elevated CRP, soluble E-selectin, IL-1ra, and IP-10 and lower eotaxin were associated with subsequent PE. The role of inflammatory factors as markers and potential mechanisms of the high prevalence of PE in T1DM merits further investigation.

OBJECTIVE Increased oxidative stress and immune dysfunction are implicated in preeclampsia (PE) and may contribute to the two- to fourfold increase in PE prevalence among women with type 1 diabetes. Prospective measures of fat-soluble vitamins in diabetic pregnancy are therefore of interest. RESEARCH DESIGN AND METHODS Maternal serum carotenoids (α- and β-carotene, lycopene, and lutein) and vitamins A, D, and E (α- and γ-tocopherols) were measured at first (12.2 ± 1.9 weeks [mean ± SD], visit 1), second (21.6 ± 1.5 weeks, visit 2), and third (31.5 ± 1.7 weeks, visit 3) trimesters of pregnancy in 23 women with type 1 diabetes who subsequently developed PE (DM PE+) and 24 women with type 1 diabetes, matched for age, diabetes duration, HbA1c, and parity, who did not develop PE (DM PE−). Data were analyzed without and with adjustment for baseline differences in BMI, HDL cholesterol, and prandial status. RESULTS In unadjusted analysis, in DM PE+ versus DM PE−, α-carotene and β-carotene were 45 and 53% lower, respectively, at visit 3 (P < 0.05), before PE onset. In adjusted analyses, the difference in β-carotene at visit 3 remained significant. Most participants were vitamin D deficient ( CONCLUSIONS In pregnant women with type 1 diabetes, low serum α- and β-carotene were associated with subsequent development of PE, and vitamin D deficiency may also be implicated.