Gemma Carreras

Plasma homocysteine is related to albumin excretion rate in patients with diabetes mellitus: a new link between diabetic nephropathy and cardiovascular disease?

Summary The high risk of cardiovascular disease in patients with diabetes mellitus, particularly in those with nephropathy, is not completely explained by classical risk factors. A high plasma homocysteine concentration is an independent risk factor for cardiovascular disease but information on its association with diabetes is limited. Fasting homocysteine concentrations were measured in the plasma of 165 diabetic patients (75 with insulin-dependent [IDDM]; 90 with non-insulin-dependent diabetes [NIDDM]) and 56 non-diabetic control subjects. Other measurements included the prevalence of diabetic complications, glycaemic control, lipid and lipoprotein levels, vitamin status and renal function tests. Patients with NIDDM had higher homocysteine levels than control subjects, whereas IDDM patients did not (9.2 ± 4.5 vs 7.7 ± 2 μmol/l, p < 0.01; and 7.0 ± 3 vs 7.4 ± 2 μmol/l, NS). Univariate correlations and multiple regression analysis showed albumin excretion rate to be the parameter with the strongest independent association with homocysteine. Patients with both types of diabetes and nephropathy had higher plasma homocysteine concentrations than those without nephropathy. Increases of homocysteine in plasma were related to increases in the severity of the nephropathy. Fasting hyperhomocysteinaemia was considered as the mean of the plasma homocysteine for all control subjects (7.5 ± 2.1 μmol/l) + 2 SD (cut-off =11.7 μmol/l). Nephropathy was present in 80 % of diabetic patients with fasting hyperhomocysteinaemia. In conclusion, increases in fasting homocysteine in diabetic patients are associated with increased albumin excretion rate, especially in those with NIDDM, thus providing a potential new link between microalbuminuria, diabetic nephropathy and cardiovascular disease. [Diabetologia (1998) 41: 684–693]

Electronegative low density lipoprotein subform is increased in patients with short-duration IDDM and is closely related to glycaemic control

Summary We evaluated the effect of improving glycaemic control with intensive insulin therapy on LDL susceptibility to oxidation, electronegative LDL proportion, and LDL subfraction phenotype in a group of 25 patients with short-duration insulin-dependent diabetes mellitus (IDDM); 25 matched healthy control subjects were also studied. LDL susceptibility to oxidation was measured by continuous monitoring of conjugated diene formation. Electronegative LDL was isolated by anion exchange chromatography, and quantified as percentage of total LDL. Six LDL subfractions were isolated by density gradient ultracentrifugation and phenotype A or B classified as the quotient (LDL1-LDL3)/(LDL4-LDL6). Compared to the control group, IDDM subjects with poor glycaemic control showed higher electronegative LDL (19.03 ± 10.09 vs 9.59 ± 2.98 %, p < 0.001), similar LDL subfraction phenotype and lower susceptibility to oxidation (lag phase 45.6 ± 8.8 vs 41.2 ± 4.7 min, p < 0.05). After three months of intensive insulin therapy, HbA1 c decreased from 10.88 ± 2.43 to 5.69 ± 1.54 % (p < 0.001), and electronegative LDL to 13.84 ± 5.15 % (p < 0.05). No changes in LDL susceptibility to oxidation or LDL subfraction phenotype were observed. Electronegative LDL appeared significantly correlated to HbA1 c and fructosamine (p < 0.01 and p < 0.001) only in poorly controlled IDDM patients. These findings suggest that high electronegative LDL in IDDM subjects is related to the degree of glycaemic control, and could therefore be due to LDL glycation rather than to LDL oxidation or changes in LDL subfraction phenotype. [Diabetologia (1996) 39: 1469–1476]

The inflammatory properties of electronegative low-density lipoprotein from type 1 diabetic patients are related to increased platelet-activating factor acetylhydrolase activity

Aims/hypothesisChemical and biological characteristics of LDL(−) from type 1 diabetic subjects were analysed. The diabetic patients were studied during poor and optimised glycaemic control.Materials and methodsTotal LDL was subfractionated into electropositive LDL(+) and electronegative LDL(−) by anion exchange chromatography and the lipid and protein composition of the two determined.ResultsLDL(−) differed from LDL(+) in that it had higher triglyceride, non-esterified fatty acids, apoE, apoC-III and platelet-activating factor acetylhydrolase (PAF-AH), as well as lower apoB relative content. No evidence of increased oxidation was observed in LDL(−). LDL(−) increased two-fold the release of interleukin 8 (IL-8) and monocyte chemotactic protein 1 (MCP-1) in endothelial cells, suggesting an inflammatory role. Optimisation of glycaemic control after insulin therapy decreased the proportion of LDL(−), but did not modify the composition of LDL subfractions, except for a decrease in PAF-AH activity in LDL(−). The possibility that LDL(−) could be generated by non-enzymatic glycosylation was studied. Fructosamine and glycated LDL content in LDL subfractions from type 1 diabetic patients was greater than in LDL subfractions isolated from normoglycaemic subjects, and decreased after glycaemic optimisation in both subfractions. However, no difference was observed between LDL(+) and LDL(−) before and after insulin therapy.Conclusions/interpretationThese results provide evidence that LDL(−) is not produced by glycosylation. Nevertheless, LDL(−) from diabetic patients displays inflammatory potential reflected by the induction of chemokine release in endothelial cells. This proatherogenic effect could be related to the high PAF-AH activity in LDL(−).

Insulin Antibody Response to a Short Course of Human Insulin Therapy in Women With Gestational Diabetes

OBJECTIVE To assess the insulin antibody (IA) response to human insulin (HI) therapy in women with gestational diabetes. RESEARCH DESIGN AND METHODS IAs were measured by a competitive radiobinding assay in 50 women with gestational diabetes before and during treatment with HI and after delivery. At delivery, 15 maternal-cord blood sample pairs were analyzed for IA. As a reference, we searched for IA in 25 new-onset type I diabetic patients, before and at 3, 6, and 12 months after insulin therapy. RESULTS Insulin autoantibodies (IAAs) were detected in 1 of 50 women with gestational diabetes and 4 of 16 type I diabetic patients (P < 0.05). At the end of pregnancy after 9.3 ± 6.8 weeks on insulin therapy, 22 of 50 (44%) women with gestational diabetes became IA+ and 4 additional women were found to be positive 2 months postpartum. After 3 months on insulin, type I diabetic patients showed a higher rate of IA positivity (92%, P < 0.001). IA titers at the end of pregnancy were associated with the cumulative insulin dose (r = 0.29, P < 0.05). Postpartum, IA disappeared slowly in most IA+ women, but two women still showed IA 2 years after delivery Titers in cord blood were strongly related to those in maternal blood (r = 0.74, P < 0.01). The rate of adverse fetal outcome did not differ in IA− and IA+ mothers (27 vs. 40%, NS). CONCLUSIONS HI is immunogenic, and a short course of HI therapy induces IA in ∼ 50% of women with gestational diabetes and 92% of type I diabetic patients. In women with gestational diabetes, insulin dose is slightly associated with IA titers. These IAs apparently cross the placenta. Fetal outcome does not differ according to the maternal IA status, and IAs disappear gradually after delivery but may remain positive for 2 years after delivery.

Effects of a short-acting insulin analog (insulin Lispro) versus regular insulin on lipid metabolism in insulin-dependent diabetes mellitus

Insulin Lispro (IL) is a short-acting insulin analog that better reproduces the physiological postprandial insulin profile. The aim of this study was to compare the effects of intensive insulin therapy on lipid metabolism using preprandial IL and regular insulin (RI) in 10 insulin-dependent diabetes mellitus (IDDM) subjects. The mean hemoglobin A1c (HbA1c) at baseline was 7.13% +/- 1.2% and did not change after both treatments. In IDDM patients, total cholesterol and triglyceride levels appeared lower after RI than after IL. The low-density lipoprotein (LDL) to high-density lipoprotein (HDL) ratio significantly decreased only after RI (baseline, 2.01 +/- 0.6; IL, 1.88 +/- 0.6; RI, 1.71 +/- 0.5, P < .05). Although no very-low-density lipoprotein (VLDL) composition abnormalities were observed at baseline, the protein content was lower (P < .05) after IL (8.13% +/- 2.93%) than after RI (11.93% +/- 3.41%). Intermediate-density lipoprotein (IDL) protein depletion at baseline (6.14% +/- 6.84%) was normalized after both treatments (IL, 11.09% +/- 12.14%; RI, 10.38% +/- 16.68%, P < .05). LDL, HDL, HDL2, and HDL3 composition abnormalities were similar after both treatments and did not normalize. IDDM and control subjects showed similar LDL subfraction distribution at baseline and after both treatments. Two-hour postprandial VLDL composition alterations, although improved after RI, completely normalized after IL (P < .05). Lipoprotein lipase (LPL) and cholesteryl ester transfer protein (CETP) activities were similar to the control group and did not change after both treatments. Hepatic lipase (HL) activity was lower in diabetic patients (39.6 +/- 35.2 v 87.0 +/- 27.1 U/L, P < .01) and remained lower after both treatments. In conclusion, in IDDM patients, IL (injected immediately before the meal) may offer small different effects on lipoprotein metabolism versus RI (injected 30 minutes before the meal) that, taken together, do not seem relevant.

Lipoprotein compositional abnormalities in type I diabetes: Effect of improved glycaemic control

Major lipoprotein mass and composition were assessed in 45 subjects with insulin dependent diabetes mellitus (IDDM), before and after 2 months of intensive insulin therapy (IIT) and in 40 healthy control subjects. As compared to the control group, diabetic subjects at baseline had higher low density lipoprotein (LDL) and lower high density lipoprotein (HDL) masses. Expressing each lipoprotein constituent as a percent of total lipoprotein mass, very low density lipoprotein (VLDL) of diabetic patients was enriched in cholesterol and phospholipid and depleted in triglyceride and protein; IDL had higher triglyceride and phospholipid and lower cholesterol and protein proportion; LDL was depleted in protein and enriched in triglyceride; HDL was depleted in protein and enriched in triglyceride, cholesterol and phospholipid. After 2 months of IIT, HbA1c fell from 10.3 +/- 2 to 7.5 +/- 2% (P < 0.0001) and so did VLDL mass, which was lower than in control subjects. In addition, LDL and HDL masses, as well as triglyceride and cholesterol proportion in IDL particles normalized. The other compositional abnormalities improved without complete normalization. Thus, intensive insulin therapy in IDDM subjects brought quantitative lipoprotein alterations to normal even subnormal range, while most of the composition abnormalities improved without reaching complete normalization.

Plasma lipoprotein(a) levels are not influenced by glycemic control in type 1 diabetes.

OBJECTIVE To determine the influence of glycemic control improvement with intensive therapy on lipoprotein(a) [Lp(a)] concentrations in type 1 diabetic patients. RESEARCH DESIGN AND METHODS A total of 105 poorly controlled type 1 diabetic patients (60 men, 45 women) without diabetic complications participated in a longitudinal study performed in a tertiary referral center, to compare lipid, lipoprotein, and Lp(a) levels before and after 3 months of intensive therapy with multiple insulin doses. Lp(a) levels were measured by the Terumo method. Differences between the two periods were assessed by the paired t test and Wilcoxons test. RESULTS After 3 months of intensive therapy, all patients exhibited improved glycemic control. HbA1c decreased from 8.9 ± 2.4 to 6.5 ± 1.6% (P < 0.0001), being ≤6% in 47% of patients. However, although a more favorable lipoprotein profile was obtained, no changes in Lp(a) concentrations were observed in the whole group of patients (16.7 ± 17.3 vs. 17.2 ± 17.7 mg/dl) or in patients with baseline Lp(a) levels above 30 mg/dl (47.1 ± 14.8 vs. 47.4 ± 18.9 mg/dl) or below 30 mg/dl (9.6 ± 7.3 vs. 10.2 ± 6.7 mg/dl). In addition, patients reaching HbA1c ≤6 or >6% presented similar Lp(a) levels (19.7 ± 18.0 vs. 15.0 ± 17.4 mg/dl), and changes in Lp(a) did not correlate with those observed in HbA1c. CONCLUSIONS These data demonstrate that the improvement of glycemic control does not influence plasma Lp(a) concentrations in type 1 diabetic patients independently of baseline Lp(a) levels and the degree of glycemic control.

Genetic analysis does not confirm non-classical congenital adrenal hyperplasia in more than a third of the women followed with this diagnosis. Results of an audit.

Objective: To assess CYP21A2 allele mutations and review clinical characteristics of patients with a diagnosis of non-classical congenital adrenal hyperplasia Design: Audit. Patients: Twenty-nine women with a diagnosis of NCCAH were recruited when they came to their regular follow-up visits at the Endocrinology, Pediatrics and Gynecology Departments. Measurements: Medical records were examined to retrieve data about initial clinical presentation, later signs and symptoms, hormonal work-up: basal and stimulated 17-hydroxyprogesterone (17OHP), and treatment at diagnosis and follow-up. A standardized database was used. Analysis of the 21-hydroxilase gene was performed through polymerase chain reaction, sequencing, and family genetic testing when possible. Results: More than a third of the women, followed and treated according to a diagnosis of NCCAH did not have a confirmatory genetic diagnosis. Conclusion: Given the lack of genetic confirmation in at least a third of patients with suspected NCCAH, careful reassessment of diagnosis ought to be undertaken.