J Boillot

Hôtel-Dieu de Paris – Hôpitaux universitaires Paris Centre, Lutetia Parisorum, Île-de-France, France

Are you J Boillot?

Claim your profile

Publications (27)105.69 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To determine the impact of the form, fibre content, baking and processing on the glycaemic, insulinaemic and lipidaemic responses of different French breads. First study: Nine healthy subjects were randomized to consume in a crossover design one of six kinds of French bread (each containing 50 g available carbohydrate): classic baguette, traditional baguette, loaf of wholemeal bread (WM-B), loaf of bread fermented with yeast or with leaven, a sandwich and a glucose challenge as reference. The glycaemic index (GI) values ranged from 57+/-9% (mean+/-s.e.m.), for the traditional baguette, to 85+/-27% for the WM-B. No significant difference was found among the different tested bread. The insulinaemic index (II), however, of the traditional baguette and of the bread fermented with leaven were lower than the other breads (analysis of variance: P<0.01). Postprandial plasma triglycerides showed similar profiles. The traditional baguette tended to decrease postprandial free fatty acids compared to levels after the classic baguette. The GI of the traditional baguette was lower than that of the classic baguette (n=8, venous blood: 70+/-4 vs 75+/-4, P=0.002; capillary blood: 69+/-5 vs 83+/-6, P=0.028, respectively). Some varieties of French bread (the TB) have lower II, in healthy subjects, and lower GI, in type 2 diabetic subjects, than that of the other varieties. These results might be due to bread processing difference rather than fibre content. SPONSORSHIPS: Supported by grants from the National French Milling Association.
    European Journal of Clinical Nutrition 02/2007; 61(2):175-83. · 2.76 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine whether a chronic low-glycemic index (LGI) diet, compared with a high-glycemic index (HGI) diet, has beneficial effects on plasma glucose control, lipid metabolism, total fat mass, and insulin resistance in type 2 diabetic patients. Twelve type 2 diabetic men were randomly allocated to two periods of 4 weeks of an LGI or HGI carbohydrate diet separated by a 4-week washout interval, in a crossover design. The LGI diet induced lower postprandial plasma glucose and insulin profiles and areas under the curve than after the HGI diet. At the end of the two dietary periods, the 7-day dietary records demonstrated equal daily total energy and macronutrient intake. Body weight and total fat mass were comparable. Four-week LGI versus HGI diet induced improvement of fasting plasma glucose (P < 0.01, Delta changes during LGI vs. HGI), HbA(1c) (P < 0.01), and whole-body glucose utilization measured by the euglycemic-hyperinsulinemic clamp (P < 0.05). LGI diet induced a decrease in fasting plasma total and LDL cholesterol (Delta changes LGI vs. HGI, P < 0.01), free fatty acids (P < 0.01), apolipoprotein B, and plasminogen activator inhibitor 1 activity. Only 4 weeks of an LGI diet was able to improve glycemic control, glucose utilization, some lipid profiles, and the capacity for fibrinolysis in type 2 diabetes. Even if changes in glycemic control were modest during the 4-week period, the use of an LGI diet in a longer-term manner might play an important role in the treatment and prevention of diabetes and related disorders.
    Diabetes Care 08/2004; 27(8):1866-72. · 7.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We have evaluated the local tolerance and the metabolic efficacy of a lyophilized nasal insulin preparation in 10 severely hyperglycaemic Type 2 diabetic patients. The study included two 4-month randomized periods: (A) three preprandial doses of nasal insulin secondarily combined with one evening subcutaneous NPH if the desired glycaemic control was not achieved; (B) two NPH injections daily. We assessed: (i) diabetes control on monthly HbA1c levels and occurrence of hypoglycaemic events; (ii) local tolerance on clinical symptoms, rhinoscopy, nasal muco-ciliary clearance and nasal biopsies; (iii) insulin absorption at months 0 and 4. One patient was withdrawn because of cough and dizziness after each nasal application. HbA1c was not significantly different at month 4 (9.4 +/- 0.5% vs. 8.8 +/- 0.2%, A vs. B). Blood glucose control remained only fair in the majority of our patients. Nasal insulin was able to replace the daytime fraction of the subcutaneous insulin with a 18% efficacy. Side-effects included transient nasal hyperactivity (pruritus, sneezing and rhinorrhoea) and chronic persistence of nasal crusts. Plasma insulin profiles were not significantly different between months 0 and 4. The utilization of nasal insulin (with or without NPH) was associated with similar diabetes control compared with NPH twice daily. Nasal insulin alone was able to achieve an adequate glycaemic control in three of the 10 patients.
    Diabetic Medicine 09/2001; 18(8):614-8. · 3.24 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Nasal insulin administration is a potential route for intensive insulin management, less invasive and more rapid than subcutaneous injections. Previous studies have shown poor bioavailability (less than 15%) with nasal insulin administration with various absorption enhancers. The aim of the study was to evaluate in type 1 diabetic patients, the metabolic efficacy and local tolerance of a new gelified sprayed nasal insulin containing glychocolate and methylcellulose as absorption promoters. The study was conducted in 16 type 1 diabetic patients (HbA1c 8.6+/-0.2%) in a cross-over trial including 2 six month randomized periods: a) NPH twice daily + 3 pre-prandial nasal insulin doses + nasal supplementation in case of unexpected hyperglycaemia; b) NPH twice daily + 3 pre-prandial regular insulin injections. End points were HbA1c levels, hypoglycaemic episodes and tolerance evaluated at month 0, 2, 6 and 8 on clinical symptoms and objective nasal assessments. Four patients were withdrawn because of nasal burning (3 cases) and persistent sinusitis (1 case), and one patient had purulent sinusitis at the month 6 examination. At month 6, HbA1c levels were comparable (8.3 +/- 0.1 vs 8.6 +/- 0.1%, m +/- SEM, NS) for nasal and subcutaneous period respectively. The number of hypoglycaemic events was identical during the 2 periods (88 episodes). Nasal tolerance with the gelified form was better than with the already reported lyophilized form but, when present, symptoms were more marked, suggesting a potentiating additional role of methylcellulose excipient on nasal intolerance. 1) Gelified nasal insulin is as efficient as subcutaneous regular insulin in type 1 diabetic patients. 2) Other galenic forms should be investigated to improve nasal tolerance and bioavailability.
    Diabetes & Metabolism 07/2001; 27(3):372-7. · 2.39 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The of this study was to evaluate the chronic effects of a high (waxy corn) vs. a low (mung beans) glycemic index starch diet on the lipogenic enzymes, fatty acid synthase (FAS) and lipoprotein lipase (LPL). Normal and diabetic (streptozotocin-injected on d 2 of life) male Sprague-Dawley rats consumed a diet containing 575 g/kg carbohydrates either as waxy cornstarch (WCS) or as mung bean starch (MBS). After 3 wk, neither body weights nor relative epididymal fat pad weights differed. In diabetic rats, the WCS diet induced high basal plasma insulin levels. Plasma triglycerides were not significantly affected by diet in either normal or diabetic rats. Adipose tissue and liver LPL activities were not modified by the type of starch in the diet. In normal rats, FAS activity and gene expression in epididymal adipose tissue but not in liver were greater in rats consuming the WCS diet than in those consuming MBS. To evaluate the implication of insulin in this regulation, two genes regulated by insulin [GLUT4 and phosphoenolpyruvate carboxykinase (PEPCK)] were also studied. The high glycemic index WCS diet compared with the low glycemic index MBS diet resulted in lower hepatic PEPCK mRNA in both normal and diabetic rats. Normal, but not diabetic rats fed WCS had greater GLUT4 gene expression in adipocytes than did those fed MBS. We conclude that the total replacement of 575 g/kg low glycemic index starch by a high glycemic index starch for 3 wk caused the following in normal rats: 1) high FAS activity and mRNA in adipose tissue but not in liver and 2) high GLUT4 gene expression in adipose tissue. In both normal and diabetic rats this same diet resulted in lower hepatic PEPCK mRNA. Therefore, high glycemic index starch diet is implicated in stimulating FAS activity and lipogenesis and might have undesirable long-term metabolic effects.
    Journal of Nutrition 12/1998; 128(11):1878-83. · 4.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study was to evaluate the effects of the chronic consumption of two starches, characterized by different glycemic indices and amylose-amylopectin content, on glucose metabolism in rat epididymal adipocytes. The two chosen starches were from mung bean (32% amylose) and cornstarch (0.5% amylose). The alpha-amylase digestibility was higher for the waxy cornstarch than that of the mung bean starch (60 +/- 4 vs. 45 +/- 3%, mean +/- SEM, respectively). The glycemic index of the waxy cornstarch diet (575 g starch /kg diet) was higher than that of the mung bean starch diet (107 +/- 7 vs. 67 +/- 5%, P < 0.01) when measured in vivo in two groups of normal rats (n = 9). In a subsequent study, normal and diabetic (streptozotocin-injected on d 2 of life) male Sprague-Dawley rats (18 per group) consumed a diet containing 575 g starch/kg diet as either waxy cornstarch or mung bean starch. After 3 wk, food intake, epididymal fat pad weights, and plasma glucose, insulin and triglyceride concentrations did not differ between diet groups. Adipocyte diameter was smaller in rats that consumed mung bean starch compared with those that consumed the waxy cornstarch diet (P < 0.01). The mung bean diet increased maximal insulin-stimulated 14C-glucose oxidation (% of basal values, P < 0. 05). In contrast, incorporation of 14C-glucose into total lipids was significantly lower in rats that consumed the mung bean diet (P < 0. 05). We conclude that in both normal and diabetic rats, the chronic replacement of a high glycemic index starch by a low glycemic index one in a mixed diet increases insulin-stimulated glucose oxidation, decreases glucose incorporation into total lipids and decreases epididymal adipocyte diameter. Thus, the type of starch mixed into the diet has important metabolic consequences at the cellular level in both normal and diabetic rats.
    Journal of Nutrition 01/1998; 128(1):35-43. · 4.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To study the effects of dietary fish oil on insulin-stimulated glucose metabolism in adipocytes of insulin-resistant rats (rats fed 50% sucrose and 30% fat), eighteen 5-wk-old Sprague-Dawley rats were fed, for 6 wk, a diet containing 30% fat as either fish oil (FO) or a mixture of vegetable and animal oils [control oils (CO)]. A third reference group was fed a standard diet (62% corn starch and 13% fat). At the end of the 6-wk period, the two experimental groups had comparable plasma glucose concentrations that were higher than that found in the reference group. FO feeding corrected the hyperinsulinemia of the experimental rats (P < 0.05) to reach values in the reference group. Plasma triacylglycerol (P < 0.01) and cholesterol (P < 0.001) concentrations were also lower in rats fed FO than in those fed CO. The body weights of FO-fed rats were similar to that of CO-fed rats, but epididymal adipose tissue weight was lower (P < 0.01). Adipocytes of FO-fed rats, compared with those of CO-fed rats, had high insulin-stimulated glucose transport (P < 0.05), oxidation (P < 0.001) and incorporation into total lipids (P < 0.05). The incorporation of (n-3) polyunsaturated fatty acids in adipocyte membrane phospholipids was higher in FO-fed rats than in those fed CO (P < 0.0001). Insulin action was positively correlated with the fatty acid unsaturation index in membrane phospholipids. Thus dietary fish oil has beneficial effects on insulinemia, plasma lipids and insulin-stimulated glucose metabolism in insulin-resistant slightly diabetic rats.
    Journal of Nutrition 09/1996; 126(8):1951-8. · 4.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aims of the study were to compare glycohaemoglobin (HbA1c) values measured by DCA (a benchtop analyzer primarily designed for within-clinic rapid HbA1c determination) to a reference HbA1c method and home blood glucose monitoring, and to explore the possibility of an uniform expression of data. A total of 103 blood samples and the corresponding mean capillary glucose values (4.4 +/- 1.2 tests/day) of the preceding 2 months were collected from 34 insulin-dependent diabetic adults. We measured the correlations and agreements using the residual plots method and regression equations between HbA1c measured by DCA and high-pressure liquid chromatography (HPLC), and between DCA and capillary glucose values. A highly significant correlation (r2 = 0.85, P < 0.001) and an acceptable agreement (97% of values within 2 SD of the mean difference of 0.9% +/- 0.4%) was found between DCA and HPLC values. The regression equation calculated on the first half of the cases was: DCA (%) = 0.72 HPLC (%) +1.38. Of DCA values expressed in HPLC terms using this equation 87% fell within a clinically acceptable confidence interval when compared with measured HPLC data. A significant correlation (r2 = 0.40, P < 0.01) was found between DCA and capillary glucose values, and the regression equation was: DCA (%) = 0.34 capillary glucose (mM) +4.44. Of glycaemic levels calculated from DCA values using this formula 82% fell within a clinically acceptable error range when compared with measured glycaemic values. We conclude that the three methods of assessment of diabetes control are well correlated and that it is possible, with a degree of precision acceptable for the clinical setting, to express all data in uniform units, e.g. mM of capillary glucose or percentage of HPLC-HbA1c, though a simple correspondence table based on our transfer equations may be clinically sufficient and more handy.
    Acta Diabetologica 09/1996; 33(3):232-5. · 4.63 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Dietary fiber intake is associated with several beneficial effects on carbohydrate metabolism. Some authors have speculated that this improvement may be due to short-chain fatty acids (SCFA) produced by the colonic fermentation of dietary fibers. To test this hypothesis, six healthy men aged 26 +/- 2 (SE) yr with a body mass index of 20.9 +/- 0.7 received on three occasions an 18-h ileal perfusion infused at a flow rate of 3.3 ml/min, containing either 90 mmol/l of SCFA (60% acetate, 25% propionate, and 15% butyrate) (A), SCFA during the first 12 h and then a saline solution (A/S), or a saline solution (S). Basal hepatic glucose production (BHGP), insulin sensitivity (3-step euglycemic-hyperinsulinic clamp), and erythrocyte insulin binding (EIB) were studied 12 h after the beginning of the ileal perfusion. There was no change in BHGP or insulin sensitivity. However, maximal EIB was significantly different: 7.1 +/- 0.1 (A), 6.8 +/- 0.1 (A/S), vs. 6.5 +/- 0.1% (S) (P = 0.03). We conclude that acute administration of SCFA does not significantly alter glucose metabolism in healthy subjects.
    The American journal of physiology 08/1996; 271(1 Pt 1):E199-204. · 3.28 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To study the cellular mechanisms underlying fructose-induced insulin resistance in rats, the effects of fructose feeding on insulin-stimulated glucose transport, oxidation and incorporation into lipids in epididymal adipocytes were evaluated in 27 normal and 27 noninsulin-dependent diabetic male Sprague-Dawley rats. Diabetes was induced by streptozotocin injection 2 d after birth. At 5 wk of age, both normal and diabetic rats were fed a diet containing 62% carbohydrate as fructose, dextrose or cornstarch. Fructose feeding for 6 wk induced glucose intolerance in normal rats (P < 0.05) and aggravated that of diabetic rats (P < 0.05). Plasma triacylglycerol concentration was higher in fructose-fed than in starch-fed or dextrose-fed rats (P < 0.05). Adipocytes of fructose-fed rats had significantly lower maximum insulin-stimulated glucose incorporation into total lipids than those of rats fed starch, and tended (P = 0.22) to have lower production of CO2 from glucose than adipocytes of the other dietary groups. Glucose transport in adipocytes of dextrose-, starch- and fructose-fed rats did not differ. We conclude that in both normal and diabetic rats, a chronic fructose-rich diet induced hypertriacylglycerolemia, glucose intolerance and insulin resistance of adipocytes.
    Journal of Nutrition 03/1995; 125(2):164-71. · 4.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Increased intake of dietary fibres is associated with several beneficial effects on carbohydrate and lipid metabolism. The colonic fermentation of dietary fibres produces short-chain fatty acids (SCFA; acetate, propionate and butyrate). Some authors have suggested that SCFA could be partly responsible for the effects of dietary fibres. The purpose of the present study was to test the effects of one of the SCFA, propionate. The effects of moderate amounts of dietary propionate on insulin sensitivity and hepatic glucose production were studied in male Sprague-Dawley rats. Two groups of twenty-one adult rats were fed for 3 weeks on a diet containing 78 g propionate/kg (P) or 78 g/kg of a poorly fermentable cellulose (control group; C). Feed intake, body weight, fasting plasma glucose, insulin, free fatty acids, alanine, lactate, glycerol and beta-hydroxybutyrate levels were measured weekly in anaesthetized rats. At the end of the feeding period basal hepatic glucose production (BHGP) was measured with a primed continuous infusion of [3-3H]glucose and the in vivo insulin sensitivity in rats was quantified by the euglycaemic-hyperinsulinaemic clamp technique (0.6 and 2 U/kg per h). At that time fasting plasma glucose measured in anaesthetized rats was significantly lower in group P than in group C: 7.7 (SE 0.2) v. 8.5 (SE 0.2) mmol/l respectively (P < 0.002); plasma insulin levels were not significantly different. Neither the BHGP (mg/min per kg; C 14.8 (SE 1.3), P 15.1 (SE 1.3); n 7, not significant) nor the basal metabolic clearance (ml/min per kg; 8.9 (SE 0.8) v. 9.9 (SE 1.1); not significant) were different between treatments. Hepatic glucose production and glucose utilization at the two insulin concentrations (approximately 500 and 1500 mU/l respectively, n 7) did not differ significantly between the two groups. These results show that dietary propionate chronically ingested by normal rats could decrease fasting glycaemia, but from our findings, no effect on hepatic glucose production and whole-body glucose utilization could be clearly demonstrated.
    British Journal Of Nutrition 02/1995; 73(2):241-51. · 3.30 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of the present study was to investigate how guar gum viscosity acts on starch digestion and glucose absorption in humans. Six healthy subjects received a mixed diet composed of 60.4% carbohydrate in the form of maize glucose or pregelatinized starch, to which was added 5.6% low- or high-viscosity guar gums. Meals were ingested or instilled in the duodenum and postprandial insulin and glucose responses were monitored for 3 h. Infusion of meals containing glucose showed that the delay in the diffusion rate to the duodenal mucosa due to bolus viscosity was not significant. Infusion of meals containing starch showed that a decrease in the digestion rate of starch in the upper small intestine accounted for part of the effect of viscosity on glycemic response, whereas the main effect of guar gum was apparently to slow gastric emptying.
    American Journal of Clinical Nutrition 05/1994; 59(4):914-21. · 6.50 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Sucrose feeding over a long period has been reported to induce glomerular basement membrane (GBM) thickening and insulin resistance in normal rats. These effects are attributed to the fructose moiety of the sucrose molecule, to Cu deprivation or both. Consequently, our aim was to evaluate the long-term effects of fructose feeding with normal or high amounts of Cu on body weight, plasma lipids, blood glucose regulation, GBM thickening and insulin binding to adipocytes. Four groups of eight Sprague-Dawley rats were fed for 10 weeks on a diet containing 570 g carbohydrate/kg supplied either as starch (S), dextrose (D), fructose (F) or fructose-starch (1:1, w/w; FS), and an adequate amount of Cu (12 micrograms Cu/g diet). A fifth group was fed on diet F supplemented with 24 micrograms Cu/g diet (FCu). After 10 weeks the epididymal adipose tissue and kidney weights expressed per 100 g body weight (relative weight) were heaviest in the F and FCu groups (P < 0.0001, ANOVA). The GBM thickness was within the normal range in the five groups but significantly higher in group D (1.95 (SE 0.04) nm and lower in group FS (1.79 (SE 0.02) nm when compared with group S (1.85 (SE 0.03) nm; P < 0.05). Insulin binding to adipocytes (expressed per cell) was lowest in the F and FCu groups, intermediate in groups D and FS and highest in group S (P < 0.05). Fasting plasma insulin level was higher in group F than in the FCu and FS groups (P < 0.05), whereas fasting plasma glucose, total cholesterol and triacylglycerol levels remained within the normal range in all groups. We conclude that in normal rats a 10-week fructose-rich diet with an adequate amount of Cu produced deleterious metabolic effects on adipose tissue, insulin binding to adipocytes, and plasma insulin, but not on GBM thickening even though kidney weight was significantly increased. However, a moderate fructose intake mixed with other sugars did not have adverse effects.
    British Journal Of Nutrition 08/1993; 70(1):199-209. · 3.30 Impact Factor
  • Journées annuelles de diabétologie de l'Hôtel-Dieu 02/1993;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Because we found previously that fructose feeding could alter lipolytic responses to isoproterenol and insulin in normal rats, we studied the effects of the same diet in neonatal, streptozotocin-diabetic rats. Twenty-seven 5-wk-old diabetic Sprague-Dawley rats were fed a diet containing 57% carbohydrate as either fructose, dextrose or starch for 6 wk. At the end of the nutritional period, plasma glucose and insulin concentrations in fed rats were similar in the three diabetic groups. Plasma triacylglycerol concentrations were higher in the fructose-fed group than in the other two groups (P < 0.05). Neither the maximal adipocyte lipolytic response (fructose = 1147 +/- 165%, starch = 1823 +/- 329% and dextrose = 1287 +/- 239% of basal values) nor the sensitivity to isoproterenol (ED50) was changed by the dietary carbohydrate exchange. The maximal antilipolytic action of insulin (starch = 68 +/- 10%, dextrose = 41 +/- 13%, fructose = 95 +/- 29% of stimulated lipolysis values) was comparable in the three diet groups. Thus, 6 wk of fructose feeding in diabetic rats increased plasma triacylglycerol concentrations, but had no detectable effect on plasma glucose or insulin concentrations, isoproterenol-induced lipolysis or the antilipolytic action of insulin.
    Journal of Nutrition 12/1992; 122(12):2361-6. · 4.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The precise effects of fructose feeding on adipose tissue is not clearly known. Consequently, we studied the effects of fructose feeding on stimulated and inhibited in vitro lipolysis. Twenty seven male Sprague Dawley rats, 5 weeks of age, were fed for 6 weeks on one of three diets containing 57% CHO as fructose (F), dextrose (D) or starch (S). At week 6 the epididymal fat pad weights showed no difference between groups. Stimulation of lipolysis by isoproterenol or theophylline showed: decreased sensitivity of adipocytes to isoproterenol, but not to theophylline, in F (p less than 0.05); the maximal responses were decreased, but NS, after stimulation by either isoproterenol or theophylline. The maximal antilipolytic responses to insulin were increased in F (27%) and D (29%) when compared to S (16%), (p less than 0.05). Only, in F there was an increase (NS) in ED50 (0.63 +/- 0.23 ng/ml) compared to D (0.45 +/- 0.18) and S (0.29 +/- 0.18), indicating decreased sensitivity. Nonfasting plasma insulin and triglycerides were increased at the 6th week in F (p less than 0.01), without any change in plasma glucose levels. However, there was no difference in 12 h fasting plasma glucose, insulin or triglycerides. In conclusion, a 6 week 57% fructose containing diet in normal rats led to: 1) decreased lipid mobilization in the epididymal adipose tissue; and 2) increased nonfasting plasma insulin and triglycerides. Thus fructose, under these experimental conditions, seems to have adverse metabolic effects in normal rats.
    Molecular and Cellular Biochemistry 03/1992; 109(2):127-32. · 2.33 Impact Factor
  • 01/1991;
  • Transplantation Proceedings 05/1990; 22(2):724-6. · 0.95 Impact Factor
  • Hormone and metabolic research. Supplement series 02/1990; 25:200-2.
  • Journées annuelles de diabétologie de l'Hôtel-Dieu 02/1989;

Publication Stats

397 Citations
105.69 Total Impact Points

Institutions

  • 1992–2007
    • Hôtel-Dieu de Paris – Hôpitaux universitaires Paris Centre
      Lutetia Parisorum, Île-de-France, France
  • 1996–1998
    • Pierre and Marie Curie University - Paris 6
      Lutetia Parisorum, Île-de-France, France
  • 1992–1996
    • French Institute of Health and Medical Research
      Lutetia Parisorum, Île-de-France, France