[Screening and diagnosis of gestational diabetes. Evaluation of the methodological quality of the guidelines of the Haute Autorité de Santé, of the American Diabetes Association, and of the World Health Organisation]. / Dépistage et diagnostic du diabète gestationnel: évaluation de la qualité méthodologique des avis de la Haute autorité de santé, de l'American diabetes association et de l'Organisation mondiale de la santé

We critically appraised the methodological quality of the clinical practice guideline (CPG) published by the Haute autorité de santé (HAS) about screening and diagnosis of gestational diabetes, and we compared its quality with that of two other CPGs, i.e. that of the American diabetes association (ADA) and that of the World health organisation (WHO). According to the AGREE criteria, HAS and ADA have produced CPGs that have approximately got the same levels of quality. Both these CPGs obtain AGREE scores that are better than those of WHO. Although the CPG of the HAS suffers from a few methodological drawbacks, regarding more particularly stakeholder involvement (AGREE domain n degrees 2), applicability (AGREE domain n degrees 5) and editorial independence (AGREE domain n degrees 6), this CPG summarises, and allows to compare most, if not all, other CPGs available with each other, with their possible benefits or harms, which may be useful for professionals involved in the care of the patient.

[Practice guidelines: let us sort them]. / Guides de pratique clinique: trions-les.

A growing number of clinical practice guidelines (CPG) is published. This is understandable because CPG are the corner stone in the evaluation of professional practices (EPP). One cannot deny that EPP is necessary. However, in order for the EPP to reach their objectives, which are to use our resources better and to improve health-care, CPG at our disposal should be of good quality, both in their form and in their content. This is not always the case. What is more, health-care professionals are often not properly trained to distinguish "good" from "not so good" CPG. In this context, the Société française de biologie clinique has created a working group on "CPG and Evidence-Based Laboratory Medicine (EBLM)". One of the main objectives of our group is to publish critical appraisals of CPG on a regular basis in the Annales de Biologie Clinique (ABC). Thus, the ABC will follow the example set by other medical journals, for example in France: Prescrire. We will more particularly appraise CPGs in relation with laboratory medicine. In this first article, we describe the methods that we will use in order to distinguish "good" from "not so good" CPG. Just like Prescrire as well as like many others, our first tool will be the AGREE instrument, which is quite consensual at an international level. The AGREE tool makes it possible to appraise quite easily, and in a reproducible way, the methodological quality of CPG. We also briefly discuss the more complicated methods that can be used to make judgments about the content of CPG, bearing in mind that equity, patients' autonomy, balancing risks and benefits, are the four universal principles of medical ethics, that is of good medicine, that is of EB(L)M.

trans-Acting factors, detoxication enzymes and hepatitis B virus replication in a novel set of human hepatoma cell lines.

A panel of four novel human hepatoma cell lines was isolated from a single tumor from a male individual. BC1, B16 and B16A2 lines were well differentiated, while cells of the B9 line were only poorly differentiated, being essentially negative for the functions analyzed. These cell lines have been surveyed for expression of a large set of plasma proteins, accumulation of liver-specific mRNAs and DNA-binding activity of ubiquitous and liver-enriched transcription factors. BC1 cells expressed the highest levels of albumin mRNA, whereas B16 and B16A2 cells accumulated the largest amounts of haptoglobin mRNA. In addition, B16 and B16A2 cells were unique in that they expressed CYP2E1 mRNA, a species absent from the available human liver cells, including HepG2 hepatoma cells, and 3-methylcholanthrene-inducible CYP1A2 mRNA. The activities of genes encoding transcription factors were evidenced in all four cell lines which expressed mRNAs for nuclear factor interleukin 6 and hepatocyte nuclear factor 1 (HNF) together with the DNA-binding activity of NFY and AP1 nuclear proteins. Strikingly, HNF-1 and HNF-4-like DNA-binding activities were restricted to BC1, B16 and B16A2 cells, supporting the idea of the potential role of these (or closely related) factors in the maintenance and/or in the establishment of the differentiated phenotype. B9 cells contained variant HNF1-like DNA-binding activity, similar to dedifferentiated rat hepatoma cells of the H5 line. CCAAT/enhancer-binding protein and HNF-3-like activities were found in all cell lines, although at a lower level and/or activity in B9 cells. Finally, transfection experiments of plasmids containing the whole hepatitis-B virus genome demonstrated that B16 cells, but not B9 cells, were able to support hepatitis-B virus replication and virion production, in agreement with the notion that HNF-1 activity is necessary for viral replication. We believe that the specific complement of transcription factors expressed in the differentiated BC1, B16 and B16A2 cells, and in the poorly differentiated B9 cells, will allow studies on the regulation of hepatic gene expression in these human lines, and will also aid the analysis of xenobiotic metabolism and the biology of hepatitis-B virus replication.

[Evaluation of assay methods of glycohemoglobin: reference method and parameter]. / Evaluation des méthodes de dosage de l'hémoglobine glyquée: méthode et paramétre de référence.

In this paper, the characteristics of the analytical procedures allowing the measurement of glycated hemoglobin products are presented. With respect to their respective performances, the authors recommend the specific measurement of HbA1c and the more reliable procedure, high pressure liquid chromatography (HPLC).

Interleukin 4 inhibits the production of some acute-phase proteins by human hepatocytes in primary culture.

Interleukin 4 (IL4) has been shown to exhibit anti-inflammatory effects by inhibiting the secretion by monocytes of proinflammatory cytokines such as interleukin 1 (IL1), interleukin 6 (IL6), and tumor necrosis factor (TNF) and by inducing the secretion of the IL1 receptor antagonist. We investigated the role of this cytokine on the production of acute-phase proteins in primary human hepatocyte cultures. Cells were exposed to either IL4 and/or IL6, the most potent mediator of hepatic acute phase proteins. IL4 led to decreased production of haptoglobin, C-reactive protein and albumin while alpha 1-antitrypsin and fibrinogen remained unaffected. These inhibitory effects of IL4 were also observed at the mRNA level. In addition, IL4 inhibited the IL6-induced production of haptoglobin although it had no effect on the induced C-reactive protein and fibrinogen. Our results demonstrate that IL4 can affect the production of a subset of acute-phase proteins by human hepatocytes and can antagonize some of the effects of IL6. These observations reinforce the notion that IL4 can be considered as an anti-inflammatory cytokine.

[Erythrocyte ferritin]. / La ferritine érythrocytaire.

Red cell ferritin is a residue of erythroblast ferritin. It reflects the balance between the iron supply to the erythroid marrow and the need for haemoglobin synthesis. Erythrocyte ferritin can be measured in haemolysates after discarding plasma and leucocytes by different methods. The decrease in erythrocyte ferritin content indicates manifest iron deficiency anaemia. Ferritin levels do not appear to be influenced by inflammation, infection, tissue necrosis or tumors and may be a reliable indicator of iron status in inflammatory diseases. Erythrocyte ferritin is markedly increased in patients with iron overload allowing early diagnosis of hereditary haemochromatosis and the monitoring of phlebotomy therapy. Finally, in some pathologies erythrocyte assay is a more reliable indicator of the iron status than serum ferritin.

[Serum ferritin assay. Value and limitations]. / Détermination de la ferritine sérique. Intérêt et limites.

There is a good correlation between serum ferritin and the amount of iron stored in the body. Reduced levels of serum ferritin are always found in iron deficiency but elevation of serum ferritin can occur without any iron overload. According to pathological situations, red cell ferritin, glycosylated ferritin or acidic ferritin measurements are useful tests for the differential diagnosis of increased serum ferritin concentration.

Alcohol induction of ferritin expression in a human hepatoblastoma cell line (HEP G2).

Hyperferritinemia, an unclear mechanism, is frequently observed in chronic alcoholics. The aim of this work was to study the effect of alcohol on ferritin expression in a human hepatoblastoma cell line, HepG2. This cell line proved to be sensitive to alcohol, since alcohol increased gamma-GT activity both in cells and media. The most striking result was the increase of ferritin in cells and media by alcohol. Moreover, this effect was specific, since it contrasted with a decrease in total protein synthesis and secretion, a decrease in transferrin excretion and a lack of effect on orosomucoid. In our model, alcohol was able to induce, in a specific manner, ferritin expression.

Iron induction of ferritin synthesis and secretion in human hepatoma cell (Hep G2) cultures.

In order to determine if iron was able to stimulate specifically ferritin synthesis and secretion in transformed human hepatocytes in culture, human hepatoma cell (HepG2) cultures were submitted to increasing doses of ferric nitrilotriacetate. Iron uptake by the cells was demonstrated by incorporation of 59 Fe and the staining method of Perls. The following results were obtained: 1. iron incorporation within the hepatocytes increased as a function of culture time; 2. during the first 24 h of treatment, ferritin synthesis increased progressively, in parallel to the iron uptake; 3. a dose-dependent significant stimulation of ferritin synthesis and secretion were observed when the medium iron concentration increased from 5 to 20 mumol/l; 4. albumin, transthyretin and transferrin secretions were unaffected. These data demonstrated that, in our hepatocyte culture model, iron load increased the expression of ferritin in a highly specific manner.

[Reference values of erythrocyte ferritin in children and adults]. / Valeurs de référence de la ferritine érythrocytaire chez l'enfant et l'adulte.

The erythrocyte ferritin content was measured in 183 healthy subjects in age from 4 to 68 years; 80 were male and 103 were female. In children between the ages of 4 and 12 years there is no significant difference in the mean value between boys and girls. In females the erythrocyte ferritin concentration is independent of age. After 12 years of age the erythrocyte ferritin content is higher in men. Reference intervals were determined by the two quantiles 0.05 and 0.95. The reference interval is 3-24 age by cell in boys between 4 and 12 and females between 4 and 63; the reference interval is 5-38 age by cell in males between 13 and 68 years of age.

[Evaluation of a qualitative test for the demonstration of microalbuminuria]. / Evaluation d'un test qualitatif pour la mise en évidence d'une microalbuminurie.

ALBU SCREEN is a commercial latex inhibition agglutination slide test to detect subclinical elevation of urinary albumin (microalbuminuria). We have appraised two different lots of latex with a sensitivity of 30 and 20 mg/l respectively. The results were compared to immunonephelemetric measurements of urinary albumin. 195 diabetic subjects urines (24 hours) were tested. 182 patients are well placed with the latex test (38 positive and 144 negative). This test could be an interesting substitute to the usual dipstick techniques, to look after potential nephropathic diabetic patients.

[Fructosamine and HbA1c during monitoring of glucose levels in diabetes]. / Fructosamine et HbA1c dans la surveillance de l'équilibre glycémique chez les diabétiques.

The authors compared the nycthemeral variations of fructosamine levels in nineteen patients with diabetes to fluctuations in their glucose levels. Thirty-five other patients with poorly controlled diabetes were given an optimal treatment then followed up for 2 months (mean glucose levels, fructosamine and HbA1c assays). Mean nycthemeral variations in fructosamine levels did not appear to be correlated with glucose levels. However, intraindividual nycthemeral variability was greater than the analytic variability. After correction for total protein levels this variability was smaller but remained significative. In this longitudinal study, fructosamine and HbA1c levels decreased when control of glucose levels improved. However, decreases were notably greater in HbA1c than in fructosamine compared at 15, 45 and 60 days of follow up. Inversely, fructosamine was correlated better with the mean glucose levels of the preceding 7 days than was HbA1c. When mean glucose levels were calculated for the 15, 30, 45 and 60 preceding days, the results favored HbA1c. The findings presented in this study show that HbA1c remains the best criteria for monitoring medium term glucose levels. However, further word is needed before this conclusion can be confirmed.

[Value of the urinary albumin/creatinine ratio in the detection of microalbuminuria]. / Intérêt et valeur du rapport albumine/créatinine urinaires dans le dépistage d'une microalbuminurie.

Microalbuminuria is currently defined as a urinary albumin excretion rate of 20 to 200 micrograms per minute, measured in the urine of 24 hours. We present an indirect approach to the urinary albumin daily excretion rate which minimizes the errors due to chronometric measurement and to the difficulty of collecting 24-hour urine. The albumin (mg/l)/creatinine (mmol/l) ratio calculated in urine collected daily indicates microalbuminuria when it is higher than 2.97 in women and 2.48 in men. The interpretation of this ratio is discussed in terms of sensitivity, specificity and predictive value.

[The coefficient of saturation of iron in transferrin]. / A propos du coefficient de saturation en fer de la transferrine.

Theoretical iron fixation capacity of transferrin (FCT) can be calculated on its immunochemical titration: (FCT (mumol/l = transferrin (g/l) x 25). Today, its reckoning is more advisable to serum total iron binding capacity measurement. The authors studied the effects of this new proceeding upon usual values interval of transferrin saturation (i.e. serum iron/FCT ratio). The mean value and the distribution of transferrin saturation appear displaced with regard to those achieved by chemical measurement of serum total iron binding capacity. We discuss interpretation of transferrin saturation related to its methods of determination and its semiological interest.

[Comparison of 2 methods of measuring microalbuminuria. Immunonephelometry and radioimmunology]. / Comparaison de deux méthodes de dosage de la microalbuminurie. L'immunonéphélémétrie et la radioimmunologie.

The early detection of microalbuminuria in insulin dependent diabetes is considered as a sign of initial stage of nephropathy (possibly reversible if glycemic balance is well maintained). This detection requires very accurate methods as radioimmunoassays. Yet, they are so slow that they represent an obstacle to systematic detection. We report an appraisement of an immunonephelemetric method. Results reveal that immunonephelemetry is a sensitive and accurate method (threshold of sensitivity 0.5 mg/l; CV intra assay less than 5%; CV inter assays less than 10%; analytical recovery: 94-104%). Moreover, immunonephelemetry and radioimmunology are significantly well correlated (r = 0.977, p less than 0.001). As a conclusion, we can say that thanks to complete automation, immunonephelemetry is a choice method to test great lines of samples.

[Reference values of serum transferrin in newborn infants, children and adults]. / Valeurs de référence de la transferrine sérique chez le nouveau-né, l'enfant et l'adulte.

The authors have studied the evolution of serum transferrin concentration in relation with age (newborn child, infant, adult). The mean concentration of serum transferrin is the lowest for neonates (2.15 g/l). At 10 months, it increases to reach a value of about 3.10 g/l then reduces at 24 (3.0 g/l) and 48 months (2.8 g/l). By the 4 to 18 years subjects, serum transferrin is approaching that of 10 months children (3.15 g/l). At last it doesn't support any modification during all the adult life. The diversity of the obtained results dependent of the used technique and/or of the laboratory, justifies that each one defines his own reference values.