Amoxicillin-clavulanic acid prescriptions at the Greater Paris University Hospitals (AP-HP).

OBJECTIVE: We aimed to document amoxicillin-clavulanic acid prescription to improve the proper use of antibiotics in hospital settings. We used three criteria: quality of medical charts, adequacy of indications, and adequacy of treatment duration. METHOD: This study was designed as a one-day point prevalence survey carried out by antibiotic lead specialists. RESULTS: We included 387 prescriptions from 32 hospitals. Immunodeficiency was recorded as a risk factor in 30% of patients. Computerized prescriptions were observed in 79% of cases. The indication was mentioned in 73% of cases and a 48/78-hour re-assessment of the antibiotic therapy was performed in 54% of cases. The antibiotic indication was primarily for pneumonia and was deemed appropriate in 75% of patients. Adult mean treatment duration was 11.1 days. Use of dual combination therapy and/or treatment duration exceeding two weeks accounted for the main reasons for an inappropriate use of antibiotics. Prescriptions recorded as having been made by senior physicians were of the shortest treatment duration (P=0.0163). CONCLUSION: Medical charts should be better filled in. Reinforcing the role of senior physicians in supervising antibiotic prescriptions is likely to result in a better control of treatment duration and ultimately in a reduced antibiotic consumption. By reinforcing the collaboration between pharmacists and antibiotic lead specialists, the improvement of computerized prescriptions at hospital level should help better detect the "at risk" prescriptions, namely those exceeding seven days or those combining antibiotics.

[Introduction of Health technology assessment at hospital]. / Mise en place de l'évaluation des technologies de santé en milieu hospitalier.

INTRODUCTION: The Medical Devices Committee (CODIMS) of the Assistance publique-Hôpitaux de Paris (AP-HP) is responsible for deciding whether innovative and costly sterile medical devices (SMD) should be adopted for the AP-HP network and for issuing recommendations on their proper use. The aim of this study was to qualify retrospectively the level of evidence of clinical studies used for the device evaluations by the CODIMS in 2012 and 2013 and to analyze the relationship between levels of evidence and decisions. MATERIAL AND METHOD: Executive summaries written in 2012 and 2013 about studied SMD was analyzed and the level of evidence of clinical studies used was qualified in high/low levels of evidence according to the scale of Sackett et al. Then, levels of evidence were correlated to decisions published by the CODIMS. RESULTS: Sixty-one files of SMD (72.1% of implantable MD) have been evaluated (225 clinical studies). Among them, only 28% of clinical studies had a high level of evidence (and 28.6% of MD at-risk) and 18% did not have any clinical studies. The CODIMS delivered an unfavourable opinion for 16 SMD: only 28 clinicals studies were available for evaluation. Among these, only 6 studies had a high level of evidence. DISCUSSION ET CONCLUSION: The amount and level of evidence of clinical studies is naturally correlated to admittance of SMD. These findings suggest that the clinical evidence used to demonstrate safety and efficacy for high-risk medical devices is based on clinical studies with poor quality data, making more difficult the evaluation of SMD in hospital. The development of a multi-criteria tool to help decision-making would improve the process of SMD evaluation by the CODIMS.

Development of a Fourier transform infrared spectroscopy coupled to UV-Visible analysis technique for aminosides and glycopeptides quantitation in antibiotic locks.

Antibiotic Lock technique maintains catheters' sterility in high-risk patients with long-term parenteral nutrition. In our institution, vancomycin, teicoplanin, amikacin and gentamicin locks are prepared in the pharmaceutical department. In order to insure patient safety and to comply to regulatory requirements, antibiotic locks are submitted to qualitative and quantitative assays prior to their release. The aim of this study was to develop an alternative quantitation technique for each of these 4 antibiotics, using a Fourier transform infrared (FTIR) coupled to UV-Visible spectroscopy and to compare results to HPLC or Immunochemistry assays. Prevalidation studies permitted to assess spectroscopic conditions used for antibiotic locks quantitation: FTIR/UV combinations were used for amikacin (1091-1115cm(-1) and 208-224nm), vancomycin (1222-1240cm(-1) and 276-280nm), and teicoplanin (1226-1230cm(-1) and 278-282nm). Gentamicin was quantified with FTIR only (1045-1169cm(-1) and 2715-2850cm(-1)) due to interferences in UV domain of parabens, preservatives present in the commercial brand used to prepare locks. For all AL, the method was linear (R(2)=0.996 to 0.999), accurate, repeatable (intraday RSD%: from 2.9 to 7.1% and inter-days RSD%: 2.9 to 5.1%) and precise. Compared to the reference methods, the FTIR/UV method appeared tightly correlated (Pearson factor: 97.4 to 99.9%) and did not show significant difference in recovery determinations. We developed a new simple reliable analysis technique for antibiotics quantitation in locks using an original association of FTIR and UV analysis, allowing a short time analysis to identify and quantify the studied antibiotics.

Genetic background of Escherichia coli isolates from patients with spontaneous bacterial peritonitis: relationship with host factors and prognosis.

Spontaneous bacterial peritonitis (SBP) is a severe complication in patients with cirrhosis and ascites. It is predominantly caused by Escherichia coli. The phylogenetic group and virulence genotype of E. coli isolates causing SBP were investigated, and the association of these characteristics with host factors and prognosis was examined. Seventy-six episodes of E. coli SBP that occurred over a 9-year period were studied. The phylogenetic group of the isolates and the presence of 36 virulence factor genes were investigated. The influence of bacterial and host factors on in-hospital mortality was assessed by multiple logistic regression. Phylogenetic groups A, B1, B2 and D were found in 26%, 4%, 46% and 24% of the isolates, respectively. Virulence factor genes were more frequent in B2 isolates than in non-B2 isolates (mean virulence score 15.4 vs. 7.3, p <10(-4)). Ciprofloxacin resistance was significantly associated with non-B2 groups and a low virulence score. Host factors independently associated with a shift from B2 to non-B2 isolates were norfloxacin prophylaxis (OR 13.01, p 0.0213) and prothrombin ratio (OR 1.04 for a 10% decrease, p 0.0211). The model for end-stage liver disease (MELD) score (OR 1.83, p 0.0007) and hospital-acquired SBP (OR 4.13, p 0.0247) were independent predictors of in-hospital mortality. In contrast, outcome was not influenced by the phylogenetic group or the virulence profile. These findings indicate that the characteristics of E. coli isolates causing SBP vary with the severity of liver disease and with fluoroquinolone prophylaxis. Host factors are more important than bacterial factors in predicting in-hospital mortality.

[Insisting on intravenous polyvalent immunoglobulin therapy in polymyositis in spite of the occurrence of sever hemolytic anemia]. / Poursuite du traitement d'une polymyosite par les immunoglobulines intraveineuses polyvalentes malgré la survenue d'une anémie hémolytique sévère.

A 23-year-old female with polymyositis received high dose intravenous immunoglobulin (IVIg) therapy. The patient suffered severe hemolytic anemia after receiving first course of IVIg infusion. This adverse reaction was likely due to allohemaglutinin A and B and from or high molecular weight IgG complexes contained in the preparation. Though this effect was observed, the treatment was repeated six times. A mild hemolysis occurred following each IVIg, with no clinical consequence. Involvement of the saturation of macrophagic receptors might explain this partial destruction of erythrocytes.

[Reference and selective list in hospital hygiene. Impact on practical improvement]. / Référentiel et liste positive en hygiène hospitalière. Impact sur l'amélioration des pratiques.

In hospitals, the infection control programs requires a basic policy on selection and use of appropriate antibiotics, as well as antiseptics and disinfectants. The selective list concept, born in Germany, spread in France during the nineties, essentially based on national microbiological standards. With the European free circulation of products, more antiseptics and disinfectants with heterogeneous status are available to customers, who must settle on the essential data, with the help of scientific societies: among these data, efficacy against multiresistant strains, adaptation to specific hospital patients, ergonomic handling and tolerance today precede economical criteria and supplement the European micro-biological standards for antiseptics and disinfectants.

Re-treatment with interferon alfa of patients with chronic hepatitis C.

Treatment of patients with chronic hepatitis C has had limited success because of relapses and nonresponse to interferon alfa therapy (currently the only established therapeutic agent). A retrospective study was done to determine the efficacy of re-treatment with interferon and the predictors of response in patients who failed to achieve sustained response after one standard course of interferon therapy (3 million units three times a week for 24 weeks). One hundred and eleven patients (47 relapsers and 64 nonresponders), mean age 45 years, were included in the study. Eighteen relapsers and 13 nonresponders received a higher dose (5 MU), and 11 relapsers and 6 nonresponders received a longer duration (48 weeks) of interferon therapy. The remaining patients received the same regimen as the first treatment. Eighty-one percent and 23% of relapsers and nonresponders, respectively, had an end-of-treatment response, and 19% and 3% of the corresponding patient groups had a sustained response to re-treatment. Two patients with breakthrough during their first treatment were the only nonresponders with sustained response after re-treatment. Sustained response was observed only in patients who received an increased dose or duration of interferon therapy. No predictor of sustained response was found. In conclusion, sustained response to re-treatment with interferon was only observed with augmentation of dose or duration of therapy in some relapsers and patients who had breakthrough. Established predictors of response to interferon in naive patients, in particular serum hepatitis C virus RNA and genotype, were not associated with sustained response to re-treatment.

Effects of ex-vivo plasma adsorption over protein A sepharose in acute leukemia.

Plasma adsorption over immobilized Staphylococcus aureus Cowan I has resulted in tumor regression in several animal and human trials. Protein A, because of its ability to bind IgGs has been considered as the effective component of Staphylococcus aureus. In 4 patients with acute leukemia, a plasma volume of 1500 cm3 was passed in an ex-vivo system over immobilized SpA-Sepharose and then reinfused. Almost all of the IgGs contained in the plasma volume could thus be removed. Toxic side-effects were mild. No significant clinical improvement could be obtained. Plasma incubation with SpA did not modify blast cell viability or leukemic progenitor cell growth. Along with others, this study shows that Protein A is probably not the mediator of the tumoricidal responses observed in studies using adsorption over Staphylococcus aureus Cowan I. The ex-vivo system prepared for this study could also be used for plasma IgG removal for the treatment of autoimmune or immune-complex related disorders.

[Relation between bacterial sensitivity to antibiotics and the consumption of antibiotics. 2 : Study carried out on Staphylococcus aureus at the Pitié-Salpêtrière Hospital from 1975 to 1980]. / Relation entre la sensibilité bactérienne aux antibiotiques et la consommation d'antibiotiques. Deuxième partie : étude effectuée sur Staphylococcus aureus dans le groupe hospitalier Pitié-Salpêtrière entre 1975 et 1980.

From 1975 to 1980, computerized data on the sensitivity of 11 342 strains of S. aureus to the main antibiotics were evaluated in relation to drug consumption and hospital activity. Sensitivity to tetracyclines and chloramphenicol increased by 7.3% and 11.4% respectively, but sensitivity to gentamicin decreased by 12%. There was no change in sensitivity to other antibiotics. In vitro response to methicillin was found to have a considerable influence on changes in sensitivity. Sensitivity to tetracyclines increased among methicillin-sensitive strains, while methicillin-resistant strains became more sensitive to chloramphenicol and less sensitive to gentamicin. During the same period, the consumption of chloramphenicol, tetracyclines and aminoglycosides decreased by 84%, 54% and 15% respectively. Hospital activity remained unchanged. The fact that methicillin-resistant strains failed to become more sensitive to the main antibiotics, with the exception of chloramphenicol, might be due to antibiotic pressure in intensive care units where these strains are usually isolated.

[Relationship between bacterial sensitivity to antimicrobial agents and their consumption (author's transl)]. / Relation entre la sensibilité des bacilles à Gram négatif aux antibiotiques et la consommation d'antibiotiques.

Computerized data on the sensitivity of 20,006 strains of Gram-negative bacilli to ampicillin, cephalothin, gentamicin, tobramycin, nalidixic acid and co-trimoxazole were collected and evaluated in relation to drug consumption and hospital activity. Despite species-related differences, global sensitivity increased from 1975 to 1978 by 44.8% for cephalotin, 14.4% for ampicillin, 7.1% for gentamicin and 4.8% for tobramycin. There was no change in sensitivity to nalidixic acid and co-trimoxazole. During the same period the consumption of cephalosporins and aminoglycosides decreased by 28% and 21% respectively. Following a decrease between 1975 and 1977, the consumption of ampicillin and of nalidixic acid (and related compounds) went up again to reach in 1978 the same level as in 1975. Co-trimoxazole consumption remained unchanged. Since hospital activity remained the same throughout the period under study, it seems justified to correlate the increase in bacterial sensitivity observed to the decrease in consumption of antimicrobial agents.