Search

Summary Background Current drug-eluting stents (c-DESs) reduce the occurrence of ischaemic events, but expose recipients to stent thrombosis and bleeding secondary to preventive antiplatelet therapy. To date, comparative data on the relative effectiveness and safety of the various c-DESs in real life are limited. Aim To compare ischaemic and bleeding risks across the major c-DESs used in France. Methods French national health insurance reimbursement and hospitalization databases were used. Patients implanted with a c-DES in 2014 were followed for 1 year. The risks of ischaemic events (revascularization, myocardial infarction and/or stroke), major bleeding events and death were compared across six c-DESs (XIENCE®, PROMUS®, RESOLUTE®, BIOMATRIX®, NOBORI® and ORSIRO®), using multilevel Cox models adjusted for baseline individual and hospital characteristics. Results A total of 52,891 subjects were included: 34.4% with XIENCE®; 27.6% with PROMUS®; 24.0% with RESOLUTE®; 8.0% with BIOMATRIX®; 5.0% with NOBORI®; and 1.0% with ORSIRO®. Among them, 9378 had at least one event (ischaemic, 6064; major bleeding, 1968; death, 2411), resulting in an overall incidence rate of 19 per 100 person-years. In the multivariable analysis, the risk of ischaemic events, major bleeding events or death did not differ between the c-DESs overall (adjusted hazard ratios between 0.85 [95% confidence interval 0.68–1.07] and 1.04 [95% confidence interval 0.98–1.10] compared with XIENCE® used as the reference) and when each outcome was considered separately. Conclusions In real life, major ischaemic and bleeding risks do not differ across the various c-DESs over the first year following implantation. Future studies are needed to assess comparative c-DES effectiveness and safety longer term.

AIMS: To assess potential change in medicine exposure and association with the risk of road traffic crash across a time period that started before the implementation of a grading system warning of the effect of medicine on driving performances. METHODS: Data from three French national databases were extracted and matched: the national health care insurance database, police reports, and the national police database of injurious crashes. Drivers involved in an injurious crash in France, from July 2005 to December 2011, and identified by their national identifier were included. Association with the risk of crash was estimated using a case-control analysis comparing benzodiazepine and z-hypnotic use among drivers responsible or not for the crash. RESULTS: 69,353 responsible and 73,410 non-responsible drivers involved in an injurious crash were included. Exposure to benzodiazepine anxiolytics was associated with an increased risk of being responsible for a road traffic crash during the pre-intervention period (OR = 1.42 [1.24-1.62]). The association disappeared in the post-intervention period, but became significant again thereafter. The risk of being responsible for a crash increased in users of z-hypnotics across the study period. CONCLUSIONS: Our results question the efficacy of the measures implemented to promote awareness about the effect of medicines on driving abilities. Prevention policies relating to the general driving population, but also to healthcare professionals, should be reviewed.This article is protected by copyright. All rights reserved. Language: en

Introduction Largement utilises dans l’insuffisance coronarienne, les stents permettent de reduire la recurrence d’evenements ischemiques mais peuvent etre a l’origine de thromboses ainsi que de complications hemorragiques en raison des traitements medicamenteux associes. Les donnees comparant les marques (gammes) de stents sont limitees a des populations selectionnees. L’objectif de cette etude etait de comparer en « vie reelle », les risques ischemiques et hemorragiques entre gammes de stents actifs, d’une part, et nus/non actifs, d’autre part. Methodes Etude sur les donnees du Sniiram incluant les sujets implantes de stents en 2014, âges d’au moins 18 ans et affilies au regime general. L’exposition etait definie par le remboursement d’au moins un stent coronaire lors d’un acte d’angioplastie. Le critere de jugement principal associait evenements ischemiques (nouvelle revascularisation, syndrome coronaire aigu et/ou AVC ischemique), evenements hemorragiques majeurs et/ou deces toutes causes. Le risque de survenue de cet evenement composite a un an a ete compare entre gammes en utilisant un modele de Cox a fragilite partagee ajuste notamment sur les comorbidites, facteurs de risque cardiovasculaire, et traitements medicamenteux a l’inclusion. Resultats Au total, 52 891 sujets etaient porteurs de stents actifs (Xience 34,4 %, Promus 27,6 %, Resolute 24,0 %, Nobori 5,0 %, Biomatrix 8,0 %, Orsiro 1,0 %) et 19 027 de stents nus/non actifs (nus 55,1 %, Energy 24,8 %, Titan 20,1 %). Le risque d’evenements ischemiques, hemorragiques et/ou de deces pour 100 patients-annee etait de 19 parmi les porteurs de stents actifs, et de 34 parmi les porteurs de stents nus/non actifs. Les risques n’etaient pas differents entre gammes : parmi les stents actifs, les HRa variaient de 0,85 a 1,04 compare a la gamme de reference (XIENCE) ; parmi les stents nus/non actifs, les HRa variaient de 0,97 a 1,03 compare aux nus. Discussion/Conclusion Le risque de complications ischemiques et hemorragiques a un an ne differe pas entre les gammes de stents en « vie reelle ». Cette etude montre l’interet du Sniiram pour l’evaluation des technologies de sante.

Studies assessing the impact of epilepsy and its medication on the risk of road traffic crashes have shown inconsistent results. The aim in this study was to assess this risk using French databases. Data from three French national databases were extracted and matched: the national health care insurance database, police reports, and the national police database of injurious crashes. Only antiepileptics prescribed predominantly in epilepsy were studied (phenobarbital, phenytoin, ethosuximide, valproic acid, vigabatrin, tiagabin, levitiracetam, zonisamide, and lacosamide). A case-control analysis comparing responsible and non-responsible drivers and a case-crossover analysis were performed. Drivers (72 685) involved in an injurious crash in France between July 2005 and May 2008, were included. Drivers exposed to prescribed antiepileptic medicines (n = 251) had an increased risk of being responsible for a crash (OR 1.74 [1.29-2.34]). The association was also significant for the most severe epileptic patients (n = 99; OR = 2.20 [1.31-3.69]). Case-crossover analysis found no association between crash risk and treatment prescription. Patients with prescription of antiepileptic drugs should be cautioned about their potential risk of road traffic crash. This risk is however more likely to be related to seizures than to the effect of antiepileptic medicines.

To assess potential change in medicine exposure and association with the risk of road traffic crash across a time period that started before the implementation of a grading system warning of the effect of medicine on driving performance.Data from three French national databases were extracted and matched: the national health care insurance database, police reports and the national police database of injurious crashes. Drivers involved in such crashes in France, from July 2005 to December 2011 and identified by their national identifier, were included. Association with the risk of crash was estimated using a case-control analysis comparing benzodiazepine and z-hypnotic use among drivers responsible or not responsible for the crash.Totals of 69 353 responsible and 73 410 non-responsible drivers involved in an injurious crash were included. Exposure to benzodiazepine anxiolytics was associated with an increased risk of being responsible for a road traffic crash during the pre-intervention period (OR = 1.42 [1.24-1.62]). The association disappeared in the post-intervention period, but became significant again thereafter. The risk of being responsible for a crash increased in users of z-hypnotics across the study period.Our results question the efficacy of the measures implemented to promote awareness about the effects of medicines on driving abilities. Prevention policies relating to the general driving population, but also to healthcare professionals, should be reviewed.

Aims To assess potential change in medicine exposure and association with the risk of road traffic crash across a time period that started before the implementation of a grading system warning of the effect of medicine on driving performance. Methods Data from three French national databases were extracted and matched: the national health care insurance database, police reports and the national police database of injurious crashes. Drivers involved in such crashes in France, from July 2005 to December 2011 and identified by their national identifier, were included. Association with the risk of crash was estimated using a case?control analysis comparing benzodiazepine and z-hypnotic use among drivers responsible or not responsible for the crash. Results Totals of 69?353 responsible and 73?410 non-responsible drivers involved in an injurious crash were included. Exposure to benzodiazepine anxiolytics was associated with an increased risk of being responsible for a road traffic crash during the pre-intervention period (OR=1.42 [1.24?1.62]). The association disappeared in the post-intervention period, but became significant again thereafter. The risk of being responsible for a crash increased in users of z-hypnotics across the study period. Conclusions Our results question the efficacy of the measures implemented to promote awareness about the effects of medicines on driving abilities. Prevention policies relating to the general driving population, but also to healthcare professionals, should be reviewed.

Objective: To estimate the risk of road traffic crash associated with prescription of antidepressants. Methods: Data from three French national databases were extracted and matched: the national health care insurance database, police reports, and the national police database of injurious crashes. A case-control analysis comparing 34,896 responsible versus 37,789 nonresponsible drivers was conducted. Case-crossover analysis was performed to investigate the acute effect of medicine exposure. Results: 72,685 drivers, identified by their national health care number, involved in an injurious crash in France over the July 2005 to May 2008 period, were included. 2,936 drivers (4.0%) were exposed to at least one antidepressant on the day of the crash. The results showed a significant association between the risk of being responsible for a crash and prescription of antidepressants (odds ratio (OR)=1.34, 95% confidence interval (CI) [1.22-1.47]). The case-crossover analysis showed no association with treatment prescription, but the risk of road traffic crash increased after an initiation of antidepressant treatment (OR=1.49 [1.24-1.79]) and after a change in antidepressant treatment (OR=1.32 [1.09-1.60]). Conclusions: Patients and prescribers should be warned about the risk of crash during periods of antidepressant medication and about particularly high vulnerability periods such as those when a treatment is initiated or modified.

Objectifs La prise en charge therapeutique des patients revascularises pour une insuffisance coronaire a evolue ces dernieres annees vers une utilisation grandissante des stents enrobes de substance pharmacologiquement active (dits « actifs ») aux depens des stents sans enrobage (dits « nus ») ou recouverts d’un enrobage non actif pharmacologiquement (dits « inactifs »). A contrario, les indications de remboursement des stents actifs restreignent leurs utilisations a certaines caracteristiques lesionnelles ou cliniques. Dans le cadre de la reevaluation des conditions de remboursement des stents coronaires par la HAS, cette etude conjointe ANSM–HAS a pour objectif de realiser un etat des lieux de la pratique medicale relative a la pose de ces dispositifs en France. Methodes L’etude a porte sur les donnees nationales de remboursement des soins par l’Assurance maladie (DCIR) chainees aux donnees d’hospitalisations (PMSI). Ont ete inclus les sujets affilies au regime de l’Assurance maladie âges de plus de 18 ans, ayant eu en 2014 une angioplastie coronaire avec pose de stent. Ont ete decrites, pour chaque patient inclus, les informations relatives a la premiere hospitalisation survenue dans l’annee avec un code d’acte medical d’angioplastie et une facturation de stent en sus du forfait hospitalier. Les informations medicales avant l’hospitalisation ont egalement ete recherchees. Resultats Au total, 114 771 patients ont ete inclus ; l’âge moyen etait de 68 ± 12 ans, 76 % etaient des hommes. La majorite des patients (76 %) avaient recu un stent actif. La duree moyenne du sejour hospitalier etait de 5 ± 5 jours avec une implantation pour 60 % d’entre eux dans les etablissements du secteur public et pour 81 % dans des centres realisant plus de 600 angioplasties par an. En moyenne, au cours du sejour, 1 ± 0,3 angioplastie (mediane : 1) etait realisee par patient avec 1,5 ± 0,8 stents poses. Pour 58 % des patients, le motif d’hospitalisation etait un syndrome coronaire aigu. Aucun antecedent de maladie coronaire n’a ete retrouve chez 46 %. Concernant les comorbidites, 70 % etaient hypertendus, 27 % etaient diabetiques, 46 % etaient traites pour une dyslipidemie, 17 % avaient un cancer actif, 8 % etaient consideres comme a haut risque hemorragique et 1,5 % etaient atteints d’une demence. Compare aux patients ayant des stents actifs, les patients porteurs de stents nus ou inactifs recevaient moins de stents, etaient plus souvent des femmes, des patients âges de plus de 80 ans, etaient plus souvent sous traitement a risque hemorragique, atteints de cancer ou de demence. Conclusion L’activite d’angioplastie en France repose essentiellement sur l’implantation de stent actif compare aux autres categories de stent. Ce resultat est en adequation avec les dernieres recommandations de la Societe europeenne de cardiologie, qui preconise l’utilisation du stent actif. Les resultats selon le type de stent pose permettent d’identifier des populations specifiques de patients eligibles a l’angioplastie chez lesquels la pose d’un stent actif n’est pas d’emblee choisie par le praticien. Cette etude qui apportera des elements utiles a la reevaluation des conditions de remboursement des stents sera poursuivie par la mesure des evenements ischemiques et hemorragiques apres la pose de stent.

BACKGROUND: In late 2012 a national pill crisis led French women to promptly change their behavior regarding contraception with a significant increase in the use of first-generation and second-generation combined oral contraceptives to the detriment of third-generation and fourth-generation products (C3Gs and C4Gs). OBJECTIVE: To assess the impact of the sharp changes in 2013 on the rate of women hospitalized for pulmonary embolism in France. METHODS: All hospitalized pulmonary embolisms in women aged 15-49 years excluding those occurring during delivery stay were identified from the French national hospital discharge databases from 2010 to 2013. Hospitalization rates overall and by age group were calculated. We compared rates in 2013 with those in 2012 and with mean rates over the preceding 3-year period (2010-2012). Two populations of non-users of contraceptives were used as control groups: men aged 15-49 years and women aged 50-69 years. The expected reduction in pulmonary embolism incidence estimated by modeling the number of expected cases based on modifications of combined oral contraceptive exposure was also considered. RESULTS: In France in 2013 as compared with 2012 the pulmonary embolism hospitalization incidence rate in women aged 15-49 years fell by 10.6% corresponding to a reduction of 322 hospitalizations (95% confidence interval -468 to -156). The expected pulmonary embolism reduction is consistent with the observed reduction in hospitalization incidence rate (-10.2% and -10.6% respectively). Such a pattern was not observed in the control groups. CONCLUSIONS: The sharp change in contraception methods with decreased use of C3Gs and C4Gs probably played a major role in the reduction in venous thromboembolism morbidity in France. (c) 2015 The Authors. Journal of Thrombosis and Haemostasis published by Wiley Periodicals Inc. on behalf of International Society on Thrombosis and Haemostasis.

Introduction La duree optimale de la bitherapie antiagregante plaquettaire (AAP) apres pose de stent reste largement debattue malgre les recommandations en vigueur. En France, la duree recommandee est d’un a trois mois apres pose de stent nu et de six a 12 mois apres pose de stent actif. En post-infarctus, une duree d’un an est preconisee quel que soit le type de stent implante. En l’absence de donnees en vie reelle, l’etude a pour objectif d’etudier la prevalence d’utilisation de la bitherapie AAP apres pose de stent a 3, 6, 12 et 18 mois et d’identifier les facteurs associes. Methodes L’etude a ete realisee a partir des donnees du systeme national d’information inter-regimes de l’assurance-maladie (Sniiram) sur la population du regime general hospitalisee pour angioplastie avec pose de stent entre le 1 er janvier 2011 et le 31 decembre 2012. Pour chaque sujet, les delivrances d’antiagregants plaquettaires etaient recherchees depuis la sortie du sejour d’hospitalisation jusqu’au 31 decembre 2013. Un sujet etait considere utilisateur de bitherapie AAP a 3, 6, 12 et 18 mois si le nombre de jours de traitement delivres couvrait au moins deux tiers de la periode etudiee. Resultats Au total, 137 195 sujets ont ete inclus. L’âge moyen etait de 66,6 ans, 74,3 % des sujets etaient des hommes, 49,4 % avaient recu un stent actif. La prevalence d’utilisation de la bitherapie AAP etait de 91,8 % a 3 mois, de 86,6 % a 6 mois, de 80,3 % a 12 mois et de 73,0 % a 18 mois. Quelle que soit la periode etudiee, la prevalence etait plus importante chez les sujets ayant recu un stent actif compares a ceux ayant recu un stent nu (95,6 % versus 86,8 % a 3 mois et 82,7 % versus 61,1 % a 18 mois). Compares aux utilisateurs de monotherapie AAP, les utilisateurs de bitherapie etaient plus jeunes, moins hypertendus et avaient utilise moins d’AAP dans l’annee precedant l’inclusion. Discussion/Conclusion L’utilisation de la bitherapie AAP au-dela des durees recommandees est tres importante en France. La balance benefice/risque de la bitherapie AAP a long terme necessite d’etre evaluee a l’echelle populationnelle.

Nephrogenic systemic fibrosis (NSF) has been related to the use of gadolinium-based contrast agents (GBCAs) in patients undergoing dialysis. The Prospective Fibrose Nephrogénique Systémique study, a French prospective study supported by the French drug regulatory agency (Agence Nationale de Sécurité du Médicament) and the French Societies of Nephrology, Dermatology, and Radiology, aimed at determining the incidence of NSF in patients undergoing long-term dialysis.Adult patients undergoing long-term dialysis receiving a magnetic resonance imaging (MRI) examination prescribed between January 15, 2009 and May 31, 2011, with or without GBCA were included. The methodology was based on a patient form intended to detect any dermatological event (DE) that may occur within 4 months after the examination. Further investigations were planned with their physicians if a DE was reported.A total of 571 patients were included. A total of 50.3% received GBCA. Among them, 93.4% received a macrocyclic GBCA, usually gadoteric acid (88.9%). All in all, 22 patients (3.9%) reported a DE. Dermatological diagnoses did not reveal any evidence of NSF.The incidence of NSF after a single dose of a macrocyclic GBCA is null in our sample of 268 patients undergoing dialysis (hemodialysis and peritoneal dialysis). This incidence is just lower than 0.5%. When contrast-enhanced MRI can be essential, or even decisive, to the diagnosis, these results are important and reassuring if physicians need to perform contrast-enhanced MRI in patients undergoing dialysis.

International audience; Background: Large data sets with many variables provide particular challenges when constructing analytic models. Lasso-related methods provide a useful tool, although one that remains unfamiliar to most epidemiologists. Methods: We illustrate the application of lasso methods in an analysis of the impact of prescribed drugs on the risk of a road traffic crash, using a large French nationwide database (PLoS Med 2010; 7:e1000366). In the original case-control study, the authors analyzed each exposure separately. We use the lasso method, which can simultaneously perform estimation and variable selection in a single model. We compare point estimates and confidence intervals using (1) a separate logistic regression model for each drug with a Bonferroni correction and (2) lasso shrinkage logistic regression analysis. Results: Shrinkage regression had little effect on (bias corrected) point estimates, but led to less conservative results, noticeably for drugs with moderate levels of exposure. Carbamates, carboxamide derivative and fatty acid derivative antiepileptics, drugs used in opioid dependence, and mineral supplements of potassium showed stronger associations. Conclusion: Lasso is a relevant method in the analysis of databases with large number of exposures and can be recommended as an alternative to conventional strategies.

Background: Opioids have been shown to impair psychomotor and cognitive functioning in healthy volunteers with no history of opioid abuse. Few or no significant effects have been found in opioid-dependant patients in experimental or driving simulation studies. The risk of road traffic crash among patients under buprenorphine or methadone has not been subject to epidemiological investigation so far. The objective was to investigate the association between the risk of being responsible for a road traffic crash and the use of buprenorphine and methadone. Methods: Data from three French national databases were extracted and matched: the national health care insurance database, police reports, and the national police database of injurious crashes. Case-control analysis comparing responsible versus non responsible drivers was conducted. Results: 72,685 drivers identified by their national health care number, involved in an injurious crash in France over the July 2005 to May 2008 period. The 196 drivers exposed to buprenorphine or methadone on the day of crash were young, essentially males, with an important co-consumption of other substances (alcohol and benzodiazepines). Injured drivers exposed to buprenorphine or methadone on the day of crash, had an increased risk of being responsible for the crash (odds ratio (OR)=2.02, 95% confidence interval (CI): 1.40, 2.91). Conclusions: Users of methadone and buprenorphine were at increased risk of being responsible for injurious road traffic crashes. This risk is probably the combined result of risky behaviors and treatments.

Directeurs de stage P. Tubert-Bitter et S. Escolano. Laboratoire d’accueil CESP, Equipe Biostatistique, Le Kremlin-Bicetre, France. Etat de la question Les etudes pharmaco-epidemiologiques realisees a partir des bases medico-administratives sont en plein essor. De par leur quasi exhaustivite, ces larges bases de donnees constituent un outil puissant. Certaines variables ne sont cependant pas recueillies, notamment les facteurs de confusion. Leur absence de prise en compte dans les modeles de regression lineaire classique conduit a une situation d’endogeneite, a l’origine d’un biais des estimateurs conventionnels. L’endogeneite est definie par une correlation non nulle entre une ou plusieurs variables explicatives et le terme d’erreur dans l’equation de regression. L’approche par variable instrumentale est une methode adaptee a son traitement. L’objectif du travail etait d’etudier le comportement de deux estimateurs par variable instrumentale proposes dans le cadre non lineaire dans un modele logistique en faisant varier la force de l’instrument. Materiel et methodes Dans un premier temps, une revue attentive de la litterature a permis de poser le cadre statistique d’utilisation des variables instrumentales, d’en exposer la methodologie et de recenser les differentes methodes d’estimation et les principaux resultats methodologiques publies. Le comportement de l’instrument « Preference de prescription du medecin » a, dans un second temps, ete explore via une etude de simulation. Deux methodes d’estimation par variable instrumentale [methode des pseudo-doubles moindres carres (2SPS) et methode avec inclusion du residu a la seconde etape (2SRI)] ont ete utilisees, et comparees a la methode d’analyse conventionnelle. Resultats En depit d’une litterature plutot foisonnante, les contours de l’approche par variables instrumentales restent flous. L’etude de simulation a montre que l’estimateur 2SRI pouvait s’averer performant en termes de reduction de biais et de convergence par rapport a l’estimateur conventionnel, mais au detriment d’une inflation de la variance, qui mene a des intervalles de confiance larges et rend leurs interpretations delicates. Conclusion L’approche par variables instrumentales peut etre envisagee comme une approche complementaire aux analyses classiques. Ses proprietes de convergence en font un outil decisif dans la comprehension des phenomenes etudies, avec la possibilite de detecter le sens du biais des estimateurs conventionnels.

Background There is a growing concern over the impact of medicines on the risk of road traffic crashes. The French Health Products Safety Agency established a classification of medicines affecting driving abilities, using a labelling system with 4 levels of risk. Objective To investigate the association between prescribed medicine use and the risk of road traffic crashes, providing risk and attributable fraction estimates, with particular focus on the French labelling system. Methods We matched data from three exhaustive nation-wide databases: the database on medicine reimbursing, police reports and the database of injurious crashes. Drivers involved in an injurious crash in France, from July 2005 to May 2008, and identified by their national healthcare number were included. We studied prescribed medicines grouped according to the four levels of the French classification (level 0 to level 3). We performed a responsibility and a case-crossover analysis. Results 72 685 drivers were included. The results showed an increased risk of being responsible for a crash for users of medicines of level 2 (OR=1.31 [1.24 to 1.40]) and level 3 (OR=1.25 [1.12 to 1.40]). The fractions of road traffic crashes attributable to levels 2 and 3 medicine use were 2.9% (2.3–3.4%) and 0.6% (0.3–0.8%) respectively. Conclusion Users of level 2 and 3 medicines are at increased risk of experiencing a road traffic crash. Further studies are needed to evaluate the impact of the French labelling system.