[Comparative relevance of declaration of side effects by patients and health professionals]. / Pertinence comparée des notifications d'effets indésirables de patients et professionnels de santé.

INTRODUCTION: Drug-induced adverse events have been accessible to patient's spontaneous reporting in France and such notifications have been steadily increasing since 2011. However, these notifications are still shrouded with medical perplexity and are sometimes subjected to partial caution in their interpretation by the patient's physician. We aimed to evaluate and compare prospectively the relevance of such spontaneous notifications with those provided by healthcare professionals to the French Pharmacovigilance Center of Nice-Alpes-Côte d'Azur. METHODS: Spontaneous reporting of drug adverse events notified by patients and health care professionals were compared in terms of critical (name, date, effect, drug involved, chronological compatibility) and non-critical (posology, dosage) information, whereas the plausibility of the cases were assessed in weekly multispecialty staffs. Each patient's notification was matched with the immediate pre- and post-notifications declared by health care providers. RESULTS: Spontaneous notifications from 61 patients were compared with 122 notifications from health care providers. Neither the critical information necessary for declaring the case in the national database (7/61 versus 16/122, P=0.75), nor the uncritical elements allowing to assess the case (30/61 versus 51/122, P=0.22), its plausibility or the causality of the drug (P=0.10) differed significantly between the two groups. 107 cases out of 122 (88%) notified by health care providers were classified as serious, as compared with 19 out of 61 (15%) of patient's ones (P<0.001). CONCLUSION: Despite concerns from pharmacovigilance specialists in France, the medical relevance of spontaneous reported drug-associated adverse events does not differ from that of health care providers.

Is anti-factor Xa chromogenic assay for Rivaroxaban appropriate in clinical practice? Advantages and comparative drawbacks.

INTRODUCTION: No routine monitoring is required with factor Xa inhibitor rivaroxaban. Yet, its titration must be adapted, and its misuse may lead to increased risk of bleeding; therefore, therapeutic rivaroxaban monitoring might help in specific situations. MATERIAL AND METHODS: As asked by clinicians of our medical center, we measured rivaroxaban plasma concentrations in real conditions of use and checked their corresponding prescriptions. Measurement of 112 samples from 94 consecutive patients was performed with a Biophen LRT® anti-Xa chromogenic assay and compared blindly to the HPLC-MSMS "gold standard" method. Rivaroxaban was effectively given to 80 out of 94 patients but a mere 57% through an adequate prescription (within the scope of indications/titration). All chromogenic measurements were over the pre-specified 30ng/ml LOQ, whereas only 98 /114 samples had quantifiable rivaroxaban with HPLC-MSMS (LOQ 1ng/ml). Correlation between the two methods and linear regression were highly significant (p<0.0001). However, chromogenic values (mean 141.6ng/ml[96.6]) overestimated HPLC-MSMS values (119.7ng/ml[79.5]) by 22ng/ml according to Bland-Altman analysis (p<0.001). After re-assessing the chromogenic LOQ at 52ng/ml, 83 quantifiable samples had a mean concentration of 176.9ng/ml as compared to 158.5ng/ml with HPLC-MSMS, with no false positive anymore. CONCLUSIONS: In our medical center, rivaroxaban concentrations could be assessed by a rapid chromogenic method. Its pre-specified LOQ proved too high after being checked "on site" against HPLC-MSMS. Prescriptions for rivaroxaban were not optimal. An overestimated LOQ may impair observance monitoring or predispose patients to either risky thrombolysis or otherwise adjournable surgery in clinical practice.

Are nilotinib-associated vascular adverse events an under-estimated problem?

Vascular adverse events have been reported with nilotinib, a tyrosine kinase inhibitor prescribed for chronic myeloid leukaemia. However, few data specify their incidence, or whether they occur in predisposed patients. Hence, we prospectively studied 30 consecutive patients to assess the frequency of such adverse reactions and determine whether the patients presenting with these adverse events bear predisposing factors. From 3 to 73 months after nilotinib initiation, 10 of the 30 patients experienced vascular events. Three patients of these 10 were devoid of any patent cardiovascular risk factor, except for age. This study points out an occurrence more frequent than expected of vascular adverse events associated with nilotinib (> 30% vs. < 1% in summary of product characteristics), and particularly of vascular events of late onset in patients with no pre-existing risk factors.

[Tuberculosis and infliximab treatment. National surveillance from January 1, 2000, through June 30, 2003]. / Tuberculoses lors de traitements par l'infliximab. Suivi national du ler janvier 2000 au 30 juin 2003.

OBJECTIVES: Analysis of the tuberculosis cases reported in France in patients treated with infliximab since its marketing approval, assessment of the effect of changes in the summary of product characteristics and national guidelines. METHODS: Based on tuberculosis reports from the national post-marketing adverse drug reaction databank of the manufacturer from January 1, 2000 through June 30, 2003, and records from the national multicenter retrospective survey on opportunistic infections with anti-TNFalpha, we analyzed all cases of tuberculosis and the impact of the changes made in December 2000 in the summary of product characteristics and the guidelines on the prevention and management of tuberculosis in patients treated with infliximab published in February 2002. RESULTS: 56 cases of tuberculosis were reported: the median interval before diagnosis was 12 weeks with a median of 3 infusions. The presence of Koch bacilli was confirmed in 32 patients; 29 patients had extrapulmonary or disseminated forms of tuberculosis. The tuberculosis rate among patients treated with infliximab was greater than among the general population and differed significantly by period (p < 0.005). CONCLUSION: Tuberculosis can occur within the first 12 weeks of treatment with infliximab. Information for practitioners must be continued, together with surveillance of the tuberculosis cases in France.