Second-degree burn induced by high-concentration topical capsaicin with mobility sequelae: A case report.

High-concentration topical capsaicin is used as a second-line treatment for neuropathic pain. Transient, mild burning sensation and erythema are expected adverse drug reactions. Here, we report the first case of second degree burn after the application of a high-concentration topical capsaicin patch with secondary mobility sequelae. Nine months after the application, neuropathic pain still remained and the patient described mobility difficulties in daily activities, preventing her from returning to work. This report aims to raise the question of the benefit/risk ratio of high concentration topical capsaicin.

Validation of Artificial Intelligence to Support the Automatic Coding of Patient Adverse Drug Reaction Reports, Using Nationwide Pharmacovigilance Data.

INTRODUCTION: Adverse drug reaction reports are usually manually assessed by pharmacovigilance experts to detect safety signals associated with drugs. With the recent extension of reporting to patients and the emergence of mass media-related sanitary crises, adverse drug reaction reports currently frequently overwhelm pharmacovigilance networks. Artificial intelligence could help support the work of pharmacovigilance experts during such crises, by automatically coding reports, allowing them to prioritise or accelerate their manual assessment. After a previous study showing first results, we developed and compared state-of-the-art machine learning models using a larger nationwide dataset, aiming to automatically pre-code patients' adverse drug reaction reports. OBJECTIVES: We aimed to determine the best artificial intelligence model identifying adverse drug reactions and assessing seriousness in patients reports from the French national pharmacovigilance web portal. METHODS: Reports coded by 27 Pharmacovigilance Centres between March 2017 and December 2020 were selected (n = 11,633). For each report, the Portable Document Format form containing free-text information filled by the patient, and the corresponding encodings of adverse event symptoms (in Medical Dictionary for Regulatory Activities Preferred Terms) and seriousness were obtained. This encoding by experts was used as the reference to train and evaluate models, which contained input data processing and machine-learning natural language processing to learn and predict encodings. We developed and compared different approaches for data processing and classifiers. Performance was evaluated using receiver operating characteristic area under the curve (AUC), F-measure, sensitivity, specificity and positive predictive value. We used data from 26 Pharmacovigilance Centres for training and internal validation. External validation was performed using data from the remaining Pharmacovigilance Centres during the same period. RESULTS: Internal validation: for adverse drug reaction identification, Term Frequency-Inverse Document Frequency (TF-IDF) + Light Gradient Boosted Machine (LGBM) achieved an AUC of 0.97 and an F-measure of 0.80. The Cross-lingual Language Model (XLM) [transformer] obtained an AUC of 0.97 and an F-measure of 0.78. For seriousness assessment, FastText + LGBM achieved an AUC of 0.85 and an F-measure of 0.63. CamemBERT (transformer) + Light Gradient Boosted Machine obtained an AUC of 0.84 and an F-measure of 0.63. External validation for both adverse drug reaction identification and seriousness assessment tasks yielded consistent and robust results. CONCLUSIONS: Our artificial intelligence models showed promising performance to automatically code patient adverse drug reaction reports, with very similar results across approaches. Our system has been deployed by national health authorities in France since January 2021 to facilitate pharmacovigilance of COVID-19 vaccines. Further studies will be needed to validate the performance of the tool in real-life settings.

Artificial Intelligence for Unstructured Healthcare Data: Application to Coding of Patient Reporting of Adverse Drug Reactions.

Adverse drug reaction (ADR) reporting is a major component of drug safety monitoring; its input will, however, only be optimized if systems can manage to deal with its tremendous flow of information, based primarily on unstructured text fields. The aim of this study was to develop an automated system allowing to code ADRs from patient reports. Our system was based on a knowledge base about drugs, enriched by supervised machine learning (ML) models trained on patients reporting data. To train our models, we selected all cases of ADRs reported by patients to a French Pharmacovigilance Centre through a national web-portal between March 2017 and March 2019 (n = 2,058 reports). We tested both conventional ML models and deep-learning models. We performed an external validation using a dataset constituted of a random sample of ADRs reported to the Marseille Pharmacovigilance Centre over the same period (n = 187). Here, we show that regarding area under the curve (AUC) and F-measure, the best model to identify ADRs was gradient boosting trees (LGBM), with an AUC of 0.93 (0.92-0.94) and F-measure of 0.72 (0.68-0.75). This model was run for external validation showing an AUC of 0.91 and a F-measure of 0.58. We evaluated an artificial intelligence pipeline that was found able to learn how to identify correctly ADRs from unstructured data. This result allowed us to start a new study using more data to further improve our performance and offer a tool that is useful in practice to efficiently manage drug safety information.

Safety of levonorgestrel 52 mg intrauterine system compared to copper intrauterine device: a population-based cohort study.

OBJECTIVE: To compare the risk of all-cause death, hospitalizations (any cause), ectopic pregnancy, pelvic inflammatory disease or infection, uterine perforation, device removal, neuro-psychiatric drugs initiation, or new psychiatric visit(s) between levonorgestrel (LNG) 52 mg intrauterine system (IUS) and copper intrauterine device (IUD) users in France. STUDY DESIGN: We identified a historical cohort of women aged 20-55 years with a first dispensing of either LNG 52 mg IUS or copper-IUD between January 1, 2010, and December 31, 2014, in the French National Claims database, SNDS. We used propensity score matching to balance the two groups on baseline sociodemographic and clinical characteristics to minimize confounding. We estimated Cox proportional hazards models to compare health outcomes between LNG 52 mg IUS and copper-IUDs users. RESULTS: We matched 9318 LNG 52 mg IUS users (mean age 36.2±6.8 years) to 10,185 copper-IUD users (mean age 35.4±7.1 years). After matching and age-adjustment, LNG 52 mg IUS users had a slightly higher risk of anxiolytic drugs initiation (HR 1.08, 95%CI 1.01-1.15) and device removal (HR 1.05, 95%CI 1.01-1.10) compared to copper-IUD users, with no differences for other studied outcomes. CONCLUSION: French IUS users report slightly more anxiolytic treatment initiation and IUD removal compared to copper-IUD users. These results are consistent with a potential pharmacovigilance signal of anxiety-related disorders in LNG 52 mg IUS users. IMPLICATIONS STATEMENT: In French LNG 52 mg IUS users, there was slightly more anxiolytic treatment initiation and IUD removal compared to copper-IUD users. No risk difference was found for all-cause death, hospitalizations, ectopic pregnancy, pelvic disorders, and uterine perforation. We cannot exclude that the associations are related to differences in characteristics of women who chose each type of type of IUD.