Automatic de-identification of French electronic health records: a cost-effective approach exploiting distant supervision and deep learning models.

BACKGROUND: Electronic health records (EHRs) contain valuable information for clinical research; however, the sensitive nature of healthcare data presents security and confidentiality challenges. De-identification is therefore essential to protect personal data in EHRs and comply with government regulations. Named entity recognition (NER) methods have been proposed to remove personal identifiers, with deep learning-based models achieving better performance. However, manual annotation of training data is time-consuming and expensive. The aim of this study was to develop an automatic de-identification pipeline for all kinds of clinical documents based on a distant supervised method to significantly reduce the cost of manual annotations and to facilitate the transfer of the de-identification pipeline to other clinical centers. METHODS: We proposed an automated annotation process for French clinical de-identification, exploiting data from the eHOP clinical data warehouse (CDW) of the CHU de Rennes and national knowledge bases, as well as other features. In addition, this paper proposes an assisted data annotation solution using the Prodigy annotation tool. This approach aims to reduce the cost required to create a reference corpus for the evaluation of state-of-the-art NER models. Finally, we evaluated and compared the effectiveness of different NER methods. RESULTS: A French de-identification dataset was developed in this work, based on EHRs provided by the eHOP CDW at Rennes University Hospital, France. The dataset was rich in terms of personal information, and the distribution of entities was quite similar in the training and test datasets. We evaluated a Bi-LSTM + CRF sequence labeling architecture, combined with Flair + FastText word embeddings, on a test set of manually annotated clinical reports. The model outperformed the other tested models with a significant F1 score of 96,96%, demonstrating the effectiveness of our automatic approach for deidentifying sensitive information. CONCLUSIONS: This study provides an automatic de-identification pipeline for clinical notes, which can facilitate the reuse of EHRs for secondary purposes such as clinical research. Our study highlights the importance of using advanced NLP techniques for effective de-identification, as well as the need for innovative solutions such as distant supervision to overcome the challenge of limited annotated data in the medical domain.

Prognostic value and predictors of the alteration of the diffusing capacity of the lungs for carbon monoxide in systemic lupus erythematosus.

OBJECTIVES: Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterized by heterogeneous manifestations and severity, with frequent lung involvement. Among pulmonary function tests (PFT), the measure of the diffusing capacity of the lungs for carbon monoxide (DLCO) is a noninvasive and sensitive tool assessing pulmonary microcirculation. Asymptomatic and isolated DLCO alteration has been frequently reported in SLE, but its clinical relevance has not been established. METHODS: This retrospective study focused on 232 SLE patients fulfilling the 2019 EULAR/ACR classification criteria for SLE. Data were collected from the patient's medical record, including demographic, clinical, and immunological characteristics while DLCO was measured when performing PFT as part of routine patient follow-up. RESULTS: At the end of follow-up, DLCO alteration (<70% of predicted value) was measured at least once in 154 patients (66.4%), and was associated with a history of smoking as well as interstitial lung disease (ILD), but was also associated with renal and neurological involvement. History of smoking, detection of anti-nucleosome autoantibodies and clinical lymphadenopathy at diagnosis were independent predictors of DLCO alteration, while early cutaneous involvement with photosensitivity was a protective factor. DLCO alteration, at baseline or anytime during follow-up was predictive of admission in intensive care unit and/or of all-cause death, both mainly due to severe disease flares and premature cardiovascular complications. CONCLUSION: This study suggests a link between DLCO alteration and disease damage, potentially related to SLE vasculopathy, and prognostic value of DLCO on death or ICU admission in SLE.

Management of Pathogenic CDH1 Variant Carriers Within the FREGAT Network: A Multicentric Retrospective Study.

OBJECTIVE: To describe the management of pathogenic CDH1 variant carriers (pCDH1vc) within the FREGAT (FRench Eso-GAsTric tumor) network. Primary objective focused on clinical outcomes and pathological findings, Secondary objective was to identify risk factor predicting postoperative morbidity (POM). BACKGROUND: Prophylactic total gastrectomy (PTG) remains the recommended option for gastric cancer risk management in pCDH1vc with, however, endoscopic surveillance as an alternative. METHODS: A retrospective observational multicenter study was carried out between 2003 and 2021. Data were reported as median (interquartile range) or as counts (proportion). Usual tests were used for univariate analysis. Risk factors of overall and severe POM (ie, Clavien-Dindo grade 3 or more) were identified with a binary logistic regression. RESULTS: A total of 99 patients including 14 index cases were reported from 11 centers. Median survival among index cases was 12.0 (7.6-16.4) months with most of them having peritoneal carcinomatosis at diagnosis (71.4%). Among the remaining 85 patients, 77 underwent a PTG [median age=34.6 (23.7-46.2), American Society of Anesthesiologists score 1: 75%] mostly via a minimally invasive approach (51.9%). POM rate was 37.7% including 20.8% of severe POM, with age 40 years and above and low-volume centers as predictors ( P =0.030 and 0.038). After PTG, the cancer rate on specimen was 54.5% (n=42, all pT1a) of which 59.5% had no cancer detected on preoperative endoscopy (n=25). CONCLUSIONS: Among pCDH1vc, index cases carry a dismal prognosis. The risk of cancer among patients undergoing PTG remained high and unpredictable and has to be balanced with the morbidity and functional consequence of PTG.

Real-world safety profiles of pirfenidone and nintedanib in idiopathic pulmonary fibrosis patients.

INTRODUCTION: While pirfenidone and nintedanib have greatly influenced the treatment of idiopathic pulmonary fibrosis (IPF), both drugs have significant early adverse drug reactions (ADRs) and almost nothing is known of their rare and delayed ADRs. We collected and analyzed pirfenidone- or nintedanib-related ADRs identified in a French rare lung disease center, recorded their profiles and identified potential safety signals. METHODS: We analyzed the medical records of IPF patients treated with pirfenidone or nintedanib between January 2011 and January 2020 at the Rennes University Hospital to estimate the incidence of serious and non-serious ADRs cases due to each drug and the incidence of ADRs involving the cardiovascular, hepatobiliary, gastro-intestinal, dermatological, and metabolic/nutritional systems. RESULTS: The 176 patients included 115 (65%) initially treated with pirfenidone and 61 (35%) given nintedanib. ADRs occurred in 78.3% of those given pirfenidone and in 70.5% of those given nintedanib. The incidence of first serious ADRs cases was about 33 per 100 person-years (100 PY) for both drugs; first non-serious pirfenidone ADRs cases were 102 per 100 PY and 130 per 100 PY for nintedanib. The incidence involving each organ system were quite similar, except for the gastro-intestinal and skin disorders. Cardiovascular disorders occurred in about 10 cases per 100 PY in both pirfenidone and nintedanib patients. DISCUSSION: Most ADRs were consistent with the expected antifibrotic drug safety profiles. As arterial and venous thromboembolic events are rare, it is important to assess the risk associated with using antifibrotics by a dedicated pharmacoepidemiological study.

A descriptive, retrospective case series of patients with factitious disorder imposed on self.

BACKGROUND: Despite cases of factitious disorder imposed on self being documented in the literature for decades, it appears to remain an under-identified and under-diagnosed problem. The present study aimed to explore factitious disorder imposed on self in a series of French patients. METHODS: Patients 18 years old and over with factitious disorder imposed on self were retrospectively included by two independent reviewers according to DSM-5 criteria in Rennes University Hospital for the period 1995 to 2019. Patients were identified from a clinical data warehouse. RESULTS: 49 patients with factitious disorder imposed on self were included. Among them, 36 (73.5%) were female. The average age at diagnosis was 38.4 years. The 16 patients with a health-related profession were all female. Direct evidence of falsification was found in 20.4% of cases. Falsification was mainly diagnosed on the basis of indirect arguments: history of factitious disorder diagnosed in another hospital (12.2%), extensive use of healthcare services (22.4%), investigations that were normal or inconclusive (69.4%), inconsistent or incomplete anamnesis and/or patient refusal to allow access to outside information sources (20.4%), atypical presentation (59.2%), evocative patient behaviour or comments (32.7%), and/or treatment failure (28.6%). Dermatology and neurology were the most frequently involved specialities (24.5%). Nine patients were hospitalized in intensive care. Some of them received invasive treatments, such as intubations, because of problems that were only reported or feigned. The diagnosis of factitious disorder imposed on self was discussed with the patient in 28 cases (57.1%). None of them admitted to making up the disorder intentionally. Two suicide attempts occurred within 3 months after the discussion of the diagnosis. No deaths were recorded. 44.9% of the patients returned to the same hospital at least once in relation to factitious disorder imposed on self. CONCLUSIONS: The present study reinforces data in favour of a predominance of females among patients with factitious disorder imposed on self. This diagnosis is difficult and is based on a range of arguments. While induced cases can be of low severity, cases that are only feigned can lead to extreme medical interventions, such as intubation.

Regression of Fibrosis Stage With Treatment Reduces Long-Term Risk of Liver Cancer in Patients With Hemochromatosis Caused by Mutation in HFE.

BACKGROUND & AIMS: Fibrosis stage can decrease following treatment in patients with hemochromatosis caused by mutations in the homeostatic iron regulator gene (HFE), but the effects on cirrhosis are not clear. We assessed regression of severe fibrosis and the ensuing risk of liver cancer after treatment. METHODS: We performed a retrospective analysis of data from 106 patients in France or Australia who were homozygous for the C282Y mutation in HFE with F3 fibrosis (n = 40) or F4 fibrosis (n = 66) at diagnosis and from whom at least 1 liver biopsy was collected during follow up. We collected data from the time of first biopsy and during follow-up period on patient demographics, treatment, smoking habits, alcohol consumption, infection with hepatitis B or C viruses, and other diseases. The median time between first and last liver biopsy was 9.5 years (range, 3.5-15.6 years). We collected results of tests for liver function, markers of iron stores, and platelet levels. Patients were followed for a median 17.6 years (range, 9.8-24.1 years) for development of liver cancer occurrence. RESULTS: At last liver biopsy, 41 patients (38.6%) had fibrosis scores of F2 or less. Liver cancer occurred in 34 patients (52.3%) with F3 or F4 fibrosis at last liver biopsy vs 2 patients (4.8%) with fibrosis scores of F2 or less at last liver biopsy (P < .001). Liver cancer incidences were 32.8 per 1000 person-years (95% CI, 22.7-45.9 per 1000 person-years) in patients with F3 or F4 fibrosis and 2.3 per 1000 person-years (95% CI, 0.2-8.6 per 1000 person-years) in patients with fibrosis scores of F2 or less (P < .001). In multivariate analysis, male sex (hazard ratio [HR], 6.09; 95% CI, 1.21-30.4), age at diagnosis (HR, 1.16; 95% CI, 1.09-1.25), presence of diabetes (HR, 3.07; 95% CI, 1.35-6.97), excess alcohol consumption (HR, 3.1; 95% CI, 1.47-6.35), serum level of ferritin at diagnosis (P < .01), and regression to fibrosis scores of F2 or less (HR, 0.08; 95% CI, 0.01-0.62) were significantly associated with risk of liver cancer. CONCLUSIONS: In a retrospective analysis of patients with hemochromatosis caused by the C282Y mutation in HFE, we found that severe liver fibrosis can regress with treatment. In patients with fibrosis regression to a stage F2 or less, the long-term risk for liver cancer is significantly reduced.

Impact of Preoperative Iron Deficiency on Blood Transfusion in Elective Cardiac Surgery.

OBJECTIVE: To evaluate the incidence and consequences of preoperative iron deficiency in elective cardiac surgery. DESIGN: A prospective observational study. SETTING: The cardiac surgery unit of a university hospital, from November 2016 to February 2017. PARTICIPANTS: All patients presenting for elective cardiac surgery during the study period, with the exclusion of noncardiac thoracic surgeries, surgeries of the descending aorta, endovascular procedures, and patients affected by an iron-metabolism disease. INTERVENTIONS: Transferrin saturation and serum ferritin levels were systematically assessed before surgery, and the care of patients was maintained as usual. MEASUREMENTS AND MAIN RESULTS: Routine analyses, clinical data, and the number of blood transfusions were recorded during the hospital stay. Among the 272 patients included, 31% had preoperative iron deficiency and 13% were anemic. Patients with iron deficiency had significantly lower hemoglobin levels throughout the hospital stay and received blood transfusions more frequently during surgical procedures (31% v 19%, p = 0.0361). Detailed analysis showed that patients with iron deficiency received more red blood cell units. There were no differences in postoperative bleeding, morbidity, or mortality. CONCLUSIONS: Iron deficiency appears to be related to lower hemoglobin levels and more frequent transfusions in elective cardiac surgery. Assessing iron status preoperatively and correcting any iron deficiencies should be one of the numerous actions involved in patient blood management for such surgeries, with the aim of reducing morbidity associated with both anemia and transfusion.

Prevalence and nature of statin drug-drug interactions in a university hospital by electronic health record mining.

AIM: Our aim was to describe prevalence, nature, and level of severity of potential statin drug-drug interactions in a university hospital. METHODS: In a cross-sectional study, statin drug-drug interactions were screened from medical record of 10,506 in-patients treated stored in the clinical data warehouse "eHOP." We screened drug-drug interactions using Theriaque and Micromedex drug databases. RESULTS: A total of 22.5% of patients were exposed to at least one statin drug-drug interaction. Given their lipophilicity and CYP3A4 metabolic pathway, atorvastatin and simvastatin presented a higher prevalence of drug-drug interactions while fluvastatin presented the lowest prevalence. Up to 1% of the patients was exposed to a contraindicated drug-drug interaction, the most frequent drug-drug interaction involving influx-transporter (i.e., OATP1B1) interactions between simvastatin or rosuvastatin with cyclosporin. The second most frequent contraindicated drug-drug interaction involved CYP3A4 interaction between atorvastatin or simvastatin with either posaconazole or erythromycin. Furthermore, our analysis showed some discrepancies between Theriaque and Micromedex in the prevalence and the nature of drug-drug interactions. CONCLUSIONS: Different drug-drug interaction profiles were observed between statins with a higher prevalence of CYP3A4-based interactions for lipophilic statins. Analyzing the three most frequent DDIs, the more significant DDIs (level 1: contraindication) were reported for transporter-based DDI involving OATP1B1 influx transporter. These points are of concern to improve prescriptions of statins.

Combining information from a clinical data warehouse and a pharmaceutical database to generate a framework to detect comorbidities in electronic health records.

BACKGROUND: Medical coding is used for a variety of activities, from observational studies to hospital billing. However, comorbidities tend to be under-reported by medical coders. The aim of this study was to develop an algorithm to detect comorbidities in electronic health records (EHR) by using a clinical data warehouse (CDW) and a knowledge database. METHODS: We enriched the Theriaque pharmaceutical database with the French national Comorbidities List to identify drugs associated with at least one major comorbid condition and diagnoses associated with a drug indication. Then, we compared the drug indications in the Theriaque database with the ICD-10 billing codes in EHR to detect potentially missing comorbidities based on drug prescriptions. Finally, we improved comorbidity detection by matching drug prescriptions and laboratory test results. We tested the obtained algorithm by using two retrospective datasets extracted from the Rennes University Hospital (RUH) CDW. The first dataset included all adult patients hospitalized in the ear, nose, throat (ENT) surgical ward between October and December 2014 (ENT dataset). The second included all adult patients hospitalized at RUH between January and February 2015 (general dataset). We reviewed medical records to find written evidence of the suggested comorbidities in current or past stays. RESULTS: Among the 22,132 Common Units of Dispensation (CUD) codes present in the Theriaque database, 19,970 drugs (90.2%) were associated with one or several ICD-10 diagnoses, based on their indication, and 11,162 (50.4%) with at least one of the 4878 comorbidities from the comorbidity list. Among the 122 patients of the ENT dataset, 75.4% had at least one drug prescription without corresponding ICD-10 code. The comorbidity diagnoses suggested by the algorithm were confirmed in 44.6% of the cases. Among the 4312 patients of the general dataset, 68.4% had at least one drug prescription without corresponding ICD-10 code. The comorbidity diagnoses suggested by the algorithm were confirmed in 20.3% of reviewed cases. CONCLUSIONS: This simple algorithm based on combining accessible and immediately reusable data from knowledge databases, drug prescriptions and laboratory test results can detect comorbidities.

Timeline representation of clinical data: usability and added value for pharmacovigilance.

BACKGROUND: Pharmacovigilance consists in monitoring and preventing the occurrence of adverse drug reactions (ADR). This activity requires the collection and analysis of data from the patient record or any other sources to find clues of a causality link between the drug and the ADR. This can be time-consuming because often patient data are heterogeneous and scattered in several files. To facilitate this task, we developed a timeline prototype to gather and classify patient data according to their chronology. Here, we evaluated its usability and quantified its contribution to routine pharmacovigilance using real ADR cases. METHODS: The timeline prototype was assessed using the biomedical data warehouse eHOP (from entrepôt de données biomédicales de l'HOPital) of the Rennes University Hospital Centre. First, the prototype usability was tested by six experts of the Regional Pharmacovigilance Centre of Rennes. Their experience was assessed with the MORAE software and a System and Usability Scale (SUS) questionnaire. Then, to quantify the timeline contribution to pharmacovigilance routine practice, three of them were asked to investigate possible ADR cases with the "Usual method" (analysis of electronic health record data with the DxCare software) or the "Timeline method". The time to complete the task and the data quality in their reports (using the vigiGrade Completeness score) were recorded and compared between methods. RESULTS: All participants completed their tasks. The usability could be considered almost excellent with an average SUS score of 82.5/100. The time to complete the assessment was comparable between methods (P = 0.38) as well as the average vigiGrade Completeness of the data collected with the two methods (P = 0.49). CONCLUSIONS: The results showed a good general level of usability for the timeline prototype. Conversely, no difference in terms of the time spent on each ADR case and data quality was found compared with the usual method. However, this absence of difference between the timeline and the usual tools that have been in use for several years suggests a potential use in pharmacovigilance especially because the testers asked to continue using the timeline after the evaluation.

Factors influencing the development of primary care data collection projects from electronic health records: a systematic review of the literature.

BACKGROUND: Primary care data gathered from Electronic Health Records are of the utmost interest considering the essential role of general practitioners (GPs) as coordinators of patient care. These data represent the synthesis of the patient history and also give a comprehensive picture of the population health status. Nevertheless, discrepancies between countries exist concerning routine data collection projects. Therefore, we wanted to identify elements that influence the development and durability of such projects. METHODS: A systematic review was conducted using the PubMed database to identify worldwide current primary care data collection projects. The gray literature was also searched via official project websites and their contact person was emailed to obtain information on the project managers. Data were retrieved from the included studies using a standardized form, screening four aspects: projects features, technological infrastructure, GPs' roles, data collection network organization. RESULTS: The literature search allowed identifying 36 routine data collection networks, mostly in English-speaking countries: CPRD and THIN in the United Kingdom, the Veterans Health Administration project in the United States, EMRALD and CPCSSN in Canada. These projects had in common the use of technical facilities that range from extraction tools to comprehensive computing platforms. Moreover, GPs initiated the extraction process and benefited from incentives for their participation. Finally, analysis of the literature data highlighted that governmental services, academic institutions, including departments of general practice, and software companies, are pivotal for the promotion and durability of primary care data collection projects. CONCLUSION: Solid technical facilities and strong academic and governmental support are required for promoting and supporting long-term and wide-range primary care data collection projects.

How to Accurately Detect Renal Replacement Therapy Weaning in Intensive Care: Data Quality and Standardization Considerations for the OMOP Common Data Model.

This study assesses the effectiveness of the Observational Medical Outcomes Partnership common data model (OMOP CDM) in standardising Continuous Renal Replacement Therapy (CRRT) data from intensive care units (ICU) of two French university hospitals. Our objective was to extract and standardise data from various sources, enabling the development of predictive models for CRRT weaning that are agnostic to the data's origin. Data for 1,696 ICU stays from the two data sources were extracted, transformed, and loaded into the OMOP format after semantic alignment of 46 CRRT standard concepts. Although the OMOP CDM demonstrated potential in harmonising CRRT data, we encountered challenges related to data variability and the lack of standard concepts. Despite these challenges, our study supports the promise of the OMOP CDM for ICU data standardization, suggesting that further refinement and adaptation could significantly improve clinical decision making and patient outcomes in critical care settings.

Secure Extraction of Personal Information from EHR by Federated Machine Learning.

Secure extraction of Personally Identifiable Information (PII) from Electronic Health Records (EHRs) presents significant privacy and security challenges. This study explores the application of Federated Learning (FL) to overcome these challenges within the context of French EHRs. By utilizing a multilingual BERT model in an FL simulation involving 20 hospitals, each represented by a unique medical department or pole, we compared the performance of two setups: individual models, where each hospital uses only its own training and validation data without engaging in the FL process, and federated models, where multiple hospitals collaborate to train a global FL model. Our findings demonstrate that FL models not only preserve data confidentiality but also outperform the individual models. In fact, the Global FL model achieved an F1 score of 75,7%, slightly comparable to that of the Centralized approach at 78,5%. This research underscores the potential of FL in extracting PIIs from EHRs, encouraging its broader adoption in health data analysis.

Monitoring Variability of Laboratory Results in a Clinical Data Warehouse Using Automatic Dashboard.

Hospital laboratory results are a significant data source in Clinical Data Ware-houses (CDW). To ensure comparability across healthcare organizations and for use in research studies, the results need to be interoperable. The LOINC (Logical Observation Identifiers, Names, and Codes) terminology provides a unique identifier for local codes for lab tests, enabling interoperability. However, in real-world, events occur over time and can disrupt the distribution of lab result values. For example, new equipment may be added to the analysis pipeline, a machine may be replaced, formulas may evolve due to new scientific knowledge, and legacy terminologies may be adopted. This article proposes a pipeline for creating an automated dashboard to monitor these events and data quality. We used automatic change point detection methods such as PELT for event detection in lab results. For a given LOINC code, we create a dashboard that summarizes the number of local codes mapped, and the number of patients (by sex, age, and hospital service) associated with the code. Finally, the dashboard enables the visualization of time events that disrupt the signal distribution. The biologists were able to explain to us the changes for several biological assays.

Intensive Care Quality Indicators Dashboard Using Observational Medical Outcomes Partnership Common Data Model.

This paper presents the development of a visualization dashboard for quality indicators in intensive care units (ICUs), using the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). The dashboard enables the user to visualize quality indicator data using histograms, pie charts and tables. Our project uses the OMOP CDM, ensuring a seamless implementation of our dashboard across various hospitals. Future directions for our research include expanding the dashboard to incorporate additional quality indicators and evaluating clinicians' feedback on its effectiveness.

Applying Machine Learning for Prescriptive Support: A Use Case with Unfractionated Heparin in Intensive Care Units.

Continuous unfractionated heparin is widely used in intensive care, yet its complex pharmacokinetic properties complicate the determination of appropriate doses. To address this challenge, we developed machine learning models to predict over- and under-dosing, based on anti-Xa results, using a monocentric retrospective dataset. The random forest model achieved a mean AUROC of 0.80 [0.77-0.83], while the XGB model reached a mean AUROC of 0.80 [0.76-0.83]. Feature importance was employed to enhance the interpretability of the model, a critical factor for clinician acceptance. After prospective validation, machine learning models such as those developed in this study could be implemented within a computerized physician order entry (CPOE) as a clinical decision support system (CDSS).

Synthesizing Clinical Data Warehouse Content for Enhanced External Collaboration: A Preliminary Proposal.

This paper introduces a novel approach aimed at enhancing the accessibility of clinical data warehouses (CDWs) for external users, particularly researchers and biomedical companies interested in developing and testing their solutions. The primary focus is on proposing a clinical data catalogue designed to elucidate the contents of CDWs, facilitating biomedical project launch and completion. The catalogue is designed to address three fundamental inquiries that external users may have regarding CDWs: "What data is available, how much data is present, and how was it generated?" Additionally, the paper showcases a prototype of the catalogue through a visualization example, utilizing data from the CDW of Rennes University Hospital.

Heterogeneity of right ventricular echocardiographic parameters in systemic lupus erythematosus among four clinical subgroups, as stratified by clinical organ involvement in observational cohort.

BACKGROUND: Systemic lupus erythematosus (SLE) is a heterogeneous autoimmune disease. Cardiac involvement in SLE is rare but plays an important prognostic role. The degree of cardiac involvement according to SLE subsets defined by non-cardiac manifestations is unknown. The objective of this study was to identify differences in transthoracic echocardiography (TTE) parameters associated with different SLE subgroups. METHODS: One hundred eighty-one patients who fulfilled the 2019 American College of Rheumatology/EULAR classification criteria for SLE and underwent baseline TTE were included in this cross-sectional study. We defined four subsets of SLE based on the predominant clinical manifestations. A multivariate multinomial regression analysis was performed to determine whether TTE parameters differed between groups. RESULTS: Four clinical subsets were defined according to non-cardiac clinical manifestations: group A (n=37 patients) showed features of mixed connective tissue disease, group B (n=76 patients) had primarily cutaneous involvement, group C (n=18) exhibited prominent serositis and group D (n=50) had severe, multi-organ involvement, including notable renal disease. Forty TTE parameters were assessed between groups. Per multivariate multinomial regression analysis, there were statistically significant differences in early diastolic tricuspid annular velocity (RV-Ea, p<0.0001), RV S' wave (p=0.0031) and RV end-diastolic diameter (p=0.0419) between the groups. Group B (primarily cutaneous involvement) had the lowest degree of RV dysfunction. CONCLUSION: When defining clinical phenotypes of SLE based on organ involvement, we found four distinct subgroups which showed notable differences in RV function on TTE. Risk-stratifying patients by clinical phenotype could help better tailor cardiac follow-up in this population.

Safety of subcutaneous versus intravenous ceftriaxone administration in older patients: A retrospective study.

BACKGROUND: Antibiotics play a central role in infection management. In older patients, antibiotics are frequently administered subcutaneously. Ceftriaxone pharmacokinetics after subcutaneous administration is well documented, but little data are available on its safety. METHODS: We compared the occurrence of adverse events associated with ceftriaxone administered subcutaneously versus intravenously in ≥75-year-old patients. We used data from a single-center, retrospective, clinical-administrative database to compare the occurrence of adverse events at day 14 and outcome at day 21 in older patients who received ceftriaxone via the subcutaneous route or the intravenous route at Rennes University Hospital, France, from May 2020 to February 2023. RESULTS: The subcutaneous and intravenous groups included 402 and 3387 patients, respectively. Patients in the subcutaneous group were older and more likely to receive palliative care. At least one adverse event was reported for 18% and 40% of patients in the subcutaneous and intravenous group, respectively (RR = 2.21). Mortality at day 21 was higher in the subcutaneous route group, which could be linked to between-group differences in clinical and demographic features. CONCLUSIONS: In ≥75-year-old patients, ceftriaxone administered by the subcutaneous route is associated with less-adverse events than by the intravenous route. The subcutaneous route, which is easier to use, has a place in infection management in geriatric settings.

Phenotyping of heart failure with preserved ejection faction using electronic health records and echocardiography.

Aims: Patients presenting symptoms of heart failure with preserved ejection fraction (HFpEF) are not a homogenous population. Different phenotypes can differ in prognosis and optimal management strategies. We sought to identify phenotypes of HFpEF by using the medical information database from a large university hospital centre using machine learning. Methods and results: We explored the use of clinical variables from electronic health records in addition to echocardiography to identify different phenotypes of patients with HFpEF. The proposed methodology identifies four phenotypic clusters based on both clinical and echocardiographic characteristics, which have differing prognoses (death and cardiovascular hospitalization). Conclusion: This work demonstrated that artificial intelligence-derived phenotypes could be used as a tool for physicians to assess risk and to target therapies that may improve outcomes.