Explore the vast range of titles available.

PurposeEarly clinical recognition of sepsis can be challenging. With the advancement of machine learning, promising real-time models to predict sepsis have emerged. We assessed their performance by carrying out a systematic review and meta-analysis.MethodsA systematic search was performed in PubMed, Embase.com and Scopus. Studies targeting sepsis, severe sepsis or septic shock in any hospital setting were eligible for inclusion. The index test was any supervised machine learning model for real-time prediction of these conditions. Quality of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology, with a tailored Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) checklist to evaluate risk of bias. Models with a reported area under the curve of the receiver operating characteristic (AUROC) metric were meta-analyzed to identify strongest contributors to model performance.ResultsAfter screening, a total of 28 papers were eligible for synthesis, from which 130 models were extracted. The majority of papers were developed in the intensive care unit (ICU, n = 15; 54%), followed by hospital wards (n = 7; 25%), the emergency department (ED, n = 4; 14%) and all of these settings (n = 2; 7%). For the prediction of sepsis, diagnostic test accuracy assessed by the AUROC ranged from 0.68–0.99 in the ICU, to 0.96–0.98 in-hospital and 0.87 to 0.97 in the ED. Varying sepsis definitions limit pooling of the performance across studies. Only three papers clinically implemented models with mixed results. In the multivariate analysis, temperature, lab values, and model type contributed most to model performance.ConclusionThis systematic review and meta-analysis show that on retrospective data, individual machine learning models can accurately predict sepsis onset ahead of time. Although they present alternatives to traditional scoring systems, between-study heterogeneity limits the assessment of pooled results. Systematic reporting and clinical implementation studies are needed to bridge the gap between bytes and bedside.

Background Antibiotic exposure is often inadequate in critically ill patients with severe sepsis or septic shock and this is associated with worse outcomes. Despite markedly altered and rapidly changing pharmacokinetics in these patients, guidelines and clinicians continue to rely on standard dosing schemes. To address this challenge, we developed AutoKinetics, a clinical decision support system for antibiotic dosing. By feeding large amounts of electronic health record patient data into pharmacokinetic models, patient-specific predicted future plasma concentrations are displayed graphically. In addition, a tailored dosing advice is provided at the bedside in real time. To evaluate the effect of AutoKinetics on pharmacometric and clinical endpoints, we are conducting the Right Dose Right Now multicenter, randomized controlled, two-arm, parallel-group, non-blinded, superiority trial. Methods All adult intensive care patients with a suspected or proven infection and having either lactatemia or receiving vasopressor support are eligible for inclusion. Randomization to the AutoKinetics or control group is initiated at the bedside when prescribing at least one of four commonly administered antibiotics: ceftriaxone, ciprofloxacin, meropenem and vancomycin. Dosing advice is available for patients in the AutoKinetics group, whereas patients in the control group receive standard dosing. The primary outcome of the study is pharmacometric target attainment during the first 24 h. Power analysis revealed the need for inclusion of 42 patients per group per antibiotic. Thus, a total of 336 patients will be included, 168 in each group. Secondary pharmacometric endpoints include time to target attainment and fraction of target attainment during an entire antibiotic course. Secondary clinical endpoints include mortality, clinical cure and days free from organ support. Several other exploratory and subgroup analyses are planned. Discussion This is the first randomized controlled trial to assess the effectiveness and safety of bedside data-driven automated antibiotic dosing advice. This is important as adequate antibiotic exposure may be crucial to treat severe sepsis and septic shock. In addition, the trial could prove to be a significant contribution to clinical pharmacometrics and serve as a stepping stone for the use of big data and artificial intelligence in the field. Trial registration Netherlands Trial Register (NTR), NL6501/NTR6689 . Registered on 25 August 2017. European Clinical Trials Database (EudraCT), 2017-002478-37. Registered on 6 November 2017.

Treatment with low-dose glucocorticoids (GCs) (≤7.5 mg prednisolone) in combination with standard care is highly effective in rheumatoid arthritis (RA), but despite 70 years of clinical experience, evidence-based information on its balance of benefit and harm is incomplete. This leads to an ongoing debate, with under- and over-use of GCs as result. The GLORIA pragmatic trial was developed to assess harm, benefit and costs of low-dose GCs added to the standard treatment of older RA patients. The objective of this abstract is to document the baseline status and frequency of comorbid conditions in the GLORIA study population. The results of the unblinded data will be submitted as late-breaking abstract. This double-blind, randomized, placebo-controlled, multicenter trial (1) was open for patients with RA according to the 1987 or 2010 (2) criteria, age ≥65 years, and disease activity score of 28 joints (DAS28) of ≥2.6. Patients were recruited from rheumatology clinics in Germany, Hungary, Italy, The Netherlands, Portugal, Romania and Slovakia. Eligible patients were randomized to two years of treatment with daily 5 mg prednisolone or matching placebo. All other medication was allowed, except for GCs. The presented data are blinded because the database is not closed yet. The population consists of 451 patients with mean disease duration 10.6 (Q1-Q3: 3-15) years. The majority (70%) is female, mean age is 72.5 (Q1-Q3: 68-76, range: 65-88) years, 66% were positive for rheumatoid factor and 56% for ACPA. Patients had a mean of 4.3 (SD 2.8) comorbidities besides RA (3.4 active) and therefore used multiple concomitant medications (3.9 (SD 3.4)) (Table 1). The most common comorbidities (provisional data of 161 patients with complete coding) in this older population are: vascular disorders (58%), musculoskeletal and connective tissue disorders (57%) and a history of surgical and medical procedures (45%). Patients were most frequently on beta blocking agents (22%, mainly metoprolol) and HMG CoA reductase inhibitors (20%, mainly simvastatin). Most patients also have an extensive history of anti-rheumatic treatment. At the start of the trial most patients (82%) were on cDMARD treatment; 15% were on bDMARDs/tsDMARDs. Almost half of the patients previously had been treated with GCs, with a mean duration of 3.4 years and a mean last dose of 4.6 mg/day. The baseline data shows that we have an older study population who have relatively many other comorbidities next to RA and who are almost all treated with multiple concomitant medications in addition to the study medication. Therefore, we expect to report a high adverse event rate. Research among older patients is urgently needed, but the frailty of this population as represented by the multiple comorbidities and concomitant medications have to be taken into account in the analyses and interpretation of the results. [1]Hartman L, Rasch LA, Klausch T, Bijlsma HWJ, Christensen R, Smulders YM, et al. Harm, benefit and costs associated with low-dose glucocorticoids added to the treatment strategies for rheumatoid arthritis in elderly patients (GLORIA trial): study protocol for a randomised controlled trial. Trials. 2018;19:67. [2]Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 2010;62:2569-81. The GLORIA project is funded by the European Union's Horizon 2020 research and innovation programme under the topic ‘‘Personalizing Health and Care“, grant agreement No 634886. None declared Table 1Comorbidities and concomitant medications at baseline in theGLORIA trial.MeanSDRangeComorbidities4.32.80-15Active3.4Past1.9Concomitant medications (count)3.93.40-15Beta blocking agents (%)22HMG CoA reductase inhibitors (%)20Platelet aggregation inhibitors (%)16ACE inhibitors (%)12Angiotensin II antagonists (%)11DAS284.521.05DAS28CRP4.060.97HAQ (0-3)1.20.7RA treatmentCurrent (%)Previous (%)cDMARD8492bDMARD/tsDMARD1522NSAID5129Glucocorticoids049

Background:Adherence is a serious problem in treatment of inflammatory diseases. To measure adherence, caps that record medication bottle openings may be superior to capsule counts (1). In the ongoing two-year GLORIA trial on the addition of low-dose (5 mg) prednisolone or placebo to standard of care in elderly patients (65+ years) with rheumatoid arthritis, adherence was measured in both ways during the whole trial.Objectives:To describe adherence patterns, and to compare adherence as assessed with adherence caps and with capsule counts in the GLORIA trial.Methods:The recorded adherence patterns of patients (blinded for treatment group) were classified according to descriptive categories. Overall adherence according to number of bottle openings was compared with adherence according to the capsule count. Good adherence was defined as 80%: i.e. for caps 80% of days one opening recorded, and for counts less than 20% of prescribed tablets returned at the subsequent visit. Each patient has a maximum of 8 periods of 90 days.Results:Trial inclusion has closed in 2018 at 452 patients; the current dataset contains adherence data of 385 patients. Mean number of recorded 90-day periods per patient was 4 (range 1-8). Based on capsule counts over all periods, 90% of the patients met the 80% threshold of adherence; based on cap data only 31% met this criterion.The four adherence patterns are shown in a calendar matrix, with yellow for zero, green for one and blue for ≥two openings on a day (Figure 1). Bottles were supposed to be opened once a day.Patients were categorized according to the opening pattern seen in at least 50% of assessed periods:32% non-use(<20% of the days an opening);26% stable use(≥80% of the days 1 opening);40% irregular use(different adherence patterns, in or between periods);2% weekly use(1 opening per week).Conclusion:In our trial of elderly rheumatoid arthritis patients, patients appeared to be mostly adherent according to conventional capsule counts. Results from adherence caps were highly discrepant with the capsule counts, with patterns suggesting patients did not use the bottle for daily dispensing, despite specific advice to do so.References:[1] El Alili M, Vrijens B, Demonceau J, Evers SM, Hiligsmann M. A scoping review of studies comparing the medication event monitoring system (MEMS) with alternative methods for measuring medication adherence. Br J Clin Pharmacol 2016;82:268-79.Acknowledgments:The GLORIA project is funded by the European Union’s Horizon 2020 research and innovation programme under the topic ‘’Personalizing Health and Care’’, grant agreement No 634886.Disclosure of Interests:Linda Hartman: None declared, Sabrina Paolino: None declared, Reinhard Bos: None declared, Daniela Opris-Belinski Speakers bureau: as declared, Marc R Kok Grant/research support from: BMS and Novartis, Consultant of: Novartis and Galapagos, Hanneke Griep-Wentink: None declared, Ruth Klaasen: None declared, Cornelia Allaart: None declared, George Bruyn: None declared, Hennie Raterman Grant/research support from: UCB, Consultant of: Abbvie, Amgen, Bristol-Myers Sqibb, Cellgene and Sanofi Genzyme, Marieke Voshaar Grant/research support from: part of phd research, Speakers bureau: conducting a workshop (Pfizer), Nuno Gomes: None declared, Rui Pinto: None declared, Thomas Klausch: None declared, WIllem Lems Grant/research support from: Pfizer, Consultant of: Lilly, Pfizer, Maarten Boers: None declared

Background:Medication adherence in the GLORIA trial, among elderly patients with rheumatoid arthritis, is measured with caps that register openings of the medication bottle. At each study visit, one or two medication bottles with cap (kits) are dispensed, each containing 90 capsules. Multiple steps are needed to come to a workable dataset to describe adherence.Objectives:To describe the steps that are needed to come from raw data to a workable dataset to analyze adherence data that are recorded by electronic caps.Methods:The medication bottle contains a cap with the ability to register cap openings. The raw dataset from the caps consist of an excel file with one opening event per row, recorded as date and time. One cap yields approximately 90 rows. First, the kit numbers were matched to the corresponding patient numbers, that are recorded in another excel file. Instances where two kits were dispensed were recorded with two kit numbers in one cell and need to be copied to two cells with one kit number. Second, the VLOOKUP function was used to combine dates and kit numbers. One row now contains all openings from one kit. Then, the number of days between first opening and each next opening date was calculated. A range of 90 days was made to calculate how many times the bottle was opened on each day of the 90-days period. The results were color-coded to visualize instances of zero, one or ≥two openings on a day.Results:The colored calendar matrix (Figure 1) can now be used to categorize adherence patterns.Conclusion:A monitoring cap seems a simple instrument to measure adherence. However, multiple steps and a lot of time are needed to come to a workable dataset for the study of adherence patterns.Acknowledgments:The GLORIA project is funded by the European Union’s Horizon 2020 research and innovation programme under the topic ‟Personalizing Health and Care’’, grant agreement No 634886.Disclosure of Interests:Linda Hartman: None declared, Elisa Alessandri: None declared, Reinhard Bos: None declared, Daniela Opris-Belinski Speakers bureau: as declared, Marc R Kok Grant/research support from: BMS and Novartis, Consultant of: Novartis and Galapagos, Hanneke Griep-Wentink: None declared, Ruth Klaasen: None declared, Cornelia Allaart: None declared, George Bruyn: None declared, Hennie Raterman Grant/research support from: UCB, Consultant of: Abbvie, Amgen, Bristol-Myers Sqibb, Cellgene and Sanofi Genzyme, Marieke Voshaar Grant/research support from: part of phd research, Speakers bureau: conducting a workshop (Pfizer), Nuno Gomes: None declared, Rui Pinto: None declared, Thomas Klausch: None declared, WIllem Lems Grant/research support from: Pfizer, Consultant of: Lilly, Pfizer, Maarten Boers: None declared

Background Rheumatoid arthritis (RA) is a chronic inflammatory disease of the joints affecting 1% of the world population. It has major impact on patients through disability and associated comorbidities. Current treatment strategies have considerably improved the prognosis, but recent innovations (especially biologic drugs and the new class of so-called “JAK/STAT inhibitors”) have important safety issues and are very costly. Glucocorticoids (GCs) are highly effective in RA, and could reduce the need for expensive treatment with biologic agents. However, despite more than 65 years of clinical experience, there is a lack of studies large enough to adequately document the benefit/harm balance. The result is inappropriate treatment strategies, i.e. both under-use and over-use of GCs, and consequently suboptimal treatment of RA. Methods The GLORIA study is a pragmatic multicentre, 2-year, randomised, double-blind, clinical trial to assess the safety and effectiveness of a daily dose of 5 mg prednisolone or matching placebo added to standard of care in elderly patients with RA. Eligible participants are diagnosed with RA, have inadequate disease control (disease activity score, DAS28 ≥ 2.6), and are ≥ 65 years. The primary outcome measures are the time-averaged mean value of the DAS28 and the occurrence of serious adverse events or adverse events of special interest. During the trial, change in antirheumatic therapy is permitted as clinically indicated, except for GCs. Cost-effectiveness and cost-utility are secondary outcomes. The main challenge is the interpretation of the trial result with two primary endpoints and the pragmatic trial design that allows co-interventions. Another challenge is the definition of safety and the relative lack of power to detect differences between treatment groups. We have chosen to define safety as the number of patients experiencing at least one serious adverse event. We also specify a decision tree to guide our conclusion on the balance of benefit and harm, and our methodology to combat potential confounding caused by co-interventions. Discussion Pragmatic trials minimise impact on daily practice and maximise clinical relevance of the results, but analysis and interpretation of the results is challenging. We expect that the results of this trial are of importance for all rheumatologists who treat elderly patients with RA. Trial registration ClinicalTrials.gov, NCT02585258 . Registered on 20 October 2015.

Over 95% of head and neck cancers are squamous cell carcinoma (HNSCC). HNSCC is mostly diagnosed late, causing a poor prognosis despite the application of invasive treatment protocols. Tumor-educated platelets (TEPs) have been shown to hold promise as a molecular tool for early cancer diagnosis. We sequenced platelet mRNA isolated from blood of 101 patients with HNSCC and 101 propensity-score matched noncancer controls. Two independent machine learning classification strategies were employed using a training and validation approach to identify a cancer predictor: a particle swarm optimized support vector machine (PSO-SVM) and a least absolute shrinkage and selection operator (LASSO) logistic regression model. The best performing PSO-SVM predictor consisted of 245 platelet transcripts and reached a maximum area under the curve (AUC) of 0.87. For the LASSO-based prediction model, 1,198 mRNAs were selected, resulting in a median AUC of 0.84, independent of HPV status. Our data show that TEP RNA classification by different AI tools is promising in the diagnosis of HNSCC.