ImportanceSelection of antiseizure medications (ASMs) for epilepsy remains largely a trial-and-error approach. Under this approach, many patients have to endure sequential trials of ineffective treatments until the “right drugs” are prescribed.ObjectiveTo develop and validate a deep learning model using readily available clinical information to predict treatment success with the first ASM for individual patients.Design, Setting, and ParticipantsThis cohort study developed and validated a prognostic model. Patients were treated between 1982 and 2020. All patients were followed up for a minimum of 1 year or until failure of the first ASM. A total of 2404 adults with epilepsy newly treated at specialist clinics in Scotland, Malaysia, Australia, and China between 1982 and 2020 were considered for inclusion, of whom 606 (25.2%) were excluded from the final cohort because of missing information in 1 or more variables.ExposuresOne of 7 antiseizure medications.Main Outcomes and MeasuresWith the use of the transformer model architecture on 16 clinical factors and ASM information, this cohort study first pooled all cohorts for model training and testing. The model was trained again using the largest cohort and externally validated on the other 4 cohorts. The area under the receiver operating characteristic curve (AUROC), weighted balanced accuracy, sensitivity, and specificity of the model were all assessed for predicting treatment success based on the optimal probability cutoff. Treatment success was defined as complete seizure freedom for the first year of treatment while taking the first ASM. Performance of the transformer model was compared with other machine learning models.ResultsThe final pooled cohort included 1798 adults (54.5% female; median age, 34 years [IQR, 24-50 years]). The transformer model that was trained using the pooled cohort had an AUROC of 0.65 (95% CI, 0.63-0.67) and a weighted balanced accuracy of 0.62 (95% CI, 0.60-0.64) on the test set. The model that was trained using the largest cohort only had AUROCs ranging from 0.52 to 0.60 and a weighted balanced accuracy ranging from 0.51 to 0.62 in the external validation cohorts. Number of pretreatment seizures, presence of psychiatric disorders, electroencephalography, and brain imaging findings were the most important clinical variables for predicted outcomes in both models. The transformer model that was developed using the pooled cohort outperformed 2 of the 5 other models tested in terms of AUROC.Conclusions and RelevanceIn this cohort study, a deep learning model showed the feasibility of personalized prediction of response to ASMs based on clinical information. With improvement of performance, such as by incorporating genetic and imaging data, this model may potentially assist clinicians in selecting the right drug at the first trial.
Objectives: To describe the clinical characteristics and evaluate the long-term treatment outcomes in older people with newly diagnosed epilepsy over the past 30 years. Methods: We included patients newly diagnosed with epilepsy and commenced on antiseizure medications (ASMs) at age 65 years or older between July 1982 and October 2012 at the Western infirmary in Glasgow, Scotland. They were followed up until April 2016 or death. Seizure freedom was defined as no seizure for at least 1 year on unchanged medication at the last follow-up. Results: A total of 201 patients (median age 73 years, 59% male) were included. The median duration from initial seizure to starting treatment was 8 months (interquartile range: 3.0-24.0 months); 42.2% (85/201) patients had more than five seizures before commencing treatment. Brain imaging showed potentially epileptogenic lesions in 19.7% (38/193) of patients and other abnormalities in 56.5% (109/193); 78.6% patients (158/201) were seizure-free at the last follow-up, of whom 94.9% were taking monotherapy. Concomitant aspirin use (n = 80) was associated with a lower probability of being seizure-free (relative risk 0.82, 95% confidence interval 0.70-0.97; P = .02). The use of second-generation ASMs as the initial monotherapy increased from 31.5% (23/73) before 2000 to 70.3% (90/128, P < .001) from 2000 onward. However, the seizure freedom rates (67.1% vs 55.5%; P = .35) and intolerable adverse-effect rates (16.4% vs 19.5%; P = .45) did not show any significant difference. Significance: There was often a long interval between seizure onset and the initiation of treatment in older people with new-onset epilepsy, although the majority responded well to ASM treatment. Brain imaging showed a high rate of abnormalities. Despite the increased use of second-generation ASMs, treatment outcomes in later-onset epilepsy have not improved over time. The possible effect of aspirin on treatment response warrants further investigation.
Objectives: To assess the temporal trends in the use of second antiseizure (ASM) regimens and compare the efficacy of substitution monotherapy and combination therapy after failure of initial monotherapy in people with epilepsy.Methods: This was a longitudinal observational cohort study conducted at the Epilepsy Unit of the Western Infirmary in Glasgow, Scotland. We included patients who were newly treated for epilepsy with ASMs between July 1982, and October 2012. All patients were followed up for a minimum of 2 years. Seizure freedom was defined as no seizure for at least 1 year on unchanged medication at the last follow up.Results: During the study period, 498 patients were treated with a second ASM regimen after failure of the initial ASM monotherapy, of whom 346 (69%) were prescribed combination therapy and 152 (31%) were given substitution monotherapy. The proportion of patients receiving second regimen as combination therapy increased during the study period from 46% in first epoch (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994) to 78% in the last (2005-2015) (RR = 1.66, 95% CI: 1.17-2.36, corrected-p = .010).Overall, 21% (104/498) of the patients achieved seizure freedom on the second ASM regimen, which was less than half of the seizure-free rate on the initial ASM monotherapy (45%, p < .001). Patients who received substitution monotherapy had similar seizure-free rate compared with those who received combination therapy (RR = 1.17, 95% CI: 0.81-1.69, p = .41). Individual ASMs used, either alone or in combination, had similar efficacy. However, the subgroup analysis was limited by small sample sizes. Significance:The choice of second regimen used based on clinical judgment was not associated with treatment outcome in patients whose initial monotherapy failed due to poor seizure control. Alternative approaches such as machine learning should be explored to aid individualized selection of the second ASM regimen.
Pituitary tumour apoplexy is a rare but potentially life threatening clinical syndrome that mostly results from haemorrhage in the pre-existent tumour. Pure ischaemic subtype of apoplexy is even rarer. The presentation can be hard to differentiate clinically from bacterial meningitis. Moreover, the presence of one does not necessarily exclude the other and early diagnosis of both conditions is imperative for timely management. We report a case of ischaemic pituitary tumour apoplexy that may have precipitated in the setting of bacterial meningitis.
Tetanus remains a significant cause of mortality especially in the developing world. Early diagnosis and institution of treatment is critical to prevent fatal complications. The diagnosis is made on clinical grounds, which may sometimes be difficult, especially in case of localised tetanus. Being able to diagnose tetanus objectively is invaluable in such cases. In this regard, masseter inhibitory reflex (MIR) is a simple neurophysiological test that can be performed at the bedside. Herein, we report a case of craniocervical tetanus that was objectively diagnosed using MIR and adequately treated.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.