Development of a Novel Statistical Model for Predicting Clinical Outcomes in Stroke Patients With Tandem Occlusions After Endovascular Therapy.

BACKGROUND: Tandem occlusions are intracranial large vessel occlusions (LVOs) with a concomitant ipsilateral extracranial internal carotid artery occlusion and can cause more severe stroke symptoms. AIM: To develop a simple, rigorously cross-validated novel tool to predict clinical outcomes following tandem occlusion in patients with acute LVO stroke, based on data that are easily available to clinicians. To have used machine learning approaches to utilize the available information from clinical and angiographic data to make predictive models able to distinguish between mortality versus survival and good (modified Rankin Scale (mRS) ≤ 2) versus unfavorable neurological outcomes (mRs ≥ 3) Materials and methods: Retrospective data from 87 consecutive patients with anterior circulation stroke and tandem occlusions who underwent mechanical thrombectomy and stenting between December 2009 and January 2020 were analyzed. Patients were stratified into three groups based on the location of their LVO, and these groups were compared using statistical tests. Predictive models were built and cross-validated 1000 times to estimate their predictive power, measured by accuracy and area under the receiver operating curve (AUROC). RESULTS: For distinguishing good outcome (mRS ≤ 2) versus poor outcome (mRS ≥ 3), the model comprised age, initial National Institute of Health Stroke Scale (NIHSS) score, Alberta Stroke Program Early CT Score (ASPECTS), NIHSS at 24 hours, NIHSS at discharge and intracranial haemorrhage and yielded an accuracy of 83% and the AUROC of 0.91. For mortality prediction, the model comprised age, initial NIHSS, intravenous thrombolysis, NIHSS at 24 hours and NIHSS at discharge and yielded an accuracy of 91% and an AUROC of 0.94. CONCLUSIONS: Models developed exhibit strong predictive performance and can distinguish between both the instances of survival versus mortality and good versus poor outcome with an aim to support clinicians in deciding on optimal management for these complex patients. The developed model will help identify those at risk of poorer outcomes and the prospective better selection of patients with acute ischaemic large vessel stroke secondary to tandem occlusions.

Impact of Extremes of BMI on Outcomes following Lung Resection.

BACKGROUND: Body mass index (BMI) has been shown to be an independent predictor of survival following lung resection surgery. This study aimed to quantify the short- to midterm impact of abnormal BMI on postoperative outcomes. METHODS: Lung resections at a single institution were examined between 2012 and 2021. Patients were divided into low BMI (<18.5), normal/high BMI (18.5-29.9), and obese BMI (>30). Postoperative complications, length of stay, and 30- and 90-day mortality were examined. RESULTS: A total of 2,424 patients were identified. Of these patients, 2.6% (n = 62) had a low BMI, 67.4% (n = 1,634) had a normal/high BMI, and 30.0% (n = 728) had an obese BMI. Overall postoperative complications were higher in the low BMI group (43.5%) when compared with normal/high (30.9%) and obese BMI group (24.3%) (p = 0.0002). Median length of stay was significantly higher in the low BMI group (8.3 days) compared with 5.2 days in the normal/high and obese BMI groups (p < 0.0001). Ninety-day mortality was higher in the low (16.1%) compared with the normal/high (4.5%) and obese BMI groups (3.7%) (p = 0.0006). Subgroup analysis of the obese cohort did not elucidate any statistically significant differences in overall complications in the morbidly obese. Multivariate analysis determined that BMI is an independent predictor of reduced postoperative complications (odds ratio [OR], 0.96; 95% confidence interval [CI], 0.94-0.97; p < 0.0001) and 90-day mortality (OR, 0.96; 95% CI, 0.92-0.99; p = 0.02). CONCLUSION: Low BMI is associated with significantly worse postoperative outcomes and an approximate fourfold increase in mortality. In our cohort, obesity is associated with reduced morbidity and mortality following lung resection surgery, confirming the existence of the obesity paradox.

Smoking Status and Outcomes following Lung Resection.

BACKGROUND: Surgical resection is the gold standard treatment for the management of early-stage lung cancer. Several modifiable factors may significantly influence postoperative morbidity and mortality. We examined the outcomes of patients following lung resection based upon preoperative smoking status to quantify the impact on postoperative outcomes. METHODS: Data from consecutive lung resections from January 1, 2012 to June 11, 2021 were included. Biopsies for interstitial lung disease and resections for emphysematous lung or bullae were excluded. Patients were divided into three cohorts: current smokers (those who smoked within 4 weeks of surgery), ex-smokers (those who stopped smoking prior to 4 weeks leading up to surgery), and nonsmokers (those who have never smoked). Patient's preoperative variables, postoperative complications, length of stay, and mortality were examined. RESULTS: A total of 2,426 patients were included in the study. A total of 502 patients (20.7%) were current smokers, 1,445 (59.6%) were ex-smokers and 479 patients (19.7%) nonsmokers. Of those smoking immediately prior to surgery 36.9% developed postoperative complications. Lower respiratory tract infections (18.1%) and prolonged air leak (17.1%), in particular, were significant higher in smokers. 90-day mortality (5.8%) was higher in the current smokers when compared with ex- and nonsmokers (5.3 and 1%, respectively). Median length of hospital stay, readmissions, and cost of hospital stay was also higher in the current smoker cohort. CONCLUSION: Smoking immediately prior to surgery is associated with an increase in morbidity, mortality, and length of stay. Not only does this have a significant individual impact, but it is also associated with a significant financial burden to the National Health Service.

Is intraprocedural intravenous aspirin safe for patients who require emergent extracranial stenting during mechanical thrombectomy?

BACKGROUND: Intraoperative antiplatelet therapy is recommended for emergent stenting during mechanical thrombectomy (MT). Most patients undergoing MT are also given thrombolysis. Antiplatelet agents are contraindicated within 24 hours of thrombolysis. We evaluated outcomes and complications of patients stented with and without intravenous aspirin during MT. METHODS: All patients who underwent emergent extracranial stenting during MT at the Royal Stoke University Hospital, UK between 2010 and 2020, were included. Patients were thrombolysed before MT, unless contraindicated. Aspirin 500 mg intravenously was given intraoperatively at the discretion of the operator. Symptomatic intracranial haemorrhage (sICH) and the National Institutes for Health Stroke Scale score (NIHSS) were recorded at 7 days, and mortality and functional recovery (modified Rankin Scale: mRS ≤2) at 90 days. RESULTS: Out of 565 patients treated by MT 102 patients (median age 67 IQR 57-72 years, baseline median NIHSS 18 IQR 13-23, 76 (75%) thrombolysed) had a stent placed. Of these 49 (48%) were given aspirin and 53 (52%) were not. Patients treated with aspirin had greater NIHSS improvement (median 8 IQR 1-16 vs median 3 IQR -9-8 points, p=0.003), but there were no significant differences in sICH (2/49 (4%) vs 9/53 (17%)), mRS ≤2 (25/49 (51%) vs 19/53 (36%)) and mortality (10/49 (20%) vs 12/53 (23%)) with and without aspirin. NIHSS improvement (median 12 IQR 4-18 vs median 7 IQR -7-10, p=0.01) was greater, and mortality was lower (4/33 (12%) vs 6/15 (40%), p=0.05) when aspirin was combined with thrombolysis, than for aspirin alone, with no increase in bleeding. CONCLUSION: Our findings based on registry data derived from routine clinical care suggest that intraprocedural intravenous aspirin in patients undergoing emergent stenting during MT does not increase sICH and is associated with good clinical outcomes, even when combined with intravenous thrombolysis.

CARRoT: R-package for predictive modelling by means of regression, adjusted for multiple regularisation methods.

We present an R-package for predictive modelling, CARRoT (Cross-validation, Accuracy, Regression, Rule of Ten). CARRoT is a tool for initial exploratory analysis of the data, which performs exhaustive search for a regression model yielding the best predictive power with heuristic 'rules of thumb' and expert knowledge as regularization parameters. It uses multiple hold-outs in order to internally validate the model. The package allows to take into account multiple factors such as collinearity of the predictors, event per variable rules (EPVs) and R-squared statistics during the model selection. In addition, other constraints, such as forcing specific terms and restricting complexity of the predictive models can be used. The package allows taking pairwise and three-way interactions between variables into account as well. These candidate models are then ranked by predictive power, which is assessed via multiple hold-out procedures and can be parallelised in order to reduce the computational time. Models which exhibited the highest average predictive power over all hold-outs are returned. This is quantified as absolute and relative error in case of continuous outcomes, accuracy and AUROC values in case of categorical outcomes. In this paper we briefly present statistical framework of the package and discuss the complexity of the underlying algorithm. Moreover, using CARRoT and a number of datasets available in R we provide comparison of different model selection techniques: based on EPVs alone, on EPVs and R-squared statistics, on lasso regression, on including only statistically significant predictors and on stepwise forward selection technique.

Bayesian networks identify determinants of outcomes following cardiac surgery in a UK population.

BACKGROUND: Traditional risk stratification tools do not describe the complex principle determinant relationships that exist amongst pre-operative and peri-operative factors and their influence on cardiac surgical outcomes. This paper reports on the use of Bayesian networks to investigate such outcomes. METHODS: Data were prospectively collected from 4776 adult patients undergoing cardiac surgery at a single UK institute between April 2012 and May 2019. Machine learning techniques were used to construct Bayesian networks for four key short-term outcomes including death, stroke and renal failure. RESULTS: Duration of operation was the most important determinant of death irrespective of EuroSCORE. Duration of cardiopulmonary bypass was the most important determinant of re-operation for bleeding. EuroSCORE was predictive of new renal replacement therapy but not mortality. CONCLUSIONS: Machine-learning algorithms have allowed us to analyse the significance of dynamic processes that occur between pre-operative and peri-operative elements. Length of procedure and duration of cardiopulmonary bypass predicted mortality and morbidity in patients undergoing cardiac surgery in the UK. Bayesian networks can be used to explore potential principle determinant mechanisms underlying outcomes and be used to help develop future risk models.

Genome-wide RNA polymerase stalling shapes the transcriptome during aging.

Gene expression profiling has identified numerous processes altered in aging, but how these changes arise is largely unknown. Here we combined nascent RNA sequencing and RNA polymerase II chromatin immunoprecipitation followed by sequencing to elucidate the underlying mechanisms triggering gene expression changes in wild-type aged mice. We found that in 2-year-old liver, 40% of elongating RNA polymerases are stalled, lowering productive transcription and skewing transcriptional output in a gene-length-dependent fashion. We demonstrate that this transcriptional stress is caused by endogenous DNA damage and explains the majority of gene expression changes in aging in most mainly postmitotic organs, specifically affecting aging hallmark pathways such as nutrient sensing, autophagy, proteostasis, energy metabolism, immune function and cellular stress resilience. Age-related transcriptional stress is evolutionary conserved from nematodes to humans. Thus, accumulation of stochastic endogenous DNA damage during aging deteriorates basal transcription, which establishes the age-related transcriptome and causes dysfunction of key aging hallmark pathways, disclosing how DNA damage functionally underlies major aspects of normal aging.

The use of hypnotherapy as treatment for functional stroke: A case series from a single center in the UK.

BACKGROUND: Functional neurological disorder is defined by symptoms not explained by the current model of disease and its pathophysiology. It is found in 8.4% of patients presenting as acute stroke. Treatment is difficult and recurrence rates are high. We introduced hypnotherapy as a therapeutic option in addition to standard stroke unit care. METHODS: This is an observational study of successive patients with functional neurological disorder presenting as acute stroke treated with hypnotherapy between 1 April 2014 and 1 February 2018. The diagnosis of functional neurological disorder was confirmed by clinical examination and computed tomography/magnetic resonance imaging. Hypnosis was delivered by a hypnotherapy trained stroke physician using imagery for induction. A positive response was defined as a National Institutes of Health Stroke score reduction to 0 or by ≥4 points posthypnotherapy. Costs were calculated as therapist time and benefits as reduction in disability/bed days. RESULTS: Sixty-eight patients (mean age 36.4 years, 52 (76%) females, mean baseline National Institutes of Health Stroke 5.0 (range 1-9)) were included. Two patients (3%) could not be hypnotized. Fifty-eight 58 (85%) responded, 47 (81%) required one treatment session, while 19% needed up to three sessions for symptomatic improvement. No adverse events were observed. Disability (modified Rankin Scale) reduced from a mean of 2.3 to 0.5 resulting in an average cost saving of £1,658 per patient. Most (n = 50, 86%) remained well without recurrence at six-month follow-up. CONCLUSIONS: In this case series, hypnotherapy was associated with rapid and sustained recovery of symptoms. A prospective randomized controlled study is required to confirm the findings and establish generalizability of the results.

Impact of COVID-19 on stroke admissions, treatments, and outcomes at a comprehensive stroke centre in the United Kingdom.

INTRODUCTION: The coronavirus disease (COVID-19) pandemic has changed routine clinical practice worldwide with major impacts on the provision of care and treatment for stroke patients. METHODS: This retrospective observational study included all patients admitted to the Royal Stoke University Hospital in Stoke-on-Trent, UK, with a stroke or transient ischaemic attack between March 15th and April 14th, 2020 (COVID). Patient demographics, characteristics of the stroke, treatment details and logistics were compared with patients admitted in the corresponding weeks in the year before (2019). RESULTS: There was a 39.5% (n = 101 vs n = 167) reduction in admissions in the COVID cohort compared with 2019 with more severe strokes (median National Institutes of Health Stroke Scale (NIHSS) 7 vs 4, p = 0.02), and fewer strokes with no visible acute pathology (21.8 vs 37.1%, p = 0.01) on computed tomography. There was no statistically significant difference in the rates of thrombolysis (10.9 vs 13.2%, p = 0.72) and/or thrombectomy (5.9 vs 4.8%, p = 0.90) and no statistically significant difference in time from stroke onset to arrival at hospital (734 vs 576 min, p = 0.34), door-to-needle time for thrombolysis (54 vs 64 min, p = 0.43) and door-to-thrombectomy time (181 vs 445 min, p = 0.72). Thirty-day mortality was not significantly higher in the COVID year (10.9 vs 8.9%, p = 0.77). None of the 7 stroke patients infected with COVID-19 died. CONCLUSIONS: During the COVID-19 pandemic, the number of stroke admissions fell, and stroke severity increased. There was no statistically significant change in the delivery of thrombolysis and mechanical thrombectomy and no increase in mortality.

Identification of a Glass Substrate to Study Cells Using Fourier Transform Infrared Spectroscopy: Are We Closer to Spectral Pathology?

The rising incidence of cancer worldwide is causing an increase in the workload in pathology departments. This, coupled with advanced analysis methodologies, supports a developing need for techniques that could identify the presence of cancer cells in cytology and tissue samples in an objective, fast, and automated way. Fourier transform infrared (FT-IR) microspectroscopy can identify cancer cells in such samples objectively. Thus, it has the potential to become another tool to help pathologists in their daily work. However, one of the main drawbacks is the use of glass substrates by pathologists. Glass absorbs IR radiation, removing important mid-IR spectral data in the fingerprint region (1800 cm-1 to 900 cm-1). In this work, we hypothesized that, using glass coverslips of differing compositions, some regions within the fingerprint area could still be analyzed. We studied three different types of cells (peripheral blood mononuclear cells, a leukemia cell line, and a lung cancer cell line) and lymph node tissue placed on four different types of glass coverslips. The data presented here show that depending of the type of glass substrate used, information within the fingerprint region down to 1350 cm-1 can be obtained. Furthermore, using principal component analysis, separation between the different cell lines was possible using both the lipid region and the fingerprint region between 1800 cm-1 and 1350 cm-1. This work represents a further step towards the application of FT-IR microspectroscopy in histopathology departments.

Integrated testing strategies can be optimal for chemical risk classification.

There is an urgent need to refine strategies for testing the safety of chemical compounds. This need arises both from the financial and ethical costs of animal tests, but also from the opportunities presented by new in-vitro and in-silico alternatives. Here we explore the mathematical theory underpinning the formulation of optimal testing strategies in toxicology. We show how the costs and imprecisions of the various tests, and the variability in exposures and responses of individuals, can be assembled rationally to form a Markov Decision Problem. We compute the corresponding optimal policies using well developed theory based on Dynamic Programming, thereby identifying and overcoming some methodological and logical inconsistencies which may exist in the current toxicological testing. By illustrating our methods for two simple but readily generalisable examples we show how so-called integrated testing strategies, where information of different precisions from different sources is combined and where different initial test outcomes lead to different sets of future tests, can arise naturally as optimal policies.