Search

Objectives To perform a systematic comparison of in-hospital mortality risk prediction post-cardiac surgery, between the predominant scoring system-European System for Cardiac Operative Risk Evaluation(EuroSCOREII), logistic regression(LR) retrained on the same variables and alternative machine learning techniques(ML)-random forest(RF), neural networks(NN),XGBoost and weighted support vector machine(SVM). Methods Retrospective analyses of prospectively routinely collected data on adult patients undergoing cardiac surgery in the UK from January 2012-March 2019. Data was temporally split 70:30 into training and validation subsets. Mortality prediction models were created using the 18 variables of EuroSCOREII. Comparisons of discrimination, calibration and clinical utility were then conducted. Changes in model performance, variable-importance over time and hospital/operation-based model performance were also reviewed. Results Of the 227,087 adults who underwent cardiac surgery during the study period there were 6,258 deaths(2.76%). In the testing cohort, there was an improvement in discrimination XGBoost(95% confidence interval(CI) Area Under the Curve(AUC): 0.834-0.834,F1 score:0.276-0.280) and RF(95%CI AUC:0.833-0.834,F1:0.277-0.281) compared with EuroSCOREII(95%CI AUC:0.817-0.818,F1: 0.243-0.245).There was no significant improvement in calibration with ML and retrained-LR compared to EuroSCOREII. However, EuroSCOREII overestimated risk across all deciles of risk and over time. The calibration drift was lowest in NN, XGBoost and RF compared with EuroSCOREII. Decision curve analysis showed XGBoost and RF to have greater net benefit than EuroSCOREII. Conclusions ML techniques showed some statistical improvements over retrained-LR and EuroSCOREII. The clinical impact of this improvement is modest at present. However the incorporation of additional risk factors in future studies may improve upon these findings and warrants further study.

Background Left ventricular free wall rupture (LVFWR) and interventricular septal rupture (VSR) are potentially catastrophic mechanical complications after acute myocardial infarction (AMI). When they occur together, “double myocardial rupture” (DMR), survival is unlikely. DMR is seen in only 0.3% of all AMIs. With or without surgical intervention, the odds are against the patient. Case presentation A 57-year-old male self-referred to the emergency department of a remote hospital 5 days after first experiencing chest pain. Investigations in ED confirmed an inferior ST-segment elevation myocardial infarction (STEMI) complicated by DMR. Coronary angiography revealed a mid-course total occlusion of the right coronary artery (RCA). He was rapidly transferred to our regional cardiac surgical unit, arriving straight into the operating theatre, in cardiogenic shock. He was briefly conscious, before arresting prior to intubation and being massaged onto bypass. Not only did he survive the all-night operation, requiring a mitral valve replacement in the process, but he survived multiple postoperative complications to be eventually transferred on postoperative day 66, neurologically intact, to a peripheral unit to complete his rehabilitation. He was subsequently discharged home 88 days after the operation and was able to ambulate with a walking frame into his first postoperative follow-up clinic appointment. Conclusions Our patient, against all odds, has survived DMR and multiple postoperative complications. We present the details of his case and the literature surrounding the condition. The patient’s mental fortitude and his supportive family played a big part, along with excellent team working in our unit.

To evaluate the impact of aortic root abscess (ARA) on the postoperative outcomes of surgically managed infective endocarditis (IE) and to inform optimal surgical approach.Between 2009 and 2020, 143 consecutive patients who underwent surgical management for aortic-valve IE were included in a retrospective cohort study. Multivariable and propensity-weighted analyses were used to adjust for demographic imbalances between those without (n = 93; NARA) and with an ARA (n = 50). Additionally, empirical subgroup analysis appraised the two most used surgical techniques; patch reconstruction (PR) and aortic root replacement (ARR).Demographic characteristics were similar between ARA and NARA except for logistic EuroSCORE, previous valve surgery, and multivalvular infection. In-hospital mortality was 8% and 12% in NARA and ARA, respectively (p = .38), with mortality rates consistently nonsignificantly higher in ARA across all time periods. The overall reoperation rate was also higher in ARA (27% vs. 14%; p = .09) and ARA was shown to be associated with late reoperation (odds ratio [OR] = 2.74; 95% confidence interval [CI] = 1.18-6.36). Patients treated with an ARR showed a 16% increase in late mortality when compared with PR (40% vs. 24%; p = .27) and a 17% lower reoperation rate (14% vs. 31%; p = .24). Propensity-weighted analysis identified ARR as a significant protective factor for reoperation (hazard ratio = 0.05; 95% CI = 0.01-0.34).The presence of an ARA in aortic valve endocarditis was not associated with significantly higher early and late mortality but is linked with a higher reoperation rate at our institution. ARR in ARA is protective from reoperation so should be considered best practice in this setting.

Ischemia and malperfusion are strong predictors of poor postoperative outcomes in type A acute aortic dissection (TAAAD). Serum lactate is an accurate surrogate point-of-care marker of malperfusion. The aim of this study is to investigate the correlation between lactate, in-hospital outcomes, and 1-year survival following TAAAD repair.One hundred and thirty-two patients underwent operative repair of TAAAD over a 4-year period at our institution 128 patients had serum lactate measurements at three stages peri-operatively-preoperatively, at the end of cardiopulmonary bypass (post-CPB) and 6 h postintensive care unit (ICU) admission. The primary outcomes were in-hospital mortality and 1-year survival. The secondary outcomes were the incidences of in-hospital morbidities.Patients were divided into two groups: 88 (68.8%) with normal lactate and 40 (31.2%) with elevated lactate (2.2 mmol/L). Lactate measured preoperatively (odds ratio 1.52, 95% confidence interval 1.17-2.07, p .01), post-CPB (1.34, 1.14-1.64, p .01) and 6 h post-ICU admission (1.29, 1.08-1.55, p .01) was an independent predictor of in-hospital mortality. Following adjustment for the Penn Classification, lactate continued to have a significant correlation with in-hospital mortality at all three timepoints. There was a higher incidence of complications in the elevated lactate group and especially hemofiltration (20% vs. 9.1%, p = .08). 1-year survival was similar in both groups (p = .23).There is a direct correlation between elevated serum lactate and postoperative mortality after TAAAD repair, which is independent of the Penn Classification status on admission.

ObjectivesThe Society of Thoracic Surgeons (STS), and EuroSCORE II (ES II) risk scores, are the most commonly used risk prediction models for adult cardiac surgery post-operative in-hospital mortality. However, they are prone to miscalibration over time, and poor generalisation across datasets and their use remain controversial. It has been suggested that using Machine Learning (ML) techniques, a branch of Artificial intelligence (AI), may improve the accuracy of risk prediction. Despite increased interest, a gap in understanding the effect of dataset drift on the performance of ML over time remains a barrier to its wider use in clinical practice. Dataset drift occurs when a machine learning system underperforms because of a mismatch between the dataset it was developed and the data on which it is deployed. Here we analyse this potential concern in a large United Kingdom (UK) database.MethodsA retrospective analyses of prospectively routinely gathered data on adult patients undergoing cardiac surgery in the UK between 2012-2019. We temporally split the data 70:30 into a training and validation subset. ES II and five ML mortality prediction models were assessed for relationships between and within variable importance drift, performance drift and actual dataset drift using temporal and non-temporal invariant consensus scoring, combining geometric average results of all metrics as the Clinical Effective Metric (CEM).ResultsA total of 227,087 adults underwent cardiac surgery during the study period with a mortality rate of 2.76%. There was a strong evidence of decrease in overall performance across all models (p < 0.0001). Xgboost (CEM 0.728 95CI: 0.728-0.729) and Random Forest (CEM 0.727 95CI 0.727-0.728) were the best overall performing models both temporally and non-temporally. ES II perfomed worst across all comparisons. Sharp changes in variable importance and dataset drift between 2017-10 to 2017-12, 2018-06 to 2018-07 and 2018-12 to 2019-02 mirrored effects of performance decrease across models.ConclusionsCombining the metrics covering all four aspects of discrimination, calibration, clinical usefulness and overall accuracy into a single consensus metric improved the efficiency of cognitive decision-making. All models show a decrease in at least 3 of the 5 individual metrics. CEM and variable importance drift detection demonstrate the limitation of logistic regression methods used for cardiac surgery risk prediction and the effects of dataset drift. Future work will be required to determine the interplay between ML and whether ensemble models could take advantage of their respective performance advantages.Central messageML performance decreases over time due to dataset drift, but remains superior to ES II. Therefore regular assessment and modification of ML models may be preferable.Prospective messageA gap in understanding the effect of dataset drift on the performance of ML models over time presents a major barrier to their clinical application. Xgboost and Random Forest have shown superior performance both temporally and non-temporally against ES II. However, a decrease in model performance of all models due to dataset drift suggests the need for regular drift monitoring.

Objectives: Ventricular septal defect (VSD) is becoming a progressively less frequent mechanical complication of myocardial infarction (MI). However, this event is still associated with high operative mortality. We aimed to describe the trends and the risk factors associated with surgical VSD repair outcomes and to provide a clinical benchmark for percutaneous VSD closure strategies.Methods: Using the UK National Adult Cardiac Surgery Audit database, we identified 1010 patients undergoing surgical VSD repair from 1996 to 2018. The primary outcome was operative mortality. Mixed-model, multivariable logistic regression was used to identify the risk factors associated with operative mortality taking into account the variation related to the centre, the surgeon and the year of the operation.Results: Both the number of surgical VSD repair and the mortality rate did not change significantly over the 23-year timeframe. Operative mortality was 38.9% overall and was higher when patients were operated within the first 6 hours (75%) or the first 24 hours (61.3%) from the index MI. Risk factors associated with higher odds of mortality were early surgery, older age, cardiogenic shock, renal failure, previous percutaneous coronary intervention and urgent/emergent operations. Moreover, the mortality rate was similar among patients undergoing isolated VSD repair and VSD repaired combined with surgical coronary revascularization alone or with concomitant mitral valve procedures.Conclusions: Post MI VSD remains a dreaded mechanical complication characterized by high surgical operative mortality. A delayed operation, whenever possible, appears to be the most beneficial strategy to reduce mortality.

Background Despite advances in cardiac surgery, observational studies suggest that females have poorer post-operative outcomes than males. This study is the first to review sex related outcomes following both coronary artery bypass graft (CABG) and valve surgery with or without combined CABG. Methods We identified 30 primary research articles reporting either short-term mortality (in-hospital/30 day), long-term mortality, and post-operative stroke, sternal wound infection and myocardial infarction (MI) in both sexes following CABG and valve surgery with or without combined CABG. Reported adjusted odds/hazard ratio were pooled using an inverse variance model. Results Females undergoing CABG and combined valve and CABG surgery were at higher risk of short-term mortality (odds ratio (OR) 1.40; 95% confidence interval (CI) 1.32–1.49; I2 = 79%) and post-operative stroke (OR 1.2; CI 1.07–1.34; I2 = 90%) when compared to males. However, for isolated AVR, there was no difference found (OR 1.19; 95% CI 0.74–1.89). There was no increased risk in long-term mortality (OR 1.04; 95% CI: 0.93–1.16; I2 = 82%), post-operative MI (OR 1.22; 95%CI: 0.89–1.67; I2 = 60%) or deep sternal wound infection (OR 0.92; 95%CI: 0.65–1.03, I2 = 87%). No evidence of publication bias or small study effect was found. Conclusion Females are at a greater risk of short-term mortality and post-operative stroke than males following CABG and valve surgery combined with CABG. However, there is no difference for Isolated AVR. Long-term mortality is equivalent in both sexes. PROSPERO Registration: CRD42021244603.

Background Data on the postoperative outcomes for patients with infective endocarditis complicated by an aortic root abscess is sparse due to the condition’s low incidence and high mortality rates. This systematic review and meta-analysis aims to evaluate existing data on the impact of aortic root abscesses on the postoperative outcomes and to inform optimal surgical approach. Methods The online databases MEDLINE, EMBASE and Cochrane library were searched from 1990 to 2022 for studies comparing cohorts of surgically managed infective endocarditis patients with and without an aortic root abscess. Data was extracted by two independent investigators and aggregated in a random-effects model. Risk of bias was assessed using an adapted version of the Newcastle-Ottawa scale. Results Six clinical studies were included in the meta-analysis ( n 1982). The abscess group was associated with increased in-hospital mortality (OR 1.74 95%: CI 1.18–2.56) and late mortality (HR 1.27 95% CI:1.03–1.58). The reoperation meta-analysis was complicated by high rates of heterogeneity (I2 = 59%) and found no significant differences in reoperation between abscess and no abscess groups (HR=1.48: 95% CI:0.92–2.40). Post-hoc scatter graph showed a strong linear relationship (r 0.998), suggesting hospitals with higher rates of aortic root replacement achieve lower rates of reoperation for aortic root abscess patients compared with patch reconstruction. Conclusions The presence of an aortic root abscess in aortic valve endocarditis is associated with elevated early and late mortality despite modern standards of care. Additionally, aortic root replacement should be considered to have a favourable postoperative profile for use in this context.

Risk stratification plays a major role in the clinical decision-making process, patient consent and clinical governance analysis. However, the calibration of current risk scores (e.g., European System for Cardiac Operative Risk Evaluation (EuroSCORE), The Society of Thoracic Surgeons (STS) risk score) has been shown to deteriorate over time – a process known as calibration drift. The introduction of new clinical scores with different variable sets typically result in disparate datasets due to different levels of missingness. This is a barrier to the full insight and predictive capability of datasets across all potentially available time ranges. Little is known about the use of ensemble learning with ensemble metrics to mitigate the effects of calibration drift and changing risk across siloed datasets and time. In this study, we evaluated the effect of various combinations of Machine Learning (ML) models in improving model performance. The National Adult Cardiac Surgery Audit dataset was used (January 1996 to March 2019, 647,726 patients). We trained six different base learner models including Logistic Regression, Neuronetwork, Random Forest (RF), Weighted Support Vector Machine, Xgboost and Bayesian Update, based on two different variable sets of either Logistic EuroScore (LogES) or EuroScore II (ES II), partitioned by the time of score adoption (1996–2011 and 2012–2019). These base learner models are ensembled using nine different combinations to produce homogeneous or heterogeneous ensembles. Discrimination, calibration, clinical effectiveness and overall accuracy were assessed using an ensemble metric, referred to as clinical effectiveness metric (CEM). Xgboost homogenous ensemble (HE) was the highest performing model (CEM 0.725) with AUC (0.8327; 95% Confidence Interval (CI) 0.8323–0.8329) followed by Random Forest HE (CEM 0.723; AUC 0.8325; 95%CI 0.8320–0.8326). Across different heterogenous ensembles, significantly better performance was obtained by combining siloed datasets across time (CEM 0.720) than building ensembles of either 1996–2011 (t-test adjusted, p = 1.67e-6) or 2012–2019 (t-test adjusted, p = 1.35e-193) datasets alone. Both homogenous and heterogenous ML ensembles performed significantly better than traditional recalibration method (Bayesian Update). Combining the metrics covering all four aspects of discrimination, calibration, clinical usefulness and overall accuracy into a single ensemble metric improved the efficiency of cognitive decision-making. Xgboost/Random Forest homogenous ensembling and a highly heterogeneous ensemble approach showed high performance across multifaceted aspects of ML performance and were superior to traditional recalibration methods. Time-dependent ensemble combination of variables, having differing qualities according to time of score adoption, enabled previously siloed data to be combined, leading to increased power, clinical interpretability of variables and usage of data. For models to guide clinicians in individual decisions, performance exceeding these multifaceted benchmarks is necessary.

Perioperative atrial fibrillation (AF) is associated with increased mortality, morbidity, and excess healthcare costs. The objective of our study was to assess if preoperative AF in patients undergoing coronary artery bypass grafting is a predictor of operative mortality, postoperative stroke, and need for postoperative dialysis by interrogating a large registry database. We included all isolated procedures performed between February 1996 and March 2019. We used a generalized linear mixed model to assess the effect of preoperative AF on mortality stroke and the need for postoperative dialysis after adjusting for the relevant confounders derived from EuroSCORE 2. Confounders considered included age, gender, neurological dysfunction, renal dysfunction, recent myocardial infarction, pulmonary disease, unstable angina, NYHA class, pulmonary hypertension, diabetes on insulin and peripheral vascular disease, and urgency of the operation. We treated the hospital and operating consultant as random effect variables. We also performed LV function subgroup analyses to assess the effect of preoperative AF on the outcomes of interest. The incidence of pre-existent AF in the cohort of patients we analyzed (N = 356,040 patients) was 3.5% (N = 12,664). In the unadjusted baseline characteristics, preoperative AF patients had more associated comorbidities. After adjustment, preoperative AF remained a significant predictor of increased mortality (odds ratio [OR]: 1.63, confidence interval [CI] 1.48-1.79, p0.001), stroke (OR: 1.33, CI 1.16-1.54, p = 0.001), and need for renal dialysis (OR:1.61, CI 1.46-1.78, p0.001). Preoperative AF was a significant predictor of adverse outcomes in patients with moderate and good LV function but not in patients with poor LV function (EF30%). Our study suggests that preoperative AF is associated with an increased risk for perioperative mortality and stroke in patients undergoing coronary artery bypass grafting.

Background Data on the postoperative outcomes for patients with infective endocarditis complicated by an aortic root abscess is sparse due to the condition’s low incidence and high mortality rates. This systematic review and meta-analysis aims to evaluate existing data on the impact of aortic root abscesses (ARA) on the postoperative outcomes of surgically managed infective endocarditis (IE) and to inform optimal surgical approach. Methods The online databases MEDLINE, EMBASE and Cochrane library were searched from 1990 to 2022 for studies comparing ARA with NARA (no ARA) in infective endocarditis. Data was extracted by two independent investigators and aggregated in a random-effects model (Review Manager version 5.3). Risk of bias was assessed using an adapted version of the Newcastle-Ottawa scale. Results Six clinical studies were included in the meta-analysis (n = 1982). The ARA group was associated with an increased risk of in-hospital mortality (OR = 1.74 96% CI 1.18-2.56) and late mortality (HR = 1.27 95% CI 1.03-1.58). The reoperation meta-analysis was complicated by high rates of heterogeneity (I = 59%) and found no significant differences in reoperation between group ARA and NARA (no ARA) (HR = 1.48; 95% CI 0.92-2.40). Post-hoc scatter graph showed a strong linear relationship (r=0.998), suggesting hospitals with higher rates of aortic root replacement (ARR) achieve lower rates of reoperation for ARA patients compared with PR. Conclusions The presence of an ARA in aortic valve endocarditis is associated with elevated early and late mortality despite modern standards of care. Additionally, ARR should be considered to have a favourable postoperative profile for use in this context.

Co-amorphization of a single drug with amino acid is a technique to improve aerosolization of inhalable spray-dried formulation for inhalation therapy. However, the incorporation of a second drug molecule into drug-amino acid co-amorphous particles to prepare combination formulations has not been explored. Here, we prepared combination powders using two model drugs, ceftazidime and roflumilast, which when concurrently used can potentially improve therapeutic outcome in non-cystic fibrosis bronchiectasis by counteracting both infection and inflammation. The study was performed using a two-step approach. The first step involved the identification of an amino acid and its concentration (%

Background Following the repair of TOF patients may be left with pulmonary regurgitation and a dilated right ventricle (RV), which in turn can lead to ventricular arrhythmias and sudden death. A prolonged QRS is a predictor of ventricular arrhythmias. However, whether subsequent pulmonary valve replacement (PVR) can reverse QRS‐prolongation is controversial. We hypothesized that changes in QRS duration following PVR are determined by preoperative QRS‐duration and RV volumes Methods A retrospective single‐center cohort study was conducted on 142 post‐TOF repair patients (mean age 25 ± 13 years) who underwent PVR between 1995 and 2019. Information on QRS duration and RV volumes measured by cardiac MRI (available in 83 patients) were collected. A linear mixed model was used to investigate the association between the preoperative QRS duration and RV volumes and the postoperative QRS duration. Results The QRS‐duration following PVR continued to increase in all subjects with a prolonged preoperative QRS‐duration(>160 ms, rate of increase of 0.87 msec ± 0.33 per year; p = .01), markedly raised RV end‐diastolic volume (RVEDV; ≥166 ml/m2, rate of increase of 2.0 msec ± 0.37 per year; p

OBJECTIVES We sought to provide further evidence on the safety and efficacy of aortic valve neocuspidization (AVNeo) using autologous pericardium in adult patients with aortic valve disease by reporting clinical and echocardiographic results from the first UK experience and performing a meta-analytic comparison with other biological valve substitutes. METHODS We reported clinical and echocardiographic outcomes of 55 patients (mean age 58 ± 15 years) undergoing AVNeo with autologous pericardium in 2 UK centres from 2018 to 2020. These results were included in a meta-analytic comparison between series on AVNeo (7 studies, 1205 patients, mean weighted follow-up 3.6 years) versus Trifecta (10 studies, 8705 patients, 3.8 years), Magna Ease (3 studies, 3137 patients, 4.1 years), Freedom Solo (4 studies, 1869 patients, 4.4 years), Freestyle (4 studies, 4307 patients, 7 years), Mitroflow (4 studies, 4760 patients, 4.1 years) and autograft aortic valve (7 papers, 3839 patients, 9.1 years). RESULTS In the present series no patients required intraoperative conversion. After mean follow-up of 12.5 ± 0.9 months, 3 patients presented with endocarditis and 1 required reintervention. The remaining patients had absent or mild aortic valve insufficiency with very low peak and mean transvalvular gradients (16 ± 3.7 and 9 ± 2.2 mmHg, respectively). Meta-analytic estimates showed non-significant difference between AVNeo and all but Magna Ease valves with regards to structural valve degeneration, reintervention and endocarditis. When compared Magna Ease valve, AVNeo and other valve substitutes showed an excess of valve-related events. CONCLUSIONS AVNeo is safe, associated with excellent haemodynamic profile. Its midterm risk of valve-related events is comparable to most biological valve substitutes. Magna Ease is potentially the best biological choice as far as risk of reintervention is concerned.

Surface-enhanced infrared absorption spectroscopy (SEIRS) using a dry-state method was established for the ultra-sensitive detection of glutathione using citrate functionalized silver nanoparticles (AgNPs) in human biological blood and serum samples. The signal intensity of glutathione was enhanced due to the hot-spots created by the aggregation of AgNPs causing the effective absorption of electromagnetic radiation in the IR region for molecular vibrations. The study revealed that the possibility of the interaction of silver with glutathione might rule out structural orientation, energy and affinity as the reason for the detection. The selective determination of glutathione was experimentally confirmed by performing controlled testing with other amino acids. The absorption peak at 3257.84 cm−1 (NH3+) was used for the quantitative analysis of glutathione when AgNPs were used as a chemical sensor in dry state SEIRS. A pre-treatment process was employed for human biological samples such as blood and serum in order to determine glutathione using AgNPs coupled with dry-state SEIRS analysis. A linear range was obtained for the quantitative determination of glutathione in human biological samples from 10 to 100 μg mL−1 with a correlation coefficient (R2) of 0.993. The limit of detection was 1.74 μg mL−1 and the limit of quantification was 5.30 μg mL−1. The advantages of using the presented dry-state SEIRS are its simplicity, selectivity and sensitivity towards the analysis of glutathione using AgNPs in human biological blood and serum samples.

OBJECTIVES Continuous improvement in the management of patients undergoing surgical aortic valve replacement (SAVR) may have considerably enhanced surgical outcomes including in-hospital mortality and perioperative complications. We aimed to analyse in-hospital mortality and morbidity trends in patients undergoing SAVR in a single centre to provide insights for future benchmarking for transcatheter aortic valve implantation indications. METHODS This was a retrospective study of prospectively collected data from patients undergoing either isolated SAVR or combined with coronary artery bypass grafting (CABG) at the Bristol Heart Institute, UK, from January 2000 to December 2017. Baseline characteristics were extracted and analysed across 3 different eras (2000–2005, 2006–2011 and 2012–2017). Risk-adjusted time trend was obtained from univariate and multivariate logistic regression including all baseline characteristics. RESULTS A total of 2719 patients (63.2%) underwent isolated SAVR, and 1581 (36.8%) underwent combined CABG and SAVR during the study period. For patients undergoing SAVR, in-hospital mortality decreased from 2.9% in 2000–2005 to 0.7% in 2012–2017 (risk-adjusted time trend 0.0001). Hospital mortality in patients aged 75–79 and ≥80 years decreased from 5.6% and 5.3% to 0.4% and 2.2%, respectively. Mortality after combined SAVR and CABG did not significantly decrease (from 3.9% in 2000–2005 to 3.5% in 2012–2017; risk-adjusted time trend = 0.62). However, in patients aged ≥80 years, index hospitalization mortality showed a decreasing non-significant trend from 9.8% to 4.8%. CONCLUSIONS Our findings support the hypothesis that mortality and morbidity rates following SAVR have significantly improved over the years, including for patients at high risk.

OBJECTIVES Interest in the clinical usefulness of machine learning for risk prediction has bloomed recently. Cardiac surgery patients are at high risk of complications and therefore presurgical risk assessment is of crucial relevance. We aimed to compare the performance of machine learning algorithms over traditional logistic regression (LR) model to predict in-hospital mortality following cardiac surgery. METHODS A single-centre data set of prospectively collected information from patients undergoing adult cardiac surgery from 1996 to 2017 was split into 70% training set and 30% testing set. Prediction models were developed using neural network, random forest, naive Bayes and retrained LR based on features included in the EuroSCORE. Discrimination was assessed using area under the receiver operating characteristic curve, and calibration analysis was undertaken using the calibration belt method. Model calibration drift was assessed by comparing Goodness of fit χ2 statistics observed in 2 equal bins from the testing sample ordered by procedure date. RESULTS A total of 28 761 cardiac procedures were performed during the study period. The in-hospital mortality rate was 2.7%. Retrained LR [area under the receiver operating characteristic curve 0.80; 95% confidence interval (CI) 0.77–0.83] and random forest model (0.80; 95% CI 0.76–0.83) showed the best discrimination. All models showed significant miscalibration. Retrained LR proved to have the weakest calibration drift. CONCLUSIONS Our findings do not support the hypothesis that machine learning methods provide advantage over LR model in predicting operative mortality after cardiac surgery.

Ion exchange resins (IERs) are solid poly-electrolytes which have both sorption and exchange capacity of several organic compounds. They have the power to separate ionic and non-ionic substances with the surrounding medium. The drug materials or substances are adsorbed on resin, which is commonly known as resinate, these features of IERs have useful applications in pharmaceutical formation (i.e., taste masking, stability and solubility enhancement, etc.) and major applications in drug delivery (i.e., oral, nasal, ophthalmic, transdarmal drug delivery). IE principles have been exploited in the investigation of numerous drug industry problems for many years. Synthetic IERs have been extensively employed in pharmacy and medicine, especially for taste masking or controlled release of drugs and have been expansively studied in the development of novel drug delivery systems and other biomedical applications. In this review, the fascinating IERs involving ion exchange processes in pharmaceutical and clinical applications and also their recent advanced uses have been discussed.

ObjectivesTo prevent the emergence of new waves of COVID-19 caseload and associated mortalities, it is imperative to understand better the efficacy of various control measures on the national and local development of this pandemic in space–time, characterise hotspot regions of high risk, quantify the impact of under-reported measures such as international travel and project the likely effect of control measures in the coming weeks.MethodsWe applied a deep recurrent reinforced learning based model to evaluate and predict the spatiotemporal effect of a combination of control measures on COVID-19 cases and mortality at the local authority (LA) and national scale in England, using data from week 5 to 46 of 2020, including an expert curated control measure matrix, official statistics/government data and a secure web dashboard to vary magnitude of control measures.ResultsModel predictions of the number of cases and mortality of COVID-19 in the upcoming 5 weeks closely matched the actual values (cases: root mean squared error (RMSE): 700.88, mean absolute error (MAE): 453.05, mean absolute percentage error (MAPE): 0.46, correlation coefficient 0.42; mortality: RMSE 14.91, MAE 10.05, MAPE 0.39, correlation coefficient 0.68). Local lockdown with social distancing (LD_SD) (overall rank 3) was found to be ineffective in preventing outbreak rebound following lockdown easing compared with national lockdown (overall rank 2), based on prediction using simulated control measures. The ranking of the effectiveness of adjunctive measures for LD_SD were found to be consistent across hotspot and non-hotspot regions. Adjunctive measures found to be most effective were international travel and quarantine restrictions.ConclusionsThis study highlights the importance of using adjunctive measures in addition to LD_SD following lockdown easing and suggests the potential importance of controlling international travel and applying travel quarantines. Further work is required to assess the effect of variant strains and vaccination measures.

Pseudomonas aeruginosa infection is common in cystic fibrosis as well as non-cystic fibrosis bronchiectasis. The pathogen presents challenges for treatment due to its adaptive antibiotic-resistance, mainly pertaining to its biofilm-forming ability, as well as limitations associated with conventional drug delivery in achieving desired therapeutic concentration in the infection site. Hence, therapeutic approach has shifted towards the inhalation of antibiotics. Ceftazidime is a potent antibiotic against the pathogen; however, it is currently only available as a parenteral formulation. Here, spray dryer was employed to generate inhalable high dose ceftazidime microparticles. In addition, the use of amino acids (valine, leucine, methionine, phenylalanine, and tryptophan) to improve aerosolization as well as chemical stability of amorphous ceftazidime was explored. The particles were characterized using X-ray diffraction, infrared (IR) spectroscopy, calorimetry, electron microscopy, particle size analyzer, and next generation impactor. The chemical stability at 25 °C/15% was assessed using chromatography. All co-spray dried formulations were confirmed as monophasic amorphous systems using calorimetry. In addition, principal component analysis of the IR spectra suggested potential interaction between tryptophan and ceftazidime in the co-amorphous matrix. Inclusion of amino acids improved aerosolization and chemical stability in all cases. Increase in surface asperity was clear with the use of amino acids which likely contributed to the improved aerosol performance, and potential interaction between amino acids and ceftazidime was plausibly the reason for improved chemical stability. Leucine offered the best aerosolization enhancement with a fine particle fraction of 78% and tryptophan showed stabilizing superiority by reducing chemical degradation by 51% over 10 weeks in 1:1 M ratio. The protection against ceftazidime degradation varied with the nature of amino acids. Additionally, there was a linear relationship between degradation protection and molar mass of amino acids or percentage weight of amino acids in the formulations. None of the amino acids were successful in completely inhibiting degradation of ceftazidime in amorphous spray-dried powder to prepare a commercially viable product with desired shelf-life. All the amino acids and ceftazidime were non-toxic to A549 alveolar cell line.

Aims Operability of type A acute aortic dissections (TAAAD) is currently based on non-standardized decision-making process, and it lacks a disease-specific risk evaluation model that can predict mortality. We investigated patient, intraoperative data, surgeon, and centre-related variables for patients who underwent TAAAD in the UK. Methods and results We identified 4203 patients undergoing TAAAD surgery in the UK (2009–18), who were enrolled into the UK National Adult Cardiac Surgical Audit dataset. The primary outcome was operative mortality. A multivariable logistic regression analysis was performed with fast backward elimination of variables and the bootstrap-based optimism-correction was adopted to assess model performance. Variation related to hospital or surgeon effects were quantified by a generalized mixed linear model and risk-adjusted funnel plots by displaying the individual standardized mortality ratio against expected deaths. Final variables retained in the model were: age [odds ratio (OR) 1.02, 95% confidence interval (CI) 1.02–1.03; P, Graphical Abstract

Visual interpretation of hand gestures is a natural method of achieving Human-Computer Interaction (HCI). In this paper, we present an approach to setting up of a smart home where the appliances can be controlled by an implementation of a Hand Gesture Recognition System. More specifically, this recognition system uses Transfer learning, which is a technique of Machine Learning, to successfully distinguish between pre-trained gestures and identify them properly to control the appliances. The gestures are sequentially identified as commands which are used to actuate the appliances. The proof of concept is demonstrated by controlling a set of LEDs that represent the appliances, which are connected to an Arduino Uno Microcontroller, which in turn is connected to the personal computer where the actual gesture recognition is implemented.

OBJECTIVES Patients presenting with more comorbidities, requiring more complex cardiac surgical procedures and an increase in public scrutiny are impacting on training programme because of the perceived risk of worse outcomes. Hence, we aimed to provide evidence that trainees as the first operator can achieve comparable results to consultants when performing isolated surgical aortic valve replacement. METHODS From 1996 to 2017, 2919 patients underwent surgical aortic valve replacement at the Bristol Heart Institute, operated on by either a consultant (n = 2220) or a trainee (n = 870) as the first operator. Propensity score matching was used to adjust for imbalance in the baseline characteristics of the 2 groups. RESULTS Over a 21-year period, the proportion of trainee cases dropped from 41.5% to 25.9%. No differences in the rates and risk of in-hospital mortality, new cerebrovascular accidents, re-exploration for bleeding, deep sternal wound infection and length of stay were found between patients operated on in the 2 groups. Also, there was a comparable risk of late death between the 2 groups (HR 0.88; 95% CI 0.73–1.06; P = 0.27) and this was present regardless of trainees career level and patients surgical risk based on the EuroSCORE. Finally, we showed an increase in patients risk profile in the latest year but, this was not associated with the worst outcomes when trainees performed the operation. CONCLUSIONS Surgical aortic valve replacement is a safe and reproducible technique and regardless of the patient’s risk profile, and no differences in the outcomes between trainees and consultant cases were found., Over the few past years, research in surgical education has bloomed and attention has been drawn to the quality and quantity of surgical training.

Introduction: Female sex is considered an independent risk factor for mortality and morbidity following cardiac surgery. This study is the first to review the UK national database to compare outcomes following surgical revascularisation and valvular cardiac procedures between females and males. Hypothesis: Females have worse outcomes following cardiac surgery than males. Methods: Using data from National Adult Cardiac Surgery Audit (NACSA), we identified all elective and urgent, isolated coronary artery by-pass grafting (CABG), aortic valve replacement (AVR) and mitral valve replacement/repair (MVR) procedures from 2010-2018. We compared baseline data, operative data and outcomes of mortality, stroke, renal failure, deep sternal wound infection, return to theatre for bleeding and length of hospital stay. Multivariable logistical/linear regression models were used to assess relationships between sex and our outcomes, adjusting for baseline characteristics. Results: Females, compared to males, had a greater odds of experiencing 30-day mortality in each procedure (CABG OR 1.76, CI 1.47-2.09, p Conclusions: Despite advances in cardiac surgery, females in the UK have an increased risk of short-term mortality after cardiac surgery compared to males. This highlights the need to focus on the understanding of the causes behind these disparities and therefore implementation of strategies to improve this.

Introduction: Despite advances in cardiac surgery, there is evidence from observational studies which suggests that females have poorer post-operative outcomes than males. This study is the first to review sex related outcomes following both coronary artery bypass graft (CABG) and valve surgery with or without combined CABG. Hypothesis: Females have worse outcomes following cardiac surgery than males. Methods: We identified 30 primary research articles reporting either short-term mortality (in-hospital/30 day), long-term mortality, post-operative stroke, sternal wound infection and post-operative myocardial infarction (MI) in both sexes following CABG and valve surgery with or without combined CABG Reported adjusted odds/hazard ratio were pooled using an inverse variance model. Results: Females were at higher risk of short-term mortality (odds ratio (OR) 1.40; 95% confidence interval (CI) 1.32-1.49; I 2 =79%) and post-operative stroke (OR 1.2; CI 1.07-1.34; I 2 =90%) when compared to males. There was no increased risk in long-term mortality (OR 1.04; 95% CI: 0.93-1.16; I 2 =82%), post-operative MI (OR 1.22; 95%CI: 0.89-1.67; I 2 =60%) or deep sternal wound infection (OR 0.92; 95%CI: 0.65-1.03, I 2 =87%). No evidence of publication bias or small study effect was found. Conclusions: Females are at a greater risk of short-term mortality and post-operative stroke than males following CABG and valve surgery with or without combined CABG. However long-term mortality is equivalent in both sexes.

OBJECTIVES Benefits of using multiple arterial grafting (MAG), over single arterial grafting in major adverse cardiac event rates and the need for repeat revascularization, have been widely reported. Several guidelines have recommended the use of MAG in selected patients. We report the trend of MAG in patients undergoing isolated coronary artery bypass grafting (CABG) in the UK. METHODS This is a retrospective analysis of a prospectively collected UK national database in patients undergoing non-emergency, isolated CABG from 1996 to 2018. Patients were divided into single arterial grafting and MAG, and trends in perioperative characteristics were analysed. RESULTS A total of 336 321 patients were included, of whom 284 003 (84.44%) received single arterial grafting and 52 318 (15.56%) received MAG. The use of MAG after an initial increase from 1996 to 2001, steadily decreased thereafter, particularly in the use of radial artery. MAG was likely to be performed in younger patients [66.72 (standard deviation: 9.22) vs 62.30 (standard deviation: 10.06), P CONCLUSIONS Our data show that the use of MAG in CABG in the UK after an initial increase from 1996 to 2001 steadily decreased thereafter until 2018. This is likely to be multifactorial and a better understanding of the main causes may contribute to establishing the best indication for MAG in everyday clinical practice.

Background Updatable estimates of COVID-19 onset, progression, and trajectories underpin pandemic mitigation efforts. To identify and characterise disease trajectories, we aimed to define and validate ten COVID-19 phenotypes from nationwide linked electronic health records (EHR) using an extensible framework. Methods In this cohort study, we used eight linked National Health Service (NHS) datasets for people in England alive on Jan 23, 2020. Data on COVID-19 testing, vaccination, primary and secondary care records, and death registrations were collected until Nov 30, 2021. We defined ten COVID-19 phenotypes reflecting clinically relevant stages of disease severity and encompassing five categories: positive SARS-CoV-2 test, primary care diagnosis, hospital admission, ventilation modality (four phenotypes), and death (three phenotypes). We constructed patient trajectories illustrating transition frequency and duration between phenotypes. Analyses were stratified by pandemic waves and vaccination status. Findings Among 57 032 174 individuals included in the cohort, 13 990 423 COVID-19 events were identified in 7 244 925 individuals, equating to an infection rate of 12·7% during the study period. Of 7 244 925 individuals, 460 737 (6·4%) were admitted to hospital and 158 020 (2·2%) died. Of 460 737 individuals who were admitted to hospital, 48 847 (10·6%) were admitted to the intensive care unit (ICU), 69 090 (15·0%) received non-invasive ventilation, and 25 928 (5·6%) received invasive ventilation. Among 384 135 patients who were admitted to hospital but did not require ventilation, mortality was higher in wave 1 (23 485 [30·4%] of 77 202 patients) than wave 2 (44 220 [23·1%] of 191 528 patients), but remained unchanged for patients admitted to the ICU. Mortality was highest among patients who received ventilatory support outside of the ICU in wave 1 (2569 [50·7%] of 5063 patients). 15 486 (9·8%) of 158 020 COVID-19-related deaths occurred within 28 days of the first COVID-19 event without a COVID-19 diagnoses on the death certificate. 10 884 (6·9%) of 158 020 deaths were identified exclusively from mortality data with no previous COVID-19 phenotype recorded. We observed longer patient trajectories in wave 2 than wave 1. Interpretation Our analyses illustrate the wide spectrum of disease trajectories as shown by differences in incidence, survival, and clinical pathways. We have provided a modular analytical framework that can be used to monitor the impact of the pandemic and generate evidence of clinical and policy relevance using multiple EHR sources. Funding British Heart Foundation Data Science Centre, led by Health Data Research UK.

OBJECTIVES Several studies have shown worse outcomes in patients operated on later in the week. We tested this hypothesis in a large UK national audit database in elective patients undergoing adult cardiac surgery. METHODS We used a generalized additive model to evaluate the effect of the day of the week on the following postoperative outcomes: 30-day mortality, stroke, need for dialysis and return to theatre for bleeding. We have adjusted for the relevant European System for Cardiac Operative Risk Evaluation (EuroSCORE) II covariates, plus responsible consultant, hospital and year of operation and performed subgroup analysis for isolated coronary artery bypass grafting (CABG) procedures. RESULTS Out of 371 500 patients, 60 555 (16.3%) underwent AVR, 36 553 (9.8%) AVR plus CABG, 238 812 (64.3%) isolated CABG, 26 517 (7.1%) isolated mitral valve repair or replacement and 9063 (2.4%) mitral valve plus CABG. A total of 13 997 (3%) had surgery over the weekend. After covariate adjustment, we found no effect of day of surgery on mortality (P = 0.081), stroke (P = 0.137) and need for postop dialysis (P = 0.732). However, across all operations, there was evidence of a lower rate of return to theatre for bleeding/tamponade at the weekend (P = 0.039). In subgroup analysis of isolated CABG, the day of the week did not affect any outcomes. CONCLUSIONS We found no effect of the day of the week on risk-adjusted short-term mortality, stroke, and the requirement for postoperative dialysis after elective cardiac surgery. Overall, the patients operated on during the weekdays were less likely to return to theatre for bleeding. In isolated CABG, the day of the week did not affect any outcomes.

Background The most used mortality risk prediction models in cardiac surgery are the European System for Cardiac Operative Risk Evaluation(EuroSCORE)(ES) and Society of Thoracic Surgeons(STS) score. There is no agreement on which score should be considered more accurate nor which score should be utilised in each population sub-group. We sought to provide a thorough quantitative assessment of these 2 models. Method We performed a systematic literature review and captured information on discrimination, as quantified by the area under the receiver operator curve(AUC), and calibration, as quantified by the ratio of observed-to-expected mortality(O:E). We performed random effects meta-analysis of the performance of the individual models as well as pairwise comparisons and sub-group analysis by procedure type, time and continent. Results The ES2(AUC 0.783[95%CI 0.765-0.800];O:E 1.102[95%CI 0.943-1.289]) and STS(AUC 0.757[95%CI 0.727-0.785];O:E 1.111[95%CI 0.853-1.447]) both showed good overall discrimination and calibration. There was no significant difference in the discrimination of the two models(Difference in AUC -0.016; 95%CI -0.034 to -0.002;p0.09). However, the calibration of ES2 showed significant geographical variations(p Conclusions ES2 and STS are both reliable predictors of short-term mortality following adult cardiac surgery in the populations from which they were derived. STS may have broader applications when comparing outcomes across continents and time periods as compared to ES2.