Deep Learning Based Long Term Mortality Prediction in the National Lung Screening Trial

In this study, the long-term mortality in the National Lung Screening Trial (NLST) was investigated using a deep learning-based method. Binary classification of the non-lung-cancer mortality (i.e. cardiovascular and respiratory mortality) was performed using neural network models centered around a 3D-ResNet. The models were trained on a participant age, gender, and smoking history matched cohort. Utilising both the 3D CT scan and clinical information, the models can achieve an AUC of 0.73 which outperforms humans at cardiovascular mortality prediction. By interpreting the trained models with 3D saliency maps, we examined the features on the CT scans that correspond to the mortality signal. The saliency maps can potentially assist the clinicians' and radiologists' to identify regions of concern on the image that may indicate the need to adopt preventative healthcare management strategies to prolong the patients' life expectancy.

Delirium and subsyndromal delirium are associated with the long-term risk of death after ischaemic stroke

Background Post-stroke delirium has a negative impact on functional outcome. We explored if there is any association between delirium, subsyndromal delirium and long-term mortality after ischaemic stroke and transient ischaemic attack. Methods We included 564 patients with ischaemic stroke or transient ischaemic attack. We assessed symptoms of delirium during the first 7 days after admission. We used Cox proportional hazards models to analyse all-cause mortality during the first 5 years after stroke. Results We diagnosed delirium in 23.4% and subsyndromal delirium in 10.3% of patients. During the follow-up, 72.7% of patients with delirium, 51.7% of patients with subsyndromal delirium and 22.7% of patients without delirious symptoms died (P < 0.001). Patients with subsyndromal delirium and delirium had higher risk of death in the multivariate analysis (HR 1.72, 95% CI 1.11–2.68, P = 0.016 and HR 3.30, 95% CI 2.29–4.76, P < 0.001, respectively). Conclusions Post-stroke delirium is associated with long-term mortality. Patients with subsyndromal delirium are at the intermediate risk of death.

Association Between Preoperative Monocyte to High-Density Lipoprotein Ratio on In-hospital and Long-Term Mortality in Patients Undergoing Endovascular Repair for Acute Type B Aortic Dissection

Aims: The monocyte to high-density lipoprotein ratio (MHR), a novel marker of inflammation and cardiovascular events, has recently been found to facilitate the diagnosis of acute aortic dissection. This study aimed to assess the association of preoperative MHR with in-hospital and long-term mortality after thoracic endovascular aortic repair (TEVAR) for acute type B aortic dissection (TBAD).Methods: We retrospectively evaluated 637 patients with acute TBAD who underwent TEVAR from a prospectively maintained database. Multivariable logistic and cox regression analyses were conducted to assess the relationship between preoperative MHR and in-hospital as well as long-term mortality. For clinical use, MHR was modeled as a continuous variable and a categorical variable with the optimal cutoff evaluated by receiver operator characteristic curve for long-term mortality. Propensity score matching was used to diminish baseline differences and subgroups analyses were conducted to assess the robustness of the results.Results: Twenty-one (3.3%) patients died during hospitalization and 52 deaths (8.4%) were documented after a median follow-up of 48.1 months. The optimal cutoff value was 1.13 selected according to the receiver operator characteristic curve (sensitivity 78.8%; specificity 58.9%). Multivariate analyses showed that MHR was independently associated with either in-hospital death [odds ratio (OR) 2.11, 95% confidence interval (CI) 1.16-3.85, P = 0.015] or long-term mortality [hazard ratio (HR) 1.78, 95% CI 1.31-2.41, P &lt; 0.001). As a categorical variable, MHR &gt; 1.13 remained an independent predictor of in-hospital death (OR 4.53, 95% CI 1.44-14.30, P = 0.010) and long-term mortality (HR 4.16, 95% CI 2.13-8.10, P &lt; 0.001). Propensity score analyses demonstrated similar results for both in-hospital death and long-term mortality. The association was further confirmed by subgroup analyses.Conclusions: MHR might be useful for identifying patients at high risk of in-hospital and long-term mortality, which could be integrated into risk stratification strategies for acute TBAD patients undergoing TEVAR.

Unstructured clinical notes within the 24 hours since admission predict short, mid & long-term mortality in adult ICU patients

Mortality prediction for intensive care unit (ICU) patients is crucial for improving outcomes and efficient utilization of resources. Accessibility of electronic health records (EHR) has enabled data-driven predictive modeling using machine learning. However, very few studies rely solely on unstructured clinical notes from the EHR for mortality prediction. In this work, we propose a framework to predict short, mid, and long-term mortality in adult ICU patients using unstructured clinical notes from the MIMIC III database, natural language processing (NLP), and machine learning (ML) models. Depending on the statistical description of the patients’ length of stay, we define the short-term as 48-hour and 4-day period, the mid-term as 7-day and 10-day period, and the long-term as 15-day and 30-day period after admission. We found that by only using clinical notes within the 24 hours of admission, our framework can achieve a high area under the receiver operating characteristics (AU-ROC) score for short, mid and long-term mortality prediction tasks. The test AU-ROC scores are 0.87, 0.83, 0.83, 0.82, 0.82, and 0.82 for 48-hour, 4-day, 7-day, 10-day, 15-day, and 30-day period mortality prediction, respectively. We also provide a comparative study among three types of feature extraction techniques from NLP: frequency-based technique, fixed embedding-based technique, and dynamic embedding-based technique. Lastly, we provide an interpretation of the NLP-based predictive models using feature-importance scores.

Sarcopenia, Precardial Adipose Tissue and High Tumor Volume as Outcome Predictors in Surgically Treatedpleural Mesothelioma

Background: We evaluated the prognostic value of Sarcopenia, low precardial adipose-tissue (PAT), and high tumor-volume in the outcome of surgically-treated pleural mesothelioma (PM). Methods: From 2005 to 2020, consecutive surgically-treated PM-patients having a pre-operative computed tomography (CT) scan were retrospectively included. Sarcopenia was assessed by CT-based parameters measured at the level of the fifth thoracic vertebra (TH5) by excluding fatty-infiltration based on CT-attenuation. The findings were stratified for gender, and a threshold of the 33rd percentile was set to define sarcopenia. Additionally, tumor volume as well as PAT were measured. The findings were correlated with progression-free survival and long-term mortality. Results: Two-hundred-seventy-eight PM-patients (252 male; 70.2 ± 9 years) were included. The mean progression-free survival was 18.6 ± 12.2 months, and the mean survival time was 23.3 ± 24 months. Progression was associated with chronic obstructive pulmonary disease (COPD) (p = <0.001), tumor-stage (p = 0.001), and type of surgery (p = 0.026). Three-year mortality was associated with higher patient age (p = 0.005), presence of COPD (p < 0.001), higher tumor-stage (p = 0.015), and higher tumor-volume (p < 0.001). Kaplan-Meier statistics showed that sarcopenic patients have a higher three-year mortality (p = 0.002). While there was a negative correlation of progression-free survival and mortality with tumor volume (r = 0.281, p = 0.001 and r = −0.240, p < 0.001; respectively), a correlation with PAT could only be shown for epithelioid PM (p = 0.040). Conclusions: Sarcopenia as well as tumor volume are associated with long-term mortality in surgically treated PM-patients. Further, while there was a negative correlation of progression-free survival and mortality with tumor volume, a correlation with PAT could only be shown for epithelioid PM.

Long-term mortality rates and associated risk factors following primary and revision knee arthroplasty

Aims The aim of this study was to determine the long-term mortality rate, and to identify factors associated with this, following primary and revision knee arthroplasty (KA). Methods Data from the Scottish Arthroplasty Project (1998 to 2019) were retrospectively analyzed. Patient mortality data were linked from the National Records of Scotland. Analyses were performed separately for the primary and revised KA cohorts. The standardized mortality ratio (SMR) with 95% confidence intervals (CIs) was calculated for the population at risk. Multivariable Cox proportional hazards were used to identify predictors and estimate relative mortality risks. Results At a median 7.4 years (interquartile range (IQR) 4.0 to 11.6) follow-up, 27.8% of primary (n = 27,474/98,778) and 31.3% of revision (n = 2,611/8,343) KA patients had died. Both primary and revision cohorts had lower mortality rates than the general population (SMR 0.74 (95% CI 0.73 to 0.74); p < 0.001; SMR 0.83 (95% CI 0.80 to 0.86); p < 0.001, respectively), which persisted for 12 and eighteight years after surgery, respectively. Factors associated with increased risk of mortality after primary KA included male sex (hazard ratio (HR) 1.40 (95% CI 1.36 to 1.45)), increasing socioeconomic deprivation (HR 1.43 (95% CI 1.36 to 1.50)), inflammatory polyarthropathy (HR 1.79 (95% CI 1.68 to 1.90)), greater number of comorbidities (HR 1.59 (95% CI 1.51 to 1.68)), and periprosthetic joint infection (PJI) requiring revision (HR 1.92 (95% CI 1.57 to 2.36)) when adjusting for age. Similarly, male sex (HR 1.36 (95% CI 1.24 to 1.49)), increasing socioeconomic deprivation (HR 1.31 (95% CI 1.12 to 1.52)), inflammatory polyarthropathy (HR 1.24 (95% CI 1.12 to 1.37)), greater number of comorbidities (HR 1.64 (95% CI 1.33 to 2.01)), and revision for PJI (HR 1.35 (95% 1.18 to 1.55)) were independently associated with an increased risk of mortality following revision KA when adjusting for age. Conclusion The SMR of patients undergoing primary and revision KA was lower than that of the general population and remained so for several years post-surgery. However, approximately one in four patients undergoing primary and one in three patients undergoing revision KA died within tenten years of surgery. Several patient and surgical factors, including PJI, were associated with the risk of mortality within ten years of primary and revision surgery. Cite this article: Bone Joint J 2022;104-B(1):45–52.

Charlson Comorbidity Index: A Critical Review of Clinimetric Properties

The present critical review was conducted to evaluate the clinimetric properties of the Charlson Comorbidity Index (CCI), an assessment tool designed specifically to predict long-term mortality, with regard to its reliability, concurrent validity, sensitivity, incremental and predictive validity. The original version of the CCI has been adapted for use with different sources of data, ICD-9 and ICD-10 codes. The inter-rater reliability of the CCI was found to be excellent, with extremely high agreement between self-report and medical charts. The CCI has also been shown either to have concurrent validity with a number of other prognostic scales or to result in concordant predictions. Importantly, the clinimetric sensitivity of the CCI has been demonstrated in a variety of medical conditions, with stepwise increases in the CCI associated with stepwise increases in mortality. The CCI is also characterized by the clinimetric property of incremental validity, whereby adding the CCI to other measures increases the overall predictive accuracy. It has been shown to predict long-term mortality in different clinical populations, including medical, surgical, intensive care unit (ICU), trauma, and cancer patients. It may also predict in-hospital mortality, although in some instances, such as ICU or trauma patients, the CCI did not perform as well as other instruments designed specifically for that purpose. The CCI thus appears to be clinically useful not only to provide a valid assessment of the patient’s unique clinical situation, but also to demarcate major diagnostic and prognostic differences among subgroups of patients sharing the same medical diagnosis.

The impact of complete atrioventricular block on in-hospital and long-term mortality in patients treated with primary percutaneous coronary intervention

Objective: To analyze the incidence and the prognostic impact of complete AV block on in-hospital and 6-year mortality in STEMI patients treated with pPCI. Method: Study included 3044 consecutive STEMI patients. Results: Complete AV block was registered only at admission in 144 (4.73%) patients; 125 (86.8%) patients with complete AV block had inferior infarction. Temporary pacemaker was implanted in 72 (50%) patients with complete AV block. No patient underwent permanent pacemaker implantation. In-hospital mortality was significantly higher in patients with complete AV block than in patients without complete AV block: 17.9%vs3.6%, respectively, p<0.001. In patients with heart block and inferior infarction inhospital mortality was 13%, whereas in patients with heart block and anterior infarction inhospital mortality was 53%. When we analyzed patients who were discharged alive from the hospital, we also found significantly higher long-term (6-year) mortality rate in those with complete AV block vs patients without AV block: 7.8%v 3.4% respectively, p<0.001. Complete AV block was an independent predictor for in-hospital and 6-year mortality: inhospital mortality OR 2.94 95%CI 1.23-5.22; six year mortality HR 1.61, 95%CI 1.10- 2.37. When subanalysis was performed, in patients with inferior STEMI, complete AV block was an independent predictor of in-hospital and 6-year mortality, while in patients with anterior STEMI, complete AV block was an independent predictor of in-hospital mortality. Conclusion: In analyzed STEMI patients complete AV block was transitory and was registered only at hospital admission. Although transitory, complete AV block remained a strong independent predictor of in-hospital and long-term mortality.