Deep-Learning-Based Survival Prediction of Patients in Coronary Care Units

Background. A survival prediction model based on deep learning has higher accuracy than the CPH model in predicting the survival of CCU patients, and it also has a better discrimination ability. We collected information on patients with various diseases in coronary care units (CCUs) from the Medical Information Mart for Intensive Care III (MIMIC-III) database. The purpose of this study was to use this information to construct a neural-network model based on deep learning to predict the survival probabilities of patients with conditions that are common in CCUs. Method. We collected information on patients in the United States with five common diseases in CCUs from 2001 to 2012. We randomly divided the patients into a training cohort and a testing cohort at a ratio of 7 : 3 and applied a survival prediction method based on deep learning to predict their survival probability. We compared our model with the Cox proportional-hazards regression (CPH) model and used the concordance indexes (C-indexes), receiver operating characteristic (ROC) curve, and calibration plots to evaluate the predictive performance of the model. Results. The 3,388 CCU patients included in the study were randomly divided into 2,371 in the training cohort and 1,017 in the testing cohort. The stepwise regression results showed that the important factors affecting patient survival were the type of disease, age, race, anion gap, glucose, neutrophils, white blood cells, potassium, creatine kinase, and blood urea nitrogen ( P < 0.05 ). We used the training cohort to construct a deep-learning model, for which the C-index was 0.833, or about 5% higher than that for the CPH model (0.786). The C-index of the deep-learning model for the test cohort was 0.822, which was also higher than that for the CPH model (0.782). The areas under the ROC curve for the 28-day, 90-day, and 1-year survival probabilities were 0.875, 0.865, and 0.874, respectively, in the deep-learning model, respectively, and 0.830, 0.843, and 0.806 in the CPH model. These values indicate that the survival analysis model based on deep learning is better than the traditional CPH model in predicting the survival of CCU patients. Conclusion. A survival prediction model based on deep learning has higher accuracy than the CPH model in predicting the survival of CCU patients, and it also has a better discrimination ability.

Management of temporary epicardial pacing wires in the cardiac surgical patient

Temporary epicardial pacing wires are used after cardiothoracic surgery to maintain a stable cardiac rhythm. They must be distinguished from the more commonly encountered transvenous temporary pacing wires, which are often used in coronary care units for the same purpose. Patients with temporary epicardial pacing wires may be transferred to hospital wards where these wires are not usually encountered, such as COVID wards, the general intensive care unit, the coronary care unit or general surgical wards if a laparotomy was required in the early period following cardiac surgery. Serious complications may arise in managing patients with temporary epicardial pacing wires, which are well known in the cardiothoracic unit but not so well known elsewhere in the hospital. This article discusses the dangers associated with the management of temporary epicardial pacing wires in adult patients, some of which are common to temporary transvenous pacing wires and others are unique to temporary epicardial pacing wires.

The impact of using “ISBAR” standard checklist on nursing clinical handoff in coronary care units

Background & Aim: Clinical handoff is the process of transmitting information, responsibility, and accountability among the health care team members. Lack of standard protocols may result in the loss of essential information and may lead to medical errors. The purpose of this study was to evaluate the impact of using a standard checklist on a clinical handoff in the coronary care unit. Methods & Materials: This quasi-experimental study was performed based on pre- and post-test design at Afshar Hospital in Yazd. There was a total of 564 handoffs with the participation of 24 nurses in two coronary care units in 2017. Before the intervention, 282 clinical handoffs were recorded and implemented. Nurses were informed about the ISBAR standard checklist and were encouraged to use it for one week. Then, 282 clinical handoffs were again recorded and implemented. The frequency of providing information during clinical handoff was determined based on the ISBAR checklist and the data were analyzed using descriptive statistics and chi-square tests. Results: Before the intervention, the frequency of providing information during clinical handoff was reported as follows: patient identity (86.9%), current position (75.1%), clinical history (52.8%), system status review (59.9%), and recommendations (92.9%). The results showed that the indexes significantly increased (P <0.001) after the intervention in all these five domains: patient identity (100%), current situation (94%), clinical history (80.1%), system status review (92.2%) and recommendations (100%). Conclusion: Transition of information based on standard checklists with a specific framework can increase the frequency of information provided during clinical handoff. Therefore, it is recommended to train nurses and nursing students about standard handoff and related tools such as ISBAR in hospitals and universities.

Post-Infectious Myocardial Infarction: Does Percutaneous Coronary Intervention Improve Outcomes? A Propensity Score-Matched Analysis

Acute infection is a frequent trigger of myocardial infarction (MI). However, whether percutaneous coronary intervention (PCI) improves post-infectious MI prognosis is a major but unsolved issue. In this prospective multicenter study from coronary care units, we performed propensity score-matched analysis to compare outcomes in patients with and without PCI for post-infectious MI with angiography-proven significant coronary stenosis (>50%). Among 4573 consecutive MI patients, 476 patients (10%) had a concurrent diagnosis of acute infection at admission, of whom 375 underwent coronary angiography and 321 patients had significant stenosis. Among the 321 patients, 195 underwent PCI. Before the matching procedure, patients without PCI had a similar age and sex ratio but a higher rate of risk factors (hypertension, diabetes, chronic renal failure, and prior coronary artery disease), pneumonia, and SYNTAX score than patients without PCI. After propensity score matching, neither in-hospital mortality (13% with PCI vs. 8% without PCI; p = 0.4) nor one-year mortality (24% with PCI vs. 19% without PCI, p = 0.5) significantly differed between the two groups. In this first prospective cohort of post-infectious MI in coronary care units, PCI might not improve short- and long-term prognosis in patients with angiography-proven significant coronary stenosis. If confirmed, these results do not argue for systematic invasive procedures after post-infectious MI.

Diferenças sociodemográficas, clínicas e de tratamento entre idosos com doenças cardiovasculares do SUS e de convênios/particulares

OBJECTIVE: To evaluate the sociodemographic, clinical characteristics and prevalence of medications used by elderly patients with cardiovascular diseases, addressing the difference between users of the single health care system (SUS) and private/covenant during hospitalization. Methods: A retrospective cross-sectional study was carried out in a large hospital in Porto Alegre, by reviewing medical records of patients hospitalized at the coronary care units in the period from January to December 2015. The sample consisted of the evaluation of 152 medical records, the collection was performed through a specific semi-structured instrument. RESULTS: A total of 68 elderly patients belonged to the SUS user group. The other 84 medical records were users of several Covenants / Individuals groups. Regarding lifestyle, it was shown that smoking was representative. The presence of a significant association Regarding the clinical profile, all patients in both groups were hospitalized in coronary care units. CONCLUSIONS: It is concluded that the patients in the private / covenants group use a greater number of drugs outside the National Essential Medicines report than those in the SUS group. The SUS group stands out for being more hypertensive and smokers and as a result, they perform more hospitalizations with more severe pathologies when compared to the other group. These comparative data highlight the importance of actions that aim to balance the differences between the modalities of health plans served in Brazil.