Inpatient palliative care referral and 9-month hospital readmission in patients with congestive heart failure: a linked nationwide analysis.

OBJECTIVE: End-stage heart failure (HF) is characterized by high symptom burden and frequent hospitalization. Palliative care (PC) is recommended for advanced HF, and there is some evidence in other diseases that this may reduce readmission rates. We attempted examine the association of an inpatient PC visit on hospital readmission for patients admitted with HF. METHODS: Retrospective linked nationwide analysis from 2013 with 9-month follow-up for all hospital readmissions for patients admitted with HF exacerbations using the Nationwide Readmission Database (NRD). The NRD gathers all hospital admissions for patients from 22 states and tracks patients throughout the year, allowing for examination of readmission statistics. A propensity score model for PC visit was made, and patients were matched in a 1 : 1 fashion. RESULTS: There were 102 746 patients who survived an admission for HF in the first 3 months of 2013. Of these, 2287 (2.2%) patients had a PC visit as inpatients. After matching based on propensity for a PC visit during the index hospitalization, 2282 patients who received a PC visit were matched to 2282 patients who did not. Those receiving a PC visit were less likely to be readmitted for HF (9.3% vs. 22.4%, P < 0.01) or for any cause (29.0% vs. 63.2%, P < 0.01) during the 9-month follow-up period. The average hospital charges during the follow-up period for the non-PC cohort were $77 643 per patient. The average charges for PC patients were $23 200 (P < 0.01). CONCLUSIONS: Patients with HF who received an inpatient PC visit had significantly lower rates of all-cause and HF-specific readmission in the subsequent 9 months. Total 9-month hospital charges were also significantly lower for patients who received an inpatient PC visit.

Association between fluid balance and survival in critically ill patients.

OBJECTIVE: Although the consequences of chronic fluid retention are well known, those of iatrogenic fluid retention that occurs during critical illness have not been fully determined. Therefore, we investigated the association between fluid balance and survival in a cohort of almost 16,000 individuals who survived an intensive care unit (ICU) stay in a large, urban, tertiary medical centre. DESIGN: Longitudinal analysis of fluid balance at ICU discharge and 90-day post-ICU survival. MEASUREMENTS: Associations between fluid balance during the ICU stay, determined from the electronic bedside record, and survival were tested using Cox proportional hazard models adjusted for severity of critical illness. RESULTS: There were 1827 deaths in the first 90 days after ICU discharge. Compared with the lowest quartile of discharge fluid balance [median (interquartile range) -1.5 (-3.1, -0.7) L], the highest quartile [7.6 (5.7, 10.8) L] was associated with a 35% [95% confidence interval (CI) 1.13-1.61)] higher adjusted risk of death. Fluid balance was not associated with outcome amongst individuals without congestive heart failure or renal dysfunction. Amongst patients with either comorbidity, however, fluid balance was strongly associated with outcome, with the highest quartile having a 55% (95% CI 1.24-1.95) higher adjusted risk of death than the lowest quartile. Isotonic fluid balance, defined as the difference between intravenous isotonic fluid administration and urine output, was similarly associated with 90-day outcomes. CONCLUSION: Positive fluid balance at the time of ICU discharge is associated with increased risk of death, after adjusting for markers of illness severity and chronic medical conditions, particularly in patients with underlying heart or kidney disease. Restoration of euvolaemia prior to discharge may improve survival after acute illness.

A machine learning-based model for 1-year mortality prediction in patients admitted to an Intensive Care Unit with a diagnosis of sepsis.

INTRODUCTION: Sepsis is associated to a high mortality rate, and its severity must be evaluated quickly. The severity of illness scores used are intended to be applicable to all patient populations, and generally evaluate in-hospital mortality. However, patients with sepsis continue to be at risk of death after hospital discharge. OBJECTIVE: To develop a model for predicting 1-year mortality in critical patients diagnosed with sepsis. PATIENTS: The data corresponding to 5650 admissions of patients with sepsis from the Medical Information Mart for Intensive Care (MIMIC-III) database were evaluated, randomly divided as follows: 70% for training and 30% for validation. DESIGN: A retrospective register-based cohort study was carried out. The clinical information of the first 24h after admission was used to develop a 1-year mortality prediction model based on Stochastic Gradient Boosting (SGB) methodology. Variable selection was addressed using Least Absolute Shrinkage and Selection Operator (LASSO) and SGB variable importance methodologies. The predictive power was evaluated using the area under the ROC curve (AUROC). RESULTS: An AUROC of 0.8039 (95% confidence interval (CI): [0.8033 0.8045]) was obtained in the validation subset. The model exceeded the predictive performances obtained with traditional severity of disease scores in the same subset. CONCLUSION: The use of assembly algorithms, such as SGB, for the generation of a customized model for sepsis yields more accurate 1-year mortality prediction than the traditional scoring systems such as SAPS II, SOFA or OASIS.

Reducing unnecessary lab testing in the ICU with artificial intelligence.

OBJECTIVES: To reduce unnecessary lab testing by predicting when a proposed future lab test is likely to contribute information gain and thereby influence clinical management in patients with gastrointestinal bleeding. Recent studies have demonstrated that frequent laboratory testing does not necessarily relate to better outcomes. DESIGN: Data preprocessing, feature selection, and classification were performed and an artificial intelligence tool, fuzzy modeling, was used to identify lab tests that do not contribute an information gain. There were 11 input variables in total. Ten of these were derived from bedside monitor trends heart rate, oxygen saturation, respiratory rate, temperature, blood pressure, and urine collections, as well as infusion products and transfusions. The final input variable was a previous value from one of the eight lab tests being predicted: calcium, PTT, hematocrit, fibrinogen, lactate, platelets, INR and hemoglobin. The outcome for each test was a binary framework defining whether a test result contributed information gain or not. PATIENTS: Predictive modeling was applied to recognize unnecessary lab tests in a real world ICU database extract comprising 746 patients with gastrointestinal bleeding. MAIN RESULTS: Classification accuracy of necessary and unnecessary lab tests of greater than 80% was achieved for all eight lab tests. Sensitivity and specificity were satisfactory for all the outcomes. An average reduction of 50% of the lab tests was obtained. This is an improvement from previously reported similar studies with average performance 37% by [1-3]. CONCLUSIONS: Reducing frequent lab testing and the potential clinical and financial implications are an important issue in intensive care. In this work we present an artificial intelligence method to predict the benefit of proposed future laboratory tests. Using ICU data from 746 patients with gastrointestinal bleeding, and eleven measurements, we demonstrate high accuracy in predicting the likely information to be gained from proposed future lab testing for eight common GI related lab tests. Future work will explore applications of this approach to a range of underlying medical conditions and laboratory tests.

Disease-based modeling to predict fluid response in intensive care units.

OBJECTIVE: To compare general and disease-based modeling for fluid resuscitation and vasopressor use in intensive care units. METHODS: Retrospective cohort study involving 2944 adult medical and surgical intensive care unit (ICU) patients receiving fluid resuscitation. Within this cohort there were two disease-based groups, 802 patients with a diagnosis of pneumonia, and 143 patients with a diagnosis of pancreatitis. Fluid resuscitation either progressing to subsequent vasopressor administration or not was used as the primary outcome variable to compare general and disease-based modeling. RESULTS: Patients with pancreatitis, pneumonia and the general group all shared three common predictive features as core variables, arterial base excess, lactic acid and platelets. Patients with pneumonia also had non-invasive systolic blood pressure and white blood cells added to the core model, and pancreatitis patients additionally had temperature. Disease-based models had significantly higher values of AUC (p < 0.05) than the general group (0.82 ± 0.02 for pneumonia and 0.83 ± 0.03 for pancreatitis vs. 0.79 ± 0.02 for general patients). CONCLUSIONS: Disease-based predictive modeling reveals a different set of predictive variables compared to general modeling and improved performance. Our findings add support to the growing body of evidence advantaging disease specific predictive modeling.

Effects of age and coronary artery disease on cerebrovascular reactivity to carbon dioxide in humans.

Alterations in cerebrovascular reactivity to CO2, an index of cerebrovascular function, have been associated with increased risk of stroke. We hypothesised that cerebrovascular reactivity is impaired with increasing age and in patients with symptomatic coronary artery disease (CAD). Cerebrovascular and cardiovascular reactivity to CO2 was assessed at rest and during hypercapnia (5% CO2) and hypocapnia (hyperventilation) in subjects with symptomatic CAD (n=13) and age-matched old (n=9) and young (n=20) controls without CAD. Independent of CAD, reductions in middle cerebral artery blood velocity (transcranial Doppler) and cerebral oxygenation (near-infrared spectroscopy) were correlated with increasing age (r = -0.68, r = -0.51, respectively, P < 0.01). In CAD patients, at rest and during hypercapnia, cerebral oxygenation was lower (P < 0.05 vs. young). Although middle cerebral artery blood velocity reactivity was unaltered in the hypercapnic range, middle cerebral artery blood velocity reactivity to hypocapnia was elevated in the CAD and age-matched controls (P < 0.01 vs. young), and was associated with age (r = 0.62, P < 0.01). Transient drops in arterial PCO2 occur in a range of physiological and pathophysiological situations, therefore, the elevated middle cerebral artery blood velocity reactivity to hypocapnia combined with reductions in middle cerebral artery blood velocity may be important mechanisms underlying neurological risk with aging. In CAD patients, additional reductions in cerebral oxygenation may place them at additional risk of cerebral ischaemia.

Quality of dying in a New Zealand teaching hospital.

OBJECTIVE: Studies suggest that there is a need to improve the way we deliver care at the end of life. Based on recommendations from end-of-life experts, metrics were identified to measure the quality of dying in Dunedin Hospital. DESIGN: A retrospective observational study was performed to assess the care provided to patients who died in the hospital in 2003. SETTING: Dunedin Hospital is a 350-bed tertiary care teaching hospital located in the South Island of New Zealand. SUBJECTS AND METHOD: Medical records of 200 consecutive decedents were reviewed to evaluate communication, interventions, and symptom control during their terminal hospitalisation. RESULTS: Mean hospital length-of-stay was 8 days; 38 patients (19%) died following an ICU admission. There was documentation of end-of-life discussion with either the patient or the family in 164 patients (82%). 74% had a DNR order. Pain status was documented in 140 patients (70%); 134 of these patients were pain-free. CONCLUSION: Overall, the results suggest that the ideals in end-of-life care pertaining to pain control, communication and avoidance of unnecessary interventions were achieved in a majority of the decedents during the study period. The socialised healthcare system, the availability of resources, societal expectations and a lack of a litigious environment are theorised to positively influence end-of-life care delivery in New Zealand.

The eICU: it's not just telemedicine.

Intensive care units (ICUs) are major sites for medical errors and adverse events. Suboptimal outcomes reflect a widespread failure to implement care delivery systems that successfully address the complexity of modern ICUs. Whereas other industries have used information technologies to fundamentally improve operating efficiency and enhance safety, medicine has been slow to implement such strategies. Most ICUs do not even track performance; fewer still have the capability to examine clinical data and use this information to guide quality improvement initiatives. This article describes a technology-enabled care model (electronic ICU, or eICU) that represents a new paradigm for delivery of critical care services. A major component of the model is the use of telemedicine to leverage clinical expertise and facilitate a round-the-clock proactive care by intensivist-led teams of ICU caregivers. Novel data presentation formats, computerized decision support, and smart alarms are used to enhance efficiency, increase effectiveness, and standardize clinical and operating processes. In addition, the technology infrastructure facilitates performance improvement by providing an automated means to measure outcomes, track performance, and monitor resource utilization. The program is designed to support the multidisciplinary intensivist-led team model and incorporates comprehensive ICU re-engineering efforts to change practice behavior. Although this model can transform ICUs into centers of excellence, success will hinge on hospitals accepting the underlying value proposition and physicians being willing to change established practices.

Adenovirus vector-based purging of multiple myeloma cells.

Adenoviruses are efficient gene delivery agents for a variety of neoplasms. In the present study, we have investigated the use of adenoviruses for the delivery of the thymidine kinase (tk) gene into multiple myeloma (MM) cells. We first demonstrated that MM cell lines and MM patient cells express both adenovirus receptors as well as the DF3/MUC1 protein, thus providing a rationale for using adenoviruses to selectively deliver genes under the control of the DF3 promoter. By using an adenoviral construct containing beta-galactosidase (beta-gal) gene driven by the DF3 promoter (Ad. DF3-betagal), we demonstrate greater than 80% transduction efficiency in OCI-My5 and RPMI 8226 MM cell lines at a multiplicity of infection of 1 to 100. Importantly, transduction with the tk gene driven by the DF3 promoter (Ad.DF3-tk) followed by treatment with 50 micromol/L ganciclovir (GCV) purged >/=6 log of contaminating OCI-My5 and RPMI 8226 MM cells within bone marrow mononuclear cells. In contrast, normal human hematopoietic progenitor cell number was unaffected under these conditions. Selectivity of DF3/MUC1 promoter was further confirmed, because Ad.DF3-betagal or Ad.DF3-tk did not transduce MUC1-negative HeLa cervical carcinoma cells. In addition, GCV treatment of Ad.DF3-tk-transduced RPMI 8226 MM cells did not induce a significant bystander effect. These findings demonstrate that transduction with Ad vectors using a tumor-selective promoter provides a highly efficient and selective approach for the ex vivo purging of MM cells.