Delirium is associated with high mortality in older adult patients with acute decompensated heart failure

Machine learning methods for developing a predictive model of the incidence of delirium in cardiac intensive care units

INTRODUCTION AND OBJECTIVES

Delirium, recognized as a crucial prognostic factor in the cardiac intensive care unit (CICU), has evolved in response to the changing demographics among critically ill cardiac patients. This study aimed to create a predictive model for delirium for patients in the CICU.

METHODS

This study included consecutive patients admitted to the CICU of the Samsung Medical Center. To assess the candidate variables for the model: we applied the following machine learning methods: random forest, extreme gradient boosting, partial least squares, and Plmnet-elastic.net. After selecting relevant variables, we performed a logistic regression analysis to derive the model formula. Internal validation was conducted using 100-repeated hold-out validation.

RESULTS

We analyzed 2774 patients, 677 (24.4%) of whom developed delirium in the CICU. Machine learning-based models showed good predictive performance. Clinically significant and frequently important predictors were selected to construct a delirium prediction scoring model for CICU patients. The model included albumin level, international normalized ratio, blood urea nitrogen, white blood cell count, C-reactive protein level, age, heart rate, and mechanical ventilation. The model had an area under the receiver operating characteristics curve (AUROC) of 0.861 (95%CI, 0.843-0.879). Similar results were obtained in internal validation with 100-repeated cross-validation (AUROC, 0.854; 95%CI, 0.826-0.883).

CONCLUSIONS

Using variables frequently ranked as highly important in four machine learning methods, we created a novel delirium prediction model. This model could serve as a useful and simple tool for risk stratification for the occurrence of delirium at the patient's bedside in the CICU.

In-hospital catatonia, delirium, and coma and mortality: Results from the delirium and catatonia prospective cohort investigation

BACKGROUND

Catatonia, a form of acute brain dysfunction typically linked with severe affective and psychotic disorders, occurs in critical illness with delirium and coma. Delirium and coma are associated with mortality, though catatonia's relationship with mortality is unclear. We aim to describe whether catatonia, delirium, and coma are associated with mortality.

METHODS

We enrolled a convenience cohort of critically ill adults (N = 378) at an academic medical center. We assessed catatonia, delirium, and coma using the Bush-Francis Catatonia Rating Scale, the Confusion Assessment Method for the Intensive Care Unit and the Richmond Agitation-Sedation Scale, respectively. We tested the associations between previous day brain dysfunction state occurrence with in-hospital and one-year mortality using multivariable time-dependent risk models. Additionally, we tested the association between brain dysfunction duration and one-year mortality.

RESULTS

Catatonia was not associated with death on the day after diagnosis during hospitalization, and neither previous catatonia occurrence nor duration was associated with one-year mortality. Delirium was not associated with death on any day following diagnosis during hospitalization, and neither previous delirium occurrence nor duration was associated with one-year mortality. The occurrence of coma was associated with death on any day after diagnosis during hospitalization (HR 2.30,CI 1.19-4.44,p = 0.014), as well as through one year following hospital discharge (HR 1.68,CI 1.09-2.59,p = 0.02).

CONCLUSIONS

Coma, but neither catatonia nor delirium, was associated with future day in-hospital and one-year mortality. More research is needed to understand catatonia's clinical impact. Delirium results differ from existing literature likely due to cohort demographics and size. Coma results highlight the prognostic significance of suppressed arousal while critically ill.

Prognostic significance of delirium in patients with heart failure: a systematic review and meta-analysis

Background

Delirium is a common symptom of heart failure (HF) and is associated with increased mortality, prolonged hospital stays, and heightened medical costs. The impact of delirium on the prognosis of HF patients is currently controversial. Therefore, we conducted a meta-analysis to evaluate the prognostic significance of delirium in HF.

Methods

Relevant articles were systematically searched in PubMed, Cochrane Library, Web of Science, and Embase based on the PRISMA guidelines. Studies that reported mortality and hospitalization-related outcomes in HF patients with or without delirium using raw or adjusted hazard ratio (HR) and odds ratio (OD) were included. Meta-analysis was then performed to evaluate the effect of delirium in HF patients. Outcomes of interest were all-cause mortality and events of the hospitalization.

Results

Of the 1,501 studies identified, 7 eligible studies involving 12,830,390 HF patients (6,322,846 males and 6,507,544 females) were included in the meta-analysis. There were 91,640 patients with delirium (0.71%) and 12,738,750 patients without delirium (99.28%). HF patients with delirium had higher OR for in-hospital mortality (1.95, 95% CI = 1.30-2.91, P = 0.135), higher pooled HR for 90-day mortality (2.64, 95% CI = 1.06-1.56, P = 0.215), higher pooled HR for 1-year mortality (2.08, 95% CI = 1.34-3.22, P = 0.004), and higher pooled HR for 30-day readmission rate (4.15, 95% CI = 2.85-6.04, P = 0.831) than those without delirium.

Conclusion

Current evidence suggests that combined delirium increases the risk of HF-related mortality and hospitalization-related outcomes in patients with HF. However, more research is needed to assess the impact of delirium on the prognosis of HF patients.

Delirium and its association with short-term outcomes in younger and older patients with acute heart failure

Younger patients (18 to 65 years old) are often excluded from delirium outcome studies. We sought to determine if delirium was associated with short-term adverse outcomes in a diverse cohort of younger and older patients with acute heart failure (AHF). We conducted a multi-center prospective cohort study that included adult emergency department patients with confirmed AHF. Delirium was ascertained using the Brief Confusion Assessment Method (bCAM). The primary outcome was a composite outcome of 30-day all-cause death, 30-day all-cause rehospitalization, and prolonged index hospital length of stay. Multivariable logistic regression was performed, adjusting for demographics, cognitive impairment without delirium, and HF risk factors. Older age (≥ 65 years old)*delirium interaction was also incorporated into the model. Odds ratios (OR) with their 95% confidence intervals (95%CI) were reported. A total of 1044 patients with AHF were enrolled; 617 AHF patients were < 65 years old and 427 AHF patients were ≥ 65 years old, and 47 (7.6%) and 40 (9.4%) patients were delirious at enrollment, respectively. Delirium was significantly associated with the composite outcome (adjusted OR = 1.64, 95%CI: 1.02 to 2.64). The older age*delirium interaction p-value was 0.47. In conclusion, delirium was common in both younger and older patients with AHF and was associated with poorer short-term outcomes in both cohorts. Younger patients with acute heart failure should be included in future delirium outcome studies.

Impact of delirium on mortality in patients hospitalized for heart failure

OBJECTIVE

Heart Failure (HF) is one of the leading causes of hospitalization in the United States accounting for ≈800,000 hospital discharges and $11 billion in annual costs. Delirium occurs in approximately 30% of elderly hospitalized patients and its incidence is significantly higher among those admitted to the critical care units. Despite this, there has been limited exploration of the clinical and economic impact of delirium in patients hospitalized with acute HF. We hypothesized that delirium in HF is associated with excess mortality and hospital costs.

METHODS

We queried the 2001-2014 Nationwide Inpatient Sample to identify hospitalizations that included a primary discharge diagnosis of HF (ICD-9-CM: 428.xx) and stratified them by presence or absence of delirium (ICD-9-CM: 239.0, 290.41, 293.0, 293.1, 348.31). Differences in in-hospital mortality, length of stay (LOS), and hospital costs were assessed using propensity-score matched cohorts.

RESULTS

Major predictors of delirium included advanced age, Caucasian race, underlying dementia or psychiatric diagnoses, higher Elixhauser Comorbidity Index, renal failure, cardiogenic shock, and coronary artery bypass surgery. In the propensity-score matched analysis of 76,411 hospitalization with delirium compared to 76,612 without delirium, in-hospital mortality (odds ratio: 1.67, 95% CI: 1.51-1.77), LOS (rate ratio [RR]: 1.47, 95% CI: 1.45-1.51), and hospital costs (RR: 1.44, 95% CI: 1.41-1.48) were all statistically higher in the presence of delirium (all p < 0.001).

CONCLUSION

In patients hospitalized with HF, delirium is an independent predictor of increased in-hospital mortality, longer LOS, and excess hospital costs despite adjustment for baseline characteristics.