Predicting ICU readmission among surgical ICU patients: Development and validation of a clinical nomogram

Predicting Unplanned Readmissions to the Intensive Care Unit in the Trauma Population

Background: Unplanned readmission to intensive care units (UR-ICU) in trauma is associated with increased hospital length of stay and significant morbidity and mortality. We identify independent predictors of UR-ICU and construct a nomogram to estimate readmission probability. Materials and Methods: We performed an IRB-approved retrospective case-control study at a Level I trauma center between January 2019 and December 2021. Patients with UR-ICU (n = 175) were matched with patients who were not readmitted (NR-ICU) (n = 175). Univariate and multivariable binary linear regressionanalyses were performed (SPSS Version 28, IBM Corp), and a nomogram was created (Stata 18.0, StataCorp LLC). Results: Demographics, comorbidities, and injury- and hospital course-related factors were examined as potential prognostic indicators of UR-ICU. The mortality rate of UR-ICU was 22.29% vs 6.29% for NR-ICU (P < .001). Binary linear regression identified seven independent predictors that contributed to UR-ICU: shock (P < .001) or intracranial surgery (P = .015) during ICU admission, low hematocrit (P = .001) or sedation administration in the 24 hours before ICU discharge (P < .001), active infection treatment (P = .192) or leukocytosis on ICU discharge (P = .01), and chronic obstructive pulmonary disease (COPD) (P = .002). A nomogram was generated to estimate the probability of UR-ICU and guide decisions on ICU discharge appropriateness. Discussion: In trauma, UR-ICU is often accompanied by poor outcomes and death. Shock, intracranial surgery, anemia, sedative administration, ongoing infection treatment, leukocytosis, and COPD are significant risk factors for UR-ICU. A predictive nomogram may help better assess readiness for ICU discharge.

Causes and Timing of Unplanned ICU Admissions Among Trauma Patients at a Level 1 Trauma Center

BACKGROUND

Unexpected ICU admissions are a key quality metric in trauma care. The purpose of this study is to identify the most common causes of unplanned ICU admissions among trauma patients at an ACS-verified level 1 trauma center.

METHODS

A retrospective review was conducted of all trauma patients with unplanned admission to the ICU at a level 1 trauma center between 2019 and 2021. Unplanned ICU admissions were categorized into (1) "bounce-backs," patients previously admitted to the ICU and (2) "upgrades," patients who had not previously been cared for in the ICU.

RESULTS

Of 300 unexpected ICU transfers, bounce-backs accounted for 69% and upgrades 31%. The most common injuries were traumatic brain injuries (40%) and rib fractures (41.3%). In-hospital mortality rate was 10% and did not significantly differ between bounce-backs and upgrades (12 vs 5%, P = .92). Respiratory distress was the most common cause of transfer (41.1%), followed by neurologic (29.6%) and cardiovascular decline (21.2%). Patients were on average 928 mL fluid positive 72 hours prior to transfer (t > 0, P < .0001), and 295 mL fluid positive in the 24 hours prior to transfer (t > 0, P .0003). Patients transferred for respiratory distress were no more fluid over-balanced than those transferred for other reasons.

CONCLUSION

We found a large percent of unplanned transfers occurring within 48 hours of admission or transfer out of the ICU suggesting under-triage as a leading cause of bounce-backs and upgrades. Respiratory distress was the leading cause of transfer. These findings highlight opportunities for targeted interventions.

The correlation between admission hyperglycemia and 30-day readmission after hip fracture surgery in geriatric patients: a propensity score-matched study

Purpose

This study aimed to investigate the association between admission hyperglycemia and 30-day readmission after hip fracture surgery in geriatric patients.

Methods

This retrospective study included 1253 geriatric hip fracture patients. Patients were categorized into normoglycemia(<6.10 mmol/L) and hyperglycemia groups(≥6.10 mmol/L) based on admission blood glucose. We performed multivariable logistic regression analyses and propensity score matching (PSM) to estimate adjusted odds ratios and 95% confidence intervals for 30-day readmission, controlling for potential confounding factors. An analysis of the dose-dependent association between admission blood glucose and the probability of 30-day readmission was performed. Additional subgroup analysis was conducted to examine the impact of other factors on the relationship between admission blood glucose and 30-day readmission.

Results

Patients with hyperglycemia had higher 30-day readmission rates than normoglycemic patients before (19.1% vs 9.7%, p<0.001) and after PSM (18.1% vs 12.3%, p=0.035). Admission hyperglycemia was an independent predictor of increased 30-day readmission risk, with an adjusted odds ratio of 1.57 (95% CI 1.08-2.29, p=0.019) after multivariable regression and 1.57 (95% CI 1.03-2.39, p=0.036) after PSM. A dose-response relationship was observed between higher glucose levels and increased readmission risk.

Conclusion

Admission hyperglycemia is an independent risk factor for 30-day readmission after hip fracture surgery in the elderly. Routine glucose testing upon admission and perioperative glycemic control may help reduce short-term readmissions in this vulnerable population.

Prognostic factors for critically ill surgical patients with unplanned intensive care unit readmission: Developing a novel predictive scoring model for predicting readmission

BACKGROUND

Unplanned readmission to the surgical intensive care unit has been demonstrated to worsen patient outcomes. Our objective was to identify risk factors and outcomes associated with unplanned surgical intensive care unit readmission and to develop a predictive scoring model to identify patients at high risk of readmission.

METHODS

We retrospectively analyzed patients admitted to the surgical intensive care unit (2020-2021) and categorized them as either with or without unplanned readmission.

RESULTS

Of 1,112 patients in the derivation cohort, 76 (6.8%) experienced unplanned surgical intensive care unit readmission, with sepsis being the leading cause of readmission (35.5%). Patients who were readmitted had significantly higher in-hospital mortality rates than those who were not. Multivariate analysis identified congestive heart failure, high Sequential Organ Failure Assessment-Hepatic score, use of carbapenem during surgical intensive care unit stay, as well as factors before surgical intensive care unit discharge such as inadequate glycemic control, positive fluid balance, low partial pressure of oxygen in arterial blood/fraction of inspired oxygen ratio, and receipt of total parenteral nutrition as independent predictors for unplanned readmission. The scoring model developed using these predictors exhibited good discrimination between readmitted and non-readmitted patients, with an area under the curve of 0.74. The observed rates of unplanned readmission for scores of <4 points and ≥4 points were 4% and 20.2% (P < .001), respectively. The model also demonstrated good performance in the validation cohort, with an area under the curve of 0.74 and 19% observed unplanned readmission rate for scores ≥4 points.

CONCLUSION

Besides congestive heart failure, clinicians should meticulously re-evaluate critical variables such as the Sequential Organ Failure Assessment-Hepatic score, partial pressure of oxygen in arterial blood/fraction of inspired oxygen ratio, glycemic control, and fluid status before releasing the patient from the surgical intensive care unit. It is crucial to determine the reasons for using carbapenems during surgical intensive care unit stay and the causes for the inability to discontinue total parenteral nutrition before discharging the patient from the surgical intensive care unit.

Intensive care unit readmission and unexpected death after emergency general surgery

Background

Intensive care unit (ICU) readmission and unexpected death are closely associated with increased length of hospitalization and total mortality. However, data about readmission or unexpected death after discharge from ICU in patients who have undergone emergency general surgery (EGS) is very limited.

Methods

In total, 1133 patients who underwent EGS were identified in the Multiparameter Intelligent Monitoring in Intensive Care IV (MIMIC-IV) database. Of these 1133 patients, 124 underwent readmission into the ICU or death unexpectedly after their initial discharge. The clinical characteristics of the patients were investigated. A logistic regression model was implemented for the analysis of the independent risk factors associated with ICU readmission or unexpected death. A nomogram model was established to predict the risk of ICU readmission or unexpected death within 72 h after EGS.

Results

Peripheral vascular disease and atrial fibrillation, vasopressor requirement, a higher respiratory rate or heart rate, a lower pulse oxygen saturation or a platelet count of <150 K/μL and a relatively low Glasgow coma scale score in the last 24 h before ICU discharge were independent risk factors for ICU readmission or death within 72 h. The nomogram had moderate accuracy with an area under the curve of 0.852, which had a stronger prediction power than the Stability and Workload Index for Transfer (SWIFT) score, a classic prediction model for ICU readmission risk.

Conclusions

In critically ill patients who undergo EGS, ICU readmission or unexpected death within 72 h can be predicted using a nomogram model based on eight parameters including physiological and laboratory test values in the last 24 h before discharge and comorbidities. ICU physicians should prudently assess patients to make effective discharge decisions.

The risk assessment tool for intensive care unit readmission: A systematic review and meta-analysis

OBJECTIVE

To review and evaluate existing risk assessment tools for intensive care unitreadmission.

METHODS

Nine electronic databases (Medline, CINAHL, Web of Science, Cochrane Library, Embase, Sino Med, CNKI, VIP, and Wan fang) were systematically searched from their inception to September 2022. Two authors independently extracted data from the literature included. Meta-analysis was performed under the bivariate modeling and summary receiver operating characteristic curve method.

RESULTS

A total of 29 studies were included in this review, among which 11 were quantitatively Meta-analyzed. The results showed Stability and Workload Index for Transfer: Sensitivity = 0.55, Specificity = 0.65, Area under curve = 0.63. And Early warning score: Sensitivity = 0.78, Specificity = 0.83, Area under curve = 0.88. The remaining tools included scores, nomograms, machine learning models, and deep learning models. These studies, with varying reports on thresholds, case selection, data preprocessing, and model performance, have a high risk of bias.

CONCLUSION

We cannot identify a tool that can be used directly in intensive care unit readmission risk assessment. Scores based on early warning score are moderately accurate in predicting readmission, but there is heterogeneity and publication bias that requires model adjustment for local factors such as resources, demographics, and case mix. Machine learning models present a promising modeling technique but have a high methodological bias and require further validation.

IMPLICATIONS FOR CLINICAL PRACTICE

Using reliable risk assessment tools is essential for the early identification of unplanned intensive care unit readmission risk in critically ill patients.  A reliable risk assessment tool must be developed, which is the focus of further research.

Explainable time-series deep learning models for the prediction of mortality, prolonged length of stay and 30-day readmission in intensive care patients

Background

In-hospital mortality, prolonged length of stay (LOS), and 30-day readmission are common outcomes in the intensive care unit (ICU). Traditional scoring systems and machine learning models for predicting these outcomes usually ignore the characteristics of ICU data, which are time-series forms. We aimed to use time-series deep learning models with the selective combination of three widely used scoring systems to predict these outcomes.

Materials and methods

A retrospective cohort study was conducted on 40,083 patients in ICU from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Three deep learning models, namely, recurrent neural network (RNN), gated recurrent unit (GRU), and long short-term memory (LSTM) with attention mechanisms, were trained for the prediction of in-hospital mortality, prolonged LOS, and 30-day readmission with variables collected during the initial 24 h after ICU admission or the last 24 h before discharge. The inclusion of variables was based on three widely used scoring systems, namely, APACHE II, SOFA, and SAPS II, and the predictors consisted of time-series vital signs, laboratory tests, medication, and procedures. The patients were randomly divided into a training set (80%) and a test set (20%), which were used for model development and model evaluation, respectively. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and Brier scores were used to evaluate model performance. Variable significance was identified through attention mechanisms.

Results

A total of 33 variables for 40,083 patients were enrolled for mortality and prolonged LOS prediction and 36,180 for readmission prediction. The rates of occurrence of the three outcomes were 9.74%, 27.54%, and 11.79%, respectively. In each of the three outcomes, the performance of RNN, GRU, and LSTM did not differ greatly. Mortality prediction models, prolonged LOS prediction models, and readmission prediction models achieved AUCs of 0.870 ± 0.001, 0.765 ± 0.003, and 0.635 ± 0.018, respectively. The top significant variables co-selected by the three deep learning models were Glasgow Coma Scale (GCS), age, blood urea nitrogen, and norepinephrine for mortality; GCS, invasive ventilation, and blood urea nitrogen for prolonged LOS; and blood urea nitrogen, GCS, and ethnicity for readmission.

Conclusion

The prognostic prediction models established in our study achieved good performance in predicting common outcomes of patients in ICU, especially in mortality prediction. In addition, GCS and blood urea nitrogen were identified as the most important factors strongly associated with adverse ICU events.

Machine learning prediction of postoperative unplanned 30-day hospital readmission in older adult

Background: Although unplanned hospital readmission is an important indicator for monitoring the perioperative quality of hospital care, few published studies of hospital readmission have focused on surgical patient populations, especially in the elderly. We aimed to investigate if machine learning approaches can be used to predict postoperative unplanned 30-day hospital readmission in old surgical patients.

Methods: We extracted demographic, comorbidity, laboratory, surgical, and medication data of elderly patients older than 65 who underwent surgeries under general anesthesia in West China Hospital, Sichuan University from July 2019 to February 2021. Different machine learning approaches were performed to evaluate whether unplanned 30-day hospital readmission can be predicted. Model performance was assessed using the following metrics: AUC, accuracy, precision, recall, and F1 score. Calibration of predictions was performed using Brier Score. A feature ablation analysis was performed, and the change in AUC with the removal of each feature was then assessed to determine feature importance.

Results: A total of 10,535 unique surgeries and 10,358 unique surgical elderly patients were included. The overall 30-day unplanned readmission rate was 3.36%. The AUCs of the six machine learning algorithms predicting postoperative 30-day unplanned readmission ranged from 0.6865 to 0.8654. The RF + XGBoost algorithm overall performed the best with an AUC of 0.8654 (95% CI, 0.8484–0.8824), accuracy of 0.9868 (95% CI, 0.9834–0.9902), precision of 0.3960 (95% CI, 0.3854–0.4066), recall of 0.3184 (95% CI, 0.259–0.3778), and F1 score of 0.4909 (95% CI, 0.3907–0.5911). The Brier scores of the six machine learning algorithms predicting postoperative 30-day unplanned readmission ranged from 0.3721 to 0.0464, with RF + XGBoost showing the best calibration capability. The most five important features of RF + XGBoost were operation duration, white blood cell count, BMI, total bilirubin concentration, and blood glucose concentration.

Conclusion: Machine learning algorithms can accurately predict postoperative unplanned 30-day readmission in elderly surgical patients.

Electronic health record machine learning model predicts trauma inpatient mortality in real time: A validation study

INTRODUCTION

Patient outcome prediction models are underused in clinical practice because of lack of integration with real-time patient data. The electronic health record (EHR) has the ability to use machine learning (ML) to develop predictive models. While an EHR ML model has been developed to predict clinical deterioration, it has yet to be validated for use in trauma. We hypothesized that the Epic Deterioration Index (EDI) would predict mortality and unplanned intensive care unit (ICU) admission in trauma patients.

METHODS

A retrospective analysis of a trauma registry was used to identify patients admitted to a level 1 trauma center for >24 hours from October 2019 to July 2020. We evaluated the performance of the EDI, which is constructed from 125 objective patient measures within the EHR, in predicting mortality and unplanned ICU admissions. We performed a 5 to 1 match on age because it is a major component of EDI, then examined the area under the receiver operating characteristic curve (AUROC), and benchmarked it against Injury Severity Score (ISS) and new injury severity score (NISS).

RESULTS

The study cohort consisted of 1,325 patients admitted with a mean age of 52.5 years and 91% following blunt injury. The in-hospital mortality rate was 2%, and unplanned ICU admission rate was 2.6%. In predicting mortality, the maximum EDI within 24 hours of admission had an AUROC of 0.98 compared with 0.89 of ISS and 0.91 of NISS. For unplanned ICU admission, the EDI slope within 24 hours of ICU admission had a modest performance with an AUROC of 0.66.

CONCLUSION

Epic Deterioration Index appears to perform strongly in predicting in-patient mortality similarly to ISS and NISS. In addition, it can be used to predict unplanned ICU admissions. This study helps validate the use of this real-time EHR ML-based tool, suggesting that EDI should be incorporated into the daily care of trauma patients.

LEVEL OF EVIDENCE

Prognostic, level III.

[Prediction of intensive care unit readmission for critically ill patients based on ensemble learning]

OBJECTIVE

To develop machine learning models for predicting intensive care unit (ICU) readmission using ensemble learning algorithms.

METHODS

A publicly accessible American ICU database, medical information mart for intensive care (MIMIC)-Ⅲ as the data source was used, and the patients were selected by the inclusion and exclusion criteria. A set of variables that had the predictive ability of outcome including demographics, vital signs, laboratory tests, and comorbidities of patients were extracted from the dataset. We built the ICU readmission prediction models based on ensemble learning methods including random forest, adaptive boosting (AdaBoost), and gradient boosting decision tree (GBDT), and compared the prediction performance of the machine learning models with a conventional Logistic regression model. Five-fold cross validation was used to train and validate the prediction models. Average sensitivity, positive prediction value, negative prediction value, false positive rate, false negative rate, area under the receiver operating characteristic curve (AUROC) and Brier score were used as performance measures. After constructing the prediction models, top 10 predictive variables based on importance ranking were identified by the model with the best discrimination.

RESULTS

Among these ICU readmission prediction models, GBDT (AUROC=0.858) had better performance than random forest (AUROC=0.827), and was slightly superior to AdaBoost (AUROC=0.851) in terms of AUROC. Compared with Logistic regression (AUROC=0.810), the discrimination of the three ensemble learning models was much better. The feature importance provided by GBDT showed that the top ranking variables included vital signs and laboratory tests. The patients with ICU readmission had higher mean arterial pressure, systolic blood pressure, diastolic blood pressure, and heart rate than the patients without ICU readmission. Meanwhile, the patients readmitted to ICU experienced lower urine output and higher serum creatinine. Overall, the patients having repeated admissions during their hospitalization showed worse heart function and renal function compared with the patients without ICU readmission.

CONCLUSION

The ensemble learning based ICU readmission prediction models had better performance than Logistic regression model. Such ensemble learning models have the potential to aid ICU physicians in identifying those patients with high risk of ICU readmission and thus help improve overall clinical outcomes.

A Tool to Predict Readmission to the Intensive Care Unit in Surgical Critical Care Patients-The RISC Score

BACKGROUND

Readmission to the Intensive Care Unit (ICU) is associated with a high risk of in-hospital mortality and higher health care costs. Previously published tools to predict ICU readmission in surgical ICU patients have important limitations that restrict their clinical implementation. We sought to develop a clinically intuitive score that can be implemented to predict readmission to the ICU after surgery or trauma. We designed the score to emphasize modifiable predictors.

METHODS

In this retrospective cohort study, we included surgical patients requiring critical care between June 2015 and January 2019 at Beth Israel Deaconess Medical Center, Harvard Medical School, MA, USA. We used logistic regression to fit a prognostic model for ICU readmission from a priori defined, widely available candidate predictors. The score performance was compared with existing prediction instruments.

RESULTS

Of 7,126 patients, 168 (2.4%) were readmitted to the ICU during the same hospitalization. The final score included 8 variables addressing demographical factors, surgical factors, physiological parameters, ICU treatment and the acuity of illness. The maximum score achievable was 13 points. Potentially modifiable predictors included the inability to ambulate at ICU discharge, substantial positive fluid balance (>5 liters), severe anemia (hemoglobin <7 mg/dl), hyperglycemia (>180 mg/dl), and long ICU length of stay (>5 days). The score yielded an area under the receiver operating characteristic curve of 0.78 (95% CI 0.74-0.82) and significantly outperformed previously published scores. The performance of the underlying model was confirmed by leave-one-out cross-validation.

CONCLUSION

The RISC-score is a clinically intuitive prediction instrument that helps identify surgical ICU patients at high risk for ICU readmission. The simplicity of the score facilitates its clinical implementation across surgical divisions.

Impact of Intensive Care Unit Readmissions on Patient Outcomes and the Evaluation of the National Early Warning Score to Prevent Readmissions: Literature Review

Background

Intensive care unit (ICU) readmissions have been shown to increase a patient’s in-hospital mortality and length of stay (LOS). Despite this, no methods have been set in place to prevent readmissions from occurring.

Objective

The aim of this literature review was to evaluate the impact of ICU readmission on patient outcomes and to evaluate the effect of using a risk stratification tool, the National Early Warning Score (NEWS), on ICU readmissions.

Methods

A database search was performed on PubMed, Cumulative Index of Nursing and Allied Health Literature, Google Scholar, and ProQuest. In the initial search, 2028 articles were retrieved; after inclusion and exclusion criteria were applied, 12 articles were ultimately used in this literature review.

Results

This literature review found that patients readmitted to the ICU have an increased mortality rate and LOS at the hospital. The sample sizes in the reviewed studies ranged from 158 to 745,187 patients. Readmissions were most commonly associated with respiratory issues about 18% to 59% of the time. The NEWS has been shown to detect early clinical deterioration in a patient within 24 hours of transfer, with a 95% CI of 0.89 to 0.94 (P<.001), a sensitivity of 93.6% , and a specificity of 82.2%.

Conclusions

ICU readmissions are associated with worse patient outcomes, including hospital mortality and increased LOS. Without the use of an objective screening tool, the provider has been solely responsible for the decision of patient transfer. Assessment with the NEWS could be helpful in decreasing the frequency of inappropriate transfers and ultimately ICU readmission.