Development and Validation of a Clinical Prediction Model for Sleep Disorders in the ICU: A Retrospective Cohort Study

The use of nomogram for detecting mild cognitive impairment in patients with type 2 diabetes mellitus

BACKGROUND

Type 2 diabetes mellitus (T2DM) is highly prevalent worldwide and may lead to a higher rate of cognitive dysfunction. This study aimed to develop and validate a nomogram-based model to detect mild cognitive impairment (MCI) in T2DM patients.

METHODS

Inpatients with T2DM in the endocrinology department of Xiangya Hospital were consecutively enrolled between March and December 2021. Well-qualified investigators conducted face-to-face interviews with participants to retrospectively collect sociodemographic characteristics, lifestyle factors, T2DM-related information, and history of depression and anxiety. Cognitive function was assessed using the Mini-Mental State Examination scale. A nomogram was developed to detect MCI based on the results of the multivariable logistic regression analysis. Calibration, discrimination, and clinical utility of the nomogram were subsequently evaluated by calibration plot, receiver operating characteristic curve, and decision curve analysis, respectively.

RESULTS

A total of 496 patients were included in this study. The prevalence of MCI in T2DM patients was 34.1% (95% confidence interval [CI]: 29.9%-38.3%). Age, marital status, household income, diabetes duration, diabetic retinopathy, anxiety, and depression were independently associated with MCI. Nomogram based on these factors had an area under the curve of 0.849 (95% CI: 0.815-0.883), and the threshold probability ranged from 35.0% to 85.0%.

CONCLUSIONS

Almost one in three T2DM patients suffered from MCI. The nomogram, based on age, marital status, household income, duration of diabetes, diabetic retinopathy, anxiety, and depression, achieved an optimal diagnosis of MCI. Therefore, it could provide a clinical basis for detecting MCI in T2DM patients.

Risk factors and prediction model of sleep disturbance in patients with maintenance hemodialysis: A single center study

Objectives

This study aimed to explore the risk factors and develop a prediction model of sleep disturbance in maintenance hemodialysis (MHD) patients.

Methods

In this study, 193 MHD patients were enrolled and sleep quality was assessed by Pittsburgh Sleep Quality Index. Binary logistic regression analysis was used to explore the risk factors for sleep disturbance in MHD patients, including demographic, clinical and laboratory parameters, and that a prediction model was developed on the basis of risk factors by two-way stepwise regression. The final prediction model is displayed by nomogram and verified internally by bootstrap resampling procedure.

Results

The prevalence of sleep disturbance and severe sleep disturbance in MHD patients was 63.73 and 26.42%, respectively. Independent risk factors for sleep disturbance in MHD patients included higher 0.1^*age (OR = 1.476, 95% CI: 1.103–1.975, P = 0.009), lower albumin (OR = 0.863, 95% CI: 0.771–0.965, P = 0.010), and lower 10^*calcium levels (OR = 0.747, 95% CI: 0.615–0.907, P = 0.003). In addition, higher 0.1^*age, lower albumin levels, and anxiety were independently associated with severe sleep disturbance in MHD patients. A risk prediction model of sleep disturbance in MHD patients showed that the concordance index after calibration is 0.736, and the calibration curve is approximately distributed along the reference line.

Conclusions

Older age, lower albumin and calcium levels are higher risk factors of sleep disturbance in MHD, and the prediction model for the assessment of sleep disturbance in MHD patients has excellent discrimination and calibration.

Development and validation of prediction models for neurocognitive disorders in adult patients admitted to the ICU with sleep disturbance

BACKGROUND

Neurocognitive disorders (NCDs) and sleep disturbance are highly prevalent in the perioperative period and intensive care unit (ICU). There has been a lack of individualized evaluation tools designed for the high-risk NCDs in critically ill patients with sleep disturbance.

OBJECTIVES

The aim of this study was to develop and validate prediction models for NCDs among adult patients with sleep disturbance.

METHODS

The R software was used to analyze the dataset of adult patients admitted to the ICU with sleep disturbance, who were diagnosed following the codes of the International Classification of Diseases, 9th Revision (ICD-9) and 10th Revision (ICD-10) using the MIMIC-IV database. We used logistic regression and LASSO analyses to identify important risk factors associated with NCDs and develop nomograms for NCDs predictions. We measured the performances of the nomograms using the bootstrap resampling procedure, sensitivity, specificity of the receiver operating characteristic (ROC), area under the ROC curves (AUC), and decision curve analysis (DCA).

RESULTS

The prediction models shared the 10 risk factors (age, gender, midazolam, morphine, glucose, diabetes diseases, potassium, international normalized ratio, partial thromboplastin time, and respiratory rate). Cardiovascular diseases were included in the logistic regression, the sensitivity was 74.1%, and specificity was 64.6%. When platelet and Glasgow Coma Score (GCS) were included and cardiovascular diseases were removed in the LASSO prediction model, the sensitivity was 86.1% and specificity was 82.8%. Discriminative abilities of the logistic prediction and LASSO prediction models for NCDs in the validation set were evaluated as the AUC scores, which were 0.730 (95% CI 0.716-0.743) and 0.920 (95% CI 0.912-0.927). Net benefits of the prediction models were observed at threshold probabilities of 0.567 and 0.914.

CONCLUSIONS

The LASSO prediction model showed better performance than the logistic prediction model and should be preferred for nomogram-assisted decisions on clinical risk management of NCDs among adult patients with sleep disturbance in the ICU.

Development and Validation of Nomogram Prediction Model for Postoperative Sleep Disturbance in Patients Undergoing Non-Cardiac Surgery: A Prospective Cohort Study

PURPOSE

To develop a risk prediction nomogram of postoperative sleep disturbance (PSD) in patients undergoing non-cardiac surgery.

PATIENTS AND METHODS

Data on 881 consecutive patients who underwent non-cardiac surgery at the Affiliated Hospital of Xuzhou Medical University between June 2020 and April 2021 were prospectively collected. Of these, we randomly divided 881 non-cardiac patients into two groups, training cohort (n = 617) and validation cohort (n = 264) at the ratio of 7:3. Characteristic variables were selected based on the data of training cohort through least absolute shrinkage and selection operator (LASSO) regression. Multivariate logistic regression was used to identify the independent risk factors associated with PSD that then were incorporated into the nomogram. The predictive performance of the nomogram was measured by concordance index (C index), receiver operating characteristic (ROC) curve, and calibration with 1000 bootstrap samples to decrease the over-fit bias.

RESULTS

PSD was found in 443 of 617 patients (71.8%) and 190 of 264 patients (72.0%) in the training and validation cohorts, respectively. The perioperative risk factors associated with PSD were female sex, anxiety, dissatisfaction of ward environment, absence of combined regional nerve block, postoperative nausea and vomiting (PONV), the longer duration stayed in post anesthesia care unit (PACU), the higher dose of midazolam and sufentanil, the higher postoperative numeric rating score for pain (NRS) score. Incorporating these 9 factors, the nomogram achieved good concordance indexes of 0.82 (95% confidence interval [CI], 0.78-0.85) and 0.80 (95% CI, 0.74-0.85) in predicting PSD in the training and validation cohorts, respectively, and obtained well-fitted calibration curves. The sensitivity and specificity (95% CIs) of the nomogram were calculated, resulting in sensitivity of 74.0% (70.0-78.2%) and 75.3% (68.4-81.7%) and specificity of 79.3% (72.5-85.2%) and 70.3% (58.4-80.7%) for the training and validation cohorts, respectively. Patients who had a nomogram score of less than 262 or 262 or greater were considered to have low or high risks of PSD presence, respectively.

CONCLUSION

The proposed nomogram achieved an optimal prediction of PSD in patients undergoing non-cardiac surgery. The risks for an individual patient to harbor PSD can be determined by this model, which can lead to a reasonable preventive and treatment measures.