Qi Li

Background: Delirium is a prevalent phenomenon among patients admitted to the geriatric intensive care unit (ICU) and can adversely impact prognosis and augment the risk of complications. Objective: We aimed to construct and validate a predictive model for postoperative delirium state in older ICU patients, providing timely and effective early identification of high-risk individuals and assisting clinicians in decision-making. Methods: The data from patients admitted to the ICU for over 24 hours were extracted from the Medical Information Marketplace for Intensive Care IV (MIMIC-IV) database and the eICU Collaborative Research Database (eICU-CRD). The MIMIC-IV data were split (7:3) into training and internal validation sets, while the eICU-CRD data served as an external validation set. Delirium predictions were conducted for the subsequent prediction windows (12 h, 24 h, 48 h, and whole stay time) using data from the first 24 hours post admission. The corresponding feature variables were subjected to Boruta feature selection, and the prediction models were constructed using logistic regression, support vector classifier, random forest classifier, and extreme gradient boosting (XGB). Subsequently, model performance was evaluated using areas under the receiver operating characteristic curves (AUCs), Brier scores, and decision curve analysis, and external validation. Results: The MIMIC-IV and eICU-CRD datasets comprised 6129 and 709 patients, respectively, who were included in the analysis. Fifty-four features were selected to construct the predictive model. Regarding internal validation, the XGB model demonstrated the most effective prediction of delirium across different prediction windows. The AUCs for the 4 prediction windows (12 h, 24 h, 48 h, and whole stay time) were 0.848 (95% CI 0.826-0.869), 0.852 (95% CI 0.831-0.872), 0.851 (95% CI 0.831-0.871), and 0.844 (95% CI 0.823-0.863), respectively, and those of the external validation set were 0.777 (95% CI 0.726-0.825), 0.761 (95% CI 0.710-0.808), 0.753 (95% CI 0.704-0.798), and 0.737 (95% CI 0.695-0.777), respectively. Furthermore, the XGB model demonstrated the most accurate calibration across all prediction windows, with values of 0.129, 0.136, 0.144, and 0.148, respectively. Additionally, decision curve analysis revealed that the XGB model outperformed the other models in terms of net gain for the majority of threshold probability values. The 6 most significant predictive features identified were the first day’s delirium assessment results, type of first care unit, minimum Glasgow Coma Scale (GCS) score, Acute Physiology Score III, acetaminophen, and nonsteroidal anti-inflammatory drugs. Conclusions: The high-performance XGB model for predicting postoperative delirium state in older adult ICU patients has been successfully developed and validated. The model predicts the delirium state at 12 h, 24 h, 48 h, and whole stay time after the first day of hospitalization within the ICU. This enables physicians to identify high-risk patients early, thus facilitating the optimization of personalized management strategies and care plans.

BACKGROUND Delirium is a prevalent phenomenon among patients admitted to the geriatric intensive care unit (ICU) and can adversely impact prognosis and augment the risk of complications. OBJECTIVE We aimed to construct and validate a predictive model for postoperative delirium state in older ICU patients, providing timely and effective early identification of high-risk individuals and assisting clinicians in decision-making. METHODS The data from patients admitted to the ICU for over 24 hours were extracted from the Medical Information Marketplace for Intensive Care IV (MIMIC-IV) database and the eICU Collaborative Research Database (eICU-CRD). The MIMIC-IV data were split (7:3) into training and internal validation sets, while the eICU-CRD data served as an external validation set. Delirium predictions were conducted for the subsequent prediction windows (12 h, 24 h, 48 h, and whole stay time) using data from the first 24 hours post admission. The corresponding feature variables were subjected to Boruta feature selection, and the prediction models were constructed using logistic regression, support vector classifier, random forest classifier, and extreme gradient boosting (XGB). Subsequently, model performance was evaluated using areas under the receiver operating characteristic curves (AUCs), Brier scores, and decision curve analysis, and external validation. RESULTS The MIMIC-IV and eICU-CRD datasets comprised 6129 and 709 patients, respectively, who were included in the analysis. Fifty-four features were selected to construct the predictive model. Regarding internal validation, the XGB model demonstrated the most effective prediction of delirium across different prediction windows. The AUCs for the 4 prediction windows (12 h, 24 h, 48 h, and whole stay time) were 0.848 (95% CI 0.826-0.869), 0.852 (95% CI 0.831-0.872), 0.851 (95% CI 0.831-0.871), and 0.844 (95% CI 0.823-0.863), respectively, and those of the external validation set were 0.777 (95% CI 0.726-0.825), 0.761 (95% CI 0.710-0.808), 0.753 (95% CI 0.704-0.798), and 0.737 (95% CI 0.695-0.777), respectively. Furthermore, the XGB model demonstrated the most accurate calibration across all prediction windows, with values of 0.129, 0.136, 0.144, and 0.148, respectively. Additionally, decision curve analysis revealed that the XGB model outperformed the other models in terms of net gain for the majority of threshold probability values. The 6 most significant predictive features identified were the first day’s delirium assessment results, type of first care unit, minimum Glasgow Coma Scale (GCS) score, Acute Physiology Score III, acetaminophen, and nonsteroidal anti-inflammatory drugs. CONCLUSIONS The high-performance XGB model for predicting postoperative delirium state in older adult ICU patients has been successfully developed and validated. The model predicts the delirium state at 12 h, 24 h, 48 h, and whole stay time after the first day of hospitalization within the ICU. This enables physicians to identify high-risk patients early, thus facilitating the optimization of personalized management strategies and care plans.

To investigate the method and effectiveness of transosseous suture in situ technique in repairing anterior cruciate ligament (ACL) avulsion injury for the multiple ligament injuries with knee dislocation (MLIKD). The clinical data of 27 patients (27 knees) with MLIKD between September 2010 and April 2016 were analyzed retrospectively. There were 21 males and 6 females, with an average age of 42 years (range, 24-60 years). The injury was caused by traffic accident in 9 cases, heavy-weight crushing in 9 cases, sports sprain in 6 cases, falling from height in 3 cases. The interval from injury to operation was 1-19 days (mean,10.8 days). There were 20 cases of femoral avulsion injury of ACL, 7 cases of tibial avulsion injury of ACL, and there were 17 cases of posterior cruciate ligament (PCL) injuries. According to the Schenck classification, there were 15 cases of KD-Ⅲ-M type, 8 cases of KD-Ⅲ-L type, and 4 cases of KD-Ⅳ type. All patients were positive in the posterior drawer test and Lachman test; 8 cases were degree Ⅲ positive in varus stress test, and 15 cases were degree Ⅲ positive in valgus stress test. The Lysholm score of knee was 27.6±6.5, the International Knee Documentation Committee (IKDC) score was 25.5±6.2, and the range of motion (ROM) of knee was (45.1±10.2)°. The injured PCL was reconstructed with a single bundle of autologous hamstring tendon. ACL was repaired with double bundle traction by transosseous suture in situ technique. Medial cruciate ligament, lateral cruciate ligament, joint capsule, and other damaged structures were repaired at the same time. All incisions healed by first intention. There were 3 cases with joint effusion and 3 cases with incomplete flexion. All patients were followed up 12-36 months (mean, 22 months). The X-ray films showed good stability in all directions. At last follow-up, the anterior and posterior drawer tests were all negative; Lachman test was degreeⅠpositive in 4 cases, valgus stress test was degreeⅠpositive in 3 cases, varus stress test was degreeⅠpositive in 1 case; and all tests were negative in the rest patients. At 1 year after operation, the ROM of knee was (119.3 ±12.6)°, Lysholm score was 87.2±6.3, and IKDC score was 87.9±6.3, showing significant differences when compared with the preoperative scores ( P<0.05). Transosseous suture in situ technique can be used to repair the ACL avulsion injury for MLIKD, which can significantly improve the stability, mobility and function of the knee joint, and obtain satisfied short-term effectiveness.

The aim of this study was to prospectively evaluate the long-term efficacy and tolerability of hyaluronic acid (HA) for the treatment of knee pain due to Kashin–Beck disease (KBD). A total of 113 patients with KBD-based knee pain were treated with a 3-week course of HA. Clinical assessments were performed for each patient at 0 (baseline), 1, 2, 4, 8, 12, 24, and 52 weeks. The primary efficacy measure was the visual analog scale (VAS) pain score. The secondary efficacy measures included the WOMAC A (pain), B (stiffness), and C (function) scores; the total WOMAC score; and the global assessments by patients and physicians. Tolerability was evaluated based on adverse events (AEs) and physician reporting. The VAS was significantly reduced within the first 4 weeks of treatment, and the reduction was maintained over 52 weeks (p < 0.001 at each endpoint). These data were supported by the secondary variables WOMAC A (all p < 0.001), WOMAC B (p = 0.002, 0.003, and 0.019, respectively), WOMAC C (all p < 0.001), total WOMAC (all p < 0.001), and the global assessments by patients and physicians at weeks 12, 24, and 52. No serious AEs were reported, and the overall incidence of AEs was 10.6 %. This study suggests that the intra-articular injection of HA is effective and well tolerated for the treatment of knee pain due to KBD as HA therapy resulted in an improvement of symptoms for at least 52 weeks. Additional randomized double-blind studies are needed to confirm our findings.