Prognostic Performance of Existing Scoring Systems among Critically Ill Patients Requiring Continuous Renal Replacement Therapy: An Observational Study

Interpretations of the Role of Plasma Albumin in Prognostic Indices: A Literature Review

This review assesses how publications interpret factors that influence the serum or plasma albumin (PA) level in prognostic indices, focusing on inflammation and nutrition. On PubMed, a search for "albumin AND prognosis" yielded 23,919 results. From these records, prognostic indices were retrieved, and their names were used as search strings on PubMed. Indices found in 10 or more original research articles were included. The same search strings, restricted to "Review" or "Systematic review", retrieved yielded on the indices. The data comprised the 10 latest original research articles and up to 10 of the latest reviews. Thirty indices had 294 original research articles (6 covering two indices) and 131 reviews, most of which were from recent years. A total of 106 articles related the PA level to inflammation, and 136 related the PA level to nutrition. For the reviews, the equivalent numbers were 54 and 65. In conclusion, more publications mention the PA level as a marker of nutrition rather than inflammation. This is in contrast to several general reviews on albumin and nutritional guidelines, which state that the PA level is a marker of inflammation but not nutrition. Hypoalbuminemia should prompt clinicians to focus on the inflammatory aspects in their patients.

Nursing Countermeasures of Continuous Renal Replacement Treatment in End-Stage Renal Disease with Refractory Hypotension in the Context of Smart Health

This work is aimed at exploring the nursing strategies and effects of continuous renal replacement therapy (CRRT) for end-stage renal disease (ESRD) with refractory hypotension under the background of smart health. 40 ESRD patients with refractory hypotension who received CRRT treatment were enrolled as the research objects and were randomly rolled into the intervention group and the control group, with 20 cases in each group. Patients in the control group received routine nursing, and those in the intervention group received individualized nursing. The incidence of hypotension, dry body weight, serous cavity effusion, renal function indicators (blood urea nitrogen (BUN) and creatinine (Cre)), and patient satisfaction were compared between the two groups. The results showed that the probability of hypotension in the intervention group was 9.38%, which was lower than that in the control group (34.38%). The probability of early termination of dialysis in the intervention group was 0%, which was lower than that in the control group (18.75%), and the difference was statistically significant (P < 0.05). The decreases of BUN and Cre in the intervention group were significantly greater than those in the control group, and the differences were statistically significant (P < 0.05). The proportion of water growth less than 10% during dialysis in the intervention group was 98.44%, which was greater than that in the control group (93.45%), and the difference was statistically significant (P < 0.05). The ultrafiltration volume after dialysis in the intervention group was 2850 ± 400 mL, which was greater than that in the control group 2350 ± 350 mL. After intervention, the proportion of patients with pleural effusion in the intervention group was 10% less than that in the control group (20%), and the difference was statistically significant (P < 0.05). The satisfaction rate of the intervention group was 97.66%, which was higher than that of the control group (65.63%). In conclusion, individualized nursing was more helpful to the recovery of ESRD patients with refractory hypotension treated with CRRT than routine nursing.

Explainable Machine Learning-Based Risk Prediction Model for In-Hospital Mortality after Continuous Renal Replacement Therapy Initiation

In this study, we established an explainable and personalized risk prediction model for in-hospital mortality after continuous renal replacement therapy (CRRT) initiation. This retrospective cohort study was conducted at Changhua Christian Hospital (CCH). A total of 2932 consecutive intensive care unit patients receiving CRRT between 1 January 2010, and 30 April 2021, were identified from the CCH Clinical Research Database and were included in this study. The recursive feature elimination method with 10-fold cross-validation was used and repeated five times to select the optimal subset of features for the development of machine learning (ML) models to predict in-hospital mortality after CRRT initiation. An explainable approach based on ML and the SHapley Additive exPlanation (SHAP) and a local explanation method were used to evaluate the risk of in-hospital mortality and help clinicians understand the results of ML models. The extreme gradient boosting and gradient boosting machine models exhibited a higher discrimination ability (area under curve [AUC] = 0.806, 95% CI = 0.770-0.843 and AUC = 0.823, 95% CI = 0.788-0.858, respectively). The SHAP model revealed that the Acute Physiology and Chronic Health Evaluation II score, albumin level, and the timing of CRRT initiation were the most crucial features, followed by age, potassium and creatinine levels, SPO2, mean arterial pressure, international normalized ratio, and vasopressor support use. ML models combined with SHAP and local interpretation can provide the visual interpretation of individual risk predictions, which can help clinicians understand the effect of critical features and make informed decisions for preventing in-hospital deaths.