Prediction Models for Postoperative Delirium of Cardiovascular Surgery (PODOCVS): Protocol for a Systematic Review
Bibliographic record
Abstract
BACKGROUND: Postoperative delirium of cardiovascular surgery (PODOCVS) is an acute brain dysfunction characterized by inattention, impaired consciousness, and cognitive disorders, and the severity and presence of these symptoms fluctuate over time. PODOCVS occurs during the early postoperative period and is associated with adverse outcomes, including prolonged mechanical ventilation, premature mortality, and so on. Advances in its early diagnosis and treatment have mitigated some of the initial adverse effects of PODOCVS, but models for predicting risk in patients who have already developed PODOCVS remain inadequate for effective secondary prevention. Developing multivariable prediction models for stratifying PODOCVS risk would enable early, personalized interventions. OBJECTIVE: This study aims to systematically review and critically evaluate the development, performance, and applicability of existing prediction models for PODOCVS. METHODS: An extensive systematic search will be performed across multiple databases, including Embase, PubMed, the Web of Science, and so on, to identify studies related to multivariate predictive models for PODOCVS. A manual search of the included studies' reference lists will also be conducted to identify any additional relevant publications. This systematic review will include studies that meet the following criteria: (1) studies with subject populations comprising adult cardiovascular surgery patients aged ≥18 years, (2) studies involving the development and internal or external validation of predictive models for PODOCVS via multivariate analysis, and (3) studies with outcome measures focused on postoperative delirium. Two researchers (ZXL and WYK) will independently extract the data and assess the included studies' model quality using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) checklist and the Predictive Model Bias Risk Assessment Tool (PROBAST). Since this study will not involve patient data, ethics approval is not required. Our findings will be published in a peer-reviewed scientific journal and the dataset will be made freely available. RESULTS: Literature searches were conducted from the inception of the database to May 20, 2024 (updated up to January 31, 2025), and data extraction and analysis are expected to be complete by the end of May 2025. We currently have a preliminary plan to publish the complete study results by August 2025, subject to any unforeseen delays or changes in the research timeline. CONCLUSIONS: We present a protocol for the systematic review of prediction models for postoperative delirium in cardiac surgery patients. Aiming to identify, summarize, and critically appraise existing risk models globally, this review seeks to provide an up-to-date reference for stakeholders involved in patients with cardiac surgery care, policy making, and research. In addition, we aim to investigate whether machine learning models for PODOCVS offer more accurate predictions than traditional statistical models. TRIAL REGISTRATION: PROSPERO CRD42024578957; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024578957. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/75368.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.044 | 0.062 |
| Meta-epidemiology (narrow) | 0.005 | 0.004 |
| Meta-epidemiology (broad) | 0.015 | 0.021 |
| Bibliometrics | 0.011 | 0.010 |
| Science and technology studies | 0.003 | 0.003 |
| Scholarly communication | 0.006 | 0.007 |
| Open science | 0.005 | 0.005 |
| Research integrity | 0.005 | 0.005 |
| Insufficient payload (model declined to judge) | 0.059 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".