From obstruction to ischemia: A systematic review and meta-analysis on the diagnostic accuracy of CT scans in identifying small and large bowel obstruction, underlying causes, and predicting critical complications in adults

• Quantify CT’s pooled sensitivity, specificity, and diagnostic odds ratio (DOR) for the diagnosis of bowel obstruction and its complications.
• Compare diagnostic performance across CT technologies, including multidetector CT (MDCT), helical CT, and dual-energy CT (DECT).
• Assess sources of heterogeneity, such as study design, imaging protocol, reference standards, and population characteristics.
• Identify methodological gaps, variability in diagnostic thresholds, and reporting limitations that hinder clinical guideline standardization.
• Propose a framework for improving diagnostic algorithms and reporting consistency in future radiological research.

Main Research Question

1. How does CT performance differ between small bowel and large bowel obstruction?
2. What CT signs or parameters most reliably predict ischemia or perforation?
3. How do diagnostic metrics vary according to CT modality (e.g., MDCT vs. helical)?
4. What are the key methodological sources of bias or heterogeneity among published studies?

Keywords and Synonyms

Source Selection Criteria (Databases)

• Primary Databases
• Ovid MEDLINE (1946 – February 2025)

• Embase (1946 – February 2025)

• Supplementary Searches
• Manual screening of reference lists from previous systematic reviews and meta-analyses on CT for bowel obstruction.

• Search of grey literature (theses, clinical reports, or institutional publications) when full-text data are available and meet eligibility standards.

• Screening of abstracts from relevant radiology and surgery conferences (e.g., RSNA, ECR, EAES).

• Search Strategy Design
• The search will combine MeSH terms and free-text keywords, including: “intestinal obstruction,” “small bowel obstruction,” “large bowel obstruction,” “computed tomography,” “CT scan,” “diagnostic imaging,” “sensitivity,” “specificity,” and “ischemia.”

• Boolean operators (AND/OR), truncations, and proximity searches will be used to optimize retrieval.

• The search strategy will be peer-reviewed by an experienced medical librarian prior to execution.

Study Selection Criteria

• Original, peer-reviewed research articles.
• Study designs: randomized controlled trials (RCTs), cohort studies, case-control studies, or cross-sectional diagnostic accuracy studies.
• Adult participants (≥18 years old) with clinically or radiologically suspected bowel obstruction.
• Studies evaluating CT (contrast-enhanced or non-contrast) as the index test for diagnosis.
• Studies reporting at least one quantitative diagnostic metric: sensitivity, specificity, positive/negative predictive values, likelihood ratios, or ROC analysis.
• Full text available in English.

Exclusion Criteria

• Studies published before 2010 (to reflect modern CT technology and diagnostic standards).
• Case reports, editorials, review articles, conference abstracts, and letters without full datasets.
• Paediatric populations (<18 years) or animal models.
• Studies focused exclusively on trauma, FAST/eFAST, MRI, or ultrasound imaging.
• Studies assessing AI algorithms or radiomics models rather than conventional CT accuracy.
• Studies limited to internal hernias without obstruction, gastric volvulus, or gastric outlet obstruction.
• Articles reporting CT findings as predictors of surgery rather than diagnostic indicators of obstruction or ischemia.

Quality Assessment Form

1. Patient Selection – Was the sampling method representative and free of inappropriate exclusions?
2. Index Test – Were CT diagnostic thresholds defined and interpreted independently of the reference standard?
3. Reference Standard – Was the comparator (e.g., surgery, contrast imaging) likely to correctly classify the target condition?
4. Flow and Timing – Was the time interval between CT and reference test appropriate, and were all patients included in the analysis?

Risk of Bias Assessment Form

Purpose

Risk Classification

• Low Risk – minimal concern for bias or confounding.
• Moderate Risk – potential bias that could influence diagnostic estimates but not invalidate results.
• High Risk – major methodological flaws that threaten the validity or reliability of conclusions.

Eligibility and Data Extraction Form

Purpose

Planned Data Synthesis

• Primary Analyses:
• Pooled sensitivity and specificity for CT in diagnosing SBO, LBO, and associated complications.

• Summary receiver operating characteristic (SROC) curve construction.

• Heterogeneity Analysis:
• I² and τ² statistics to quantify heterogeneity.

• Subgroup analyses by CT modality (MDCT vs helical vs DECT), bowel region (small vs large), and study design (prospective vs retrospective).

• Publication Bias:
• Evaluated using Deeks’ funnel-plot asymmetry test and visual inspection.

• Software:
• Statistical analyses will be conducted using OpenMetaAnalyzer and IBM SPSS Statistics v29.

Expected Outcomes

• Provide the most comprehensive synthesis of diagnostic accuracy metrics for CT in bowel obstruction to date.
• Quantitatively confirm CT’s superiority over traditional modalities and delineate its diagnostic limitations in ischemia detection.
• Identify heterogeneity in study design, reference standards, and CT techniques.
• Establish evidence-based recommendations for standardized CT interpretation criteria and reporting practicesin diagnostic research.
• Highlight future research directions, including the role of AI, radiomics, and dual-energy imaging in improving diagnostic precision.

Ethics and Dissemination

• Peer-reviewed journal publication in a radiology or surgery outlet.
• International conference presentations (e.g., RSNA, EAES, ECR).
• Open-access sharing of extracted datasets (upon request) for reproducibility and secondary analysis.

Funding and Competing Interests

Summary of Expected Results