Published in:
01-11-2021 | Magnetic Resonance Imaging | Hollow Organ GI
Abbreviated MRI in patients with suspected acute appendicitis in emergency: a prospective study
Published in: Abdominal Radiology | Issue 11/2021
Login to get accessAbstract
Purpose
To determine the diagnostic performance of an abbreviated non-contrast MRI protocol in diagnosing acute appendicitis.
Methods
Prospectively, a total of 67 consenting consecutive patients with clinical suspicion of acute appendicitis (Alvarado score ≥ 5) were evaluated with an abbreviated three-sequence non-contrast MRI protocol (axial T2WI, coronal T2WI, axial DWI) at a single tertiary care center. MRI was interpreted by two radiologists blinded to the clinical details, other investigations, and outcome of the patients. Diagnostic performance of MRI was determined using either histopathological examination (HPE) results as the reference standard in surgical cases (n = 39), or final clinical diagnosis at discharge and 3-months follow-up in non-operatively managed cases (n = 28).
Results
Sixty-seven patients comprising 42 males, 25 females including 1 pregnant patient were enrolled (median age 24 years; age range 6–70 years). The median acquisition duration of the MRI protocol was 12.5 min. In the analysis of the complete cohort including both surgical and non-operatively managed cases (n = 67), MRI showed sensitivity of 93.3% (95% CI 81.7–98.6%), specificity of 86.4% (95% CI 65.1–97.1%), and diagnostic accuracy of 91.0% (95% CI 81.5–96.6%) (p < 0.001). In the subset of surgical cases with HPE as the reference standard (n = 39), MRI showed sensitivity of 97.1% (95% CI 84.7–99.9%), specificity of 100% (95% CI 47.8–100%), and diagnostic accuracy of 98% (95% CI 87.5–100%) (p < 0.001).
Conclusion
MRI may be performed to diagnose acute appendicitis or alternative causes of right iliac fossa pain. An abbreviated MRI protocol consisting of only three sequences without IV contrast, patient preparation, or antiperistaltic agents could shorten the examination duration while retaining diagnostic accuracy.
Graphic abstract
×
![]()