What is the diagnostic accuracy of red flags related to cauda equina syndrome (CES), when compared to Magnetic Resonance Imaging (MRI)? A systematic review

Abstract Objective To review and statistically pool available evidence on the diagnostic accuracy of red flags to clinically identify MRI confirmed Cauda Equina Syndrome (CES). Study design Systematic review. Data sources Embase, Scopus, Ovid Medline, Ovid Healthstar, Amed and CINAHL from inception to January 30, 2018 and a grey literature search. Inclusion criteria Primary diagnostic studies, published in English; comparing red flags for CES; to Magnetic Resonance Imaging (MRI) as reference standard; in humans; older than 18 years. Methods Data extraction, assessment of study quality using a modified QUADAS-2 tool and the use of GRADE to synthesize the results for each test was performed by three independent assessors. Diagnostic accuracy statistics applied to the identified data andpooled analysis performed using Meta-DiSc, version 1.4. Moderator analyses planned for pooled results. Results Seven studies (total N = 569 participants) were included. Potential signs or symptoms of CES were compared to MRI findings. Diagnostic data could be pooled for reduced anal tone, leg pain, back pain, saddle anaesthesia, urinary retention, urinary incontinence and bowel incontinence from six of seven studies. The pooled sensitivity for the signs and symptoms ranged from 0.19 (95% CI 0.09 to 0.33) to 0.43 (95% CI 0.30 to 0.56) while the pooled specificity ranged from 0.62 (95% CI 0.59 to 0.73) to 0.88 (95% CI 0.85 to 0.92). Conclusion Red flags used to identify potential CES appear to be more specific than sensitive. As such, when these are present, they should be considered justification for prompt diagnostic workup.