Objective:To describe clinical features relevant to diagnosis, mechanism and aetiology in patients with ‘scan-negative’ cauda equina syndrome (CES).Methods:We carried out a prospective study of consecutive patients presenting with the clinical features of CES to a regional neurosurgery centre comprising semi-structured interview and questionnaires investigating presenting symptoms, neurological examination, psychiatric and functional disorder comorbidity, bladder/bowel/sexual function, distress and disability.Results:198 patients presented consecutively over 28 months. 47 were diagnosed with ‘scan-positive’ CES (mean age 48yrs, 43% female). 76 ‘mixed’ category patients had nerve root compression/displacement without CES compression, (mean age 46yrs, 71% female) and 61 patients had ‘scan-negative’ CES (mean age 40yrs, 77% female). An alternative neurological cause of CES emerged in 14/198 patients during admission and 4/151 patients with mean duration 25 months follow up.Patients with ‘scan-negative’ CES had more positive clinical signs of a functional neurological disorder (11%‘scan positive’ CES v. 34%mixed and 68%‘scan-negative’, p<0.0001), were more likely to describe their current back pain as ‘worst ever’ (41% vs. 46% and 70%, p=0.005) and have symptoms of a panic attack at onset (37% vs. 57% and 70%, p=0.001). Patients with ‘scan positive’ CES were more likely to have reduced/absent bilateral ankle jerks (78% ‘vs. 30% and 12%, p=<0.0001). There was no significant difference between groups in the frequency of reduced anal tone and urinary retention.Conclusions:The first well phenotyped, prospective study of ‘scan-negative’ CES supports a model in which acute pain, medication, and mechanisms overlapping with functional neurological disorder may be relevant.