Objective Cauda equina syndrome (CES) is a neurosurgical emergency, identified by a range of clinical symptoms including back pain coupled with saddle anaesthesia, bladder or bowel symptoms. A high index of suspicion for CES warrants an emergency lumbar MRI. However, up to 48% of patients undergoing such scans have no radiological correlate for their symptoms but are clinically indistinguishable from those with established CES (Rooney et al, 2009, J Neurol). Relatively little is known about this scan-negative patient group. We therefore examined a cohort of “CES” patients from a single regional centre and examined radiological and referral outcomes.
Method 179 patients who had a lumbar MRI at Kings College Hospital, London, after attending A and E in a 12 month period ending December 2015 were retrospectively identified. 160 patients met the inclusion criteria: over 18; had lumbar MRI performed specifically to exclude suspected CES; and had an admission entry in the medical notes detailing their history. Electronic patient records were then assessed for age, gender, presenting symptoms and follow up.
Results CES was confirmed in 16% of lumbar MRI scans (n=25), 84% (n=135) were ‘scan-negative’ with no structural correlate for CES although 44% showed some evidence of root involvement. Fisher’s exact test (two-sided) was used to compare the groups’ symptomology. No significant differences were found for back pain (p=0.59), urinary incontinence (p=0.98), urinary retention (p=0.16), altered urinary sensation (p=0.57), bowel incontinence (p=0.80), leg weakness (p=0.67), sciatica (p=0.57), leg paresthesia (p=0.37). There was an observed difference in the demographics: the scan-negative group was predominantly female [71% vs 48% (p=0.023)] and significantly younger [42.7 vs 54.7 years (p<0.005)]. 90% of the confirmed CES underwent surgical intervention. Follow up for the scan-negative group was variable and included surgery (22%, n=30), neurosurgery clinic (16%, n=22), neurology clinic (13%, n=17), physiotherapy (24%, n=33), pain clinic (16% n=22), psychiatry (4%, n=5), GP (22% n=30) and no follow up (13%, n=18).
Conclusion These findings replicate established difficulties in identifying CES from clinical symptoms alone. Demographic differences between scan-positive and negative patients indicate heterogeneity requiring further investigation. It is likely that a proportion of the scan negative patients have a functional neurological disorder although the precise figure has yet to be determined. Routine follow up was inconsistent. More detailed neuropsychiatric evaluation and longer follow-up are required to characterise this patient group which would inform a range of clinical care pathways addressing their medical and psychiatric needs.