Introduction Renal pelvis contrast opacification (RPO) in CT myelography has been proposed as a specific sign of CSF leak in spontaneous intracranial hypotension (SIH). However, published studies have been limited in size and differences in prevalence of this sign in distinct CSF leak aetiologies remain of uncertain significance.

Aim of Study This study aimed to re-evaluate the significance of RPO in SIH, and investigate any correlation between RPO, time from contrast administration to identification of RPO, and aetiology of CSF leak.

Methods All CT myelography during a 4 year period at a single neuroscience centre was investigated. Time of intrathecal contrast injection, time of first evidence of RPO, contemporaneous renal function, and aetiology of CSF leak (if identified) were recorded and analysed.

Results 169 studies were included. 26 studies were for non-SIH indications – none demonstrated RPO. 46/143 SIH studies were positive for RPO: 52% were dural CSF leaks and 39% were CSF-venous fistulae. The remaining 9% had no aetiology identified but either had confirmed low intracranial pressure or responded to treatment. In the 97 RPO negative SIH studies, 54% had no CSF leak aetiology identified. 41% were dural leaks and only 5% were CSF-venous fistulae. No significant correlation between time of first evidence of RPO and CSF leak aetiology was identified.

Conclusion This study confirms renal pelvis opacification is a highly specific marker of CSF leak with a 100% positive predictive value for SIH. The findings suggest the sign is more sensitive for CSF-venous fistulae than dural CSF leaks.

Disclosure of Interest Nothing to disclose