Clinical and video head impulse test in the diagnosis of posterior circulation stroke presenting as acute vestibular syndrome in the emergency department.

INTRODUCTION: Head impulse test (HIT) is the critical bedside examination which differentiates vestibular neuritis (VN) from posterior circulation stroke (PCS) in acute vestibular syndrome (AVS). Video-oculography based HIT (vHIT) may have aadditional strength in making the differentiation.

METHODS: Patients admitted to the emergency department of a tertiary-care medical center with AVS were studied. An emergency specialist and a neurologist performed HIT. vHIT was conducted by an neuro-otology research fellow.

RESULTS: Forty patients 26 male, 14 female with a mean age of 49 years were included in the analyses. Final diagnoses were VN in 24 and PCS in 16 patients.In the VN group, clinical HIT was assessed as abnormal in 19(80%) cases by the emergency specialist and in 20(83%) by the neurologist. In all PCS patients, HIT was recorded as normal both by the emergency specialist and the neurologist (100%).On vHIT, patients with VN had significantly low gain values for both the ipsilesional and contralesional sides when compared with the healthy controls, with significantly lower figures for the ipsilesional side (p < 0.001). All patients in this group had normal DWI-MRI.PCS patients had bilaterally low gain (p < 0.05) on vHIT. However, gain asymmetry was not significant. Subgroup analyses according to presence of brainstem involvement revealed bilateral low gain (p < 0.05) in patients with brainstem infarction (anterior inferior cerebellar artery-posterior inferior cerebellar artery stroke, AICA-PICA stroke) whereas patients with pure cerebellar infarction (posterior inferior cerebellar artery-superior cerebellar artery stroke, PICA-SCA stroke) had gain values similar to healthy controls.With a gain cut-off ≤0.75 and gain asymmetry cut-off ≥17%, as determined by ROC analysis, 100% of PCS patients and 80% of VN patients were correctly diagnosed.

CONCLUSIONS: Clinical HIT, either performed by an emergency specialist or neurologist is equivalent to vHIT gain and gain asymmetry analysis as conducted by neuro-otologist in the diagnosis of PCS, albeit mislabeling about 20% of VN patients. vHIT does not appear to yield additional diagnostic information. These findings indicate the strength of clinical HIT. Pure gain-based vHIT analysis seems limited and needs to be incorporated with saccade analysis.