VEMP testing in BPPV

Timothy C. Hain, MD • Page last modified: February 25, 2022 Return to testing index

See also: oVEMP testingVEMP testing. •BPPV


One would expect that decreased cVEMPs might be weakly correlated with BPPV, as BPPV is thought to be caused by utricular damage, and what damages the utricle might also damage the saccule. One would expect that oVEMPs would have a stronger correlation, as oVEMPs are thought to be produced by the utricle.

Few studies have been done. Fife et al (2017) stated that "No role for VEMP in the diagnosis of BPPV has been established". However, the lack of evidence is not the same as evidence of lack.

cVEMPs in BPPV -- small effect

One might think that as VEMP's assess the saccule (an otolith organ), and because BPPV is thought to be due to another otolithic disease (loose otoconia from the utricle), that there might be abnormal VEMP's in persons with BPPV.

Several authors have reported increased latency of VEMPs in patients with BPPV (Yang et al, 2008; Akkuzu et al, 2006; Eryaman et al, 2012). Korres et al (2011) found no difference. Overall, it would seem that most report slightly increased latency. As we are very dubious that latency is a useful measure at all, we are also dubious about the use of cVEMP's in BPPV. Nevertheless, there is likely an underlying mechanism for increased latency in BPPV, perhap reflecting the idea advanced above that BPPV reflects an underlying disorder of the otolithic organs.

Boleas-Aguirre et al (2007) reported a small effect on amplitude. Hong et al (2008) reported that "(24.5%) showed abnormal VEMP responses on the affected side when compared with their age-related control subgroup. " Kim et al (2015) reported abnormalities in about 20% of both cVEMP and oVEMP. We think this is reasonable.

Murofushi et al (1996) took the opposite position and reported that if VEMPs are absent in vestibular neuritis, BPPV is unlikely to develop. While somewhat reasonable, again this appears to us to be a small effect.

So to summarize the situation, it appears that there is likely a small reduction of VEMP's of any type in BPPV, presumably just correlated with the greater likelihood of ear disease in persons with BPPV. VEMP's are not useful for diagnosis of BPPV.

oVEMPs in BPPV -- slightly larger effect

The general theme of these papers is that oVEMP is better than cVEMP. This is logical as one would expect that BPPV would be correlated with utricular damage.

Hoseinabadi et al (2016) reported that abnormal cVEMP or oVEMP, quality of life is more compromised than those without abnormality. Or in other words, more abnormalities are correlated with more problems. Sounds reasonable to us.

Xu et al (2016) reported that there were far more abnormal oVEMP and cVEMP responses in patients with BPPV, and suggested that this reflected utricular dysfunction.

Zhou et al (2015) suggested that recurrent BPPV patients had more abnormal oVEMPs. Here abnormal was defined as lack of an oVEMP at all, or asymmetry greater than 29%. In our opinion, these criteria seem vulnerable to bias.

Singh and Apeksha (2015) studied 31 patients with BPPV. They stated that " The results demonstrated no significant group difference on any of the cVEMP parameters (p > 0.05). A similar trend was noticed for the latency-related parameters of oVEMP. However, the peak-to-peak amplitude was significantly smaller in the affected ears of individuals with BPPV than their unaffected ears and the ears of healthy controls (p < 0.05). " They concluded that the sensitivity and specificity of VEMP was "far superior" to cVEMP.

On the other hand, there is a literature suggesting that oVEMPs are helpful in BPPV.