Subclinical vestibular dysfunction in type 1 diabetes mellitus

Abstract

Objective Metabolic changes that occur in diabetes mellitus (DM) can interfere with the energy supply for the metabolic active inner ear, resulting in vestibular impairments. The purpose of this study was to evaluate the effects of type 1 DM on the vestibular system using a vestibular test battery. Methods Vestibular function of fifteen asymptomatic type 1 DM patients with a mean age of 28 years (SD= 5.80) and 16 healthy controls with a mean age of 26 years (SD = 2.86), respectively, was assessed using video-head impulse test (vHIT), ocular and cervical vestibular evoked myogenic potential (o/cVEMP), and subjective visual vertical and horizontal (SVV/SVH). Results vHIT response was pathological, in approximately 6.66% of canals. The peak-to-peak p1-n1 amplitude in cVEMP and p1 latency of oVEMP were significantly smaller in the left ear (p = .018) and lengthened in both ears (p = .004) of DM participants compared to the healthy group, respectively. The mean deviation of dynamic SVV (Opto +40/s) was significantly increased (p = .022). Conclusion Asymptomatic DM can influence the vestibular system, especially the otolith organs, indicating the presence of subclinical vestibular dysfunction (VD) prior to the onset of vestibular manifestations, which can be detected using appropriate vestibular tests. Furthermore, DM might have selective or chronological order effects on the human vestibular system, with the otolith system damage preceded the semicircular canals. Thus, early and periodic vestibular assessment of DM patients for detecting possible latent VD and preventing further problems is advisable. However, further clinical studies are required.

When, where, and why should we look for vestibular dysfunction in people with diabetes mellitus?

“…Of these, all but one small case-control study ( N = 16) ( 52 ) found abnormal utricular responses in people with DM. Significantly delayed latency, or a greater number of abnormal latencies, were the most common finding in people with DM compared to controls, and in one study comparing DMPN to DM and controls ( 45 , 48 , 55 , 57 ). Perhaps due to differences in approaches, two studies identified low amplitude responses in people with DM vs. controls ( 48 , 50 ).…”

“…Given this chronological distribution and more recent clinical implementation of cervical and ocular vestibular evoked myogenic potential (VEMP) and multi-canal head thrust tests, the horizontal semicircular canal (SCC) was the most frequently assessed vestibular end-organ in people with Type 1 DM. In these studies, some degree of abnormal caloric responses (low frequency) were consistently reported (5/5 studies), 1/3 reported vestibulo-ocular reflex (VOR) abnormalities with higher frequency SCC tests ( 35 , 36 , 39 – 42 , 55 , 56 ), and 2/2 reported abnormal saccule and/or utricle function ( 44 , 55 ). In contrast, the otolith received more attention than the SCCs, and SCC assessments were across a range of frequencies (low, middle, high; Table 2 ) ( 17 ) in people with Type 2 DM.…”

“…Reviewing literature regarding the location of peripheral vestibular dysfunction in people with DM may provide clues to elevate efficiency of clinical testing. However, different study samples, designs, testing approaches, and the reality that only 4/23 studies considered each end-organ pathway within the same sample, will be reflected in our ability to make recommendations ( 48 , 50 , 52 , 55 ). Nevertheless, studies are fairly balanced by end-organ as 10 studies assessed utricle function, 10 assessed saccule function, and 16 assessed SCC function ( Table 4 ).…”

“…Six studies employed oVEMP testing with a median DM group sample size of 27.5 and age of 48.5 years ( 45 , 48 , 50 , 52 , 55 , 57 ). Tone burst VEMPs were the most commonly employed stimuli, although intensity varied across studies.…”

“…Four studies conducted tests within the subjective visual vertical (SVV) paradigm with a median DM group sample size of 47 and age of 57.5 years ( 46 , 49 , 53 , 55 ). Three studies failed to detect differences in static SVV ( 46 , 53 , 55 ).…”