Risk factors related balance disorder for patients with dizziness/vertigo

The study was approved by Ethics Committee of Beijing Tainan Hospital and performed according to the Declaration of Helsinki guidelines. Informed consent was obtained from the participants. We reviewed the medical data of 1002 dizziness/vertigo patients who registered in vertigo/balance disorder registration database of the Clinical Center for Vertigo and Balance Disturbance of Beijing Tiantan Hospital. The definition of dizziness and vertigo was in accordance with the Bárány Society consensus [11]. Dizziness is the sensation of disturbed or impaired spatial orientation without a false or distorted sense of motion. Vertigo is the sensation of self-motion (of head/body) when no self-motion is occurring or the sensation of distorted self-motion during an otherwise normal head movement. The eligible patients who met the following criteria for inclusion in this study were: 1) 18 ≤ age ≤ 80 years; 2) complaint of dizziness or vertigo symptom; 3) ability to walk independently without assistance; 4) completed Romberg test, videonystagmography (VNG) and limits of stability (LOS) of computed dynamic posturography (CDP). Exclusion criteria were as follows: 1) contraindication of VNG; 2) any disease history that precludes patients from evaluating of balance state and vestibular function including dementia, visual impairment, acute stroke, movement disorders, musculoskeletal disease, and disease of extraocular muscle. The detailed flowchart figure was shown in the Fig. 1.

Comprehensive history taking questionnaires including demographic and clinical characteristics were performed on all participants. The demographic information was consisted of age and gender. The clinical characteristics referred to symptoms, medical history, and risk factors for atherosclerosis (AS). The medical history included benign paroxysmal positional vertigo (BPPV), Ménière’s disease (MD), sudden deafness, vestibular neuronitis (VN), motion sickness, migraine, hyperthyroidism, anemia, orthostatic hypotension, obstructive sleep apnea-hypopnea syndrome (OSAHS), and traumatic brain injury (TBI). The AS risk factors included age ≥ 60 years, hypertension, diabetes, hyperlipidemia, coronary heart disease (CHD), and stroke.

All participants underwent detailed neurological examination including consciousness, speech, cranial nerve, sensory, and motor. Cerebellum function was assessed: index-nose test, and heel-knee test. Meanwhile, we also performed Romberg test which can reflect the functions of somatosensory system, cerebellum function, and muscular strength. The participants were asked to maintain balance in the natural standing position with feet together and arms beside the body. The eyes were closed. The definition of subjective imbalance was patient's complained of postural symptoms (one of unsteadiness, directional pulsion, balance-related near fall, and balance-related fall) when performing the Romberg test.

All enrolled patients adopted VNG to evaluate the vestibular function. The VNG test (Micro-medical, VisualEyes 20,178,624, America) included spontaneous nystagmus test, gaze test, saccade test, smooth pursuit test, head shaking test (HST), optokinetic nystagmus test, positioning test (Roll test or Dix-Hallpike test), and caloric test. To test spontaneous nystagmus, the subject was seated in the upright position with looking straight ahead. The eye movements were recorded for at least 20–30 s under two conditions: open eyes in normal bright room and dark room. The subject looked at the four eccentric optotypes in turn, respectively 30 degrees on the left/right and up/down 25 degrees to test gaze. Each direction was recorded for at least 20 s. When nystagmus occurred, the observation record was 60s. Spontaneous nystagmus with slow phase velocity (SPV) >3°/sec and gaze-evoked nystagmus >6°/sec were considered abnormal. To test saccade, the subject took a sitting position with head fixed in the middle position, and eyes were 1.2 m away from the visual target. The subject was asked to look at the target and record their eye movements. Each test lasts at least 1 min. The subject looked at and swing in a sine wave with the visual target. The smooth pursuit test was divided into four types according to the eye movement curve: I and II were considered normal, III and IV were abnormal. To test optokinetic nystagmus, firstly the subject was instructed to gaze at a fixed spot projected on the center of a uniform white hemispherical full- field screen (diameter 140 cm) in a dark room. A random dot pattern (black and white with uniform acceleration and deceleration) was projected (ranging in size from 0.5 to 2.0 cm in diameters, covering 87% of the visual field) on the screen. The subject was presented with a pattern that filled the greater part of his peripheral and foveal ranges of vision. The subjects were in sitting position, head fixed, facing the horizontal target in front of them. The subjects were asked to keep an eye on the target and count the number of targets silently. The subjects were exposed optokinetic stimulation at speeds of 60 degrees/s preceded by acceleration at 6 degrees/s². Bilateral asymmetry was considered abnormal. HST was performed with the patient sitting in a clinical chair with the head leaning down by 30°. The patient’s head was vigorously rotated for 20 times on the horizontal plane with a maximum amplitude around 30–40°. Post HST nystagmus was recorded for 1 min and was considered positive when nystagmus lasting at least 5 s was detected. The HST was performed to evaluate the vestibulo-ocular reflex, which represents the control of eye movements by the vestibule. A peripheral etiology was related to horizontal or horizontal-rotatory nystagmus, indicative of a vestibulo-ocular pathway abnormality. The supine-roll test was performed with the neck flexed forward about 30° in order to bring the lateral semicircular canal into alignment with the vertical. Then examiner turned the patient’s head to the right or left in the supine position, and eye movements were recorded for 30–60 s. To test Dix-Hallpike, patient sited upright. The examiner turned the patient’s head 45 degrees to the patient’s right. Then examiner quickly changed the posture from sitting position to the supine position with the head hanging below the top end of the examination table at an angle of 30 degrees. The test was abnormal if three or more than beats of nystagmus were observed. BPPV patients showed a few seconds of the incubation period, short-term dizziness and vertical rotation nystagmus occurrence. Caloric test was performed with the head supine and tilted forward 30°. Two methods were applied to perform caloric test. One irrigated the right and left ears with cold water (30 °C) and hot water (44 °C). The other irrigated the right and left ears, respectively with cold air (24 °C) and hot air (50 °C). The canal paresis (CP) value greater than 25% was defined as abnormal according to the Jongkees Index formula calculation.

The dynamic balance function evaluation was performed using the SMART EquiTest balance assessment system (NeuroCom International, Clackamas, OR, USA). The LOS test can evaluate an individual’s ability to move their center of gravity (COG) to the given eight direction in space. There is a stick figure in the system screen representing the subject’s live-time dynamic center of pressure (COP). The figure is surrounded by eight targets that represent the maximal theoretical stability of patients. When the COP of patients moves, the figure on the screen will move accordingly. Once the voice prompt sounds, subjects need to move their body to the target that appears on the screen and keep balance as much as possible. A total of eight trials were measured, each with 8 s. Every trial was incorporated in the equation of the following measurements, which are mean composite (eight directions/trials averaged) scores: 1) Reaction time (RT) is the time from the start of trial to patient’s first movement (seconds); 2) Movement velocity (MVL) is the mean speed of COG movement (degrees/second); 3) Endpoint excursion (EXE) is the distance of COP displacement when initial lean towards target (% of maximal LOS); 4) Maximum excursion (MXE) is farthest distance of COP displacement during the eight trials (% of maximal LOS); 5) Directional control (DCL) is a comparison of the amount of movement in the planned direction (towards the target) to the amount of extraneous movement (away from the target). When the MVL, DCL, one of the EXE and MXE were both abnormal according to the results of LOS, we considered it as balance disorder.

Continuous variable namely age was described as mean ± standard deviation (SD). Categorical variables were presented as frequency and percentage. Baseline characteristics between normal balance and balance disorder were compared with chi-squared test for categorical and Mann-Whitney U test for age. Multivariable logistic regression analysis was adopted to analyze the risk factors related balance disorder. We incorporated the variable of P<0.2 in the univariate analysis into the multivariable analysis. The receiver operating characteristic (ROC) curve was performed to evaluate sensitivity and specificity of regression model. The “optimum” cut-off value for ROC curve was defined as the value associated with the maximal sum of sensitivity and specificity. Statistical analysis was performed in SPSS 24.0 (IBM, Chicago, IL, USA). Values with P < 0.05 were regarded as statistically significant.

We reviewed 1002 dizziness/vertigo patients who registered in the vertigo/balance disorder registration database of Clinical Center for Vertigo and Balance Disturbance of Beijing Tiantan Hospital from Jul 2019 to Jul 2020. After excluding patients with incomplete clinical data, a total of 553 patients who could walk independently were included in the final analysis. Five hundred fifty-three patients both completed VNG and LOS examinations. According to abnormal results of LOS, participants were divided into 334 (60%) normal balance and 219 (40%) balance disorder.

The demographics and baseline clinical characteristics were shown in Table 1. There was no significant difference in sex between normal balance and balance disorder groups. Compared to patients with normal balance group, patients with balance disorder were older (P = 0.045) and had more risk factors for atherosclerosis (P<0.0001, Fig. 2). When we performed the Romberg test, more patients in the balance disorder group complaint of imbalance by verbal report to the physician (P<0.0001). No significant difference was found in other medical history which was related to dizziness/vertigo symptom between two groups.

All included patients had received the VNG examination. The detailed results were indicated in Table 2. The abnormality of optokinetic nystagmus test was more common in the balance disorder group (4.1% vs 0.6%, P = 0.004). Other testes of VNG including saccade, spontaneous nystagmus, gaze, smooth pursuit test, head shaking, positioning, and caloric testes, didn’t have significant difference between two groups.

Based on the results of univariable analysis, we performed multivariable analyses to investigate the risk factors for balance disorder in patients with dizziness/vertigo who could walk independently. The consequences were shown in Table 3. The results showed that subjective imbalance (OR 4.835, 95% CI 3.047–7.673, P<0.0001), abnormality of optokinetic nystagmus (OR 8.308, 95% CI 1.576–43.789, P = 0.013), and risk factors for AS (OR 1.494, 95% CI 1.198–1.863, P<0.0001) were associated with balance disorder. The smooth pursuit test, positioning test, medical history of MD and motion sickness were not related to balance disorder. To evaluate the utility of regression model in identifying the risk factors for balance disorder, ROC curve was depicted in Fig.3. The sensitivity and specificity of regression model were 71 and 63% respectively (95% CI 0.678–0.764, P<0.0001). The area under the curve (AUC) was 0.721.