Objective: To assess the effect of caffeine in the following vestibular function tests: Cervical  Vestibular Evoked Potential (cVEMP), Ocular Vestibular Evoked Potential (oVEMP) and Caloric  Test. 
Methods: Randomized, prospective triple-blind, placebo controlled clinical trial. All  participants underwent otoscopy, tympanometry and responded to the Profile of Mood State  (POMS). They were submitted to the cVEMP, oVEMP and caloric tests. After that they received  placebo capsule (maize starch) or caffeine capsule (300mg) and repeated the procedures 45  minutes later. 
Results: There were no statistically significant differences in latencies, peak to peak  amplitudes, asymmetry ratio or rate of change in cVEMP. A statistically significant difference  was observed in the caffeine group (p15 latency of left ear) in oVEMP. The Non-caffeine group  showed statistically significant difference between the relative values in caloric test. No variable  of any test was influenced by caffeine intake. 
Conclusions: Moderate caffeine consumption does not significantly alter vestibular function  tests.
Significance: This study provides the evidence that cVEMP, oVEMP and caloric test do not  suffer influence from moderate caffeine consumption.

• Electronystagmography

• Vestibular evoked myogenic potentials

• Vertigo

• Diseases

• Vestibular

• Caffeine

Losno Ledesma AL, De Souza Chelminski Barreto1 MA, De Oliveira CACP, Bahmad F (2016) Effect of Caffeine Intake in Vestibular Function: A Randomized, Triple-Blind, Controlled Trial. Ann Otolaryngol Rhinol 3(12): 1155.

In vertigo patients vestibular tests are indicated in order to identify the presence of changes and find the affected area. Thus a thorough investigation of the peripheral and central vestibular organs is required [1,2].

Caloric test is an important tool in vestibular evaluations and it investigates the horizontal semicircular canals [3,4]. The application of different temperatures to the external auditory canal (warm or cold) generates an endolymphatic movement inducing the appearance of nystagmus [5]. Cervical vestibular evoked myogenic potentials (cVEMP) is a manifestation of the vestibulo-colic reflex [6-8]. It is an inhibitory electromyographic potential, due to acoustic stimuli of high intensity, which excites the saccule and inferior vestibular nerve (sacculocollic pathway). Since the first description it has become a well-established clinical test of vestibular function [9].

Ocular vestibular evoked myogenic potentials (oVEMP) reflect predominantly utricular -otolith function and crossed vestibuloocular pathway - superior division of the vestibular nerve [10- 12]. The response consists of a series of waves, beginning with a negative peak followed by a positive one [13].

The posterior labyrinth is a highly sensitive organ to changes in other organs and systems, so many of these changes first manifest with vestibular symptoms [14]. Caffeine is a psychoactive substance very commonly used in the world, found in the most diverse products such as food and drugs. Being a stimulant of the CNS it is believed that it excites the labyrinth, but no strong scientific evidence of this relationship exists [15-18]. While there is no consensus on this interaction, each service adopts recommendations that seem convenient, making it difficult to compare the exams. In addition, patients are subjected to a strict diet that can have systemic effects on habitual caffeine users, such as severe headaches, making it difficult to cooperate during the tests.

 The aim of this study is to assess the effect of caffeine in the following vestibular tests: cVEMP, oVEMP and caloric test.

Study design

This is a randomized, prospective triple-blind, placebo controlled clinical trialapproved by the institutional review boardunder protocol number 1.399.322/16 and registered in Clinical Trial. Informed consent was obtained after a full explanation of the experimental procedure.

Eligibility

The sample was made up of medical students who agreed to participate as volunteers and had no hearing and/or vestibular complaints or other health problems that could affect the homeostasis of the vestibular system (problems in the cervical spine, cardiovascular problems, migraine, metabolic disorders, hormonal changes, psychiatric disorders, neurological diseases, use of prescription drugs continuously, were smokers, alcoholics, or illegal drug users).It consisted of 32 healthy young subjects randomly divided into two parallel groups: caffeine [18] and non-caffeine [14], similar to each other in terms of age and daily caffeine intake (Figure 1).

Figure 1: Patients election.

Randomization and allocation concealment

At baseline, study participants were randomized to receive caffeine or placebo at a 1:1 ratio, by independent statistician. We used a completely randomized design for two treatmentswith PLAN procedure (SAS 9.4). The allocation was made for a collaborator, who identified the capsules in identical packages labeled with the name of the participants. The active (300mg caffeine) and placebo (maize starch) capsules were identical in color, size, weight, and packaging.

Blinding

The collaborator responsible for labeling the capsules with the names of the participants had no contact with them, the audiologist, or the statistician responsible for the data analysis. Thus,the participants were not aware of what was contained in the capsule, as well as the audiologist responsible for collecting and analyzing the exams and questionnaires. The statistician was independent and was also unaware of the substance to which the two groups had been exposed.

Intervention

Subjects were invited to participate in the study through telephone contact and were advised to abstain from products that contained caffeine 24 hours before their participation. The same detailed guidelines were sent by email. All participants were questioned about compliance with the recommendations for the exams and those who declared they had not followed, however, still wished to participate in the study, had their exams re-scheduled. All participants underwent otoscopy and tympanometry, responded the Profile of Mood State (POMS),and were submitted to the cVEMP, oVEMP and caloric tests in that order. After that they received placebo or caffeine capsule. After 45 minutes they again responded to the POMS, repeated the cVEMP, oVEMP and caloric test.The procedures were repeated 45 minutes after ingestion of the capsule because it corresponds to the peak plasma concentration of caffeine [19]. The tests were performed in a center for rehabilitation of hearing and balance by the same audiologist. Data collection time was 3 months.

Caffeine consumption

To determine caffeine consumption a questionnaire which sought to investigate eating habits was applied. They have to include for each item (coffee, teas, soft drinks, chocolate, powdered chocolate, powdered guarana, food supplements and energy drink) the amount, kind (i.e. espresso, filter coffee) and brand. Based on these data, the consumption was calculated using a preview standardization [20]. Those who had higher intakes equal to 500mg / dayor more (heavy or very heavy) were excluded.

POMS

The questionnaire was developed with the objective of investigating mood states and their fluctuations in psychiatric patients and in normal adults, becoming a widely used tool in caffeine research [21].

Self-administered instrument that consists of 65 items describing feelings; to them values should be assigned from 0 to 4 according to the Likert scale of 5 points (0 - no, 1 - a little, 2 - more or less, 3 - well, 4 - extremely). The POMS aims to evaluate six factors: tension-anxiety (9 items), depression-dejection (15 items), anger-hostility (12 items), vigor-activity (8 items), fatigue-inertia (7 items), confusion-bewilderment (7 items), and a total mood disturbance score (TMD) [22].

cVEMP

VEMP was performed in response to air-conducted stimuli, using alternate 500Hz tone bursts presented at 120 dB pSPL and rate of 5.1 stimuli per second [23]. Band pass filter of 10 Hz to 1500 Hz was used, 50ms window and 200 stimulations in each track were standard. Two stimulations were recorded on each side, in order to observe the replicability[24]. The stimuli were presented via insert earphones ER-3A.

A cleaning of the skin with abrasive paste was performed and surface electrodes were affixed using conductive paste. The non inverting electrode was placed in the middle part of the sternocleidomastoid [25], the ground electrode in lower forehead and the inverting on the upper forehead. The impedance of the electrodes should be less than or equal to 5hms. Subjects were positioned seated and instructed to maintain maximum head turn to the side contra lateral to the stimulus throughout stimulation [26].

The latency of p13 and n23 and the amplitude peak to peak were analyzed in each ear and an assymmetry rate was also analyzed. For the last one we used the formula ?amplitude of right ear – amplitude of left ear?/?amplitude of right ear + amplitude of left ear?x100 [27]. To compare the amplitudes in each ear before and after tests we used the rate of change formula, consisting of ?amplitude of ear before – amplitude of ear after?/?amplitude of ear before + amplitude of ear after?x100 [28].

We considered normal p13 latencies between 13.9 to 19.2 and from 22.9 to 30.3 n23 latencies [29] and the assymmetry rate and rate of change of up to 28% [28].

oVEMP

We used the same parameters of stimulation and analysis described in cVEMPas previously proposed [8]. Non inverting electrodes were placed on the face just inferior to each eye (inferior oblique muscle), reference electrode was placed 1-2cm below contra lateral to the stimulus and ground electrode in the forehead. The subject was placed in the sitting position and instructed to look upward [8,9]. The latencies of initial negative (n1) and positive (p1) peak were measured. We considered normal n1 latencies from 10.2 to 11.8 and from 14.7 to 17.3 p1 latencies [9].

Caloric test

To perform the caloric test we used an air oto calorimeter, and carried out four stimulations of 60 seconds each, with a flow of 8 liters per minute: 500 C in the right ear, 500 C on the left, 240 C on the left ear and 240 C in the right ear, in this order. The absolute values and percentages of unilateral weakness or directional preponderance were analyzed for each subject in each test. They were considered normal when they showed absolute values from 3 to 450 , a range of 30% or less in directional preponderance and of less than 25% in the unilateral weakness[1].

Outcomes

The primary outcome was the latency of waves for cVEMP and oVEMP and absolute and relative values for caloric tests. These parameters were chosen for analysis of the primary outcome for being the most well-established parameters for use in clinical practice. The secondary outcomes were asymmetry ratio and ratio of change for cVEMP. There were no secondary outcomes for oVEMP and Caloric test.

Statistical analysis

Non-parametric tests were used seeking statistically describe and compare the two groups - caffeine and no caffeine (MannWhitney test), the two time points - before then within each group (Wilcoxon signed-rank test) and study the statistical effect of caffeine consumption on the other variables (JonckheereTerpstra test). ACI of 95%was used.

The spread sheet MS-Excel was used in the version of MSOffice 2013 for the organization of data, and IBM SPSS (Statistical Package for Social Sciences), in its version 23.0, to obtain the results.

The sample was composed of 27 females and 5 males ranging in age from 18 to 42 years (mean: 26,88 ±SD:6,98 yr). Subjects reported consuming an average of 65,63mg/day (SD: 68,33) of caffeine. The distribution of the groups is shown in Figure (2).

Figure 2: Distribution of sample according to age and caffeine intake.

The choice by young individuals occurred to exclude degenerative processes of vestibular system structures that occurs at advanced age as a part of the multiple sensorial loss of aged individuals [30].

In the literature we found few studies that estimated the intake of caffeine: most estimated the coffee intake. In this study the average consumption of caffeine was 65,63mg / day, consumption much lower than reported by other studies [31- 33].The low habitual caffeine consumption reported by the participants would make us to believe that they would suffer more the effects of abrupt consumption of moderate dose of caffeine.

The groups showed no statistically significant difference either in POMS or TMD comparing caffeine and non-caffeine. Comparing the two periods, there was statistically significant improvement in depression-dejection and anger-hostility in the caffeine group and in tension-anxiety, anger-hostility and confusion-bewilderment in the placebo group see Table (1).

Table 1: Comparing the values before and after capsule intake in the Profile of Mood State in each group.

Decrease in depression-dejection was reported in another study [33]. However, the results disagree with the majority of studies that found increase in vigor-activity [17,34-37]and reduction in fatigue-inertia [34,36,38]after caffeine intake. This fact is attributed to the low consumption of caffeine in the habitual diet of the participants.

Regarding the caffeine intake we found a strong relationship between consumption and the TMD: the higher the caffeine intake the greater was the change in mood, a fact previously reported [39].Evidenced by the Jonckheere-Terpstra test with p-value of 0.022.

In cVEMP, there were no statistically significant differences in latencies (p13 and n23), peak to peak amplitudes, asymmetry ratio or rate of change between the caffeine and non-caffeine groups. Comparing the two moments (before and after the capsule) there was also no statistically significant difference in any of the parameters analyzed. This was also observed in other studies investigating the action of caffeine in healthy individuals [28,40].

With respect to oVEMP, there were no statistically significant differences in latencies (n10 and p15) between the caffeine and non-caffeine groups. Comparing the two moments a statistically significant difference was observed in p15 latency of left earin the caffeine group. This finding did not affect the clinical outcome of this test. We did not find other study that tried to analyze the effect of caffeine on oVEMP. Considerable research efforts have led to better understanding oVEMP.

In the last years, they focus primarily on determining the origin of responses, standardization of the stimulus, electrode placement and position of the patient[13,41]. We did not find studies suggesting some preparation for the exam: diet, medication usage restrictions, etc [9,12].This demonstrates the need for future studies. Other analyzed parameters did not show this difference (Table 2).

Table 2: Comparison of before and after capsule intake in the caffeine and non caffeine groups in oVEMP and cVEMP parameters.

In caloric tests one of the non-caffeine group participants presented neurovegetative exacerbated symptoms during the first examination, and was not submitted to the second test. As a consequence the caloric test statistical analysis was performed with 31 participants: caffeine group (18) and non-caffeine group (13).

In caloric tests the averages of the Peak slow velocity (PSV) in all stimulations were similar in both groups. Likewise, in the caffeine group all stimulations were statistically similar comparing the before and after caffeine intake. All participants had the same type of analysis (unilateral weakness or directional preponderance) in the two moments. In this group, one of the participants presented hyper reflexia during the warm and cool stimulation of the left ear at both times. The results were similar to those found previously[40,41].

The non-caffeine group showed statistically significant difference between the relative values before and after capsule intake (Table 3).

Table 3: Comparison of before and after the caffeine group and non caffeine regarding Caloric test parameters: PSV for each irrigation, percentage of unilateral weakness or directional preponderance.

It can be noted that all participants got the same report in the tests performed before and after placebo capsule, showing that the differences were of no clinical value. These differences can be explained by the low test–re-test reliability of the caloric test [32]. It is important to point out that variance for calorics is large and is dependent on several factors, such as attention, the effectiveness of caloric stimulation, and size of ear canal [40].

No variable of any test (cVEMP, oVEMP, caloric test) was influenced by caffeine intake, showing that habitual caffeine consumption had little effect on the parameters analyzed in these tests.

Moderate caffeine consumption does not significantly alter the clinical interpretation of the results obtained in the vestibular tests: cVEMP, VEMP and caloric test. This way we can infer that it is possible to perform vestibular exams following the usual patterns of caffeine consumption in young individuals.

1. Eza-Nuñez P, Fariñas-Alvarez C, Fernandez NP. Comparison of three diagnostic tests in detecting vestibular deficit in patients with peripheral vestibulopathy. J LaryngolOtol. 2016; 130:145-150.

2. Neuhauser HK, Von Brevern M, RadtkeA, Lezius F, Feldmann M, Ziese T, et al. Epidemiology of vestibular vertigo: a neurotologic survey of the general population. Neurology. 2005; 65: 898-904.

3. Burle NLO, Abreu ACP, Santos JN et al. The impact of dizziness on the quality of life of 235 Individuals who completed vestibular testing in Brazil. Int Arch Otorhinolaryngol. 2016; 20: 54-60.

4. Chiarovano E, Vidal P, Magnani C, Lamas G, Curthoys IS, de Waele C, et al. Absence of rotation perception during warm water caloric irrigation in some seniors wit postural instability. Front Neurol 2016; 7:1-8.

5. Maheu M, Houde MS, Landry SP, Champoux F. The Effects of Aging on Clinical Vestibular Evaluations. Front Neurol. 2015; 6: 205.

6. Colebatch JG, Halmagyi GM. Vestibular evoked potentials in human neck muscles before and after unilateral vestibular deafferentation. Neurology. 1992; 42:1635-1636.

7. Kantner C, Gürkov R. Characteristics and clinical applications of ocular vestibular evoked myogenic potentials. Hear Res. 2012; 294: 55-63.

8. Rosengren SM1, Welgampola MS, Colebatch JG. Vestibular evoked myogenic potentials: past, present and future. Clin Neurophysiol. 2010; 121: 636-51.

9. Chiara Y, Iwasaki S, Ushio M, Murofushi T. Vestibular-evoked extraocular potentials by air-conducted sound: Another clinical test for vestibular function. ClinNeurophysiol. 2007; 118: 2745-2751.

10. Felipe L, Kingma H. Ocular vestibular evoked myogenic potentials. Int Arch Otorhinolaryngol. 2014; 18: 77-79.

11. Rosengren SM, McAngus Todd NP, Colebatch JG. Vestibular-evoked extraocular potentials produced by stimulation with bone-conducted sound. Clin Neurophysiol. 2005; 116: 1938-1948.

12. Sandhu JS, Stefan RG, Peter AR. The effect of electrode positioning on the ocular vestibular evoked myogenic potential to air-conducted sound. Clin Neurophysiol. 2013; 124: 1232-1236.

13. Weber KP, Rosengren SM. Clinical utility of ocular vestibular-evoked myogenic potentials (oVEMPs). Curr Neurol Neurosci Rep. 2015; 15: 22.

14.Cal R, Bahmad Jr F. Vestibular evoked myogenic potentials: an overview. Braz J Otorhinolaryngol. 2009; 75: 456-462.

15.Chou T. Wake up and smell the coffee. Caffeine, coffee, and the medical consequences. West J Med. 1992; 157: 544-53.

16. agarde D, Batéjat D, Sicard B, Trocherie S, Chassard D, Enslen M, et al. Slow-release caffeine: a new response to the effects of a limited sleep deprivation. Sleep. 2000; 23: 651-661.

17. Schuh KJ, Griffiths RR. Caffeine reinforcement: the role of withdrawal. Psychopharmacology (Berl). 1997; :320-326.

18. Silverman K, Evans SM, Strain EC, Griffiths RR. Withdrawal Syndrome after the Double-Blind Cessation of Caffeine Consumption. N Engl J Med. 1992; 327: 1109-1114.

19. Fredholm BB, Bättig K, Holmén J, Nehlig A, Zvartau EE. Actions of caffeine in the brain with special reference to factors that contribute to its widespread use. Pharmacol Rev. 1999; 51: 83-133.

20. Camargo MCR, Toledo MCF. Caffeine content of commercial brazilian coffee. Food Science and Technology 1998;18: 421-424.

21. McNair DM, Lorr M, Droppleman LF. Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service. 1971.

22. Olson CA, Thornton JA, Adam GE, Lieberman HR. Effects of 2 adenosine antagonists, quercetin and caffeine, on vigilance and mood.J ClinPsychopharmacol. 2010; 30: 573-578.

23. Akin FW, Murnane OD, Medley TM. The effects of click and tone burst stimulus parameters on the vestibular evoked myogenic potential (VEMP). J Am AcadAudiol. 2003; 14: 500-508.

24. ilburg MJV, Hermann BS, GuinanJr JJ, Rauch SD. Normalization Reduces Intersubject Variability in Cervical Vestibular Evoked Myogenic Potentials. Otol. Neurotol 2014; 35: 222-227.

25. Piker EG, Jacobson GP, McCaslin DL, Hood LJ. Normal Characteristics of the Ocular Vestibular Evoked Myogenic Potential. J Am AcadAudiol. 2011; 22: 222-230.

26. Alves AM, Tateyama T, Jakarius F, Silva RC, Kasse CA, Doriguetto RS, et al. Vestibular Evoked Myogenic Potentials: Recording Methods and Cilinical Applications. Reces. 2014; 6: 67-71.

27. Welgampola MS, Colebatch JG. Characteristics and clinical applications of vestibular-evoked myogenic potentials. Neurology. 2005; 64:1682- 1688.

28. Sousa AMA, Suzuki FA. Effect of caffeine on cervical vestibular-evoked myogenic potential in healthy individuals. Braz j otorhinolaryngol. 2014; 80: 26-30.

29. Blakley BW1, Wong V. Normal Values for Cervical Vestibular-Evoked Myogenic Potentials. Otol Neurotol. 2015; 36: 1069-73.

30. Nguyen KD, Welgampola MS, Carey JP. Test-Retest Reliability and AgeRelated Characteristics of the Ocular and Cervical Vestibular Evoked Myogenic Potential Tests. Otol. Neurotol. 2010; 31: 793-802.

31.Claire LS, Stothart G, Mckenna L, Rogers PJ. Caffeine abstinence: an ineffective and potentially distressing tinnitus therapy. Int. J. Audiol. 2010; 49: 24-29.

32. Eny KM, Orchard TJ, Miller RG, Maynard J, Grant DM, Costacou T, et al. Paterson A.D. Caffeine Consumption Contributes to Skin Intrinsic Fluorescence in Type 1 Diabetes. Diabetes TechnolTher. 2015; 17: 726-734.

33. Richards G, Smith A. Caffeine consumption and self-assessed stress, anxiety, and depression in secondary school children. JPsychopharmacol. 2015; 29:1236-1247.

34. Strain EC, Mumford GK, Silverman K, Griffths RR. Caffeine dependence syndrome: evidence from case histories and experimental evaluations. JAMA. 1994; 272:1043-1048.

35. Ali A, O’Donnell J, Von Hurst P, Foskett A, Holland S, Starck C, et al. Caffeine ingestion enhances perceptual responses during intermittent exercise in female team-game player. J Spots Med. 2016; 34: 330-341.

36. Griffiths RR, Evans SM, Heishman SJ, Preston KL, Sannerud CA, Wolf B, et al. Low-dose caffeine discrimination in humans. J Pharmacol Exp Ther. 1990; 252: 970-978.

37. James JE, Gregg ME. Effects of dietary caffeine on mood when rested and sleep restricted. Hum PsychopharmacolClin Exp. 2004; 19: 333- 341.

38. Evans SM, Griffiths RR. Caffeine withdrawal: a parametric analysis of caffeine dosing conditions. J Pharmacol Exp Ther. 1999; 289: 285-294.

39. Heatherley SV, Hancock KM, Rogers PJ. Psychostimulant and other  effects of caffeine in 9- to 11-year-old children. J Child Psychol Psychiatry. 2006; 47:135-142.

40. McNerney K, Coad ML, Burkard R. The influence of caffeine on calorics and cervical vestibular evoked myogenic potentials (cVEMPs). J Am Acad Audiol. 2014; 25: 261-267.

41. Felipe P, Simes LC, Gon alves DU, Mancini PC. Evaluation of the caffeine effect in the vestibular test. Braz J Otorhinolaryngol. 2005; 71: 758-762