Anesth Analg 1999;88:1064-1068
© 1999 International Anesthesia Research Society
AMBULATORY ANESTHESIA
The Effects of Midazolam on Pure Tone Audiometry, Speech Audiometry, and Audiological Reaction Times in Human Volunteers
Dermot J. Kelly, MRCPI, FFARCSI*,
Fergus Walsh, MRCPI, FFARCSI*,
Gary S. Norman, MSc
, and
Anthony J. Cunningham, MD, FFARCSI*
Departments of
*Anaesthesia and
Clinical Audiology, Beaumont Hospital, Royal College of Surgeons in Ireland, Dublin, Ireland
Address correspondence and reprint requests to Dermot J. Kelly, MRCPI, FFARCSI, Department of Anaesthesia, St. Vincent's Hospital, Elm Park, Dublin 4, Ireland.
 |
Abstract
|
|---|
Auditory evoked potentials are effected by benzodiazepines, as is cortical processing of auditory stimuli. The effect of benzodiazepines on auditory sensitivity has not, however, been studied. We designed the present study to investigate the effect of sedative doses of midazolam on pure tone and speech audiometry and on audiological reaction times in healthy volunteers. Thirty volunteers underwent baseline audiological assessment for pure tones and speech and had their audiological reaction times measured at 10 and 50 dB above their threshold hearing level at a frequency of 1 kHz. Subjects were then randomly assigned to one of two groups. Group A (n = 15) received midazolam (0.04 mg/kg) IV, and Group B (n = 15) received a similar volume of placebo IV. The audiological tests were repeated 5 min later, and performance was compared with baseline data. Scheffé post hoc tests were used to assess the significance of changes in each group. There was no pre- to posttest change in audiological performance in either the placebo group (P = 0.194) or the midazolam group (P = 0.957). Speech audiometry performance was likewise unaffected by midazolam (P = 0.154). Reaction time at the 10-dB and 50-dB sensation levels were both significantly prolonged after midazolam administration (P = 0.023 and P = 0.012, respectively). In this study, we demonstrate that sedation with midazolam (0.04 mg/kg) does not alter pure tone or speech audiometric thresholds, but it does significantly delay the reaction time to auditory stimuli. Medical practitioners should advise midazolam-sedated patients of their impaired reaction to auditory warning signals (e.g., traffic and car horns) as part of the day-ward discharge recommendations.
Implications: In this study, we demonstrate that sedation of healthy volunteers with the benzodiazepine midazolam, in the common clinical dosage, does not affect their hearing capability as measured by pure tone and speech audiometry. However, one's ability to react to auditory signals is impaired after midazolam, which may have implications for patients after day-case procedures.
|
Introduction
|
|---|
Benzodiazepines affect processing of acoustic stimuli by the primary auditory cortex (1) and impair higher cognitive processing of auditory signals (2). They also affect auditory evoked potentials (AEPs), by increasing the latency and decreasing the amplitude of the cortical response (2,3). The amplitude of the cortical response is directly related to the intensity of the auditory signal (4,5). However, despite numerous studies evaluating the effects of benzodiazepines on the various electrophysiological components of the auditory pathway (1–3), their effect on auditory sensitivity has not been studied. Nonetheless, many studies investigating the effect of benzodiazepines on memory and/or intraoperative recall use verbal associations or visual images (explained by verbal commands) as a means of testing memory function (6–8).
Pure tone audiometry measures the threshold of hearing in response to pure tone signals. The pure tones are delivered monoaurally at a range of frequencies and a variety of volume intensity levels; the subject acknowledges detection of the signal by pressing a button. The auditory threshold level is noted at each frequency and is a measure of auditory sensitivity. Unlike its pure tone counterpart, speech audiometry gives a measure of the clinical handicap produced by a given hearing deficit. Word lists are delivered monoaurally to the subject at varying volume intensities, and he/she repeats the words on recognition. The performance on this test gives an estimation of the impairment of speech perception.
The objectives of this study were to measure the effects of a sedative dose of midazolam (0.04 mg/kg) on pure tone and speech audiometry and on audiological reaction time and to determine whether midazolam impairs auditory perception in healthy volunteers.
|
Methods
|
|---|
After institutional ethics committee approval and informed, written consent, 30 healthy volunteers were selected for inclusion in the study. All subjects underwent a clinical ear examination, including otoscopy. Those with a clinical hearing abnormality and those taking medications thought to interact with midazolam were excluded. All subjects were asked to refrain from alcohol for 48 h before testing. Unilateral testing was performed in each subject. Because clinical hearing was normal in each ear, the test ear was randomly selected in all cases.
A double-walled soundproofed room (Industrial Acoustics Corporation, Stains, Middlesex, UK) meeting specifications for permissible ambient noise (9) served as the test environment. Auditory sensitivity was measured using standard clinical pure tone audiometry in accordance with the British Society of Audiology standard (10). An audiometer recently calibrated with earphones equipped with cushions was used. Thresholds for air-conducted (headphone) signals were measured in decibels hearing level over 10 frequencies (125, 250, 500, 1000, 1500, 2000, 3000, 4000, 6000, and 8000 Hz).
Speech audiometry was performed to measure auditory discrimination using phonetically balanced prerecorded monosyllabic word lists (11). A cassette recorder connected to an audiometer delivered the speech material in calibrated volume intensity levels. The word lists, comprising 12 lists of 12 monosyllables (consonant-vowel-consonant; for example dog, toy, cat) were chosen to reflect the normal range of speech frequencies. The lists were delivered monoaurally via headphones at varying volume intensity levels. The percentage of phonemes correctly heard and repeated was recorded for each intensity used; results were then plotted as a function of performance versus intensity, producing a speech audiogram.
Using the speech audiogram, the intensity level of the speech stimulus was measured at the point at which performance was half of the maximal discrimination score obtained; this is known as the half peak-level (12). The difference between the expected half-peak level (from population studies) and the measured half peak level is termed the half-peak level elevation (HPLE) (12). The HPLE is a measure of the reduction in an individual's auditory sensitivity to the spoken word, and it gives an estimation of the clinical handicap produced by a given hearing deficit (12).
Audiological reaction times were measured for subjects to a 1-kHz pure tone signal at two intensity levels: 10 and 50 dB above their measured threshold level (i.e., 10 and 50 dB sensation levels). Four measurements were made at each intensity level using variable waiting length intervals to minimize subject anticipation. The duration of the delivered stimulus was the same in all cases. The audiometer was modified to measure the time from stimulus presentation, initiated by the audiologist, to the subject response of pressing a button similar to that used in conventional audiometry. This reaction time was recorded on a digital clock in milliseconds.
After instruction on the procedure involved, each subject underwent an initial test period comprised of pure tone audiometry, speech audiometry, and audiological reaction time measurement. Subjects were then randomly assigned to one of the two groups. Subjects in Group A (n = 15) received midazolam (0.04 mg/kg) IV over a 2-min period; Group B (n = 15) subjects received a similar volume of placebo (isotonic sodium chloride solution) IV over the same time period. Pure tone audiometry, speech audiometry, and audiological reaction times were remeasured 5 min later because midazolam has a rapid onset of hypnotic activity after IV injection (1–2 min) (13). The subjects, the investigator administering the IV drug, and the testing audiologist were blinded to the group assignment. Testing was completed within 25 min of midazolam/placebo administration in all cases.
Data were analyzed using DataDesk® Version 6 (Data Description Inc., Ithaca, NY) to construct random effects general linear models (sometimes referred to as a repeated measures by nesting). In these models, variation among subjects is explicitly entered as a random factor, allowing the application of post hoc tests, which are not available after a conventional repeated-measures analysis. Before analysis, data were checked for distribution. The pure tone audiometry values were close to a normal distribution, with a correlation of 0.962 between scores and expected normal values. The distribution of HPLE scores also correlated well with the normal distribution (P = 0.991). Reaction times, however, were markedly non-normal in distribution. They were reexpressed as reciprocals, which can be thought of as reaction speeds (1/time). The resulting variables correlated well with the expected normal distribution (r = 0.995 for reaction time at 50 [RT50] and 10 dB [RT10] above the auditory threshold). The effect of treatment was evaluated using the two-way interaction between treatment and time. Scheffé post hoc tests were used to assess the significance of changes in each group.
|
Results
|
|---|
Audiological testing data were complete in 30 volunteers. Two reaction times were omitted from analysis. Reaction times of 0.138 s and 1.32 s were associated with extreme externally studentized residuals of –6.5 and 7.0, respectively. These were therefore omitted from the analysis as being unlikely to be true reaction times. The patient demographics were similar in both groups (Table 1). The pre- and posttest scores of the subjects are included in Table 2. The audiometry scores were averaged over all frequencies, and the reaction time measures were averaged over all three of the repeat measures made at each intensity level.
Audiometric performance was analyzed with frequency, treatment (placebo or midazolam), and time (pre- or posttreatment) as fixed factors and subject as a random factor nested within treatment. Although frequency had a highly significant effect on response (F = 13.3, df = 9,559; P < 0.0001), there was no significant three-way interaction among time, treatment, and frequency, which indicates that the effect of treatment (i.e., midazolam) was not frequency-dependent. There was no interaction between treatment and time (F = 0.8, df = 1,559; P = 0.379). In fact, Scheffé post hoc tests revealed that there was no pre- to posttest change in audiological performance (across all frequencies) in either the placebo group (P = 0.194) or the midazolam group (P = 0.957). Limiting the analysis to frequencies up to 2000 Hz (the frequencies used in everyday speech) did not affect the results.
There was no significant interaction between treatment and time (F = 2.1, df = 1,28; P = 0.154) for HPLE performance, which indicates that the change in scores from pre- to posttest was not related to treatment (midazolam or placebo).
For RT10, there was a significant interaction between time and treatment (F = 11.5, df = 1,204; P = 0.0008), which indicates that the change in reaction times differed between midazolam and placebo patients. Although the placebo group's reaction times decreased significantly from pretest to posttest measurements (Scheffé post hoc test P = 0.023), the reaction times of the midazolam group were significantly longer (Scheffé post hoc test P = 0.012). With RT50, there was also a significant treatment x time interaction (F = 14.2, df = 1,202): the placebo showed no significant change (Scheffé post hoc test P = 0.309), whereas the midazolam group showed a significant lengthening (Scheffé test P < 0.0001).
|
Discussion
|
|---|
In this study, we demonstrate that there is no significant alteration in pure tone or speech audiometric performance after IV midazolam administration; consequently, there is no change in peripheral auditory function, as measured by audiometry, with this dosage of the benzodiazepine. However, testing of the pure tone and speech audiogram thresholds was performed in 5-dB increments, as is recommended procedure (10). It may be that a subtle difference in hearing after the administration of midazolam would only be detected by smaller incremental increases in stimulation intensity. However, a 20-dB loss of sensitivity is conventionally accepted as the point at which an auditory disability begins (14,15). Alternatively, a larger dose of midazolam might produce a more exaggerated reduction in auditory thresholds, although the concomitant increase in the degree of sedation would diminish subject cooperation, making accurate measurement of audiological performance difficult. In addition, midazolam is administered in a dosage of 0.04 mg/kg (approximately 3 mg IV in a 70-kg person) to provide anxiolysis in clinical practice. All of the repeat audiometric testing was completed within 25 min of the placebo/midazolam administration, as this approximates a single redistribution half-life of the latter drug (16). However, plasma concentrations of midazolam would undoubtedly have varied during the time course of the study.
Assessment of hearing for pure tones provides valuable information regarding sensitivity but only limited information concerning receptive auditory communication ability (i.e., the functional handicap produced by a given decrease in auditory performance). Testing of speech discrimination by audiometry offers a means of assessing speech understanding under controlled conditions. A normal performance on such tests does not necessarily imply a normal ability to hear and discriminate speech and sound in everyday life (12). Tests that require speech discrimination tasks in the presence of background noise, such as speech babble, could be a better indicator of the patient's ability to comprehend relevant auditory stimuli in the clinical environment (12).
The results of this study support the use of verbal associations or commands when investigating the effects of midazolam on memory, provided that such studies are performed under the appropriate control conditions. External stimuli (audiological, visual, tactile, etc.) were carefully controlled in the course of this study, thereby permitting investigation of the effect of midazolam on auditory perception under ideal conditions. Until further work is completed in this area, such conditions should exist when testing the effect of midazolam on memory using auditory transmitted information.
The change in audiological reaction time after the administration of midazolam is consistent with other studies, which have shown that it causes a slowing of reaction time and a deterioration in psychomotor performance (3,17–19). The delay in reaction time demonstrated in this study may be caused by midazolam-induced motor incoordination, given that audiological performance was unimpaired. However, midazolam causes a delay in sensing the auditory stimulus, which may account, at least in part, for the prolongation of the reaction time. This is supported by studies showing an increased latency period of the AEP after the administration of midazolam (3,19,20). Our results show that both groups responded faster to the 50-dB stimulus and that the reaction time was prolonged in the midazolam group for both the 10-dB and 50-dB stimuli (Table 2, Fig. 1). However, the placebo group responded faster at both intensity levels compared with their control performance, which was presumably the effect of learning. The prolongation of auditory reaction times has implications for the discharge of patients after sedation, as their ability to respond to auditory warning signals (traffic, verbal instructions, etc.,) is impaired. The exact duration of such impairment was not examined in this study, nor was the dose-response relationship of the effect.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. Reaction time (RT) results. The mean RT in each group is shown at each of the two intensity levels tested in the form of a boxplot. The placebo group reacted faster at both intensity levels postinjection, presumably the effect of learning. The midazolam group, however, reacted significantly slower postinjection.
|
|
In summary, sedative doses of midazolam do not alter audiometric performance, but they do delay subject response to audiologic stimuli for an unknown length of time. Whether normal audiometry equates with normal auditory perception in everyday life requires further investigation. Based on the information available, using auditory stimuli or instructions is appropriate when investigating the effect of midazolam on memory, provided that the appropriate conditions exist. Medical practitioners should advise midazolam-sedated patients of their impaired reaction to auditory warning signals (e.g., traffic and car horns) as part of the day-ward discharge recommendations.
|
Footnotes
|
|---|
Presented in part at the annual meeting of the American Society of Anesthesiologists, October 1996, New Orleans, LA.
|
References
|
|---|
-
Schwender D, Klasing S, Madler C, et al. Effects of benzodiazepines on mid-latency auditory evoked potentials. Can J Anaesth 1993;40:1148–54.[Abstract/Free Full Text]
-
Reinsel RA, Veselis RA, Heino R, et al. Effect of midazolam on the auditory event-related potential : measures of selective attention. Anesth Analg 1991;73:612–8.[Abstract/Free Full Text]
-
Lader M. Effects of psychotropic drugs on auditory evoked potentials in man. In: Desmedt JE, ed. Auditory evoked potentials in man: psychopharmacology correlates of evoked potential. Basel:Karger, 1977:142–57.
-
Hall JW III. Effect of stimulus factors. In: Hall JW III, ed. Handbook of auditory evoked responses. Needham Heights, MA: Allyn & Bacon, 1992.
-
McPherson D. Neuropathological findings of the long latency auditory evoked potentials. In: McPherson D, ed. Late potentials of the auditory system. San Diego, CA:Singular Publishings, 1996:101–22.
-
Eisenberg L, Taub HA, Burana A. Memory under diazepam-morphine neuroleptanesthesia in male surgical patients. Anesth Analg 1974;53:488–95.[Abstract/Free Full Text]
-
Russel IF. Midazolam-alfentanil : an anaesthetic? An investigation using the isolated forearm technique. Br J Anaesth 1993;70:42–6.[Abstract/Free Full Text]
-
De Roode A, Jelicic M, Bonke B, Boville JG. The effect of midazolam premedication on implicit memory activation during alfentanil-nitrous oxide anaesthesia. Anaesthesia 1995;50:191–4.[ISI][Medline]
-
American National Standards Institute.Criteria for permissible ambient noise during audiometric testing. ANSI 1997;53:1–4.
-
British Society of Audiology.Recommended procedures for pure tone audiometry using a manually operated instrument. Br J Audiol 1981;15:213–6.[Medline]
-
Boothroyd A. Developments in speech audiometry. Sound 1968;2:3–10.
-
Lutman ME. Diagnostic audiometry. In: Kerr AG, ed. Scott-Brown's otolaryngology. Vol 2. London: Butterworths, 1987.
-
Smith MT, Eadie MJ, O'Rourke Brophy T. The pharmacokinetics of midazolam in man. Clin Pharmacol 1981;19:271–8.
-
Hall JW III. Basic audiometry, handbook of auditory evoked responses. Needham Heights, MA:Allyn & Bacon, 1992:337.
-
Neary WJ, Newton VE, Laoide-Kemp SN, et al. A clinical, genetic and audiological study of patients and families with unilateral vestibular schwannomas. II. Audiological findings in 93 patients with unilateral vestibular schwannomas. J Laryngol Otol 1996;110:1120–8.[ISI][Medline]
-
Allonen H, Ziegler G, Klotz U. Midazolam kinetics. Clin Pharmacol Ther 1981;30:653–61.[ISI][Medline]
-
Bill KM, Fee JPH, Clare RSJ. Psychomotor and clinical assessment of flumazenil as an antagonist of midazolam. J R Soc Med 1991;84:277–9.[Abstract]
-
Thapar P, Zacny JP, Choi M, Apfelpaum JL. Objective and subjective impairment from often used sedative/analgesic combinations in ambulatory surgery, using alcohol as a benchmark. Analg 1995;80:1092–8.[Abstract]
-
Domino EF, French J, Pohorecki R, et al. Further observations on the effects of subhypnotic doses of midazolam in normal volunteers. Psychopharmacol Bull 1989;25:460–5.[ISI][Medline]
-
Sarma SK, Bradshaw EG, Pandit SK, et al. The relation between lorazepam-induced auditory amnesia and auditory evoked potentials. Anesth Analg 1988;67:526–33.[Abstract/Free Full Text]
Accepted for publication January 20, 1999.
Top
Abstract
Introduction
