Comparison of conventional pure tone audiometry and 2f1f2 distortion product otoacoustic emissions i

1
Comparison of conventional pure tone audiometry
and 2f1-f2 distortion product otoacoustic
emissions in TB sufferers receiving
aminoglycosides

• Lucretia Petersen Shajila Singh
• Division of Communication Sciences and Disorders
• University of Cape Town

2
Introduction

• Identification and management of TB national
  priority in SA
• In 1993, TB most commonly reported infectious
  disease in SA with a national prevalence of 223
  per 100 000
• Prevalence in Western Cape (1993) of 703 per 100
  000

3
Introduction (cont.)

• WHO Report (2003) prevalence in 2001 gt 300 per
  100 000 nationally
• Epidemic of TB boosted by concurrent HIV
  infection (WHO, 2003)
• HIV increases susceptibility for TB infection
• Chances of MDR-TB greater with increased
  likelihood of reinfection

4
Introduction (cont.)

• MDR-TB TB that is resistant to isoniazid and
  rifampicin which form part of standard TB
  treatment regimen (Fätkenhauer et al. 1999)
• Increasing bacterial resistance to standard
  anti-TB drugs has led to more frequent reliance
  on aminoglycosides like streptomycin and
  kanamycin
• Known side effects nephrotoxicity, ototoxicity
• Despite side effects aminoglycosides are still
  widely used due to effectiveness against
  gram-negative bacteria causing TB, as well as low
  cost

5
Introduction (cont.)

• These advantages have lead to persistence of
  aminoglycoside use, especially in developing
  countries like SA
• With respect to the inner ear, aminoglycosides
  may either injure the cochlea or vestibular
  system
• In SA streptomycin and kanamycin form part of
  drug regimen administered to MDR-TB sufferers
• Streptomycin primarily vestibulotoxic, with
  minor cochleotoxic effect (Bennett, 1996)
• Kanamycin predominantly cochleotoxic (Voogt
  Schoeman, 1996)

6
Introduction (cont.)

• Incidence of ototoxicity varies between 0-20
  depending on type of aminoglycoside
• Aminoglycosides affect outer hair cell (OHC)
  integrity in the cochlea, with initial damage in
  the basal area that could progress to the more
  apical region
• Hearing changes due to OHC destruction are most
  likely irreversible
• Damage to auditory system can continue even after
  cessation of drug administration (Hall, 2000)

7
Introduction (cont.)

• Risk factors include genetic susceptibility,
  prior noise exposure, young or old age,
  pre-existing sensory hearing loss, renal
  problems, concomitant treatment of nephrotoxic or
  other ototoxic drugs, and prior aminoglycoside
  treatment
• Risk for developing cannot be predicted due to
  intersubject variability, thus monitoring the
  auditory functioning of all individuals receiving
  aminoglycosides is imperative
• An auditory monitoring procedure that permits
  early identification of ototoxicity (prior to
  subjective detection by individual) could
  potentially prevent permanent communication
  deficits

8
Introduction (cont.)

• Until recent years ototoxicity could only be
  monitored by conventional pure tone audiometry,
  i.e. between 250-8000Hz
• Behavioural test audiologist has to rely on
  active participation of patient, which can be a
  problem in very ill patients
• Could potentially be a time-consuming procedure,
  especially where responses are unreliable
• Late detection of cochlear damage, i.e. when
  frequencies in speech range (lt 8000 Hz) are
  affected, whereas initial damage occurs beyond 10
  kHz

9
Introduction (cont.)

• Also likely that normal behavioural sensitivity
  to pure tones in quiet circumstances can depend
  on responsiveness of only a few OHC, IHC and
  associated eighth nerve fibres
• Thus deterioration in frequency selectivity of
  the cochlea can occur before pure tone threshold
  elevation, therefore conventional pure tone
  audiometry of limited clinical value in
  ototoxicity monitoring
• High frequency audiometry involves obtaining
  behavioural auditory thresholds of frequencies
  between 8-20 kHz
• HFA has been established as a sensitive method
  for early detection of ototoxicity (Voogt
  Schoeman, 1996)

10
Introduction (cont.)

• Special audiometers and transducers needed that
  are capable of producing high frequency stimuli -
  more expensive than standard audiometers and
  transducers a problem in a country like South
  Africa where resources are limited
• HFA cannot be used to the exclusion of
  conventional pure tone audiometry, because normal
  or unchanged high frequencies do not guarantee
  unchanged thresholds in the conventional range
• Time-consuming procedure problem with ill
  patients

11
Introduction (cont.)

• HFA might also yield problems with the elderly,
  because a vast portion of the geriatric
  population might not have sufficient high
  frequency hearing for high frequency monitoring,
  due to presbyacusis
• Behavioural procedure, thus has to rely on active
  participation of testee
• Otoacoustic emissions (OAEs) constitute the only
  non-invasive means of objective cochlear
  investigation to date (Stavroulaki et al., 1999)
• OAEs are sounds that originate due to activity of
  the OHCs in the cochlea, and are measured in the
  external auditory meatus (Kemp, 1978)

12
Introduction (cont.)

• The two types of evoked OAEs used most often
  clinically are transient evoked OAEs (TEOAEs) and
  distortion product OAEs (DPOAEs)
• Both are low level sounds emitted from the
  cochlea as a result of acoustic stimulation
  (Hall, 2000)
• The presence of evoked OAEs is associated with
  normal, or near normal, mechanically active,
  functioning of the (OHCs)
• TEOAEs are elicited by brief broadband stimuli,
  e.g. clicks or tonebursts, which simultaneously
  stimulate the cochlea across a wide frequency
  region thus representing an averaged overview of
  cochlear activity

13
Introduction (cont.)

• DPOAEs are defined as acoustic energy in the ear
  canal arising from the non-linear interaction of
  two simultaneously applied pure tones, of
  frequency f1 and f2, within the cochlea
• F1 and f2 are known as primaries, with
  intensities L1 and L2 respectively.
• Several combination tones may be predicted
  mathematically with such stimulation, but the
  tone at the frequency 2f1-f2 is the most robust
  and the easiest to detect in human ears (Beattie
  Bleech, 2000)

14
Introduction (cont.)

• Generation of 2f1-f2 depends upon active
  non-linear processes within the cochlea at the
  frequency place where f1 and f2 interact. When
  this cochlear region is altered, then DPOAEs are
  either reduced in amplitude or eliminated
• Changes in DPOAE-amplitude are specific to the
  frequencies affected by the damage, and the
  amplitude of DPOAEs generated in other frequency
  regions remains unchanged
• Research is suggesting that DPOAEs are more
  frequency-specific than TEOAEs for determining
  minor cochlear dysfunction, thus better tool to
  use in monitoring ototoxicity

15
Introduction (cont.)

• Currently 2 methods of DPOAE-measurement are used
  in clinical applications, namely DPgram and DP
  input/output function (growth function) (Ozturan
  Lam, 1998)
• With appropriate stimulus parameters, DPOAEs can
  be measured in nearly all normally hearing ears,
  providing that middle ear status is normal (Hall,
  2000).
• Stimulus parameters influence the amplitude of
  the DPOAE, as well as the region of the cochlea
  being stimulated
• High stimulus levels (? 70 dB SPL) tend to
  stimulate passive cochlear processes, whereas
  moderate level stimuli elicit active cochlear
  responses

16
Introduction (cont.)

• Relationship of L1 and L2 L1gtL2 yields highest
  DP amplitudes, thus sensitivity to cochlear
  deficits is enhanced
• Stimulus levels of L165 and L255 are generally
  used in clinical applications
• Frequency separation of the primaries also plays
  a critical role in DPOAE measurement, an f2/f1
  ratio of 1.22 yields maximum DPOAE amplitude
  between 1-4 kHz
• Clinically effective f2/f1 ratio is between 1.20
  and 1.23 (Hall, 2000).

17
Introduction (cont.)

• Support for the notion that OAEs act as early
  detectors of imminent cochlear dysfunction, are
  clinical observations in humans of reduced OAEs
  in normal-hearing patients, who typically have
  been exposed to noise, or potentially ototoxic
  drugs, and who complain of hearing difficulties,
  which are not picked up by conventional pure tone
  audiometry. Therefore the assumption is made that
  the DPOAE amplitude indicates the summed activity
  of a substantial number of OHCs, rather than only
  a few OHCs as in the case with conventional pure
  tone audiometry (Arnold et al., 1999).

18
Introduction (cont.)

• To date 122 institutions in Western Cape where
  aminoglycoside treatment is provided to TB
  sufferers ototoxicity monitoring takes place at
  only one of these
• Because of limited resources monitoring takes
  place once a month at this one institution, using
  conventional pure tone audiometry ototoxicity
  often detected late
• Even less frequent testing with ill patients
• Purpose of this study compare use of DPOAEs and
  conventional pure tone audiometry as ototoxicity
  monitoring tools, to develop a more efficient
  testing protocol

19
Methodology

• The study set out to determine
• the time difference between changes in
  conventional pure tone audiometry and DPOAEs due
  to cochlear changes during ototoxicity monitoring
  
• the sensitivity of the DPgram in comparison to
  that of the DP I/O function in ototoxicity
  monitoring
• whether the clinical use of DP I/O function is
  feasible

20
Methodology (cont.)

• Design one-tail repeated measures-
  aminoglycoside intervention study
• Subjects were recruited from a TB-hospital in the
  Western Cape after consent was received from the
  superintendent of the institution in question
• The resident audiologist weekly provided a list
  of newly admitted patients with normal hearing
  who were treated with either streptomycin or
  kanamycin. If patient consented to participation
  and fitted the selection criteria, they were
  included in the study

21
Methodology (cont.)

• Selection criteria
• In-patient at the TB-institution, and scheduled
  for aminoglycoside treatment for minimum of 14
  days
• Normal otoscopy, middle ear functioning and
  hearing thresholds
• Able to yield reliable responses during pure tone
  audiometry
• DPOAEs present for at least 5 out of 9
  frequencies
• 16 years or older

22
Methodology (cont.)

• Subjects
• N 17
• 11 females 6 males
• Average age 31years range 16-58 years
• HIV ve 5
• MDR-TB 8
• Prior aminoglycoside treatment 9
• Prior noise exposure 3

23
Methodology (cont.)

• Ethics
• Informed consent of superintendent and all
  subjects
• Could withdraw at any stage, without treatment
  being affected
• Ethical consent from UCT Health Sciences Ethics
  Committee

24
Methodology (cont.)

• Procedure
• Non-diagnostic otoscopy, screening tympanometry,
  conventional pure tone audiometry, DPgram, DP I/O
  function performed to obtain baseline measures
• Repeated once every two weeks for two months
  total of 5 test sittings
• DPgram stimulus parameters
• 1000-6000 Hz
• F2/f1 ratio1.22
• L165 L255 dB SPL

25
Methodology (cont.)

• Procedure
• DP I/O stimulus parameters
• 1000-6000 Hz
• F2/f1 ratio1.22
• L1gtL2, with 10 dB difference
• Intensity range 45-75 dB

26
Methodology (cont.)

• Equipment
• Heine Minilux otoscope
• Madsen Screen-Tymp
• Amplaid audiometer with TDH-39 earphones
• Otodynamics ILO 92
• All testing in sound-treated booth

27
Methodology (cont.)

• Data analysis
• Descriptive statistics mean, mode, range,
  standard deviation
• Future analysis will involve non-parametric
  statistics, including Friedmans ANOVA, factor
  analysis, and multiple regression analysis
• SPSS statistical package
• Qualitative analysis will involve examining data
  for trends relating to changes over time in pure
  tone thresholds, amplitudes of DPgram and DP I/O
  function, as well as in DP I/O thresholds

28
Results

• Preliminary results visual comparison of mean
  changes over time in pure tone audiometry, DPgram
  and DP I/O function

29
Results (cont.)
30
Results (cont.)
31
Results (cont.)
32
Results (cont.)
33
Results (cont.)
34
Results (cont.)
35
Discussion

• Changes in pure tone thresholds seen at Test 5
• With DPgram cochlear damage visible at Test 4,
  thus earlier detection than with conventional
  pure tone audiometry
• Consistent with previous research that found
  DPOAE testing to be more effective in early
  detection of cochlear damage due to ototoxicity
• DP I/O results inconclusive, with no clear
  patterns/trends observable
• More research with lower stimulus levels in DP
  I/O function might yield different/clinically
  useful results

36
Discussion

• Study with more regular test intervals, and
  greater subject sample will provide more precise
  information regarding ideal test protocol
• More homogenous subject sample advised

37
Conclusion

• The use of aminoglycosides for tuberculosis
  treatment is unavoidable in most cases
• It is therefore the responsibility of the
  audiologist to ensure that frequent monitoring of
  these patients occurs using reliable techniques
• Thus we need to constantly explore more efficient
  ways of monitoring, in order to do more with the
  limited resources in our country
• Only then will ototoxicity be detected early and
  the negative side effects avoided or alleviated

38
Acknowledgements
NIH