Ι Ο J O U R N A L O F T H E S O U T H A F R I C A N L O G O P E D I C S O C I E T Y O C T O B E R The Role of Speech Audiometry in Practical Otology By D. R. HAYNES, M.B., B.Ch. (Rand), D.L.O. (R.C.P. & S.) Registrar, Johannesburg General Hospital. It is only recently that a method for measur- ing hearing in terms of speech discrimination has p a s s e d from the field of laboratory re- search to the practical field within easy reach of the practising otologist. We may well ask how large a part machines are entitled to play in the practice of medicine and I have heard audiometry condemned by one of the eminent American otologists on the grounds that it does not make a diagnosis and it is not necessary in the treatment of ear disease. In both respects I think he is wrong, but his being wrong is not quite a s obvious a s his being right appears to be. The majority of cases of Meniere's disease may be diagnosed on the history alone. No- body, however, would think of making a diagnosis of the disease without carrying out a complete clinical examination of the patient, including tuning fork tests. Nobody should diagnose the condition without carrying out tests of labyrinthine function. And, when all this has been done, there will remain a small number of c a s e s in which audiometry gives the final clue to diagnosis when the other tests have failed. The course of disease is of tremendous im- portance in its treatment and in assessing the progress of hearing or of deterioration of hearing, audiometry plays perhaps an even greater part than it does in the diagnosis. Modern neuro-otology, a s this branch of surgery is called by Cawthorne, would scarcely be justified if it were not' for our ability to measure and record patients' hear- ing. Speech audiometry by itself is not a com- plete measurement of hearing ability and in the first place pure-tone audiometry should be employed. There are a great many advant- a g e s to the use of speech audiometry but un- fortunately it cannot provide all the answers usually required in assessing hearing. The disadvantages of speech audiometry are for the most part of a technical nature. It is easy, for instance, to obtain a perfectly satisfactory pure-tone audiogram from an un- intelligent subject. In fact, my experience has been that the unintelligent subject usually gives a more consistently reproducible audio- gram than the intelligent, introspective type of person. If standard, comparable results are to be obtained in speech audiometry, the subjects must be reasonably intelligent and, to a cer- tain extent, educated. In this regard, I must mention the speech audiograms that were taken of a professor of l a n g u a g e s at one of the American universities. His mother tongue was German but he taught in English and French and spoke all three l a n g u a g e s fluently. In spite of this the speech audiograms in the three l a n g u a g e s showed a very much better discrimination for the German word lists than for either of the other two. Here w a s a highly educated, intelligent man whose audiograms for the extra languages might have been interpreted a s showing a hearing loss when there was none in fact. In exactly the same way, differences in dia- lect may give results in speech audiometry for uneducated subjects which are not com- mensurate with the pure-tone audiograms. To overcome these difficulties to a certain sxtent, the monitored live voice may be em- ployed instead of calibrated records but this does not altogether obviate the undersirable features of the test. It becomes obvious now that assessment of hearing must be done with pure-tone and speech audiograms side by side. A good exercise is to forecast the speech audiogram after examining the amount of residual hearing for the accepted speech fre- quencies on the pure-tone audiogram. This can usually be done with a fair degree of accuracy in the c a s e of conductive deafness. With many cases of perceptive deafness, how- ever, the forecast is found to be grossly in- accurate and this applies particularly in those cases exhibiting the phenomenon of recruit- ment. Here it is of interest to mention that an in- telligent adult can obtain a high discrimina- tion score even if he has no hearing for fre- quencies above 500 c.p.s. Using the master hearing aid such a s that used for research in hearing aids by the British Medical Research Council, it is possible to cut out all frequencies above 500 c.p.s. and yet obtain good speech audiograms, even with p.b. word lists. In practice, speech audiograms can be done quickly, especially when a rough assessment R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) J O U R N A L O F T H E S O U T H A F R I C A N L O G O P E D I C S O C I E T Y of hearing for speech is made according to the pure-tone audiogram. In c a s e s of con- ductive deafness, the curve will be parallel to the normal curve. In perceptive deafness, the curve flattens out a s the percentage of discrimination increases and it may never reach the 100% level. When the phenomenon of recruitment is present, the percentage dis- crimination actually falls with the increase of intensity above a certain level and it is there- fore essential in cases of recruitment to com- plete the audiogram in order to get the whole picture. CONDUCTIVE DEAFNESS. From the patient's point of view, the most important aspect of hearing is his hearing for speech. With conductive deafness, the speech audio- gram follows a regular pattern and can be a s s e s s e d with a fair degree of accuracy from the pure-tone audiogram. Our experience with fenestration surgery has led us to believe that if we wish accur- ately to determine the improvement or other- wise of hearing following surgery we must have speech audiograms a s well a s pure tone audiograms. The best illustration of this is a feature that I have noticed repeatedly with fenestrations, regardless of how expert the surgeon might be. In attempting to a s s e s s the suitability of c a s e s for fenestration, the otologist usually regards the patient with a rising curve in the high frequencies on pure tone audiograms a s having the better prospects of success than the patient with a straight curve or with a curve dropping in the high frequencies. Yet there is a very large proportion of fenestrated cases whose post-operative pure- tone audiograms show a very successful rise in the hearing over the speech frequencies but a marked! drop in the high frequencies. In other words, the post-operative curve is reversed. If an a g g r e g a t e of all frequency changes were to be made in these cases, the sum would probably b e zero. The speech audiograms give a completely different picture in these successful fenestra- tions and the curve is merely moved bodily to the left which is much more in keeping with the general clinical expectations. This fact is then a very important one in making pre- and post-operative speech audio- grams an essential part of fenestration surgery. PERCEPTIVE DEAFNESS. The attempt to match hearing aids to pure- tone audiograms has been a failure. This may b e possible in cases of conductive deaf- ness but with perceptive deafness trial and error becomes a necessity for fitting a hearing aid. Once again, the speech audiogram is in itself a very much better indication of suit- ability for hearing aid than is the pure-tone audiogram by itself. If recruitment is shown to be present on a speech audiogram, then it is at once obvious that pure amplification of sound over all frequencies will not benefit the patient. These are the c a s e s in which there is an indication for'modification of the hearing aid by means of automatic loudness control. The speech audiometer in a c a s e of this sort acts a s a master hearing aid and if the speech audiogram shows a marked degree of recruitment it may be assumed that the patient will not benefit from the use of a hear- ing aid. RECRUITMENT. Speech audiograms in themselves may give a very much better indication of the degree of recruitment present than the more tedious and difficult tests of loudness balance. They have the further a d v a n t a g e of indicating recruit- ment when both ears are affected, whereas the pure-tone audiograms are of use only when one ear is affected and the other ear normal. If loudness balance tests are to be reliable, they must be carried out over several frequen- cies and the time necessary to do this is con- siderable. Furthermore, the patient is easily fatigued and usually two or three attempts are necessary to complete the test. The only condition in which recruitment reg- ularly occurs is Meniere's Disease and for this reason speech audiometry is an essential part of the investigation of this condition. In spite of the fact that most modern audio- meters are fitted with loudness balance attach- ments, the only satisfactory way of carrying out accurate tests for loudness balance is by the use of two separate, equally calibrated or corrected audiometers or their equivalent. The tests should be done for at least three fre- quencies and because of this the test is scarcely to be regarded a s of great value to the practising otologist. Speech audiometry is readily available nowadays to any otologist and speech audio- grams are easily done so that here is a prac- tical indication of recruitment that has a tre- mendous diagnostic value. Occasionally, a mild degree of recruitment is present in cases of presbyacusis and it may, of course, occur immediately after trauma to R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) J O U R N A L O F T H E S O U T H A F R I C A N L O G O P E D I C S O C I E T Y O C T O B E R the labyrinth. In the latter condition, the recruit- ment tends to disappear with the p a s s a g e of time after the trauma. CHILDREN. Speech audiometry has a special place in the assessment of hearing in children. Pure- tone audiometry is notoriously inaccurate in children and it is extremely difficult to repro- duce accurate audiograms. Speech audiograms, however, may be very successfully employed in children of a v e r a g e intelligence over the a g e of four years. If records are made with a carrier phrase, such a s "You will now say . . .," there is very little difficulty in obtaining accurate audio- grams which maintain the interest of the child. These audiograms are used in children up to the a g e of 10 or 12 years and in the occas- ional c a s e s of Meniere's Disease encountered in this a g e group, the above remarks on the assessment of recruitment apply even more strongly. SOCIAL ADEQUACY INDEX. The Social Adequacy Index is designed to give an indication of the intelligibility oi speech for a particular subject. It is a com- bined index of hearing ability and interpre- tation and the pure-tone audiogram plays no part in arriving at the index. While there is no practical value to assess- ment of percentage of hearing, the Social Adequacy Index has a definite value and gives a truer reflection of "social hearing ability" than does a percentage figure b a s e d upon pure-tone audiometry. Unfortunately, there are far too many factors which influence hearing and the S.A.I, takes into account only two of them, so that it can- not truly be regarded a s an index of social adequacy. The index can b e determined if speech dis- crimination does not reach the 100% level even at the highest amplification. It cannot be determined, however, if recruitment is pre- sent and the percentage discrimination actually decreases with an increase in ampli- fication. MALINGERING. Numerous tests for feigning deafness have been devised and naturally some tests are better than others, but the malingerer simulat- ing deafness may be difficult to detect with certainty. A new test is being carried out in Sweden in which the speech audiometer is used and it a p p e a r s to b e almost impossible to simulate deafness effectively when this test is em- ployed. The suspected malingerer is given an article to read and a tape recording of his speech is made. This recording is played back to the patient after a time lag of one third of a second. If he is in fact malingering, he finds it impossible to concentrate on what he is reading and begins to falter and stammer and becomes utterly confused within a few seconds. CONCLUSION. I have attempted to show that speech audio- metry should be regarded a s an essential part of modern otology. Only a few years ago, otology consisted for the most part of surgery of suppurative ear disease. In the last 15 years, great advances in otologic surgery have coincided with ad- vances in the treatment of infections of all kinds. Contrary to frequently expressed opinion, the relief from treatment of suppurative ear conditions has made possible a greater con- centration on the surgical treatment of deaf- ness and has thereby widened the scope of otology. Lempert designed the modern fenestration operation and thereby brought practical hear- ing within the reach of millions of sufferers from otosclerosis. Hallpike described the pathology of Meniere's Disease and Caw- thorne designed his operation for labyrinth- ectomy which can be regarded a s a cure for unilateral Meniere's Disease. There is a considerable emphasis at the present time on the plastic surgery of the tympanic membrane and it is to be hoped that these procedures will b e productive of more useful hearing in the future. We have to await the results of the opera- tion for mobilization of the stapes in otos- clerosis and we may be on the brink of a further big advance here. Because of the advances in otologic sur- gery, new standards have been set in all the auxiliary services and we have now reached the stage where speech audiometry is an essential auxiliary service. I do not suggest that b e c a u s e we have speech audiometry, that the older methods of investigation and examination may be dis- carded. On the contrary, I emphasise that complete examination of hearing now con- sists of clinical examination, including tuning fork tests, pure-tone and speech audiograms. It is only by accepting this a s a new stand- ard in practical otology that we can hope to achieve the best results from management of the deaf patient, ALL THE TIME. R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2)