A Normative Study of Distortion Product Otoacoustic Emissions in Six-year-old School Children

By Clare O'Rourke,Carlie Driscoll, Joseph Kei and Veronica Smyth
Department of Speech Pathology and Audiology,
The University of Queensland,
Australia

carlie.driscoll@uq.edu.au

Phone Int +617 3365 3095 ,
Fax Int +617 3365 1788

Introduction

     Distortion product otoacoustic emission (DPOAE) technology presents as a viable tool for the hearing screening of entry-level school children. Conducting screening at this age is recommended by health care authorities in order to capture those children with hearing impairment that is fluctuating, progressive or late-onset (European Consensus Statement, 1998; NIH Consensus Statement, 1993). The current, typical protocols involve combinations of otoscopy, pure tone screening and tympanometry. However, these are subjective tests that may not be time-efficient or suitable for all children. DPOAEs exhibit great promise as a screening technique due to their objectivity, high frequency sensitivity and frequency specificity, timeliness, non-invasiveness and ease of test administration (Moulin, Bera, & Collet, 1994).
     Prior to accepting DPOAEs as a mass hearing screening method for six-year-old (entry-level) school children, it is necessary that we clearly define the characteristics of DPOAE results in children of this age, including any normally occurring effects on the DPOAE spectrum (e.g., gender, ear asymmetry and history of ear infection). The creation of highly descriptive normative data is, therefore, needed and may also assist in the development of appropriate pass/fail criteria. The current study aimed to investigate the range of DPOAE values found in a large group of six-year-old children in schools.

Materials & Methods

     From Year 1 classes in 23 schools in Brisbane, Australia, 1003 subjects were recruited for this study. All participation was voluntary. The total subject number was later reduced for statistical analysis and normative database construction, using an age restriction (5.5-6.5 years), as age-related effects on DPOAE results have been reported (Lonsbury-Martin et al., 1994; Prieve et al., 1997). Therefore, the total working subject group was comprised of 788 subjects (407 males and 381 females; 1576 ears), with a mean age of 6.0 years (SD = 0.4).

      The Grason-Stadler GSI-60 Distortion Product system, connected to a laptop computer, was used for all DPOAE testing and analysis [see Zapala (1998) for details of this system]. Tympanometry was also performed using the Madsen Zodiac 901 Middle Ear Analyzer. Analysis of DPOAE results was made with the inclusion of subjects who had failed tympanometry, in order to obtain data that was representative of the general six-year-old population and to allow valid investigation of history effects. For comparative purposes, the analysis was repeated with the exclusion of these failing subjects and may be viewed in O'Rourke et al. (2002). An audiologist tested all subjects individually, in a seated position, in non-sound treated rooms within each school. Background noise levels ranged from 34-51 dB A.
       A series of simultaneous pure tone pairs, of frequencies f1 and f2, at intensities of 65 dB SPL (L1) and 55 dB SPL (L2), respectively, were delivered to the test ear and produced a DP-gram. For further details regarding data acquisition, refer to O'Rourke et al. (2002). The measurement parameters of interest in this study included both amplitude of the distortion product inclusive of the noise floor (DP-amp) and signal-to-noise ratio (SNR), defined as DP-amp minus noise floor, at f2 frequencies of 1.1, 1.5, 1.9, 2.4, 3.0, 3.8, 4.8 and 6.0 kHz. However, it is the SNR parameter that will now receive attention, as it is often utilized in clinical practice and it allows us to focus on DPOAE generation without the inclusion of artifactual noise. To investigate the effects of gender, ear asymmetry and ear infection history on SNR, a factorial model, which included 3 factors [gender (male/female), ear (left/right) and ear infection history] and all interactions, was fitted to the data and the significance of any term was assessed using the analysis of variance (ANOVA) for each of the f2 frequencies tested.

Results

     

Table 1 displays a summary of the occurrence of ear asymmetry, gender and history of ear infection effects on DPOAE SNR. It reveals a pattern of effects that differs for the high and low frequencies. In particular, the high frequencies (3.0-6.0 kHz) are characterized by ear asymmetry, gender and history effects, while the low frequencies (1.1-2.4 kHz) are mainly characterized by the history effect.

Frequency f2 (kHz)	Ear effect	Gender effect	History effect
1.1			*
1.5			*
1.9	*		*
2.4			*
3.0	*	*	*
3.8	*	*	*
4.8		*	*
6.0	*	*	*

Table 1: Summary of significant effects on DPOAE SNR results of 1576 ears of six-year-old subjects. Note: * = significant at p<0.05

      Table 2 contains the corresponding means and standard deviation values for DPOAE SNR at each of the f₂ frequencies. This normative data table has been constructed in accordance with the various significant effects that were summarized above. As an example of the mentioned high and low frequency pattern of effects, SNR results at 6.0 and 2.4 kHz will be examined. Results for 6.0 kHz showed a significant difference in SNR across ears [F(1,783) = 8.32, p<0.01]. Right ears exhibited higher SNR values than left ears. For SNR across gender, females displayed greater SNRs than males [F(1, 783) = 12.424, p<0.01]. A significant difference in SNR values was also found across the history condition [F(1,783) = 12.531, p<0.01], with those children with a negative history of ear infection displaying higher values than those with a positive history. SNR results for 2.4 kHz also indicated a significant difference across the history condition [F(1,784) = 25.322, p<0.01], with those children with a negative history of ear infection displaying greater SNR values than those with a positive history. However, all other interactions were not significant at this frequency.

Frequency f2 (kHz)	Condition	Mean (dB)	SD	Mean–1SD	Mean–2SD	No. Ears
1.1	P	0.99	11.93	-10.94	-22.87	710
	N	3.36	10.42	-7.06	-17.48	866
1.5	L/P	16.52	7.23	9.29	2.06	355
	L/N	16.61	6.87	9.73	2.86	433
	R/P	15.71	7.43	8.28	0.85	355
	R/N	17.38	7.01	10.37	3.37	433
1.9	L/M/P	7.52	9.78	-2.26	-12.04	190
	L/M/N	11.32	7.46	3.86	-3.60	217
	L/F/P	9.87	8.36	1.51	-6.85	164
	L/F/N	12.31	8.34	3.96	-4.77	216
	R/M/P	10.07	9.07	1.00	-8.07	190
	R/M/N	12.18	6.77	5.41	-1.37	217
	R/F/P	10.15	9.09	1.06	-8.03	164
	R/F/N	12.27	8.68	3.59	-5.08	216
2.4	P	13.66	7.14	6.51	-0.63	710
	N	15.73	6.21	9.51	3.31	866
3.0	L/M/P	11.68	6.66	5.02	-1.64	189
	L/M/N	13.03	4.42	8.61	4.19	217
	L/F/P	12.55	5.67	6.87	1.20	165
	L/F/N	14.22	4.74	9.48	4.75	216
	R/M/P	12.31	6.11	6.20	0.09	189
	R/M/N	13.64	3.87	9.77	5.89	217
	R/F/P	13.16	6.32	6.83	0.51	165
	R/F/N	14.40	5.51	8.89	3.38	216
3.8	L/M/P	12.64	5.88	6.75	0.87	190
	L/M/N	13.42	5.38	8.04	2.66	217
	L/F/P	13.73	6.97	6.77	-0.20	165
	L/F/N	15.60	5.05	10.55	5.51	216
	R/M/P	12.95	7.03	5.92	-1.11	190
	R/M/N	14.84	4.45	10.39	5.94	217
	R/F/P	14.72	6.59	8.13	1.53	165
	R/F/N	15.77	6.02	9.75	3.72	216
4.8	M/P	13.78	7.56	6.22	-1.34	380
	M/N	15.16	5.64	9.51	3.87	434
	F/P	14.76	8.49	6.27	-2.21	330
	F/N	16.56	6.64	9.92	3.28	432
6.0	L/M/P	10.34	7.55	2.79	-4.76	190
	L/M/N	12.34	6.40	5.94	-0.46	217
	L/F/P	12.96	7.53	5.42	-2.11	164
	L/F/N	13.87	6.48	7.38	0.90	216
	R/M/P	11.51	7.51	3.99	-3.51	190
	R/M/N	13.76	5.08	8.68	3.60	217
	R/F/P	13.12	8.30	4.82	-3.47	164
	R/F/N	14.05	6.90	7.15	0.26	216

Table 2: Effects of ear asymmetry, gender and history of ear infection on signal-to-noise ratio (SNR) at f2 frequencies. L, R, M, F, N and P represent left, right, male, female, negative history and positive history, respectively.

Discussion

     DPOAE results of six-year-old children in school settings have been shown in this study to be influenced by ear asymmetry, gender and history effects. Right ears generally displayed greater SNR values than left ears. Previous studies of DPOAEs in neonates, infants and adults have failed to demonstrate a significant difference between ears (Gordts et al., 2000; Lonsbury-Martin et al., 1997; Molloy et al., 1999).
     Females of the six-year-old group also displayed greater SNRs in general than males, a finding that has also been suggested for adults (Lonsbury-Martin et al., 1997). In comparison, studies of neonates and infants have not yielded any significant DPOAE gender effects (Molloy et al., 1999; Sheppard, Brown, & Russell, 1996). The gender effect was more prominent in the higher frequencies (3.0-6.0 kHz), a pattern that was also found in the examination of ear asymmetry effects on SNR. Conceivably, the general lack of ear asymmetry and gender effects on SNR at the lower test frequencies could be related to higher ambient and physiologic noise levels in this particular frequency range in the non-sound treated test rooms.
     Children with positive history of ear infection evidenced lower SNR values across the frequency spectrum than children with negative history. It is possible that those subjects of the present study with a positive history of ear infections were more likely to have middle ear dysfunction of some degree on the day of testing (as evidenced by a significant difference in tympanometry failure rates across history), leading to reduced DPOAE recordings.
     Mean SNRs ranged from approximately 11 to 16 dB (with the exception of 1.1 kHz). SNR values at 1.1 kHz were very low on average. SNR standard deviation values were large at 1.9 kHz and mean-1SD values were less than 3 dB. The range of mean SNRs contained in this investigation appeared to be slightly lower than those reported for neonates and infants (Molloy et al., 1999). Given the discovery of significant ear asymmetry, gender and history of ear infection effects upon DPOAE results, the use of separate norms for six-year-old children in school settings should be considered.

References

European Consensus Statement. (1998). Scandinavian Audiology, 27: 259-260.

Gordts, F., Naessens, B., Mudde, C.A., & Clement, P.A.R. (2000). Scandinavian Audiology, 29: 79-82.

Lonsbury-Martin, B., Martin, G., McCoy, M., & Whitehead, M. (1994). American Journal of Otology, 15(suppl.1): 13-20.

Molloy, J., Kei, J., Smyth, V., McPherson, B., Young, J., Tudehope, D., Maurer, M., Rankin, G., Latham, S., & Loscher, J. (1999). Australian & New Zealand Journal of Audiology, 21(2): 65-76.

Moulin, A., Bera, J., & Collet, L. (1994). Audiology, 33: 305-326.

NIH Consensus Statement. (1993). NIH, 11: 1-24.

Prieve, B.A., Fitzgerald, T.S., Schulte, L.E., & Kemp, D.T. (1997). Journal of the Acoustical Society of America, 102: 2871-2879.

Sheppard, S., Brown, A., & Russell, P. (1996). British Journal of Audiology, 30: 261-274.

Zapala, D.A. (1998). Journal of the American Academy of Audiology, 9: 332-341.