|Year : 2017 | Volume
| Issue : 2 | Page : 98-103
Comparison of performance across transcranial contralateral routing of signal, trimmer digital and digital bone-anchored hearing implant (with headbands) in children with unilateral hearing loss
N Devi, Suma Chatni, K. J. S. Ramadevi, D Baba Fakruddin
Department of Audiology, All India Institute of Speech and Hearing, Mysore, Karnataka, India
|Date of Web Publication||14-Jun-2017|
All India Institute of Speech and Hearing, Mysore, Karnataka
Source of Support: None, Conflict of Interest: None
Background: Children with unilateral hearing loss (UHL) of severe profound degree find it challenging to locate a sound source and to comprehend speech in the presence of noise, like in classrooms. The objectives of the present study were to compare the available rehabilitative options for children with UHL and arrive at a probably appropriate option. Methods: Fifteen children with unilateral severe sensorineural or profound hearing loss participated in the study. The sound field thresholds, speech perception in the presence of noise and in quiet were assessed with the digitally programmable bone-anchored hearing implant (BAHI), the trimmer digital BAHI and the transcranial contralateral routing of signal (T-CROS). Results: BAHI and T-CROS both provide significant benefit for the children with UHL. However, the children performed better with the BAHIs than the T-CROS both in quiet and noise in most of the measures. It was also noted that the digitally programmable and the trimmer digital BAHI both gave similar performance. Conclusions: BAHI was a better option in both quiet and noise than the T-CROS as a rehabilitation option for children with UHL. The current study helps in prioritizing the rehabilitation options while considering trial and also helps in narrowing down at the appropriate rehabilitative option for the children with UHL.
Keywords: Bone-anchored hearing implant, children and speech perception, transcranial contralateral routing of signal, unilateral hearing loss
|How to cite this article:|
Devi N, Chatni S, Ramadevi K, Fakruddin D B. Comparison of performance across transcranial contralateral routing of signal, trimmer digital and digital bone-anchored hearing implant (with headbands) in children with unilateral hearing loss. Indian J Otol 2017;23:98-103
|How to cite this URL:|
Devi N, Chatni S, Ramadevi K, Fakruddin D B. Comparison of performance across transcranial contralateral routing of signal, trimmer digital and digital bone-anchored hearing implant (with headbands) in children with unilateral hearing loss. Indian J Otol [serial online] 2017 [cited 2020 Aug 13];23:98-103. Available from: http://www.indianjotol.org/text.asp?2017/23/2/98/208032
| Introduction|| |
Unilateral hearing loss (UHL) refers to having normal hearing sensitivity in one ear and hearing loss in the other ear. This loss can range from mild to severe form of degree. The impacts of the UHL are more in children compared to adults. The children with UHL might find difficulty in various conditions such as listening to speech in the noisy environment, localizing a sound source that arrives from the direction of the poorer ear, and understanding speech from a distant source. These children will also have poorer performance in their academics,, particularly with language-based subjects such as reading, writing, and spelling. The children will also have difficulty maintaining attention, and following classroom instructions, these may in turn have a devasting effect on their psychological, social and academic progress, inattention, distractibility, lack of cooperation, and aggression., However, a hearing aid (HA) might be an option for the child identified with UHL.
McKay  carried out a retrospective survey to determine whether fitting an HA for a child with mild to moderately severe UHL improved the quality of life. All parents of the participants were satisfied with the performance of their child with the HA. It was concluded that fitting a child with UHL with an HA will improve the child's quality of life when compared to no hearing amplification. There are other different options to fit in amplification for an individual with UHL such as frequency modulation (FM) system, contralateral routing of signal (CROS), HA, bone-anchored hearing implant (BAHI), and cochlear implantation (CI). The latest FM transmitters possess technologies such as digital noise reduction and directional microphone which provide relief to the individuals with UHL in adverse listening conditions. However, research is limited to assess the usefulness of FM devices in individuals with UHL.
CROS HA has its clinical limitation as it requires the individual to wear behind-the-ear (BTE) units on both ears, which many would not prefer. Apart from cosmetic demerits, it has also been found that children wearing CROS HAs perform poorer in noise than those wearing FM systems or those without any hearing device but who are given preferential seating in the classroom. This can be because some children wearing CROS HAs might lack the ability to recognize when the microphone placed on the poorer HA is facing the noise sources and spontaneously correct their position. Updike  compared the efficacy of conventional air conduction (AC) HAs, conventional CROS HAs, and FM devices in the rehabilitation of six children with UHL ranging in age from 5 to 13 years. The word recognition ability and sound discrimination ability of the participant were measured in quiet and at 6 dB signal-to-noise ratio (SNR) in unaided condition and with conventional AC HAs, conventional CROS HAs, and FM devices. Results reveal that fitting a conventional AC HA to a participant with UHL of any degree higher than mild does not significantly improve word recognition ability both in quiet and may affect adversely in the presence of noise. Similar results were obtained with the conventional CROS with word recognition scores decreasing in the presence of noise. However, results showed that the participants benefited with FM system both in quiet and in the presence of noise and the participants with greater loss reaped more benefit.
Bosman et al. compared the conventional CROS and implanted BAHI in nine participants with unilateral sensorineural hearing loss after a month's habituation of each of the devices. Localization, speech in noise, and subjective evaluations were carried out. The localization test results revealed no difference between the aided and the unaided conditions. It was also observed that the speech perception performance in the presence of noise with the BAHI and the CROS was comparable and better than unaided condition. Moreover, the subjective evaluation results revealed that most of the participants preferred the BAHI. Hol et al. carried out a similar study with twenty participants and compared the performance of conventional CROS and BAHI after a habituation period of 1 month with both devices. The BAHI was given for preimplantation trial with the headband for 1–2 weeks. Sound localization and lateralization performance were assessed using a 9 speaker array, SNR-50 was measured using Dutch sentences presented in the presence of spectrally shaped noise. The subjective evaluation using abbreviated profile of hearing aid benefit was carried out after a month of usage with each of the devices. Results showed that there was no difference between the aided and unaided conditions in terms of localization performance. Overall, participants performed better with the BAHI as compared to the conventional CROS in the presence of noise and poorest in unaided condition. Subjectively also BAHI scored better than the conventional CROS.
Recently, CI is also gaining increased acceptance as a rehabilitation option in children with UHL as it tries to provide some of the benefits of binaural hearing. Távora-Vieira and Rajan  evaluated three children with congenital hearing loss and one with a sudden UHL for speech perception in noise and sound localization ability after implantation. The authors reported that CI may benefit in improving binaural hearing if implantation occurs within the critical period.
As there is a dearth of studies which have analyzed the means of rehabilitation for children with UHL, the most appropriate rehabilitation option still remains a question. Even though the BAHI is used for children with UHL, not much research has been done to check its benefit and more in-depth research is required to generalize the results to the complete population of UHL. In addition, with the rapid development in the technology, more and more new rehabilitation devices and options need to be thoroughly examined before being recommended to the target population. A comparison of the available rehabilitative options is also required to arrive at the best possible rehabilitation means for an individual in particular. However, in case of individuals with UHL, there is a shortage of such reports. Therefore, the current study was formulated with the aim of comparing the performance of the children with UHL with three rehabilitative options, namely, the digitally programmable BAHI, the trimmer digital BAHI, and the transcranial CROS (T-CROS). Since it was not feasible to assess the outcome with the CI with the available resources and clinical setup, the present study aimed at assessing the performance with BAHI attached to headband and CROS.
The present study specifically aimed at comparing the efficacy of the digitally programmable BAHI, trimmer digital BAHI, and the T-CROS in improving the listening performance in children with UHL.
| Methods|| |
The participants consisted of 15 children (9–14 years, mean age =12 years, 8 males and 7 females) who had UHL with either severe sensorineural hearing loss or profound hearing loss in the poorer ear and normal or near normal hearing sensitivity (pure-tone audiometry [PTA] <25 dBHL) in the better ear. The causes of UHL in these participants were mumps, measles, and idiopathic. The participants were native speakers of Kannada language with adequate speech and language skills, and none of them had a prior experience with hearing amplification. Audiological testing, as well as the recording of the test stimulus, was conducted in an acoustically treated room with noise levels within permissible limits. The PTA and speech audiometry were carried out in a double room suite.
In the present study, all the testing procedures were approved by the Institutional Review Board. The procedures involved in the present study were noninvasive, and all the procedures were explained to their family members before testing, and informed consent was taken from the family members of the participants.
Optimizing the amplification devices for each participant
Measurement of transcranial contralateral routing of signal
A high gain digitally programmable BTE HA was used to aid the poorer ear along with a silicone earmold custom-made for the participant. The HA was programmed to obtain maximum output using the desired sensation level (input/output) prescriptive formula. Whenever there was any trace of feedback, a feedback test was run to eliminate it.
The attenuation due to the head shadow effect has been found to be lower at frequencies below 1 kHz and higher at frequencies above approximately 1.5 kHz.,, Consequently, in individuals with single-sided deafness (SSD), sounds of frequencies below 1.5 kHz are heard by the better ear without marked attenuation when a sound source is placed at the side of the poorer ear. However, the sounds which are above 1.5 kHz, from the same source, undergo significant attenuation due to the head shadow effect. Therefore, in the present study, HA gain for the low-frequency sounds below 0.75 kHz was attenuated. The microphone of the HA was set to the omnidirectional mode, and the noise reduction feature was switched off to keep the features across the devices uniform as BAHI 2 lacked both directionality and noise reduction feature. Following programming, it was verified that the HA gain was enough to cross over and be heard to the better ear. The procedure for measuring the transcranial thresholds (TCTs) and the real-ear-aided response (REAR), which was measured to assess the benefit of the T-CROS, is as used by Valente et al. Instructions were given to the participant to keep their head steady during the testing with T-CROS.
Measurement of transcranial thresholds
As a first step, leveling of the probe microphone from Fonix 7000 (Frye Electronics, Tigard, OR, USA) a real-ear analyzer was carried out. Subsequently, the custom-made soft earmold was placed in the ear canal of the poorer ear along with the probe tube, placed at 5 mm from the earmold's tip. The tube of an insert earphone (ER-3A) was connected to the tubing of the earmold along with an adapter and pure tones of frequencies 0.5, 1, 1.5, 2, 3, and 4 kHz were routed through the insert earphone from a calibrated audiometer (Maico MA-52), and thresholds were obtained. Further, the Fonix 7000 analyzer was set to “calibrate probe microphone” mode, and at the threshold values that were obtained, the sound pressure level (SPL) in the poorer ear canal was measured with a continuous signal being presented. The SPL values obtained represented the TCTs, which helped in assessing if the REARs exceeded the TCTs at input levels of 50, 70, and 80 dB SPL.
Measurement of real-ear-aided response
The custom ear mold was placed in the poorer ear along with the probe tube at 5 mm from the tip of the mold. The HA was connected to the tube of the earmold. Subsequently, the speaker of the Fonix 7000 real-ear analyzer was leveled, and the obtained TCTs were entered into the analyzer as the hearing threshold levels. The target gain was displayed according to the NAL-NL1 prescriptive formula. Further, the REAR was obtained for digi speech presented at 50, 70, and 80 dB SPL through a loudspeaker placed at ear level at a distance of one foot from the head.
Programming of bone-anchored hearing aids
Bone-anchored hearing implant 1
The BAHI was programmed to the better ear thresholds of the participant. As previously, mentioned, the low-frequency attenuation was incorporated into BAHI as well with a cutoff frequency of 0.75 kHz. This step was further supported by the finding that BAHIs are likely to induce more of distortion at lower frequencies. Pfiffner et al. reported that low-frequency attenuation till 1.5 kHz in the BAHI sound processor will not affect the performance in individuals with SSD when the noise comes from the front; however, in turn brings down the unfavorable effects of noise coming from the BAHI side. The authors observed that this reduction of the harmful effects increased along with the increase in low-frequency attenuation. The microphone was set to omnidirectional mode, and noise reduction feature was switched off to keep the features of the devices uniform. The volume control was switched off and position compensation was turned on. Further, before testing, the BAHI was connected to the headband and kept on the mastoid of the poorer ear.
Bone-anchored hearing implant 2
The BAHI 2 was optimized based on the Ling sounds identification for each participant. The frequencies below 0.75 kHz were attenuated using the trimmer controls in the device.
Assessing the usefulness of the devices
Sound field warble tone thresholds
Sound field thresholds were obtained for warble tones of 0.5, 1, 2, and 4 kHz presented through a loudspeaker placed at 45°and 1 m distance from the participant toward the test ear. The participant was instructed to indicate whenever the tone was heard, no matter how soft it was. Threshold was considered to be the lowest intensity at which the participant responded positively 50% of the time. The sound field warble tone thresholds were obtained in unaided and three-aided conditions. Both in aided and unaided conditions, the better ear was blocked by both ear plug and ear muffs (noise reduction rating = 39 dB). Blocking the better ear was considered a better option to avoid the direct participation of the better ear since presenting noise to the better ear would adversely affect the performance in case of the transcranial conduction, in turn reducing the performance of the amplification devices. The amplification devices used in the study include T-CROS, with BAHI 1 and BAHI 2. The aided conditions were randomized to avoid the order effect. In total, four sets of warble tone thresholds were obtained for each participant.
Speech identification score in quiet
Speech identification score (SIS) was obtained in unaided and each of the aided conditions using the PB word list developed by Vandana and Yathiraj. The loudspeaker was placed at a distance of 1 m from the participant at 45°azimuth toward the side of the test ear. The better ear was blocked by ear plug and ear muffs to avoid its participation. The word list was presented keeping the microphone of the audiometer at a distance of 6–7 inches from the tester at an intensity of 40 dB HL. The participant was asked to repeat the words understood. The SIS was calculated based on the total number of words correctly repeated.
Signal to noise ratio-50 in direct and indirect conditions
SNR-50 is the difference between the intensity of the speech noise and the intensity of the speech presentation in dB when the participant repeats back at least two words in a set of three words presented in the presence of competing speech noise. The SNR-50 was obtained for each of the aided and unaided modes, in two conditions (Direct: Signal presented toward the better ear and noise toward the poorer ear and Indirect: Signal presented toward the poorer ear and noise towards the better ear).
The participant was positioned at a distance of 1 m from the loudspeaker and asked to sit on a chair. The SNR-50 was obtained using a list of forty sets of Kannada bi-syllabic words. The list was presented at 40 dB HL, and the speech noise was initially presented at an intensity 15 dB lower than the speech signal. Further, the speech noise was altered in 1 dB steps. The participant was asked to repeat the words heard, where at each level of noise, a set of three words were presented. If the participant repeated at least two of the three words, then the level of noise was increased by 1 dB. If they failed to do so, then the noise level was decreased by 2 dB. This manipulation of the noise level was carried out until the tester found out the highest level of noise where the participant could repeat two of the three words correctly. At that point, the difference between the level of speech and the speech noise was considered as SNR-50.
| Results and Discussion|| |
The mean and standard deviation values for each of the measures (sound field thresholds, SIS, and SNR-50) obtained in each of the conditions (unaided and three-aided conditions) were calculated and have been tabulated in [Table 1].
|Table 1: Mean and standard deviation for the sound field thresholds, speech identification score, and signal to noise ratio-50 scores obtained in unaided and three-aided conditions|
Click here to view
From [Table 1], it is evident that the mean sound field thresholds for the aided condition are better than those for the unaided condition. The table also shows that among the aided conditions, the thresholds are the best with the BAHI 1 and are slightly better at lower frequencies with all the devices. It can also be observed that the SIS in quiet condition is better with BAHI 1 compared to all other devices. It is known that lower SNR values in the SNR-50 test indicate better performance. It can be noted that SNR-50 values are lower in the direct condition than the indirect. In the direct condition, it can be seen that the T-CROS had better SNR values than other devices, which indicates that the T-CROS might pose lesser interference in terms of speech perception when the signal arrives from the better ear and the noise from the device's side. However, there are no noticeable differences seen across the devices in indirect condition.
Nonparametric tests were considered for the analysis since the data were variable indicating heterogeneity. To check whether the hearing devices were significantly different from each other and unaided versus aided conditions, analysis of the sound field thresholds, SIS, and the SNR-50 values was carried out using the Freidman test. Consequently, wherever Friedman test showed significance, pairwise comparison was conducted using Wilcoxon's signed rank test. The results of the Freidman test analyzing the sound field thresholds, SIS, and SNR-50 values in children have been tabulated in [Table 2].
|Table 2: Comparison of the unaided and aided conditions on the basis of sound field thresholds, speech identification score, and signal to noise ratio-50 values in children|
Click here to view
It can be inferred from [Table 2] that the listening conditions were significantly different (P < 0.05) from each other as measured in terms of sound field thresholds, SIS obtained in quiet condition, and SNR-50 direct. However, the listening conditions did not differ significantly in terms of SNR-50 indirect measure. This indicates that the aided condition was not significantly different from the unaided condition in the SNR-50 indirect condition, and there was no significant benefit from the hearing devices in the presence of noise. Further, Wilcoxon's signed-rank test was applied wherever significant differences were obtained across the listening conditions. [Table 3] shows the results of the Wilcoxon's signed-rank test for the sound field thresholds.
|Table 3: Pairwise comparison of the unaided and aided conditions for sound field thresholds in children using Wilcoxon's signed-rank test|
Click here to view
Wilcoxon's signed-rank test for the sound field thresholds revealed that the unaided sound field thresholds were significantly different from those of aided condition for all the measures with the aided scores being better. However, the results varied across frequencies for the aided condition. It was observed that at 500 Hz both the BAHI devices performed better than the T-CROS device. Similar results were obtained by Shastri, who compared the performance of BAHI attached to a headband and AC HA in individuals with conductive and mixed hearing loss. At 1000 and 2000 Hz, all the devices performed more or less to the same extent, and at 4000 Hz, BAHI 1 scored better than the BAHI 2 and T-CROS.
The results of Wilcoxon's signed rank test for the SIS obtained in the quiet condition and for the SNR-50 obtained in the direct condition for the children are as displayed in [Table 4].
|Table 4: Pairwise-comparison of the unaided and aided conditions for speech identification score and signal to noise ratio-50 direct values in children|
Click here to view
[Table 4] represents the finding that the unaided SIS scores were significantly different from the aided scores. It is also evident that the scores with both the BAHI devices were significantly different than those with the T-CROS HA, with BAHI scoring better. Niparko et al. also obtained significantly higher scores with the BAHI compared to the conventional CROS in individuals with unilateral hearing as measured in quiet condition. With respect to the SNR-50 measured in direct condition, only BAHI 1 made a significant difference from the unaided condition. In addition, BAHI 1 differed significantly from the BAHI 2 and T-CROS; however, there was no significant difference between BAHI 2 and T-CROS. From [Table 1], it can be inferred that the children required higher SNR with BAHI 1 to achieve the 50% score compared to the unaided and the other aided conditions. This means that the BAHI 1 was having a significant adverse effect on the speech perception when the signal was given to the better ear and noise toward the poorer ear.
In a similar study, Bosman et al., who compared the conventional CROS with the implanted BAHI in nine participants with unilateral sensorineural hearing loss, also reported that both the BAHI and the conventional CROS-yielded better responses in terms of speech perception in noise compared to the unaided condition. Further, there was no difference noted between the speech perception in noise performances of BAHI and the conventional CROS. A similar result was obtained in the present study where there was no significant difference in speech perception in noise scores with BAHI 2 and T-CROS, and the aided performance with all the three devices was better than the unaided condition.
| Conclusions|| |
Aided thresholds were significantly better than those of unaided condition at all frequencies. Participants performed better with both the BAHIs at 500 Hz when compared to the T-CROS. At 1000 and 2000 Hz, all the devices performed more or less to the same extent, and at 4000 Hz, BAHI 1 scored better than the BAHI 2 and T-CROS. Aided SIS in quiet was significantly better than unaided with both the BAHI devices being more beneficial than the T-CROS. No significant difference was observed between the two BAHI devices. In terms of SNR-50 direct measure in children, only BAHI 1 was significantly different from the unaided condition. In case of SNR-50 indirect measure, the listening conditions did not differ significantly. However, the above results were obtained using BAHI attached to headband. Hence, these results should be viewed in light of the fact that, while using BAHI attached to the headband, there would be attenuation in the amount of energy transmitted owing to the skin-dampening. Therefore, it can be considered that after implantation of the BAHI, the performance would be further enhanced.
The present study pertaining to the children with UHL throws light on the possible rehabilitation means for the children with UHL. It also helps in deciding the order or the importance to be given to each of the devices while considering trial. The study assists in arriving at the appropriate rehabilitation device for the children since it involves comparing the available rehabilitation options for children with UHL. The results of the study also help in counseling the parents/caregivers of the child regarding the comparative performance of the digitally programmable BAHI, the trimmer digital BAHI, and the T-CROS. It can be deduced from the results of the study that both the BAHI and the T-CROS provided sufficient usefulness compared to the unaided condition. However, it was found that BAHI was better than T-CROS. It can be concluded that the T-CROS can be considered an option for rehabilitation in children with UHL if the parents/caregivers consider the BAHI to be expensive and prefer a non-invasive rehabilitation option.
The authors would like to thank the Director, All India Institute of Speech and Hearing, Mysore, for granting the funds for the study and the participants for their cooperation.
Financial support and sponsorship
These data are a part of a project funded by the AIISH Research Fund
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bess FH, Tharpe AM, Gibler AM. Auditory performance of children with unilateral sensorineural hearing loss. Ear Hear 1986;7:20-6.
Bess FH, Tharpe AM. Unilateral hearing impairment in children. Pediatrics 1984;74:206-16.
Bess FH, Tharpe AM. Case history data on unilaterally hearing-impaired children. Ear Hear 1986;7:14-9.
Brown AS, Holstrum J, Ringwalt SS. Early intervention. Semin Hear 2008;29:196-211.
Holstrum WJ, Gaffney M, Gravel JS, Oyler RF, Ross DS. Early intervention for children with unilateral and mild bilateral degrees of hearing loss. Trends Amplif 2008;12:35-41.
Kenworthy OT, Klee T, Tharpe AM. Speech recognition ability of children with unilateral sensorineural hearing loss as a function of amplification, speech stimuli and listening condition. Ear Hear 1990;11:264-70.
Updike CD. Comparison of FM auditory trainers, CROS aids, and personal amplification in unilaterally hearing impaired children. J Am Acad Audiol 1994;5:204-9.
Bosman AJ, Hol MK, Snik AF, Mylanus EA, Cremers CW. Bone-anchored hearing aids in unilateral inner ear deafness. Acta Otolaryngol 2003;123:258-60.
Hol MK, Bosman AJ, Snik AF, Mylanus EA, Cremers CW. Bone-anchored hearing aid in unilateral inner ear deafness: A study of 20 patients. Audiol Neurootol 2004;9:274-81.
Távora-Vieira D, Rajan GP. Cochlear implantation in children with congenital and noncongenital unilateral deafness: A case series. Otol Neurotol 2015;36:235-9.
ANSI. Maximum Permissible Ambient Noise Levels for Audiometric Test Rooms. American National Standards Institute, S3.1-1999. New York: American National Standards Institute, Ine.; 1999.
Shaw EA. Transformation of sound pressure level from the free field to the eardrum in the horizontal plane. J Acoust Soc Am 1974;56:1848-61.
Kompis M, Dillier N. Simulation transfer functions in a reverberant room including source directivity and head shadow effect. J Acoust Soc Am 1993;93:2779-87.
Algazi VR, Duda RO, Duralswami R, Gumerov NA, Tang Z. Approximating the head-related transfer function using simple geometric models of the head and torso. J Acoust Soc Am 2002;112(5 Pt 1):2053-64.
Valente M, Potts LG, Valente M, Goebel J. Wireless CROS versus transcranial CROS for unilateral hearing loss. Am J Audiol 1995;4:52-9.
Pfiffner F, Kompis M, Flynn M, Asnes K, Arnold A, Stieger C. Benefits of low-frequency attenuation of Baha®
in single-sided sensorineural deafness. Ear Hear 2011;32:40-5.
Wazen JJ, Spitzer J, Ghossaini SN, Kacker A, Zschommler A. Results of the bone-anchored hearing aid in unilateral hearing loss. Laryngoscope 2001;111:955-8.
Vandana S, Yathiraj A. Speech Identification Test in Kannada for Children. Dissertation, University of Mysore; 1998.
Sahgal A, Manjula P. A Comparative Study of Proprietary and Generic Prescriptive Procedures for Nonlinear Hearing Aids. Dissertation, University of Mysore; 2005.
Shastri U. Comparison of Performance Between Air Conduction Hearing Aid and Bone Anchored Hearing Aid Attached to the Headband. Dissertation, University of Mysore; 2010.
Niparko JK, Cox KM, Lustig LR. Comparison of the bone anchored hearing aid implantable hearing device with contralateral routing of offside signal amplification in the rehabilitation of unilateral deafness. Otol Neurotol 2003;24:73-8.
[Table 1], [Table 2], [Table 3], [Table 4]