- Email: [email protected]
Influence of pulsed laser irradiation on the morphology and function of the guinea pig cochlea
Hearing Research 144 (2000) 97^108 www.elsevier.com/locate/heares
In£uence of pulsed laser irradiation on the morphology and function of the guinea pig cochlea S. Jovanovic a
a;
*, Y. Jamali b , D. Anft b , U. Scho«nfeld a , H. Scherer a , G. Mu«ller
c
ENT Department, University Medical Center Benjamin Franklin, Free University of Berlin, Hindenburgerdamm 30, D-12200 Berlin, Germany b Clinic for Otorhinolaryngology, Face and Neck Surgery, Martin Luther University of Halle-Wittenberg, Halle-Wittenberg, Germany c Institute for Medical/Technical Physics and Laser Medicine, University Medical Center Benjamin Franklin, Free University of Berlin, Berlin, Germany Received 3 June 1999; accepted 14 March 2000
Abstract Recent experimental and clinical studies have demonstrated that several pulsed laser systems are also suitable for stapedotomy. The aim of the study was to investigate morphological and functional inner ear changes after irradiating the basal turn of the guinea pig cochlea with two pulsed laser systems of different wavelengths. The Er:YSGG (V = 2.78 Wm) and Ho:YAG (V = 2.1 Wm) lasers were used applying the laser energies necessary for perforating a human stapes footplate. The cochleas were removed 90 min, 1 day, 2 weeks, or 4 weeks after laser application. Acoustic evoked potentials (compound action potentials) were measured before and after laser application and at the above times immediately before removal of the cochleas. The organ of Corti was examined by scanning electron microscopy. Application of Er:YSGG laser parameters effective for stapedotomy had no adverse effects on Corti's organ in the guinea pig cochlea. On the other hand, effective Ho:YAG laser parameters cause damage to the outer hair cells with fusion of stereocilia and formation of giant cilia leading to partial or total cell loss. The inner hair cells and supporting cells were usually normal. These morphological data show a good correlation with the electrophysiological measurements. Our results clearly demonstrate that, besides achieving efficient bone management, the Er:YSGG laser has high application safety. On the other hand, the Ho:YAG laser is not well tolerated in our animal study. Its use in stapedotomy would be unreliable and dangerous for the inner ear. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Laser stapedotomy; Er:YSGG laser; Ho:YAG laser; Guinea pig cochlea; Scanning electron microscopy; Compound action potential
1. Introduction Applying the laser for surgery in the middle ear enables greater precision and control than the manual technique. In the speci¢c case of stapes footplate perforation (stapedotomy), however, the cochlear structures can be directly or indirectly damaged by laserinduced thermal or acoustic transfer of energy. To be safely applied, the laser must therefore be examined for detrimental e¡ects on Corti's organ.
* Corresponding author. Tel.: +49 (30) 8445-2440; Fax: +49 (30) 8445-4141; E-mail: [email protected]
Recent experimental studies have demonstrated that, apart from the continuous wave (cw) lasers, several pulsed laser systems (holmium:YAG (yttrium^aluminum^garnet), erbium :YAG, erbium :YSGG (yttrium^ scandium^gallium^garnet)) may be suitable for stapedotomy (Kautzky et al., 1991; Stubig et al., 1993; Romano et al., 1993 ; Pfalz, 1995; Pratisto et al., 1996; Jovanovic et al., 1990, 1995, 1997, 1998). In a range of short pulse durations and high power densities (about 108 W/cm2 ), the occurrence of so-called non-linear processes leads to a change in the e¡ect of the laser on tissue. The resultant phenomena di¡er markedly from the purely thermal e¡ects of the laser application and lead to changes in the ablation mechanism and marginal e¡ects. This process, which is designated photoablation, takes place at energy densities of
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 0 ) 0 0 0 5 8 - 7
HEARES 3438 12-5-00
98
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
about 0.1^10 J/cm2 and laser pulse durations in the nano- and microsecond range. With this process, the exposure times and thus the duration of the temperature increase are so short that heat conduction is negligible. Tissue ablation is thus achieved with low heating of adjacent structures and thus there are only small thermal-in£uenced zones around the ablated tissue. Preliminary experiments were performed to examine the laser tissue e¡ect in the human petrous bone preparation and the bovine compact bone model and to record the temperature and pressure parameters in the simpli¢ed cochlea model (Jovanovic et al., 1996, 1997, 1998 ; Scho«nfeld et al., 1994). The results of these studies show that the tissue-ablating e¡ect of pulsed laser systems permits precise and controlled stapes footplate perforation through low and readily reproducible ablation rates. An adequately large perforation can generally only be achieved by multiple shots at the same application site, since only a small amount of tissue is ablated per application. Because of a higher ablation rate the Er:YSGG laser (wavelength 2.78 Wm) requires not only a lower pulse count (about ¢ve pulses) but also less energy to achieve an adequately large perforation than the Ho:YAG laser (wavelength 2.1 Wm) which needs about 10 pulses and an approximately ¢ve times higher total energy. The absorption properties and thus footplate e¡ects of the Er:YAG laser (wavelength 2.94 Wm) are comparable to those with the Er:YSGG laser. With these pulsed lasers the extent of thermal side e¡ects at the footplate is lower in comparison to the purely thermal acting cw and superpulse laser systems. Also the quality (shape and structure) of perforations is generally better with pulsed than with cw lasers (Jovanovic et al., 1996, 1997). Investigations of the transmission of thermal energy into the vestibule indicated that thermal loading of the inner ear is based not only on the initial temperature developed at the footplate but also on the heat exchange processes in the perilymph (Jovanovic et al., 1998). With pulsed lasers the ¢rst non-perforating applications of a pulse series do not irradiate the £uid directly, which results in less heating. Direct irradiation of the £uid after perforation then leads to more marked temperature elevations. Through a summation e¡ect, increasing the number of pulses raises the maximal temperature. A higher repetition rate has the same e¡ect, since there is less time for cooling between the pulses, and the initial temperature is not regained before the next application. With multiple application, the highest temperature increments in the £uid of the cochlea model at a distance of 2 mm behind the perforation were measured with the Ho:YAG laser and, in the e¡ective laser range, amounted to 29.9³C (median 26.1³C) above the initial temperature of 37³C. Despite short thermal
exposure times ( 6 10 ms), inner ear irritations cannot be excluded at these high temperatures (67³C). With the Er:YSGG and Er:YAG lasers, on the other hand, the maximal temperature increases in the e¡ective energydensity range are 3.6³C (median 3.6³C) and 5.9³C (median 5.5³C) respectively and thus these appear safe for the inner ear in view of the short heat exposure time (Jovanovic et al., 1998). Furthermore, a short-term local warming with vaporization of the perilymph and photoablation causes turbulent convective £ow in the cochlea and the formation of gas or vapor bubbles whose implosions (cavitation) trigger pressure pulses in the inner ear that spread spherically from the site of origin. These pressure pulses are a stimulus to the basilar membrane and Corti's organ and are thus analogous to the sound input to the inner ear via the tympanic membrane and middle ear and consequently to the mechanism of impulse noise. For pulsed laser systems, the pressure^time course shows a single short pressure pulse whose duration, when considering the `minus 10 dB e¡ective duration' used for such impulse noise, lies in the order of magnitude of the laser pulse length (Er :YSGG and Ho :YAG laser ca. 500 Ws, Er:YAG laser ca. 180 Ws). The comparable peak sound pressure levels are higher than with the CO2 lasers in cw and superpulse mode. The lowest levels were found at 153 dB(SPL) with the Er:YSGG laser and the highest at 175 dB(SPL) with the Er:YAG laser, which has the shortest pulse length and the wavelength most strongly absorbed in water. Pressure levels were 164 dB(SPL) with the Ho :YAG laser and thus 11 dB higher than with the Er:YSGG laser. These results can be interpreted according to the damage risk criteria for impulse noise (according to Pfander, 1975). The `noise dose' for single applications does not exceed the critical values for permanent hearing loss with any of the laser systems. However, these values are sometimes exceeded by multiple applications such as those needed for adequate footplate perforation, since the pulse durations add up to a total exposure time. Due to their high peak levels, the Er:YAG and Ho :YAG lasers exceed the limiting curve for hearing damage, whereas the Er:YSGG laser remains markedly below and can thus be rated as safe. The aim of this animal experiment was to determine whether in vivo application of these lasers may cause inner ear damage and, if so, to what extent. Scanning electron microscopy (SEM) enables precise morphological evaluation of Corti's organ in all turns of the cochlea and thus provides data on the possible type, extent and site of damage. Results obtained by measuring acoustic evoked potentials (compound action potentials (CAP)) after laser application enable a comparison of morphological and electrophysiological data.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
2. Materials and methods The investigations were performed in 48 female albino guinea pigs. Before starting the experiments, Preyer's re£ex was tested, and otoscopy was performed to exclude animals with otitis media or otitis externa from further study. Anesthesia was carried out with intramuscular application of xylazine (Beyer, Leverkusen, Germany; 13 mg/kg body wt.) and ketamine (Parke Davis, Freiburg, Germany; 87 mg/kg body wt.). To avoid hypothermia during anesthesia, all examinations took place on a hot plate adjusted to 37³C. The basal turn of the cochlea at a distance of 1 mm from the round window was chosen as the application site for the laser beam, since its thickness is similar to that of the human stapes footplate and is easily accessible surgically. An Er:YSGG and a Ho:YAG laser (Spektrum Co., Berlin, Germany) were used. Speci¢cations of the Er: YSGG laser included a wavelength of V = 2.78 Wm and a pulse duration [full width at half maximum (FWHM)] of t = 500 Ws and of the Ho:YAG laser a wavelength of V = 2.1 Wm and a pulse duration of t = 500 Ws (FWHM). Since the Er:YSGG laser applied was not designed as a medical laser, a micromanipulator could not be used in the experiments. The irradiation of the Er:YSGG laser was transmitted into the operating area via a light ¢ber with an outer diameter of 400 Wm and a length of about 20 cm (Q-Q-IR 434, Schott Co., Mainz, Germany). The transmission system of the Ho:YAG laser irradiation was a photoconductor (low-OH ¢ber) with a diameter of 400 Wm. With both systems, a non-contact procedure was used for application at a distance of 2 mm from the basal turn of the guinea pig cochlea with a resultant spot diameter of about 550 Wm. Multiple shots were applied at the same application site in order to achieve an adequately large perforation diameter of 500^600 Wm. The treatment groups are listed in Table 1. Laser e¡ects were examined after perforation of the basal turn (Er :YSGG laser group 1) and subsequent irradiation of the open cochlea (other groups). Of particular interest were the e¡ective laser range (Er :YSGG laser group 2 and Ho:YAG laser groups 1 and 2) determined
99
by preliminary experiments in the human stapes for a footplate perforation diameter of 500^600 Wm (Jovanovic et al., 1997), the safety range of the two laser systems during treatment (i.e., in£uence of higher repetition rates (Ho:YAG laser group 2) and the total energy absorbed (Er:YSGG and Ho :YAG laser groups 3)). A control group of eight animals was used to determine whether or nor any intervention and the course of anesthesia caused inner ear changes. With the exception of laser treatment, control animals underwent all operative procedures and measurements. The cochleas were removed 90 min, 1 day, 2 weeks or 4 weeks after laser application (Table 2). They were chosen in such a way as to di¡erentiate reversible from irreversible damage and to detect acute and late changes. The removed and exposed cochlea was perfused with 2.5% glutaraldehyde at room temperature, then ¢xed in a 4% formalin and 1.25% glutaraldehyde solution cooled to 4³C (McDowell and Trump, 1976) and, after 2 days, preserved in 70% ethanol cooled to 4³C until the day of the SEM examination. The preserved cochlea was rinsed twice for 30 min with 0.2 mol cacodylate bu¡er. Dehydration of the cochlea was performed in an acetone series of increasing concentrations (25%, 50%, 75% and 100% acetone) for 1 h in each solution (Soudijn, 1976). Finally it was rinsed three times (10^15 min each) in a ¢nished solution of hexamethyldisilazane for drying. Preparation of the cochlea was done under the stereomicroscope with a micrometer eyepiece at magni¢cation factors of 2.0^6.4-fold. The bony capsule was circularly opened from the apex to the basal turn. The membranous part of the cochlea was left intact. With ¢ne forceps, the spiral ligament was separated from the three upper turns with the exception of the ligament at the level of the basal turn. This remained intact; only its apical margin was removed. Corti's organ could thus be inspected from the helicotrema to the round window. For microscopic visualization, the cochlea was mounted on an aluminum plate with a ¢xative (Go«cke's Leit-C, Neubauer Co., Germany), a thin gold layer was sputtered onto it (MED 010, Balzer Co., Germany). The cochlea was then examined with a scanning elec-
Table 1 Applied laser systems and treatment groups Laser
Group
Spot size (Wm)
Energy Q (mJ)
Energy density H (J/cm2 )
Number of pulses
Total energy (J)
Repetition rate Number of (Hz) animals
Er:YSGG 2.78 Wm, 500 Ws
1 2 3 1 2 3
550 550 550 550 550 550
85 85 85 210 210 210
36 36 36 90 90 90
1 5 25, 50 10 10 20
0.085 0.425 2.1, 4.2 2.1 2.1 4.2
^ 1, 5 5 1 5 5
Ho:YAG 2.1 Wm, 500 Ws
HEARES 3438 12-5-00
8 7 8 7 6 4
100
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Table 2 Number of examined guinea pig cochleas by SEM at di¡erent posttreatment intervals Laser
Group
90 min
1 day
2 weeks 4 weeks
Er:YSGG 2.78 Wm, 500 Ws
1 2 3 1 2 3
2 2 1 1 1 1
2 1 ^ 2 ^ ^
2 2 3 2 1 ^
Ho:YAG 2.1 Wm, 500 Ws
2 2 4 2 4 3
tron microscope (950 SEM, Zeiss Co., Oberkochen, Germany). CAPs were measured to obtain data on possible laser-induced functional e¡ects on the cochlea. Function of the contralateral ear in each animal was eliminated by transtympanic opening of the cochlea to exclude recording action potentials from the non-test ear as well as to facilitate measuring acoustic evoked potentials. The needle electrodes were inserted subcutaneously. One electrode was placed subaurally with the reference electrode on the ipsilateral mastoid, and one hind leg served for grounding. The electrode impedance was kept below 5 k6. Recordings were done in a sound-attenuated and electromagnetically shielded chamber. The electrical impulse (click half width 125 Ws) was acoustically transformed by earphones (DT 48, Beyerdynamic, Heilbronn, Germany). Stimulation was applied at levels of 80, 60, 40, 30 and 20 dB (HL) (humans). The recorded signal (stimulation window 12.5 ms; about 500 averages; ¢lter 200^3000 Hz) was analyzed with respect to the threshold, the threshold shift and the latency of the ¢rst negative CAP peak N1 . Measurements were performed at the following time intervals: before and after opening the bulla as well as 4, 30, 60 and 90 min after laser irradiation and also on the 1st, 2nd, 7th, 14th and 28th postoperative days. Postoperative measurements were carried out up to the day on which the cochlea was removed and prepared for the SEM examinations. The care and use of animals in this study supported by the German Research Foundation (Jo 180/1-1) were approved by the Animal Care Committee of the Ministry of Health in Berlin (TVV 23/91) in accordance with the guidelines of the Declaration of Helsinki. 3. Results 3.1. Control animals The SEM examination of all eight control animals yielded normal ¢ndings for the stereocilia and cuticular plates of the inner and outer hair cells of the basal, second and third turns. In the apex, the stereocilia on
Fig. 1. SEM image of the basal turn of a cochlea 2 weeks after applying Er:YSGG laser irradiation (energy: 5 times 85 mJ; repetition rate: 1 Hz). The stereocilia and cuticular plates of the inner and outer hair cells and supporting cells show normal con¢guration. (A) Inner and outer hair cells. Magni¢cation: 1500-fold. (B) Outer hair cells. Magni¢cation: 3750-fold. (C) Inner hair cell. Magni¢cation: 10 875-fold.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
101
Fig. 2. CAPs of an animal before and after applying Er:YSGG laser irradiation (energy: 5 times 85 mJ; repetition rate: 1 Hz). Normal CAP thresholds and latencies throughout the entire examination period.
the second and third rows of the outer hair cells showed a partial contortion and a collapse as normal variants. The supporting cells had a normal con¢guration in all examined turns. The electrophysiological recordings were likewise normal throughout the entire examination period. 3.2. Er:YSGG laser 3.2.1. Perforation of the basal turn Group 1: The basal turn of eight animals was irradiated once with an energy of 85 mJ (energy density : 36 J/cm2 , t = 500 Ws (FWHM)). As in the human stapes, these laser parameters were su¤cient to create a small perforation (50^100 Wm) in the guinea pig cochlear bone, but not to achieve the required bone perforation diameter of 500^600 Wm. After laser application, the animals showed no SEM abnormalities. The partially irregular structure of the cilia on the second and third rows of outer hair cells with normal visualization of the inner hair cells in the apex is considered to be a normal variant. The electrophysiological recordings showed a normal time course for all animals examined during a period extending from directly after the laser application to 4 weeks later. 3.2.2. Perforation of the basal turn and irradiation of the open cochlea Group 2: The basal turn of seven animals was irradiated with an energy of ¢ve times 85 mJ (total energy : approx. 0.5 J, energy density : 36 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 1 Hz and 5 Hz. The ¢rst application already led to a small perforation,
the others at the same site to enlargement of the perforation and irradiation of the open cochlea. A ¢vefold application with these parameters is necessary for an adequate perforation (diameter 500^600 Wm) of the human stapes footplate. All animals showed normal SEM ¢ndings for the hair and supporting cells in the examined areas of the cochlea (Fig. 1A^C). The electrophysiological examination likewise showed no changes (Fig. 2). 3.2.3. High total energies Group 3: To determine the application safety of the laser system, the basal turn of four animals was irradiated using an energy of 85 mJ/pulse with 25 applications (total energy : approx. 2.1 J, energy density : 36 J/ cm2 , t = 500 Ws (FWHM)) and a repetition rate of 5 Hz. The SEM examination yielded normal ¢ndings for the inner and outer hair cells in all turns of the cochlea. None of the animals showed a threshold shift or latency prolongation of the CAP at the electrophysiological examination. In four animals, the basal turn was irradiated using an energy of 85 mJ/pulse with 50 applications (total energy : approx. 4.2 J) and a repetition rate of 5 Hz. In two animals, the outer hair cells of all turns examined showed fusion of the stereocilia tips with formation of giant cilia (Fig. 3A,B). Findings were normal for the inner hair cells in all turns examined. At the electrophysiological examination, these animals had threshold shifts of 40 dB starting directly after laser application and persisting until the 28th postoperative day. The other two animals had lost all inner and outer hair cells throughout the cochlea. The missing cells had been re-
HEARES 3438 12-5-00
102
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
placed by polygonal epithelial cells (Fig. 4). Electrophysiologically, no CAP could be recorded from 30 min after laser application until the 28th day (Fig. 5). 3.3. Ho:YAG laser 3.3.1. Perforation of the basal turn and irradiation of the open cochlea Group 1: The basal turn of seven animals was irradiated applying 10 laser pulses with an energy of 210 mJ per pulse (total energy : approx. 2.1 J, energy density : 90 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 1 Hz. A 10-fold irradiation of the same application site with the above parameters is necessary for an adequately large perforation of the human footplate. SEM revealed normal ¢ndings for the hair and support-
Fig. 3. SEM image of a cochlea after applying Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Two weeks after laser application, stereocilia tips of outer hair cells in the second and third rows of all turns of the cochlea show giant cilium formation with retraction of the cuticular plates. (A) Second turn. Magni¢cation: 1500-fold. (B) Third turn. Magni¢cation: 3750-fold.
Fig. 4. SEM image of a cochlea after applying Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Four weeks after application, there is a loss of all outer hair cells in the basal turn and replacement by polygonal epithelial cells. Magni¢cation: 750-fold.
ing cells in the entire cochlea in three animals. These animals showed a normal electrophysiological course starting directly after laser application and extending to the 28th postoperative day. The other four animals had fusion of the stereocilia tips with formation of giant cilia, retraction of the cuticular plates and loss of individual groups of outer hair cells. The stereocilia of the inner hair cells were partially splayed and disarrayed (Fig. 6). Electrophysiologically, the latter four animals evidenced a threshold shift of 40 dB starting directly after laser application and extending to the 28th postoperative day (Fig. 7). Group 2: The basal turn of six animals was irradiated applying 10 laser pulses with an energy of 210 mJ per pulse and a repetition rate of 5 Hz. The aim was to examine the in£uence of a higher repetition rate on cochlea morphology and function. SEM revealed changes of di¡erent extent in all animals examined at the various time points. On the one hand, pathological ¢ndings, which included fusion of the stereocilia tips on the outer and inner hair cells, retraction of the cuticular plate and loss of individual outer hair cells, were mainly detected in the basal and second turns (Fig. 8). On the other hand, the slight damage limited to the lasered area was accompanied by the loss of all outer hair cells in all turns, while the inner hair cells showed fusion of the stereocilia tips (Fig. 9). Electrophysiologically, all animals had a threshold shift of 30^40 dB starting directly after laser application. On the ¢rst and second postoperative days, two animals had a severe hearing loss with a threshold shift of v 60 dB. Twenty-eight days after laser treatment, an extreme CAP alteration persisted in four animals (Fig. 10).
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
103
Fig. 5. CAPs of an animal before and after application of Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Recordings show strong CAP threshold shift starting 30 min after applying Er:YSGG laser irradiation.
3.3.2. High total energies Group 3: To determine the application safety of the laser system, an energy of 210 mJ was selected to irradiate the basal turn of four animals with 20 applications (total energy: approx. 4.2 J, energy density: 90 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 5 Hz. After 20 laser applications, the basal and second turns of the cochleas showed a loss of the entire organ of Corti and its replacement by polygonal epithelial cells (Fig. 11). There were irregularly structured hair cells and a loss of individual cells in the third turn. Electrophysiologically, CAP alterations occurred directly after the laser application and led to a hearing loss by the 28th postoperative day (Fig. 12).
iants. About 20% of the cochlea at the apex should therefore be disregarded at the SEM evaluation. CAPs were determined as a measure of inner ear function in guinea pigs. These were readily reproducible with respect to the threshold and latency. The non-invasive technique of needle electrode recording applied here had several advantages over the invasive method of permanent electrode implantation at the round window used for recording cochlear microphonics (CM): it considerably shortened the preparation time and markedly reduced the risk of infection. This enabled reliable measurements over a longer period of time (here
4. Discussion SEM examinations were performed to detect morphological changes in the cochlea after laser application. This method is particularly e¡ective for investigating changes in Corti's organ with the sensory and supporting cells. SEM provides good visualization of the apical surfaces of the outer and inner hair cells with their stereocilia. The apex of the cochlea was not included in the evaluation, since both the controls and laser-treated animals evidenced repeated changes in the stereocilia of the outer and inner hair cells in this part of the cochlea. Findings included torsions, a collapse and irregular structure of the stereocilia. Soudijn (1976) and Rydmarker et al. (1987) also observed such changes in the apex when evaluating Corti's organ after exposure of the guinea pig cochlea to impulse noise. These changes were interpreted as normal var-
Fig. 6. SEM image of the second turn of the cochlea 90 min after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 1 Hz). Outer hair cells show fusion of stereocilia with partially giant cilium formation and loss of outer hair cells in the third row. The stereocilia of the inner hair cells are partially splayed and disarrayed. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
104
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Fig. 7. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 1 Hz). The recordings show moderate hearing impairment that started immediately after applying Ho:YAG laser irradiation and persisted up to 4 weeks thereafter.
4 weeks). The measurements were performed at close intervals after laser application in order to record the immediate e¡ects of laser irradiation on the cochlea and to classify them according to their severity and reversibility. The broad-band acoustic stimulation necessary to evoke the CAP, however, does not permit frequencyspeci¢c di¡erentiation of possible hair cell damage, Fast N1 peaking of CAP mostly re£ects high-frequency components in the basal part of the guinea pig cochlea which can be stimulated by frequencies up to 40 kHz. Cochlear responses from middle and apical areas have
Fig. 8. SEM image of the second turn of the cochlea 2 weeks after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). Fusion of stereocilia tips of the second and third rows of the outer hair cells with retraction of the cuticular plate and loss of individual outer hair cells. The disarrayed stereocilia of the inner hair cells are an artefact. Magni¢cation: 1500-fold.
only a slight in£uence on CAP. More precise information on possible inner ear impairments in adjacent areas can be gained by performing histological and SEM examinations of laser-treated cochleas. On the other hand, the data obtained by recording frequency-speci¢c CM is not accurate enough. Laser application in the basal turn of the guinea pig cochlea was the model selected for laser stapedotomy, as already described by Thoma et al. (1986). It should be noted, however, that anatomic and pathophysiological conditions in laser stapedotomy di¡er from those in this animal experiment. When perforating the human stapes footplate, the most exposed parts of the inner
Fig. 9. SEM image of the basal turn of the cochlea 4 weeks after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). Loss of all outer hair cells and fusion of stereocilia tips of the inner hair cells. The ruptures of the cuticular plate are shrinking artefacts due to ¢xation. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
105
Fig. 10. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). The recordings show a threshold shift of 40 dB starting immediately after laser application, reaching a maximum of s 60 dB on the ¢rst post-treatment day and still persisting as a 60 dB threshold shift on the 28th post-treatment day.
ear are the saccule, the utricle and the cochlear duct, lasering the basal turn of the guinea pig cochlea, however, involves a higher risk to structures of Corti's organ like the basilar membrane, the stria vascularis, the outer and inner hair cells, etc. Since this probably results in higher vulnerability of these anatomic structures in the animal experiment, inner ear damage is more likely to result from laser application in our animal model than would be the case when lasering the human stapes footplate and the vestibulum. If no alterations are found in animal experiments, it is probable that the same laser parameters will cause no injury to human inner ear structure. The Er:YSGG laser (pulse half width: t = 500 Ws (FWHM), spot size approx. 550 Wm) applied once with an energy of 85 mJ did not morphologically change Corti's organ in any cochlear areas examined by SEM. CAP recordings were unchanged (i.e., thresholds and latencies) at all recovery times. A ¢vefold application of Er:YSGG laser irradiation with these parameters (total energy : approx. 0.4 J) is necessary to achieve an adequate perforation of the human stapes footplate (Jovanovic et al., 1997). In this connection, Corti's organ of the cochleas examined in all animals did not show detectably altered sensory and supporting cells in any of the turns. The CAP recordings likewise did not change. The in£uence of higher pulse repetition rates (5 Hz) on inner ear morphology and function was also examined. Here too, Corti's organ of the cochleas examined did not evidence morphological changes in the sensory or supporting cells of any of the turns. The CAP recordings were likewise normal. These results in animals demonstrate the safety of applying Er: YSGG laser irradiation to the basal cochlea with an
energy density of 36 J/cm2 and a total energy of approx. 0.4 J, which are parameters e¡ective for human footplate perforation. To determine the range of safe Er:YSGG laser application, the basal turn of the guinea pig cochlea was irradiated with a much higher total energy than that required for adequate perforation of the human footplate. A 25-fold laser application (¢ve-fold higher total energy : approx. 2.1 J) did not cause pathological changes in any of the cochleas examined by SEM. There were also no CAP changes. It was only after a 50-fold laser application (10 times higher total energy: approx. 4.2 J) that Corti's organ showed pathologically altered sensory and supporting cells in the cochleas ex-
Fig. 11. SEM image of the second turn of the cochlea 4 weeks after applying Ho:YAG laser irradiation (energy: 20 times 210 mJ; repetition rate: 5 Hz). The loss of all hair and supporting cells and their replacement by polygonal epithelial cells. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
106
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Fig. 12. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 20 times 210 mJ; repetition rate: 5 Hz). A threshold shift s 60 dB occurred directly after laser irradiation. No CAP could be recorded 1 day after laser treatment.
amined. There were more marked changes of outer than of inner hair cells, particularly in the second and third rows. These alterations included an irregular structure of the stereocilia and their fusion with formation of giant cilia that exceeded the normal stereocilia in length and thickness. Some of the pathological changes were so marked that they resulted in a total loss of the sensory cells with replacement by polygonal epithelial cells 4 weeks after laser treatment. The cuticular plates of the sensory cells were deformed in the majority of cells. These changes were equally detectable in all cochleas examined at the various post-treatment times. The electrophysiological examination showed moderate to severe CAP alterations, some of which were partially irreversible. In contrast to the inner hair cells, the stereocilia of the outer hair cells are in direct contact with the tectorial membrane. The irregular structure and partial loss of stereocilia can be explained by extreme shearing movements of the stereocilia by the tectorial membrane. This extreme movement can result from impulse noise exposure and may lead to destruction of the tectorial membrane at the site of the inserted stereocilia and of the stereocilia tips themselves (Hamernik et al., 1984 ; Saunders et al., 1985). There may also be a rupture of the reticular lamina with opening of the endolymphatic space followed by mixture of high-potassium endolymph with low-potassium perilymph. This leads to further degeneration and even death of the hair cells (Bohne and Rabbitt, 1983; Duvall and Rhodes, 1967 ; Konishi and Salt, 1979). The additional loss of stereocilia we observed 4 weeks after laser application may be due to this degeneration. The dead hair cells are ultimately discarded into the endolymphatic space. The
defects in the reticular lamina are closed by expansion of the supporting cells (Gao et al., 1992 ; Spoendlin, 1971). The Er:YSGG laser enables nearly athermal bone ablation and only minimally increases the perilymph temperature by about 3.6³C at a distance of 2 mm behind the laser application site (Jovanovic et al., 1997, 1998). On the other hand, footplate perforation involves explosion-like pressure impulses generating pressure and shock waves in the cochlear perilymph which are comparable to peak sound pressure levels of about 153 dB(SPL) (Scho«nfeld et al., 1994). The pathological ¢ndings for hair and supporting cells in our animal experiment closely resemble the morphological changes in these structures after impulse noise exposure and are thus probably due to the pressure impulses rather than to the thermal e¡ect of the laser irradiation (Jovanovic et al., 1998). Ho :YAG laser irradiation (pulse half width: t = 500 Ws (FWHM), spot size approx. 550 Wm), on the other hand, exhibits high damaging potential in guinea pigs. E¡ective laser parameters (total energy : approx. 2.1 J) already altered the inner and outer hair cells in more than 50% of the animals. The changes included torsion and collapse of the stereocilia and fusion of the stereocilia tips with giant cilia formation. The electrophysiological examination revealed moderate to severe CAP alterations. A doubling of the total energy to approx. 4.2 J caused morphological changes extending to a loss of all sensory cells and their replacement by polygonal epithelial cells. Electrophysiologically, the animals thus treated showed severe irreversible CAP alterations. In comparison to Er:YSGG laser irradiation, the morphological inner ear changes after the Ho :YAG
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
laser application are more pronounced. One reason for this high damaging potential may be the poorer ablation capacity of Ho:YAG laser irradiation in bone. This necessitates high energy densities and total energies for an adequately large footplate perforation diameter of 500^600 Wm. Thus, compared to the Er:YSGG laser, the energy density is increased nearly threefold and the total energy ¢vefold. This raises the perilymph temperature about 30³C above the initial value of 37³C and to higher peak sound pressure levels of about 164 dB(SPL). Both in vitro values are critical for the inner ear structures. Another reason may be that the higher optical penetration depth of Ho:YAG laser irradiation in the perilymph could also cause immediate damage to inner ear structures by direct exposure. The optical penetration depth of the Ho:YAG wavelength (V = 2.1 Wm) in water is about 200 Wm and is therefore 200 times higher than that of the Er:YAG wavelength (V = 2.94 Wm). In electrophysiologically controlled animal experimental studies (measuring CM potentials in guinea pigs), Pfalz et al. (1994) and Pfalz (1995) investigated the e¡ects of the Er:YAG laser irradiation. Comparable to our Er:YSGG laser results they also postulated a good tolerance and high safety of the Er:YAG laser for application to the meatal lining near the limbus or to the manubrium mallei. Apart from a single greater temporary threshold shift with 50 applications, pulse counts of below 500 did not involve any irregularities in the CM potentials. Changes in the CM a¡ecting all frequencies with a maximum of 2 kHz were only seen after 500 applications of Er:YAG laser irradiation with an energy of 50 mJ (total energy : 25 J), a pulse duration of 250 Ws (FWHM), a pulse sequence of 2 Hz and a beam diameter of 200 Wm (energy density 160 J/cm2 ). A complete recovery of the CM was observed 90 min after laser application. When applying laser irradiation to the middle ear and auditory ossicles, the author considers even pulse counts of 10 000 or more permissible, which indicates the high suitability of the Er:YAG laser for middle ear surgery. These results are not applicable to the footplate, however, because of other damaging mechanisms involved in direct irradiation of the inner ear. Our results demonstrate that, with the tested parameters, the pulsed Er:YSGG laser shows good tolerance and high application safety in these animal experiments. Its high ablation rate and low damaging potential could render it a useful alternative to the thermal CO2 laser for stapedotomy (Jovanovic et al., 1996, 1999). The applied Ho:YAG laser, on the other hand, has a high damaging potential in our animal experiments. Its application in stapedotomy appears to be unpredictable and dangerous for the inner ear. Thus the conclusions drawn in our in vitro experi-
107
ments (Jovanovic et al., 1997, 1998) regarding the damaging mechanism and the risk of laser damage to the inner ear have been con¢rmed in vivo in relation to the laser parameters applied. However, these results are only valid for the lasers examined in this study, since their applicability to other pulsed laser systems, particularly those with shorter pulse half widths, is still questionable due to possibly higher pressure impulses. Acknowledgements The study was supported by the German Research Foundation (Jo 180/1-1). The authors wish to thank Dr. V. Stiglic, a veterinarian at the Berlin Central Animal Laboratories, for assistance in performing the animal experiments.
References Bohne, B.A., Rabbitt, K.B., 1983. Holes in the reticular lamina after noise exposure. Implications for continuing damage in the organ of Corti. Hear. Res. 11, 41^53. Duvall, A.J., Rhodes, V.T., 1967. Ultrastructure of the organ of Corti following intermixing of cochlear £uids. Ann. Otol. Rhinol. Laryngol. 76, 688^708. Gao, W.Y., Ding, D., Zheng, X., Ruan, F., Liu, Y., 1992. A comparison of changes in the stereocilia between temporary and permanent hearing losses in acoustic trauma. Hear. Res. 62, 27^41. Hamernik, R.P., Turrentine, G., Roberto, M., Salvi, R., Henderson, D., 1984. Anatomical correlates of impulse noise- induced mechanical damage in the cochlea. Hear. Res. 13, 229^247. Jovanovic, S., Scholz, C., Berghaus, A., Scho«nfeld, U., 1990. Experimenteller Vergleich zwischen kurzgepulsten und kontinuierlich strahlenden Lasern in der Stapeschirurgie ^ histologische-morphologische Ergebnisse. Arch. Otorhinolaryngol. II (Suppl.), 73^75. Jovanovic, S., Anft, D., Scho«nfeld, U., Berghaus, A., Scherer, H., 1995. Experimental studies on the suitability of the erbium laser for stapedotomy in an animal model. Eur. Arch. Otorhinolaryngol. 252, 422^427. Jovanovic, S., Scho«nfeld, U., Prapavat, V., Berghaus, A., Fischer, R., Scherer, H., Mu«ller, G., 1996. E¡ects of continuous-wave laser systems on stapes footplate. Lasers Surg. Med. 19, 424^432. Jovanovic, S., Scho«nfeld, U., Prapavat, V., Berghaus, A., Fischer, R., Scherer, H., Mu«ller, G., 1997. E¡ects of pulsed laser systems on stapes footplate. Lasers Surg. Med. 21, 341^350. Jovanovic, S., Scho«nfeld, U., Fischer, R., Do«ring, M., Prapavat, V., Mu«ller, G., Scherer, H., 1998. Thermal e¡ects in the `vestibule' during laser stapedotomy with pulsed laser systems. Lasers Surg. Med. 23, 7^17. Jovanovic, S., Anft, D., Scho«nfeld, U., Berghaus, A., Scherer, H., 1999. In£uence of CO2 laser irradiation of the guinea-pig cochlea on compound action potentials. Am. J. Otol. 20, 1^8. Kautzky, M., Tro«dhan, A., Susani, M., Schenk, P., 1991. Infrared laserstapedotomy. Eur. Arch. Otorhinolaryngol. 248, 449^451. Konishi, T., Salt, A.N., 1979. E¡ects of exposure to noise on ion movement in guinea pig cochlea. Hear. Res. 1, 325^342. McDowell, E.M., Trump, B.F., 1976. Histologic ¢xatives suitable for diagnostic light and electron microscopy. Arch. Path. Lab. Med. 100, 405^414.
HEARES 3438 12-5-00
108
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Pfalz, R., 1995. Eignung verschiedener Laser fu«r Eingri¡e vom Trommelfell bis zur FuMplatte (Er:YAG-, Argon-, CO2 s.p.-, HoYAGLaser). Laryngol. Rhinol. Otol. 74, 21^25. Pfalz, R., Nagel, D., Bald, N., Hibst, R., 1994. Vergleich von BERASchwellen und cochlea«ren Mikrophonpotentialen (CM) vor und nach Er:YAG-Laserung im Mittelohr narkotisierter Meerschweinchen. Eur. Arch. Otorhinolaryngol. II (Suppl.), 36. Pfander, F., 1975. Das Knalltrauma. Springer, Berlin. Pratisto, H., Frenz, M., Ith, M., Romano, V., Felix, D., Grossenbacher, R., Altermatt, H., Weber, H., 1996. Temperature and pressure e¡ects during erbium laser stapedotomy. Lasers Surg. Med. 18, 100^108. Romano, V., Rodriguez, R., Altermatt, H., Frenz, M., Weber, H., 1993. Bone microsurgery with IR-lasers: A comparative study of the thermal action at di¡erent wavelengths. Proc. SPIE 2077, 87^ 97. Rydmarker, S., Nilsson, P., Grenner, J., 1987. Morphological and functional changes in the organ of Corti after noise exposure. Acta Otolaryngol. 441, 3^43.
Saunders, J.C., Dear, S.P., Schneider, M.E., 1985. The anatomical consequences of acoustic injury: A review and tutorial. J. Acoust. Soc. Am. 78, 833^860. Scho«nfeld, U., Fischer, R., Jovanovic, S., Scherer, H., 1994. `La«rmbelastung' wa«hrend der Laserstapdotomie. Eur. Arch. Otorhinolaryngol. II (Suppl.), 244^246. Soudijn, E.R, 1976. Scanning electron microscopic study of the organ of Corti in normal and sound-damaged guinea pigs. Ann. Otol. Rhinol. Laryngol. 29 (Suppl.), 1^46. Spoendlin, H.H., 1971. Primary ultrastructural changes in the organ of Corti after acoustic overstimulation. Acta Otolaryngol. 71, 166^ 176. Stubig, I.M., Reder, P.A., Facer, G.W., Rylander, H.G., Welch, A.J., 1993. Holmium:YAG laser stapedotomy: preliminary evaluation. Proc. SPIE 1876, 10^19. Thoma, J., Mrowinski, D., Kastenbauer, E., 1986. Experimental investigations on the suitability of the carbon dioxide laser for stapedotomy. Ann. Otol. Rhinol. Laryngol. 95, 126^131.
HEARES 3438 12-5-00
In£uence of pulsed laser irradiation on the morphology and function of the guinea pig cochlea S. Jovanovic a
a;
*, Y. Jamali b , D. Anft b , U. Scho«nfeld a , H. Scherer a , G. Mu«ller
c
ENT Department, University Medical Center Benjamin Franklin, Free University of Berlin, Hindenburgerdamm 30, D-12200 Berlin, Germany b Clinic for Otorhinolaryngology, Face and Neck Surgery, Martin Luther University of Halle-Wittenberg, Halle-Wittenberg, Germany c Institute for Medical/Technical Physics and Laser Medicine, University Medical Center Benjamin Franklin, Free University of Berlin, Berlin, Germany Received 3 June 1999; accepted 14 March 2000
Abstract Recent experimental and clinical studies have demonstrated that several pulsed laser systems are also suitable for stapedotomy. The aim of the study was to investigate morphological and functional inner ear changes after irradiating the basal turn of the guinea pig cochlea with two pulsed laser systems of different wavelengths. The Er:YSGG (V = 2.78 Wm) and Ho:YAG (V = 2.1 Wm) lasers were used applying the laser energies necessary for perforating a human stapes footplate. The cochleas were removed 90 min, 1 day, 2 weeks, or 4 weeks after laser application. Acoustic evoked potentials (compound action potentials) were measured before and after laser application and at the above times immediately before removal of the cochleas. The organ of Corti was examined by scanning electron microscopy. Application of Er:YSGG laser parameters effective for stapedotomy had no adverse effects on Corti's organ in the guinea pig cochlea. On the other hand, effective Ho:YAG laser parameters cause damage to the outer hair cells with fusion of stereocilia and formation of giant cilia leading to partial or total cell loss. The inner hair cells and supporting cells were usually normal. These morphological data show a good correlation with the electrophysiological measurements. Our results clearly demonstrate that, besides achieving efficient bone management, the Er:YSGG laser has high application safety. On the other hand, the Ho:YAG laser is not well tolerated in our animal study. Its use in stapedotomy would be unreliable and dangerous for the inner ear. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Laser stapedotomy; Er:YSGG laser; Ho:YAG laser; Guinea pig cochlea; Scanning electron microscopy; Compound action potential
1. Introduction Applying the laser for surgery in the middle ear enables greater precision and control than the manual technique. In the speci¢c case of stapes footplate perforation (stapedotomy), however, the cochlear structures can be directly or indirectly damaged by laserinduced thermal or acoustic transfer of energy. To be safely applied, the laser must therefore be examined for detrimental e¡ects on Corti's organ.
* Corresponding author. Tel.: +49 (30) 8445-2440; Fax: +49 (30) 8445-4141; E-mail: [email protected]
Recent experimental studies have demonstrated that, apart from the continuous wave (cw) lasers, several pulsed laser systems (holmium:YAG (yttrium^aluminum^garnet), erbium :YAG, erbium :YSGG (yttrium^ scandium^gallium^garnet)) may be suitable for stapedotomy (Kautzky et al., 1991; Stubig et al., 1993; Romano et al., 1993 ; Pfalz, 1995; Pratisto et al., 1996; Jovanovic et al., 1990, 1995, 1997, 1998). In a range of short pulse durations and high power densities (about 108 W/cm2 ), the occurrence of so-called non-linear processes leads to a change in the e¡ect of the laser on tissue. The resultant phenomena di¡er markedly from the purely thermal e¡ects of the laser application and lead to changes in the ablation mechanism and marginal e¡ects. This process, which is designated photoablation, takes place at energy densities of
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 0 ) 0 0 0 5 8 - 7
HEARES 3438 12-5-00
98
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
about 0.1^10 J/cm2 and laser pulse durations in the nano- and microsecond range. With this process, the exposure times and thus the duration of the temperature increase are so short that heat conduction is negligible. Tissue ablation is thus achieved with low heating of adjacent structures and thus there are only small thermal-in£uenced zones around the ablated tissue. Preliminary experiments were performed to examine the laser tissue e¡ect in the human petrous bone preparation and the bovine compact bone model and to record the temperature and pressure parameters in the simpli¢ed cochlea model (Jovanovic et al., 1996, 1997, 1998 ; Scho«nfeld et al., 1994). The results of these studies show that the tissue-ablating e¡ect of pulsed laser systems permits precise and controlled stapes footplate perforation through low and readily reproducible ablation rates. An adequately large perforation can generally only be achieved by multiple shots at the same application site, since only a small amount of tissue is ablated per application. Because of a higher ablation rate the Er:YSGG laser (wavelength 2.78 Wm) requires not only a lower pulse count (about ¢ve pulses) but also less energy to achieve an adequately large perforation than the Ho:YAG laser (wavelength 2.1 Wm) which needs about 10 pulses and an approximately ¢ve times higher total energy. The absorption properties and thus footplate e¡ects of the Er:YAG laser (wavelength 2.94 Wm) are comparable to those with the Er:YSGG laser. With these pulsed lasers the extent of thermal side e¡ects at the footplate is lower in comparison to the purely thermal acting cw and superpulse laser systems. Also the quality (shape and structure) of perforations is generally better with pulsed than with cw lasers (Jovanovic et al., 1996, 1997). Investigations of the transmission of thermal energy into the vestibule indicated that thermal loading of the inner ear is based not only on the initial temperature developed at the footplate but also on the heat exchange processes in the perilymph (Jovanovic et al., 1998). With pulsed lasers the ¢rst non-perforating applications of a pulse series do not irradiate the £uid directly, which results in less heating. Direct irradiation of the £uid after perforation then leads to more marked temperature elevations. Through a summation e¡ect, increasing the number of pulses raises the maximal temperature. A higher repetition rate has the same e¡ect, since there is less time for cooling between the pulses, and the initial temperature is not regained before the next application. With multiple application, the highest temperature increments in the £uid of the cochlea model at a distance of 2 mm behind the perforation were measured with the Ho:YAG laser and, in the e¡ective laser range, amounted to 29.9³C (median 26.1³C) above the initial temperature of 37³C. Despite short thermal
exposure times ( 6 10 ms), inner ear irritations cannot be excluded at these high temperatures (67³C). With the Er:YSGG and Er:YAG lasers, on the other hand, the maximal temperature increases in the e¡ective energydensity range are 3.6³C (median 3.6³C) and 5.9³C (median 5.5³C) respectively and thus these appear safe for the inner ear in view of the short heat exposure time (Jovanovic et al., 1998). Furthermore, a short-term local warming with vaporization of the perilymph and photoablation causes turbulent convective £ow in the cochlea and the formation of gas or vapor bubbles whose implosions (cavitation) trigger pressure pulses in the inner ear that spread spherically from the site of origin. These pressure pulses are a stimulus to the basilar membrane and Corti's organ and are thus analogous to the sound input to the inner ear via the tympanic membrane and middle ear and consequently to the mechanism of impulse noise. For pulsed laser systems, the pressure^time course shows a single short pressure pulse whose duration, when considering the `minus 10 dB e¡ective duration' used for such impulse noise, lies in the order of magnitude of the laser pulse length (Er :YSGG and Ho :YAG laser ca. 500 Ws, Er:YAG laser ca. 180 Ws). The comparable peak sound pressure levels are higher than with the CO2 lasers in cw and superpulse mode. The lowest levels were found at 153 dB(SPL) with the Er:YSGG laser and the highest at 175 dB(SPL) with the Er:YAG laser, which has the shortest pulse length and the wavelength most strongly absorbed in water. Pressure levels were 164 dB(SPL) with the Ho :YAG laser and thus 11 dB higher than with the Er:YSGG laser. These results can be interpreted according to the damage risk criteria for impulse noise (according to Pfander, 1975). The `noise dose' for single applications does not exceed the critical values for permanent hearing loss with any of the laser systems. However, these values are sometimes exceeded by multiple applications such as those needed for adequate footplate perforation, since the pulse durations add up to a total exposure time. Due to their high peak levels, the Er:YAG and Ho :YAG lasers exceed the limiting curve for hearing damage, whereas the Er:YSGG laser remains markedly below and can thus be rated as safe. The aim of this animal experiment was to determine whether in vivo application of these lasers may cause inner ear damage and, if so, to what extent. Scanning electron microscopy (SEM) enables precise morphological evaluation of Corti's organ in all turns of the cochlea and thus provides data on the possible type, extent and site of damage. Results obtained by measuring acoustic evoked potentials (compound action potentials (CAP)) after laser application enable a comparison of morphological and electrophysiological data.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
2. Materials and methods The investigations were performed in 48 female albino guinea pigs. Before starting the experiments, Preyer's re£ex was tested, and otoscopy was performed to exclude animals with otitis media or otitis externa from further study. Anesthesia was carried out with intramuscular application of xylazine (Beyer, Leverkusen, Germany; 13 mg/kg body wt.) and ketamine (Parke Davis, Freiburg, Germany; 87 mg/kg body wt.). To avoid hypothermia during anesthesia, all examinations took place on a hot plate adjusted to 37³C. The basal turn of the cochlea at a distance of 1 mm from the round window was chosen as the application site for the laser beam, since its thickness is similar to that of the human stapes footplate and is easily accessible surgically. An Er:YSGG and a Ho:YAG laser (Spektrum Co., Berlin, Germany) were used. Speci¢cations of the Er: YSGG laser included a wavelength of V = 2.78 Wm and a pulse duration [full width at half maximum (FWHM)] of t = 500 Ws and of the Ho:YAG laser a wavelength of V = 2.1 Wm and a pulse duration of t = 500 Ws (FWHM). Since the Er:YSGG laser applied was not designed as a medical laser, a micromanipulator could not be used in the experiments. The irradiation of the Er:YSGG laser was transmitted into the operating area via a light ¢ber with an outer diameter of 400 Wm and a length of about 20 cm (Q-Q-IR 434, Schott Co., Mainz, Germany). The transmission system of the Ho:YAG laser irradiation was a photoconductor (low-OH ¢ber) with a diameter of 400 Wm. With both systems, a non-contact procedure was used for application at a distance of 2 mm from the basal turn of the guinea pig cochlea with a resultant spot diameter of about 550 Wm. Multiple shots were applied at the same application site in order to achieve an adequately large perforation diameter of 500^600 Wm. The treatment groups are listed in Table 1. Laser e¡ects were examined after perforation of the basal turn (Er :YSGG laser group 1) and subsequent irradiation of the open cochlea (other groups). Of particular interest were the e¡ective laser range (Er :YSGG laser group 2 and Ho:YAG laser groups 1 and 2) determined
99
by preliminary experiments in the human stapes for a footplate perforation diameter of 500^600 Wm (Jovanovic et al., 1997), the safety range of the two laser systems during treatment (i.e., in£uence of higher repetition rates (Ho:YAG laser group 2) and the total energy absorbed (Er:YSGG and Ho :YAG laser groups 3)). A control group of eight animals was used to determine whether or nor any intervention and the course of anesthesia caused inner ear changes. With the exception of laser treatment, control animals underwent all operative procedures and measurements. The cochleas were removed 90 min, 1 day, 2 weeks or 4 weeks after laser application (Table 2). They were chosen in such a way as to di¡erentiate reversible from irreversible damage and to detect acute and late changes. The removed and exposed cochlea was perfused with 2.5% glutaraldehyde at room temperature, then ¢xed in a 4% formalin and 1.25% glutaraldehyde solution cooled to 4³C (McDowell and Trump, 1976) and, after 2 days, preserved in 70% ethanol cooled to 4³C until the day of the SEM examination. The preserved cochlea was rinsed twice for 30 min with 0.2 mol cacodylate bu¡er. Dehydration of the cochlea was performed in an acetone series of increasing concentrations (25%, 50%, 75% and 100% acetone) for 1 h in each solution (Soudijn, 1976). Finally it was rinsed three times (10^15 min each) in a ¢nished solution of hexamethyldisilazane for drying. Preparation of the cochlea was done under the stereomicroscope with a micrometer eyepiece at magni¢cation factors of 2.0^6.4-fold. The bony capsule was circularly opened from the apex to the basal turn. The membranous part of the cochlea was left intact. With ¢ne forceps, the spiral ligament was separated from the three upper turns with the exception of the ligament at the level of the basal turn. This remained intact; only its apical margin was removed. Corti's organ could thus be inspected from the helicotrema to the round window. For microscopic visualization, the cochlea was mounted on an aluminum plate with a ¢xative (Go«cke's Leit-C, Neubauer Co., Germany), a thin gold layer was sputtered onto it (MED 010, Balzer Co., Germany). The cochlea was then examined with a scanning elec-
Table 1 Applied laser systems and treatment groups Laser
Group
Spot size (Wm)
Energy Q (mJ)
Energy density H (J/cm2 )
Number of pulses
Total energy (J)
Repetition rate Number of (Hz) animals
Er:YSGG 2.78 Wm, 500 Ws
1 2 3 1 2 3
550 550 550 550 550 550
85 85 85 210 210 210
36 36 36 90 90 90
1 5 25, 50 10 10 20
0.085 0.425 2.1, 4.2 2.1 2.1 4.2
^ 1, 5 5 1 5 5
Ho:YAG 2.1 Wm, 500 Ws
HEARES 3438 12-5-00
8 7 8 7 6 4
100
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Table 2 Number of examined guinea pig cochleas by SEM at di¡erent posttreatment intervals Laser
Group
90 min
1 day
2 weeks 4 weeks
Er:YSGG 2.78 Wm, 500 Ws
1 2 3 1 2 3
2 2 1 1 1 1
2 1 ^ 2 ^ ^
2 2 3 2 1 ^
Ho:YAG 2.1 Wm, 500 Ws
2 2 4 2 4 3
tron microscope (950 SEM, Zeiss Co., Oberkochen, Germany). CAPs were measured to obtain data on possible laser-induced functional e¡ects on the cochlea. Function of the contralateral ear in each animal was eliminated by transtympanic opening of the cochlea to exclude recording action potentials from the non-test ear as well as to facilitate measuring acoustic evoked potentials. The needle electrodes were inserted subcutaneously. One electrode was placed subaurally with the reference electrode on the ipsilateral mastoid, and one hind leg served for grounding. The electrode impedance was kept below 5 k6. Recordings were done in a sound-attenuated and electromagnetically shielded chamber. The electrical impulse (click half width 125 Ws) was acoustically transformed by earphones (DT 48, Beyerdynamic, Heilbronn, Germany). Stimulation was applied at levels of 80, 60, 40, 30 and 20 dB (HL) (humans). The recorded signal (stimulation window 12.5 ms; about 500 averages; ¢lter 200^3000 Hz) was analyzed with respect to the threshold, the threshold shift and the latency of the ¢rst negative CAP peak N1 . Measurements were performed at the following time intervals: before and after opening the bulla as well as 4, 30, 60 and 90 min after laser irradiation and also on the 1st, 2nd, 7th, 14th and 28th postoperative days. Postoperative measurements were carried out up to the day on which the cochlea was removed and prepared for the SEM examinations. The care and use of animals in this study supported by the German Research Foundation (Jo 180/1-1) were approved by the Animal Care Committee of the Ministry of Health in Berlin (TVV 23/91) in accordance with the guidelines of the Declaration of Helsinki. 3. Results 3.1. Control animals The SEM examination of all eight control animals yielded normal ¢ndings for the stereocilia and cuticular plates of the inner and outer hair cells of the basal, second and third turns. In the apex, the stereocilia on
Fig. 1. SEM image of the basal turn of a cochlea 2 weeks after applying Er:YSGG laser irradiation (energy: 5 times 85 mJ; repetition rate: 1 Hz). The stereocilia and cuticular plates of the inner and outer hair cells and supporting cells show normal con¢guration. (A) Inner and outer hair cells. Magni¢cation: 1500-fold. (B) Outer hair cells. Magni¢cation: 3750-fold. (C) Inner hair cell. Magni¢cation: 10 875-fold.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
101
Fig. 2. CAPs of an animal before and after applying Er:YSGG laser irradiation (energy: 5 times 85 mJ; repetition rate: 1 Hz). Normal CAP thresholds and latencies throughout the entire examination period.
the second and third rows of the outer hair cells showed a partial contortion and a collapse as normal variants. The supporting cells had a normal con¢guration in all examined turns. The electrophysiological recordings were likewise normal throughout the entire examination period. 3.2. Er:YSGG laser 3.2.1. Perforation of the basal turn Group 1: The basal turn of eight animals was irradiated once with an energy of 85 mJ (energy density : 36 J/cm2 , t = 500 Ws (FWHM)). As in the human stapes, these laser parameters were su¤cient to create a small perforation (50^100 Wm) in the guinea pig cochlear bone, but not to achieve the required bone perforation diameter of 500^600 Wm. After laser application, the animals showed no SEM abnormalities. The partially irregular structure of the cilia on the second and third rows of outer hair cells with normal visualization of the inner hair cells in the apex is considered to be a normal variant. The electrophysiological recordings showed a normal time course for all animals examined during a period extending from directly after the laser application to 4 weeks later. 3.2.2. Perforation of the basal turn and irradiation of the open cochlea Group 2: The basal turn of seven animals was irradiated with an energy of ¢ve times 85 mJ (total energy : approx. 0.5 J, energy density : 36 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 1 Hz and 5 Hz. The ¢rst application already led to a small perforation,
the others at the same site to enlargement of the perforation and irradiation of the open cochlea. A ¢vefold application with these parameters is necessary for an adequate perforation (diameter 500^600 Wm) of the human stapes footplate. All animals showed normal SEM ¢ndings for the hair and supporting cells in the examined areas of the cochlea (Fig. 1A^C). The electrophysiological examination likewise showed no changes (Fig. 2). 3.2.3. High total energies Group 3: To determine the application safety of the laser system, the basal turn of four animals was irradiated using an energy of 85 mJ/pulse with 25 applications (total energy : approx. 2.1 J, energy density : 36 J/ cm2 , t = 500 Ws (FWHM)) and a repetition rate of 5 Hz. The SEM examination yielded normal ¢ndings for the inner and outer hair cells in all turns of the cochlea. None of the animals showed a threshold shift or latency prolongation of the CAP at the electrophysiological examination. In four animals, the basal turn was irradiated using an energy of 85 mJ/pulse with 50 applications (total energy : approx. 4.2 J) and a repetition rate of 5 Hz. In two animals, the outer hair cells of all turns examined showed fusion of the stereocilia tips with formation of giant cilia (Fig. 3A,B). Findings were normal for the inner hair cells in all turns examined. At the electrophysiological examination, these animals had threshold shifts of 40 dB starting directly after laser application and persisting until the 28th postoperative day. The other two animals had lost all inner and outer hair cells throughout the cochlea. The missing cells had been re-
HEARES 3438 12-5-00
102
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
placed by polygonal epithelial cells (Fig. 4). Electrophysiologically, no CAP could be recorded from 30 min after laser application until the 28th day (Fig. 5). 3.3. Ho:YAG laser 3.3.1. Perforation of the basal turn and irradiation of the open cochlea Group 1: The basal turn of seven animals was irradiated applying 10 laser pulses with an energy of 210 mJ per pulse (total energy : approx. 2.1 J, energy density : 90 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 1 Hz. A 10-fold irradiation of the same application site with the above parameters is necessary for an adequately large perforation of the human footplate. SEM revealed normal ¢ndings for the hair and support-
Fig. 3. SEM image of a cochlea after applying Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Two weeks after laser application, stereocilia tips of outer hair cells in the second and third rows of all turns of the cochlea show giant cilium formation with retraction of the cuticular plates. (A) Second turn. Magni¢cation: 1500-fold. (B) Third turn. Magni¢cation: 3750-fold.
Fig. 4. SEM image of a cochlea after applying Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Four weeks after application, there is a loss of all outer hair cells in the basal turn and replacement by polygonal epithelial cells. Magni¢cation: 750-fold.
ing cells in the entire cochlea in three animals. These animals showed a normal electrophysiological course starting directly after laser application and extending to the 28th postoperative day. The other four animals had fusion of the stereocilia tips with formation of giant cilia, retraction of the cuticular plates and loss of individual groups of outer hair cells. The stereocilia of the inner hair cells were partially splayed and disarrayed (Fig. 6). Electrophysiologically, the latter four animals evidenced a threshold shift of 40 dB starting directly after laser application and extending to the 28th postoperative day (Fig. 7). Group 2: The basal turn of six animals was irradiated applying 10 laser pulses with an energy of 210 mJ per pulse and a repetition rate of 5 Hz. The aim was to examine the in£uence of a higher repetition rate on cochlea morphology and function. SEM revealed changes of di¡erent extent in all animals examined at the various time points. On the one hand, pathological ¢ndings, which included fusion of the stereocilia tips on the outer and inner hair cells, retraction of the cuticular plate and loss of individual outer hair cells, were mainly detected in the basal and second turns (Fig. 8). On the other hand, the slight damage limited to the lasered area was accompanied by the loss of all outer hair cells in all turns, while the inner hair cells showed fusion of the stereocilia tips (Fig. 9). Electrophysiologically, all animals had a threshold shift of 30^40 dB starting directly after laser application. On the ¢rst and second postoperative days, two animals had a severe hearing loss with a threshold shift of v 60 dB. Twenty-eight days after laser treatment, an extreme CAP alteration persisted in four animals (Fig. 10).
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
103
Fig. 5. CAPs of an animal before and after application of Er:YSGG laser irradiation (energy: 50 times 85 mJ; repetition rate: 5 Hz). Recordings show strong CAP threshold shift starting 30 min after applying Er:YSGG laser irradiation.
3.3.2. High total energies Group 3: To determine the application safety of the laser system, an energy of 210 mJ was selected to irradiate the basal turn of four animals with 20 applications (total energy: approx. 4.2 J, energy density: 90 J/cm2 , t = 500 Ws (FWHM)) and a repetition rate of 5 Hz. After 20 laser applications, the basal and second turns of the cochleas showed a loss of the entire organ of Corti and its replacement by polygonal epithelial cells (Fig. 11). There were irregularly structured hair cells and a loss of individual cells in the third turn. Electrophysiologically, CAP alterations occurred directly after the laser application and led to a hearing loss by the 28th postoperative day (Fig. 12).
iants. About 20% of the cochlea at the apex should therefore be disregarded at the SEM evaluation. CAPs were determined as a measure of inner ear function in guinea pigs. These were readily reproducible with respect to the threshold and latency. The non-invasive technique of needle electrode recording applied here had several advantages over the invasive method of permanent electrode implantation at the round window used for recording cochlear microphonics (CM): it considerably shortened the preparation time and markedly reduced the risk of infection. This enabled reliable measurements over a longer period of time (here
4. Discussion SEM examinations were performed to detect morphological changes in the cochlea after laser application. This method is particularly e¡ective for investigating changes in Corti's organ with the sensory and supporting cells. SEM provides good visualization of the apical surfaces of the outer and inner hair cells with their stereocilia. The apex of the cochlea was not included in the evaluation, since both the controls and laser-treated animals evidenced repeated changes in the stereocilia of the outer and inner hair cells in this part of the cochlea. Findings included torsions, a collapse and irregular structure of the stereocilia. Soudijn (1976) and Rydmarker et al. (1987) also observed such changes in the apex when evaluating Corti's organ after exposure of the guinea pig cochlea to impulse noise. These changes were interpreted as normal var-
Fig. 6. SEM image of the second turn of the cochlea 90 min after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 1 Hz). Outer hair cells show fusion of stereocilia with partially giant cilium formation and loss of outer hair cells in the third row. The stereocilia of the inner hair cells are partially splayed and disarrayed. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
104
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Fig. 7. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 1 Hz). The recordings show moderate hearing impairment that started immediately after applying Ho:YAG laser irradiation and persisted up to 4 weeks thereafter.
4 weeks). The measurements were performed at close intervals after laser application in order to record the immediate e¡ects of laser irradiation on the cochlea and to classify them according to their severity and reversibility. The broad-band acoustic stimulation necessary to evoke the CAP, however, does not permit frequencyspeci¢c di¡erentiation of possible hair cell damage, Fast N1 peaking of CAP mostly re£ects high-frequency components in the basal part of the guinea pig cochlea which can be stimulated by frequencies up to 40 kHz. Cochlear responses from middle and apical areas have
Fig. 8. SEM image of the second turn of the cochlea 2 weeks after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). Fusion of stereocilia tips of the second and third rows of the outer hair cells with retraction of the cuticular plate and loss of individual outer hair cells. The disarrayed stereocilia of the inner hair cells are an artefact. Magni¢cation: 1500-fold.
only a slight in£uence on CAP. More precise information on possible inner ear impairments in adjacent areas can be gained by performing histological and SEM examinations of laser-treated cochleas. On the other hand, the data obtained by recording frequency-speci¢c CM is not accurate enough. Laser application in the basal turn of the guinea pig cochlea was the model selected for laser stapedotomy, as already described by Thoma et al. (1986). It should be noted, however, that anatomic and pathophysiological conditions in laser stapedotomy di¡er from those in this animal experiment. When perforating the human stapes footplate, the most exposed parts of the inner
Fig. 9. SEM image of the basal turn of the cochlea 4 weeks after applying Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). Loss of all outer hair cells and fusion of stereocilia tips of the inner hair cells. The ruptures of the cuticular plate are shrinking artefacts due to ¢xation. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
105
Fig. 10. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 10 times 210 mJ; repetition rate: 5 Hz). The recordings show a threshold shift of 40 dB starting immediately after laser application, reaching a maximum of s 60 dB on the ¢rst post-treatment day and still persisting as a 60 dB threshold shift on the 28th post-treatment day.
ear are the saccule, the utricle and the cochlear duct, lasering the basal turn of the guinea pig cochlea, however, involves a higher risk to structures of Corti's organ like the basilar membrane, the stria vascularis, the outer and inner hair cells, etc. Since this probably results in higher vulnerability of these anatomic structures in the animal experiment, inner ear damage is more likely to result from laser application in our animal model than would be the case when lasering the human stapes footplate and the vestibulum. If no alterations are found in animal experiments, it is probable that the same laser parameters will cause no injury to human inner ear structure. The Er:YSGG laser (pulse half width: t = 500 Ws (FWHM), spot size approx. 550 Wm) applied once with an energy of 85 mJ did not morphologically change Corti's organ in any cochlear areas examined by SEM. CAP recordings were unchanged (i.e., thresholds and latencies) at all recovery times. A ¢vefold application of Er:YSGG laser irradiation with these parameters (total energy : approx. 0.4 J) is necessary to achieve an adequate perforation of the human stapes footplate (Jovanovic et al., 1997). In this connection, Corti's organ of the cochleas examined in all animals did not show detectably altered sensory and supporting cells in any of the turns. The CAP recordings likewise did not change. The in£uence of higher pulse repetition rates (5 Hz) on inner ear morphology and function was also examined. Here too, Corti's organ of the cochleas examined did not evidence morphological changes in the sensory or supporting cells of any of the turns. The CAP recordings were likewise normal. These results in animals demonstrate the safety of applying Er: YSGG laser irradiation to the basal cochlea with an
energy density of 36 J/cm2 and a total energy of approx. 0.4 J, which are parameters e¡ective for human footplate perforation. To determine the range of safe Er:YSGG laser application, the basal turn of the guinea pig cochlea was irradiated with a much higher total energy than that required for adequate perforation of the human footplate. A 25-fold laser application (¢ve-fold higher total energy : approx. 2.1 J) did not cause pathological changes in any of the cochleas examined by SEM. There were also no CAP changes. It was only after a 50-fold laser application (10 times higher total energy: approx. 4.2 J) that Corti's organ showed pathologically altered sensory and supporting cells in the cochleas ex-
Fig. 11. SEM image of the second turn of the cochlea 4 weeks after applying Ho:YAG laser irradiation (energy: 20 times 210 mJ; repetition rate: 5 Hz). The loss of all hair and supporting cells and their replacement by polygonal epithelial cells. Magni¢cation: 1500-fold.
HEARES 3438 12-5-00
106
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Fig. 12. CAPs of an animal before and after application of Ho:YAG laser irradiation (energy: 20 times 210 mJ; repetition rate: 5 Hz). A threshold shift s 60 dB occurred directly after laser irradiation. No CAP could be recorded 1 day after laser treatment.
amined. There were more marked changes of outer than of inner hair cells, particularly in the second and third rows. These alterations included an irregular structure of the stereocilia and their fusion with formation of giant cilia that exceeded the normal stereocilia in length and thickness. Some of the pathological changes were so marked that they resulted in a total loss of the sensory cells with replacement by polygonal epithelial cells 4 weeks after laser treatment. The cuticular plates of the sensory cells were deformed in the majority of cells. These changes were equally detectable in all cochleas examined at the various post-treatment times. The electrophysiological examination showed moderate to severe CAP alterations, some of which were partially irreversible. In contrast to the inner hair cells, the stereocilia of the outer hair cells are in direct contact with the tectorial membrane. The irregular structure and partial loss of stereocilia can be explained by extreme shearing movements of the stereocilia by the tectorial membrane. This extreme movement can result from impulse noise exposure and may lead to destruction of the tectorial membrane at the site of the inserted stereocilia and of the stereocilia tips themselves (Hamernik et al., 1984 ; Saunders et al., 1985). There may also be a rupture of the reticular lamina with opening of the endolymphatic space followed by mixture of high-potassium endolymph with low-potassium perilymph. This leads to further degeneration and even death of the hair cells (Bohne and Rabbitt, 1983; Duvall and Rhodes, 1967 ; Konishi and Salt, 1979). The additional loss of stereocilia we observed 4 weeks after laser application may be due to this degeneration. The dead hair cells are ultimately discarded into the endolymphatic space. The
defects in the reticular lamina are closed by expansion of the supporting cells (Gao et al., 1992 ; Spoendlin, 1971). The Er:YSGG laser enables nearly athermal bone ablation and only minimally increases the perilymph temperature by about 3.6³C at a distance of 2 mm behind the laser application site (Jovanovic et al., 1997, 1998). On the other hand, footplate perforation involves explosion-like pressure impulses generating pressure and shock waves in the cochlear perilymph which are comparable to peak sound pressure levels of about 153 dB(SPL) (Scho«nfeld et al., 1994). The pathological ¢ndings for hair and supporting cells in our animal experiment closely resemble the morphological changes in these structures after impulse noise exposure and are thus probably due to the pressure impulses rather than to the thermal e¡ect of the laser irradiation (Jovanovic et al., 1998). Ho :YAG laser irradiation (pulse half width: t = 500 Ws (FWHM), spot size approx. 550 Wm), on the other hand, exhibits high damaging potential in guinea pigs. E¡ective laser parameters (total energy : approx. 2.1 J) already altered the inner and outer hair cells in more than 50% of the animals. The changes included torsion and collapse of the stereocilia and fusion of the stereocilia tips with giant cilia formation. The electrophysiological examination revealed moderate to severe CAP alterations. A doubling of the total energy to approx. 4.2 J caused morphological changes extending to a loss of all sensory cells and their replacement by polygonal epithelial cells. Electrophysiologically, the animals thus treated showed severe irreversible CAP alterations. In comparison to Er:YSGG laser irradiation, the morphological inner ear changes after the Ho :YAG
HEARES 3438 12-5-00
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
laser application are more pronounced. One reason for this high damaging potential may be the poorer ablation capacity of Ho:YAG laser irradiation in bone. This necessitates high energy densities and total energies for an adequately large footplate perforation diameter of 500^600 Wm. Thus, compared to the Er:YSGG laser, the energy density is increased nearly threefold and the total energy ¢vefold. This raises the perilymph temperature about 30³C above the initial value of 37³C and to higher peak sound pressure levels of about 164 dB(SPL). Both in vitro values are critical for the inner ear structures. Another reason may be that the higher optical penetration depth of Ho:YAG laser irradiation in the perilymph could also cause immediate damage to inner ear structures by direct exposure. The optical penetration depth of the Ho:YAG wavelength (V = 2.1 Wm) in water is about 200 Wm and is therefore 200 times higher than that of the Er:YAG wavelength (V = 2.94 Wm). In electrophysiologically controlled animal experimental studies (measuring CM potentials in guinea pigs), Pfalz et al. (1994) and Pfalz (1995) investigated the e¡ects of the Er:YAG laser irradiation. Comparable to our Er:YSGG laser results they also postulated a good tolerance and high safety of the Er:YAG laser for application to the meatal lining near the limbus or to the manubrium mallei. Apart from a single greater temporary threshold shift with 50 applications, pulse counts of below 500 did not involve any irregularities in the CM potentials. Changes in the CM a¡ecting all frequencies with a maximum of 2 kHz were only seen after 500 applications of Er:YAG laser irradiation with an energy of 50 mJ (total energy : 25 J), a pulse duration of 250 Ws (FWHM), a pulse sequence of 2 Hz and a beam diameter of 200 Wm (energy density 160 J/cm2 ). A complete recovery of the CM was observed 90 min after laser application. When applying laser irradiation to the middle ear and auditory ossicles, the author considers even pulse counts of 10 000 or more permissible, which indicates the high suitability of the Er:YAG laser for middle ear surgery. These results are not applicable to the footplate, however, because of other damaging mechanisms involved in direct irradiation of the inner ear. Our results demonstrate that, with the tested parameters, the pulsed Er:YSGG laser shows good tolerance and high application safety in these animal experiments. Its high ablation rate and low damaging potential could render it a useful alternative to the thermal CO2 laser for stapedotomy (Jovanovic et al., 1996, 1999). The applied Ho:YAG laser, on the other hand, has a high damaging potential in our animal experiments. Its application in stapedotomy appears to be unpredictable and dangerous for the inner ear. Thus the conclusions drawn in our in vitro experi-
107
ments (Jovanovic et al., 1997, 1998) regarding the damaging mechanism and the risk of laser damage to the inner ear have been con¢rmed in vivo in relation to the laser parameters applied. However, these results are only valid for the lasers examined in this study, since their applicability to other pulsed laser systems, particularly those with shorter pulse half widths, is still questionable due to possibly higher pressure impulses. Acknowledgements The study was supported by the German Research Foundation (Jo 180/1-1). The authors wish to thank Dr. V. Stiglic, a veterinarian at the Berlin Central Animal Laboratories, for assistance in performing the animal experiments.
References Bohne, B.A., Rabbitt, K.B., 1983. Holes in the reticular lamina after noise exposure. Implications for continuing damage in the organ of Corti. Hear. Res. 11, 41^53. Duvall, A.J., Rhodes, V.T., 1967. Ultrastructure of the organ of Corti following intermixing of cochlear £uids. Ann. Otol. Rhinol. Laryngol. 76, 688^708. Gao, W.Y., Ding, D., Zheng, X., Ruan, F., Liu, Y., 1992. A comparison of changes in the stereocilia between temporary and permanent hearing losses in acoustic trauma. Hear. Res. 62, 27^41. Hamernik, R.P., Turrentine, G., Roberto, M., Salvi, R., Henderson, D., 1984. Anatomical correlates of impulse noise- induced mechanical damage in the cochlea. Hear. Res. 13, 229^247. Jovanovic, S., Scholz, C., Berghaus, A., Scho«nfeld, U., 1990. Experimenteller Vergleich zwischen kurzgepulsten und kontinuierlich strahlenden Lasern in der Stapeschirurgie ^ histologische-morphologische Ergebnisse. Arch. Otorhinolaryngol. II (Suppl.), 73^75. Jovanovic, S., Anft, D., Scho«nfeld, U., Berghaus, A., Scherer, H., 1995. Experimental studies on the suitability of the erbium laser for stapedotomy in an animal model. Eur. Arch. Otorhinolaryngol. 252, 422^427. Jovanovic, S., Scho«nfeld, U., Prapavat, V., Berghaus, A., Fischer, R., Scherer, H., Mu«ller, G., 1996. E¡ects of continuous-wave laser systems on stapes footplate. Lasers Surg. Med. 19, 424^432. Jovanovic, S., Scho«nfeld, U., Prapavat, V., Berghaus, A., Fischer, R., Scherer, H., Mu«ller, G., 1997. E¡ects of pulsed laser systems on stapes footplate. Lasers Surg. Med. 21, 341^350. Jovanovic, S., Scho«nfeld, U., Fischer, R., Do«ring, M., Prapavat, V., Mu«ller, G., Scherer, H., 1998. Thermal e¡ects in the `vestibule' during laser stapedotomy with pulsed laser systems. Lasers Surg. Med. 23, 7^17. Jovanovic, S., Anft, D., Scho«nfeld, U., Berghaus, A., Scherer, H., 1999. In£uence of CO2 laser irradiation of the guinea-pig cochlea on compound action potentials. Am. J. Otol. 20, 1^8. Kautzky, M., Tro«dhan, A., Susani, M., Schenk, P., 1991. Infrared laserstapedotomy. Eur. Arch. Otorhinolaryngol. 248, 449^451. Konishi, T., Salt, A.N., 1979. E¡ects of exposure to noise on ion movement in guinea pig cochlea. Hear. Res. 1, 325^342. McDowell, E.M., Trump, B.F., 1976. Histologic ¢xatives suitable for diagnostic light and electron microscopy. Arch. Path. Lab. Med. 100, 405^414.
HEARES 3438 12-5-00
108
S. Jovanovic et al. / Hearing Research 144 (2000) 97^108
Pfalz, R., 1995. Eignung verschiedener Laser fu«r Eingri¡e vom Trommelfell bis zur FuMplatte (Er:YAG-, Argon-, CO2 s.p.-, HoYAGLaser). Laryngol. Rhinol. Otol. 74, 21^25. Pfalz, R., Nagel, D., Bald, N., Hibst, R., 1994. Vergleich von BERASchwellen und cochlea«ren Mikrophonpotentialen (CM) vor und nach Er:YAG-Laserung im Mittelohr narkotisierter Meerschweinchen. Eur. Arch. Otorhinolaryngol. II (Suppl.), 36. Pfander, F., 1975. Das Knalltrauma. Springer, Berlin. Pratisto, H., Frenz, M., Ith, M., Romano, V., Felix, D., Grossenbacher, R., Altermatt, H., Weber, H., 1996. Temperature and pressure e¡ects during erbium laser stapedotomy. Lasers Surg. Med. 18, 100^108. Romano, V., Rodriguez, R., Altermatt, H., Frenz, M., Weber, H., 1993. Bone microsurgery with IR-lasers: A comparative study of the thermal action at di¡erent wavelengths. Proc. SPIE 2077, 87^ 97. Rydmarker, S., Nilsson, P., Grenner, J., 1987. Morphological and functional changes in the organ of Corti after noise exposure. Acta Otolaryngol. 441, 3^43.
Saunders, J.C., Dear, S.P., Schneider, M.E., 1985. The anatomical consequences of acoustic injury: A review and tutorial. J. Acoust. Soc. Am. 78, 833^860. Scho«nfeld, U., Fischer, R., Jovanovic, S., Scherer, H., 1994. `La«rmbelastung' wa«hrend der Laserstapdotomie. Eur. Arch. Otorhinolaryngol. II (Suppl.), 244^246. Soudijn, E.R, 1976. Scanning electron microscopic study of the organ of Corti in normal and sound-damaged guinea pigs. Ann. Otol. Rhinol. Laryngol. 29 (Suppl.), 1^46. Spoendlin, H.H., 1971. Primary ultrastructural changes in the organ of Corti after acoustic overstimulation. Acta Otolaryngol. 71, 166^ 176. Stubig, I.M., Reder, P.A., Facer, G.W., Rylander, H.G., Welch, A.J., 1993. Holmium:YAG laser stapedotomy: preliminary evaluation. Proc. SPIE 1876, 10^19. Thoma, J., Mrowinski, D., Kastenbauer, E., 1986. Experimental investigations on the suitability of the carbon dioxide laser for stapedotomy. Ann. Otol. Rhinol. Laryngol. 95, 126^131.
HEARES 3438 12-5-00