- Email: [email protected]
Alterations in the stria vascularis in relation to cisplatin ototoxicity and recovery
Available online at www.sciencedirect.com R
Hearing Research 185 (2003) 49^56 www.elsevier.com/locate/heares
Alterations in the stria vascularis in relation to cisplatin ototoxicity and recovery Steven Sluyter, Sjaak F.L. Klis, John C.M.J. de Groot, Guido F. Smoorenburg
Hearing Research Laboratories, Department of Otorhinolaryngology, University Medical Center Utrecht, Room G.02.531, P.O. Box 85.500, 3508 GA Utrecht, The Netherlands Received 11 March 2003; accepted 11 August 2003
Abstract We have investigated whether or not cisplatin-induced depression of the endocochlear potential (EP), and its subsequent recovery, possesses a morphological correlate in the stria vascularis. Guinea pigs implanted with round window electrodes were treated daily with cisplatin (1.5 mg/kg/day) until the compound action potential showed a profound hearing loss ( v 40 dB at 8 kHz after 5^18 days). Animals were either sacrificed immediately after the shift in hearing threshold (‘SHORT’ group) or allowed to recover for v 4 weeks and subsequently sacrificed (‘LONG’ group). Control animals (‘CONTROL’ group) were not treated with cisplatin. Using stereological methods we measured the total strial cross-sectional area together with the areas occupied by the different strial components: the marginal, intermediate and basal cells. The total strial cross-sectional area in the basal turn of the LONG group was found to be significantly smaller than that of the SHORT and the CONTROL groups, whereas the EP was normal in the LONG group (in comparison to the CONTROL group) and markedly decreased in the SHORT group. The smaller area in the LONG group was mainly due to a decrease in the area occupied by the intermediate cells and to a lesser extent to a decrease in the marginal cell area. The area occupied by the basal cells did not change. Thus, the marked decrease in EP after 5^18 days of cisplatin administration was not related to shrinkage of the stria vascularis. Moreover, 4 weeks later the EP showed full recovery, whereas the stria vascularis had shrunk markedly. @ 2003 Elsevier B.V. All rights reserved. Key words: Cisplatin; Ototoxicity; Stria vascularis; Endocochlear potential; Guinea pig
1. Introduction Cisplatin [cis-diamminedichloroplatinum(II)] is an important anticancer drug, which is used in the treatment of various types of malignant tumors. Ototoxicity, however, is a serious side e¡ect of cisplatin. Extensive >experimental research has shown that cisplatin exerts a toxic e¡ect on the outer hair cells (OHCs) (Schweitzer, 1993; De Groot et al., 1997; Cardinaal et al., 2000) and
* Corresponding author. Tel.: +31 (30) 2507581; Fax: +31 (30) 2541922. E-mail address: [email protected] (G.F. Smoorenburg). Abbreviations: A1, lower apical turn; A2, second apical turn; A3, upper apical turn; ANOVA, analysis of variance; B1, lower basal turn; B2, upper basal turn; CAP, compound action potential; EP, endocochlear potential; M1, lower middle turn; M2, upper middle turn; OHC, outer hair cell
the stria vascularis (Tange and Vuzevski, 1984; Kohn et al., 1991; Meech et al., 1998; Campbell et al., 1999; Cardinaal et al., 2000), resulting in permanent sensorineural hearing loss. However, recovery from cisplatininduced hearing loss has been reported occasionally in both humans (Aguilar-Markulis et al., 1981; Laurell and Jungnelius, 1990) and guinea pigs (Nakai et al., 1982; Stengs et al., 1997). In albino guinea pigs, Klis et al. (2000, 2002) studied recovery from cisplatin-induced hearing loss, measuring the compound action potential (CAP) via implanted round window electrodes. The animals were treated daily with i.p. injections of cisplatin (1.5^2.0 mg/kg/day) until a profound hearing loss occurred (criterion : v 40 dB shift in CAP threshold at 8 kHz, CAP threshold de¢ned as the 3 WV iso-response level). CAP threshold gradually recovered in the ¢rst 2 weeks after cessation of treatment. In the studies of Klis et al. (2000, 2002) a
0378-5955 / 03 / $ ^ see front matter @ 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0378-5955(03)00260-0
HEARES 4766 27-10-03
50
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
similar recovery was observed in the endocochlear potential (EP) that was found to be markedly depressed immediately after the profound CAP threshold shift occurred. In addition, histologically, OHC loss was found with the losses predominantly in the basal turns. OHC counts did not di¡er between a group of animals processed before they had a chance to recover (within 3 days after termination of treatment) and a group of animals allowed to recover for up to 100 days. Together, these ¢ndings suggested that the loss and subsequent recovery of the CAP is related to loss and recovery of the EP. Since the EP is generated through strial activity (Wangemann and Schacht, 1996), it can be surmised that loss and recovery of the EP would have a morphological correlate in the stria vascularis. Thus, as a logical follow-up on our previous studies (Klis et al., 2000, 2002), we have performed a quantitative analysis of several morphological parameters of the stria vascularis in relation to cisplatin ototoxicity and its recovery in albino guinea pigs. The approach of this study was to measure the total cross-sectional area of the stria vascularis and to determine the area occupied by the di¡erent strial components (marginal, intermediate and basal cells as well as the strial capillaries and the intercellular space). Furthermore, we have compared these morphological data with EP data in each individual animal.
2. Materials and methods 2.1. Animals and experimental design Healthy, female albino guinea pigs (strain Dunkin Hartley), weighing 250^350 g, were used. They were housed under standard laboratory conditions and fed ad libitum. Cisplatin (Platosin0 ; Pharmachemie, Haarlem, The Netherlands) was diluted with physiological saline (pH 7.4) to a ¢nal concentration of 0.1 mg/ml. The high dilution was chosen in order to stimulate diuresis and, thus, to minimize renal e¡ects. All animals were treated with daily i.p. injections of cisplatin at a dose of 1.5 mg/kg until a shift of 40 dB or more in the CAP threshold at 8 kHz occurred. The 8 kHz CAP threshold was chosen as a global cochlear sensitivity indicator because of the high reproducibility at this frequency, but losses occurred at higher and lower frequencies as well (Klis et al., 2002). When criterion loss was reached, cisplatin administration was terminated and the animals were processed for histological examination of the cochlea, either within 1^3 days after termination of cisplatin administration (SHORT group; n = 5) or after 4 weeks or more (LONG group; n = 14). These time intervals were based upon the earlier work of Klis et al. (2000, 2002). In both groups the EP was
measured before harvesting the cochleas. The two groups were complemented by a CONTROL group (n = 7) which was not treated with cisplatin but subjected to the same measurements. The Animal Care and Use Committee of the Medical Faculty of the University of Utrecht approved the care and use of animals in this study under numbers 91035, 89007, and 89055 for electrode implantation, electrocochleographic techniques and histological techniques, respectively. 2.2. Electrophysiology Details about the surgical techniques and electrophysiologic measurements have been described previously and will not be outlined here in detail (Van Emst et al., 1997; Klis et al., 2000, 2002). The method for chronic round window recording was adopted from Aran and Erre (1979). Under sterile conditions the bulla of the right ear was opened retro-aurically and an electrode was placed on the round window. Electrophysiological measurements were performed di¡erentially with the round window electrode as the active electrode and two screws on the skull as reference and ground electrodes respectively. CAP threshold (iso-response level at 3 WV) was measured daily, after which injections with cisplatin were given. In all animals, the EP in the second turn from the base of the right cochlea was measured with a standard glass pipette microelectrode ¢lled with 150 mM KCl. The EP was measured at this location for technical reasons, but the EP at this location can be taken to be representative for the EP in the entire cochlea because of the good conducting properties of the endolymphatic compartment. 2.3. Histology Immediately following the EP measurement, the cochleas of both ears were ¢xed by intralabyrinthine perfusion with a tri-aldehyde ¢xative (3% glutaraldehyde, 2% formaldehyde, 1% acrolein, and 2.5% dimethyl sulfoxide in 0.08 M sodium cacodylate bu¡er, pH 7.4; De Groot et al., 1987) followed by immersion in the same ¢xative for 3 h at room temperature. Following several rinses (2U15 min) in 0.1 M sodium cacodylate bu¡er (pH 7.4), the cochleas were decalci¢ed in 10% EDTAW2Na for 5 days at room temperature. Next, the specimens were post¢xed in 1% OsO4 containing 1% K4 Ru(CN)6 in 0.1 M sodium cacodylate bu¡er (pH 7.4) for 2 h at 4‡C (De Groot et al., 1987). Dehydration was performed in a graded ethanol series and the cochleas were embedded in toto in Spurr’s low-viscosity resin. The cochleas were divided into two cochlear halves along a midmodiolar plane. The orientation of the midmodiolar plane was the same in all cochlear
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
51
preparations, using the attachment point of the cochlea onto the medial wall of the bulla and both the oval and round window as anatomical landmarks. After division along this midmodiolar plane, the cochleas were re-embedded in the same resin. For light microscopic assessment, semi-thin (1 Wm) sections were cut with a Histo diamond knife on a Reichert-Jung 2050 microtome, collected on glass slides and stained with 1% methylene blue and 1% azur II in 1% sodium tetraborate.
line lattice with a 10-mm distance between the lines. The number of points overlying each strial component was separately counted as well as the number of points overlying the entire stria vascularis and converted to fractions of cross-sectional area (in Wm2 ). In addition, the same method was used to assess the area that is taken by the interdigitations formed by marginal and intermediate cells, relative to the total strial cross-sectional area.
2.4. Determination of the total strial cross-sectional area
2.6. Statistical analysis
Midmodiolar sections of the right cochleas (in which the EP had been measured) from both experimental groups and the control group were examined using a Zeiss Axiophot light microscope. Images of the individual transections of the respective half turns (two transections each for the basal [B1, B2], middle [M1, M2] and apical [A1, A2] turns) were captured with a JVC TK-1281 color videocamera and digitally stored (TIFF format) for o¡-line processing. We used two adjacent midmodiolar sections and measured all strial transections twice on two separate occasions. Thus, the actual ¢gure for each strial cross-sectional area is based on four measurements. Digitized images were imported into the public domain NIH Image program (developed at the US National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image). Next, the boundaries of the stria vascularis were outlined with a pressure-sensitive stylus on a Wacom UD0608-R digitizer interfaced to a Macintosh computer. Finally, the strial cross-sectional area (in Wm2 ) demarcated by these boundaries was calculated.
The quantitative data were statistically analyzed using analysis of variance (ANOVA; Statistica Software Package). Post-hoc comparisons were made with Tukey’s HSD (honestly signi¢cant di¡erences) test for unequal n (Spjotvoll/Stoline). The criterion for statistical signi¢cance was taken at P = 0.05.
2.5. Determination of the areas of the strial components Ultrathin (90 nm) sections of the re-embedded lower basal turn (B1; located approximately 2.4 mm from the round window) from the left cochleas were cut with a Diatome diamond knife on an LKB Ultrotome Nova and collected on Pioloform-coated, single-slot copper grids. Sections were contrast-stained with 7% uranyl acetate in 70% methanol and Reynolds’ lead citrate, and examined in a JEOL 1200EX transmission electron microscope operating at 80 kV. The entire stria vascularis was photographed in a series of electron micrographs that were printed at a ¢nal magni¢cation of 5500 times. The electron micrographs were assembled into a complete photomontage of the stria vascularis. To determine the relative areas of the di¡erent strial components (i.e., marginal cells, intermediate cells, basal cells, capillaries, and intercellular space) a stereological point-counting method was used (Santi and Lakhani, 1983; Lohuis et al., 1990). Each photomontage was overlaid with a transparent sheet bearing a square
3. Results 3.1. EP recordings The data with regard to loss and recovery of cochlear sensitivity (CAP thresholds) in the LONG and SHORT groups have been published earlier in Klis et al. (2002). CAP thresholds remained stable during the ¢rst days of treatment with cisplatin. Then, after a period which varied considerably between animals (5^18 days from treatment onset), thresholds suddenly increased within 1 or 2 days after which the criterion threshold shift was reached. Subsequent to termination of cisplatin treatment thresholds recovered spontaneously over a period of about 2 weeks, after which no further recovery of thresholds was observed. At the lower frequencies recovery was almost complete, whereas at the higher frequencies recovery was usually poor. The EP values of the LONG, SHORT and CONTROL groups, measured in the second turn, are presented in Fig. 1 (averaged data) and Fig. 3 (individual data). The EP in the
Fig. 1. Mean values of the second turn EP in mV U S.E.M. in the CONTROL, SHORT, and LONG groups.
HEARES 4766 27-10-03
52
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
Fig. 2. Mean strial cross-sectional area in Wm2 ( U S.E.M.) based on measurements in semi-thin midmodiolar sections of the right cochleas from the CONTROL, SHORT, and LONG groups. The cochlear turns are related to frequencies (Greenwood, 1996). B1 relates to 26.2 kHz; B2 to 10.4 kHz; M1 to 5.1 kHz; M2 to 2.7 kHz; A1 to 1.3 kHz; and A2 to 0.7 kHz. *P 6 0.05; **P 6 0.01.
Fig. 3. Strial cross-sectional area in Wm2 based on a point-counting method performed in ultrathin sections of the lower basal (B1) turn in relation to the EP and comparing the CONTROL, SHORT, and LONG groups.
Fig. 4. Light micrographs of the stria vascularis in the lower basal turn (B1) of the CONTROL (A) and LONG (B) groups. Note that the stria vascularis in the LONG group is thinner than that in the CONTROL.
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
LONG group did not di¡er from that in the CONTROL group (82 U 1 mV and 83 U 2 mV, respectively). The EP in the SHORT group was signi¢cantly lower (55 U 4 mV; P 6 0.001). The present EP values in the CONTROL and LONG groups were in the normal range for albino guinea pigs (Gill and Salt, 1997). 3.2. Total strial cross-sectional area Fig. 2 shows the total strial cross-sectional area for all individual transections present in midmodiolar sections of the cochlea. Unexpectedly, the strial cross-sectional area in the LONG group, the group that showed normal EPs, appeared to be smaller than those in both the SHORT and CONTROL groups, especially in the lower basal turn (B1 ; Fig. 4). ANOVA over the entire data set con¢rmed signi¢cant main e¡ects of group (SHORT, LONG, CONTROL) and of cochlear turn. Tukey’s post-hoc test showed that the LONG group was primarily responsible for the main e¡ect of group in the lower basal turn (B1 ; P 6 0.01 with respect to both the SHORT and CONTROL groups). In all other half-turns the LONG group had, on average, the small-
53
est total cross-sectional area, but the statistical criterion of 5% was reached only in the lower middle turn (M1). 3.3. Relation between EP and strial cross-sectional area Fig. 3 shows the relation between the EP in individual right ears and the total strial cross-sectional area in the lower basal turn (B1). This graph clearly shows three distinct clusters of data points. The SHORT group is comparable to the CONTROL group with respect to strial cross-sectional area, but the EPs are markedly lower in the SHORT group. In contrast, the LONG group is comparable to the CONTROL group with respect to the EP, but, on average, the LONG group has a much smaller strial cross-sectional area. 3.4. Determination of the area of the strial components Since the largest decrease in strial cross-sectional area was found in the lower basal turn (B1), we examined this turn in more detail using transmission electron microscopy (Figs. 5 and 6). The areas occupied by the di¡erent cell types (marginal, intermediate and basal
Fig. 5. Transmission electron micrograph of the stria vascularis in the lower basal (B1) turn from a CONTROL animal.
HEARES 4766 27-10-03
54
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
Fig. 6. Transmission electron micrograph of the stria vascularis in the lower basal (B1) turn of an animal from the LONG group.
cells) and by the capillaries and intercellular space were determined for the three groups. Mean values ( U S.E.M.) are presented in Table 1 and Fig. 7. ANOVA showed that the total area occupied by the intermediate cells is signi¢cantly smaller in the LONG group, as compared to both the SHORT (P 6 0.01) and CONTROL groups (P 6 0.05). The area occupied by the marginal cells was signi¢cantly smaller in the LONG group when tested against the CONTROL group (P 6 0.05), but not when tested against the SHORT group (P = 0.30). There was no di¡erence between the three groups with respect to the areas occupied by the basal cells, the capillaries and the intercellular space. Table 2 shows the relative areas occupied by the interdigitations between the marginal and intermediate cells. There were no signi¢cant di¡erences between the two groups.
4. Discussion We hypothesized that the decrease in EP immediately after cisplatin treatment (Klis et al., 2000, 2002) could have a morphological correlate in the stria vascularis. However, there were no obvious morphological changes
in the stria vascularis shortly after termination of cisplatin treatment (SHORT group). In contrast, animals that appeared to recover several weeks after termination of treatment (judging from the electrophysiological recordings: CAP and EP) demonstrated a decreased strial cross-sectional area, especially in the basal turn (Figs. 1 and 2). Further ultrastructural analysis of this decrease in the basal turn showed that it was mainly due to a decrease in the area occupied by the intermediate cells and to a lesser extent to a decrease in marginal cell area. Finally, inspection of the stria vascularis showed no bulging of luminal membranes and no differences in vacuolization between the three di¡erent groups. Our results collected immediately after termination of cisplatin treatment (SHORT group), showing no morphological changes in the stria vascularis, di¡er markedly from those of Meech et al. (1998). They performed a semi-quantitative analysis of the e¡ect of cisplatin on the basal turn of the stria vascularis in the rat shortly after treatment and reported that the strial cross-sectional area in the cisplatin-treated group was signi¢cantly larger than in the controls. Also, they observed speci¢c alterations in the marginal cells, consisting of bulging, rupture and/or compression of the cell body
Table 1 Cross-sectional strial areas in Wm2 ( U S.E.M.) of the di¡erent strial components, as measured in ultrathin sections of the lower basal turn (B1) Group
Marginal cells
Intermediate cells
Basal cells
Strial capillaries
Intercellular space
CONTROL SHORT LONG
2336 U 120 2196 U 83 1941 U 69
2140 U 80 2294 U 98 1574 U 114
600 U 82 525 U 57 661 U 59
633 U 96 616 U 121 358 U 37
18 U 8 21 U 6 67 U 26
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
and depletion of cytoplasmic organelles. The di¡erences between both studies may be explained by the species di¡erences (rats vs. guinea pigs) but, more importantly, probably by the di¡erent treatment protocols. Meech et al. (1998) used a single high dose, well above the lethal level as evidenced by the loss of 50% of the animals in their experimental group. We, however, used repeated low-dose cisplatin injections, which are not lethal but still result in auditory e¡ects and allow the animals to recover from these e¡ects. Thus, on the basis of experimental outcome, i.e., non-lethality and small-to-moderate, partly reversible hearing loss, our protocol closely resembles the commonly used human protocols. With respect to di¡erences in treatment protocol, possibly resulting in di¡erences in the strial e¡ect of the drug in guinea pigs, the argument can be extended to earlier work of Tange and Vuzevski (1984) who found no strial changes in guinea pigs after 5^10 days of cisplatin administration (1.5 mg/kg/day) but did observe speci¢c alterations after prolonged treatment with higher cumulative doses. Also, Kohn et al. (1991) reported morphological changes in the stria vascularis in both albino and pigmented guinea pigs shortly after cisplatin treatment, but only at doses higher than those used in our study (2.5 mg/kg/day ; 7 days). We could attribute the decrease in total strial crosssectional area in the LONG group to the reduction in the areas occupied by the intermediate cells and, to a lesser extent, by the marginal cells. The question has to be addressed what might have caused this decrease in the LONG group and why this morphological alteration is not re£ected in the EP measurements. Interestingly, Forge et al. (1987) reported a signi¢cant decrease in strial width 4 weeks after cessation of gentamicin treatment, but not shortly after. In contrast to the present ¢ndings this decrease could be attributed to a decreased volume density of the marginal cells. A decrease in total strial cross-sectional area has also been described in connection with presbycusis (Gratton et al., 1997). A possible explanation would be that hair cell loss leads to less demand for strial activity with subsequent atrophy of the stria vascularis. An argument in favor of this hypothesis would be the local character of the decrease in strial cross-sectional area, both after cisplatin (this study) and after gentamicin treatment (Forge et al., 1987). The largest decrease is found in the basal turn while hair cell loss is also most prominent
Table 2 Mean area of the interdigitations formed by the marginal and intermediate cells, relative to the strial cross-sectional area ( U S.E.M.) Group
Marginal cells
Intermediate cells
CONTROL LONG
0.22 U 0.03 0.17 U 0.01
0.16 U 0.03 0.13 U 0.01
55
Fig. 7. Mean cross-sectional areas in Wm2 ( U S.E.M.) of the di¡erent strial components in the CONTROL, SHORT and LONG groups. *P 6 0.05; **P 6 0.01.
in this region (Forge and Schacht, 2000; Klis et al., 2000, 2002). Acceptance of this hypothesis would imply that strial atrophy in the LONG group is secondary to OHC loss. However, in view of the observations from the SHORT group, we are still facing the issue of the primary target for cisplatin in the cochlea. Although histologically determined strial atrophy may be secondary to OHC loss, disturbances in strial function, as re£ected in the EP, occur simultaneously with OHC loss.
Acknowledgements This study was supported by grants from the Heinsius-Houbolt Fund. The authors wish to thank E.G.J. Hendriksen for his assistance in histological processing of the cochleas.
References Aguilar-Markulis, N.V., Beckley, S., Mettlin, C., 1981. Auditory toxicity e¡ects of long-term cis-dichlorodiammineplatinum II therapy in genitourinary cancer patients. J. Surg. Oncol. 16, 111^123. Aran, J.M., Erre, J.P., 1979. Long-term recording of cochleo-neural potentials in the guinea pig. In: Beagley, H.A. (Ed.), Auditory Investigation. The Scienti¢c and Technological Basis. Clarendon Press, Oxford, pp. 233^261. Campbell, K.C., Meech, R.P., Rybak, L.P., Hughes, L.F., 1999. D-Methionine protects against cisplatin damage to the stria vascularis. Hear. Res. 138, 13^28. Cardinaal, R.M., De Groot, J.C.M.J., Huizing, E.H., Veldman, J.E., Smoorenburg, G.F., 2000. Dose-dependent e¡ect of 8-day cisplatin administration upon the morphology of the albino guinea pig cochlea. Hear. Res. 144, 135^146. De Groot, J.C.M.J., Veldman, J.E., Huizing, E.H., 1987. An improved ¢xation method for guinea pig cochlear tissues. Acta Otolaryngol. (Stockh.) 104, 234^242. De Groot, J.C.M.J., Hamers, F.P.T., Gispen, W.H., Smoorenburg, G.F., 1997. Co-administration of the neurotrophic ACTHð49Þ
HEARES 4766 27-10-03
56
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
analogue, ORG 2766, may reduce the cochleotoxic e¡ects of cisplatin. Hear. Res. 106, 9^19. Forge, A., Schacht, J., 2000. Aminoglycoside antibiotics. Audiol. Neurootol. 5, 3^22. Forge, A., Wright, A., Davies, S.J., 1987. Analysis of structural changes in the stria vascularis following chronic gentamicin treatment. Hear. Res. 31, 253^265. Gill, S.S., Salt, A.N., 1997. Quantitative di¡erences in endolymphatic calcium and endocochlear potential between pigmented and albino guinea pigs. Hear. Res. 113, 191^197. Gratton, M.A., Schulte, B.A., Smythe, N.M., 1997. Quanti¢cation of the stria vascularis and strial capillary areas in quiet-reared young and aged gerbils. Hear. Res. 114, 1^9. Greenwood, D.D., 1996. Comparing octaves, frequency ranges, and cochlear map curvature across species. Hear. Res. 94, 157^162. Klis, S.F.L., O’Leary, S.J., Hamers, F.P.T., De Groot, J.C.M.J., Smoorenburg, G.F., 2000. Reversible cisplatin ototoxicity in the albino guinea pig. NeuroReport 11, 623^626. Klis, S.F.L., O’Leary, S.J., Wijbenga, J., De Groot, J.C.M.J., Hamers, F.P.T., Smoorenburg, G.F., 2002. Partial recovery of cisplatininduced hearing loss in the albino guinea pig in relation to cisplatin dose. Hear. Res. 164, 138^146. Kohn, S., Fradis, M., Podoshin, L., Ben David, Y., Zidan, J., Robinson, E., Nir, I., 1991. Toxic e¡ects of cisplatin alone and in combination with gentamicin in stria vascularis of guinea pigs. Laryngoscope 101, 709^716. Laurell, G., Jungnelius, U., 1990. High-dose cisplatin treatment: Hearing loss and plasma concentrations. Laryngoscope 100, 724^734.
Lohuis, P.J.F.M., Patterson, K., Rarey, K.E., 1990. Quantitative assessment of the rat stria vascularis. Hear. Res. 47, 95^102. Meech, R.P., Campbell, K.C., Hughes, L.P., Rybak, L.P., 1998. A semiquantitative analysis of the e¡ects of cisplatin on the rat stria vascularis. Hear. Res. 124, 44^59. Nakai, Y., Konishi, K., Chang, K.C., Ohashi, K., Morisaki, N., Minowa, Y., Morimoto, A., 1982. Ototoxicity of the anticancer drug cisplatin. Acta Otolaryngol. (Stockh.) 93, 227^232. Santi, P.A., Lakhani, B.N., 1983. The volume density of cells and capillaries of the normal stria vascularis. Hear. Res. 11, 7^22. Schweitzer, V.G., 1993. Cisplatin-induced ototoxicity: The e¡ect of pigmentation and inhibitory agents. Laryngoscope 103 (Suppl. 59), 1^52. Stengs, C.H.M., Klis, S.F.L., Huizing, E.H., Smoorenburg, G.F., 1997. Cisplatin-induced ototoxicity: Electrophysiological evidence of spontaneous recovery in the albino guinea pig. Hear. Res. 111, 103^133. Tange, R.A., Vuzevski, V.D., 1984. Changes in the stria vascularis of the guinea pig due to cis-platinum. Arch. Otorhinolaryngol. 239, 41^47. Van Emst, M.G., Klis, S.F.L., Smoorenburg, G.F., 1997. Identi¢cation of the nonlinearity governing even-order distortion products in cochlear potentials. Hear. Res. 114, 93^101. Wangemann, P., Schacht, J., 1996. Homeostatic mechanisms in the cochlea. In: Dallos, P., Fay, R.R., Popper, A.N. (Eds.), Springer Handbook of Auditory Research. Volume 8. The Cochlea. Springer-Verlag, New York, pp. 130^185.
HEARES 4766 27-10-03
Hearing Research 185 (2003) 49^56 www.elsevier.com/locate/heares
Alterations in the stria vascularis in relation to cisplatin ototoxicity and recovery Steven Sluyter, Sjaak F.L. Klis, John C.M.J. de Groot, Guido F. Smoorenburg
Hearing Research Laboratories, Department of Otorhinolaryngology, University Medical Center Utrecht, Room G.02.531, P.O. Box 85.500, 3508 GA Utrecht, The Netherlands Received 11 March 2003; accepted 11 August 2003
Abstract We have investigated whether or not cisplatin-induced depression of the endocochlear potential (EP), and its subsequent recovery, possesses a morphological correlate in the stria vascularis. Guinea pigs implanted with round window electrodes were treated daily with cisplatin (1.5 mg/kg/day) until the compound action potential showed a profound hearing loss ( v 40 dB at 8 kHz after 5^18 days). Animals were either sacrificed immediately after the shift in hearing threshold (‘SHORT’ group) or allowed to recover for v 4 weeks and subsequently sacrificed (‘LONG’ group). Control animals (‘CONTROL’ group) were not treated with cisplatin. Using stereological methods we measured the total strial cross-sectional area together with the areas occupied by the different strial components: the marginal, intermediate and basal cells. The total strial cross-sectional area in the basal turn of the LONG group was found to be significantly smaller than that of the SHORT and the CONTROL groups, whereas the EP was normal in the LONG group (in comparison to the CONTROL group) and markedly decreased in the SHORT group. The smaller area in the LONG group was mainly due to a decrease in the area occupied by the intermediate cells and to a lesser extent to a decrease in the marginal cell area. The area occupied by the basal cells did not change. Thus, the marked decrease in EP after 5^18 days of cisplatin administration was not related to shrinkage of the stria vascularis. Moreover, 4 weeks later the EP showed full recovery, whereas the stria vascularis had shrunk markedly. @ 2003 Elsevier B.V. All rights reserved. Key words: Cisplatin; Ototoxicity; Stria vascularis; Endocochlear potential; Guinea pig
1. Introduction Cisplatin [cis-diamminedichloroplatinum(II)] is an important anticancer drug, which is used in the treatment of various types of malignant tumors. Ototoxicity, however, is a serious side e¡ect of cisplatin. Extensive >experimental research has shown that cisplatin exerts a toxic e¡ect on the outer hair cells (OHCs) (Schweitzer, 1993; De Groot et al., 1997; Cardinaal et al., 2000) and
* Corresponding author. Tel.: +31 (30) 2507581; Fax: +31 (30) 2541922. E-mail address: [email protected] (G.F. Smoorenburg). Abbreviations: A1, lower apical turn; A2, second apical turn; A3, upper apical turn; ANOVA, analysis of variance; B1, lower basal turn; B2, upper basal turn; CAP, compound action potential; EP, endocochlear potential; M1, lower middle turn; M2, upper middle turn; OHC, outer hair cell
the stria vascularis (Tange and Vuzevski, 1984; Kohn et al., 1991; Meech et al., 1998; Campbell et al., 1999; Cardinaal et al., 2000), resulting in permanent sensorineural hearing loss. However, recovery from cisplatininduced hearing loss has been reported occasionally in both humans (Aguilar-Markulis et al., 1981; Laurell and Jungnelius, 1990) and guinea pigs (Nakai et al., 1982; Stengs et al., 1997). In albino guinea pigs, Klis et al. (2000, 2002) studied recovery from cisplatin-induced hearing loss, measuring the compound action potential (CAP) via implanted round window electrodes. The animals were treated daily with i.p. injections of cisplatin (1.5^2.0 mg/kg/day) until a profound hearing loss occurred (criterion : v 40 dB shift in CAP threshold at 8 kHz, CAP threshold de¢ned as the 3 WV iso-response level). CAP threshold gradually recovered in the ¢rst 2 weeks after cessation of treatment. In the studies of Klis et al. (2000, 2002) a
0378-5955 / 03 / $ ^ see front matter @ 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0378-5955(03)00260-0
HEARES 4766 27-10-03
50
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
similar recovery was observed in the endocochlear potential (EP) that was found to be markedly depressed immediately after the profound CAP threshold shift occurred. In addition, histologically, OHC loss was found with the losses predominantly in the basal turns. OHC counts did not di¡er between a group of animals processed before they had a chance to recover (within 3 days after termination of treatment) and a group of animals allowed to recover for up to 100 days. Together, these ¢ndings suggested that the loss and subsequent recovery of the CAP is related to loss and recovery of the EP. Since the EP is generated through strial activity (Wangemann and Schacht, 1996), it can be surmised that loss and recovery of the EP would have a morphological correlate in the stria vascularis. Thus, as a logical follow-up on our previous studies (Klis et al., 2000, 2002), we have performed a quantitative analysis of several morphological parameters of the stria vascularis in relation to cisplatin ototoxicity and its recovery in albino guinea pigs. The approach of this study was to measure the total cross-sectional area of the stria vascularis and to determine the area occupied by the di¡erent strial components (marginal, intermediate and basal cells as well as the strial capillaries and the intercellular space). Furthermore, we have compared these morphological data with EP data in each individual animal.
2. Materials and methods 2.1. Animals and experimental design Healthy, female albino guinea pigs (strain Dunkin Hartley), weighing 250^350 g, were used. They were housed under standard laboratory conditions and fed ad libitum. Cisplatin (Platosin0 ; Pharmachemie, Haarlem, The Netherlands) was diluted with physiological saline (pH 7.4) to a ¢nal concentration of 0.1 mg/ml. The high dilution was chosen in order to stimulate diuresis and, thus, to minimize renal e¡ects. All animals were treated with daily i.p. injections of cisplatin at a dose of 1.5 mg/kg until a shift of 40 dB or more in the CAP threshold at 8 kHz occurred. The 8 kHz CAP threshold was chosen as a global cochlear sensitivity indicator because of the high reproducibility at this frequency, but losses occurred at higher and lower frequencies as well (Klis et al., 2002). When criterion loss was reached, cisplatin administration was terminated and the animals were processed for histological examination of the cochlea, either within 1^3 days after termination of cisplatin administration (SHORT group; n = 5) or after 4 weeks or more (LONG group; n = 14). These time intervals were based upon the earlier work of Klis et al. (2000, 2002). In both groups the EP was
measured before harvesting the cochleas. The two groups were complemented by a CONTROL group (n = 7) which was not treated with cisplatin but subjected to the same measurements. The Animal Care and Use Committee of the Medical Faculty of the University of Utrecht approved the care and use of animals in this study under numbers 91035, 89007, and 89055 for electrode implantation, electrocochleographic techniques and histological techniques, respectively. 2.2. Electrophysiology Details about the surgical techniques and electrophysiologic measurements have been described previously and will not be outlined here in detail (Van Emst et al., 1997; Klis et al., 2000, 2002). The method for chronic round window recording was adopted from Aran and Erre (1979). Under sterile conditions the bulla of the right ear was opened retro-aurically and an electrode was placed on the round window. Electrophysiological measurements were performed di¡erentially with the round window electrode as the active electrode and two screws on the skull as reference and ground electrodes respectively. CAP threshold (iso-response level at 3 WV) was measured daily, after which injections with cisplatin were given. In all animals, the EP in the second turn from the base of the right cochlea was measured with a standard glass pipette microelectrode ¢lled with 150 mM KCl. The EP was measured at this location for technical reasons, but the EP at this location can be taken to be representative for the EP in the entire cochlea because of the good conducting properties of the endolymphatic compartment. 2.3. Histology Immediately following the EP measurement, the cochleas of both ears were ¢xed by intralabyrinthine perfusion with a tri-aldehyde ¢xative (3% glutaraldehyde, 2% formaldehyde, 1% acrolein, and 2.5% dimethyl sulfoxide in 0.08 M sodium cacodylate bu¡er, pH 7.4; De Groot et al., 1987) followed by immersion in the same ¢xative for 3 h at room temperature. Following several rinses (2U15 min) in 0.1 M sodium cacodylate bu¡er (pH 7.4), the cochleas were decalci¢ed in 10% EDTAW2Na for 5 days at room temperature. Next, the specimens were post¢xed in 1% OsO4 containing 1% K4 Ru(CN)6 in 0.1 M sodium cacodylate bu¡er (pH 7.4) for 2 h at 4‡C (De Groot et al., 1987). Dehydration was performed in a graded ethanol series and the cochleas were embedded in toto in Spurr’s low-viscosity resin. The cochleas were divided into two cochlear halves along a midmodiolar plane. The orientation of the midmodiolar plane was the same in all cochlear
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
51
preparations, using the attachment point of the cochlea onto the medial wall of the bulla and both the oval and round window as anatomical landmarks. After division along this midmodiolar plane, the cochleas were re-embedded in the same resin. For light microscopic assessment, semi-thin (1 Wm) sections were cut with a Histo diamond knife on a Reichert-Jung 2050 microtome, collected on glass slides and stained with 1% methylene blue and 1% azur II in 1% sodium tetraborate.
line lattice with a 10-mm distance between the lines. The number of points overlying each strial component was separately counted as well as the number of points overlying the entire stria vascularis and converted to fractions of cross-sectional area (in Wm2 ). In addition, the same method was used to assess the area that is taken by the interdigitations formed by marginal and intermediate cells, relative to the total strial cross-sectional area.
2.4. Determination of the total strial cross-sectional area
2.6. Statistical analysis
Midmodiolar sections of the right cochleas (in which the EP had been measured) from both experimental groups and the control group were examined using a Zeiss Axiophot light microscope. Images of the individual transections of the respective half turns (two transections each for the basal [B1, B2], middle [M1, M2] and apical [A1, A2] turns) were captured with a JVC TK-1281 color videocamera and digitally stored (TIFF format) for o¡-line processing. We used two adjacent midmodiolar sections and measured all strial transections twice on two separate occasions. Thus, the actual ¢gure for each strial cross-sectional area is based on four measurements. Digitized images were imported into the public domain NIH Image program (developed at the US National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image). Next, the boundaries of the stria vascularis were outlined with a pressure-sensitive stylus on a Wacom UD0608-R digitizer interfaced to a Macintosh computer. Finally, the strial cross-sectional area (in Wm2 ) demarcated by these boundaries was calculated.
The quantitative data were statistically analyzed using analysis of variance (ANOVA; Statistica Software Package). Post-hoc comparisons were made with Tukey’s HSD (honestly signi¢cant di¡erences) test for unequal n (Spjotvoll/Stoline). The criterion for statistical signi¢cance was taken at P = 0.05.
2.5. Determination of the areas of the strial components Ultrathin (90 nm) sections of the re-embedded lower basal turn (B1; located approximately 2.4 mm from the round window) from the left cochleas were cut with a Diatome diamond knife on an LKB Ultrotome Nova and collected on Pioloform-coated, single-slot copper grids. Sections were contrast-stained with 7% uranyl acetate in 70% methanol and Reynolds’ lead citrate, and examined in a JEOL 1200EX transmission electron microscope operating at 80 kV. The entire stria vascularis was photographed in a series of electron micrographs that were printed at a ¢nal magni¢cation of 5500 times. The electron micrographs were assembled into a complete photomontage of the stria vascularis. To determine the relative areas of the di¡erent strial components (i.e., marginal cells, intermediate cells, basal cells, capillaries, and intercellular space) a stereological point-counting method was used (Santi and Lakhani, 1983; Lohuis et al., 1990). Each photomontage was overlaid with a transparent sheet bearing a square
3. Results 3.1. EP recordings The data with regard to loss and recovery of cochlear sensitivity (CAP thresholds) in the LONG and SHORT groups have been published earlier in Klis et al. (2002). CAP thresholds remained stable during the ¢rst days of treatment with cisplatin. Then, after a period which varied considerably between animals (5^18 days from treatment onset), thresholds suddenly increased within 1 or 2 days after which the criterion threshold shift was reached. Subsequent to termination of cisplatin treatment thresholds recovered spontaneously over a period of about 2 weeks, after which no further recovery of thresholds was observed. At the lower frequencies recovery was almost complete, whereas at the higher frequencies recovery was usually poor. The EP values of the LONG, SHORT and CONTROL groups, measured in the second turn, are presented in Fig. 1 (averaged data) and Fig. 3 (individual data). The EP in the
Fig. 1. Mean values of the second turn EP in mV U S.E.M. in the CONTROL, SHORT, and LONG groups.
HEARES 4766 27-10-03
52
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
Fig. 2. Mean strial cross-sectional area in Wm2 ( U S.E.M.) based on measurements in semi-thin midmodiolar sections of the right cochleas from the CONTROL, SHORT, and LONG groups. The cochlear turns are related to frequencies (Greenwood, 1996). B1 relates to 26.2 kHz; B2 to 10.4 kHz; M1 to 5.1 kHz; M2 to 2.7 kHz; A1 to 1.3 kHz; and A2 to 0.7 kHz. *P 6 0.05; **P 6 0.01.
Fig. 3. Strial cross-sectional area in Wm2 based on a point-counting method performed in ultrathin sections of the lower basal (B1) turn in relation to the EP and comparing the CONTROL, SHORT, and LONG groups.
Fig. 4. Light micrographs of the stria vascularis in the lower basal turn (B1) of the CONTROL (A) and LONG (B) groups. Note that the stria vascularis in the LONG group is thinner than that in the CONTROL.
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
LONG group did not di¡er from that in the CONTROL group (82 U 1 mV and 83 U 2 mV, respectively). The EP in the SHORT group was signi¢cantly lower (55 U 4 mV; P 6 0.001). The present EP values in the CONTROL and LONG groups were in the normal range for albino guinea pigs (Gill and Salt, 1997). 3.2. Total strial cross-sectional area Fig. 2 shows the total strial cross-sectional area for all individual transections present in midmodiolar sections of the cochlea. Unexpectedly, the strial cross-sectional area in the LONG group, the group that showed normal EPs, appeared to be smaller than those in both the SHORT and CONTROL groups, especially in the lower basal turn (B1 ; Fig. 4). ANOVA over the entire data set con¢rmed signi¢cant main e¡ects of group (SHORT, LONG, CONTROL) and of cochlear turn. Tukey’s post-hoc test showed that the LONG group was primarily responsible for the main e¡ect of group in the lower basal turn (B1 ; P 6 0.01 with respect to both the SHORT and CONTROL groups). In all other half-turns the LONG group had, on average, the small-
53
est total cross-sectional area, but the statistical criterion of 5% was reached only in the lower middle turn (M1). 3.3. Relation between EP and strial cross-sectional area Fig. 3 shows the relation between the EP in individual right ears and the total strial cross-sectional area in the lower basal turn (B1). This graph clearly shows three distinct clusters of data points. The SHORT group is comparable to the CONTROL group with respect to strial cross-sectional area, but the EPs are markedly lower in the SHORT group. In contrast, the LONG group is comparable to the CONTROL group with respect to the EP, but, on average, the LONG group has a much smaller strial cross-sectional area. 3.4. Determination of the area of the strial components Since the largest decrease in strial cross-sectional area was found in the lower basal turn (B1), we examined this turn in more detail using transmission electron microscopy (Figs. 5 and 6). The areas occupied by the di¡erent cell types (marginal, intermediate and basal
Fig. 5. Transmission electron micrograph of the stria vascularis in the lower basal (B1) turn from a CONTROL animal.
HEARES 4766 27-10-03
54
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
Fig. 6. Transmission electron micrograph of the stria vascularis in the lower basal (B1) turn of an animal from the LONG group.
cells) and by the capillaries and intercellular space were determined for the three groups. Mean values ( U S.E.M.) are presented in Table 1 and Fig. 7. ANOVA showed that the total area occupied by the intermediate cells is signi¢cantly smaller in the LONG group, as compared to both the SHORT (P 6 0.01) and CONTROL groups (P 6 0.05). The area occupied by the marginal cells was signi¢cantly smaller in the LONG group when tested against the CONTROL group (P 6 0.05), but not when tested against the SHORT group (P = 0.30). There was no di¡erence between the three groups with respect to the areas occupied by the basal cells, the capillaries and the intercellular space. Table 2 shows the relative areas occupied by the interdigitations between the marginal and intermediate cells. There were no signi¢cant di¡erences between the two groups.
4. Discussion We hypothesized that the decrease in EP immediately after cisplatin treatment (Klis et al., 2000, 2002) could have a morphological correlate in the stria vascularis. However, there were no obvious morphological changes
in the stria vascularis shortly after termination of cisplatin treatment (SHORT group). In contrast, animals that appeared to recover several weeks after termination of treatment (judging from the electrophysiological recordings: CAP and EP) demonstrated a decreased strial cross-sectional area, especially in the basal turn (Figs. 1 and 2). Further ultrastructural analysis of this decrease in the basal turn showed that it was mainly due to a decrease in the area occupied by the intermediate cells and to a lesser extent to a decrease in marginal cell area. Finally, inspection of the stria vascularis showed no bulging of luminal membranes and no differences in vacuolization between the three di¡erent groups. Our results collected immediately after termination of cisplatin treatment (SHORT group), showing no morphological changes in the stria vascularis, di¡er markedly from those of Meech et al. (1998). They performed a semi-quantitative analysis of the e¡ect of cisplatin on the basal turn of the stria vascularis in the rat shortly after treatment and reported that the strial cross-sectional area in the cisplatin-treated group was signi¢cantly larger than in the controls. Also, they observed speci¢c alterations in the marginal cells, consisting of bulging, rupture and/or compression of the cell body
Table 1 Cross-sectional strial areas in Wm2 ( U S.E.M.) of the di¡erent strial components, as measured in ultrathin sections of the lower basal turn (B1) Group
Marginal cells
Intermediate cells
Basal cells
Strial capillaries
Intercellular space
CONTROL SHORT LONG
2336 U 120 2196 U 83 1941 U 69
2140 U 80 2294 U 98 1574 U 114
600 U 82 525 U 57 661 U 59
633 U 96 616 U 121 358 U 37
18 U 8 21 U 6 67 U 26
HEARES 4766 27-10-03
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
and depletion of cytoplasmic organelles. The di¡erences between both studies may be explained by the species di¡erences (rats vs. guinea pigs) but, more importantly, probably by the di¡erent treatment protocols. Meech et al. (1998) used a single high dose, well above the lethal level as evidenced by the loss of 50% of the animals in their experimental group. We, however, used repeated low-dose cisplatin injections, which are not lethal but still result in auditory e¡ects and allow the animals to recover from these e¡ects. Thus, on the basis of experimental outcome, i.e., non-lethality and small-to-moderate, partly reversible hearing loss, our protocol closely resembles the commonly used human protocols. With respect to di¡erences in treatment protocol, possibly resulting in di¡erences in the strial e¡ect of the drug in guinea pigs, the argument can be extended to earlier work of Tange and Vuzevski (1984) who found no strial changes in guinea pigs after 5^10 days of cisplatin administration (1.5 mg/kg/day) but did observe speci¢c alterations after prolonged treatment with higher cumulative doses. Also, Kohn et al. (1991) reported morphological changes in the stria vascularis in both albino and pigmented guinea pigs shortly after cisplatin treatment, but only at doses higher than those used in our study (2.5 mg/kg/day ; 7 days). We could attribute the decrease in total strial crosssectional area in the LONG group to the reduction in the areas occupied by the intermediate cells and, to a lesser extent, by the marginal cells. The question has to be addressed what might have caused this decrease in the LONG group and why this morphological alteration is not re£ected in the EP measurements. Interestingly, Forge et al. (1987) reported a signi¢cant decrease in strial width 4 weeks after cessation of gentamicin treatment, but not shortly after. In contrast to the present ¢ndings this decrease could be attributed to a decreased volume density of the marginal cells. A decrease in total strial cross-sectional area has also been described in connection with presbycusis (Gratton et al., 1997). A possible explanation would be that hair cell loss leads to less demand for strial activity with subsequent atrophy of the stria vascularis. An argument in favor of this hypothesis would be the local character of the decrease in strial cross-sectional area, both after cisplatin (this study) and after gentamicin treatment (Forge et al., 1987). The largest decrease is found in the basal turn while hair cell loss is also most prominent
Table 2 Mean area of the interdigitations formed by the marginal and intermediate cells, relative to the strial cross-sectional area ( U S.E.M.) Group
Marginal cells
Intermediate cells
CONTROL LONG
0.22 U 0.03 0.17 U 0.01
0.16 U 0.03 0.13 U 0.01
55
Fig. 7. Mean cross-sectional areas in Wm2 ( U S.E.M.) of the di¡erent strial components in the CONTROL, SHORT and LONG groups. *P 6 0.05; **P 6 0.01.
in this region (Forge and Schacht, 2000; Klis et al., 2000, 2002). Acceptance of this hypothesis would imply that strial atrophy in the LONG group is secondary to OHC loss. However, in view of the observations from the SHORT group, we are still facing the issue of the primary target for cisplatin in the cochlea. Although histologically determined strial atrophy may be secondary to OHC loss, disturbances in strial function, as re£ected in the EP, occur simultaneously with OHC loss.
Acknowledgements This study was supported by grants from the Heinsius-Houbolt Fund. The authors wish to thank E.G.J. Hendriksen for his assistance in histological processing of the cochleas.
References Aguilar-Markulis, N.V., Beckley, S., Mettlin, C., 1981. Auditory toxicity e¡ects of long-term cis-dichlorodiammineplatinum II therapy in genitourinary cancer patients. J. Surg. Oncol. 16, 111^123. Aran, J.M., Erre, J.P., 1979. Long-term recording of cochleo-neural potentials in the guinea pig. In: Beagley, H.A. (Ed.), Auditory Investigation. The Scienti¢c and Technological Basis. Clarendon Press, Oxford, pp. 233^261. Campbell, K.C., Meech, R.P., Rybak, L.P., Hughes, L.F., 1999. D-Methionine protects against cisplatin damage to the stria vascularis. Hear. Res. 138, 13^28. Cardinaal, R.M., De Groot, J.C.M.J., Huizing, E.H., Veldman, J.E., Smoorenburg, G.F., 2000. Dose-dependent e¡ect of 8-day cisplatin administration upon the morphology of the albino guinea pig cochlea. Hear. Res. 144, 135^146. De Groot, J.C.M.J., Veldman, J.E., Huizing, E.H., 1987. An improved ¢xation method for guinea pig cochlear tissues. Acta Otolaryngol. (Stockh.) 104, 234^242. De Groot, J.C.M.J., Hamers, F.P.T., Gispen, W.H., Smoorenburg, G.F., 1997. Co-administration of the neurotrophic ACTHð49Þ
HEARES 4766 27-10-03
56
S. Sluyter et al. / Hearing Research 185 (2003) 49^56
analogue, ORG 2766, may reduce the cochleotoxic e¡ects of cisplatin. Hear. Res. 106, 9^19. Forge, A., Schacht, J., 2000. Aminoglycoside antibiotics. Audiol. Neurootol. 5, 3^22. Forge, A., Wright, A., Davies, S.J., 1987. Analysis of structural changes in the stria vascularis following chronic gentamicin treatment. Hear. Res. 31, 253^265. Gill, S.S., Salt, A.N., 1997. Quantitative di¡erences in endolymphatic calcium and endocochlear potential between pigmented and albino guinea pigs. Hear. Res. 113, 191^197. Gratton, M.A., Schulte, B.A., Smythe, N.M., 1997. Quanti¢cation of the stria vascularis and strial capillary areas in quiet-reared young and aged gerbils. Hear. Res. 114, 1^9. Greenwood, D.D., 1996. Comparing octaves, frequency ranges, and cochlear map curvature across species. Hear. Res. 94, 157^162. Klis, S.F.L., O’Leary, S.J., Hamers, F.P.T., De Groot, J.C.M.J., Smoorenburg, G.F., 2000. Reversible cisplatin ototoxicity in the albino guinea pig. NeuroReport 11, 623^626. Klis, S.F.L., O’Leary, S.J., Wijbenga, J., De Groot, J.C.M.J., Hamers, F.P.T., Smoorenburg, G.F., 2002. Partial recovery of cisplatininduced hearing loss in the albino guinea pig in relation to cisplatin dose. Hear. Res. 164, 138^146. Kohn, S., Fradis, M., Podoshin, L., Ben David, Y., Zidan, J., Robinson, E., Nir, I., 1991. Toxic e¡ects of cisplatin alone and in combination with gentamicin in stria vascularis of guinea pigs. Laryngoscope 101, 709^716. Laurell, G., Jungnelius, U., 1990. High-dose cisplatin treatment: Hearing loss and plasma concentrations. Laryngoscope 100, 724^734.
Lohuis, P.J.F.M., Patterson, K., Rarey, K.E., 1990. Quantitative assessment of the rat stria vascularis. Hear. Res. 47, 95^102. Meech, R.P., Campbell, K.C., Hughes, L.P., Rybak, L.P., 1998. A semiquantitative analysis of the e¡ects of cisplatin on the rat stria vascularis. Hear. Res. 124, 44^59. Nakai, Y., Konishi, K., Chang, K.C., Ohashi, K., Morisaki, N., Minowa, Y., Morimoto, A., 1982. Ototoxicity of the anticancer drug cisplatin. Acta Otolaryngol. (Stockh.) 93, 227^232. Santi, P.A., Lakhani, B.N., 1983. The volume density of cells and capillaries of the normal stria vascularis. Hear. Res. 11, 7^22. Schweitzer, V.G., 1993. Cisplatin-induced ototoxicity: The e¡ect of pigmentation and inhibitory agents. Laryngoscope 103 (Suppl. 59), 1^52. Stengs, C.H.M., Klis, S.F.L., Huizing, E.H., Smoorenburg, G.F., 1997. Cisplatin-induced ototoxicity: Electrophysiological evidence of spontaneous recovery in the albino guinea pig. Hear. Res. 111, 103^133. Tange, R.A., Vuzevski, V.D., 1984. Changes in the stria vascularis of the guinea pig due to cis-platinum. Arch. Otorhinolaryngol. 239, 41^47. Van Emst, M.G., Klis, S.F.L., Smoorenburg, G.F., 1997. Identi¢cation of the nonlinearity governing even-order distortion products in cochlear potentials. Hear. Res. 114, 93^101. Wangemann, P., Schacht, J., 1996. Homeostatic mechanisms in the cochlea. In: Dallos, P., Fay, R.R., Popper, A.N. (Eds.), Springer Handbook of Auditory Research. Volume 8. The Cochlea. Springer-Verlag, New York, pp. 130^185.
HEARES 4766 27-10-03