The Vestibular
Aqueduct and
Endolymphatic Sac and Duct in Endolymphatic Hydrops Koos F.
Plantenga, MD, George
G.
Browning, MD, ChB, FRCS
\s=b\ The histologic features of the endolymphatic sac and duct in 23 serially sectioned temporal bones with idiopathic or secondary endolymphatic hydrops were blindly compared with 22 randomly selected, normal temporal bones. In idiopathic hydrops, the pars rugosa of the endolymphatic sac extended out of the vestibular aqueduct into the dura in 29% of bones, compared with none of normal bones (P < .01). In the other 71%, the pars rugosa in the vestibular aqueduct was surrounded by dura more commonly
than
normal.
Functional
studies
are
required to assess the relationship of these findings to hydrops. In secondary hydrops (eg, due to labyrinthitis), the endolymphatic duct was obliterated in the isthmus of the vestibular aqueduct by bone or fibrosis in seven of nine bones. Because of similar ossification and fibrosis elsewhere in the vestibular labyrinth, a direct relationship with hydrops cannot be assumed.
(Arch Otolaryngol 105:546-552, 1979)
hydrops Endolrecognized ymphatic pathologie
is
that
a
well-
entity
in association with congenital syphilis, labyrinthitis, and chronic otitis media,1 but which occurs more often without any recognizable predisposing pathologic features. In a certain proportion of the latter idio¬ pathic group, an association can be made with clinical symptoms of episodic vertigo and/or fluctuating hearing loss, the symptom complex commonly referred to as Meniere's disease. Although the histologie find¬ ings in temporal bones with endolym¬ phatic hydrops are well documented, the underlying physiologic or bio¬ chemical abnormality causing these changes is not understood. Interest has focused on the role of the endo¬ lymphatic sac and duct, mainly be¬ cause of the reliability with which endolymphatic hydrops can be pro¬ duced in guinea pigs by surgically can
Accepted
occur
publication Sept 21, 1978. Department of Otorhinolaryngology, University of Glasgow, Glasgow, Scotland (Dr Browning), and the Ear, Nose, and Throat Department, Wilhelmina Gasthuis, Amsterdam (Dr Plantenga). Reprint requests to Glasgow Royal Infirmary, 84 Castle St, Glasgow G4 OSF, Scotland (Dr Browning). From the
for
obstructing
the vestibular aqueduct,-' and because of a few human case reports in which obliteration or occlu¬ sion of the vestibular aqueduct has been found in association with endo¬ lymphatic hydrops.1 The concept of pathologic features in the vestibular aqueduct is also supported by polytomographic evidence that suggests that in at least 40% of patients with Meniere's disease, the distal portion of the vestibular aqueduct is narrow or cannot be visualized, compared with 5% of normal individuals.'" However, narrowing or nonvisualization of the vestibular aqueduct also occurs in 50% of patients with chronic otitis media,14 and the issue as to whether there is a relationship between the dimensions of the vestibular aqueduct and endo¬ lymphatic hydrops could perhaps be answered by a temporal bone study. Measurements of the width of the isthmus of the vestibular aqueduct, "about 1.5 mm" from the vestibular end, have been reported in serial sections of temporal bones with endo¬ lymphatic hydrops, but no difference was found compared with controls." This last study would benefit from being extended to include the whole vestibular aqueduct, as the radio-
Downloaded From: http://archotol.jamanetwork.com/ by a Georgetown University Medical Center User on 05/24/2015
'
studies were concerned with the distal vestibular aqueduct. However, the histologie evidence to date suggests that in the majority of temporal bones with endolymphatic hydrops there is no obstruction of the vestibular aqueduct."1" Evidence to support the concept of altered func¬ tion of the endolymphatic sac and duct has been looked for and, although the normal function of the endolymphatic sac and duct has not yet been clearly defined, both temporal bone studies1" and biopsies of the endolymphatic sac taken at decompression surgery1"11 showed various abnormalities, partic¬ ularly "fibrosis" of the subepithelial connective tissue. However, none of these studies was well controlled, this
graphie
being obviously important, as even Hallpike and Cairns1" in their original description of endolymphatic hydrops
drew attention to the fact that absence of loose perisaccular connec¬ tive tissue was a "normal variation." The object of the present study was twofold. The first was to reconstruct, on plexiglass sheets, the bony vestibu¬ lar aqueduct in serially sectioned temporal bones with endolymphatic hydrops and to take measurements for comparison with normal temporal bones. The second object was to perform a controlled, blind study of the histologie features of the endo¬ lymphatic sac and duct in the same temporal bones. METHOD
Twenty-three temporal bones with endo¬ lymphatic hydrops were studied, endolym¬ phatic hydrops being an increase in the volume of endolymph manifested by distention of the enclosing membranes with a corresponding diminution of the volume of perilymph. The temporal bones
had been removed at autopsy, immersed in Heidenhain Susa solution, decalcified, de¬ hydrated, and embedded in celloidin. Sections of 20 µ thickness were then prepared, every tenth section being stained with hematoxylin-eosin for study
by light microscopy.
In 14 of these 23 temporal bones, the endolymphatic hydrops could be considered idiopathic; in nine, it was secondary. In the latter nine temporal bones, there was evidence suggestive of congenital syphilis in four, of labyrinthitis in three, of congen¬ ital abnormalities in one, and of chronic otitis media in one. The total of 23 tempo-
ral bones came from 17 individuals: six male, 11 female, with a mean age of 70 years and a range of 42 to 97 years. In these individuals, the endolymphatic hy¬ drops was bilateral in six and unilateral in 11. Excluded from the study were an addi¬ tional 23 temporal bones with endolym¬ phatic hydrops, preparation and postmor¬ tem artifact being severe in ten, the dura and endolymphatic sac having been stripped off during temporal bone removal in eight, only selected sections being avail¬ able in three, and the plane of section not being horizontal in two. For comparison, 22 temporal bones with¬ out any gross pathologic features were studied. These temporal bones had been selected from a numbered list of normal temporal bones by taking random numbers from a randomizing pocket calculator. These temporal bones came from 22 indi¬ viduals with no history of ear disease; eight were male, 14 were female, with a mean age of 63 years and a range of 41 to 84 years. The vestibular aqueduct was recon¬ structed after the method of Waldorf (1974)," using sets of plexiglass sheets 35.0 30.0 0.16 cm, in which four align¬ ment holes for bolts had been drilled near the corners. The appropriate sections containing the vestibular aqueduct were projected (20 times) onto these sheets using a microprojector. The bony outline of the vestibular aqueduct, posterior cranial fossa, facial nerve canal, and vestibular labyrinth were drawn with a writing set. The drawings of contiguous sections could then be aligned on the bolts using standard washers of 0.8 and 0.6 mm thickness to space the plexiglass sheets at the appro¬ priate distances apart, taking into account the magnification factor and the nine intervening unmounted temporal bone sec¬ tions of 20 µ thickness. After assembly, the total length of the vestibular aqueduct was trigonometrically calculated from direct measurements on the model. The vestibular aqueduct was considered to commence at a line drawn perpendicular from the anterior wall to the lip of the posterior wall of the internal aperture, and to terminate in the section above the one in which the vestibular aque¬ duct opened into the foveate fossa. Alter¬ native methods for determining the endpoint were evaluated, but were unreliable because of the absence of a definite anatomical point in the foveate fossa. Taking the most lateral point of the vestib¬ ular aqueduct as a reference point on each sheet, the diagonal distance between them could be calculated by knowledge of the vertical distance between sheets and the horizontal distance apart measured on the
most superior sheet. By summating the diagonal distances between contiguous sheets, the length of the vestibular aque¬ duct could be calculated. Having calculated the total length of the vestibular aqueduct,
the maximum horizontal width at 1-mm intervals along the vestibular aqueduct was measured directly from the appro¬ priate plexiglass sheet. To assess the histologie features of the endolymphatic sac and duct, the name of the individual and the labyrinth were covered with tape in all sets of slides. Each set was then assessed in a random order, the presence or absence of endolymphatic hydrops being unknown. For analysis, the endolymphatic sac was anatomically div¬ ided into the pars rugosa and pars intraduralis; the endolymphatic duct was divided into the sinus and isthmus.4 In each of these divisions, several parameters were assessed and graded by reference to 35 mm color transparencies previously taken of a wide range of normal endolymphatic sacs and ducts. Findings outside this range were also recorded. Particular attention was paid to the fibrous elements in the connective tissue surrounding the endo¬ lymphatic sac and duct, this assessment being facilitated by the use of polarized
light microscopy.
RESULTS Reconstruction of the Vestibular Aqueduct
The vestibular aqueduct was oblit¬ erated by bone in five of the 23 tempo¬ ral bones with endolymphatic hydrops that were studied, an example of which is shown in Fig 1. The bony obstruction started on average 1.8 mm (range, 1.1 to 2.3 mm) from the start of the vestibular aqueduct, and was on average 2.3 mm (range, 1.8 to 2.7 mm) long. These five bones came from three individuals in whom the endo¬ lymphatic hydrops could be considered secondary. In two of these individuals (four bones), there was evidence of congenital syphilis based on the clini¬ cal history and the histologie appear¬ ances of the temporal bones, which included labyrinthine bone résorption and fibrosis within the semicircular canals. In the temporal bone of the other individual, there was ossifica¬ tion and fibrosis of the perilymphatic spaces of the semicircular canals suggestive of labyrinthitis. Thus, the bony obstruction of the vestibular aqueduct was in every case part of a generalized involvement of the vestib-
Downloaded From: http://archotol.jamanetwork.com/ by a Georgetown University Medical Center User on 05/24/2015
ular labyrinth by fibrosis and/or heterotopic ossification. The mean length, and one standard error of the mean, of the vestibular aqueduct in the temporal bones with endolymphatic hydrops was 6.7 ± 0.4 mm, as opposed to 6.9 ± 0.4 mm in normal temporal bones, there being no significant difference (P > .5). There was also no significant difference (P > .5) in the mean length in tempo¬ ral bones where the endolymphatic
hydrops was idiopathic (6.5 ± 0.5 mm) or secondary (7.1 ± 0.6 mm). The mean length of the five temporal
bones with obliteration of the vestibu¬ lar aqueduct was 6.9 ± 0.5 mm. To assess the variability of the method of reconstruction and mea¬ surement of the vestibular aqueduct length, four sets of temporal bones were reconstructed on three separate occasions and the length of the vestib¬ ular aqueduct was calculated from each. The maximum difference be¬ tween the three lengths for each of the four temporal bones was 0.53 mm, and the mean was 0.33 mm. The maxi¬ mum variation from the mean of the three lengths was 11%. The bone surrounding this vestibu¬ lar aqueduct was assessed as being either dense or cellular. The size of the cellular spaces was classified as being either small or large, and their contents were classified as being ei¬ ther marrow or air. There was consid¬ erable variation between bones when analyzed in this manner, and this was found to be dependent on whether the
posterosuperior, posteromedial, hypotympanic, and retrofacial cell tracts were present and the relative position of the jugular bulb. In general, the
small cellular spaces were filled with inactive marrow and the larger cellu¬ lar space were filled equally with inac¬ tive marrow or air. Unfortunately, this detailed analysis had to be summarized otherwise the resultant groups would have been too small. When this was done, no difference could be found between normal bones and those with either idiopathic or
secondary hydrops (Table 1). The mean length of the vestibular aqueduct when the surrounding bone was
dense was 6.5 ± 1.9 mm; when small cellular spaces it was
having
Fig 1.—Horizontal section through temporal bone with secondary endolymphatic hydrops secondary to labyrinthitis showing bony obliteration of isthmus of vestibular aqueduct. Table
1.—Type
of Bone
Surrounding
the Vestibular No.
Normal
Idiopathic hydrops Secondary hydrops Total hydrops
Total 22 14 23
Dense 12 (54)
hydrops. Figure
2 illustrates the horizontal widths of the vestibular aqueduct as calculated at 1-mm intervals along its length in normal temporal bones and in those with idiopathic and secondary hydrops. There was no difference evident in the width of the temporal bones with idiopathic endolymphatic hydrops compared with normal bones. In the temporal bones with secondary hydrops, there was the already stated bony obstruction in five, but in the other four, including two with fibrous obliteration of the endolymphatic duct
(%)
Small Cellular
7(32) 8(57) 0(0) 8(35)
5(36) 7(78) 12(52)
6.3 ± 1.2 mm; and when having large cellular spaces it was 8.8 ± 1.5 mm. Statistical analysis of these figures showed that the vestibular aqueduct was significantly longer (P < .01) in the temporal bones with large cellular spaces compared with the others. However, this difference was not confined to either the normal tempo¬ ral bones or those with endolymphatic
Aqueduct
Large Cellular 3(14) 1(1) 2(22) 3(13)
lumen, there
was no obvious differ¬ in the width of the vestibular aqueduct compared with normal.
ence
Histologie Features of the Endolymphatic Sac Pars Rugosa.—In the normal tempo¬ ral bones and those with secondary hydrops, the pars rugosa was invari¬ ably within the vestibular aqueduct. This was the case in ten (71%) of the 14 temporal bones with idiopathic
hydrops,
a
statistically significant
.01). In the other four (29%) bones, the pars rugosa extended
difference (P
Aqueduct and
Endolymphatic Sac and Duct in Endolymphatic Hydrops Koos F.
Plantenga, MD, George
G.
Browning, MD, ChB, FRCS
\s=b\ The histologic features of the endolymphatic sac and duct in 23 serially sectioned temporal bones with idiopathic or secondary endolymphatic hydrops were blindly compared with 22 randomly selected, normal temporal bones. In idiopathic hydrops, the pars rugosa of the endolymphatic sac extended out of the vestibular aqueduct into the dura in 29% of bones, compared with none of normal bones (P < .01). In the other 71%, the pars rugosa in the vestibular aqueduct was surrounded by dura more commonly
than
normal.
Functional
studies
are
required to assess the relationship of these findings to hydrops. In secondary hydrops (eg, due to labyrinthitis), the endolymphatic duct was obliterated in the isthmus of the vestibular aqueduct by bone or fibrosis in seven of nine bones. Because of similar ossification and fibrosis elsewhere in the vestibular labyrinth, a direct relationship with hydrops cannot be assumed.
(Arch Otolaryngol 105:546-552, 1979)
hydrops Endolrecognized ymphatic pathologie
is
that
a
well-
entity
in association with congenital syphilis, labyrinthitis, and chronic otitis media,1 but which occurs more often without any recognizable predisposing pathologic features. In a certain proportion of the latter idio¬ pathic group, an association can be made with clinical symptoms of episodic vertigo and/or fluctuating hearing loss, the symptom complex commonly referred to as Meniere's disease. Although the histologie find¬ ings in temporal bones with endolym¬ phatic hydrops are well documented, the underlying physiologic or bio¬ chemical abnormality causing these changes is not understood. Interest has focused on the role of the endo¬ lymphatic sac and duct, mainly be¬ cause of the reliability with which endolymphatic hydrops can be pro¬ duced in guinea pigs by surgically can
Accepted
occur
publication Sept 21, 1978. Department of Otorhinolaryngology, University of Glasgow, Glasgow, Scotland (Dr Browning), and the Ear, Nose, and Throat Department, Wilhelmina Gasthuis, Amsterdam (Dr Plantenga). Reprint requests to Glasgow Royal Infirmary, 84 Castle St, Glasgow G4 OSF, Scotland (Dr Browning). From the
for
obstructing
the vestibular aqueduct,-' and because of a few human case reports in which obliteration or occlu¬ sion of the vestibular aqueduct has been found in association with endo¬ lymphatic hydrops.1 The concept of pathologic features in the vestibular aqueduct is also supported by polytomographic evidence that suggests that in at least 40% of patients with Meniere's disease, the distal portion of the vestibular aqueduct is narrow or cannot be visualized, compared with 5% of normal individuals.'" However, narrowing or nonvisualization of the vestibular aqueduct also occurs in 50% of patients with chronic otitis media,14 and the issue as to whether there is a relationship between the dimensions of the vestibular aqueduct and endo¬ lymphatic hydrops could perhaps be answered by a temporal bone study. Measurements of the width of the isthmus of the vestibular aqueduct, "about 1.5 mm" from the vestibular end, have been reported in serial sections of temporal bones with endo¬ lymphatic hydrops, but no difference was found compared with controls." This last study would benefit from being extended to include the whole vestibular aqueduct, as the radio-
Downloaded From: http://archotol.jamanetwork.com/ by a Georgetown University Medical Center User on 05/24/2015
'
studies were concerned with the distal vestibular aqueduct. However, the histologie evidence to date suggests that in the majority of temporal bones with endolymphatic hydrops there is no obstruction of the vestibular aqueduct."1" Evidence to support the concept of altered func¬ tion of the endolymphatic sac and duct has been looked for and, although the normal function of the endolymphatic sac and duct has not yet been clearly defined, both temporal bone studies1" and biopsies of the endolymphatic sac taken at decompression surgery1"11 showed various abnormalities, partic¬ ularly "fibrosis" of the subepithelial connective tissue. However, none of these studies was well controlled, this
graphie
being obviously important, as even Hallpike and Cairns1" in their original description of endolymphatic hydrops
drew attention to the fact that absence of loose perisaccular connec¬ tive tissue was a "normal variation." The object of the present study was twofold. The first was to reconstruct, on plexiglass sheets, the bony vestibu¬ lar aqueduct in serially sectioned temporal bones with endolymphatic hydrops and to take measurements for comparison with normal temporal bones. The second object was to perform a controlled, blind study of the histologie features of the endo¬ lymphatic sac and duct in the same temporal bones. METHOD
Twenty-three temporal bones with endo¬ lymphatic hydrops were studied, endolym¬ phatic hydrops being an increase in the volume of endolymph manifested by distention of the enclosing membranes with a corresponding diminution of the volume of perilymph. The temporal bones
had been removed at autopsy, immersed in Heidenhain Susa solution, decalcified, de¬ hydrated, and embedded in celloidin. Sections of 20 µ thickness were then prepared, every tenth section being stained with hematoxylin-eosin for study
by light microscopy.
In 14 of these 23 temporal bones, the endolymphatic hydrops could be considered idiopathic; in nine, it was secondary. In the latter nine temporal bones, there was evidence suggestive of congenital syphilis in four, of labyrinthitis in three, of congen¬ ital abnormalities in one, and of chronic otitis media in one. The total of 23 tempo-
ral bones came from 17 individuals: six male, 11 female, with a mean age of 70 years and a range of 42 to 97 years. In these individuals, the endolymphatic hy¬ drops was bilateral in six and unilateral in 11. Excluded from the study were an addi¬ tional 23 temporal bones with endolym¬ phatic hydrops, preparation and postmor¬ tem artifact being severe in ten, the dura and endolymphatic sac having been stripped off during temporal bone removal in eight, only selected sections being avail¬ able in three, and the plane of section not being horizontal in two. For comparison, 22 temporal bones with¬ out any gross pathologic features were studied. These temporal bones had been selected from a numbered list of normal temporal bones by taking random numbers from a randomizing pocket calculator. These temporal bones came from 22 indi¬ viduals with no history of ear disease; eight were male, 14 were female, with a mean age of 63 years and a range of 41 to 84 years. The vestibular aqueduct was recon¬ structed after the method of Waldorf (1974)," using sets of plexiglass sheets 35.0 30.0 0.16 cm, in which four align¬ ment holes for bolts had been drilled near the corners. The appropriate sections containing the vestibular aqueduct were projected (20 times) onto these sheets using a microprojector. The bony outline of the vestibular aqueduct, posterior cranial fossa, facial nerve canal, and vestibular labyrinth were drawn with a writing set. The drawings of contiguous sections could then be aligned on the bolts using standard washers of 0.8 and 0.6 mm thickness to space the plexiglass sheets at the appro¬ priate distances apart, taking into account the magnification factor and the nine intervening unmounted temporal bone sec¬ tions of 20 µ thickness. After assembly, the total length of the vestibular aqueduct was trigonometrically calculated from direct measurements on the model. The vestibular aqueduct was considered to commence at a line drawn perpendicular from the anterior wall to the lip of the posterior wall of the internal aperture, and to terminate in the section above the one in which the vestibular aque¬ duct opened into the foveate fossa. Alter¬ native methods for determining the endpoint were evaluated, but were unreliable because of the absence of a definite anatomical point in the foveate fossa. Taking the most lateral point of the vestib¬ ular aqueduct as a reference point on each sheet, the diagonal distance between them could be calculated by knowledge of the vertical distance between sheets and the horizontal distance apart measured on the
most superior sheet. By summating the diagonal distances between contiguous sheets, the length of the vestibular aque¬ duct could be calculated. Having calculated the total length of the vestibular aqueduct,
the maximum horizontal width at 1-mm intervals along the vestibular aqueduct was measured directly from the appro¬ priate plexiglass sheet. To assess the histologie features of the endolymphatic sac and duct, the name of the individual and the labyrinth were covered with tape in all sets of slides. Each set was then assessed in a random order, the presence or absence of endolymphatic hydrops being unknown. For analysis, the endolymphatic sac was anatomically div¬ ided into the pars rugosa and pars intraduralis; the endolymphatic duct was divided into the sinus and isthmus.4 In each of these divisions, several parameters were assessed and graded by reference to 35 mm color transparencies previously taken of a wide range of normal endolymphatic sacs and ducts. Findings outside this range were also recorded. Particular attention was paid to the fibrous elements in the connective tissue surrounding the endo¬ lymphatic sac and duct, this assessment being facilitated by the use of polarized
light microscopy.
RESULTS Reconstruction of the Vestibular Aqueduct
The vestibular aqueduct was oblit¬ erated by bone in five of the 23 tempo¬ ral bones with endolymphatic hydrops that were studied, an example of which is shown in Fig 1. The bony obstruction started on average 1.8 mm (range, 1.1 to 2.3 mm) from the start of the vestibular aqueduct, and was on average 2.3 mm (range, 1.8 to 2.7 mm) long. These five bones came from three individuals in whom the endo¬ lymphatic hydrops could be considered secondary. In two of these individuals (four bones), there was evidence of congenital syphilis based on the clini¬ cal history and the histologie appear¬ ances of the temporal bones, which included labyrinthine bone résorption and fibrosis within the semicircular canals. In the temporal bone of the other individual, there was ossifica¬ tion and fibrosis of the perilymphatic spaces of the semicircular canals suggestive of labyrinthitis. Thus, the bony obstruction of the vestibular aqueduct was in every case part of a generalized involvement of the vestib-
Downloaded From: http://archotol.jamanetwork.com/ by a Georgetown University Medical Center User on 05/24/2015
ular labyrinth by fibrosis and/or heterotopic ossification. The mean length, and one standard error of the mean, of the vestibular aqueduct in the temporal bones with endolymphatic hydrops was 6.7 ± 0.4 mm, as opposed to 6.9 ± 0.4 mm in normal temporal bones, there being no significant difference (P > .5). There was also no significant difference (P > .5) in the mean length in tempo¬ ral bones where the endolymphatic
hydrops was idiopathic (6.5 ± 0.5 mm) or secondary (7.1 ± 0.6 mm). The mean length of the five temporal
bones with obliteration of the vestibu¬ lar aqueduct was 6.9 ± 0.5 mm. To assess the variability of the method of reconstruction and mea¬ surement of the vestibular aqueduct length, four sets of temporal bones were reconstructed on three separate occasions and the length of the vestib¬ ular aqueduct was calculated from each. The maximum difference be¬ tween the three lengths for each of the four temporal bones was 0.53 mm, and the mean was 0.33 mm. The maxi¬ mum variation from the mean of the three lengths was 11%. The bone surrounding this vestibu¬ lar aqueduct was assessed as being either dense or cellular. The size of the cellular spaces was classified as being either small or large, and their contents were classified as being ei¬ ther marrow or air. There was consid¬ erable variation between bones when analyzed in this manner, and this was found to be dependent on whether the
posterosuperior, posteromedial, hypotympanic, and retrofacial cell tracts were present and the relative position of the jugular bulb. In general, the
small cellular spaces were filled with inactive marrow and the larger cellu¬ lar space were filled equally with inac¬ tive marrow or air. Unfortunately, this detailed analysis had to be summarized otherwise the resultant groups would have been too small. When this was done, no difference could be found between normal bones and those with either idiopathic or
secondary hydrops (Table 1). The mean length of the vestibular aqueduct when the surrounding bone was
dense was 6.5 ± 1.9 mm; when small cellular spaces it was
having
Fig 1.—Horizontal section through temporal bone with secondary endolymphatic hydrops secondary to labyrinthitis showing bony obliteration of isthmus of vestibular aqueduct. Table
1.—Type
of Bone
Surrounding
the Vestibular No.
Normal
Idiopathic hydrops Secondary hydrops Total hydrops
Total 22 14 23
Dense 12 (54)
hydrops. Figure
2 illustrates the horizontal widths of the vestibular aqueduct as calculated at 1-mm intervals along its length in normal temporal bones and in those with idiopathic and secondary hydrops. There was no difference evident in the width of the temporal bones with idiopathic endolymphatic hydrops compared with normal bones. In the temporal bones with secondary hydrops, there was the already stated bony obstruction in five, but in the other four, including two with fibrous obliteration of the endolymphatic duct
(%)
Small Cellular
7(32) 8(57) 0(0) 8(35)
5(36) 7(78) 12(52)
6.3 ± 1.2 mm; and when having large cellular spaces it was 8.8 ± 1.5 mm. Statistical analysis of these figures showed that the vestibular aqueduct was significantly longer (P < .01) in the temporal bones with large cellular spaces compared with the others. However, this difference was not confined to either the normal tempo¬ ral bones or those with endolymphatic
Aqueduct
Large Cellular 3(14) 1(1) 2(22) 3(13)
lumen, there
was no obvious differ¬ in the width of the vestibular aqueduct compared with normal.
ence
Histologie Features of the Endolymphatic Sac Pars Rugosa.—In the normal tempo¬ ral bones and those with secondary hydrops, the pars rugosa was invari¬ ably within the vestibular aqueduct. This was the case in ten (71%) of the 14 temporal bones with idiopathic
hydrops,
a
statistically significant
.01). In the other four (29%) bones, the pars rugosa extended
difference (P
