Search

The middle fossa approach is useful for the removal of small acoustic tumors when hearing preservation is possible. This approach provides complete exposure of the contents of the internal auditory canal and positive facial nerve identification. Because access to the posterior fossa is limited, the middle fossa approach is most appropriate for tumors with less than 5 mm extension into the cerebellopontine angle. With the introduction of gadolinium-enhanced magnetic resonance imaging, very small acoustic tumors are diagonosed more frequently, and the middle fossa approach is well-suited for the removal of these tumors.

The present studywas done to evaluate the performance of short electrode, single channel, low cost cochlear implant working with analogue strategy.Both prelingual and post lingual patients were selected for the study. Results were assessed using standard protocol established by us evaluation of these patients in the local languagues.For two years follow up, overall results for post lingual patients are satisfactory.Incase of prelingual patients results are very encouraging as well.If we consider the cost v/s. the performance, these low cost implents can be the ideal implant for the Indian patients.

We present the 20-year case report of an auditory implant placed on the dorsal cochlear nucleus with long-term electrical stimulation in a patient with neurofibromatosis 2. The patient has continued to use her implant daily for 20 years, and it has greatly enhanced her quality of life. There have been no adverse sequelae.

Cochlear implants are now an accepted treatment in the rehabilitation of sensorineural hearing impairment. An overview of our understanding of how implants work, the clinical indications, and the current expected results of implantation are briefly reviewed. This is not a comprehensive review but rather a brief summary on the current knowledge of cochlear implants.

Seventeen patients with petroclival meningioma were operated on through a middle fossa transpetrous approach. This approach exposes the anterior cerebellopontine angle through a middle fossa craniotomy with removal of the petrous apex medial to the cochlea and petrous carotid artery. This approach may be enlarged by transection of the superior petrosal sinus and tentorium. The surgical technique and application of the middle fossa transpetrous approach for petroclival meningiomas is presented.

The entire endolymphatic duct and sac as well as the vestibular epithelia were obtained from four patients with Meniere's disease during translabyrinthine (TL) eighth nerve section and from 12 patients undergoing TL resection of acoustic schwannomas. After these specimens were processed for routine transmission electron microscopy (TEM), they were studied for morphologic evidence of viral infection. Although no virus particles were identified, numerous regularly occurring cell components and artifacts were found to morphologically mimic viruses. An atlas of these structures is presented.

Hearing conservation surgery for small acoustic neuromas is well accepted. At present, two approaches are primarily used: the suboccipital and the middle fossa. The middle fossa approach to the internal auditory canal has the advantage of using bony landmarks to identify and protect the facial nerve. Because of anatomic constraints presented by the superior semicircular canal however, its uses are limited to intracanalicular tumors or tumors protruding only slightly into the cerebellopontine angle. By extending the approach through the superior semicircular canal, a wide access to the cerebellopontine angle can be safely obtained. In this study three procedures, two through the superior semicircular canal and one through the posterior semicircular canal, were utilized for hearing conservation surgery. By immediately sealing off the canal ends, hearing preservation was accomplished in two out of three of these cases. The dictum that labyrinthine opening invariably leads to anacusis should be reconsidered. By utilizing approaches through the semicircular canal, it is possible that morbidity from this surgery may be reduced.

The middle fossa approach is well suited for the removal of small acoustic tumors with possible hearing preservation. The most appropriate candidates have tumors with less than 5 mm extension into the cerebellopontine angle and good preoperative hearing (speech reception threshold less than or equal to 30 dB, speech discrimination score greater than or equal to 70%). Measurable postoperative hearing can be preserved in 31% to 59% of patients, and normal or near normal facial function occurs in 86% to 89%. Serious postoperative complications are rare with this approach. With the advent of gadolinium-enhanced MRI, it is now possible to diagnose acoustic tumors reliably when small and before hearing has been significantly affected. The middle fossa approach provides excellent access for the removal of these small tumors.

The endolymphatic sac (ES) may play a crucial role in the pathophysiology of Ménière's disease. This paper presents a technique to obtain and process fresh human specimens of the endolymphatic duct (ED) and the presumably more active intraosseous portion of the ES obtained at surgery. The specimens are preserved with an intact bony shell around the ED and the intraosseous ES. This allows ultrastructural histopathological evaluation of the intraluminal contents, the epithelium and the subepithelial tissue as well as the mutual relationships of these structures. Various factors influencing ES ultrastructure are discussed. The results obtained from this method may increase our understanding of the possible role of the ES in the etiology of different inner ear disorders.

This paper presents results of a histologic study of 16 temporal bones with cochlear implants from 13 subjects. Damage caused by electrode insertion in the basal turn of the cochlea was evaluated. Dendrite and spiral ganglion cell populations were compared to clinical performance scores to determine structures necessary for stimulation and the minimum number needed for electrical stimulation. Results show that damage from insertion of long electrodes was located mainly at the most anterior part of the basal turn; that despite total degeneration of dendrites in the area near the electrode, some spiral ganglion cells remained; and that spiral ganglion cells or possibly axons are the stimulated structures and that fewer of them than previously thought are necessary to achieve a hearing sensation from electrical stimulation.

Residual hearing conservation may be important in cochlear implantation of children and of adults with disabling tinnitus responsive to extracochlear stimulation. Damage to the neural elements of the cochlea during electrode insertion may have a negative impact on residual hearing conservation. Histologic analysis of eight temporal bones with cochlear implants reveals trauma at essentially two locations: the round window insertion site and along the basal turn of the cochlea. In four of the bones, insertion at the round window resulted in damage to the osseous spiral lamina and the electrode was inserted through the scala media into the scala vestibuli. Evidence of secondary reactive osseous formation was also noted in these bones. This paper relates the surgical anatomy of the round window to histologic findings and microanatomical dissections. Recommendations for implantation surgery include creation of a cochleostomy by removal of the floor of the round window niche and a superior-to-inferior angle of electrode insertion.

Twenty-two temporal bones and one brain stem from 13 cochlear implant patients were examined histologically. Sixteen temporal bones had undergone one or more implant procedure. Results of analysis suggested that the ganglion cells were the responding elements to the implant and that useful auditory sensation could result from as few as 10 percent of the normal number of ganglion cells. All implanted bones exhibited varying amounts of fibrosis (some ossified) in the basal turn of the cochlea and beyond in some cases. Usually there was damage to the surviving elements of the organ of Corti and the dendrites throughout the extent of the electrode insertion. However, the ganglion cell population was not affected. Prolonged electrical stimulation (up to 14 years) did not affect ganglion cell survival in three cases, and had no effect on the cochlear nerves in two cases or on cochlear nuclei in one case.

The results of a light and transmission electron microscopic analysis of an endolymphatic sac (ES) from a patient suffering from episodic vertigo, tinnitus, and hearing loss are presented. A biopsy of the intraosseous portion of the ES was obtained during a translabyrinthine approach to section the vestibular nerve in the internal acoustic meatus. The material consisted mainly of tubular epithelial structures filled with heavily stained material. Pathologically dilated and degranulated rough endoplasmic reticuli and disaggregation of polyribosomes with accumulation of solitary ribosomes in the cytosol and endoplasmic reticulum suggested a disturbed epithelial cell protein synthesis. Ultrastructural evidence of an increased merocrine secretion of glycoprotein conjugates into the ES was noted. This made it possible to analyze the presumed intracellular secretory pathways. An increased number of intraepithelial lymphocytes and monocytes was observed. Since the inner ear had been subjected to surgical intervention before the vestibular nerve section, no conclusions can be drawn as to whether the patient's symptoms were related to the disturbed protein metabolism and hypersecretion of glycoprotein conjugates into the ES. The findings support earlier experimental results that indicate that the ES has not only a resorptive function but also a secretory one.

A surgical technique is presented to obtain the entire vestibular aqueduct, containing the endolymphatic duct and sac as well as the vestibular epithelia, from the maculae and cristae during labyrinthectomy. The inner ear tissue was fixed in 3% glutaraldehyde, decalcified in 0.1 mol/L Na-EDTA, and routinely processed for transmission electron microscopy, including post-fixation with osmium tetroxide. Postembedding protein A-colloidal gold immunoelectron microscopy was performed after exposure of cellular antigenic sites by sodium metaperiodate. To validate this technique, thin sections from the crista ampullaris and the endolymphatic sac were incubated with antibodies raised against S-100, a protein present in specific types of neural tissue. Specific immunoreactivity was observed in the crista ampullaris, but not in the endolymphatic sac. The surgical biopsy technique described provides a means for the neuro-otologist to collect inner ear tissue from surgical patients that can be used for pathological studies using immunohistochemistry and electron microscopy.

Insertion of electrodes into the cochlea can damage the spiral ligament, organ of Corti, osseous spiral lamina, and dendrites. The amount of damage depends on the method of insertion and the amount of surgical trauma. Damage is limited to the area immediately adjacent to the electrode. Damage in the cochlea and prolonged electrical stimulation do not affect ganglion cell population. The ganglion cell population responds directly to the electrical stimulus. Examination of three auditory nerves from three cases and the cochlear nuclei from one case also reveals no effect of prolonged electrical stimulation.

The growing number of options in the surgical management of skull base disease has renewed interest in the microvascular anatomy of the lateral temporal region. We studied this anatomy by injecting colored solutions of methyl methacrylate into the major blood vessels of six human cadaver heads or selectively into their major branches. We used several techniques to see the vascular anatomy and to study its relationship to the layers of the scalp. Results revealed that every anatomically named blood vessel was accompanied by a finer, deeper blood vessel supplying the periosteum and outer table of the skull. These vessels arborized into a network of capillaries in the periosteum adherent to the outer bony cortex, from which we saw tiny perforators entering bone. This layered blood supply has direct implications for both ablative and reconstructive surgery for skull base disease. We also saw a previously undescribed arterial plexus accompanying the commonly described venous plexus in the infratemporal fossa.

Hearing improvement after removal of an acoustic tumor is an uncommon occurrence. Hearing improvement was observed in 8.5% of acoustic tumor removals performed using the middle fossa approach in an attempt to preserve hearing. Improvement in speech discrimination occurred most frequently and was of greatest magnitude compared with changes in the speech reception threshold or pure-tone average. No preoperative factors were predictive of postoperative hearing improvement. Findings support the conclusion that candidates for hearing preservation surgery should be chosen on the merits of their existing preoperative hearing and not on the basis of anticipated improvement.

Arachnoid cysts of the posterior fossa are rare. When arachnoid cysts are encountered, the presenting symptoms are frequently otologic, with hearing loss and imbalance occurring commonly. Three cases are presented with a previously unreported otologic symptom, that of bilateral hearing loss, which in one case was fluctuant. None of the patients had the common symptoms of unilateral hearing loss and headache. With the advent of computed tomography and magnetic resonance imaging, these cysts may be readily identified, usually with diagnostic imaging alone. Unfortunately there is often a delay in diagnosis because of the vague and fleeting nature of the symptoms. Because no single diagnostic symptom pattern is able to characterize all cases, it is believed computed tomography or magnetic resonance imaging or both are indicated in patients with long-standing otologic complaints--even in the absence of unilateral symptoms. Treatment of posterior fossa arachnoid cysts primarily consists of surgical procedures designed to decompress the cyst. In this series, treatment with diuretics alone resulted in improvement of symptoms during several years of followup, with no evidence of enlargement of the cysts.

The auditory brain stem implant is an investigational device designed to provide hearing sensations to patients without functioning auditory nerves. We analyzed results from 17 implants in 15 patients to determine if tumor size or preoperative hearing level might be related to proper device function. We found no significant correlation between preoperative hearing level or tumor size and device function. We also found no significant correlation between preoperative hearing level and tumor size in these 15 patients.

Unilateral acoustic neuromas in only-hearing ears and bilateral acoustic neuromas (NF-2) are separate entities, but both pose a common problem because surgical removal has the potential to leave the patient totally deafened. A middle fossa decompression of the internal auditory canal (IAC) was performed in 8 patients (5 with NF-2 tumors and 3 with neuromas in an only-hearing ear). In 5 of the 8, the speech discrimination scores at the 6-month follow-up were better than preoperative scores. After 6 months, however, hearing regressed at variable rates. Although not a definitive therapeutic treatment, decompression of the IAC appears to improve and perhaps prolong useful hearing, which gains valuable time for rehabilitation. Rigid follow-up by computed tomography scans or magnetic resonance imaging is essential.

We performed quantitative assays for estrogen, progesterone, and testosterone receptors in 19 acoustic neuroma specimens from ten men and nine women. No patient received preoperative or intraoperative glucocorticoids, mineralocorticoids, or sex hormones. All tumors were unilateral and removed by the translabyrinthine approach. No tumor specimen was positive for estrogen or testosterone receptors. Three of ten men and seven of nine women had tumors positive for progesterone receptors (10 fmol/mg cytosol protein or greater). This between-sex difference was of borderline significance by the Mann-Whitney U test (p = 0.08). This finding indicates the need for further investigations of endocrinologic therapy as a possible treatment of acoustic neuromas.

The role of endolymphatic sac (ES) dysfunction in the etiopathogenesis of Meniere's disease has remained controversial since the early 1900s. The first reports of the ultrastructural (transmission electron microscopy, TEM) pathology of the human ES in Meniere's disease have been published only in the last decade. These studies have been based on biopsies of the extraosseous (intradural) ES and in no cases has the TEM appearance of the intraosseous ES been described. Likewise the control material used has been from biopsies of extraosseous ES taken from patients with acoustic schwannomas. To date, no reports have compared the ultrastructure of the intrasosseous ES from normal control patients to patients with Meniere's disease. Since the intraosseous ES is believed to be the most active portion of the entire ES, studies were made of the ultrastructure of ten normal interosseous human ESs fixed immediately after death and obtained at autopsy (control material). Fourteen patients undergoing translabyrinthine (TL) neurotologic procedures (10, TL resection of acoustic schwannoma; 4, TL eighth cranial nerve section for Meniere's disease) had the entire vestibular aqueduct, containing the endolymphatic duct and the intraosseous ES, removed and processed for TEM. The roles of the epithelium, subepithelial space, and vasculature were morphologically studied to evaluate possible ES pathology in Meniere's disease and in patients with acoustic schwannoma. Wide anatomic variation in the distribution and density of the subepithelial connective tissue was observed in all groups. There was no difference in the TEM appearance of the intraosseous ES from normal controls and patients with eighth nerve schwannoma, nor was there any difference in the ES collagen deposition in patients with Meniere's disease. The ESs from two patients with Meniere's disease showed evidence of abnormal glycoprotein metabolism; one with possible hypersecretion and one with possible alteration of degradation of resorbed glycoprotein. The results of this preliminary study suggest that "perisaccular fibrosis" of the intraosseous ES was not a pathologic feature in these four cases of Meniere's disease and that alteration of ES glycoprotein secretion/resorption may be of etiopathologic significance.

Eight investigational sites are involved in clinical trials of the 3M/House cochlear implant for children. Statistical analyses have been performed to examine differences between investigators in patient populations selected for implantation and in the results achieved. Important characteristics of the subjects from different investigational sites were compared for House, Maddox, and all "others" combined. There were no differences between investigators in sex, age at surgery, or duration of deafness of the children enrolled in the study. For age at onset, House and Others had higher proportions of postlingually deaf subjects than Maddox (16% for House and 18% for Others, compared with less than 2% for Maddox). However, all investigators had some subjects in every category: postlingual, prelingual, and congenital. Auditory discrimination and speech production scores for subjects from the three groups were compared using analysis of variance techniques. A few interaction effects between time interval and investigator indicated that not all investigator groups performed similarly over time for all measures. However, postimplant scores were always significantly higher than preimplant scores for all investigators. Differences that did occur appeared to be attributable to differences in subject characteristics.

This report examines the durability of preserved postoperative hearing in 25 middle fossa acoustic tumor patients with a minimum follow-up of 3 years. The mean follow-up time for this group was more than 8 years, with a maximum of almost 20 years. The initial postoperative audiogram was compared to the most recent audiogram for each patient, with change in the nonoperated ear serving as the control. Fourteen of the patients (56%) had a significant loss of the preserved hearing in the operated ear over time. The mean loss of speech discrimination was 25%, and the mean loss of speech reception threshold was 12 dB. Only one of the 14 patients had a similar loss in the contralateral ear. No recurrent tumors were identified. Good preoperative hearing is an obvious criterion for selection of candidates for hearing preservation surgery. Results of this study emphasize that a good initial postoperative hearing level is necessary to offset the potential deterioration of hearing that may occur over time.

We have performed a histological examination of the pars petrosa of 13 implanted patients. Morphological data were confronted with the clinical findings. The results of this study show that the nerve elements stimulated by the electrode are the cells of the spiral ganglion, and that a small number of ganglionic cells, as few as 10% of the normal figure, is compatible with auditory perception by electrical stimulation. Furthermore, all the implanted pars petrosa were the seat of a fibrosis and neo-ossification reaction around and sometimes even beyond the tip of the electrode. The trauma caused by the insertion of the electrode into the cochlea involved the sensorial cells of the organ of Corti and the dendritic populations in the lamina spiralis, but the cells of the spiral ganglion were spared. Prolonged electrical stimulation (14 years in 1 case) does not appear to have any deleterious effect on the survival of the spiral ganglion, of the cochlear nerve or of the cochlear nucleus.

The cochlear implant offers an alternative to selected profoundly deaf patients in whom conventional treatments or prostheses have been unsuccessful. Substantial benefits have been obtained by patients, with apparently minimal risk. There have been no serious surgical complications and only a few postoperative problems, which were remedied satisfactorily. The device has continued to function in patients over a period of years, and in cases of internal coil failure it has been removed and replaced. Replacement devices seem to function as well as the patient's first device. There has been no evidence of any decrement in performance after years of electrical stimulation or any measurable adverse effects. The cochlear implant is now widely available and should be considered an option to be offered to appropriate patients.

Intraosseous endolymphatic sacs obtained from patients with acoustic neuromas who had undergone total labyrinthectomy during tumor removal were examined for the presence of T helper/inducer and T suppressor/cytotoxic lymphocytes, B lymphocytes, plasma cells, and macrophages. Immunoperoxidase staining of cryostat sections revealed the presence of T helper/inducer lymphocytes, T suppressor/cytotoxic lymphocytes, and macrophages. The number of B lymphocytes and plasma cells was much smaller than the number of T lymphocytes. The number of T suppressor/cytotoxic lymphocytes was higher than the number of T helper/inducer lymphocytes. This study supports the notion of local immune responsiveness in the human inner ear. This is the first immunohistochemical study to analyze lymphocyte subpopulations; specifically, to provide insight into T-cell function in the endolymphatic sac.

We suggest a new explanation for the lack of auditory response to electric stimulation in children with cochlear implants: the very narrow internal auditory canal, 1 to 2 mm in diameter, and the probable absence of the cochlear nerve. This defect can be seen on high-resolution computed tomographic x-ray studies and may represent aplasia of the auditory-vestibular nerve. We report on eight children with this anomaly, three of whom have received implants and failed to respond with a sensation of sound. Identification of this problem on screening x-ray films is a contraindication to cochlear implantation for auditory stimulation.

The histopathologic findings in the temporal bones of two patients who underwent a cryo-otic-perotic shunt procedure are reported. Morphologic changes in the inner ear were limited to the area of cryoprobe application on the promontory. The hook region of the cochlear duct showed severe atrophy of the spiral ligament, stria vascularis, and sensorineural structures. The ampulla of the posterior semicircular canal showed fibrosis in the perilymphatic space and ruptured membranous walls. No changes were found in the rest of the vestibular labyrinth or in the facial nerve. The long-term improvement in these patients' symptoms, i.e., eight years in one and four years in the other, could only be explained on the basis that endolymph and perilymph mix across intact, nonfunctioning membranes of the cochlear duct or through, presumably antemortem, ruptures of the posterior semicircular canal ampulla.

Five temporal bones and three surgical cochlear nerve specimens from patients who previously underwent labyrinthectomy were studied by light and electron microscopy. All cochleas had nearly total loss of sensory cells, extensive or total loss of both cochlear nerve fibers in the osseous spiral lamina, and spiral ganglion cells. In spite of this severe distal degeneration of the cochlear nerve, it processes within the internal auditory canal appeared normal in four of the five temporal bones, and the majority of the myelinated nerve fibers had survived in all three surgical specimens. Most of the central processes of the primary neurons of the cochlear nerve appear to have resisted retrograde degeneration when it was induced by a lesion primarily causing degeneration of the organ of Corti. This peculiar situation of a few nerve cell bodies and disproportionately large numbers of central axons is discussed.

The death in 1976 of a patient who had a multiple-electrode cochlear implant in one cochlea and a single-electrode cochlear implant in the other provided a unique opportunity for both clinical and histological study of the effects of such implants in a human an subject. The patient was 68 years old at the time of his death. In 1969, at the age of 61, and after 11 years of profound deafness, apparently as a result of lues, the patient was implanted on the left side with a hard-wired 5-wire multiple-electrode cochlear implant. Five years later, his right side was implanted with a single-electrode induction coil cochlear implant. Audiological and psychoacoustic studies were performed by a number of investigators. The results are reviewed. Following the patient's death, his temporal bones were obtained and examined using microdissection techniques followed by serial sectioning. The histological findings are described.

A multiple-electrode cochlear implant comprised of three intracochlear and four extracochlear electrodes is described. A percutaneous plug allowed for direct electrical measurements and presentation of stimuli to selected electrode combinations. Electrical impedance measurements revealed functions that decreased as frequency increase to about 20 kHz. Absolute impedance magnitude varied across electrodes. Equal loudness contours were similar in shape for all three cochlear electrodes, although dynamic range was significantly reduced for the most apical electrode. Periodicity and place pitch were demonstrated; effects were most salient for low frequency stimuli. A 3-channel sound processor was built and multiple-channel stimulation was compared to single-channel stimulation. Significant performance differences were not found between single-channel and multiple-channel systems for environmental or speech sounds. However, the patient subjectively reported superior quality with the multiple-channel system.

Extensive clinical investigation of the cochlear implant, both in adults and children, has provided evidence of significant benefit with relatively minimal risk. Work is in progress world-wide to develop improved cochlear implant device. Meanwhile, an auditory prosthesis is now available for the profoundly deaf patient who could not be successfully treated in the past. It is the result of interaction between medicine and engineering, and represents a new approach to the treatment of profound deafness.

The clinical-histopathological correlation between Menière's disease and endolymphatic hydrops was done to explain, as much as possible, causes of symptoms of Menière's disease. Twenty-three temporal bones with endolymphatic hydrops from 17 patients were reviewed and clinical and histopathological findings were correlated. Histopathological examination revealed frequent, severe deformities in the labyrinthine walls and permanent changes in Reissner's membranes. Evidence of rupture was difficult to assess. Of 21 ears of patients with the clinical diagnosis of Menière's disease, 93% had endolymphatic hydrops. A statistical correlation between increased area of the cochlear duct and hearing loss was found. Some correlation was also found between frequency of vertigo and results of electronystagmography with histopathological findings. Consequently, the mechanical effect of endolymphatic hydrops seems to have greater significance in the production of symptoms of Menière's disease than the biochemical effect of ruptures. In a review of ten unusual cases from this series of 17 patients, traumatic neuromas or remnants of vestibular structures were found after an incomplete labyrinthectomy. Histopathological findings of four patients who had undergone endolymphatic subarachnoid shunt surgery are described.

The rugose portion of two human endolymphatic sacs--one from a Meniere's disease patient, the other from a patient who underwent removal of an eighth nerve tumor and had no other evidence of inner ear disease--has been examined by scanning and transmission electron microscopy. The only morphological finding that distinguished the normal from the Meniere's disease specimen was more folds in the epithelium of the normal sac. Two types of cells have been distinguished in the two sacs through both scanning and transmission electron microscopy: cuboidal-to-columnar cells with a convex luminal surface and microvilli, and squamous cells with a flat luminal surface and microvilli. Qualitative differentiation of the two specimens according to frequency of cellular type or difference in cellular or microvillar morphology was not possible. Observations of this study on the epithelial morphology of the rugose portion did not correspond with descriptions of the epithelial morphology in animals. It is, therefore, tentatively concluded that the morphology of the rugose portion of the human endolymphatic sac is different from that of animals. This study suggests that scanning electron microscopy can be used to determine the number and location of cell types and/or the quantity of microvilli in the rugose portion of the endolymphatic sac. Comparison of measurements of normal endolymphatic sacs and those from Meniere's disease patients may demonstrate significant differences.

Vestibular nerve specimens obtained at neurectomy were studied by light and electron microscopy in 150 patients with various neuro-otological disorders. The vestibular nerve was considered morphologically normal in only 20% of the cases but the changes observed in most cases were interpreted as artifacts or physiological changes of aging. In nearly half of the cases there was excessive fibrosis of the perineurium or endoneurium of the vestibular nerve, or occurrence of glial tissue within or near Scarpa's ganglion. When these changes are of moderate or severe degree, they may be of pathological significance. Severe nerve cell destruction with subsequent fibrosis, indicating primary neuropathy, was found in four cases.