Search

The results of tympanoplasty performed in 72 ears with cholesteatoma from 1988 to 1992 are analyzed. The patients were followed for 6 to 60 months (average 21.4 months). The canal down method was used in 32 ears, the canal up method in 8 ears, and posterior wall reconstruction in 32 ears. Mastoid obliteration was done in 61 ears. Staged and second look operations were performed in 19 and 14 ears, respectively. Fifteen of the 33 ears with staged or second look operations had residual cholesteatoma. Recurrence was observed in one ear treated with the canal up method and in 3 ears treated with posterior wall reconstruction. No recurrence was observed in ears treated with the canal down method. Tympanoplasty type I, type II, type II, columella type III, columella type IV were performed in 8, 1, 1, 48 and 14 ears, respectively. Autologous cartilage from the auricle was used in most ears as a columella. The rate of success of hearing preservation was 82% according to the Criteria of the Japan otological Society. There was no large difference in the rate of success of hearing preservation among the ears treated with the canal down method, canal up method and posterior wall reconstruction.

Dendritic cells, macrophages and lymphocytes infiltrating the epithelia of sino-nasal lesions were investigated immunohistochemically using antibody S-100 protein for dendritic cells, Ki-M1P for macrophages, MT1 for T-cells and L26 for B-cells, respectively. Seventy-five specimens of sino-nasal mucosae involving 51 cases of metaplastic squamous epithelium (MSE), 12 of inverted papilloma, 1 of atypical hyperplasia and 11 of squamous cell carcinoma (SCC) were studied. Dendritic cells were observed in 12 cases of MSE (23.5%), in 4 of inverted papilloma (33.3%), and in 6 of SCC (54.5%). More dendritic cells were found infiltrating MSE of the infective and fibrotic types than of the mild and edematous types, histopathologically. Furthermore, infiltration correlated with the degree of squamous differentiation in MSE, but not in SCC. The number of macrophages was significantly smaller in MSE than in columnar ciliated epithelium, while that in SCC was significantly greater than that in MSE. With dendritic cell infiltration macrophages in MSE as well as SCC and T-cells in and around the carcinoma were often increased. These findings suggest that dendritic cells may be attracted to maturative squamous epithelium, and may play an important role in the immunologic defence mechanisms of MSE. Dendritic cells, in contrast, may play a role in the cellular immune response of T-cells against the carcinoma.

Argyrophilic nucleolar organizer regions (AgNORs) were determined by a semiautomatic image analyzer in a group of 28 patients with T2 or T3 maxillary sinus squamous cell carcinoma. The group was separated into recurrent and non-recurrent groups, metastatic and nonmetastatic groups, as well as groups strongly or weakly positive for epidermal growth factor receptor (EGFR). The mean number and area of AgNOR dots per nucleus and the mean size of each dot were calculated. The mean number and area were significantly greater than those in normal cells (P < 0.001), but not in recurrent and non-recurrent groups, metastatic and non-metastatic groups and groups separated by EGFR. There was also no difference found in the mean sizes of tumor and normal cells. These results suggest that AgNOR is related to malignant transformation, but not prognosis.

Adenocarcinoma of the middle ear is seldom encountered in clinical practice. In the past 20 years, nearly 20 cases of primary adenocarcinoma of the middle ear have been reported, and most cases of these tumors have been described without histologic presentation. To our knowledge, only one case of adenocarcinoma originating in the postoperative middle ear has been reported. We describe a case of primary poorly differentiated adenocarcinoma in the middle ear which was surgically treated for chronic otitis media 50 years ago. Planning diagnostic and therapeutic procedures in connection with middle ear tumors is also discussed.

Pharyngeal fricatives and affricatives in cleft palate speech due to velopharyngeal incompetence have been considered to be produced in the pharynx. Fluorovideoscopic and fiberscopic examinations in five patients revealed that so-called pharyngeal fricatives and affricatives, as judged on the basis of auditory impression, were both produced at a constriction formed by backward tilting of the epiglottis and elevation of the arytenoid. The velopharyngeal port remained open during production of these sounds but was almost closed during stop and vowel productions. These faulty articulations are adequately termed laryngeal fricatives and affricatives rather than pharyngeal fricatives and affricatives.

The ultrastructure and barrier properties of the endolymphatic duct (ED) were examined by light and electron microscopy. ED epithelial cells were classified into two types: type I and type II duct cells. The type I duct cells were cuboidal or low columnar and were characterized by a convex apical surface and a few basal processes. The type II duct cells were squamous and were characterized by a flat apical and basal membrane, many small vesicles and a number of small pits along the basal membranes. After electrophoretic horseradish peroxidase (HRP) injection into the ED lumen, no HRP uptake into ED epithelial cells of either type was observed. There was no reaction product either in the lateral intercellular spaces beyond the tight junctions between ED epithelial cells or in ED subepithelial tissues. The ED epithelial cells were considered to play no active role in apical macromolecular absorption and to be impermeable to intra-luminal macro molecules.

Enophthalmos is usually a consequence of orbital trauma resulting in a blowout fracture of the orbital floor. We report here the case of a 35-year-old male who had a radical operation of the maxillary sinus and the ethmoidal sinus, and 2 months after the operation, developed spontaneous enophthalmos and ocular pain. The left orbit was enophthalmic by 5mm according to Hertel enophthalmometry. Surgical correction was performed under general anesthesia. The left orbit was explored through a transcutaneous incision of the left lower lid which revealed marked periosteal adhesions and absence of the bony floor of the orbit. A rib cartilage graft was used to reconstruct the floor of the orbit, and the globe was brought forward by means of retrobulbar placement. Enophthalmos and ocular pain disappeared postoperatively.

The serum and middle ear tissue concentration of enoxacin, a pyridone carboxylic acid derivative, was measured in 20 patients with chronic otitis media. The drug was given orally in a dose of 600mg 3 times a day for one to 24 days (average 6.3 days) on an average of 13.1 hours before the serum and tissue samples were obtained during tympanoplasty. In the 14 patients, for whom data of both serum and tissue concentrations were obtained, the concentrations were 0.90±0.66μg/ml and 3.32±3.50μg/kg, respectively. The tissue concentration was 384±236% times the serum concentration.

Veloplasty was performed for two patients with micrognathia and difficulty in opening the mouth. They complained of heavy snoring and obstructive sleep apnea. The micrognathia of the patients was congenital in one and acquired in the other. Although they had had operations of the temporomandibular joint, they could open the mouth only one finger breadth between the teeth. Because of the difficulty in opening the mouth, veloplasty was done instead of uvulopalatopharyngoplasty, with complete relief of snoring and obstructive sleep apnea syndrome.

The capillary permeability and the basolateral endocytic pathway of the mouse endolymphatic sac (ES) epithelium were examined in vivo using intravenous injection of horseradish peroxidase (HRP). The capillaries of the ES were classified as either fenestrated or non-fenestrated. Because dense reaction products were observed soon after injection of HRP in macrophages near both types of capillaries, they were both considered to be permeable to macromolecules. In non-fenestrated capillaries, the basement membrane and small vesicles of the endothelium of the ES were stained with reaction product. These non-fenestrated capillaries were considered to be of muscle type. After 15 min, ES epithelial cells absorbed HRP basolaterally, and the multivesicular bodies and lysosomes of epithelial cells were stained with reaction product. The process of basolateral absorption in the ES epithelium was similar to that in the intestinal epithelium. Our results provide further evidence that the ES is a metabolically active organ which plays an important role in fluid transport in the inner ear.

To test the possibility that catecholamines have an influence on the function of the endolymphatic sac (ES), which is known to absorb the endolymph, the effect of catecholamines on the ES direct current potential (ESP) was examined in the guinea pig. The ESP was recorded extradurally by the posterior occipital approach. Isoproterenol (Iso), epinephrine (Epi), and norepinephrine (NE) were administered intravenously for 4 min using an infusion pump. Catecholamines reversibly depressed the ESP with the rank order of potency Iso greater than Epi greater than NE. The threshold doses were 0.021, 0.27, and 1.94 micrograms/kg for Iso, Epi, and NE, respectively. beta-Antagonist (propranolol) blocked the action of catecholamines on the ESP but alpha-antagonist (phentolamine) did not. This indicates that beta-adrenergic action induces the ESP change. This study implies that catecholamines may regulate the function of the ES.

The effect of canrenoate, an aldosterone antagonist, on the ampullar endolymphatic potential (AEP) was examined to assess a possible role for aldosterone in the ampulla. Intravenous administration of canrenoate increased the AEP amplitude in a dose-dependent manner, with a significant reduction of the AEP negative component induced by anoxia. Pretreatment with aldosterone attenuated the AEP change produced by canrenoate. The results suggest that aldosterone may be involved in the modulation of ampullar function.

The aim of this retrospective study was to investigate a correlation between histopathological differentiation and metastasis in patients with carcinoma of the nasal cavity and paranasal sinuses to improve the survival. Between 1984 and 1992, 52 cases of sino-nasal malignant tumors were treated at Kagawa Medical School Hospital. Squamous cell carcinoma was evaluated in 29 patients and undifferentiated carcinoma in five patients. Treatment in most cases consisted of radiotherapy, 50 Gy/25 fr/5 weeks, continuous intra-arterial infusion of 5-FU 2000 mg/4 weeks, and an extensive Denker's operation 2-3 weeks after the completion of radiotherapy. The 5-year survival rate was 53%, and the local control rate was 82%. Patients died of local failure (six), distant metastases (six), lymph node metastases (one) and other causes (three). Twenty-eight patients with local control were classified into G1-2 (well- and moderately-differentiated) and G3-4 (poorly-differentiated and undifferentiated) groups. Lymph node metastasis was unrelated to the degree of differentiation, but distant metastasis was significantly correlated to the degree of differentiation (P = 0.014). The 5-year survival rates in lymph node metastasis (n = 7), non-lymph node metastasis (n = 21), distant metastasis (n = 7), non-distant metastasis (n = 21), and the G1-2 (n = 17) group and the G3-4 group (n = 11) with local control were 57%, 63%, 14%, 78% (P = 0.002), 72% and 44% (P = 0.0384), respectively. The results of combination therapy were poor in patients with poorly-differentiated or undifferentiated carcinoma because of distant metastasis, and adjuvant chemotherapy may be necessary in such patients to prevent distant metastasis.

Expression of mineralocorticoid type I receptor (MR) gene in the rat cochlea was determined using molecular biological techniques. We synthesized complementary DNA (cDNA) from rat cochlear total RNA and then amplified MR cDNA fragments by polymerase chain reaction (PCR). The amplified cDNA fragments were subcloned into an expression vector and the nucleotide sequence was analyzed to confirm the expression of mRNA encoding MR in the cochlea. We then synthesized digoxigenin-labeled riboprobes with this cloned DNA template and examined the localization of MR mRNA in the cochlea by in situ hybridization. The amino acid sequence of MR cDNA expressed in the cochlea was identical to that of the MR first cloned in the rat hippocampus. In situ hybridization showed the expression of MR mRNA in marginal cells of the stria vascularis, suggesting that aldosterone may regulate microhomeostasis of the endolymph, presumably by modulating Na, K-ATPase activity. Intense MR signal was also identified in spiral ganglion cells, the function of which remains to be determined.

Nine oral and oropharyngeal squamous cell carcinomas were removed with a contact Nd-YAG laser. Clinical and morphological studies on wound healing and effects of the Nd-YAG laser were performed. The wounds were not sutured and patients did not complain of any severe symptoms. The wound healing was prolonged to 3–6 weeks. Cells and glands degenerated at a depth of 2–3 mm by the direct effects of the laser. The epithelium exhibited acantholysis 5 mm from the site of laser injury. Scanning electron microscopy showed no coagulation or other coated material in the wound created by a surgical knife, while the wound surface by laser consisted of carbonization, well-coated and non-coated areas. These findings varied with the laser power. The coagulation and thermal effects of the Nd-YAG laser resulted in autolysis of tumor cells at least within 5 mm of the cutting surface. The contact Nd-YAG laser was useful for the removal of the malignant tumor in the oral cavity and oropharynx where it is difficult to maintain a generous safety margin.

A constant volume of horseradish peroxidase (HRP) was injected directly into the endolymphatic sac (ES) lumen of the guinea pig to investigate the detailed absorption activity and the barrier properties of the ES. The reaction products were analyzed using an ultrastructural and morphometric method 1 to 10 h after the injection of this tracer. The epithelial cells in the proximal portion did not absorb the intraluminal HRP at any intervals after the tracer injection. The epithelial cells of the intermediate portion were classified clearly into two types according to their absorption activity: active-absorptive, and non-active cells. Uptake in the active-absorptive cells reached its maximal rate 8 h after the injection and then decreased. The active-absorptive cells in teh intermediate portion are considered to play a major role in the macromolecular absorption in the ES epithelium. The non-active cells scarcely absorbed the intraluminal HRP, suggesting that these cells are not involved in the macromolecular absorption. The absorption activity in the distal portion was lower than that in the active-absorptive cells in the intermediate portion 1 to 8 h after the HRP injection, while higher 10 h after the injection. Not only the intermediate portion but also the distal portion may play an active role in the macromolecular absorption. In no portion of the ES did intraluminal HRP penetrate beyond the junctional complexes between epithelial cells or through the cytoplasm. It is conceivable that there is a tight barrier to the intraluminal macromolecules in the epithelial linings of the ES.

The localization of carbonic anhydrase (CA) in mouse and guinea pig endolymphatic sac (ES) was determined by immunohistochemistry. In the distal and intermediate portions of the ES of both species, epithelial cells reacted with anti-CA antiserum. All epithelial cells in the distal portion stained for CA. In the intermediate portion, immunoreactivity was prominent in about half the epithelial cells, but less intense in the remaining epithelial cells. The proximal portion of the ES lacked immunoreactivity. CA may play a crucial role in ES electrolyte transport.

The expression of epidermal growth factor receptor (EGFR) was determined immunohistochemically in two groups of maxillary sinus squamous cell carcinomas, one with recurrences at the primary site after combination therapy with radiotherapy, chemotherapy and/or surgery and one without local recurrences. Using a four-graded scale (-, +, + +, + + +), 9 of the 10 recurrent carcinomas had a staining intensity and proportion of stained cells of + + or more. A comparable staining intensity was found in 9 of the 18 non-recurrent carcinomas. This difference is statistically significant (Fisher's exact probability test, P0.05). These results indicate that an increased expression of EGFR may influence the recurrence rate of squamous cell carcinoma of the maxillary sinus after combined therapy.

The effect of acetazolamide known as a potent carbonic anhydrase inhibitor of the endolymphatic sac DC potential (ESP) was examined in the guinea pig. Intravenous administration of acetazolamide produced a dose-dependent decrease in ESP amplitude at doses of less than 10 mg/kg. the effect of acetazolamide on ESP was saturated at doses exceeding 10 mg/kg. the maximum reduction of ESP induced by acetazolamide was approximately 50% of the original ESP amplitude. the results suggest that carbonic anhydrase is involved in the generation of ESP.

The ultrastructure of the endolymphatic sac (ES) in the mouse was examined by light and electron microscopy. This organ was divided into three parts: proximal, intermediate and distal. In the proximal portion of the ES, the epithelium consisted of thin squamous cells. The epithelial cells had acquired basolateral processes, numerous small vesicles and well-developed Golgi apparatus. In the intermediate portion, the epithelium consisted of columnar or cuboidal cells. The epithelial cells could be classified into two types: type I and type II. The type I cells had abundant microvilli, pinocytotic vesicles, vacuoles, multivesicular bodies, lysosomes and mitochondria. The type II cells had fewer numbers of these organelles. A few free-floating cells could be observed in the lumen of this intermediate portion, most of which were macrophages. In the distal portion, the epithelium consisted of squamous or cuboidal cells. The epithelial cells had a few cytoplasmic organelles. In the ultrastructural study, each portion of the mouse ES was found to have a very distinct morphological feature. It was suggested that each of these three portions has a different function.

The effect of aldosterone and canrenoate, known as an aldosterone antagonist, on the DC potential in the endolymphatic sac (ESP) was examined in the guinea pig. Intravenous administration of aldosterone (1 mg/kg) induced no change in the ESP for 60 minutes after the injection. However, canrenoate, an aldosterone antagonist, produced a dose-dependent decrease in the ESP. Pretreatment with aldosterone attenuated the decrease in the ESP by canrenoate. The results suggest the possibility that aldosterone may play a role in the endolymph absorption of the endolymphatic sac.

Symptoms at first onset and on initial examination were analyzed in 845 patients with maxillary sinus carcinoma. About 20–30 per cent of the patients became aware of swelling of cheek, nasal obstruction, nasal discharge and epistaxis as first symptoms. About 50–80 per cent of the patients complained of such symptoms on initial examination. Symptoms of maxillary sinus carcinoma are similar to those of chronic sinusitis which results in a delay in diagnosis. Such symptoms should be recognized to improve survival rate. CT and MRI scans are useful for early diagnosis and determination of tumour extension. Sinoscopy with biopsy is most important to detect the earliest tumour stage.

We investigated the effects of furosemide, a loop diuretic, and canrenoate, an aldosterone antagonist, on the endocochlear potential (EP) and the endolymphatic sac potential (ESP) in the guinea pig. Furosemide produced no significant change in the ESP at a dose of 100 mg/kg after an intravenous infusion for 20 min. However, this dose decreased the EP to a negative level. Canrenoate produced no significant change in the EP at an intravenous dose of 300 mg/kg for 20 min, but it did decrease the ESP. The differences in the EP and ESP in the repsonse to the diuretics indicate a dissimilarity of the origin of both d.c. potentials in the endolymphatic space.

In order to examine a blood-labyrinth barrier in the vestibular region, capillary permeability of the vestibular organs, i.e. the sacculus, the utriculus and the ampullae of semicircular canals, were studied light and electron microscopically by the horseradish peroxidase (HRP) tracer method. Intravenously infused HRP was not observed outside capillaries because of the tightness of endothelial tight junctions and the paucity of pinocytotic vesicles. Furthermore, the effect of drug-induced acute hypertension and acute hypotension on the capillaries of the vestibular organs was studied morphologically using the tracer technique. In each experiment, there was no extravasation of HRP from capillaries, thus agreeing with the results of the experiment made under normal conditions. These findings mean that capillaries of the vestibular region are of ordinary brain capillary type and contribute to the blood-labyrinth barrier to macromolecules such as HRP both under normal conditions and following abrupt changes in blood pressure.

Five patients with cerebellar infarction consulted one of us (Otolaryngologist) at an early stage of the disease. Dizziness or gait disturbance was the chief complaint in four of the five. Cerebellar signs were seen in three, but only for a short period. Muscular weakness and unequal tendon reflexes were noted in only two patients and hypesthesia in one. Thus, symptoms were similar to those of patients with inner ear lesions. Neuro-otologic examination revealed bilateral gaze nystagmus in all five patients for one to two weeks. Positional and paroxysmal nystagmus tests showed down beat nystagmus in two patients and apogeotropic direction changing positional nystagmus in two. No hearing disturbance occurred in any of them. Brain CT examination should be performed if any of the above-mentioned signs are present in dizzy patients.

The complement receptor (CR1) in the maxillary sinus mucosa of normal patients and in cases of chronic sinusitis was studied with the peroxidase-antiperoxidase, avidin-biotin peroxidase and immunofluorescent methods. CR1 was localized on the ciliary surface and in the cytoplasm of the covering epithelium in both normal controls and the cases of chronic sinusitis. CR1 tended to be denser in the mucosa of chronic sinusitis than in normal mucosa. CR1-binding capacity was also studied with the immunofluorescent method, using C3b-conjugated zymosan. Although CR1 did not bind to C3b in vivo, it was found to bind to C3b in the normal maxillary mucosa when it was treated with C3b-conjugated zymosan. CR1-binding capacity could not be detected in the mucosa from cases with chronic sinusitis, indicating that CR1 was already bound to activated C3b in these cases.

Histopathological findings of chronically inflamed maxillary sinus mucous membrane were treated using grading codes for computer processing. The criteria prepared for this purpose were as follows: 1) thickness of mucous membrane, 2) infiltration of inflammatory cells, 3) infiltration of eosinophils, 4) edema, 5) polypoid change, 6) fibrosis, 7) vascularization, 8) numerical presence of proper mucosal glands, 9) increase in goblet cells, 10) formation of pseudoglands, 11) sloughing of epithelium, 12) stratification of epithelial cells, 13) thickening of basement membrane, 14) squamous metaplasia, and 15) dysplasia of epithelium.In each criterion, the grade of pathological change was coded in the range of 1 to 4.The mucous membranes studied in this investigation were taken from 923 maxillary sinuses of 582 patients. Correlation coefficients for all possible pairs of the 14 histopathological criteria were calculated and factor analysis was applied. From these results, a new system of classification was proposed in relation to the histopathological findings. It consisted of four main types, i.e., infectious type, edematous type, fibrous type, and mild type. Two subtypes of secretory type and non-secretory type were also suggested. Furthermore, allergic reaction was found by factor analysis as an independent factor.The correlations of the histopathological type with the age of the patient, clinical findings, macroscopic findings obtained at the time of operation and in the postoperative course were also investigated.

A case of malignant fibrous histiocytoma of the temporomandibular joint in a 21-year-old Japanese female was presented. The patient died of the tumor with local aggressive growth but without any organ metastasis, despite the intensive treatment including radical operation of the tumor, irradiation, and chemotherapy. This case suggested that radical curative operation was the only method of the treatment for this tumor. The clinicopathological features of the tumor are discussed, making reference to the malignant fibrous histiocytoma.

Therapeutic results of 130 primary cases of maxillary sinus carcinoma from1967 to 1969 and 80 primary cases in 1972 and 1973, were analysed in December 1975.The 2-year and the 5-year crude survival rates for the former were 42% and 30%, and the 2-year crude survival for the latter was 53%. The treatment policy over these periods consisted of combined irradiation and intraarterial infusion given as the first step and maxillectomy performed in only recurrent cases. Therefore, a numerical comparison could be made between survival rate in the initial combined therapy alone and that in the following maxillectomy.The survival rates of both groups from 1967 to 1969 were almost equal, but, in 1972 and 1973, the therapeutic weight of the initial combined therapy became much heavier than that of the following operation. A randomized study, carried out from 1967 to 1969, concerning 25 cases treated by intraarterial infusion of 5-FU alone, 35 cases by intraarterial infusion with other combined drugs and 45 cases by irradiation alone, resulted in the significantly better survival rate and the local nonrecurrence rate of the 5-FU intraarterial infusion group than that of the other groups.In a trial undertaken in 1972 and 1973, 63 cases, 70% of the total cases, were treated by the combination of intraarterial infusion of 5-FU and irradiation with a total irradiation dose of 5000 rads in 10 fractions during 5 weeks or 3500 rads in 7 fractions during 3.5 weeks.There was no significant difference of the rates concerned between the two groups. At the same time, therapeutic results of the group in which tumor tissue was removed from the maxillary sinus by curettage were compared with those of group treated without curettage and no significant difference crude survival rates was found between the two groups.However, the 2-year non-recurrence rate by initial treatment combined with curettage was significantly better than that without curettage.This could be evaluated as a worth treatment for preservation of both the figure and the function.On the other hand, this mode of treatment left us such problems as difficulty in following maxillectomy and increase in occurrence of distant metastases. The authors concluded that in the series of trials for the principle treatment policy of maxillary sinus carcinoma, the best results were obtained by initial treatment by 5000 rads irradiation of Co-60 gamma-ray in 25 fractions during 5 weeks combined with continuous intraarterial infusion of 5-FU in the amount of 2000mg during 4 weeks, and antrostomy in all cases and curettage, if necessary.The following treatment composed of maxillectomy as a rule and the intracavital or external irradiation additionally depending upon the situation

Six patients of olfactory neuroblastoma were treated, and two of them were studied with electron microscopy and immunohistochemical staining. These patients, four male and two female, ranged in age from 37 to 73 years old. Five patients had nasal bleeding, and only one patient had frontal pain and optic disorder without nasal bleeding. None of the patients had metastasis. However, three patients had local recurrence and invasion into the intracranial region. The tumors of three patients were confined to the nasal cavity and one or more paranasal sinuses. The tumors of the other three patients extended beyond the nasal cavity and paranasal sinuses. Two patients, treated with radiotherapy and cryosurgery, died of the disease 7 and 21 years after treatment, respectively. Two patients, treated with radiotherapy alone or a combination of radiotherapy and chemotherapy, died of other causes without recurrence 2 and 3 years later, respectively. A patient treated with craniofacial resection alone is still alive after 4 years. However, he has local recurrence and has been treated with radiotherapy and adjuvant chemotherapy. The last patient, treated with craniofacial resection following radiotherapy, is still alive after 3 years without any symptoms of recurrence. The two most recent cases were studied with electron microscopy, stains for PAS, Grimelius and Bodian, and immunohistochemical staining for S-100 protein, neuron-specific enolase, and glial fibrillary acidic protein. The tumor cells of one case were well differentiated because they contained neurosecretory granules and neurites, and the tumor cells of the other case were poorly differentiated because they contained only neurites-like structures.

A case of malignant fibrous histiocytoma (MFH) in the right maxillary sinus is reported. Patient was a 30-year-old male, and was diagnosed as pleomorphic type of MFH by biopsy before therapy. The tumor shrunk and changed into xanthoma-like tumor after radiotherapy. A case like the present case has not been reported previously in the literature.