Cytology in the diagnosis of retinoblastorna wasrecentl yexperienced in 4 cases. These 4 cases werefemale children and the diagnosis was clinicallysuspected. Specimen was obtained from each caseby aspiration of either aquous humor in the anteriorchamber, vitreous body or subretinal fluid. Cytological observation was made under Papanicolaoustain. Typical cells of retinoblastoma were detectedon smears. These cells were morphologically analyzed in terms of rossette formation, size and formof cells, cytoplasmic size, nuclear form and rim, chromatin distribution and shape of nucleoli. Undifferentiated type showed no rossette formation, small sized cells and cytoplasms, irregular nuclearrims, increased amount of chromatin and increasednumber of nucleoli, whereas differentiated type tended to form rossette with fairly large sized cellsand cytoplasms. Of 4 cases, 1 case was cytologicallyclassified undifferentiated or well differentiated type, respectively, and the rest of 2 cases were classifiedmoderately differentiated type. These findings wereconfirmed by histological findings of the lesionswhich were surgically obtained after the cytologicalexamination.
A new auto-cytoscreener has been developed bythe authors for mass screening of vaginal cytolsm.ears to detect mahgnancies of the uterine cervix.Thisdevice utilizes the size and density of nucleistained by Feulgen's reaction to ascertain rnalignantcells. The histological type of cancer of the cervix, most common in Japan, is squamous cell carcinoma.Hence, enlargement and hyperchromatism of nucleiare characteristics most expediently utilized in aut.omatic cytoanalysis. However, the conventional processing by Feulgen's technique requires more than two hours, and itis very difficult to achieve equal staining reactionsof large numbers of samples. Therefore, the necessity of a less time consurning and rnore expedienttechnique yielding uniform staining reactions instigated the authors to develop anautomation of Feulgen's reaction. By system-programing a novel Feulgen's procedure was devised with each steprequiring a constant and equal amount of time.This constancy of all steps in the procedure facilitates readily the preparation of unlimited numbersof specimens. The staining process by Schiff's reagent could bereduced by changing the HCl concentration in thereagent and the time of incubation. The most optimal condition of staining process was ascertainedto be at 37°C, for ten minutes, and 0.07N HCl forthe Schiff's reagent. The optimal concentration ofHCl for suitable hydrolysis was determined to be2.9 normal at 37°C. Moreover, the entire stainingprocess requires merely forty minutes, and consistedfour equal steps each of ten minutes duaration, namely, hydrolysis, staining, rinse, and clearingfor mounting. Each step was followed by a double rinse fortwenty sec. to forestall contamination, preparingthe srnear for next step. Cervical cyto-srnears prepared by this newly developed autornated modification of Feulgen's staining method were compared with those prepared bythe conventional technique, ernploying on ZeissU.M.S.P. Thirty nuclei of superficial cells were scannedin each stained specimen at the wave length of570n.m.and the integrations of absorption werecornputed. Differences in the quality of stainingbetween specirnens processed by either proceduredid not occur. The shortened modification of the technique proved to either be as efficient and reliable as thelonger conventional procedure.
At the present time, the cytological diagnosis isestablished by the subjective recognition of the morphological differences between the normal and themalignant cells utilizing the optical microscope. However, it is quite difficult with this method to managea large number of specimens both quickly and efficiently. Therefore, the automation of cytological diagnosishas been anxiously sought for by analizing the characteristics of neoplastic cells quantitatively. In orderto materialize the automatic diagnosis, first of all, it is mandatory for us to pick up all the necessarycharacteristics of the malignant cells and arrangethem into really objective criteria. Fluorospectrophotometry will be able to determinethe degree of fluorescence being adsorbed by thenuclear DNA. Of a few obvious characteristics ofmalignant cells which may be obtained quantitatively, the nuclear density, or the DNA content ofthe neucleus, will be determined by the above technique. Thus, the author actually counted the degree of fluorescence concerning with the nuclearDNA of various cells in the total of 134 smear specimens with AO staining by means of the Nikon'sfluoromicroscope, a modified type of SUR-F at thewave-length of 530 mu. The followings are the results. 1) The area of determination was 15 square microns and the time required for each single cell wasless than one minute. The degree of fluorescence atthe background and that of a leucocyte were adjusted as 0 and 10 respectively, then the relative valueof fluorescence was calculated by means of the smallsampling theory: The upper limits of rejection were20.37 for normal cells, 37.63 for dyskaryotic cellsand 65.30 for malignant cells. 2) 40 cases of invasive cancer all revealed thedegree of fluorescence over 20.37 and 38 cases ofwhich (i.e. 95%) disclosed the value above 37.63.10 cases of carcinoma in situ all revealed the fluorescence over 20.37 and 7 cases of which (i.e. 70%) disclosed the value above 37.63. 6 cases of displasiaall revealed the fluorescence over 20.37 and 2 casesof which (i.e. 33.3%) were above 37.63. 3) The author classified those cases with the degree of fluorescence less than 20.37 as negative, cases ranging from 20.37 to 37.63 as suspicious andabove 37.63 as positive respectively. Blind test wascarried out on 50 cases applying the above classification in comparison with the degree of ordinaryPapanicolaou smear. 94.5% of negative cases byfluorospectrophotometry accorded with the negativePapanicolaou smear. Likewise, both suspicious andpositive cases by fluorospectrophotometry exactlyaccorded with those by Papanicolaou smear. 4) By establishing the uppper limits of rejectionas to the degree of fluorescence for nuclear DNA ofvarious cells which appear in the smear specimensobtained from the uterine cervix, the fluorospectrophotometry will be able to differentiate the normal, dyskaryotic and malignant cells quite objectively, thus providing a promising future for the automation of cytological diagnosis.
Contamination of the epithelial cells exfoliatedfrom the oropharynx often makes us misdiagnosein oral cytology. Therefore cytologic and histologicstudies of the superficial area of the borderline between ciliated epithelium and squamous epitheliumwere made in order to overcome such misdiagnosisand to assure accurate and reliable oral and oropharyngeal cytologic diagnosis. Smears of normal exfoliated cells were taken from33 healthy individuals and 8 corpses, while those ofexfoliated malignant tumor cells were taken from 9cases of oropharyngeal malignant neoplasm. Smears were stained with Papanicolaou stain andhistologic sections taken simultaneously were stainedwith hematoxylin-eosin. 1. Smaller squamous cells were exfoliated fromthe epipharyngeal posterior wall and the pharyngealsurface of the soft palate, middle cells from themesopharyngeal posterior wall and the palatine tonsil, and larger cells from the oral surface of the softpalate and the buccal mucosa as observed in controlcases. 2. Cells from oropharynx consisted mainly of thebasophilic middle squamous cells and they were lesskeratinized in comparison with those from the oralmucosa. 3. Exfoliation of many ciliated cells were observed on the smears from the epipharyngeal posteriorwall and the pharyngeal surface of the soft palate, and they were classified morphologically into thefollowing four types: ciliated cell, goblet cell, typeI and type II cells. 4. A borderline between ciliated epithelium andsquamous epithelium of the pharynx was distinctand it was located approximately between epipharynx and mesopharynx. 5. Five cases of squamous cell carcinoma, onecancer (indefinite histologic type), and one reticulum cell sarcoma among 9 cases of oropharyngealmalignant tumor were positive and 2 cases of adenocarcinoma were false negative in cytology. 6. Malignant cells of squamous cell carcinomashowed a decrease of cytoplasmic-nuclear ratio (C/Nratio) in comparison with normal squamous cells. 7. As the result of cytologic observations of malignant tumors of the oropharynx, no specific findingof oropharyngeal carcinoma in cytology could benoticed in a series of the author's cases.
Serial cytological examination of pleural effusionstaken from the cases with primary pulmonary carcinomas; all cases have not be done surgical resection; are studied. After diagnosis are determined, combination chemotherapy is conducted systemicallyand also intrapleuraly. 1, 000 cells of one preparation are observed anddevided to (1) mesothelial cells (2) phagocytic cells (3) lynphocytes (4) neutrophilic leukocytes (5) others. Cell component ratio are compared with clinicalcondition, i.e. appetite, body weight, pain, fever, general felling, X-ray finding, biochemical date ofserum etc.Identification of cells are done by Papanicolaoustaining, Giemsa staining, PAS reaction, Peroxydasereaction, neutral red Janus green supravital staining, acid phosphatase staining, In addition to thesemethod, identification of phagocytic cells (histiocyte, monocyte) are done by means of carbon particlemethod. Conclusions are as follows: (1) In generally (before treatment), if phagocyticcells are found over 50% of all cells of the effusions, these effusion may contain the cancer cells. On the contrary, if lymphocytes are found over50% of all cells of effusions, effusions may containno cancer cell. (2) Cancer cells of the effusions are decreased bychemotherapy, but it does not always mean thegood clinical course. (3) In the period when the effusion has lymphocyteas a prominant cellular component, clinical condition shows good one. In these state, the X-ray levelof effusion does not show any change. (4) linical symptom shows rather critical duringphagocytic cell is prominant in effusion. Rapid increasing of phagocytic cells means the clinical courcemay turn for the worse process. (5) We can find the changes of cellular componentof pleural effusions before the manifestation of clinical symptoms. So, the cellular changes may be calledas a indicator for clinical course. (6) Rate of lymphocyte in pleural effusion does notrelate to the rate of lymphocyte in peripheralblood. (7) A ll cellular findings above mentioned do notrelate pH of pleural effusions.
Evaluation of cytologic effects caused by various.xogenous hormonal preparations has been made on: he endometrium as well as endocervix. This stuliesalso include endometrial and endocervical cellhanges by oral contraceptive pills or injectable progestational agent. Those cytologic investigationaas been associated with determination of B.B.T.and hormonal evaluation such as pregnandiol andstrogen in urine. On the other hand, endocervical changes duringabnormal pregnancy were studied. The results areas follows: (1) Cyclic cellular changes of the endometriumclearly observed on normally menstruating women.Endocervical cell changes are also cyclicallyseen though they are not so noteworthyas those in the endometrium. (2) Estrogen preparations may stimulate the proliferative-like change. Giant or atypical nucleiare occasionally noted. (3) Progestational agents may temporarily causea secretory-like cell pattern. (4) Neither testosterone nor gonadotropin preparationsreveal significant cellular configulation. (5) Endometrium of normally ovulating womenwho were given oral contraceptive pills revealssecretorylike presentation below middle dosagesand atrophic appearance in large and longcontinuingadministration. Similar findings arenoted by the injectable progestational agent.These findings are sooner or later reversible incessation of the administration. (6) Pregnandiol in urine generally diminishes incases of injectable DMPA administered. Levelof estrogen fractions, however, fluctuates onlyprobably within limit of error. (7) Endometrial cell pattern shows conventionalgestational features even in abnormal pregnancy.Mild nuclear atypism may be observed.
Endometrial glycogen and nidation have an importantrelationship. Nidation is known to be quiteimpossible, if the glycogen level of endometrial cellis low, even in the presence of a histologicallynormal endometrium. Glycogen can be demonstrated by the periodic acidschiff reagent (PAS) staining technique. I have examined the amount of glycogen in theendometrial cells by staining endometrial aspirationsmear with this PAS reagent. Functional infertility patients have a normal infertilitywork up with a histologically normal lookingsecretory endometrium. However, by the PASstaining of the endometrial aspiration smears, Ihave demonstrated, a lowering of glycogen, contentat the cellular level in these women. In functional infertility group of 441 cases. Negativeglycogen was present in the early secretoryphase in 32.7% of these cases compared to only11.9% in the controls. Negative glycogen was present in the late secretoryphase in 41.6% of these cases compared to10.8% in the controls. In the 164 cases of anovalatory infertility glycogenwas negative in 131 cases (79.8%). This lowering of glycogen is a key point in functionalinfertility making this PAS staining techniquean important test. It can be carried out routinelyand repeatedly with great simplicity.
The cytological results of 16 severe dysplasias, 58 in situ carcinomas, and 23 minimal invasive carcinomaswere presented in this study. The hystologicaldiagnosis was confirmed by conization specimenor removed uterine cervix. 1) Of 16 severe dysplasias, 4 were positive incytology, 10 were suspicious, and 2 were negative.Of 81 in situ or minimal invasive carcinomas, 53 (65.4%) were positive in cytology, 27 (33.3%) weresuspicious, and one was false negative. 2) Biopsy diagnosis of 80 in situ or minimalinvasive carcinomas were as follows 13 were dysplasia, 62 were in situ carcinoma, and 5 were minimalinvasive carcinoma. One in situ carcinomawas found in a removed uterine cervix, that wasnegative in cytology. 16 severe dysplasias weredecided the same diagnosis by biopsy. 3) Incidence of in situ carcinoma associated withdysplasia was 65.5%, and that of minimal invasivecarcinoma in the same relation was 34.8%. 4) Of 67 in situ or minimal invasive carcinomasdecided the same diagnosis by biopsy, 48 (71.6%) were positive in cytology and 19 (28.4%) weresuspicious. As the results of re-examination ofsmears, 11 suspicious smears were corrected to bepositive. Therefore in this group the positive rateof cytology became high incidence of 88.1%. 5) On the other hand, of 13 in situ or minimalinvasive carcinomas diagnosed as dysplasia by biopsy, 5 (38.5%) were positive in cytology and 8 (61.5%) were suspicious. As 4 suspicious smearswere corrected to be positive after re-examination, the positive rate of cytology in this group became69.2%. 6) 4 in situ carcinomas, suspicious in cytologyeven after re-examination and also its biopsy diagnosesbeing the dysplasia, had relatively small areaof in situ lesion. We discussed the significance of suspicious smear (class III) and the value of cytology for detectingearly cancer of the uterine cervix.
Whether the imprint smear cytology is availableor not as a tool of the gynecosurgical pathology wasstudied, and the following results were obtained. 1. By the use of imprint smear technique satisfactorilythin and beautiful specimens can be obtainedeasily in such ablative tissues as lymph nodes. 2. The imprint smear cytology is available ingynecological surgery, especially during the lymphnode dissection. 3. As the staining methods of the imprint smearspecimens, hematoxylin-Eosin and Giemsa are themost recommendable. 4. An erroneous diagnosis was given in the imprintsmear cytology in 6 per cent of the lymphnode specimens, and in 11 per cent of the ovarianand uterine tumors. On the contrary, by the useof frozen section histology an erroneous diagnosiswas given only in two percent of the examinedcases. From these results the author reached a conclusionthat the imprint smear cytology should be usedmore widely in gynecosurgical practice, especiallyin hospitals where not so well-equipped. The frozensection histology, however, seems to be in generalmore recommendable than the imprint smear cytologyas the pathological tool in gynecological surgery.
A case of double cancer of the uterine cervix wasreported, which we diagnosed cytologically with thevaginal smear from the patient of uterine cervicalcancer (Stage IIb). In cytological observation of the vaginal smear, we noticed both a group of well-differentiated malignantcells of squamous origin and a malignant cellsof adenomatous nature, which were independent eachother in their arrangement. We considered that thepresent case was obviously different from adenoacanthoma, since the latter is known to be free ofa well-differentiated malignant cells of squamousorigin and is also known to include mixed arrangementof those different types of malignant cells.The cytological findings of the present case, therefore, indicates a double cancer of uterine cervix. Punch biopsy of the uterine cervix (before operation) as well as pathological examination by serialsectioning of the same tissue (after operation) werecarried out. Those histological studies also provedthe coexistence of two different types of malignantcells; i.e., a part of well-differentiated squamous cellcarcinoma and one of adenocarcinoma. The specimensshowed obvious features of so-called ‘collisions-tumor’ of two different nature of malignant cells, whichnever intermingled, The same findings were again noticed in the metastaticlymph nodes.
The cytologic findings of a hemangioendotheliomaof the uterine cervix are reported. Vaginal and cervical smears were initially diagnosed as suspiciousfor malignancy, then the atypical cells in these specimen were diagnosed as the tumor cell by a induction from observation of imprint and histologicalsections of the cervical hemangioendothelioma. The cytomorphological findings of the tumor cellin vaginal and cervical smears are as follows: 1. Structure of nucleus: Round or oval nucleiwith thin nuclear membranes and small unequalchromatin granules. A few small nucleoli are bluishred. 2. Nuclear size: 12.5-10.0μ×10.0-7.5μ. 3. N/C ratio: Moderately increased. 4. Shape of cell: Polymorphic. Occasionally it ispossible to find a pseudopod-like projection. 5. Appearance and staining reaction of cytoplasm;Thin and greensh-blue.