Sarcoidosis: histopathological definition and clinical diagnosis.

Sarcoidosis is best defined in histopathological terms as 'a disease characterised by the presence in all of several affected organs and tissues of non-caseating epithelioid-cell granulomas, proceeding either to resolution or to conversion into hyaline connective tissue'. Although the defining characteristics are thus histopathological, diagnosis during life depends largely upon clinical, radiological, and immunological findings. The amount of support required from histology varies greatly from case to case. Though histology from one site cannot in itself establish the diagnosis of sarcoidosis, a generalised disease, detailed histological study of biopsy tissue makes an important and often essential contribution. In many instances, complete lack of necrosis, an intact reticulin pattern, and failure to demonstrate infective agents permit an unequivocal statement of compatibility with this diagnosis; however, a compatible clinical picture and absence of evidence of known causes of local granulomatous reactions or of other generalised granulomatous diseases are required for definitive diagnosis. In some, the histological pattern deviates in some particular from the accepted 'typical' pattern; there may be a little necrosis, the follicular pattern of the granuloma may be less than perfect, and exclusion of known infective agents can never be absolute. In such instances, subsequent surveillance, including possible response to treatment, may show a clinical course justifying a diagnosis of sarcoidosis, and necropsy may establish it; but it must be recognised that in a few cases, particularly those in which the clinical evidence of disease is confined to one organ, diagnosis is likely to remain in doubt for long periods. Reports on the histology of the Kveim test should be made without knowledge of clinical findings and in terms of the presence and quality of granulomatous response. A granulomatous reaction to a validated test suspension makes a contribution to diagnosis similar to the finding of granulomas in an additional organ or tissue.

Survey of the sputum cytology service in England and Wales.

Of the 231 laboratories carrying out sputum cytology in England and Wales, information on the staffing, laboratory procedures, workload, records, and results for the year 1971 was obtained from 228 (98.7%) by means of a questionnaire, which was completed in 1973 in 81% of the laboratories and at the end of 1972 in the remainder. At the time the questionnaire was completed, the number of laboratories per region offering a service ranged from 6 to 25. Most of the processing of specimens was done by technicians and non-medical screeners (technicians without recognized qualifications) and most of the microscopy by pathologists and technicians. Papanicolaou and/or haematoxylin and eosin stains were used in 97% of the laboratories. During 1971, the number of specimens tested per laboratory ranged from 9 to 6000. A median of 381 specimens and 165 patients were tested per laboratory. Almost all specimens were submitted by hospitals and chest clinics and only 1% by general practitioners. An estimated median of 6.5% of all cytological specimens were sputum, most of the remainder being cervical smears. Laboratories varied greatly in their positive (0%-24%) and doubtful positive (0%-27%) rates. An estimated 63 000 patients had their sputum tested in 1971, and about 6300 had positive reports, rather less than one-quarter of all patients diagnosed as having lung cancer in England and Wales in 1971. The cost of the service was estimated to be of the order of pound375 000 at 1974 prices; that is, pound2.50 per specimen or pound6.00 per patient tested. An improved service might best be provided by district general hospital laboratories testing at least 1000 specimens per year so that a high level of accuracy in reporting could be assurred. A large proportion of the microscopy might be done by non-medical graduates and technicians under the supervision of pathologists.

An assessment of the World Health Organization classification of the histologic typing of lung tumors applied to biopsy and resected material.

In a study of the World Health Organization classification of the histologic typing of lung tumors, sections from a total of 740 patients in the Medical Research Council Study of Cytotoxic Chemotherapy, 182 of whom also had positive preoperativebronchial biopsies and 231 involved lymph nodes in the resected specimens, and from 30 patients in the Medical Research Council Trial of Surgery and Radiotherapy in Small or Oat-celled Carcinoma of the Bronchus have been assessed. Of the 740 primary tumors from the Study of Cytotoxic Chemotherapy, 71% were placed in Type I, 12% in Type II, 9% in Type III, and 7% in Type IV. Only 2 primary tumors could not be typed. A blind comparison of the type of primary tumor and bronchial biopsy showed that the biopsy was a good indicator of the type of the primary tumor. A bind comparison of the primary tumor and involved lymph node also showed a close degree of agreement. However, when the type of the primary tumor was assessed in the presence of the involved node. Tumors were placed in Type IV far less frequently than when assessed blind. It is concluded that the World Health Organization classification is applicable to primary tumor, bronchial biopsy, and involved node, that the biopsy is a valuable indicator of the type of the primary tumor, and that the apparent type of the involved node should not be allowed to over-influence the pathologist in deciding on the type of the primary tumor when both are assessed together.