J. Benezra et al., COMPARISON OF THE CLINICAL MICROSCOPY LABORATORY WITH THE CYTOPATHOLOGY LABORATORY IN THE DETECTION OF MALIGNANT-CELLS IN BODY-FLUIDS, American journal of clinical pathology, 102(4), 1994, pp. 439-442
The clinical microscopy (fluids) laboratory evaluates almost every bod
y fluid that is obtained in the hospital. Because the fluids laborator
y functions at all hours, it is often the first laboratory to receive
a body fluid. In addition to its primary purpose of quantitating categ
ories of cells, the medical technologist in this laboratory has an opp
ortunity to identify malignant cells. To our knowledge, no formal stud
y has ever been undertaken to evaluate the performance of the fluids l
aboratory in detecting malignancy. The authors therefore retrospective
ly identified 998 body fluids (pleural, peritoneal, cerebrospinal, and
miscellaneous) over a 2-year period that had undergone simultaneous c
ytologic examination in our cytopathology laboratory and body fluid an
alysis in our fluids laboratory. Of these, a cytologic diagnosis of ma
lignancy was made by the cytopathology laboratory in 107 cases; 821 we
re considered to be benign. No false-positive results were rendered by
the fluids laboratory (100% specificity), but only 26 of the 107 mali
gnant cases were identified (24% sensitivity); the overall accuracy wa
s 93%. Factors contributing to the inability of the fluids laboratory
to identify malignant cells included (1) too few cells to warrant a cy
tocentrifuge preparation, especially in cerebrospinal fluid specimens;
(2) differences in the processing of specimens; (3) differences in st
aining procedures; and (4) differences in the training of personnel. T
he authors conclude that although the fluids laboratory correctly iden
tifies neoplastic cells in approximately one fourth of the cases in wh
ich they are present, it should not be expected to detect malignant ce
lls in every cytologically malignant case. Unless these contributing f
actors can be addressed and corrected, the fluids laboratory should ne
ver replace the cytopathology laboratory in the evaluation of body flu
ids for the presence of malignant cells.