PART ONE
General Cytology
Fig. 5.14
True tissue fragment of
papillary urothelial carcinoma
in
Fig. 5.16 Warthin's tumor
in parotid FNA. Cell block (H&E x Mp).
voided urine. Surepath (papanicolaou x Mp).
Fig. 5.15 Metastatic ovarian carcinoma
in peritoneal fluid showing clean
background and cells with good nuclear detail. Cytospin (papanicolaou x Mp).
urine, pleural fluid, pancreatic/biliary tract, soft tissue, breast, and
thyroid FNA. The recommendations of the various authors differ
somewhat in favoring one processing technique over the other. A
thorough review of the relevant literature, however, leads one to
the conclusion that direct smears, cytocentrifugation, and filtra-
tion techniques are worthy of routine use with comparable diag-
nostic and cost parameters for most nongynecologic specimens. In
the end the decision to utilize one or more of the processing tech-
niques depends upon weighting the service demands against the
financial and labor resources available in a particular laboratory.
The following highlights some observations in the literature
concerning these processing modalities:
Cellular yield
was found to be superior by the ThinPrep method
which retained three times as many cells as cytocentrifugation.39
There were no statistically significant differences in unsatisfac-
tory rates, sensitivity, specificity, or positive predictive value in
both FNA and body cavity fluid specimens processed by Thin-
Prep and direct smears methods.40
Overall technical
quality was
reported to be improved by ThinPrep processing when compared
Fig. 5.17 Metastatic lung adenocarcinoma
in pleural fluid. Cell block
(TTF1 immunostain positivity x Mp).
to direct smears on split FNA specimens due to cleaner back-
ground and better monolayer formation.41,42 Some authors,
however, advised caution to avoid diagnostic errors when inter-
preting ThinPrep slides due to the increased incidence of follow-
ing
cytologic artifacts
(Table 5.1) : disruption of tissue fragments,
formation of cell clusters, aggregation of lymphocytes, cellu-
lar shrinkage, attenuation of nuclear details, and exaggeration
nucleolar prominence.42 In comparison to SurePath processed
specimens ThinPrep slides demonstrated increased cellular
shrinkage, flattening, and fragmentation of large cellular sheets
and nuclear chromatin patterns were reportedly more difficult
to evaluate.43 SurePath slides were also found to contain larger
branched three-dimensional tissue fragments.43
Separate studies involving FNAs of thyroid nodules, breast
and salivary gland lesions, and pancreatic and soft-tissue tumors
reported somewhat conflicting results in terms of unsatisfactory
rates, quality of nuclear details, diagnostic accuracy, and rela-
tive prevalence of artifacts when ThinPrep-processed slides were
compared to direct smears.44-53 Among the artifacts attributed
70
previous page 74 ComprehensiveCytopathology 1104p 2008 read online next page 76 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off