Glandular Lesions of the Uterine Cervix
Fig. 9.22 Examples of cases interpreted as "atypical endocervical cells, favor neoplasia" show the spectrum of changes that can be associated with
reactive and/or neoplastic follow-up. (A) A hyperchromatic crowded group of columnar cells with a ‘hint” of feathering at the margin. The nuclei are large,
show irregular contours, and coarse chromatin. This case showed AIS on follow-up. (B) A honeycombed, flat group with prominent nuclear size and shape
pleomorphism. Prominent nucleoli are present in each cell. This case was found to be AIS on histologic examination. (C) A feathered, densely cellular group,
with isolated bizarre enlarged nuclei. Follow-up showed only tubal metaplasia. (D) Low numbers of cells, some of which were arranged in mini-rosettes as
seen here. The nuclei show coarsely granular hyperchromasia and the background contains a clinging granular diathesis. Histologic examination revealed an
invasive endocervical adenocarcinoma (A, B, D, liquid-based preparation; C, conventional smear; Papanicolaou x (A, B, D) HP, (C) MP)).
As with squamous atypia, not all of these criteria need exist to
arrive at an interpretation of glandular atypia, and other features
may come into play as will be illustrated. Suffice it to say that
atypical endocervical cells is a category utilized by the cytologist
when features that lead to an increased, but not absolute, sus-
picion of a neoplastic process are present. No one feature, cell,
or cell grouping may account for a final interpretation of AEC.
As in squamous atypias, designations of glandular atypia rep-
resent "composite" interpretations derived from a cumulative
summing of otherwise non-diagnostic features gleaned from the
overall slide examination.
There are a number of key indicators that must be appreci-
ated when considering the broad applicability of endocervical
atypia designations. First is the prevalence of interpretations
that fall into this category. In a recent review of 22 studies
reporting the prevalence of AGC after the widespread adoption
of TBS 1991, Chhieng and Cangiarella noted that the range of
prevalence was 0.11 to 2.1% with a mean and median of 0.48
and 0.27%, respectively.62 Since the widespread introduction of
TBS 2001, studies of AGC have consistently shown prevalence
rates to be less than 0.5%, and often less than 0.2%, indicat-
ing generally falling or plateauing AGC rates.63-69 Interestingly,
these reports come from all over the world, indicating, not only
the utility of a widely recognized terminology and classification
scheme (TBS), but also the improvement in specificity that may
have accrued from the refinements made to the classifications
and criteria for AGC in TBS 2001. Compared with a prevalence
range of approximately 2-9% for ASC-US (College of American
Pathologists survey figures for 10th to 90th percentiles),70 AGC is
a relatively infrequent interpretation in most laboratories. Mon-
itoring of this prevalence rate is an essential quality assurance
exercise, predominantly to check for overusage of this category if
rates are above 0.5% (without identifiable mitigating factors).
The second important factor to consider with AGC is the
follow-up expected upon rendering this interpretation. Again,
from Chhieng and Cangiarella's review article, the likelihood of
previous page 230 ComprehensiveCytopathology 1104p 2008 read online next page 232 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off