8
Benign Proliferative Reactions, Intraepithelial Neoplasia, and Invasive cancer of the Uterine cervix
optimal, the consulting cytopathologist has an obligation to
inform the referring physician of such a conclusion and the rea-
sons for it. A significantly lower number of epithelial abnormali-
ties were diagnosed in smears without endocervical cells present
than in smears that did show endocervical cells.70,72,128,258
In a cohort study, we showed that the absence of endocervical
cells could not be ascribed to a physiologic phenomenon but
was a result of inadequate sampling of the uterine cervix.72 It is
generally assumed that obtaining a cervical smear is an easily
executed, clinically simple procedure. This is not true. In our
experience, well-trained paramedical sample-takers perform far
better than most physicians who are relatively untrained.26 Tak-
ing duplicate smears could possibly reduce the error, but this is
a very costly procedure when routinely practiced in large-scale
screening programs.279
Many variables influence the detection rate of abnormali-
ties in cervical smears, including the sampling technique used,
the preparation of the patient, the fixation and staining of the
smears, the accuracy of the screening, the daily workload, and
the interpretation of morphologic changes. All smears should
be screened according to a strict laboratory protocol. We have
adopted the procedure that smears with diagnoses of even
minor abnormalities made by a cytotechnologist with short-
term experience are always checked by a cytotechnologist with
long-term experience. A protocol for multiple screening by more
than one observer of specimens at high risk for a false diagnosis
should also be instituted. Included in this high-risk category are
smears from women in postmenopause, women with clinical
symptoms of irregular bleeding, and women with a previous
history of cancer.
Smears of unsatisfactory quality for cytologic diagnosis
should also be screened by more than one observer, because in
cases of severe epithelial lesions the proportion of unsatisfactory
smears is significantly higher.122,124,276
As a procedure for quality control, it is often advocated that
10% of smears initially judged to be negative be rescreened.
Considering that 10% of smears at most show an abnormal-
ity, rescreening normal specimens is a very unrewarding and
time-consuming procedure. It may take several years before an
unreliable observer is detected.118,124,128,184,271,272,280 Much more
effective in reducing the number of screening errors are quality-
promoting procedures such as good supervision of daily work-
load to avoid too large a volume to be screened per working
day; a screening protocol for multiple screening of selected spec-
imens, which because of anamnestic, clinical, or cytomorpho-
logic reasons are thought to be at high risk for a false diagnosis;
and good supervision and training of both cytotechnologists
and cytopathologists.121,122,124,128,276
The cytologic report should contain properly worded recom-
mendations for further patient evaluation.21,74,75,122,124,128 To be
able to provide proper recommendations, all clinical informa-
tion is essential to a cytopathologist. A clinician should take the
advice of a cytopathologist as a strong recommendation but
should institute a different treatment strategy when intercurrent
clinical information or special circumstances warrant this.124
Review
A useful procedure for quality control is a review of previous
cytologic and histologic material.124 The degree of abnormal-
ity may differ between the cytologic sample and the subsequ-
ently taken biopsy. Nevertheless, it is usually very possible to
determine the causes of discrepancies. Because cytologic diag-
nosis is highly dependent on the observer's ability to refer the
cellular and nuclear changes to a complex histologic architec-
ture of the lesion, comparison between cytologic and histo-
logic characteristic is very instructive. With Koss and Hicklin,
we consider a comparison of cytologic findings with histo-
logic findings and vice versa and with long-term follow-up of
patients one of the best methods of quality control.281
Inadequate sampling of the cervix can account for about
half of the false-negative cytologic diagnoses. Taking smears
should be avoided during the first 4 days of the menstrual
cycle. In women using oral contraceptives, the number of inad-
equate or less reliable smears is significantly increased. In all
instances, the best collection time is from day 5 through day 14
of the menstrual cycle.26,282,283 The best smears submitted to our
laboratory were taken with a relatively recently developed
ectocervical brush (Cervex Brush; Rovers B.V., Oss, the Neth-
erlands). The percentage of smears containing endocervical
columnar cells was higher, and the number of unsatisfactory
smears was significantly lower than those of spatula-made
smears151,275 and combined spatula and endocervical brush
smears.284
Concluding Remarks
Worldwide, cancer of the cervix is the second most common
malignancy in women. In developing countries, this type of can-
cer even ranks first in frequency, whereas in developed countries
it ranks as high as tenth. In most developed countries, the fre-
quency and the mortality rate for cervical cancer have declined
since the 1950s. Cervical smears have played an important role
in its prevention.
Many epidemiological and clinicopathological studies have
demonstrated that HPV is associated with the development of
intraepithelial and invasive cervical lesions, and the mecha-
nisms by which it affects the carcinogenic sequence are being
carefully elucidated.
Most reactions occurring in the cervix can be recognized on
the basis of cellular changes alone. The number of cells diagnos-
tic for a specific, benign proliferative reaction depends on the
extent of the epithelial change and the method used for collect-
ing the cells.
Carcinogenic changes in the uterine cervix are considered to
be part of a field effect neoplastic process. This means that the
carcinogenic stimulus does not exert its action on an isolated
cell that becomes the stem cell of a malignant proliferation, but
actions occur on a larger area of the epithelium. This may be the
explanation for the common observation that next to a severe
lesion, changes of different, often lesser degrees of dysplasia are
found. It also explains the common multifocal occurrence of
epithelial abnormalities.
The evolution of invasive squamous cell cancer involves a
number of stages with increasing intraepithelial abnormality
designated as dysplasia, carcinoma in situ, and microinvasive
carcinoma.
The morphologic variations that can be observed when
studying the spectrum of lesions directly associated with car-
cinoma in situ suggest a common pathway in the develop-
ment of dysplasia and carcinoma in situ. The basic factor in
the mechanism of the genesis of carcinoma in situ is reserve
cell hyperplasia.
205
previous page 206 ComprehensiveCytopathology 1104p 2008 read online next page 208 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off