Ministry of the
Flemish Community
Department of Public Health
Flemish Steering Group on Cervical Cancer Screening
CERVICAL CANCER SCREENING IN THE FLEMISH COMMUNITY
___________________________________________________________________________
A TECHNICAL GUIDELINE:
COLLECTION OF ADEQUATE PAP SMEARS OF THE
UTERINE CERVIX
Translation of the final report of the Working Group“Sampling”
Dactylography and lay-out by:
Veerle Eygenraam
Cathérine Vân Phan
Isabelle Moorkens
Marc Arbyn and Rika De Cock
Flemish Co-ordination cell for Cervical Cancer Screening,
Scientific Institute for Public Health
J. Wytsmanstraat 14
1050 Brussels
2
Ministry of the Flemish Community
Department of Public Health
Flemish Steering Group on Cervical Cancer Screening
CERVICAL CANCER SCREENING IN THE FLEMISH COMMUNITY
___________________________________________________________________________
A TECHNICAL GUIDELINE:
COLLECTION OF ADEQUATE PAP SMEARS OF THE
UTERINE CERVIX
Translation of the final report of the Working Group “Sampling”
Dr. G. Albertyn, Dr. M. Arbyn, Dr. C. Bourgain, Prof. Dr. F. Buntinx, Prof. Dr. M. Dhont,
Prof. Dr. M. Drijkoningen, Dr. P. Neven, Dr. L. Thienpont, Prof. Dr. P. Vandam,
Prof. Dr. H. Van Oyen, Prof. Dr. I. Vergote
Marc Arbyn and Rika De Cock
Flemish Co-ordination cell for Cervical Cancer Screening,
Scientific Institute for Public Health
J. Wytsmanstraat 14
1050 Brussels
S_eng1.doc
19/09/01 -
3
Contents
1. Reports from the Working Group “Sampling”............................................................................ 4
2. Technical guidelines for the collection of a Papanicolaou smear ................................................. 5
3. Accompanying illustrations ....................................................................................................... 14
4. Standard request form................................................................................................................ 14
5. Thin-layer cytology ................................................................................................................... 15
6. References ................................................................................................................................ 16
7. Annexes.................................................................................................................................... 19
1. Accompanying illustrations............................................................................................... 20
2. Quality judgement of a Papanicolaou smear according to The Bethesda System ................. 34
3. Standard request form ....................................................................................................... 36
4. Summary: Collection of an adequate Papanicolaou smear of the uterine cervix................... 37
5. Comparison of the quality of cervical smears obtained with the spatula &
CytobrushR versus the Cervex-BrushR ............................................................................... 42
6. Members of the Sampling Working Group ........................................................................ 51
7. List of used abbreviations.................................................................................................. 53
S_eng1.doc
19/09/01 -
4
1. Reports from the Working Group “Sampling”
The conclusions of the working group will be published in the form of the following outputs:
a) Final report: A technical guideline: Collection of adequate Pap smears of the uterine cervix.
b) Scientific paper (possibly with illustrations): "Technical guideline for the collection of a
Papanicolaou smear", to be submitted for publication in a Belgian medical journal.
c) Booklet: "Uniform guidelines for cervical cancer screening in Flanders:
(1) collection of Pap smears, (2) cytological interpretation, (3) follow up of abnormal results”.
Clear, readable, illustrated booklet summarising the conclusions of the three working groups
(consistency in cervix cytology methods, follow-up, sampling).
To be distributed to all general practitioners and gynaecologists in Flanders (~8000 copies).
The “Working Group Sampling” guidelines are summarised in Appendix 4.
d) Technical record card:
Three boxed laminated cards on specimen collection, interpretation and follow-up.
e) Accompanying colour illustrations: overhead transparencies, slides, and files in a standard
graphic format for electronic transmission.
f) Presentation of the final report as a downloadable Acrobat file on the Scientific Institute for
Public Health's Website.
S_eng1.doc
19/09/01 -
5
2. Technical guidelines for the collection of a Papanicolaou smear
Introduction
In late 1996, the Cervical Cancer Screening Steering Group, a subcommittee of the Flemish Advisory
Board for Cancer Prevention (VACK, “Vlaamse Advies Commissie voor Kankerpreventie”), requested
to set up a working party “the Sampling Working Group” which was to formulate guidelines for the
collection of a Papanicolaou smear. Since the VACK was closed down in 1997, the Steering Group no
longer operates. Therefore, the Sampling Group has been idle for some time now. A new Steering
Group is scheduled to be resurrected in September 1999. The guideline below can be assessed and
validated at this forum.
The diagnostic value of cervical smears for detecting cytological precursors of uterine cervical cancer
is determined by the quality of the sampling technique used. Several experts even claim that
inadequate sampling, more so than cytological interpretation errors, is responsible for false-negative
results [Richart, 1965; Frost, 1969; McGooghan, 1997].
The anatomy and histology of the uterine cervix and more specifically the transition zone between the
squamous and columnar epithelia is described briefly, as understanding of this is essential for the
collection of a representative smear.
In addition, the criteria for quality judgement of a smear are discussed. These are specified using The
Bethesda System for reporting cervical smears [Lundberg, 1989; Luff, 1992; Bethesda, 1993; Kurman,
1994], modified by the Working Group on Uniformisation of Cervical Cytology [WUCC, 1996].
A summary is provided of the factors that affect the interpretation of a smear. Physicians who take
smears should take account of these as far as possible.
Finally, a technical guideline is provided on all the steps of the sampling process:
- preparing the patient and equipment;
- visualising the cervix;
- scraping cellular material from the uterine cervix with appropriate sampling devices;
- transferring them on a slide;
- fixing the specimen;
- completing the standard request form;
- transporting the sample to a cytology laboratory.
The choice of sampling device is discussed in detail. Data in the literature show that the combination
of a spatula and an endocervical brush produce significantly better results than a spatula alone. The
Cervex-BrushR allows a representative smear to be obtained with one single device. In a meta-analysis
we have investigated whether this latter method produces results comparable to those from the
combination of spatula/endocervical brush.
There is also a short explanation of thin-layer cytology, a new sampling method in which the cellular
material is immersed in a container with transport fluid from which uniform thin smears can be
prepared in the laboratory.
This guideline on the collection of a Papanicolaou smear completes a series of earlier published
instructions on uniform cytology reporting and follow-up of screen-positive lesions, which were
formulated by the Working Group on Uniformity in Cervix Cytology [WUCC, 1996] and the Followup Working Group [Arbyn, 1996].
The illustrations in this guideline are made by Herman Vanvinckenroye
S_eng1.doc
19/09/01 -
6
Anatomy and histology of the uterine cervix
See Figures 1a-d, 2a-d and 3 in Annex 1.
The exocervix or portio protrudes into the vagina and is generally covered with multilayer squamous
epithelium. Via the external os, the exocervix internally becomes the narrow endocervical canal. This
canal branches into the endocervical crypts and is lined with monolayer columnar mucus-producing
epithelium. The transition between the two epithelia is called the squamocolumnar junction (SCJ). In
prepuberty, this SCJ is usually situated inside the endocervical canal, close to the ostium externum.
After puberty, as a consequence of a change in shape of the uterine cervix, with use of the
contraceptive pill and still under the influence of the first pregnancy, the endocervical columnar
epithelium everts. This cervical ectopy or ectropion is perfectly physiological during the reproductive
period. It is visible as a red zone (erroneously termed cervical erosion) contrasting with the pinker
squamous portio. The extent of the ectropion varies from person to person [Ferenczy, 1994].
Gradually, this exocervical columnar epithelium is replaced by metaplastic squamous epithelium. Two
junctions can be distinguished: the distal transition between squamous and metaplastic epithelium (the
primary SCJ) and the more proximal junction between metaplastic squamous and columnar epithelium
(the secondary or physiological squamocolumnar junction [Boon, 1993; Ferenczy, 1994]. This area
between the original and the current junction is called the transformation zone (TZ) or transition zone.
The shape of the TZ is often irregular and at first comprises multifocal islands, which gradually go on
to coalesce. Sometimes, there are still residual gland openings visible. Later, the metaplastic squamous
epithelium may mature into fully mature multilayer squamous epithelium. The functional SCJ
migrates inwards with increasing age. In post-menopausal women, this junction is almost always
inside the endocervical canal.
Virtually all squamous cell neoplasias of the cervix arise initially around the physiological SCJ and the
preferential location for dysplasia corresponds with the topographical distribution of the TZ
[Burghardt, 1970; Richart, 1973]. Therefore, locating the transformation zone and sampling its entirety
is essential to obtaining a representative smear.
Therefore, according to several authors, , an adequate smear should contain metaplastic and/or
columnar endocervical cells as well as squamous cells [Elias, 1983; Vooijs, 1985; Boon, 1993;
Kurman, 1994].
Evaluating the quality of a smear
The Bethesda System for reporting smears provides three categories for evaluating the quality of the
smear: “satisfactory for evaluation”, “satisfactory for evaluation but limited by … .” and
“unsatisfactory for evaluation .… ”
The assessment is based on the following criteria:
- presence of sufficient number of cells;
- presence of metaplastic and/or columnar cells as well as squamous cells;
- fixation of the cell material;
- absence of excessive numbers of erythrocytes or inflammatory cells;
- absence of extensive cytolysis of the squamous cells;
- even, thin spread of the cellular material, absence of aggregations of cells;
A smear suitable for cytological evaluation should also be labelled with identifying details. These refer
to a request form that contains adequate clinical information. The specimen must, of course, not have
been irreparably broken.
S_eng1.doc
19/09/01 -
7
Various authors report on the correlation between these quality criteria on the one hand and the
sensitivity and specificity on the other [Elias, 1983; Vooijs, 1985; Mitchell, 1991 and 1992; Editorial
Acta Cytologica, 88; Buntinx, 1992a en 1992b; Boon, 1993; Henry, 1996]. Opinions differ on the
clinical significance of the absence of endocervical cellsa.
With these guidelines, the Working Group Sampling, hopes to help minimizing the percentage of
unsatisfactory smears. (see Annex 2)
Factors that adversely affect smear quality
Table 1 on the next page lists a number of factors that adversely affect the quality of a smear. These
are then discussed and recommendations given to limit the detrimental elements.
Table 1 Factors that adversely affect smear quality
Menstruation, blood loss or breakthrough bleeding
Vaginal inflammation/infection
Severe genital atrophy (menopause)
Pregnancy (and lactation)
Foregoing digital examination
Disinfectant cream or fluid, lubricating jelly
Vaginal medication (less than 48 hours before)
Vaginal douche (less than 24 hours before)
Foregoing colposcopy with acetic acid (less than 24 hours before)
Previous smear (less than 3 months before)
Cervical surgery (less than 3 months before)
Radiotherapy
Effect of menstrual cycle on smear quality
The best time for a smear to be taken is between the seventh and fifteenth day of the menstrual cycle
[Vooijs, 1987]. At this time, there is no longer menstrual blood or cell debris and there is virtually no
cytolysis of squamous cells. However, this does not mean that a doctor should propose a woman to
come back if she presents at a different time in her cycle. The presence of a small amount of blood
does not necessarily have any adverse effect on the quality of the smear, but smears should not be
taken during menstruation itself or when there is significant blood loss or breakthrough bleeding, as
the presence of excess blood obscures a clear view of the epithelial cells causing reduction of the
sensitivity of the screening test.
a
Thompson [1989] considers the finding of squamous metaplasia as an important indicator that a sample has been taken from the
target zone. Moreover, the presence of endocervical columnar cells indicates that cellular material beyond the transformation zone has been
included in the specimen. However, the presence of squamous metaplastic and/or columnar cells as well as squamous cells is no proof that
the entire transformation zone has been sampled [Kurman, 1994; Solomon, 1995].
Recent longitudinal studies do not support the thesis that the absence of endocervical cells is associated with a higher risk of false-negative
results [Kivlahan, 1986; Mitchell, 1991]. Mitchell [1992] was able to confirm this correlation only for the presence of metaplastic, but not
for columnar endocervical cells. Boon [1993] holds with the contention that the presence of immature metaplastic cells and/or columnar
cells is associated with the detection of high grade intraepithelial lesions. If low grade lesions are also included in the definition of a positive
smear, this association disappears.
Boon [1993] also states that the progressive character of squamous lesions increases the closer they are to the metasquamo-columnar
junction and therefore also pleads for a double endocervical and exocervical sample.
The exclusive presence of columnar cells (without pavement cells) indicates a non-representative smear.
S_eng1.doc
19/09/01 -
8
Inflammation/infection
Purulent infections/inflammations that are clinically obvious, should be treated appropriately before a
smear is taken. The presence of large numbers of inflammatory cells inhibits a clear microscopic view
on the epithelial cells. Inflammation or infection can sometimes give rise to atypia, which usually
disappears after treatment [WHO, 1988].
Atrophy
A highly atrophic cervix is best treated with hormones before a smear is taken.
Pregnancy and postpartum
During pregnancy and up to 3 months postpartum, screening smears should not be taken [NHG,
1996]b. The cytology of the cervix is affected by pregnancy and lactation and difficult to interpret
[Ziekenfondsraad, 1993]. Several authors report an increased risk of false-negative results or underinterpretation of epithelial lesions [Hellberg, 1987; Ritter, 1995].
With pregnant women who have not had a previous smear, and for whom the chance of future
participation in screening is low (for example in the case of low socio-economic status), it may be
opportune to take a smear as a precaution at the prenatal or postnatal consultation.
Mechanical manipulation or chemical irritation of the cervix epithelium
A smear should not be taken if vaginal medication has been administered less than 48 hours before, or
less than 24 hours following a vaginal douche.
Diagnostic epithelial cells can be removed or damaged by scraping with a sampling device, by
vaginal digital examination or as a result of application of acetic acid prior to a colposcopic
examination [Griffiths, 1989]. It can take several weeks for such lesions to heal. Any vaginal digital
examination, or colposcopic examination should therefore be done after the smear has been takenc.
A cytological check, indicated in case of ASCUS, L-SIL or unsatisfactory quality judgement should
only be done after a minimum interval of three months, since lesions can be temporarily absent, or
more difficult to detect after a smear, thus increasing the likelihood of false-negative results [FU
working group, 1995]. There is little correspondence between the assessment of two smears taken
from the same woman at an interval of less than 3 months [Meisels, 1990].
In addition, after surgical intervention (biopsy, conisation, partial excision, laser or cryo therapy) one
should wait three months before taking another smear, as reparative changes can give rise to
false-positive results.
Radiotherapy can in itself give rise to fairly extensive cell abnormalities.
Contra-indications for screening smears
Hysterectomy
Screening smears should no longer be taken after total hysterectomyd.
Following total hysterectomy, because of cervical neoplasia, the need for follow-up smears (or other
investigations) is determined on an individual basis.
b
The Follow-up Working Group recommends that screening smears should not be taken from pregnant or lactating women until
6 months after delivery or breast-feeding [Follow-up Working Group, 1995; Arbyn, 1996]. The Sampling Working Group revised this
recommendation and reduced the interval to three months post delivery.
c
A supplementary smear during colposcopic examination has very little diagnostic value and is therefore not recommended
[Beckwith, 1995].
d
Following total hysterectomy no further invitations should be sent.
S_eng1.doc
19/09/01 -
9
Macroscopically suspect lesions
If there are macroscopic suspect abnormalities of the cervical surface (irregular elevated lesion with
ulceration, covered with blood or necrotic material) no smear is taken, but the patient is referred to a
gynaecologist for biopsy under colposcopic examination [NCCLS, 1994; NHG, 1996; NHSCSP,
1998].
Information to the woman
The woman should be given adequate information on how the smear will be taken [Buntinx, 1989],
why it is being taken and the possible consequences of any abnormalities [NHG, 1996]. The advice is
given that one correct smear every three years - in the absence of gynaecological complaints and if
previous results were negative - is almost always sufficient to exclude the appearance of uterine
cervical cancer or to enable prompt detection and treatment.
It is explained that a smear sometimes has to be repeated because of technical problems relating to the
quality of the smear beyond any suspicion of (pre-)cancerous lesions. [NHG, 1996].
It is recommended that the woman be systematically questioned on whether a copy of the results may
be sent to another doctor (her general practitioner, or her attending gynaecologist)e [Flemish Steering
Group on Cervical Cancer Screening, 1996].
A formal arrangement is made concerning when and how the patient will be notified of the result. In
that way, it can be agreed that no message within two weeks means that the smear is normal. The
woman may, if she so wishes, make a telephone call herself to find out the result.
If a repeat smear, other additional examination or referral to a gynaecologist is required, the doctor
himself will do whatever is necessary to contact the patient [NHG, 1996; Follow-up Working Group,
1995].
Preparation of the smear
The examination couch should be equipped for gynaecological examination. A suitable light-source
must be available. The doctor taking the smear should be familiar with gynaecological examinations
and trained in taking cervical smears.
All the equipment should be laid out ready and within hand-reach so that the interval between taking
the smear and fixing it is as short as possible (at most a few seconds) (see Figure 4).
Any digital vaginal examination should be postponed until after the smear has been taken.
e
We cite here four clauses from the consensus reached within the Flemish Steering Group on Cervical Cancer Screening, Discussion
concerning Communication between General Practitioners and Specialists, - Brussels 9 February 1996:
Clause 2: Gynaecologists support the contention that the general practitioner is the manager of the central medical file of his patient. They
also agree that if the screening is done by a gynaecologist, a copy of the protocol should be sent to the general practitioner. Feedback on
diagnosis or treatment of a referred patient is part of good medical practice.
Clause 3: The general practitioner in his turn will notify the gynaecologist on the further (gynaecological) course of any patient who has
been monitored and/or treated in the past.
Clause 6: A section "send copy to Dr. ........" is to be inserted into the request form for cytological investigation that accompanies the smear.
Clause 7: This section will only be filled in if the woman gives her consent. The physician must ask this question systematically.
S_eng1.doc
19/09/01 -
10
Equipment required:
- specula (various sizes, virgo-specula)
- gloves
- microscope slides
- pencil
- forceps with swabs for removing any excess mucus
- atraumatic forceps
- sampling devices: spatula + endocervical brush (CytobrushR) or Cervex-BrushR
- fixative: aerosol spray or alcohol solution.
Slides, sampling devices and aerosol spray are supplied by most cytological laboratories for physicians
who take smears. It is recommended that the top from the aerosol spray be removed and to check that
the spray is neither clogged nor empty.
The microscope slide is labelled on the frosted end with lead pencil with the name or reference number
(see Figure 6).
The speculum is preferably inserted without specific lubricant and the portio cervicis is visualised as
clearly as possibly. For the comfort of the patient, some people recommend that the speculum first be
warmed for example in tepid water or in a warming drawer. Contact between the speculum and the
cervix should be avoided as far as is possible. In the event of extreme retroversion or anteversion of
the uterus, the cervix can be manipulated with closed atraumatic forceps towards the centre to enable a
smear to be taken. The position and aspect of the transformation zone is checked visually (see Figure
7).
If there is excessive mucus, discharge or blood, it should be very carefully removed with a swab .
Number of specimens
One or two microscope slides can be used for smears of exocervical and endocervical mucus [Buntinx,
1989]. The use of one slide leads to less weight and volume for transport and archiving, and to less
surface and time for reading. Some older experiments show that the sensitivity of the Pap test can be
increased by a double smear spread over two slides [Sedlis, 1974; Shulman, 1975; Luthy, 1978;
Beilby, 1982]. However, there is not enough evidence on whether this increase in sensitivity is due to
the effect of a second reading or to the examination of more cell material.
Moreover, these studies report nothing about the value of a double exocervical/endocervical smear
spread over one slide. The reports also show that in routine use, the two smears are not examined with
the same care if two smears are taken from each patient without indication of a special reason for so
doing [A. Hanselaer, personal communication].
The Working Group therefore pleads for the use of one slide only.
Sampling devices
Cervical screening always requires an endocervical and an exocervical specimen taken with the
appropriate sampling devices. Only in that way is there a reasonable chance that the transformation
zone is sampled satisfactorily [Boon, 1993; Buntinx, 1996]. The presence of an ectropion does not in
any way change the requirement for dual sampling [Garite, 1978; Buntinx, 1995].
Endocervical sampling alone, for example just with a CytobrushR, is to be avoided just as much as is
an exocervical smear alone, for example with an Ayre-spatula [Spurett, 1989; Buntinx, 1991].
Comparable detection percentages are found if either a spatula + endocervical brush, the sharp + blunt
S_eng1.doc
19/09/01 -
11
end of a combination spatulaf or a Cervex-BrushR [Buntinx, 1996]. The first two methods result in a
larger number of smears with endocervical cells [Buntinx, 1994], which is considered to be an
indicator of good quality [Elias, 1983; Vooijs, 1985; Boon, 1993]. The good performance of the
combination spatula + endocervical brush depend little on the experience of the smear-taker [Boon,
1985; 1986a; Szarewski, 1990]. The likelihood of detection of glandular lesions is also greater if the
endocervical brush is used [Boon, 1986]. Therefore, the original preference of the Sampling Working
Group was for the combined use of the appropriate end of the combination spatula and the
endocervical brush (CytobrushR). Recent experience of a few pathologists in the Working Group,
however, shows that the Cervex-BrushR scores very highly in terms of technical interpretability of the
smear. A supplementary literature survey reveals that the detection rate for cytological lesions and the
presence of metaplastic squamous cells and columnar cells does not differ significantly between the
Cervex-BrushR and the combination of spatula + CytobrushR [Arbyn, 99; Annex 5]. The cost of the
sampling devices are comparable in the two methods (~10 BEF). Given these arguments, the Working
Group has decided that both the Cervex-BrushR and the combination of the spatula with the CytobrushR
can now be recommended (see Figure 5).
The possibility of obtaining a representative smear with the Cervex-BrushR, i.e. with a single device
offers a definite saving in time, so the specimen can be fixed more quickly. Moreover, the
Cervex-BrushR is highly suitable for thin layer cytology (see Section 5).
The endocervical brush often results in a very slight bleeding, which usually causes minor if any
problems [Buntinx, 1989; Boon, 1989]. In the case of pregnancyg or a cervix that bleeds easily, it is
best to use the Cervex-BrushR.
After surgical intervention on the cervix, in post-menopausal women, and in follow-up of glandular
lesions, the CytobrushR is often required for obtaining endocervical cells. The combination method
with an endocervical brush is then preferred [Thompson, 1989; Szarewski, 1991].
Sampling technique
1. Cervex-BrushR sampling (see Figure 8)
Endocervical cells and exocervical cells are sampled simultaneously. The long bristles are applied to
the endocervix to pick up endocervical cells, while the short bristles are put into contact with the
exocervical region. The handle of the brush is rotated between thumb and forefinger a few times
(ideally 5) in a clockwise direction, under gentle pressure,h [Ferris, 1992]. The sample is spread onto
the slide with a painting action, using both sides of the brush (see Figure 9) and fixed immediately (see
Figure 10).
f
The combination spatula, as shown in Figure 5a, is a combination of a classical spatula as designed by Ayre (blunt end) and a
an Aylesbury spatula (sharp end with elongated tip).
g
As mentioned earlier, screening smears should not be taken from pregnant women.
h
The bristles of the Cervex-BrushR are constructed such that the cells are collected if the instrument is rotated clockwise
(comparable to a water-wheel effect). (Personal communication from the manufacturer).
S_eng1.doc
19/09/01 -
12
2. Combination sampling with spatula and endocervical brush
-
2.a. Spatula sampling (see Figure 11)
In theory, the end of the spatula is used that is most suitable to the anatomy of the portio. In nullipara,
the narrow end (Aylesbury spatula, with elongated tip) is usually used; in multipara the wider end is
used (Ayre spatula, with shorter tip). With the tip at the ostium, the spatula is rotated 360°under gentle
pressure. It will be necessary to change the hold at least once. The tip scrapes cells from the ostium
while the less protruding part scrapes the surface of the portio. Care should be taken to scrape the
(physiological) squamocolumnar junction over its whole circumference. If there is extensive ectropion,
the outer edge of the portio is scraped separately.
After the sample has been taken, the spatula is put aside with the specimen face-up. The danger of
drying out is less if the cell material plus mucus remain in contact with the sampling device. Only after
the CytobrushR has been used the material is spread onto the slide.
-
2.b. CytobrushR sampling (see Figure 12)
Then, the brush is inserted two thirds into the endocervical canal and gently rotated 90 to 180°. A full
turn as with the spatula sampling is not necessary since the bristles come in contact with the entire
surface of the endocervical canal [NCCLS, 1994]. Rotating over a larger angle only increases the
likelihood of cell damage and endocervical bleeding.
The CytobrushR is immediately rolled (not wiped) over the outer third of the slide. The sample from
the spatula is then smeared over the central third as quickly as possible (see Figure 13). Fixation is
done immediately. The interval between sampling and fixation should be as short as possible, because
drying artefacts can appear after just a few seconds.
The rolling and spreading is done is a single movement (not in a zigzag) and without pressure, so that
an even thin smear is obtained.
Because of blood loss associated with use of the CytobrushR, the spatula sample should be taken before
the one with brush.
If the double sample is spread over two slides, the spatula sample is spread and fixed before
proceeding to the endocervical brush sampling.
Fixation
Immediate fixation kills bacteria, denatures enzymes and eliminates the likelihood of drying artefacts
in the cells on later staining [Boon, 1993].
Epithelial cells stay moist for longer if they are left on the sampling device [Boon, 1993]. Once spread
onto the slide, they dry out very quickly, which leads to alterations in the chromatin pattern of the cells
if they are not fixed immediately (air drying effects) [Boon, 1993]. The speed of fixation is very
important, particularly if the specimen contains little or no mucus. Endocervical and immature
squamous cells, reserve cells, atrophic and dysplastic cells are extremely sensitive to drying. Mature
squamous cells are somewhat more resistant.
S_eng1.doc
19/09/01 -
13
Fixation is done with a special spray. Commercial hairsprays are not recommended given the variety
in composition.
It is best to shake the aerosol can well and remove its top before starting to take the smear. The
specimen is fixed by spraying the slide at right angles from 20 cm (see Figure 10). If the distance is
less, the cells may be blasted away or frozen; if on a slant, the material is blown into ridges. Formation
of droplets should be avoided: so do not spray too much [Buntinx, 1989].
Fixation can also be done by immersion for 5 to 30 minutes in 95% ethanol [NCCLS, 1994]. The top
of the container should be removed beforehand so that no precious seconds are lost.
Inadequate fixation reduces the specificity of the examination [Buntinx, 1992a; Arbyn, 1997]. The
sensitivity is also affected adversely.
Transport to the cytology laboratory
After fixation (spray fixation or by immersion), the specimen is allowed to dry fully in air. It is then
put into a cardboard or plastic holder for transport to the laboratory. A wet specimen can stick at the
edges if it is put into a plastic folder. The holder is labelled with identifying details matching those on
the request form (see Figure 14).
Biopsies in formalin are not sent in the same package as smears because the vapour can affect the
fixation of the smears.
Feedback on the quality evaluation of the specimen
Research has demonstrated that systematic feedback to the physician on the quality of the smears
he/she has taken can significantly improve the quality [Boon, 1986; Buntinx, 1993]. Periodic
notification of the average quality score of an individual doctor compared with the general or regional
distribution of smears taken by all doctors is an important incentive element in quality assurance of the
screening process [Buntinx, 1993; Wilson, 1999].
All cytology laboratories are also encouraged to provide regular feedback. In the future, a central
register can take on the task of providing physicians with feedback.
Training of physicians
The experience and dedication of the person taking the smear also plays a crucial role in achieving a
good smear [Lundberg, 1989b; Boon, 1993; Szarewski, 1993; Bar-Am, 1997]. According to some
authors, the collector effect is even more important than the choice of equipment [Wolfendale, 1991].
This is insufficiently stretched in comparative literature. Theoretical and practical training in taking
Pap smears needs to be a permanent feature of training programmes in the medical faculties and in
post-graduate courses [NHSCSP, 1998].
S_eng1.doc
19/09/01 -
14
3. Accompanying illustrations
These guidelines are illustrated with figures (see Annex 1), made by H. Vanvinckenroye, graphic artist
at the Institute of Tropical Medicine in Antwerp.
These illustrations are available in colour in the form of slides and overhead transparencies.
4. Standard request form
Clinical details are important for accurate interpretation of a smear [Craig, 1985; Bethesda: Lundberg,
1989; Luff, 1992; Kurman, 1994; NVVP, 1996; WUCC, 1996].
The Working Group Sampling presents a standard request form with a minimum list of essential
details that can contribute to the diagnostic value of the cytology report and the pertinence of any
follow-up advice.
Annex 3 contains a standard request form for cervico-vaginal cytology.
This form can be used for screening purposes, and also for follow-up or clinical indications.
Correctly completed identification and dating of the request and specimen are obviously indispensable
to the processing of the sample. The identification details of the patient can also be provided in the
form of a sticker or badge.
The cytopathologist needs relevant clinical information in order to increase the reliability of the
cytology result. This information is essential to the interpretation of certain cell pictures (e.g. presence
of endometrial cells according to the cycle, menopausal status, hormone therapy). Based on the
clinical data, specimens can also be selected for specific review (e.g. in the case of follow-up of low
grade lesions). Advice relating to further management depends both on the cytological appearance and
on clinical details. If this information is lacking, a specimen cannot be processed optimally. In this
case, it is recommended that the cytopathologist reports as follows: “satisfactory for evaluation but
limited by: lack of clinical data” [Luff, 1992; NVVP, 1996]. The Working Group Sampling
recommends the use of specific request forms for cervix cytology. Requests for clinical biological
investigation are usually unsuitable for this particular purpose and provide insufficient space for the
clinical details required.
Date of birth and date of last menstrual period must always be provided; other details should be
included if they apply to the patient. The use of the standard request form (Annex 3) allows all the
pertinent details to be listed simply and efficiently.
The minimum clinical data needed (origin of sample, gynaecological status, interventions) can be
provided by means of a check-box system. In this way, data relating to screening history and reason
for smear can also be specified.
So as not to overload the request form, space is provided for free text for reporting relevant data that
may influence the cell picture but are less frequently encountered (e.g. radiation, chemotherapy,
topical treatment of condylomas).
Finally, correct identification and signature of the requesting doctor are required. To aid
communication between the first and second echelon, the requesting doctor is recommended to
indicate whether a colleague (general practitioner or attending gynaecologist) is to receive a copy of
the report.
S_eng1.doc
19/09/01 -
15
5. Thin-layer cytology
Thin-layer cytology is a new technique for transferring the cellular material to the microscope slide.
The Cervex-BrushR is usually the device recommended for taking the sample. However the
combined use of spatula and endocervical brush is also an option.
The smear is not transferred in the usual way onto a slide. The sampling device i carrying the
material is immersed in a container with a special liquid transport medium. The container is then
sent to a specially equipped laboratory. At present, there are several commercial systems available.
Only ThinprepR (Cytic) and CytorichR (Autocyte) have so far been approved in the United States by
the Food and Drug Administration. With the first system, the liquid is sucked through a membrane
and the cellular material sticks to the filter, which is then stamped onto a slide in the form of a
monolayer. With the more automated AutocyteR machine, the material is sedimented through a
density gradient [Howel, 1998].
A primary advantage of these methods is that almost all the sampled cells are rinsed into the liquid
while with the conventional smear a selective portion of the cellular material may remain stuck to
the sampling device [Rubio, 1977; Hutchinson, 1993]. Transfer by means of a liquid increases the
likelihood of representative smears. In addition, fixation of the cell material is optimal. The altered
background requires a degree of adaptation on the part of the cytologist, however. Red blood cells
and mucus are for the most part absent and leukocytes are more evenly distributed. Epithelial
fragments, which are difficult to interpret on a classical smear, are for the most part disaggregated
during the preparation, while diagnostic clusters of columnar or metaplastic cells are usually
preserved. The microscopic visualisation of a calibrated thin line of beautifully distributed cells is
more comfortable for cytological interpretation, which should improve the evaluability and
diagnostic quality of the investigation [Linder, 1997; Austin, 1998].
Multiple smears can be made or additional investigations done (e.g. HPV-DNA detection) with the
residual material [Ferenczy, 1997; Sherman, 1997].
A significant disadvantage is the high cost - both the capital investment and the operating costs. The
cost-benefit ratio needs further investigation before this technique can be recommended for general
use.
i
The head of the Cervex-BrushR can easily be removed from the handle. The brush is then immersed in the transport fluid. The
container (brush plus fluid) is then sent off to the laboratory. The top of the Cytobrush and spatula should be broken off. The Cytobrush
needs a pair of wire-clippers. Recently, endocervical brushes have also become available with a notch, which facilitates snapping.
S_eng1.doc
19/09/01 -
16
6. References
Arbyn M., Albertyn A., De Groof V., 1996a. Consensus omtrent follow-up adviezen bij cytologische screening naar baarmoederhalskanker. Tijdschr. voor Geneeskunde,
52, 709-715.
Arbyn M., Wallyn S., Van Oyen H., Cannoodt L., Mulkers A., Calcoen P., 1997. Nationwide screening of cervical cytology, a practical approach. Second Symposium on
Cervical Cancer Screening, Belgian Society of Clinical Cytology, Flemish Community, Europe Against Cancer, Ghent 22th of March.
Austin R.M., Ramzy I., 1998. Increased detection of epithelial cell abnormalites by liquid-based gynecologic cytology preparations. A review of accumulated data. Acta
Cytol., 42, 178-184.
Backwith B.A., Abu-Jawdeh G.M., Burke L., Wang H.H., 1995. Lack of utility of Pap-smears obtained at colopscopy with tissue sampling. J. Gynecol. Surgery, 11, 85-89.
Bloom W., Fawcett D.W., 1975. Chapter 33: The female reproductive system. In: A Textbook of Histology, 10th edition, W.B Saunders Company, Philadelphia; pp 858906.
Boon M.E., 1985. De Cytobrush-methode: een middel ter verbetering van de kwaliteit van door huisartsen gemaakte uitstrijken. Medisch Contact, 28, 28-30.
Boon M.E., Alons-Van Kordelaar J.J.M., Rietveld-Scheffers P.E.M., 1986. Consequences of the introduction of the combined spatula and cytobrush sampling for cervical
cytology. Improvements in smear quality and detection rates. Acta Cytol., 30, 264-270.
Boon M.E., De Graaf-Guilloud J.C., Rietveld W.J., 1989. Analysis of five sampling methods for the preparation of cervical smears. Acta Cytol., 33, 843-848.
Boon M. E., Suurmeijer A.J.M., 1993. The Pap Smear, 2nd edition, Coulomb Press Leyden, Leiden; pp. 255-264.
Buntinx F., Essed G., 1989. Het maken van uitstrijkjes. Huisarts Nu, 18, 291-296.
Buntinx F., Boon M.E., Beck S., Knottnerus J.A., Essed G.G.M., 1991. Comparison of cytobrush sampling, spatula sampling and combined cytobrush-spatula sampling of
the uterine cervix. Acta Cytol., 35, 64-68.
Buntinx F., Knottnerus J.A., Crebolder H.F.J.M., Essed G.G.M., Schouten H., 1992a. Relation between quality of cervical smears and probability of abnormal result. BMJ,
304, 1224.
Buntinx F., 1992b. De afnamekwaliteit van cervixuitstrijkjes. Doctoraatsproefschrift Rijksuniversiteit Limburg te Maastricht, Thesis Publischers, Amsterdam.
Buntinx F., Knottnerus J.A., Crebolder HFJM, Seegers T., Essed G.G.M., Schouten H., 1993. Does feedback improve the quality of cervical smears? A randomized
controlled trial. Brit. J. Gen. Pract., 43, 194-198.
Buntinx F., Knottnerus J.A., André J., Crebolder H.F.J.M., 1994. The effect of different sampling devices on the presence of endocervical cells in cervical smears. A
systematic literature review. Eur. J. Cancer Prev., 3, 23-30.
Buntinx F., Essed G.G., Van der Putten H.W., Buchholz R.F., Arends J.W., Knottnerus J.A., 1995. Composition of cervical smears in patient with and without cervical
ectropion. J. Clin. Pathol., 48, 408-409.
Buntinx F., Brouwers M., 1996. Relation between sampling device and detection of abnormality in cervical smears: a meta-analysis of randomised and quasi-randomised
studies. BMJ, 313, 1285-1290.
Burghardt E., 1970. Latest aspects of precancerous lesions in squamous and columnar epithelium of the cevix. Int. J. Gynec. & Obstet., 8, 573-580.
Burghardt E., Pickel H., Girardi F., Ostör A.G., Tamussino K., 1998. Colposcopy – Cervical Pathology: Textbook and Atlas, 3th ed. Ed: Thieme, Stutgart, New York, 1998.
Cannon J.M., Blythe J.G., 1993. Comparison of the Cytobrush plus plastic spatula with the Cervex brush for obtaining endocervical cells. Obstet. Gynecol., 82, 569-572.
Craig S., 1985. The smear test. Aust. Family Physician, 14, 1092-1094.
Cuvelier C., Arbyn M., Bourgain C., De Roy G., Drijkoningen M., 1998. The organisation of cervical cancer screening in Flanders: uniform cytology reporting. Acta
Cytol., 42, 468.
Editorial Acta Cytologica, 1988. Replies to questions on quality assurance measures in cytopathology. Acta Cytol., 32, 913-939.
Elias, A., Linthorst G., Bekker B, Vooijs P.G., 1983. The significance of endocervical cells in the diagnosis of cervical epithelial changes. Acta Cytol., 27, 225-229.
Ferenczy A., Wright T.C., 1994. Ch 5: Anatomy and histology of the cervix. In: Blaustein's Pathology of the Female Genital Tract, ed. Kurman, R.J., 4 th edition, SpringerVerlag; pp 185-201.
Ferenczy A., Franco E.L., 1997. Human Papillomavirus DNA testing using liquid-based cytology. In: New Developments in Cervical Cancer Screening and Prevention, Ed:
E. Franco & J. Monsonego; Blackwell Science, Cambridge, pp 348-353.
Ferris D.G., Berrey M.M., Ellis K.E., Petry L.J., Voxnaes J., Beatie R.T., 1992. The optimal technique for obtaining a Papanicolaou smear with the Cervex brush. J Family
Practice, 34, 276-280.
Fokke H.E., Salvatore C.M., Schipper M.E.I., Bleker O.P., 1993. A randomised trial of three methods of obtaining Papanicolaou smears. Eu. J. Obstet. Gynecol. Reprod.
Biol., 48, 103-106.
Frost J.K., 1969. Diagnostic accuracy of cervical smears. Obstet. Gynec. Survey, 24, 893-908.
Garite T.J., Feldman M.J., 1978. An evalutation of cytologic sampling techniques: a comparative study. Acta Cytol., 22, 83-85.
Griffiths M., Turner M, Partington C.K., Soutter W.P., 1989. Should smears in a colposcopy clinic be taken after the application of acetic acid? Acta Cytol., 33, 324-326.
Henry J.A., Wadehra V., 1996. Influence of smear qualtity on the rate of detecting significant cervical cytologic abnormalities. Acta Cytol., 40, 529-535.
Howell L.P., Davis R.L., Belk T.I., Agdigos R., Lowe J., 1998. The Autocyte preparation system for gynecologic cytology. Acta Cytol., 42, 171-177.
Hutchinson M., Fertitta L., Goldbaum B., Hamza M., Vanerian S., Isenstain L., 1991. Cervex-brush and Cytobrush: comparison of their ability to sample abnormal cells for
cervical smears. J. Reprod. Med., 36, 581-586.
S_eng1.doc
19/09/01 -
17
Hutchinson,M.L., Isenstein,L.M., Goodman,A., Hurley,A., Douglas K.L., Mui K.K., Patten F.W., Zahniser D.J., 1993. Homogeneous sampling accounts for the
increased diagnostic accuracy using the ThinprepR processor. Am J Clin Pathol, 101, 215-219.
Järvi K., 1997. Cervex brush versus vaginal-cervical-endocervical (VCE) triple smear techniques in cervical sampling. Cytopathology, 8, 282-288.
Kavak Z.N., 1995. A ranomized comparison of the 3 Papanicolaou smear collection methods. Aus. N.Z. J. Obstet. Gynecol, 35, 446-449.
Kivlahan C., Ingram E., 1986. Papanicolaou smears without endocervical cells. Are they inadequate? Acta Cytol., 30, 258-260.
Kurman R.J., Solomon D., 1994. The Bethesda System for Reporting Cervical/Vaginal Cytologic Diagnoses. Springer Verlag, New York.
Linder J. 1997. Liquid-based cytology: comparison of ThinPrep 2000 with conventionally prepared Pap smears. In: New Developments in Cervical Cancer Screening and
Prevention, Ed: E. Franco & J. Monsonego; Blackwell Science, Cambridge, pp 284-293.
Luff R.D, 1992. The Bethesda system for reporting cervical/vaginal cytologic diagnosis: report of the 1991 Bethesda Workshop. Acta Cytol., 46, 719-721.
Lundberg G. D., 1989a. The 1988 Bethesda system for reporting cervical/vaginal cytological diagnoses. National Cancer Institute Workshop. JAMA, 262, 931-934.
Lundberg G.D., 1989b Quality assurance in cervical cytology. The Papanicolaou smear. JAMA, 262, 1672-1679.
McCord M.L., 1992. Cervical cytology: a randomized comparison of four sampling methods. Am. J. Obstet. Gynecol., 166, 1772-1779.
McGooghan E., 1997. Cervical cancer following negative smears. In: New Developments in Cervical Cancer Screening and Prevention; Ed: E. Franco & J. Monsonego;
Blackwell Science, Cambridge, pp 169-177.
Meisels A., 1990. Are two smears better than one? Acta Cytol., 34, 459-460.
Mitchell H., Medley G., 1991. Longitudinal study of women with cervical smears according to endocervical status. Lancet, 337, 265-267.
Mitchell H., Medley G., 1992. Influence of the endocervical status on the cytologic prediction of cervical intraepithelial neoplasia. Acta Cytol., 36, 875-880.
NCCLS, 1994. Papanicolaou technique; approved guideline. National Comity for Clinical Laboratory Standards, NCCLS Document GP15-A, vol 14 N°8, Pensylvania,
USA.
Neinstein L.S., 1992. Comparison of Cytobrush with Cervex-brush for endocervical cytologic sampling. J. Adolesc. Health, 13, 520-523.
Nielsen M.L., Davey D.D., Kline T.S., 1993. Specimen adequacy evaluation in gynecologic cytopahtology: current laboratory practice in the College of American Pathologists
Interlaboratory Comparison Program and tentative guidelines for future practice. Diagn Cytopathol., 9, 394-403.
NHG, 1996. NHG-Standaard Cervixuitstrijken (eerste herziening). Huisarts en Wetenschap, 39, 134-141.
NHSCSP, 1995. Achievable standards, benchmarks for reporting and criteria for evaluating cervical cytopathology. Working Party set by the Royal College of Pathologists,
the British Society for Clinical Cytology and the National Health System Cervical Screening Programme, NHSCSP Publication N°1, Sheffield, UK.
NHSCSP, 1996. Quality assurance guidelines for the cervical cancer screening programme. Working Party of the National Health System Cervical Screening Programme,
NHSCSP Publication N°3, Sheffield, UK.
NHSCSP, 1997. Improving the quality of the written information sent to women about cervical cancer screening: guidelines on the presentation and content of letters and
leaflets. Austoker J., Davey C., Jansen C., Working Party of the National Health System Cervical Screening Programme, NHSCSP Publication N°5, Sheffield, UK.
NHSCSP, 1998. Resource pack for training smear takers. Working Party of the National Health System Cervical Screening Programme, NHSCSP Publication N°9,
Sheffield, UK.
NVVP, 1996. Praktijkrichtlijn voor het opzetten van een kwaliteitssysteem voor cyto-pathologisch onderzoek van de baarmoederhals. Taakgroep Bevolkingsonderzoek
Baarmoederhalskanker.
Richart R. M., Valliant H.W. 1965. Influence of cell collection techniques upon cytologic diagnosis. Cancer, 11, 1474-1478.
Risberg B., Andersson A., Zetterberg C., Nordin B., 1997. Cervex-brush vs. spatula and Cytobrush. A cytohistologic evaluation. J Reprod Med, 42, 405-408.
Ritter J., Baldauf J.J., Dreyfus M., Sader N., 1995. The abnormal cervix in pregancy. In: Papillomavirus in Human Pathology; Ed. J. Monsonego; Ares Serono Symposia
Publications, Rome.
Rubio C.A., 1977. The false negative smear: II The trapping efffect of collecting instruments. Obstet. Gynecol., 49, 576-580.
Sedlis A., Walters A.T., Balin H., Hontz A., Lo Sciuto L., 1974. Evaluation of two simultaneously obtained cervical cytological smears: a comparison study. Acta Cytol.,
18, 291-296.
Sherman M.E., Shiffman M.H., Lorincz A.T., Herrero R., Hutchison M.L., Bratti C., Morales J., Hildesheim H., Helgesen K., Kelly D., Alfaro M., Mena F., Balmaceda I.,
Mango L., Greenberg M., 1997. Cervical specimens collected in liquid buffer are suitable for both cytologic screening and ancillar human papillomavirus testing. Cancer,
81, 89-97.
Sparrow M.J., 1997. A trial of two methods of taking cervical smears: the Aylesburry spatula plus cytology brush. N.Z. Med J., 110, 356-358.
Spurrett B., Ayer B., Pacey N.F., 1989. The inadequacies of instruments used for cercial screening. Aust. NZ. J. Gyneacol., 29, 44-46.
Solomon D., Henry M., 1995. Specimen adequacy. In Compendium on Qualtity Assurance, Proficiency Testing and Workload Limitations in Clinical Cytology. Ed: Wied
G.L., Keebler C.M., Rosenthal D.L., Schenck U., Somrak T.M., Vooijs G.P. Series: Tutorials of Cytology, Chicago, Illinois, USA. Pp 90-94.
Szarewski A., Cuzick J., Nayagam M., Thin R.N., 1990. A comparison of four cytological sampling techniques in a genitourinary medicine clinic. Genitourin Med, 66,
439-443.
Szarewski A., Cuzick J., Singer A., 1991. Cervical smears following laser treatment. Acta Cytol., 35, 76-78.
Szarewski A., Curran G., Edwards R., Cuzick J., Kocjan G., Bounds W., Guillebaud J., 1993. Comparison of four cytologic sampling techniques in a large family planning
centre. Acta Cytol., 37, 457-460
S_eng1.doc
19/09/01 -
18
Thompson D.W., 1989. Adequate "Pap" smears. A Guide for Sampling Techniques in Screening for abnormalities of the Uterine Cervix. Laboratory Proficiency Testing
Programme, Canada.
Vlaamse Stuurgroep Cervixkankeropsporing, 1996. Overleg omtrent communicatie tussen huisartsen en specialisten. Administratie Gezondheidzorg van de Vlaamse
Gemeenschap, Brussel, 9 februari 1996.
Vooijs GP, Elias A., Van der Graaf Y., Veling S., 1985. Relationship between the diagnosis of epithelial abnormalities and the composition of cervical smears. Acta Cytol.,
29, 323-328.
Vooijs G.P., van der Graaf Y., Elias A.G., 1987. Cellular composition of cervical smears in relation to the day of the menstrual cycle and the method of contraception. Acta
Cytol., 31, 417-26.
Werkgroep Follow-up (Abertyn G., Arbyn M., Bonte J., Buntinx F., De Groof V., Dhont M., Michielsen P., Thienpont L., Van Dam P., Van Oyen H., Vergote I., Weyler J.),
1995. Consensus omtrent follow-up advies bij cytologische screening naar baarmoederhalskanker. Koepel Cervix-screening, IHE/Episerie N° 10, 30 november 1995,
Brussel.
WHO, 1988. Cytological screening in the control of cervical cancer: technical guidelines. WHO, Geneva.
Willbur D.C., Cibas E.S., Merrir S., James L.P., Berger B.M., Bonfiglio T.A., 1994. Thinprep processor clinical trials demonstrate an increased detection rate of abnormal
cervical cytology specimens. Am J Clin Pahtol, 101, 209-214.
Williamson S.L.H., Hair T., Wadehra V., 1997. The effects of different sampling techniques on smear quality and the diagnosis of cytological abnormalities in cervical
screening. Cytopathology, 8, 188-195.
Wilson J.D., French R., Branch T., Sutton J. 1999. Inadequate cervical cytology – the need to audit individual smear takers’ inadequate inadequate rates. Cytopathology,
10, 107-111.
Wolfendale M., 1991. Cervical samplers. The most important variable is probably the operator’s skill. BMJ, 302, 1554-1555.
WUCC (Bourgain C., De Groof V., Drijkoningen V.), 1996. Naar een uniforme verslaggeving van de cervicale/vaginale cytologische diagnose gebaseerd op het Bethesda
systeem. Brochure gerealiseerd door de Werkgroep Uniformisatie van Cervixcytologie (WUCC) en voorgesteld op het Congres van de Belgische Vereniging van Klinische
Cytologie, Aalst 8 juni 1996.
Ziekenfondsraad, 1993. Uitgangspunten herstructurering bevolkingsonderzoek naar baarmoederhalskanker, 592, Amstelveen, Nederland.
S_eng1.doc
19/09/01 -
19
7. Annexes
1.
2.
3.
4.
5.
Accompanying illustrations
Quality judgement of the Papanicolaou smear according to The Bethesda System.
Standard request form
Summary of the guidelines: “Collection of an adequate Pap smear of the uterin cervix”
Literature review: Comparison of the quality of cervical smears obtained with
spatula/CytobrushR versus Cervex-BrushR
6. Members of the Working Group Sampling
7. List of used abbreviations
S_eng1.doc
19/09/01 -
20
Annex 1
Accompanying illustrations.
Fig 1 a - d
S_eng1.doc
Fig 2 a – d
19/09/01 -
21
Figure 1. Schematic presentation of the ontogenesis of the transformation zone
(after Ferency, 1994).
1a: The cervix is completely covered with squamous epithelium. The endocervical canal and the
underlying crypts are covered by mucin-secreting columnar epithelium. The original squamocolumnar junction (SCJ) is situated at the external ostium.
1b: Cervical ectropion: columnar epithelium has migrated more distally; the squamo-columnar
junction is situated at the exocervix.
1c: The exocervical columnar epithelium is replaced by metaplastic squamous epithelium. This is
termed the transformation zone (TZ). This TZ constitutes the predilection place for the
development of squamous intra-epithelial neoplastic lesions detectable by the Papanicolaou smear
test.
1d: The transformation zone is fully matured and covered with squamous epithelium. There is
only one physiological squamo-columnar junction left. This physiological SCJ migrates inwards
and is situated in the endocervical canal in post-menopausal women.
Figure 2. Different aspects of the cervical portio.
2a: View of the immature cervix: the squamo-columnar junction at the ostium
2b: Cervical ectropion
2c: View of the complete transformation zone (both junctions are visible)
2d: The squamo-columnar junction is situated inside the endocervical canal, only squamous
epithelium is visible from outside.
The visualisation of the portio and the localisation of the transformation zone are essential
for the collection of an adequate smear.
²
S_eng1.doc
19/09/01 -
22
Figure 3.
A: Squamous cells
B: Metaplastic cells
C: Columnar cells
An exo- and endocervical smear give the best guarantee to find cytological abnormalities. Such a
smear contains squamous and metaplastic and/or columnar cells.
S_eng1.doc
19/09/01 -
23
Figure 4.: Required material for sampling.
Microscope slide, pencil, gloves, sampling devices, fixative, slide holder and request form.
S_eng1.doc
19/09/01 -
24
Figure 5. Sampling devices.
5a: Combined spatula with an Ayre-pole (under) and an Aylesbury pole (on top)
5b: CytobrushR
5c: Cervex-BrushR
S_eng1.doc
19/09/01 -
25
Figure 6. Labelling the slide.
Identification data are written by pencil on the frosted end of the slide.
S_eng1.doc
19/09/01 -
26
Figure 7. Visualising the cervix.
After insertion of the vaginal speculum the cervix is visualised and the position of the
transformation zone is checked.
S_eng1.doc
19/09/01 -
27
Figure 8. Specimen collection using Cervex-BrushR
The rubber bristles of the Cervex-BrushR are flexible and adapt themselves to most of the
cervixes. This instrument permits to take an exo- and endocervical smear in one manoeuvre.
The central bristles of the Cervex-BrushR are inserted with gentle pressure in the endocervical
canal until the lateral bristles bend against the exocervix. The broom is then turned five times
360° by rolling the handle clockwise between thumb and index finger.
S_eng1.doc
19/09/01 -
28
Figure 9. Transferring the sample to the slide using Cervex-BrushR
The sample is spread onto the slide with a painting action, using both sides of the brush.
S_eng1.doc
19/09/01 -
29
Figure 10. Fix immediately.
The sample is fixed as soon as possible with a fixation spray. The spray is directed at right angles
at about 20 cm distance.
S_eng1.doc
19/09/01 -
30
Figure 11.Specimen collection using spatula.
An exocervical smear should be taken using the end of the spatula which corresponds best with
the anatomy of the cervix (the Aylesbury pole in nullipara and the Ayre pole in multipara).
The point of the spatula is inserted in the ostium. The whole circumference of the exocervix is
sampled under gentle pressure with the broad end of the spatula.
S_eng1.doc
19/09/01 -
31
Figure 12. Collection of an endocervical smear with the CytobrushR
The brush is inserted for two thirds in the endocervical canal and gently rotated 90° to 180°.
S_eng1.doc
19/09/01 -
32
Figure 13. Transferring the sample to the slide using CytobrushR and spatula.
The CytobrushR is rolled over the outer third of the slide.
The material on the spatula is spread in a thin layer in one movement over the middle third of the
slide.
Fixation is done immediately in the same way as prescribed in Figure 10.
S_eng1.doc
19/09/01 -
33
Figure 14. Standard request form
The fixed specimen is left to air dry and subsequently packed in the holder. The request form is
completely filled in.
S_eng1.doc
19/09/01 -
34
Annex 2
Quality judgement of a Papanicolaou smear according to The Bethesda System
The Bethesda System for reporting cervical smears, developed by the National Cancer Institute, awards
a significant role to quality evaluation of the specimen [Kurman, 1994; Nielsen, 1993]. The Flemish
standard protocol of the Working Group on Uniformisation of Cervical Cytology (WUCC) is based on
this American model [Cuvelier, 1998]. Three categories are distinguished:
- satisfactory (synonyms: adequate, optimal)
- satisfactory for evaluation but limited by … (qualify with reason) (synonym: suboptimal)
- unsatisfactory (qualify with reason), (synonym: inadequate or poor)
Four kinds of criteria are used for the quality evaluation of the smear: identity details of the woman and
specimen, availability of adequate and pertinent clinical data, cellular composition and technical
interpretability.
The quality of the smear to a great extent determines the accuracy of the cytological interpretation.
Below is the definition of the three categories of quality evaluation:
Satisfactory
-
Satisfactory labelling of the specimen matching the identification details on the request form.
Relevant clinical information.
Adequate number of well-preserved and clearly visible squamous cells: visible over more than 10%
of the specimen surface.
Presence of an endocervical/transformation zone (EC/TZ) component: at least two clusters of at
least 5 endocervical or metaplastic sqamous cells.
The requirement for the presence of a transformation zone/endocervical component does not change
with age and applies equally for pre and post menopausal women. If there is extensive atrophy,
however, metaplastic and endocervical cells can scarcely be distinguished from parabasal cells and
absence of a TZ/EC-component is not reported.
The presence of metaplastic squamous cells and/or columnar cells as well as squamous cells or mucus
indicate that the transformation zone has been sampled. However, they do not prove that the entire TZ
was included in the sampling. On the basis of the cellular composition alone, the cytologist cannot
guarantee that the entire circumference of the transformation zone has been sampled. It is ultimately up
to the doctor who takes the smear to integrate the information on the composition of the specimen with
the clinical data (examination, history) and thus assess the adequacy of a smear from an individual
woman [Kurman, 1994; NHSCSP, 1995].
Satisfactory for evaluation but limited by …
-
-
Between 50% and 75% of the specimen cannot be read because of interfering elements (blood,
inflammation, aggregation of cells, defective fixation, drying artefacts, cytolysis, contamination
etc.).
Absence of a component from the transformation zone: two clusters of at least 5 endocervical or
metaplastic cells (applies to all women, irrespective of age).
Absence of pertinent clinical information.
Unsatisfactory
-
Too few squamousl cells: well-preserved and recognisable squamous cells covering less than 10%
of the area of the smear.
S_eng1.doc
19/09/01 -
35
≥
-
75 % of the specimen unreadable because of blood, inflammation elements, areas that
are too thick, drying artefacts because of insufficient fixation, cytolysis or contaminants
[Kurman, 1994].
Identity data lacking on slide or missing on the accompanying request form.
Broken, irreparable slide.
Smears in which abnormalities of the epithelial cells are discovered are not labelled “unsatisfactory”.
With the specification of “satisfactory for evaluation but limited by… ” and ”unsatisfactory” the reason
is given (see Table 2).
Table 2: quality evaluation of Papanicolaou smears from the uterine cervix and summary of the possible reasons for less than satisfactory and
poor quality [WUCC, 1996].
Quality evaluation
0 Satisfactory
0 Satisfactory but limited by …
0 Unsatisfactory
Reasons:
exclusively endocervical cells
no endocervical cells
poor or unsatisfactory fixation
too little cellular material
too much blood, pus
extensive cytolysis
poor smear technique
other: .............................................
Thompson [1989] considers the finding of squamous metaplasia to be a strong indication that a sample
has been taken from the target zone. The presence of endocervical columnar cells proves that cellular
material from beyond the transformation zone has been included in the specimen. The presence of
squamous metaplastic and/or columnar cells as well as pavement cells, however, does not prove that
the entire transformation zone was sampled. [Kurman, 1994; Solomon, 1995]. Many observations
support the thesis that the presence of endocervical or metaplastic squamous cells increase the
likelihood of finding cytological lesions. And, vice versa, the absence of such cells leads to more
false-negative results.
A few recent longitudinal studies do not support this contention [Kivlahan, 1986; Mitchell, 1991].
Mitchell [1992] was able to confirm this correlation only for the presence of metaplastic, but not for
columnar endocervical cells. Boon [1993] stands by the contention that the presence of immature
metaplastic cells and/or columnar cells is associated with the detection of high-grade intra-epithelial
lesions. If low grade lesions are also included in the definition of a positive smear, this connection
disappears.
Boon [1993] also reports that the progressive nature of squamous cell lesions increases the closer they
are to the metasquamo-columnar junction and therefore also pleads for double
endocervical/exocervical sampling. Histological examination of conisation material and colposcopic
observations confirm this latest thesis [Burghardt, 1998].
The exclusive presence of columnar cells (with no squamous cells) indicates a non-representative
smear.
S_eng1.doc
19/09/01 -
36
Annex 3
AANVRAAGFORMULIER VOOR CERVICO-VAGINALE CYTOLOGIE
Identificatie patiënt
Naam en voornaam:................................
................................
................................
..............................
Adres: ................................
................................
................................
................................
.................
Geboortedatum: ................................
................................
................................
................................
...
Mutualiteitsgegevens: ................................
................................
................................
..........................
__________________________________________________________________________________
Datum afname staal: . . / . . / . . . .
Oorsprong staal £ cervix
£ vagina
Gynaecologische status
Datum laatste menses: . . / . . / . . . .
£ Anticonceptie
£ pil: ................................
.......................
£ spiraal
£ andere: ................................
.................
£ Zwanger . . . weken
£ Postpartum
£ Menopauzaal, beginjaar: . . . .
£ Substitutie
£ nee
£ ja
Vorig uitstrijkje
£ Ongekend
£ Nooit onderzocht
£ < 1 jaar
£ 1 tot 3 jaar
£ > 3 jaar
Reden tot onderzoek
£ Georganiseerde screening
£ Opportunistische screening
£ Klacht of klinisch letsel: .............................
................................
................................
..
£ Follow-up onderzoek
reden: ................................
........................
Gynaecologische ingrepen
£ Hysterectomie
£ cervix aanwezig
£ cervix afwezig
£ Conisatie, lusexcisie of analoge ingrepen
£ Andere:
Afname-instrument
£ Gecombineerde spatel
£ Spatel en Cytobrush
£ Cervex-Brush
£ Andere:(preciseer)… … … … … … … … … …
Gevolgde therapie
£ Chemotherapie
£ Pelvische radiotherapie
£ Anti-oestrogenen
Andere relevante inlichtingen uit de anamnese of het klinisch onderzoek
................................
................................
................................
................................
............................
................................
................................
................................
................................
............................
................................
................................
................................
................................
............................
Handtekening, datum en stempel van de aanvragende arts
Kopij verslag aan Dr. ................................
................................
................................
...........................
S_aanvr.doc
37
Annex 4
Summary: Collection of an adequate Papanicolaou smear of the uterine cervix
Importance and objective of the guideline on sampling
The correct sampling of a cervical smear with appropriate equipment and in the approved way
contributes to a significant extent to the diagnostic value of the screening test. A smear taken
unsatisfactorily is a significant cause of false negative and false positive results.
Below is a concise summary of “A technical guideline: collection of adequate Pap smears of the
uterine cervix” compiled by the technical Sampling Working Group. The aim is to propagate the
consensus amongst all general practitioners, gynaecologists, pathologists and other healthcare workers
involved in the detection of cervical cancer in Flanders.
Indicators of the quality of a smear
The cytological precursors of cervical cancers arise mainly in the transformation zone (TZ) between
the multilayer squamous epithelium and the endocervical columnar epithelium. Therefore, it is
important that cell material be sampled primarily from this zone. An optimum smear contains
metaplastic and columnar endocervical cells as well as squamous cells. The presence of significant
quantities of blood cells or inflammatory cells can reduce the quality of the smear. The cellular
material must be spread over the slide in a homogeneous thin layer. The specimen must be fixed as
quickly as possible. Quality evaluation of the smear (satisfactory, satisfactory but limited and
unsatisfactory) is an essential component of the cytology report.
Contra-indications for screening smears
Total hysterectomy, cervical amputation, and the presence of a suspect, macroscopically visible lesion
in the area of the cervix are contra-indications for screening . In the latter case, the woman must be
referred for colposcopic examination and/or biopsy.
Factors adversely affecting the quality of a smear
The quality of a smear is adversely affected by:
•
menstruation, blood loss, breakthrough bleeding
•
vaginal inflammation/infection
•
severe genital atrophy (menopause)
•
pregnancy and lactation
•
physical manipulation or chemical irritation such as: preceding digital vaginal examination,
disinfectant cream or liquid, lubricating jelly, vaginal medication (less than 48 hours before),
vaginal douche (less than 24 hours before), prior colposcopy with acetic acid (less than 24 hours
before), previous smear (less than 3 months before), cervical surgery (less than 3 months before)
•
radiotherapy
It is essential to recognise these factors and reduce their effect to a minimum. Relevant clinical
information must be recorded on the request form.
Information to the woman
The woman is informed of the aim of the smear and its procedure. She is informed that sometimes the
examination has to be repeated within 3 to 6 months, if the smear was not of satisfactory quality. The
doctor makes a clear arrangement how the woman will be notified of the laboratory result.
S_eng1.doc
19/09/01 -
38
Preparation of the smear
•
The equipment required for the smear is laid out ready in advance: specula (various sizes),
gloves, microscope slides, pencil, tampon tongs plus swabs, atraumatic forceps, sampling
devices: spatula + endocervical brush (CytobrushR or Cervex-BrushR), fixative spray or alcohol
solution, request form
The doctor should take special care to keep the interval between taking the smear and fixing it as
short as possible. So: remove top from aerosol can, check that the can is not blocked or empty.
•
The slide is labelled on the frosted end in lead pencil with the name or reference number.
•
The speculum is inserted preferably without any lubricating jelly and the cervix is visualised as
clearly as possible. If there is extreme retro or anteversion of the uterus, the cervix can be
manipulated into a good position with a pair of closed atraumatic forceps. The position and
aspect of the transformation zone is checked visually. Contact with the portio should be avoided
as far as is possible. Excessive mucus, discharge or blood can be removed carefully with a swab
(not wiped away).
Number of specimens
A single specimen is usually sufficient for a smear. The exocervical and endocervical material can be
spread over a single slide.
Sampling devices
Cervical screening always requires an endocervical and an exocervical sample, taken with the
appropriate instruments.
Two methods are recommended as the best choice:
•
Cervex-BrushR or endocervical brush (Fig. 1c)
•
Combination of a spatula (Figure 1a) for the exocervical sample and the CytobrushR (Figure 1b)
for the endocervical sample.
The wooden spatula (Figure 1a) with an Ayre end and Aylesbury end (with elongated tip) is best.
S_eng1.doc
19/09/01 -
39
The Cervex-BrushR is best if the woman is pregnant or has a cervix that bleeds easily. The combination
method is best if the squamocolumnar junction is high (often post menopausal), after cervical surgery or
if there is extensive ectropion of the columnar epithelium.
Technique for taking the smear
1.
Cervex-BrushR
Endocervical cells and exocervical cells are sampled simultaneously - the long bristles pick up
endocervical cells while the short bristles collect exocervical cells.
•
The long bristles are positioned endocervically.
•
With gentle pressure the brush is turned five times 360° by rolling the handle clockwise between
thumb and forefinger.
•
The sample is spread onto the slide with a painting action, using both sides of the brush.
•
The specimen is fixed immediately by being sprayed at right angles from a distance of 20 cm. If
closer, the cells are blown away or frozen, if on a slant, the material is blown into aggregates.
Droplet formation should be avoided: so do not use too much fixative. A very fast fixation, within
a few seconds, is essential to prevent drying artefacts.
S_eng1.doc
19/09/01 -
40
2.
•
•
•
•
•
•
•
•
•
Combination of spatula and CytobrushR
Take the spatula. The end of the spatula should be used that is most appropriate to the anatomy of
the portio. For nullipara, this is usually the Aylesbury end, for multipara the broader Ayre end.
The spatula is then rotated 360º under slight pressure with the point at the level of the ostium. The
hand-hold will need to be changed at least once.
The tip scrapes the ostium while the less protruding part scrapes the surface of the portio. Special
care should be taken to scrape the squamocolumnar junction as fully as possible. If there is
extensive ectropion, the outer part of the portio should be scraped separately.
After the sample has been taken, the spatula is put aside with the specimen face-up. The danger of
drying out is less if the cell material and mucus remain in contact with the sampling device. Only
after the CytobrushR has been used the material is spread onto the slide.
Take the CytobrushR. Insert it for two thirds into the endocervical canal and rotate gently 90 to
180º.
The CytobrushR is immediately rolled (not wiped) over the outer third of the microscope slide.
the spatula is then streaked as quickly as possible onto the central third.
The rolling and wiping should be done in one movement (not in a zigzag) and without pressure,
so that an even thin smear is obtained.
If a double sample is taken, spread over two slides (rarely necessary), a primary fixation is done
immediately after the spatula sample has been taken, before proceeding to taking the endocervical
sample.
S_eng1.doc
19/09/01 -
41
Transport to the cytology laboratory
After fixation, the specimen is allowed to dry completely. Then it is put into a cardboard or plastic
holder for transport to the laboratory. A wet specimen can stick at the edges if it is put into a plastic
folder. The holder is labelled with identification details matching those on the request form.
Standard request form for cytology laboratory
Clinical details are important for accurate interpretation of a smear. The standard form (see Annex 1) is
filled in as fully as possible. The identification details required for recording the cytological data must
not be missing. It is indicated whether the result of the examination should be sent to a different doctor.
Feedback on the quality evaluation of the specimen
The cytological report on the smear, evaluated according to the guidelines of the Working Group on
Uniformity in Cervix Cytology (WUCC) always includes an assessment of the quality of the smear.
Every doctor periodically receives a summary of the quality evaluations of the smears he/she has taken
compared with general distributions. This feedback, provided by the laboratory or a central register,
should help to improve the average quality of smears.
S_eng1.doc
19/09/01 -
42
Annex 5
Comparison of the quality of cervical smears obtained with the spatula & CytobrushR versus the
Cervex-BrushR
In order to study the differences in smear quality obtained with the two methods a review of the
cytological literature was undertaken. The following outcome measures were studied: ratios of
detection rates of intraepithelial lesions and the presence of endocervical cilindrical/metaplastic
squamous cells in Papanicoloau smears obtained with the considered sampling methods. Crude
relative risks for individual studies and weighted relative risks for the pooled studies were calculated
applying the same procedure as in Buntinx’ meta-anlysis [1996].
1. Combination spatula with extended tip & CytobrushR versus Cervex-BrushR
1. A. Detection rate of cervical lesions (Table 1)
1A1. Low SIL:
- No separate study showing any significant association.
- Crude pooled analysis: significantly more lesions found with the combination method:
RR=1.24 (CI=1.09-1.41).
- Weighted pooled analysis: the contrary is observed: Mantel-Haenszel RR =0.94
(CI=0.83-1.07). Nevertheless: the adjusted association is not significant.
1A2. High SIL:
- No separate study showing any significant association.
- Crude pooled analysis: no significant relation RR=0.80 (CI=0.58-1.09).
- Weighted pooled analysis: no significant relation: Mantel-Haenszel RR =1.01 (CI=0.731.38).
1. B. Presence of endocervical cells/transformation zone component (Table 2)
1B1. Presence of endocervical cells (EC):
- 2 separate studies (Hutchinson 92 & Svarevski 93) showing a significant association:
more EC found with the combination method.
- Crude pooled analysis: significantly more EC found with the combination method:
RR=1.08 (CI=1.07-1.08).
- Weighted pooled analysis: same result: RR =1.07 (CI=1.06-1.08).
1B2. Presence of metaplastic squamous cells (MSC):
- 1 separate study (Järvi 97) showing significantly more MSC with the Cervex BrushR.
- Crude pooled analysis: significant relation RR=0.90 (CI=0.85-0.95).
- Weighted pooled analysis: same result: Mantel-Haenszel RR =0.90 (CI=0.84-0.94).
1B3. Presence of metaplastic squamous/endocervical cells (MSC/EC):
- No separate study showing any significant relation.
- Crude pooled analysis: not significant relation RR=1.01 (CI=0.98-1.03).
- Weighted pooled analysis: same result: Mantel-Haenszel RR =1.01 (CI=0.98-1.03).
S_eng1.doc
19/09/01 -
43
2. Combination spatula & CytobrushR versus Cervex-BrushR
2. A. Detection rate of cervical lesions (Table 4)
2A1. ASCUS and low SIL:
- 1 study showing significantly more l-SIL and ASCUS with combination (Williamson 98).
- Crude pooled analysis: significantly more l-SIL lesions found with the combination
method: RR=1.13 (CI=1.03-1.24). Especially due to one study.
- Weighted pooled analysis: same result.
2A2. High SIL:
- No separate study showing any significant association.
- Crude pooled analysis: more h-SIL found with the combination, but not significant
relation RR=1.18 (CI=0.93-1.51).
- Weighted pooled analysis: idem.
2. B. Presence of endocervical cells/transformation zone component (Table5)
2B1. Presence of endocervical cells (EC):
- 1 separate study (Neinstein 92) showing a significant association: more EC found with
the combination method.
- Crude pooled analysis: significantly more EC found with the combination method:
RR=1.10 (CI=1.04-1.16).
- Weighted pooled analysis: same result: RR =1.06 (CI=1.00-1.06).
2B2. Presence of metaplastic squamous cells (MSC):
No studies.
2B3. Presence of metaplastic squamous/endocervical cells (MSC/EC):
- 2 studies (Svarevski 90 & 91) showing more MSC/EC with the combination.
- Crude pooled analysis: significantly more MSC/EC with the combination RR=1.04
(CI=1.03-1.06).
- Weighted pooled analysis: idem.
2. C. Quality judgement of the smear (Table 6)
-
-
Contradictory results: 2 studies (Fokke 93, Williamson 98) show more inadequate smears
with the combination (1st not significant); 2 other studies (Szarevski 90 & 91) show more
inadequate smears with the Cervex-BrushR.
Pooled analysis: the cervex smears have significantly less inadequate (adj RR=0.74;
CI:0.68-0.81).
2. D. Sensitivity, specificity (compared with histology) (Table 7)
Risberg found higher specificity and comparable sensitivity for the cervex, resulting in higher
positive predictive value and positive likelihood ratio.
S_eng1.doc
19/09/01 -
44
Conclusion:
More low SIL or ASCUS and smears with EC found with the combination method. Differences are
not spectacular. More smears found with metaplastic cells with the Cervex-BrushR. The CervexBrushR provides also less inadequate smears and is a little bit more specific.
I propose that the Cervex-BrushR should be considered as almost equivalent to the combination
methods.
Comfort: Cervex-BrushR offers more comfort, sampling is somehow quicker, less risk for fixation
problems.
Price of the Cervex-BrushR comparable to the combination (9-10 BEF).
One first choice cannot be recommended. It is to the lab or clinician to choose.
Marc Arbyn
Brussels, December 1998.
S_eng1.doc
19/09/01 -
07/02/00
Table 1: detection of cervical lesions in smears obtained with Cytobrush & spatula with extented tip (1) versus Cervex (2)
=> RR<1 means cervex is more effective
=> RR>1 means combination is more effective
combination
-ve
% +ve
Outcome
Boon, '89
Boon, '89
Hutchinson, '91
Hutchinson, '91
Acta Cytol
Mild
Severe
Mild
Mild
29
9
18
26
4.113
4.133
195
146
0,7%
0,2%
8,6%
15,0%
14
4
21
35
2.613
2.623
232
132
0,5%
0,2%
8,1%
21,0%
1,31
1,43
1,06
0,71
0,70
0,44
0,54
0,39
Szarewski, '93
Szarewski, '93
Acta Cytol
Acta Cytol
Mild
Severe
312
64
3.029
3.277
9,3%
1,9%
380
75
3.448
3.753
9,9%
2,0%
0,94
0,98
0,82
0,7
McCord, '92
Am J Obst Gyn
Mild
29
472
5,8%
31
474
6,1%
0,94
0,58
Järvi, '97
Cytopathol
Mild
2
1.062
0,2%
4
955
0,4%
0,45
0,08
J Repr Med
J Repr Med
+ve
% +ve
RR
CI low
Ref
Acta Cytol
+ve
cervex
-ve
Author, Year
crude RR
TOTAL mild
TOTAL severe
416
73
9.017
7.410
4,4%
1,0%
485
79
7.854
6.376
5,8%
1,2%
0,76
0,80
CI low
0,67
0,58
CI up Design
Setting
2,48
4,63
1,88 randomised
1,05 randomised
randomised/
1,08 period
"
1,36
pop screening
pop screening
screening
high risk population
fam plan, mostyly young
women
"
outpat clinic, pregn & npregn women
experienced sm takers
experienced sm takers
2 experienced clinicians
FU smears
experienced sm takers
"
1,54 randomised
randomised/
gynecol,GP, technicians
2,45 day
CI low
0,86
1,09
M-H RR
0,93
1,01
CI low
CI low
0,82
0,73
1,06
1,38
45
07/02/00
Table 2: detection of a TZ/EC component in smears obtained with Cytobrush & spatula with extended tip (1) versus Cervex (2)
=> RR<1 means cervex-sample contains more EC
=> RR>1 means combination-sample contains more EC
Author, Year
Ref
Outcome
McCord, '92
Am J Obst Gyn
Hutchinson, '91 J Repr Med
Hutchinson, '91 J Repr Med
EC
EC
MSC
Szarewski, '93
Acta Cytol
Järvi, '97
EC+
combination
EC%+
EC+
cervex
EC-
%+
RR
CI low
465
247
129
36
12
130
92,8%
95,3%
50,0%
454
251
166
51
57
142
89,9%
81,6%
53,9%
1,03
1,17
0,93
0,99
1,1
0,79
EC
3.097
244
92,7%
3.202
626
83,6%
1,11
1,09
Cytopathol
EC
714
30
96,0%
640
31
95,4%
1,01
0,98
Järvi, '97
Cytopathol
EC
270
48
84,9%
230
60
79,3%
1,07
0,99
Järvi, '97
Cytopathol
MSC
490
254
65,9%
518
153
77,2%
0,85
0,8
Järvi, '97
Cytopathol
MSC
192
126
60,4%
185
105
63,8%
0,95
0,84
Järvi, '97
Cytopathol
EC/MSC
717
27
96,4%
648
23
96,6%
1,00
0,98
Järvi, '97
Boon, '89
Cytopathol
EC/MSC
EC present
274
4.110
44
32
86,2%
99,2%
241
2.512
49
115
83,1%
95,6%
1,04
1,04
0,97
1,03
8.903
811
991
402
509
71
95,7%
61,5%
93,3%
7.289
869
889
940
400
72
88,6%
68,5%
92,5%
Acta Cytol
crude RR
TOT EC present
Tot Metapl present
TOT EC/metapl present
1,08
0,90
1,01
CI low
1,07
0,85
0,98
CI up Design
Setting
1,07 randomised
1,24
1,08
randomised/
1,13 period
randomised/
1,03 day
randomised/
1,15 day
randomised/
0,91 day
randomised/
1,07 day
randomised/
1,02 day
randomised/
1,11 day
1,05
CI low
1,08
0,95
1,03
M-H RR
1,07
0,9
1,01
outpat clinic, pregn & npregn women
fam plan, mostyly young
women
experienced sm takers
gynecol,GP,technicians
<50y
gynecol,GP,technicians
>=50 y
gynecol,GP,technicians
<50y
gynecol,GP,technicians
>=50 y
gynecol,GP,technicians
<50y
gynecol,GP,technicians
>=50 y
CI low
CI low
1,06
0,84
0,98
1,08
0,94
1,03
46
07/02/00
Table 3: distribution of quality judgments of smears obtained with Cytobrush & spatula with extended tip (1) versus Cervex (2)
=> RR<1 means cervex-sample less inadequate
=> RR>1 means combination-sample less inadequate
Author, Year
Ref
Outcome
Szarewski, '93
Acta Cytol
% inad
combination
n inad
% inad
inad
79
3.262
2,4%
inad
119
cervex
n inad % inad
3.709
3,1%
RR
1,31
CI low
0,99
CI up Design
Setting
randomised/
1,74 period
GP'practice, mostly screening, sometimes clinic+
47
07/02/00
Table 4: detection of cervical lesions in smears obtained with Cytobrush & spatula (1) versus Cervex (2)
=> RR<1 means cervex is more effective
=> RR>1 means combination is more effective
Author, year
Journal
Outcome
Fokke, '93
Szarewski, '90
Szarewski, '91
Szarewski, '91
Williamson, '97
Williamson, '97
Williamson, '97
Eu J Obst Gyn & Repr Biol
Mild dyspl.
Mild dyspl.
LSIL
HSIL
ASCUS
LSIL
HSIL
Genitourin Med
Acta Cytol
Acta Cytol
Cytopathol
Cytopathol
Cytopathol
+ve
5
400
23
9
1.165
409
129
combination
-ve
% +ve
87
1.369
358
372
13.034
13.790
14.070
5,4%
22,6%
6,0%
2,4%
8,2%
2,9%
0,9%
+ve
8
401
30
9
992
321
111
cervex
-ve
82
1.400
391
412
13.554
14.225
14.435
% +ve
8,9%
22,3%
7,1%
2,1%
6,8%
2,2%
0,8%
RR
0,61
1,02
0,85
1,10
1,20
1,31
1,19
crude RR
TOTAL ASCUS
TOTAL LSIL
TOTAL hSIL
1.165
837
138
13.034
15.604
14.442
8,2%
5,1%
0,9%
992
760
120
13.554
16.098
14.847
6,8%
4,5%
0,8%
1,20
1,13
1,18
CI low
0,21
0,9
0,5
0,44
1,11
1,13
0,92
CI low
1,11
1,03
0,93
CI up
1,8
1,15
1,43
2,75
1,31
1,51
1,53
CI low
1,31
1,24
1,51
Design
Setting
randomised/periodmostly young women
genitourin clinic
alternated/period FU after laser treatment
exp takers
"
"
NR
GP'practice, mostly screening, sometimes clinic+
"
"
"
"
M-H RR
1,13
1,18
CI low
1,03
0,93
CI low
1,24
1,51
48
07/02/00
Table 5: detection of a TZ/EC component in smears obtained with Cytobrush & spatula (1) versus Cervex (2)
=> RR<1 means cervex-sample contains more EC
=> RR>1 means combination-sample contains more EC
Author, year
Journal
Outcome
Neinstein, '92
Risberg, '97
Cannon, '93
Cannon, '93
Cannon, '93
Cannon, '93
Kavak, '95
Kavak, '95
J Ad Health
Au NZ J Obst Gyn
EC
EC
EC
EC >=mod
EC
EC >=mod
EC
EC >=mod
Szarewski, '90
Genitourin Med
Szarewski, '90
Fokke, '93
Acta Cytol
Sparrow, '97
combination
EC-
EC+
%+
EC+
cervex
EC-
%+
RR
CI low
CI up
77
77
115
86
43
37
80
57
7
19
17
46
3
9
31
54
91,7%
80,2%
87,1%
65,2%
93,5%
80,4%
72,1%
51,4%
59
84
144
107
58
50
88
73
22
33
33
70
9
17
22
37
72,8%
71,8%
81,4%
60,5%
86,6%
74,6%
80,0%
66,4%
1,26
1,12
1,07
1,08
1,08
1,08
0,90
0,77
1,09
0,96
0,97
0,91
0,96
0,88
0,76
0,62
1,46
1,3
1,18
1,28
1,22
1,32
1,01
0,97
EC/MSC
1.672
97
94,5%
1.639
162
91,0%
1,04
1,02
1,06
Eu J Obst Gyn & Repr Biol
EC/MSC
EC
360
82
21
10
94,5%
89,1%
374
77
47
13
88,8%
85,6%
1,06
1,04
1,02
0,93
1,11
1,16
NZ Med J
EC
95
2
97,9%
76
5
93,8%
J Reprod Med
Obstet Gynecol
Obstet Gynecol
Obstet Gynecol
Obstet Gynecol
Au NZ J Obst Gyn
1,04
crude RR
TOT EC present
TOT EC>=moderate
TOT EC/metapl present
569
180
2.032
96
109
118
85,6%
62,3%
94,5%
586
230
2.013
169
124
209
77,6%
65,0%
90,6%
1,10
0,96
1,04
0,98
CI low
1,04
0,85
1,03
1,11
CI low
1,16
1,08
1,06
Design
Setting
randomised
Adolescent clinic
randomised
Referred cases for biopsy
randomised
clinic gyneco/obstet (>1/3 pregnant women)
"
"
"
pregnant women (part of women in previous records)
"
"
randomised
"
alternated/
period
alternated/
period
clinic gyneco/obstet (>1/3 pregnant women)
"
mostly young women
genitourin clinic
FU after laser treatment
exp takers
n randomised
M-H RR
1,06
0,97
1,04
Sexual health service clinic
CI low
CI low
1,00
0,86
1,03
1,12
1,09
1,06
49
07/02/00
Table 6: distribution of quality judgments of smears obtained with Cytobrush & spatula (1) versus Cervex (2)
=> RR<1 means cervex-sample less inadequate
=> RR>1 means combination-sample less inadequate
combination
n inad
% inad
cervex
n inad
CI up
Journal
Outcome
Williamsom, '98
Acta Cytol
% inad
1.154
13.045
8,1%
827
13.719
5,7%
0,70
0,64
0,76
Szarewski, '90
Genitourin Med
% inad
35
1.734
2,0%
62
1.739
3,4%
1,74
1,16
2,62
Szarewski, '91
Fokke, '93
Acta Cytol
% inad
% poor
1
6
380
86
0,3%
6,5%
17
2
404
88
4,0%
2,2%
15,38
0,34
2,06
0,07
115,06
1,64
Eu J Obst Gyn & Repr Biol
inad
inad
% inad
RR
CI low
Author, year
crude RR
TOTAL POOR/INAD
1.196
15.245
7,3%
908
15.950
5,4%
1,35
CI low
0,21
Design
Setting
not randomised
alternated/
period
alternated/
period
GP'practice, mostly screening, sometimes clinic+
randomised
gynecol clinic
CI low
1,80
M-H RR
mostly young women
genitourin clinic
FU after laser treatment
exp takers
CI low
0,74
CI low
0,68
0,81
Table 7: sensitivity, specificity, positive predictive value and positive likelihood ratio of smears obtained with Cytobrush & spatula (1) versus Cervex (2)
for the detection of high grade SIL as compared with histology
Author, year
Risberg, '97
"
Journal
J Reprod Med
Outcome
Cyt+His+
HSIL Cervex
Sp/Cybr
46
35
Cyt-His+ Cyt+His- Cyt-His28
2
41
21
3
37
p for H0: cervex & spat/cyt.brush smears equally valid
TOT
117
96
SE
62,2%
62,5%
SP
95,3%
92,5%
PPV
95,8%
92,1%
0,97
0,59
0,46
Design
PLR
13,4 randomised
8,3
"
Setting
Referred cases for biopsy
"
50
51
Annex 6.
Members of the Sampling Working Group
Dr. Georges Albertyn
Voorzitter van de Werkgroep
Gynaecoloog
Academisch Ziekenhuis St.-Augustinus, Antwerpen-Wilrijk
Acacialaan 5, 2020 Antwerpen
Dr. MSc. Marc Arbyn
Verslaggever van de Werkgroep en auteur van het eindverslag
Epidemioloog
Wetenschappelijk Instituut voor Volksgezondheid
J. Wytsmanstraat 14
1050 Brussel
Dr. Claire Bourgain
Patholoog
Laboratorium voor Cytodiagnostiek
Vrije Universiteit Brussel
Laarbeeklaan 101
1080 Brussel
Prof. Dr. Frank Buntinx
Huisarts
Eenheid Klinische Epidemiologie
Academisch Centrum voor Huisartsgeneeskunde
Katholieke Universiteit Leuven
Kapucijnenvoer 33
3000 Leuven
Prof. Dr. Marc Dhont
Gynaecoloog
Diensthoofd Gynaecologie en Verloskunde
Universitair Ziekenhuis Gent
De Pintelaan 185
9000 Gent
Prof. Dr. Maria Drijkoningen
Patholoog
Laboratorium voor Pathologie
St-Rafaël Ziekenhuis, Katholieke Universiteit Leuven
Minderbroederstraat 12
3000 Leuven
S_eng1.doc
19/09/01 -
52
Dr. Patrick Neven
Gynaecoloog
Kliniek St.-Jan
Broekstraat 114
1000 Brussel
Dr. Louis Thienpont
Patholoog
Onze-Lieve-Vrouwe Ziekenhuis
Moorselbaan 164
9300 Aalst
Prof. Dr. Peter Vandam
Gynaecoloog-oncoloog
Academisch Ziekenhuis St.-Augustinus
Acacialaan 5
2020 Antwerpen
Prof. Dr. Herman Van Oyen
Epidemioloog
Diensthoofd Afdeling Epidemiologie
Wetenschappelijk Instituut voor Volksgezondheid
J. Wytsmanstraat 14
1050 Brussel
Prof. Dr. Ignace Vergote
Gynaecoloog-oncoloog
Universitair Ziekenhuis Leuven – Gasthuisberg
Herenstraat 49
3000 Leuven
S_eng1.doc
19/09/01 -
53
Annex 7
List of used abbreviations
AGUS
ASCUS
EC
HPV
H-SIL
L-SIL
MSC
RR
TZ
SCJ
VACK
WIV
WUCC
S_eng1.doc
atypical glandular cells of unspecified significance
atypical squamous cells of unspecified significance
endocervicaal
Human Papillomavirus
High grade squamous intraepithelial lesion
Low grade squamous intraepithelial lesion
Metaplastic squamous cells
Relative risk
transformation zone
squamo-columnaire junction
Vlaamse Advies-Commissie voor Kankerpreventie
Wetenschappelijk Instituut voor Volksgezondheid
Werkgroep voor Uniformisatie van Cervix-Cytologie
19/09/01 -