The importance of proper specimen collection cannot be overemphasized. Although no formal studies have demonstrated that educating clinicians on the optimal technique of obtaining cervical cytology samples improves specimen quality, there is considerable anecdotal evidence that this is important (Krieger et al., 1998). One half to two thirds of false negative cervical cytology results are attributable to either poor patient conditions at the time the cervical specimen is collected or the manner in which it is collected (Morell et al., 1982; Gay et al., 1985; Vooijs et al., 1985; Agency for Health Care Policy and Research, 1999). Therefore it is important that clinicians and nurses obtaining specimens be adequately trained in specimen collection and that they avoid situations that may reduce the performance of the test (McGoogan et al., 1998). This is especially important in low-resource settings, where women may undergo screening only once or twice in their lifetime.
Preparing the woman
Whenever possible, appointments for a cervical cytology examination should be scheduled approximately two weeks after the first day of the last menstrual period. Patients should be instructed to avoid sexual intercourse and douching for 24 to 48 hours before having the cytology specimen collected. In addition, women should not use any intravaginal products or medicine for several days before the smear is taken. Women using an intravaginal estrogen product should discontinue its use several days before the examination. Circumstances that may interfere with the interpretation of a cervical cytology test include active menstruation, significant cervical or vulvovaginal infections and a timing less than eight weeks post-partum. When a woman is actively menstruating, blood and cellular debris from the endometrium tend to obscure the cells on the smear, particularly during the first few days. Similarly, a cytology specimen should not be obtained when an abnormal vaginal or cervical discharge is observed. Women with a discharge should be evaluated for cervicitis and vaginitis using appropriate tests and be treated before the cytology specimen is taken, otherwise the specimen may be compromised by the inflammatory exudates or mildly reactive cells may be misinterpreted as a significant cytological abnormality. There is controversy as to the ideal timing of post-partum smears. Smears obtained less than eight weeks postpartum are often difficult to interpret because of marked inflammation and reparative changes, so a high rate of mild cytological abnormalities may be diagnosed. Another factor that can adversely affect the interpretation of cervical cytology specimens is severe atrophy. Although one should strive to collect specimens under ideal conditions, failure to comply with suggested screening intervals presents a greater risk to women. For previously noncompliant women, particularly those at risk for cervical neoplasia, a smear obtained under less than ideal conditions is preferable to no smear at all.
To collect a conventional cervical cytological specimen, the equipment required is a speculum, a light source, a collection device, a glass slide and fixative. Since most cervical cancer precursors and invasive cancers occur in the transformation zone, the use of specially designed devices that sample this area is recommended. The most common is a wooden or plastic spatula that conforms to the curvature of the portio. It is critical that the endocervical canal be sampled in order to obtain reasonable sensitivity (Martin Hirsch et al., 1999) and many spatula-type devices have extended tips designed to collect cells from this area. Either a moistened cotton swab or a brush-type endocervical sampler device (e.g., cytobrush) can be used to collect a second sample directly from the endocervical canal after the portio has been sampled (Koonings et al., 1992; Kohlberger et al., 1999). Recently developed collection devices that sample the endocervix and exocervix simultaneously do not provide a significantly lower false negative rate than the combination of spatula and a conical cervical brush (Szarewski et al., 1993). There is no consensus as to whether a single-slide technique, with both samples of the ectocervix and endocervix placed on the same slide, or a technique in which the two samples are put on two separate slides is preferable. Comparative studies of the two techniques have reported similar results (Saitas et al., 1995; Quackenbush, 1999). The single-slide approach has the advantage of reducing screening time and laboratory workload and it decreases the storage space required for archiving slides. When a single-slide technique is utilized, there also is no consensus on whether the specimens from the ectocervix and endocervix should be mixed together on the slide or kept separate as in the V (vagina) C (ectocervix) E (endocervix) technique.
Collecting the sample
A conventional cytology specimen is typically obtained using a spatula and conical cervical brush. The slide must first be labelled with the woman's name or number. Laboratories should have a written protocol specifying what is considered adequate labelling and should not accept inadequately labelled specimens. The person collecting the specimen should ensure that a test requisition is accurately and legibly filled out before collecting the specimen. The information most commonly requested by laboratories includes:
• Woman's name and indication if there has been a name change in the last five years. Some laboratories also use unique patient identifier numbers
• Date of birth or age
• Menstrual status (date of last menstrual period, whether the woman is pregnant, post partum, on hormone replacement therapy, or has had a hysterectomy)
• Previous history of abnormal cervical cytology, or treatment for CIN or cancer
• Whether the clinician considers the woman to be at high risk for developing CIN or cancer. Possible risk factors include smoking, infection with HIV, lack of previous screening and multiple partners.
• Specimen source – vaginal or cervical
Good visualization of the cervix is important for obtaining an adequate specimen. Cervical cytology specimens are generally collected with the woman in the dorsolithotomy position. A sterilized or single-use bivalve speculum of appropriate size is inserted into the vagina in such a manner as to allow complete visualization of the cervical os and as much of the transformation zone as possible. The cervix should not be contaminated with lubricant or water-soluble gel that may obscure the smear. Therefore the smear must be obtained before any bimanual examination. Gentle removal of excess mucus and discharge from the cervix with a large cotton-tipped applicator can produce a better-quality smear (Kotaska & Matisic, 2003), but vigorous cleansing may remove many of the most easily exfoliated cells. Saline should not be used to help clear debris from the surface of the cervix. It is also preferable not to apply 3–5% acetic acid to the cervix before taking the cytology specimen, as this can reduce the cellularity of the smear and produce poor staining (Griffiths et al., 1989; Cronje et al., 1997). Before the specimen is collected, the cervix should be carefully inspected with the naked eye for grossly visible masses or ulcerations that may indicate an invasive cervical cancer. If a grossly visible lesion isidentified, the woman should be referred for further confirmation. In many cases, the lesion can be directly sampled and the cellular sample obtained can be submitted separately for cytological assessment. The procedure for collecting cells from the cervix varies depending on the type of device used and the number of slides to be prepared. If a spatula and conical cervical brush are utilized, the first step is to place the spatula firmly against the ectocervix with the long projection extending into the endocervical canal. The spatula is then rotated several times 360° around the portio and removed. It is important to ensure that the entire squamocolumnar junction is sampled, since this is the site where most CIN lesions develop. In most women, the spatula will come into contact with the squamocolumnar junction if the pointed end is placed in the os, but in young women with a large ectopy, the spatula may need to be moved laterally to sample a peripherally positioned squamocolumnar junction. When rotating the spatula, it is easy to miss part of the cervix; this can be alleviated by directly visualizing the cervix while sampling. Transfer is best performed by using the spatula to thinly spread the cells onto the glass slide. It is important to ensure that as much cellular material as possible is transferred from both sides of the spatula. The endocervical canal is then sampled, using a conical cervical brush, which is placed in the endocervical canal so that the last few bristles remain visible and then gently rotated 90° to 180° once. One such rotation will adequately sample the endocervical canal and generally does not produce bleeding. Material from both sides of the spatula should be spread onto the slide. If collection devices that simultaneously sample both the endocervix and the ectocervix are used, the manufacturers' directions should be followed for each type of device. Cell fixation must be performed within a few seconds of specimen collection in order to prevent air-drying, which obscures cellular detail and hinders interpretation (Somrak et al., 1990). Immersing the slide in alcohol or spraying it with a specially formulated spray fixative can prevent air-drying. With immersion fixation, the slide is either immersed in alcohol and transferred to the laboratory in the container of alcohol or allowed to fix for 20 to 30 minutes in the alcohol, removed and allowed to air-dry. Various different spray fixatives are available. Only spray fixatives specifically designed for cytological specimens should be used and the manufacturer's instructions for a given product must be followed. The fixative should be liberally applied such that the slide appears moist over its entire surface. In order to prevent disruption of the cellular layer on the slide, the container of spray fixative should generally be held 15–25 cm from the slide during application.
Performance of conventional cytology
Despite the proven effectiveness of cervical cytological screening in reducing the incidence of cervical cancer, over the last decade the accuracy of cervical cytology has been questioned. Two factors need to be considered when assessing the accuracy of any screening or diagnostic test. One is whether the test is specific in detecting a given condition; the other is the sensitivity of the test for detecting the condition. Several large meta-analyses have indicated that both the sensitivity and specificity of cervical cytology are lower than previously thought (Fahey et al., 1995; McCrory et al., 1999; Nanda et al., 2000). [The Working Group considered the estimates of cytology test performance obtained through these meta-analyses to be of concern, given current cytology practices. In particular, it felt that it is very unlikely that specificities as low as 60–70% would be observed in a modern cytological screening practice.]. Even within the confines of research studies, a wide range of performance has been reported.