CHAPTER 7: Quality assurance guidelines for pathology in cervical screening based on draft of 19. Feb 2003 of R. Horvat, and on draft by A Hernández of 4. March 2003
Executive summary and introduction
Cervical cancer is in the vast majority of cases caused by an infection with high risk types of human papillomavirus (HPV), which is transmitted by sexual intercourse.
Transformation of normal cervical epithelium to invasive cancer is a process taking several years or even more than a decade in most cases, involving several stages of premalignant precursor lesions.
Despite of various treatment efforts, cervical cancer, detected in advanced stages has still a considerable mortality, while lethality of cervical cancers detected at early stages is small, and premalignant precursor lesions might be cured completely with preservation of the uterus.
In combination with the fact that the human cervix is an easily accessable structure, cervical screening is aimed to and also able to detect especially premalignant lesions or very early stage cervical cancer, which are cureable in the vast majority of cases. So nationwide cervical screening programs have been initiated in the last decades, which dramatically reduced invasive cervical cancer worldwide, but nevertheless, cervical cancer it is still one of the most common malignant diseases in women, especially in developing countries.
As a result of cervical screening programs, the number of cervical pre-malignant precursor lesions detected has raised significantly. In addition, patients presenting with cervical lesions have become considerable younger over the last 30 years due to changes of sexual habits.
Vaccacinations against HPV are currently under development, and there might be hope that one day HPV-caused cervical cancer might be exstinguished like smallpox. But even if a successful vaccacine is developed today and all children are inoculated now, it will take decades until an effect on the incidence of cervical cancer will be seen in the general population.
Cervical cytology does not have a final diagnostic significance, but has to be considered as important screening tool. So cervical cytology should always be considered as closely related to -and cooperating with- colposcopy and microscopical study of cervical biopsy.While a suspicion of a cervical lesion might be raised by cytological findings, a final diagnosis, on which therapeutic interventions are based on, must be achieved by means of a biopsy. Microscopical study of the biopsy will confirm or discard cytological findings, and therefore allow clinicians to make a final decision about the follow-up and treatment of the patient.
Colposcopy is the procedure to be undertaken in order to locate suspicious cervical areas. A good colposcopy is an absolute need for a cervical biopsy. For a highest efficiency in biopsy taking, several requirements should be met:
A properly trained colposcopist, preferably an able and experienced gynecologist specially dedicated to study and treatment of the lower genital area. The colposcopist should be able to distinguish efficiently between normal, benign and abnormal colposcopical changes.
Enough material for a proper histological study must be obtained, while avoiding any bleeding or other nuisances to the patient.
Colposcopy and biopsy do not require any anesthesia and should be performed in an outpatient regime.
Quality of the histopathological report will heavily depend on the biopsy quality. Since these specimens are often very small, careful handling and work-up will be required.
If suspicious changes or obvious dysplasia can be seen in the cytological slide, but not in the biopsy, the pathologist should always consider that dysplastic lesions might be rather small and difficult to reach, e.g. due to a endocervical localisation, and therefore might have been missed by a biopsy. For that reason histology and cytology should be closely correlated in such cases, so as to give the gynecologist a clear impression of the individual diagnostic situation: wether the changes found in the biopsy might explain the cytological changes, or wether performance of a new biopsy should be considered.
A special case of cervical biopsy is the cone biopsy. It is aimed to complete removal of dysplastic lesions, confirmed by a previous biopsy and cytology. Important information that pathologists should give to gynecologists should include a clear diagnosis of the primary lesion and wether the resection margins are affected by it. Since micro-invasion has a severe impact on clinical management of patients, complete work-up of specimens in serial sections is recommended.
Additional immunohistochemistry could be useful in order to support a diagnosis of possible microinvasion or vessel involvement.
7.1 Cervical biopsies
They include: exocervical punch biopsy, Loop- electrosurgical- Excision Procedure (LEEP), cone biopsy (Large Loop Excision of the Transformation Zone (LLETZ), by surgical blade-so called “cold-knife conization”, or with laser), endocervical curettage. Endocervical curettage, diathermal loop biopsies and cone biopsies are referred to in points 7.2 and 7.3 of this chapter.
High grade (High-grade SIL, in situ carcinoma or invasive carcinoma)
Persistent low grade
Persistent unclear cytological findings (ASCUS)
Suspicious colposcopy: As previously referred, biopsy must be performed under colposcopical control. Colposcopical changes suggestive of a cytological abnormality include those suggestive of an invasive carcinoma, and were listed as such in the colposcopical listings of the VII World Congress of I.F.C.P.C (1990), modified in the german version of the same year.
7.1.2 Diagnostic goal
Cervical cytology is not diagnostic, but a screening tool. In order to know if abnormal cytological findings are really related to an underlying pathological process in the cervical tissue, it is necessary to fulfill a colposcopical and bioptical study. By this procedure, the grade and extension of the lesion will be revealed and these findings will enable clinicians to perform adequate treatment.
Colposcopically directed biopsy will result in:
Histological evaluation of suspicious lesions, which requires subesquent treatment if necessary,
In most cases of cone biopsy or LEEP, the biopsy might also be therapeutic by achieving complete removal of the suspicious lesion.
7.1.3 Technique Cervical punch biopsy is the most frequently used method to obtain small fragments of cervical tissue. This kind of biopsy does not require any kind of anesthesia, and only causes a small ulcer, scarcely a bleeding one.
It is always performed under colposcopical control, being based on the removal of small fragments of colposcopically abnormal areas.
There are several kinds of instruments for this procedure, all of them composed by a fixed part and a cutting mobile one.
Women with a positive screening (cytological slide with high-grade changes or persistent low-grade or unclear cytological changes) must undergo colposcopical evaluation of the lower genital area. After locating the abnormal areas (white-coloured after application of acetic acid, dotted, mosaic, irregular vascularization or leucoplasia), a biopsy instrument is used on these areas to obtain a small tissue fragment. The material obtained this way should be immideatly transferred into a fixation solution. Formaline, alcohol or Bouin’s solution are the most commonly used fixatives, but formaline should be preferred, allowing a better immunohistochemical staining if necessary. The sepecimen should be sent to the pathological laboratory as soon as possible after biopsy.
Number, diameter and color of specimens should be documented.
Small specimens should be embedded in toto, for later work-up in serial sections. For paraffin embedding, it is necessary to remember that the cutting plane level should be at a normal angle to the surface of the specimen.
Larger specimens should be divided in half, and should also be worked up in serial sections.
Since the differences between LEEP and LLETZ are quite floating in clinical practice, LEEP specimens should be treated and processed like LLETZ-cone-biopsies (see below)
It should always include:
Histological type and grade of the lesion (CIN I- III)
Circumference of the lesion
When required by each case, reference will also be made to:
HPV –typical changes
Is normal epithelium present?
Invasion. If present: uni/ multifocal?, depth of invasion?, circumference of invasion?, stromal reaction? Dripping off of tumor cells? Lymphovascular invasions?
7.2 Endocervical curettage
It is the main procedure to obtain of endocervical tissue
Endocervical curettage should be compulsory in cases of positive screening with a normal or unsatisfactory colposcopy (those in which the transformation zone cannot be seen)
It is necessary when the screening citology is reported as:
Containing abnormal glandular cells (AGUS)
Suggestive of Adenocarcinoma
A fractionated curettage must be performed when an endometrial adenocarcinoma is suspected.
7.2.2 Diagnostic goal
Endocervical curettage is aimed to:
Evaluate wether the lesion has reached the endocervical channel. Treatment of cervical high-grade lesions depends on the lesional status of the endocervix. If the epithelial changes have reached the endocervix, a full cone biopsy would become necessary.
Diagnosis of endocervical adenocarcinoma
Discard a possible cervical involvement by an endometrial carcinoma.
7.2.3 Technique(Serial sections)
Endocervical curettage obtains tissue from the endocervical channel by using a sharp spoon, after cervical dilatation with a Hegar stalk (number 4). Tisssue from all four sides of the cervical channel must be obtained. Endocervical evaluation can also be performed by a microcolpohysteroscopia (direct visualizing of the endocervix after application of a color stain – lugol, methylene blue, waterman blue or toluidin blue).
Serial sections of the biopsy specimen are strongly recommended. If specimens are very small, they might be embedded wrapped in paper.
It is specially important to describe the various types of tissues found in the curettage:
Special attention should also be payed to necrotic tissue found in an endocervical curettage.
7.3 Cone biopsy
Cone biopsy can be defined as a resection of cone of cervical tissue with a truncated apex. Its basal circumference should include the iodine-negative area found in the Schiller test and its walls should include the longest possible extension of the cervical channel. In contrast to cervical biopsies mentioned in chapter 7.1, cone biopsy should be performed under general anesthesia, and hospitalization of patients is recommended.
Women with a positive screening for high-grade lesions and a transitional zone not evident in the colposcopy
Cytology suggestive of an early invasive carcinoma. Cone biopsy should not be performed in cases of clinically evident invasive carcinoma.
Positive screening for high-grade lesions, with colposcopically evident lesions of the transitional zone extending into the endocervical channel.
Persistence of AGUS or cytology suggestive of in situ adenocarcinoma.
7.3.2 Diagnostic goals
A cone biopsy is aimed to diagnosis of the cervical lesions. Nevertheless, in most cases this procedure will also have a therapeutic value (whenever the whole lesion is included in the specimen and the surgical margins are lesion-free).
Surgical blade conisation
The first step will be to stain the cervix with a lugol solution to visualize the dysplastic area. Afterwards, it will be held by two clamps (at 3 and 9 hours) externally to the lesion and infiltrated with vasopresin or epinephrin for hemosthasia, in four to six different points distant from the lesion. The surgical incision shall also be external to the lesion and will be deep enough to include most of the endocervical channel. During this procedure the fragment can be held by a clamp. After resection of the cone, bleeding can be stopped by means of electrocoagulation or by a Sturmdorf plastic.
As a general rule, these specimens are quite large and can be easily orientated. Nevertheless, as a consensus requirement, the specimen should be marked at 12 o´clock by the gynecologist in order to allow for a better orientation. Today this technique has been mostly replaced by LLETZ-conisation.
It is performed under colposcopical control, with the very same indications used for surgical blade conisation, using a CO2 laser instead of a surgical blade. Depth and width of the resection will be defined by the topography of the lesion (as seen in colposcopy) and of the cervix. A previous infiltration of epinephrin for hemosthasia must be used. The specimen shall be oriented by the gynecologist by placing surgical stitches at 12 o’clock.
Resection margins are in most cases damaged by the heat. This technique is rarely performed today, and has been mostly replaced by LLETZ-conisation.
LLETZ (Large Loop Excision of the Transformation Zone)
This is a complete biopsy of the transitional zone using a diathermal loop (a cutting instrument formed by an active electrode that is half circular or semioval).
It requires exposal and cleaning of the cervix, followed by an application of a iodine solution. This will evidence the areas to be resected. The loop must be placed on the external limit of the iodine-negative area. After beginning the emission of high-frequency electricity the electrode will be progressively and delicately introduced into the cervix, without any hesitations. It will be targeted to reach the internal limit of the lesion. Experience of the performer will define the speed and depth of the loop advance. Too fast speed would not allow for a correct hemosthasia, and too low speed would cause significant thermal damage in the remaining (helathy) cervical tissue. The same procedure is subesquently used for resection of the lower parts of the cervical channel.
So the specimen is sent to the pathological laboratory mostly in two portions, an ectocervical and an endocervical one. There should be marks for orientation at 12 o´clock at the ectocervical specimen, and at the endocervical margin of the endocervical piece. The gynecologist should specially try to retrieve the ectocervical specimen intact, i.e. with central cervical channel and well preserved circumference. This is the only way to enable the pathologist to determine if a lesion (dysplasia) has been resected in sano.
The gynecologist should also try not to expand the cervical channel prior to conisation.
7.3.4 Macroscopic description:
It should include: a-p diameter, height of the specimen, location of the cervical channel (central, paracentral, marginal), any macroscopically visible lesion, and the position of marks. If the circumference is not intact, this should also be noted.
7.3.5 Adequate orientation of cone specimens to define resection margins
prior to paraffin embedding
Ectocervical specimen: The mark (at 12 o´clock) should be replaced by ink. If required, another clearly distinguishable marking might be applied using an additional color.
It is recomended to cut both specimens, ectocervical and endocervical, in half.
The ectocervical specimen should be cut and divided from 12 to 6 o´clock, running through the cervical channel.
In case of a small ectocervical specimen, both parts might be placed in one cup. In this case, we recommend to mark the specimens also at 6 o´clock with a different color to allow an easier orientation.
As for the endocervical specimen, both halves are usually put together in a single cup.
For the sectioning of the specimen, we recommend serial sections of 0.1-0.2. mm thickness. This method would result in 100-200 sections/specimen, enabling the pathologist to assess possible microinvasion and vascular involvement in a more reliable way. We also recommend to place each 10th section on a silanized slide, thus allowing immunhistochemitry if required. If not, this slide might also be processed for H& E staining. Removal of coverslips is required before performing immunohisochemistry.
7.3.7 Immunohistochemistry to support a diagnosis of microinvasion or
Microinvasion: Immunohistochemical markers (antibodies directed to extracellular matrix components of the basal membrane) could be used in selected cases for the assessment of microinvasion.
Vascular invasion: Several studies have shown that routine H&E slides are not always adequate for the detection of vascular invasions, specially if there is a strong inflammatory stromal reaction. Antibodies against endothelial marker proteins (such as Factor VIII-related antigen), stain both lymphatic and blood vessel endothelium and therefore represent a useful tool for detection of lymphovascular invasion in cervical cancer. For a more selective assessment of blood vessels, CD31 can be recommended. For detection of lymph vessels invasion, immunostaining with newly recognized lymphendothelial proteins (like podoplanin) could be performed.
7.3.8 Histological diagnosis:
Features to be reported should include:
histological type and grade of the lesion (CIN I- III)
circumference and localisation of the lesion
distance to the endocervical and ectocervical margins
In cases where microinvasion is evidenced by the histological study, additional features to be included in the report would be:
The endocervical margin is a specially interesting point for the gynecologist: If ectocervical margins are not lesion-free, re-conisation is possible in most cases. Conversely, a dysplasia reaching very high areas of the cervical channel might also make hysterectomy necessary, if the lesion cannot be completely removed by cone biopsy.
7.4 Quality Assessment Women who have shown cytological disorders in a cervical screening program require a quality follow-up and monitoring:
For the first place, any of these patients require adequate information. The standards to assess the quality of the information provided are
Rate of women who have been given the information about their clinical symptoms at the same moment in which they were informed of their morphological disorders.
Rate of women who considered that the information they were given together with the diagnosis of their smear was satisfactory.
Any woman whose cytological slide has shown signs of a high-grade cervical lesion must undergo a later colposcopical and biopsic study. A quality monitoring shall be identified by
Rate of women with a cytological report of a high-grade lesion who have undergone colposcopy and biopsy.
c) Any woman showing morphological changes in the cervical screening program require an adequate follow-up. The guideline to evaluate the quality of this follow-up will be
Rate of women with a cytological report of morphological changes who have been seen by a specialist in a maximum of three months after the cytology.
In addition, it is recommended to establish e.g. weekly meetings of pathologists, cyto-technicians and gynecologists, were cytological slides, colposcopic pictures and histological slides are shown and discussed together.
Birner P, Obermair A, Schindl M, Kowalski H, Breitenecker G, Oberhuber G: Selective immunohistochemical staining of blood and lymphatic vessels reveals independent prognostic influence of blood and lymphatic vessel invasion in early-stage cervical cancer; Clin Cancer Res 2001 Jan;7(1):93-7.
Costa MJ, Grimes C, Tackett E, Naib ZM.: Cervicovaginal cytology in an indigent population. Comparison of results for
1964, 1981 and 1989; Acta Cytol 1991 Jan-Feb;35(1):51-6.
Gunasekera PC, Phipps JH, Lewis BV. Large loop excision of the transformation zone (LLETZ) compared to carbon dioxide laser in the treatment of CIN: a superior mode of treatment. Br J Obstet Gynaecol. 1990 Nov;97(11):995-8.
Heatley MK. How many histological levels should be examined from tissue blocks originating in cone biopsy and large loop excision of the transformation zone specimens of cervix? J Clin Pathol. 2001 Aug;54(8):650-1.
Horn LC, Riethdorf L, Loning T. Leitfaden für die Präparation uteriner Operationspräparate. Pathologe. 1999 Jan;20(1):9-14.
Kobak WH, Roman LD, Felix JC, Muderspach LI, Schlaerth JB, Morrow CP. The role of endocervical curettage at cervical conization for high-grade dysplasia. Obstet Gynecol. 1995 Feb;85(2):197-201.
Obermair A, Wanner C, Bilgi S, Speiser P, Reisenberger K, Kaider A, Kainz C, Leodolter S, Breitenecker G, Gitsch G.. The influence of vascular space involvement on the prognosis of patients with stage IB cervical carcinoma: correlation of results from hematoxylin and eosin staining with results from immunostaining for factor VIII-related antigen. Cancer 1998 Feb 15;82(4):689-96.
Prendiville W, Cullimore J, Norman S.: Large loop excision of the transformation zone (LLETZ). A new method of management for women with cervical intraepithelial neoplasia. Br J Obstet Gynaecol. 1989 Sep;96(9):1054-60.
Robboy SJ, Bentley RC, Russell P, Anderson MC: Cutup- the gross desription, processing and reporting of specimens. In: Robboy SJ, Anderson MC, Russel P (Eds.): Pathology of the Female Reproductive Tract., pp. 861- 877, Churchill Livingstone London 2002.
Robboy SJ, Kraus FT, Kurman RJ: Gross Desription, Processing and Reporting of Gyencologic and Obstetric Specimens. In: Kurman RJ (Ed.): Blaustein´s Pathology of the Female Genital Tract 4th Ed.., pp. 1225- 1240, Springer-Verlag New York Berlin Heidelberg 1994.
Spitzer M, Chernys AE, Shifrin A, Ryskin M. Indications for cone biopsy: pathologic correlation. Am J Obstet Gynecol. 1998 Jan;178(1 Pt 1)
Vutuc C, Haidinger G, Waldhoer T, Ahmad F, Breitenecker G: Prevalence of self-reported cervical cancer screening and impact on cervical cancer mortality in Austria; Wien Klin Wochenschr 1999 May 7;111(9):354-9.