By Author
  By Title
  By Keywords

May 1993, Volume 43, Issue 5

Original Article

ROLE OF COLPOSCOPY IN THE DIAGNOSIS AND OUTPATIENT TREATMENT OF CERVICAL INTRAEPITHELIAL NEOPLASIA

Tasneem Aslam Tariq  ( Department of Gynaecology, St. Mary’s/Samaritan Hospital, London. )
William Peter Mason  ( Department of Gynaecology, St. Mary’s/Samaritan Hospital, London. )
Tracey Steele  ( Department of Gynaecology, St. Mary’s/Samaritan Hospital, London. )

ABSTRACT

One hundred and seventy nine women with abnormal smears referred for colposcopy to the Samaritan Hospital were studied. After confirming CIN on repeat smear, colposcopy and histology, cone biopsy was carried out. One hundred and twenty one patients (68%) had treatment either with electrical loop diathermy under local anaesthesia (105 patients, 59%) or laser excision of transforma­tion zone under general anaesthesia as a day case (16 patients, 9%). Cone biopsy was done in 121 patients (32%). We conclude that colposcopy is a valuable tool in combination with cytology and histology for diagnosis and delineating the extent of CIN, thus avoiding unnecessary cone biapsies particularly in women of child bearing age. The majority of patients were treated by an outpatient procedure, which is easy to learn, safe, effective and inexpensive (JPMA 43:86, 1993).

INTRODUCTION

Preinvasive cervical carcinoma was first described by Sir John Williams in 18861, long before exfoliative cytology was introduced. ‘WHO classification2 of mild, moderate, severe dysplasia and carcinoma in situ has been replaced with CIN 1, CIN 2, CIN 3. This system of gradingis well established and persists for the time being. It is convenient for treating patients. Hinselmann3 in 1925 first introduced the colpo­scope which made it easier to diagnose the preinvasive stages. Although cytology was introduced long after colposcopy, it was used widely for detecting preinvasive lesions well before colposcopy became a routine method of assessment for woman with abnormal cervical smears. Early diagnosis of CIN is suggested by cytology using the method described by Papanicolaou and Traut4 and may be confirmed by colposcopic examination. If required biopsy can be undertaken using colposcopic control for histopathology. In places where colposcopic expertise is not available, the standard management of an abnormal smear is still knife cone biopsy. When a colposcope is accessible, it is an invaluable aid to approach the clinical problem of CIN in which there are no symptoms or clinical findings on naked eye inspection of the cervix. Over the past 20 years the number of young women with GEM has increased dramatically5-7 in the UK. With the help of the colposcope it has become easier, not only to reduce the number of conizations, but also to treat patients with a cone biopsy tailored according to the extent of the lesion as an outpatient procedure. In this study we report our experience of outpatient colposcopy in the management of CIN using a protocol which allows for accurate diagnosis of cervical lesions with treatment tailored to each case as required. The methods are easy to learn, safe and well accepted.

PATIENTS AND METHODS

One hundred and seventy nine patients with abnormal  Samaritan Hospital’s colposcopy clinic over a period of 6 months from January - June, 1991. The referring smear was CIN1 in 55 patients (30.5%), CIN2 in 70 patients (39%) and CIN3 in 54 patients (30%). Colposcopy was performed in the outpatient clinic by colposcopists with experience ranging from 1-10 years, using a Zeiss colposcope OPMI 1 FC. Video attachment was helpful in demonstrating to the patient if she wished. The cervixwas exposed with a Cusco’s speculum and a repeat Papanicolaou smear was taken. The cervix was then examined through the colposcope to see any abnormal vascular pattern, before cleaning it with cotton ball soaked in 3% acetic acid solution. The transformation zone should be seen clearly and in cases where it was not visible, Kaurihaura forceps were used to visualize the canal and upper limit of the lesion. Following colpos­copy,  Lugol’s iodine was applied to the cervix to demarcate the abnormal area. Eppendorfer or Pistofidis forceps were used to take a punch biopsy from any visible suspicious areas for histology. All biopsies were serially sectioned and histologi­cally evaluated. Urgent histology results were obtained in 24 hours and the rest in a week’s time. The patient was informed of the results in 2-3 weeks time. A management plan was made depending on the smear results, colpos­copy and histology findings. CIN1 was generally not treated, whereas CIN2 and CIN3 were treated within 1-2 months. Colposcopy was described as satisfactory when the transformation zone was seen fully including the upper limit of the lesion. It was deemed unsatisfactory if the squamo columnar junction or upper limit of the lesion was not visible. Where colposcopy was satisfactory, large loop excision of the transformation zone was carried out with electrical loop diathermy under local anaesthesia using a dental syringe with 27 gauge needle, three 2.2 ml ampoules of Xylotox (lidocaine hydrochloride 24.7 mg/ml and adrenaline 22.7 mcg/ml) were injected at 6,9, 12 and 3 o’clock position. No other systemic or oral analgesic was given. Where colposcopy was unsatisfactory, laser cone excision was performed under general anaesthesia as a day case. The cone biopsy was fixed in formalin and sent for histology. Women with CIN1 were followed every 6 months by smear and colposcopy. The majority returned to normal and after 2 normal smears they were discharged back to the general practitioner for yearly smears. Women with consistent CIN1 smears were offered the option of either treatment or further follow-up. After treatment women were seen at 4 and 10 months for colposcopy and cytology. With two normal smears they were referred back to their general practitioner for yearly smears for 5 years.

RESULTS

The age of the patients ranged from 18-78, mean age was 31.6 years. Fifty nine percent of the patients were under the age of 30 years. Eighty percent of the patients had no living children. Parity ranged from zero to nine, average being 1.6. Colposcopic examination was satisfac­tory in 157 patients (88%) and unsatisfactory in 22 patients (12%). Outpatient treatment with large loop excision of transformation zone under local anaesthesia was done in 105 patients (59%). Inpatient treatment with laser excision of transformation zone under general anaesthesia was carried out in 16 patients (9%). The total number of patients treated was 121 patients (68%) (Table I).

Punch biopsy was carried out in 156 patients (87%). CIN1 was in 29 patients (18.58%), CIN2 in 42 patients (26.92%), CIN3 in 45 patients (28.84%) and no GIN in 40 patients (25.64%) (Table II).

Punch biopsy, when compared with the final cone biopsy agreed in 91 patients (75%) with one degree above or below (Table III).

Seventeen patients (14%) had no GIN on cone biopsy. Two patients had an unsuspected focus of micro invasion. Cytology had shown severe dyskaryosis and a punch biopsy in each case had revealed CIN3 only. Invasion extended to 1.1 mm in one case and 3.5 mm in depth and 1.1 mm horizontally in the other case. Both lesions were completely excised. They were of 33 and 24 years age with no living children. No further treatment has been given and follow-up has been normal. It is worthwhile seeing patients with persistent GIN 1 smears as in our series, 10 out of 15 had GIN 1 or worse on cone biopsy (Table IV). Excision was complete in 111 patients (91.73%) and incomplete in 11 patients (9.0%). In 6 patients lesion extended up to the endocervical margin and in 5 it extended up to the ectocervical margin. The majority of the patients tolerated the procedure quite well, no case was abandoned because of pain. No patient had excessive bleeding needing transfusion or was readmitted for secondary haemorrhage. One 70 year old patient had an unsatisfactory colposcopy. The vagina and cervix was atrophic and the cervix was flush with the ngina. Laser cone was not possible, hence cold cone knife biopsy was resorted to. The pouch of Douglas was opened accidentally and vaginal hysterectomy was car­ried out. Two patients who had complete excision of CIN3 showed recurrence of CIN2 confirmed on repeat cone biopsy at 10 months follow-up.

DISCUSSION

Colposcopy is a recognized useful clinical method in the early diagnosis of preclinical carcinoma of the cervix when used in conjunction with cytology. It compensates greatly for any error of cytology8. Cytology alone, gives no idea of the site and size of the lesion which can vary from a very small abnormality to a lesion extending to the vaginal fornices. Colposcopy can help in selecting patients with an abnormal smear of significance and can define the extent of the lesion. Patients with abnormal cervical cytology should have a colposcopically directed biopsy from the most significant area of their lesions before embarking on definitive treatment. Random punch biopsies may fail to determine the extent and severity of the disease9. The error of these random punch biopsies range from 6 to 34%10,11. A high degree of accuracy has been reported when using colposcopically directed punch biopsies to diagnose CIN12-15. Because of varying skills of the colpos­copist the biopsy can be unreliable and hence to improve accuracy multiple punch biopsy should be taken from suspicious areas. Widelyvarying discrepancies in correla­tion of colposcopic appearance and histopathology have been reported and this is attributed to different levels of expertise. Colposcopic appearance alone may have a poor analytical value. Surprisingly in our series 26% of punch biopsies had no CIN and in 14% of cases the histology of cone biopsy was negative. The flaw of treating patients based on colposcopic appearances alone cannot be over emphasised16 Recognising microinvasive carcinoma or early in­vasion is not easy as colposcopic features for it are not specific or diagnostic17-18. Choo19 in 1984 on reviewing 26 patients with microinvasion found that only 27% were correctly recognised at colposcopy. Bender20 reported a 15.9% failure to recognise microinvasion and 10.4% of occult invasive carcinoma by colposcopy. Biopsy taken should be big enough for a reliable opinion21. In our series 2 patients had GIN 3 on biopsy and colposcopic evaluation but on cone biopsy were found to have microinvasion. The main aim of the treatment should be to exclude invasion as failure to recognize may lead to under treatment. Under diagnosis is a more serious problem than over diagnosis. The only reliable way to exclude invasion is by excision biopsy22. This not only gives the whole specimen for examination and assessment but it is a safe and reliable way of curing the disease23. There are many ways of treating CJN such as cryotherapy, coagula­tion electrodiathermy, cold coagulation, laser vaporisa­tion, excision of transformation zone either by LLETZ or laser excision. The method of treatment can be chosen according to the available facilities in different countries. In places where colposcopy is not available the knife cone biopsy is still the standard procedure carried out in patients with abnormal pap smear, although it carries its own hazards of hospitalisation, general anaesthesia and late complications. Knife conization when compared with laser conization had more blood loss during the operative, early and late postoperative period. In Larsson’s series the blood loss during the operation was 0.4 to 155.4 mls in laser conization group and 5.6 to 1,570.9 mls in cold knife group. The incidence rate for bleeding complications requiring surgical intervention was 14.6% for knife cone and 1.8% for laser conization24. Stenosis after knife cone was 4.7% and after laser cone was 0.8% in Larsson’s series25. In Luesley’s series, 6% had primary and 7% had secondary haemorrhage, stenosis was in 17% and infertility in 4% of cases who had cone biopsies. Complication rate is greatly influenced with the size of the cone biopsy26. Davies in 1972 reviewed 400 consecutive cold knife conization and reported 11% morbidity and 9% incidence of significant bleeding. Gold cone biopsy may, however, be required in cases of post menopausal cervix totally flushed with vagina. Where colposcopy is available the initial step should be colpo­scopic assessment and not a knife cone biopsy. Cartier has long advocated low voltage diathermy loop as a means of investigating and treating CIN. LLETZ is being used more frequently now because it is a quick procedure, easy to learn and can be performed in the outpatient under local anaesthesia. It is well accepted by the patients and is generally pain free. Laser cone biopsy on the other hand takes longer to learn and perform. Treatment should be carried out carefully by an expert colposcopist after vigilant assessment of the patient. The objective in treating the patient with an abnormal cervical smear is to treat her as a prioritysafely and in-expensively. In these days the rising costs and busy operating lists have increased theimportance of safe outpatient treatment. In conclusion colposcopy should be used as an inter­mediate step between cytology and histology for evaluat­ing GIN. Correctly performed, colposcopy and colpo­scopic cone biopsy are important aids in the diagnosis and treatment of GIN allowing the identification of patients with insignificant changes only and avoids unnecessary treatment of these patients. In our study cone biopsy was avoided in 58 patients (32%). The extent of the abnormality is delineated by colposcopy was many of these patients can be safely treated on an outpatient basis only.

ACKNOWLEDGEMENTS

Our thanks go to Library Services, Royal College of Obstetricians and Gynaecologists, London for providing literature search, Angus J. McIndoe, senior Registrar, St. Mary’s Hospital for revising the manuscript and Tariq K. Shah, Research Fellow and Lecturer, University of Lon­don, Institute of Urology for help in preparing the manuscript.

REFERENCES

1. William, Sir John. cancer of the uterus, Harveian Lectures for 1886. London, Lewis, 188k.
2. Poulson, H.E., Taylor, C.W., Sobin, L.H. etal. International histological classification of turnouts 13. Geneva, World Health Organization, 1975.
3. Hinselmann, H. Verbesserung der Inspektionsmoglichkeit von Vulva, Vagina and Portio. Munchner Medizinische Wochenschrift, 1925;77:1733.
4. Papanicolaou, G. and Traut, H.P. The diagnosis of uterine cancer by thevaginal smear. New York, Commonwealth Fund, 1943.
5. Shingleton, ELM., Gore, H. and Austin, J.M. Jr. Outpatient evaluation of patients with atypical papanicolau smears. contribution of endocervical curretage. Am.J.Obstet.Gynecol., 1976;126:122-28.
6. Wolfendale, MR., King, S. and Ushetwood, M. Abnormal cervical smears; arewe in for an epidemic? Br.Med.J., 1983;287:526-28.
7. Yarnoz, M.C., Cortes, J., Liompart, M., Torrecabota, J., Rossello, J.J. and Amengual, E. The colpoacopyand thecone biopsy in the diagnoeia, treatment and follow-up of 81 cases of cervical intraepithelial neoplasia. Eur.J. Gynaecol. Oncol., 1988;9:345-49.
8. Navratil, E., Burghardt, E., Bajardi, F. and Naah, W. Simultaneoue colposcopy and cytology used in screening for carcinoma of the cervix. Am.J.Obstet.Gynecol., 1958;75:1292-97.
9. Townsend, D.E., Ostergard, D.R., Mishell, DR. and Hirose, F.M. Abnormal papanicolaou smear. Evaluation by colposcopy, biopsies and endocervical curettage. Am. J. Obstet. Gynecol., 1970;1O8:429-34,
10. Silbar, EL. and Woodruff, 3D. Evaluation of biopsy, coneand hysterectomy. Sequence in intrsepithelial carcinoma of the cervix. Obatet. GynecoL, 1966;27:89-97.
11. Anderaon, S.G. and Linton, E.B. The diagnostic accuracy ofeervical biopsy and cervical conization. Am.J.Obstet.Gyecol., 1967;99:113-16.
12. Ostergard, D.R. and Gondos, B. Outpatient therapy of preinvaaive cervical neoplasia: selection of patientawith theuse ofcolposcopy. Am.J.Obstet.Gynecol., 1973;115:783-85.
13. Bendet, 3.1..., Boyes, D.A., Nichols, T.M. and Millner, A. Colposcopic evaluation of patientswith abnormal cervical cytology. Br.3.Obstet. Gynecol.. 1976;83:177-82.
14. Kirkup, W. and Hill, A.S.The accuracyofcolposcopicallydirected biopsy in patients with suspected intraepithelisl neoplasia of the cervix. Br.3. Obstct. Gynecol., 1980;87:1-4.
15. Cinel, A.. Oselladore, Insacco, E. and Minucci, D. The accuracy of colposcopically directed biopsyin diagnosisof CIN. Eur.3.Gynecol. Oncol., 1990;443-7.
16. Edebri, A.A. The relative aignificance of colposcopic descriptive appearances in the diagnosis cervical intraepithelial neoplasia. Int.J. Gynecol. Obstet., 1990;33:23-29.
17. van Meir, J.M., Wielenga, G. and Dragendijk, A. Colpoacopy analysis of2Opatientswith microinvasive carcinoma of uterine cervix. Obstet. Gynecol Sure., 1979;34:872-3.
18. Sillman, F., Boyce, 3. and Fruchtcr, R. The significance of atypical vessels and neovascularization in cervical neoplasia. Am.J.Obstet.Gynecol., 1981;139:154-59.
19. Choo, Y.C., Chan, O.L.Y., Hsu, C. and Ma, H.K. Colposcopyin microinvasivecarcinoma of the cervix an enigma of diagnosis. Br.J., Obstet-Gynecol., 1984;91:1156-60.
20. Bender, J.L., Anderson, G.M. and Boyes, D.A. Colposcopicaccuracyin the diagnosisof microinvasive and occult invasivecarcinoma of thecervix. Obatet. Gynecol., 1985;65:557-62.
21. Anderson, M.C. and Hartley, RB. Cervical ctypt involvement by intrscpithelial neoplaaia. Obatet. Gynecol, 1980;55:546-50.
22. Howe, D.T. and Vincenti, A.Ct Is large loop excision of the transformation zone (LLETZ) more accurate than colpoacopically directed punch biopsy in the diagnosis of cervical intraepithelial neoplasia? Br.J.Obstet.Gynecol., 1991;98:588.91.
23. McIndoe, G.A.J., Robson, MS., Tidy, J.A., Mason, W.P. and Anderson, M.C. Laser excision rather than vaporization: the treatment of choice for cervical intracpithelial ncoplasia. Br.J. Obstet. Gynecol., 1989;74:165-68.
24. Larsaon, G., Per. AIm and Hans, Grundscll. Laser conization versus cold knife conization. Surg. Gynecol. Obstet., 1982;154:59-61.
25. L.arsson, G. and Grundsell, H. A comparison of complications of laser and cold knife conization. Obstet. Gynecol., 1983;62:213- 17.
26. Luesly, D.M., McCrum, A., Terry, P.B., Wade-Evans, T., Mylotte, M.J. and Emens, J.M. Complications of cone biopsy related to the dimensions of the cone and the influence of prior colpoacopicaasessment. Br.J.Obstet. Gynecol., 1985;92:158-64.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: