Wide local excision (WLE) for vaginal intraepithelial neoplasia (VAIN)

Copyright C Acta Obstet Gynecol Scand 1999

Acta Obstet Gynecol Scand 1999; 78: 648–652 Printed in Denmark ¡ All rights reserved

Acta Obstetricia et Gynecologica Scandinavica ISSN 0001-6349

ORIGINAL ARTICLE

Wide local excision (WLE) for vaginal intraepithelial neoplasia (VAIN) DANNY CHENG, TONG-YOW NG, HEXTAN Y. S. NGAN AND LING-CHUI WONG From the Department of Obstetrics and Gynecology, The University of Hong Kong, Queen Mary Hospital, Hongkong, Peoples’ Republic of China

Acta Obstet Gynecol Scand 1999; 78: 648–652. C Acta Obstet Gynecol Scand 1999 Background. To assess the use of wide local excision in the treatment of VAIN. Methods. A retrospective review on 40 WLE procedures for VAIN. Results. The mean age was 60 years. Thirty-six (90%) patients had previous treatment for genital tract cancer or pre-cancer. The median duration and blood loss during operation was 45 minutes and 50 mls respectively. Fifteen complications affected 11 patients. Only one of five patients with ‘vaginal cancer’ was diagnosed prior to WLE. Of the 35 patients treated for VAIN 3, 12 (34%) developed abnormal cytology during follow up – three had residual VAIN 3, five had recurrent VAIN 3 and four had invasive cancer diagnosed. The remaining 23 (66%) patients were disease free at a median follow up of 44 months. Conclusions. Ablative therapy for VAIN 3 is unsafe as occult invasive foci can be missed by biopsy. WLE is efficacious in treating high grade VAIN 3. Long term surveillance of the lower genital tract is needed to diagnose metachronous lesions. Key words: excision; occult cancer; treatment of VAIN Submitted 20 October, 1998 Accepted 8 February, 1999

Primary invasive and preinvasive neoplasia of the vagina are rare. In the well screened population, they are usually detected by abnormal cytology. Patients previously treated for genital tract dysplasia or malignancy have an increased risk of vaginal neoplasia, therefore, they should have regular clinical and cytologic follow up. Ablative therapy is commonly used to treat cervical intra-epithelial neoplasia (CIN). However, a histologic discrepancy between biopsy and loop excisional specimens of 47% is reported (1). On a review of 124 smears and 70 corresponding biopsies in 45 women diagnosed to have vaginal intraepithelial neoplasia (VAIN), the authors reported a cyto-histologic discrepancy of more than 2 Abbreviations: 5-FU: 5 fluorouracil; CIN: cervical intraepithelial neoplasia; N: number of patients; NRT: non-irradiated; RT: irradiated; VAIN: vaginal intraepithelial neoplasia; WLE: wide local excision. C Acta Obstet Gynecol Scand 78 (1999)

grades (out of 3) in 30% of cases (2). This caused concern over the use of ablative therapy in the treatment of VAIN. VAIN had been treated by partial vaginectomy (3–5), local means like 5-fluorouracil (5-FU) cream (6–13), laser treatment (10, 14–19), and irradiation (20–23) with variable degrees of success and morbidity. Recently, more investigators realized the importance of exclusion of occult invasion and reverted to the use of excisional biopsy. Reports of loop diathermy (24) or laser superficial vaginectomy (25) suggested their superior healing, ease of use and with few complications. We studied the utilization of wide local excision (WLE) of the vaginal vault in the treatment of high grade VAIN 3. The operative complications, incidence of occult cancer, correlation between the cytologic abnormalities and the worst histologic diagnoses, recurrence rate and the risk of developing ‘vaginal cancer’ are presented.

649

Wide local excision for VAIN Materials and methods

The case notes of all patients who had WLE of the vaginal vault performed from June 1982 to May 1998 were reviewed. Pathology slides were reviewed by a panel of gynecologic pathologists in conference with the clinicians. Except for those with malignant cells, all patients had at least 2 abnormal smears prior to further assessment. Colposcopic directed biopsies were performed when possible. Under anesthesia, the vagina was painted with Lugol’s iodine to highlight areas with no uptake. Any agglutination was broken down bluntly and the inferior margin of excision marked by scalpel. A combination of sharp and gauze dissection were used for en bloc removal of the mucosa of the upper vagina. Bleeding points were cauterized and the vagina packed overnight for persistent oozing. Prophylactic antibiotics and skin grafting were not used. The patient was discharged on day 2 unless there were complications. A vaginal mould was prescribed when required. The patients were reviewed 3 monthly the first year after treatment and thereafter 6 monthly by physical examination and cytology. Treatment failure (residual disease) was defined as persistent dyskaryosis or VAIN during the first 12 months of WLE and recurrence as disease reappearing when adequate follow up had been negative for more than 12 months (26). Follow up was censored on December 31, 1998. Results

Forty patients underwent WLE for VAIN during the study period. The mean age and parity for these cases was 60. One year (range 32–79) and 4.6 (range 0–10), respectively. Eleven (44%) of the 25 cases whose sexual function was recorded in the charts led an active sexual life and 14 did not. Twenty-nine (72.5%) patients had a past history of genital tract cancer (26 cervical and three endometrial) and 22 (55%) had been treated by pelvic irradiation (RT). Nine (22.5%) gave a past history of intraepithelial neoplasia of the lower genital tract (cervical seven, vaginal five, and three had both) while 22 (55%) had a prior hysterectomy (total abdominal 21, subtotal 1). Four (10%) gave no previous history of genital tract cancer or precancer. The median interval from previous genital tract treatment to the appearance of abnormal cytology was 24.1 months (range 1.4–331). This interval was significantly longer in the group of patients who had received pelvic irradiation (108 months vs 9.3 months, Mann-Whitney test, p∞0.0001). The colposcopic findings were available in 34 pa-

tients (Table I). One patient had no discernible abnormalities. Acetowhite epithelium and punctation were more commonly seen in the non-irradiated (NRT) group. Of the 12 patients with atypical vessels, two turned out to have invasive cancer (one had previous RT and the other had not). Thirty-nine (97.5%) of the procedures were carried out under general anesthesia and one (2.5%) under spinal anesthesia. The median duration and blood loss during operation was 48 minutes (range 15–165 min) and 50 mls (5–1000), respectively. There was no significant difference (p±0.05) between the irradiated (RT) (50 minutes, 40 mls) and non-irradiated (NRT) (45 minutes, 50 mls) groups. The mean size of the excised vaginal mucosa measured 2.3¿3.5 cm (range 0.5 to 5.5 cm¿1.0 to 7.0 cm). The median hospital stay was 3.0 days for both groups. The median duration for wound healing was 8 weeks (range 2–61) and this did not differ significantly (pΩ0.39) whether the patient had received prior irradiation (8.3 weeks) or not (7.1 weeks). Fifteen complications affected 11 (27.5%) patients – eight had 1 complication, two had 2, one had 3 (Table II). There was no case of procedure related cardio-pulmonary decompensation. The complication rate in irradiated patients (36%) was twice that of non-irradiated patients (17%) but this difference was statistically insignificant (pΩ0.29, 2tailed chi square test). One patient, from the RT group, suffered from a massive secondary hemorrhage 7 months after excision. She had suffered from persistent vault necrosis and abscess. Lig-

Table I. Colposcopic features* in the irradiated (RT) and non-irradiated (NRT) cases

Acetowhite epithelium Punctation Atypical vessels

RT group (nΩ19)

NRT group (nΩ15)

Total (nΩ34)

9 (47%) 3 (16%) 9 (47%)

14 (88%) 5 (41%) 3 (18%)

23 (68%) 8 (24%) 12 (35%)

* Some patients have more than one feature.

Table II. Complication rates between irradiated (RT) and non-irradiated (NRT) groups RT group (nΩ22) Primary hemorrhage Secondary hemorrhage* Fever Urinary tract infection Other†

1 2 4 2 3

(4.5%) (9.1%) (18.2%) (9.1%) (13.6%)

NRT group (nΩ18) 1 (5.5%) 0 0 2 (11.1%) 0

* See text for description. † One patient was taken back to the operating theater for hemostasis.

C Acta Obstet Gynecol Scand 78 (1999)

650

D. Cheng et al.

Table III. Correlation between cytology and worst histologic diagnosis Worst histology Cytology (total) Atypia (6) Severe dyskaryosis (13) Suspicious/malignant (21)

High grade VAIN

Invasion

6 12 17

0 1 4

nosed to have invasive cancer of the lower genital tract (Table IV). The remaining 23 patients showed no signs of disease – a median follow up of 44 months (range 4 to 132 months). This gives a failure rate of 34% (12/35), with a third (4/12) of them having invasive disease. Discussion

ation of internal iliac artery was required to control bleeding from an eroded pudendal artery. She was disease free with a duration of follow up of 69 months. Three more patients from the RT group had complications. One patient each had her bladder and the peritoneal cavity entered inadvertently. Both wounds were closed primarily without sequelae. The third patient had secondary hemorrhage on day 2. Examination under anesthesia and hemostasis was secured in the operating theater. She developed a vesicovaginal fistula 6 months postoperatively. An ileal conduit was performed for urinary diversion. The correlation between the worst cytologic and histologic diagnoses is shown in Table III. Three of five patients with ‘vaginal cancer’ had hysterectomy (two for cervical cancer and one for CIN 3) and two had pelvic irradiation (both for cervical cancer). Four had undergone colposcopic directed biopsy – all showed VAIN 3 but a small focus of early stromal invasion was identified in one patient. She had malignant vaginal cytology but declined further treatment. Here, ‘vaginal cancer’ is used to describe an invasive focus in the vagina. Twelve of 35 patients with VAIN 3 developed abnormal cytology at a median follow up of 13 months (range 1.1 to 124 months). On investigation, three had residual VAIN (two VAIN 3 and one VAIN 1) and five recurrent VAIN 3. Of the seven patients with VAIN 3, two underwent repeat local excision and both were disease free at 45 and 56 months of follow up, one was treated with 5-FU (Efudex) cream and four had no further treatment because of general debility. One patient with VAIN 1/HPV infection had a spontaneous regression of cytologic abnormalities. Four patients were diag-

Despite being exposed to the same oncogenic viruses associated with CIN (27), VAIN is uncommon. It was thought to be the absence of the squamocolumnar junction, where CIN commonly arises, that accounted for the rarity of VAIN (2). Over half of patients with VAIN had previously been treated for a genital tract neoplasm (2, 27). The cyto-pathologic classification of VAIN had been extrapolated from that of their cervical counterpart. With limited data, its natural history was assumed to be similar to that of CIN. Two reports addressed the pre-malignant potential of VAIN. The respective authors found epidemiological and molecular similarities between patients with VAIN and vaginal cancer leading them to conclude the malignant potential of the former (28, 29). Most gynecologists, therefore, manage abnormal vaginal cytology similarly to abnormal cervical cytology i.e. colposcopy, biopsy and treatment. In the present series where 55% had prior pelvic irradiation, difficulties with cytological diagnosis was highlighted by the 81% overcall rate (17/21) in those with malignant smears. In one series where 27% (14/45) of patients had received treatment for genital tract malignancy, only one of five patients with malignant cytology had histologic proof of invasive vaginal cancer i.e. an over-diagnosis rate of 80% (2). This contrasts with the 0.3% false positive rate in cervical cytology in population screening (30). The 8% (1/13) incidence of cytologic under-diagnosis is also comparable with that of cervical screening. Changes brought on by previous therapy, like vault pockets, vault agglutinations, telangiectasia, atrophy, could have camouflaged abnormal foci from the colposcopist. Atypical vessels, a cardinal

Table IV. Details of patients who developed invasive foci in the vagina after WLE for VAIN 3

Patient

Age

Antecedent diagnosis and treatment

WYL TK LNM LWH

58 65 56 64

Cervical cancer treated by hysterectomy and irradiation Cervical cancer treated by primary irradiation Cervical cancer treated by primary irradiation CIN3* treated by hysterectomy and vaginal cuff

* Cervical intra-epithelial neoplasia grade 3.

C Acta Obstet Gynecol Scand 78 (1999)

Site

Interval from primary treatment (months)

Interval from WLE (months)

Vault Vault Vault Paraurethral

133 115 111 52

39 85 24 4

Wide local excision for VAIN feature of stromal invasion, was only predictive of cancer in two of 12 patients – 11% of irradiated and 33% of non-irradiated patients. Additionally, the fibrotic epithelium might make an adequate biopsy difficult. Consequently, the pathologist is reluctant to diagnose invasion if only the superficial epithelium is submitted. All these make colposcopic biopsies unreliable in diagnosing occult invasion in patients with VAIN. Radiotherapy had the lowest failure rate of 0 to 14% (20–23, 26, 31, 32) in treating VAIN. However, it is contraindicated in previously irradiated patients because of its high local morbidity. Various regimens of local 5-FU cream have been proposed but its use was associated with poorly tolerated introital ulceration (6) The reported failure rate was up to 25% with a mean of 15% (6–11, 33). Many gynecologists excise CIN 3 in order to exclude early cancer missed by colposcopic biopsy, estimated to have an incidence of 0.5–1% (1, 34). Most colposcopists see few cases of VAIN because of its rarity. In five series totalling 101 patients, the rate of occult vaginal cancer uncovered on vaginectomy was up to 32%, with a mean of 23% (3–5, 24, 35). The fact that four of five cases of vaginal cancer in this series were undiagnosed pre-operatively adds weight to the argument that high grade VAIN should also be treated by excision. In two series of surgical excision, the mean failure rate was 5% (3, 4). With a follow up of between 3 months to 11 years, there was no failure or recurrence after transabdominal vaginectomy (3). However, significant morbidity included blood loss, hematoma formation, loss of coital function, bladder atony and stress incontinence. A laparotomy could tax cardio-pulmonary reserves, particularly in the elderly. In irradiated women, it might be impossible to dissect the bladder from the fibrotic anterior vaginal wall therefore, forcing one to abandon the procedure. Using transvaginal partial colpectomy, two of 12 patients recurred after a follow up period between 9 to 99 months (4). Laser vaginectomy seems to produce a specimen that is relatively free from thermal artifacts with rapid healing (25). Its use is associated with expensive equipment investment and protracted training and its failure rate remains to be reported. The rapidity of using electrical loop diathermy excision of VAIN was stressed as an advantage in a small series (24). However, not all gynecologists are experienced in WLE of the vaginal mucosa, particularly in irradiated patients. Despite the fact that irradiation did not significantly increase the complication rate, when they did occur to the irradiated patients, major operative interventions were required. One returned to the operating room for hemostasis, one internal iliac artery ligation, one ileal conduit,

651

one suturing of cystotomy. To minimize complications and morbidities, it is useful to refer the occasional case of VAIN, particularly in an irradiated patient, to experienced centers for management. The lack of major complications in other series (4, 5) might be accounted for by a low incidence of previous irradiation (0–16%) comparing to our patients (55%). Local ablative laser and diathermy treatment have the highest failure rate of up to 23% (10, 12, 14–17, 19, 26, 36). This was attributed to anatomical sanctuaries not easily accessible by the straight laser beams. The development of invasive cancer after ablative treatment for VAIN has been described (22, 26, 36). This ranged from 9 to 14% of vaginal cancer 6 to 36 months following laser ablation (18, 36). The case of para-urethral cancer could be explained by inadequate colposcopic evaluation of a pre-existing lesion. The other three cases of vaginal vault cancer occurred at 24, 39 and 195 months, with the latter two probably representing neoplastic field change rather than true recurrence. Conclusions

Persistent abnormal vaginal cytology is associated with significant pathology. Invasive foci can be missed by colposcopic directed biopsy. Ablative therapy for high grade VAIN is unsafe as occult cancer could be under treated. Cardiopulmonary morbidity of WLE is less than that of trans-abdominal vaginectomy but other surgical complications can be severe, particularly in irradiated patients. WLE is efficacious in treating high grade VAIN, with 66% of patients remaining disease free at a median follow up of 44 months. Long term surveillance of the whole vagina and vulva is important as these patients are at risk of vaginal cancer or metachronous high grade VAIN. References 1. Buxton EJ, Luesley DM, Shafi MI, Rollason M. Colposcopically directed punch biopsy: a potentially misleading investigation. Br J Obstet Gynaecol 1990; 98: 1273–6. 2. Sherman ME, Paull G. Vaginal intraepithelial neoplasia. Reproducibility of pathologic diagnosis and correlation of smears and biopsies. Acta Cytol 1993; 37: 699–704. 3. Ireland D, Monaghan J. The management of the patient with abnormal cytology following hysterectomy. Br J Obstet Gynaecol 1989; 95: 973–5. 4. Curtis P, Shepherd JH, Lowe DG, Jobling T. The role of partial colpectomy in the management of persistent vaginal neoplasia after primary treatment. Br Obstet Gynaecol 1992; 99: 587–9. 5. Hoffman MS, DeCasare SL, Roberts WS, Fiorica JV, Finan MA, Cavanagh D. Upper vaginectomy for in situ and occult, superficially invasive carcinoma of the vagina. Am J Obstet Gynecol 1992; 166: 30–3. C Acta Obstet Gynecol Scand 78 (1999)

652

D. Cheng et al.

6. Woodruff JD, Parmley TH, Julian CG. Topical 5-fluorouracil in the treatment of vaginal carcinoma in situ. Gynecol Oncol 1975; 3: 124–32. 7. Ballon SC, Roberts JA, Lagasse LD. Topical 5-fluorouracil in the treatment of intrapithelial neoplasia of the vagina. Obstet Gynecol 1979; 54: 163–6. 8. Piver MS, Barlow JJ, Tsukada Y, Gamarra M, Sandecki A. Postirradiation squamous cell carcinoma in situ of the vagina: Treatment by topical 20 percent 5-fluorouracil cream. Am J Obstet Gynecol 1979; 135: 377–80. 9. Daly JW, Ellis GF. Treatment of vaginal dysplasia and carcinoma in situ with topical 5-fluorouracil. Obstet Gynecol 1980; 55: 350–2. 10. Petrilli ES, Townsend DE, Morrow CP, Nakao CY. Vaginal intraepithelial neoplasia: Biologic aspects and treatment with topical 5-fluorouracil and the carbon dioxide laser. Am J Obstet Gynecol 1980; 138: 321–8. 11. Caglar H, Hertzog RW, Hreshchyshyn MM. Topical 5fluorouracil treatment of vaginal intraepithelial neoplasia. Obstet Gynecol 1981; 58: 580–3. 12. Krebs H. Treatment of vaginal intraepithelial neoplasia with laser and topical 5-fluorouracil. Obstet Gynecol 1989; 73: 657–60. 13. Sillman FH, Boyce JG, Macasaet MA, Nicastri AD. 5fluorouracil/chemosurgery for intraepithelial neoplasia of the lower genital tract. Obstet Gynecol 1998; 58: 356–60. 14. Stafl A, Wilkinson EJ, Matting RJ. Laser treatment of cervical and vaginal neoplasia. Am J Obstet Gynecol 1977; 128: 128–36. 15. Capen CV, Masterton BJ, Magrina JF, Calkins JW. Laser therapy of vaginal intraepithelial neoplasia. Am J Obstet Gynecol 1982; 142: 973–6. 16. Townsend DE, Levine RU, Crum CP, Richart RM. Treatment of vaginal carcinoma in situ with the carbon dioxide laser. Am J Obstet Gynecol 1982; 143: 565–8. 17. Jobson VW, Homesley HD. Treatment of vaginal intraepithelial neoplasia with the carbon dioxide laser. Obstet Gynecol 1983; 62: 90–3. 18. Stanhope CR, Phibbs GD, Stuart GC, Reid R. Carbon dioxide laser surgery. Obstet Gynecol 1983; 61: 624–7. 19. Stuart GC, Flagler EA. Nation JG, Duggan M, Robertson DI. Laser vaporization of vaginal intraepithelial neoplasia. Am J Obstet Gynecol 1988; 158: 240–3. 20. Rutledge F. Cancer of the vagina. Am J Obstet Gynecol 1967; 635–55. 21. Woodman CB, Mould JJ, Jordan JA. Radiotherapy in the management of vaginal intraepithelial neoplasia after hysterectomy. Br J Obstet Gynaecol 1988; 95: 976–9. 22. MacLeod C, Fowler A, Dalrymple C, Atkinson KA, Elliott PM, Carter J. Highdose-rate brachytherapy in the management of high-grade intraepithelial neoplasia of the vagina. Gynecol Oncol 1997; 65: 74–7. 23. Ogino I, Kitamura T, Okajima H, Matsubara S. High-dose-

C Acta Obstet Gynecol Scand 78 (1999)

24. 25. 26. 27. 28. 29.

30. 31. 32.

33. 34.

35. 36.

rate intracavitary radiotherapy in the management of cervical and vaginal intraepithelial neoplasia. Int J Radiat Oncol Biol Phys 1998; 40: 881–7. Patsner B. Treatment of vaginal dysplasia with loop excision: Report of five cases. Am J Obstet Gynecol 1993; 169: 179–80. Julian TM, O’Connell BJ, Gosewehr JA. Indications, techniques, and advantages of partial laser vaginectomy. Obstet Gynecol 1992; 80: 140–3. Lenehan PM, Meffe F, Lickrish GM. Vaginal intraepithelial neoplasia: biologic aspects and management. Obstet Gynecol 1986; 68: 333–7. Aho M, Vesterinen E, Meyer B, Purola E, Paavonen J. Natural history of vaginal intraepithelial neoplasia. Cancer 1991; 68: 195–7. Brinton LA, Nasca PC, Mallin K et al. A case control study of in situ and invasive carcinoma of the vagina. Gynecol Oncol 1990; 38: 49–54. Punnonen R, Kallioniemi OP, Mattila J, Vayrnen M, Koivula T. Primary invasive and in situ vaginal carcinoma. Flow cytometric analysis of DNA aneuploidy and cell proliferation from archival paraffin-embedded tissue. Eur J Obstet Gynecol Reprod Biol 1989; 32: 247–51. Morrison BJ, Coldman AJ, Boyes DA, Anderson GH. Forty years of repeated screening – the significance of carcinoma in situ. Br J Cancer 1996; 74: 814–19. Gallup DG, Morley GW. Carcinoma in situ of the vagina. A study and review. Obstet Gynecol 1975; 46: 334–40. Hernanandez-Linares W, Puthawala A, Nolan JF, Jenrstrom PH, Morrow CP. Carcinoma in situ of the vagina: past and present management. Obstet Gynecol 1980; 56: 356–60. Oliver JA, Jr. Severe dysplasia and carcinoma in situ of the vagina. Am J Obstet Gynecol 1979; 134: 133–7. Bigrigg MA, Codling BW, Pearson P, Read MD, Swingler GR. Colposcopic diagnosis and treatment of cervical dysplasia at a single clinic visit. Experience of low-voltage diathermy loop in 1000 patients. Lancet 1990; 336: 229–31. Choo YC, Anderson DG. Neoplasms of the vagina following cervical carcinoma. Gynecol Oncol 1982; 14: 125–32. Woodman CBJ, Jordan JA, Wade-Evans T. The management of vaginal intraepithelial neoplasia after hysterectomy. Br J Obstet Gynaecol 1984; 91: 707–11.

Address for correspondence: Danny Cheng, F.R.A.C.O.G. Department of Obstetrics and Gynecology The University of Hong Kong Queen Mary Hospital Hong Kong Peoples’ Republic of China