Vulval squamous cell carcinoma is relatively rare; however, up to 20% of women have significant vulval symptoms during their lifetime. Formal screening programmes for vulval disease have not been established. The evidence for the use of vulval cytology and vulvoscopy is reviewed. No randomised-controlled trials have compared follow-up regimens, and although a few consensus documents have been published, formal guidelines are lacking in Grade A evidence. With increasing pressure on healthcare resources, the possibility of identifying high-risk groups to optimise the use of follow up in specialist clinics is explored.
Vulval disease is uncommon and there is no evidence that screening would decrease incidence. If high-risk groups can be identified, follow up should take place in specialised vulval clinics with experienced clinicians who are trained in vulval disease. Women with uncomplicated vulval conditions should be discharged to patient-initiated follow up or primary care. Central to the reduction of mortality and morbidity is increased awareness of vulval conditions among women and improved education of healthcare professionals, with particular understanding of the importance of physical examination.
Screening for vulval disease
The incidence of vulval cancer is about two in 100,000, and a life-time risk has been calculated as 1 in 293 women. The incidence of vulval skin disorders is not known. A community-based survey has estimated that 20% of women have vulval symptoms lasting more than 3 months during their lifetime.
Cervical cancer screening programmes have been effective in identifying premalignant lesions and reducing the incidence of cervical cancer. Is it possible to screen for vulval cancer or its precursors? If not, is it possible to identify women with existing vulval skin disorders who are at increased risk of developing cancer, and concentrate follow up on this group? In the current climate of reduced clinical resources, is there justification for prolonged specialist input, or would patient-initiated follow up or primary care surveillance be equally effective? In this chapter, I examine evidence that could provide a framework for follow-up guidelines, and focus in particular on vulval premalignancies.
Screening in primary care
Self-examination
The Vulval Health Awareness Campaign, launched 10 years ago, promotes vulva health awareness and provides a dedicated helpline for women with any vulval disorder in the UK. In collaboration with the National Lichen Sclerosus Support Group and Vulval Pain Society, it encourages self-examination once a month between periods, and provides information on the changes to look for in vulval skin. Education of the public and health professionals is important. A former President of the International Society for the Study of Vulvovaginal Disease commented: ‘Perhaps the role of the GP to assess and refer on selected cases will act as the ‘gate keeper’, but only if we continue to educate medical students and GP trainees so they are competent in screening, examination and selective referral.’ Inspection of the vulva, as a basic level of screening, can also take place when women attend for cervical smears in primary care; however, this must be coupled with increased awareness by women and clinicians.
Sixty years ago, clinician and patient delay was reported in the diagnosis of vulval cancer, more so than other gynaecological malignancies, and things have not changed. Women who present with vulval cancer have often had symptoms for many years, and evidence shows that delay in diagnosis is common. In a retrospective review of 102 women with vulval cancer, 31% had more than three consultations relating to vulval symptoms more than 6 months before the diagnosis of cancer. Women in whom a biopsy was carried out were more likely to present with an earlier stage cancer. Ninety-four per cent had symptoms of chronic irritation and 85% had abnormal skin adjacent to the tumour. The investigators concluded that suboptimal primary and secondary care and the failure of knowledge and proper examination contributed to development of these cancers; however, they acknowledge that the incidence of vulval cancer and its precursor lesions is low and screening the general population is not recommended.
Vulval cytology
Cervical cytology in organised screening programmes has had a huge effect on the incidence of cervical cancer. Although it is not diagnostic of preinvasive disease, it provides a trigger for further investigation. Whether vulval cytology is useful is debatable, and various techniques have been explored, with varying results.
Dennerstein described the use of vulval cytology in 1968 using a scalpel to scrape the surface cells from the vulva. Samples were taken from women with squamous and basal cell carcinomas of the vulva, vulval intraepithelial neoplasia (VIN) and Paget’s disease, and were found to give sufficient sampling for diagnosis. Dennerstein has since argued that subsequent studies with lower success rates used poor collection techniques, and that scalpel skin scrapings are reliable.
Nauth and Schilke correlated exfoliative cytology with histology using moistened cotton buds on samples from 464 women, including 111 histologically proven malignant lesions. They found only half of these exfoliated tumour cells through the superficial layer of the vulva sufficient for cytological diagnosis.
In a study by Nauth and Böger, 58% of smears showed dyskeratotic cells in precancerous lesions and 66% in malignant conditions. They suggested that, although cytology should not replace biopsy, it may provide a reason for taking an early biopsy. Bae-Jump et al. used a spatula to sample the entire mucosal surface of the vulva and subsequent smearing onto a glass slide. They correlated cytology using Bethesda criteria and concomitant histology on 50 patients in an attempt to use smears to prevent repeated biopsy. Only 32% of the biopsy-proven vulval VIN or vulval carcinoma had a positive smear. Negative correlation improved, however, with 96% of benign lesions having a negative smear. Given that the women with biopsy-proven disease all had visible lesions, and that the smear was not accurate in women with an increased likelihood of disease, they suggested that its usefulness as a screening test in low-risk patients was doubtful. The suggestion that the results were hampered by air drying and the lack of mucus in the vulva (indicating that that smears were difficult to fix) was not borne out by the six women in the study who underwent ThinPrep smears, as half of these still had a negative smear with biopsy-proven high grade VIN.
Promising results have been produced with the use of a brushing technique and cytospin analysis. Levine et al. used a nylon brush rotated over the vulva 20 times and subsequently placed directly into tissue culture fluid. Twenty-three women were sampled, of which 22 had a visible vulval lesion. The pathologist was blinded to the clinical diagnosis, but correctly identified VIN in 91% of cases, and identified the one patient with an invasive squamous cell carcinoma. The 15 women with no dyskaryotic cells on cytology did not show VIN on subsequent histology; however, conditions such as lichen sclerosus or lichen simplex could not be diagnosed specifically. The sensitivity was 100% with no false negatives and the investigators suggested this as a useful method for distinguishing between neoplastic and non-neoplastic vulval disease.
A pilot study using a new type of brush, the Vulvabrush ® , compared cytology and histology. It showed a high sensitivity (96.6%) and negative predictive value (90.9%) when differentiating between benign and malignant lesions. Cellularity of the smears were improved by 82% when the brush was moistened.
The use of vulval cytology obtained from existing visible lesions will not replace the need for biopsy and is unlikely to produce reliable results as a screening tool when the vulva appears normal. The use of cytology on a keratinised surface cannot be as accurate as a full thickness biopsy of vulval skin, and histology should be regarded as the gold standard.
It has been argued that vulval cytology is a useful adjunct in managing vulval disease, although it does not have the same effect as in cervical disease, in which the preinvasive lesion may often be asymptomatic and not visible to the naked eye.
Examination of the vulva, particularly in primary care, and appropriate referral and biopsy remains the management of choice for identifying vulval skin disorders, and is not likely to be replaced by vulval cytology as a screening test.
Screening in primary care
Self-examination
The Vulval Health Awareness Campaign, launched 10 years ago, promotes vulva health awareness and provides a dedicated helpline for women with any vulval disorder in the UK. In collaboration with the National Lichen Sclerosus Support Group and Vulval Pain Society, it encourages self-examination once a month between periods, and provides information on the changes to look for in vulval skin. Education of the public and health professionals is important. A former President of the International Society for the Study of Vulvovaginal Disease commented: ‘Perhaps the role of the GP to assess and refer on selected cases will act as the ‘gate keeper’, but only if we continue to educate medical students and GP trainees so they are competent in screening, examination and selective referral.’ Inspection of the vulva, as a basic level of screening, can also take place when women attend for cervical smears in primary care; however, this must be coupled with increased awareness by women and clinicians.
Sixty years ago, clinician and patient delay was reported in the diagnosis of vulval cancer, more so than other gynaecological malignancies, and things have not changed. Women who present with vulval cancer have often had symptoms for many years, and evidence shows that delay in diagnosis is common. In a retrospective review of 102 women with vulval cancer, 31% had more than three consultations relating to vulval symptoms more than 6 months before the diagnosis of cancer. Women in whom a biopsy was carried out were more likely to present with an earlier stage cancer. Ninety-four per cent had symptoms of chronic irritation and 85% had abnormal skin adjacent to the tumour. The investigators concluded that suboptimal primary and secondary care and the failure of knowledge and proper examination contributed to development of these cancers; however, they acknowledge that the incidence of vulval cancer and its precursor lesions is low and screening the general population is not recommended.
Vulval cytology
Cervical cytology in organised screening programmes has had a huge effect on the incidence of cervical cancer. Although it is not diagnostic of preinvasive disease, it provides a trigger for further investigation. Whether vulval cytology is useful is debatable, and various techniques have been explored, with varying results.
Dennerstein described the use of vulval cytology in 1968 using a scalpel to scrape the surface cells from the vulva. Samples were taken from women with squamous and basal cell carcinomas of the vulva, vulval intraepithelial neoplasia (VIN) and Paget’s disease, and were found to give sufficient sampling for diagnosis. Dennerstein has since argued that subsequent studies with lower success rates used poor collection techniques, and that scalpel skin scrapings are reliable.
Nauth and Schilke correlated exfoliative cytology with histology using moistened cotton buds on samples from 464 women, including 111 histologically proven malignant lesions. They found only half of these exfoliated tumour cells through the superficial layer of the vulva sufficient for cytological diagnosis.
In a study by Nauth and Böger, 58% of smears showed dyskeratotic cells in precancerous lesions and 66% in malignant conditions. They suggested that, although cytology should not replace biopsy, it may provide a reason for taking an early biopsy. Bae-Jump et al. used a spatula to sample the entire mucosal surface of the vulva and subsequent smearing onto a glass slide. They correlated cytology using Bethesda criteria and concomitant histology on 50 patients in an attempt to use smears to prevent repeated biopsy. Only 32% of the biopsy-proven vulval VIN or vulval carcinoma had a positive smear. Negative correlation improved, however, with 96% of benign lesions having a negative smear. Given that the women with biopsy-proven disease all had visible lesions, and that the smear was not accurate in women with an increased likelihood of disease, they suggested that its usefulness as a screening test in low-risk patients was doubtful. The suggestion that the results were hampered by air drying and the lack of mucus in the vulva (indicating that that smears were difficult to fix) was not borne out by the six women in the study who underwent ThinPrep smears, as half of these still had a negative smear with biopsy-proven high grade VIN.
Promising results have been produced with the use of a brushing technique and cytospin analysis. Levine et al. used a nylon brush rotated over the vulva 20 times and subsequently placed directly into tissue culture fluid. Twenty-three women were sampled, of which 22 had a visible vulval lesion. The pathologist was blinded to the clinical diagnosis, but correctly identified VIN in 91% of cases, and identified the one patient with an invasive squamous cell carcinoma. The 15 women with no dyskaryotic cells on cytology did not show VIN on subsequent histology; however, conditions such as lichen sclerosus or lichen simplex could not be diagnosed specifically. The sensitivity was 100% with no false negatives and the investigators suggested this as a useful method for distinguishing between neoplastic and non-neoplastic vulval disease.
A pilot study using a new type of brush, the Vulvabrush ® , compared cytology and histology. It showed a high sensitivity (96.6%) and negative predictive value (90.9%) when differentiating between benign and malignant lesions. Cellularity of the smears were improved by 82% when the brush was moistened.
The use of vulval cytology obtained from existing visible lesions will not replace the need for biopsy and is unlikely to produce reliable results as a screening tool when the vulva appears normal. The use of cytology on a keratinised surface cannot be as accurate as a full thickness biopsy of vulval skin, and histology should be regarded as the gold standard.
It has been argued that vulval cytology is a useful adjunct in managing vulval disease, although it does not have the same effect as in cervical disease, in which the preinvasive lesion may often be asymptomatic and not visible to the naked eye.
Examination of the vulva, particularly in primary care, and appropriate referral and biopsy remains the management of choice for identifying vulval skin disorders, and is not likely to be replaced by vulval cytology as a screening test.
Vulvoscopy
Colposcopy has a vital role in the screening of cervical abnormalities, however can we assume that colposcopic examination of the vulva, known as vulvoscopy, is as important in the assessment of vulval disease? Recent guidelines from the Royal Society of Obstetricians and Gynaecologists have suggested that this is not generally the case, but colposcopy should be available for women with VIN.
The use of a colposcope and staining of vulval lesions has been recommended in numerous studies over the years, with some investigators considering its use mandatory. Other opinions vary, however, from being uncertain whether vulvoscopy has the same predictive value as colposcopy to the feeling that it adds little or nothing and that its use is a ‘result of misinterpretation and inappropriate transfer of gynecologic and colposcopic knowledge from the cervix to the vulva’.
Routine use of vulvoscopy may not add to clinical assessment alone; however, it is of value in the examination of women with preinvasive disease. It also offers the advantages of improved light, optimal patient positioning and magnification for siting biopsies.
The use of stains has been assessed more objectively than vulvoscopy itself. Acetic acid has been used with the assumption that the spectrum of VIN follows the same pattern as the spectrum of CIN, and that high-grade abnormalities will stain more readily. We now know, however, that HPV-related usual type VIN does not follow this continuum of disease, and that VIN1 as a term has been disregarded, as its pathological diagnosis is unreliable. Acetowhite is more apparent on mucous membranes, and the vulval epithelial thickness and cornification theoretically would lead to a less prominent effect on the vulva.
Studies specifically focusing on the use of 5% acetic acid on the vulva do not seem to support its use as either a diagnostic or screening test.
Jonsson et al. evaluated acetowhite changes on the vulva as a predictor of subclinical HPV infection in a population-based study of 535 women. Sensitivity was only 44%, specificity 68%, with a positive predictive value of only 26%. Another study of normal women without vulval symptoms, found that 30% showed significant acetowhite staining on normal vulval epithelium outside of the vestibule, with all women staining positive within the vestibule.
The use of toluidine blue was first described nearly 50 years ago and used as a screening tool for vulval cancer in order to select a biopsy site. The investigators reported some false–positive results when the vulva stained blue, but no cancer or intraepithelial disease was found, these women mainly had other inflammatory conditions. No false–negative results were reported, as all invasive and preinvasive disease stained positive with toluidine blue.
The addition of colposcopy to the use of toluidine blue was studied when 1071 women without vulval lesions were screened as part of family planning and cancer screening. One-third of the cohort was aged 35–44 years, with the remainder older than 45 years. A high level of false–positive results and a low incidence of premalignancy (only 0.56%) were reported, and therefore was not useful as screening for the general population.
More recently, a study has investigated whether toluidine blue can be used to distinguish VIN from other disorders. Ninety-six women in a vulval clinic underwent vulvoscopy, staining with 1% toluidine blue and vulval biopsy, and the degree of staining was assessed. Vulval intraepithelial neoplasia was classified in this study under the old grading system of 1, 2 and 3. They found 92% sensitivity, 88% specificity and 96% negative predictive value for VIN3. The authors suggest that this test is reliable for separating VIN3 from hyperplastic non-neoplastic areas and helps in selecting a biopsy site. This is in contrast to a previous study using colposcopy, toluidine blue and biopsy on 93 women with various vulval lesions. When combining pre-invasive and invasive disease, the false positive rate was 26.9% and the false negative rate 37.5%. For VIN alone, the false negative rate was a remarkable 85.8%.
Broen et al. also noticed that leukoplakia did not stain blue and that thickened plaques had variable uptake caused by the layer of overlying keratin. Hyperkeratosis and hyperplastic skin changes are more likely to be associated with lichen sclerosus that progresses to cancer, and many cases historically assumed to be lichen sclerosus may well be differentiated VIN, with a higher malignant potential than usual type VIN, and therefore a negative toluidine blue test may not be reassuring.
There is no evidence that vulvoscopy, as a screening tool for identifying vulval disorders, is effective, but there is a role for judicious use, particularly where early invasion is suspected and as a tool in the follow up of complicated women with distorted vulvas from multiple previous treatments. Use of acetic acid on the vulva is non-specific, unhelpful clinically and may cause the patient discomfort and should not be used routinely. Toluidine blue may be helpful in delineating lesions more clearly, but should not be assumed to be diagnostic, and often these lesions will be visible without additional staining.
Given the risk of multicentric disease in women with VIN, examination of the vagina and cervix should be remembered, in which case a colposcope and subsequent use of acetic acid is justified and important.
In conclusion no systematic screening is available for vulval cancer, nor are there reliable screening methods for identifying malignant precursors. The identification of these conditions, therefore, relies on the reporting of vulval symptoms by the patient, and the knowledge and clinical acumen of the healthcare professional.
Follow up of vulval skin disorders
Routine follow up has been part of medical management for years. In times of limited resources, however, we should critically examine available evidence to determine the benefits.
The purpose of follow up of women with vulval disease is multifactorial. The clinician can evaluate symptom control and treatment compliance, and chart the course of the disease. Identification of early malignant change and subsequent management may prevent unnecessary morbidity and the woman is reassured by regular visits offering physical and psychosexual support. A multidisciplinary vulval clinic offers these services with the benefit of several skilled clinicians, with adequate examination and biopsy facilities. Although this sounds ideal, this arrangement is not necessary for all women, and a combination of patient-initiated follow up and attendance in primary care may be appropriate for a large number of women. The question lies in whether we can identify who requires specialist follow up and who is at risk of poorly controlled disease or malignant progression.
Lichen sclerosus
The prevalence of lichen sclerosus is debatable, with estimates between 1 in 300 to 1 in 1000 women ; however, one study has suggested that, in the general elderly population, this may rise to 1 in 30.
No randomised-controlled trials of regimens have been published, although recently several institutions have issued consensus guidelines.
Follow up of lichen sclerosus aims to monitor control of patient symptoms, identify and prevent malignancy, scarring and loss of function. If possible, it aims to identify groups at greater risk, and then tailored follow-up protocols may be possible.
The widely quoted risk of malignancy is 3–5%, with an average duration of antecedent disease of 10 years. Lichen sclerosus has been found adjacent to two-thirds of vulval cancers.
Increasingly, it has been recognised that all women with lichen sclerosus do not have the same risk of malignancy and that lichen sclerosus can be divided into simple atrophic lichen sclerosus with lichenification and lichen sclerosus with differentiated VIN; therefore, follow up could be tailored.
The use of ultrapotent topical steroids has revolutionised the treatment of lichen sclerosus, with symptom improvement in up to 96% of women. Sixty-six per cent became symptom free and a further 30% had a partial response. Skin changes were completely reversed in 20% of women, but six women developed a cancer. Another prospective study over 20 years showed that the chance of disease remission is more likely in younger women. Remission at 3 years was 72% in women younger than 50 years, 23% in women aged between 50 and 70 years, and 0% in women over the age of 70 years. The incidence of relapse of the disease, despite treatment, was 50% at 16 months and 84% at 4 years, with no differences between the age groups. Eight cancers developed in women with untreated or irregularly treated lichen sclerosus, suggesting that regular treatment with a potent steroid may protect against malignancy, although numbers were too small to be significant. Similar findings of carcinoma in undertreated women have been reported, emphasising the role of follow up ensuring treatment compliance.
Treatment compliance has also been suggested as a reason for long-term follow up in an Australian review. Clobetasol propionate is not available in Australia, and standard practice is a less potent topical steroid. The investigators found a 98% symptom remission in women compliant with treatment, which reduced to 75% if not compliant. Women were reviewed at 6 weeks, 3 and 6 months and then annually. Progression of scarring was not seen in the compliant women, but occurred in 35% of non-compliant women. In the compliant group, no women had malignant progression, whereas five women (11% of the non-compliant group) developed vulval carcinoma. In this study, long-term treatment prevented scarring, and the investigators suggested that long-term follow up is particularly important in sexually active women.
In an attempt to identify women at risk, investigators of a study in New Zealand compared 46 women with lichen sclerosus who had developed cancer with 213 who had not. They found that symptoms, duration of symptoms and signs of loss of architecture were not predictive; however, age and the presence of hyperplasia were independently associated with carcinoma. Eighty-seven per cent of the women with carcinoma were over 60 years of age, and they suggested that routine follow up should be considered in this age group. The proposal that vulval cancer is potentially preventable as it tends to arise in visible lesions is valid, and highlights the importance of good examination. It is likely that the areas of hyperplasia are differentiated VIN and, the investigators warned that the efficacy of topical steroids in symptom control can give rise to false reassurance if there are clinical hyperplastic changes, and these areas should be excised and closely followed up. Recent guidelines have suggested groups that require specialist follow up ( Table 1 ).
| Difficulty with symptom control. |
| Clinical evidence of localised skin thickening. |
| Previous vulval squamous cell carcinoma. |
| Previous treatment for usual or differentiated type vulval intraepithelial neoplasia. |
| Pathological concern of differentiated vulval intraepithelial neoplasia on biopsy, even if not diagnostic. |
| Use of more than 30 g steroid in 6 months. |
The International Society for the Study of Vulvovaginal Disease consensus document defines a specialist as a consultant gynaecologist, dermatologist or genito-urinary physician who has had additional and dedicated training in managing vulval disease.
Other guidelines are in agreement. The British Association of Dermatologists suggest secondary care should be reserved for women with complicated lichen sclerosus, and that women with uncomplicated disease should be seen at 3 and 6 months and then discharged to primary care, accompanied with written instructions and reasons for re-referral. They must also undergo annual review. Unfortunately, this advice is not always followed. One study has shown that 1 year after discharge from a vulval clinic, 38% had not seen their GP for follow up and, of those that had been seen, 17% were not examined. The American College of Obstetricians and Gynecologists suggest annual review if disease is controlled, and more frequently if not.
Lichen sclerosus in children
Lichen sclerosus has bimodal incidence, with older women the most affected, but a second peak occurring in childhood. Suggested prevalence is 1 in 900, and it may be unreported or recognised.
Earlier research suggested that lichen sclerosus resolved at puberty, but this has now been contested. This obviously has implications for follow up. The natural history and long-term outcome, including risk of malignancy of lichen sclerosus progressing from child to adulthood, is unclear, as most studies use different treatment regimens, have limited follow up past puberty, and teenage girls are lost to follow up because of embarrassment.
Women may improve symptomatically, but lichen sclerosus does not resolve at puberty. In a UK study of 75 prepubescent girls, 21 were followed after puberty. Seventy-five per cent reported an improvement in their symptoms, but one-half still had sufficient pruritus to require regular treatment. One-quarter had complete resolution of physical signs, but 75% still had pallor, atrophy, labial resorption and purpura. The investigators suggest that these signs do not equate with disease activity, and that if follow up is discontinued, particularly in asymptomatic cases, the continuing disease activity will perpetuate the risk of cancer later in life. Included in this study is a woman with difficult to control lichen sclerosus in childhood, which had apparently resolved at puberty who later died of metastatic vulval cancer aged 32 years.
Similar findings were reported in Australia, with 75% (nine out of 12 women) of their cohort still having active disease after puberty and continuing to require maintenance therapy after menarche. Women who have achieved remission before puberty, remained so 3 years later. Fifty per cent experienced alteration in vulval architecture, albeit without the introital stenosis often seen in older women. Some cases of children as young as 3 years have presented with labial fusion.
Concerns have been raised about the use of ultrapotent steroids in children; however, evidence supports their use as safe and effective, and that early aggressive treatment will usually lead to improvement within 3 months.
Childhood lichen sclerosus should be managed in specialist clinics, and a good relationship should be built with the child and parents early in childhood. Long-term follow up every 6–12 months is appropriate, and this should continue into adulthood and beyond. The risk of girls becoming lost to follow up is particularly real in this group, and so honest and realistic advice should be given to parents to improve the chances of teenagers understanding their condition and continuing to attend annually.
Vulval intraepithelial neoplasia
Vulval intraepithelial neoplasia can be classified into two types: usual type VIN and differentiated VIN. Each have different clinical features and risk of progression, and two separate pathways leading to vulval cancer. Vulval intraepithelial neoplasia was originally assumed to follow the same pattern as cervical intraepithelial neoplasia of the cervix and so was graded in the same way: VIN1, 2 and 3. It is now understood, however, that the continuum of disease is different in the vulva, and this grading, therefore, is inappropriate. It has been shown that VIN1 is not reproducible on pathological diagnosis and is thought to be merely an expression of HPV infection with no malignant potential. Inter-observer variation was demonstrated as only 5% agreement of opinion was obtained with VIN1 specimens, whereas concordance increased to 75% for VIN2 and 3 ( Table 2 ).
