Diagnosis of female external genital warts: reality and prospects

The most common manifestations of human papillomavirus infection (HPV) in the genital area are exophytic condylomas. In the Russian Federation, official registration of HPV manifested as genital warts (GW) started in 1993 by order of the Ministry of Health of the Russian Federation No. 286 dated 12/7/1993. From 2004 to 2014, the proportion of GW in the structure of sexually transmitted infections increased from 6% to 11% [1-3]. In 2014, the incidence of anogenital warts in the Russian Federation and Moscow was 21.8 per 100,000 and 33.6 per 100,000, respectively [2]. Based on estimates from a systematic review, the global annual incidence of GW (including new and recurrent cases) ranges from 160 to 289 cases per 100,000, with an average of 194.5 per 100,000, and the average annual incidence of new GW cases was 120.5 per 100,000 among women. The rates were highest among women aged 24 years [4]. Low-oncogenic HPV types 6 and 11 types account for 90 - 100% of genital warts.

GWs are most often diagnosed based on their clinical appearance. They typically present as single or multiple finger-shaped protrusions on the skin or mucosa of the external genitalia with a typical multicolored or loop-like pattern [5]. GWs are usually located in areas of maceration, including labia minora, vagina, cervix, external urethral orifice, anus, and skin. In some women, GWs are asymptomatic, while in others, they may cause psychological discomfort due to a cosmetic defect, cancerphobia, and even negatively affect patients’ sex life [6–8]. Although the diagnosis of GWs can be made by visual inspection during a gynecological examination, most often they are diagnosed during vulvoscopy and colposcopy [7, 9, 10]. Colposcopy is considered as the most sensitive method for detecting subclinical forms of HPV [10, 11]. Colposcopy carries the risk of false-positive (from 4 to 33%) or false-negative (from 15 to 62%) results, which limits its diagnostic performance [12]. Histologic examination of biopsy specimens is considered a gold standard in the diagnosis of subclinical and clinical manifestations of HPV, but its utility is limited due to technical difficulties and invasiveness. Therefore, the search for new diagnostic tools providing advantages over the above modalities is promising and relevant.

Today, one of the new promising non-invasive imaging technologies for the diagnosis of GW is a high-frequency ultrasound (HF-US) of the skin and mucosa of the external genitalia. Skin ultrasound has long been inaccessible due to the low resolution of images acquired with probes of a 3-10 MHz frequency range [7, 13]. The first attempt to use ultrasonography in dermatology was reported in 1979. Since then, continuous medical-technological advances with the advent of high frequency and resolution equipment and sensors have made possible clear visualization of different layers of the skin.

In the early 90s of the twentieth century, German company Taberna Pro Medium initiated large-scale production of ultrasonic equipment for skin investigation. Piezo crystals generating ultrasonic vibrations of much higher frequency than in traditional ultrasonic scanners - over 20 MHz - became available. Such characteristics of piezo crystals allowed a high-resolution image of the skin structures to be obtained [13]. The principle of operation of the high-frequency ultrasonic scanner is the following: the generator produces a short electrical pulse that is converted by a piezo crystal into an ultrasonic signal with a frequency of 22 to 100 MHz, depending on the sensors used. When transmitted to the skin, these signals are reflected, absorbed, and dispersed due to the heterogeneity of the skin tissues (cellular structures, the intercellular space, etc.). The reflected echo-signals get reflected back and transformed again into electrical impulses that are converted into a visible image using a specialized computer. This imaging modality allows the examination of the skin at different time intervals with documented fixation of all the features, including quantitative parameters such as the thickness of the epidermis and dermis, the size of lesions in the area of interest, and the acoustic density of the epidermis, dermis and subcutaneous tissue [7, 13-15]. In the ultrasound image, the epidermis is presented as a hypoechoic homogeneous line with regular contours. The dermis has heterogeneous echogenicity. This method is widely used for noninvasive assessment and monitoring of the treatment effectiveness of various skin diseases, including dermatosis, systemic scleroderma, melanoma [16–19]. When used for diagnosis of GW, HF-US allows visualization of changes in the internal structure of the skin and mucosa visible as homogeneous hypoechoic or anechoic lesions with pathognomonic echo signs [7, 20].

This study aimed to optimize the diagnosis of vulvar GWs by identifying the diagnostic potential of HF-US of the skin and vulvar mucosa in comparison with the vulvoscopy and morphological study.

Material and methods

The study comprised 187 women who sought medical care at the Department of Aesthetic Gynecology and Rehabilitation and the Gynecological Department of Rehabilitation, V.I. Kulakov NMRC for OG&P from October 2013 to September 2017. We selected 159 patients who met the inclusion criteria and signed informed consent to take part in the study. Of them, 57% (n = 90) had GWs of the external genitalia and were included in the study group, while the remaining 43% (n = 69) of women without visible manifestations of GW constituted the control group.

The diagnosis of GW was based on clinical and visual assessment, vulvoscopy with the acetic acid test, and morphological study of GW biopsy specimens taken from 90 patients. Baseline clinical assessment included general clinical and gynecological examination, patients’ history taking, HF-US of the skin and mucosa of the external genitalia, and advanced colposcopy. Skin HF-US was performed with a DUB - Digital Ultraschall Bildsystem - tpm machine with DUB-Skin Scan software. Evaluation of the skin structure was carried out using contact ultrasonic imaging. The study used a scanning transducer of the device, generating an ultrasonic wave with a frequency of 22-75 MHz. The patients in the study group also underwent GW photo registration, vulvoscopy after the acetic acid test, and morphological examination of the GW biopsy tissue. A GW biopsy was performed when GW size allowed sufficient material for histological examination (≥3 mm) to be obtained; simultaneously, surgical GW destruction was carried out using a DermoPunch skin biopsy instrument.

Statistical analysis and graphical display of data were performed using IBM SPSS software version 22 and Microsoft Excel 2013. All obtained quantitative data were analyzed by standard statistical methods. Quantitative variables were expressed as means (M) and standard deviation (SD) and presented as M (SD). Qualitative variables were summarized as counts (n) and percentages (%). Continuous variables showing normal distribution were compared between two groups using the Student’s t-test and among three groups with one-way analysis of variance (ANOVA). Categorical variables were compared by the Chi-squared test and odds ratios (OR). Differences between the groups were considered statistically significant at p<0.05.

Sensitivity, specificity, and positive and negative predictive values for detecting GWs were calculated, as well as false-positive and false-negative results. Receiver operating characteristic (ROC) curve analysis was performed by evaluating the area under the ROC curve (AUC) to assess the diagnostic potential of vulvoscopy and HF-US.

Results

Women of reproductive age (25-35 years) constituted 51% of the whole study cohort. The mean age of women in the study and control group was 27.3 (2.2) * and 25.4 (1.2) * years, respectively [data are presented as mean (M) and standard deviation (SD)]. There were no statistically significant differences between the groups regarding age (p = 0.09). Anthropometric data, such as height, weight, body mass index, menstrual history including the age of menarche, duration and regularity of menstruation, the use of contraceptives also did not differ significantly between the two groups. Most of the study participants were women of reproductive age who had a regular menstrual cycle. Early-onset sexual activity (16 years) was reported by 18.9% and 7.24% of women in the study and control group, respectively. Women in the control group statistically significantly later entered into sexual relations than women in the study group (p = 0.03). There were significant differences in the number of sexual partners: about ¾ (80%) of the women in the study group had more than one sexual partner (p = 0.018).

Results of extended colposcopy/vulvoscopy

All patients (n = 159) underwent extended colposcopy/vulvoscopy; 102 (64%) and 14 (9%) patients had a normal and abnormal colposcopic pattern, respectively. Patients in the study group were statistically significantly more likely to have cervical epithelial changes seen at colposcopy (p <0.05). Mild changes were registered in 10 (11%) and 1 (1.4%) patient in the study and control group. Pronounced changes in the epithelium were detected only in 3 (3.3%) patients in the study group.

Along with extended colposcopy, patients underwent vulvoscopy. When lesions suggestive of GW were detected, 3% acetic acid was applied to the surface to improve the visualization of abnormal areas. At vulvoscopy, the diagnosis of GW was confirmed in 90 (100%) in the study group patients, of whom 61 (68%) had lesions of both the vagina and vulva and 5 (5.5%) had cervical condylomas. Vulvoscopy revealed subclinical GW lesions that were invisible to the naked eye in 2 (3%) patients in the control group (Table 1).

Typical vulvoscopic findings in the study group patients were whitish epithelial lesions of an irregular shape with finger-like protrusions. The most important diagnostic criterion was the presence of the normal capillary network in the protrusions, which was detected by application of 3% acetic acid to improve the visualization. Application of acetic acid results in GW vessels’ vasoconstriction, and they become less visible, which gives the site with GW a characteristic whitish appearance. We systematized the results and compared HF-US and morphological findings.

The sensitivity and specificity of vulvoscopy for the detection of vulvar GW were 94.4% (95% CI 89.7-99.2) and 90% (95% CI 82.7-97), respectively. Positive and negative predictive values were 92.4 (95% CI 87 - 97.8) and 92.5 (95% CI 86.2 - 98.5), respectively. The predictive accuracy of the method was 92.4% (95% CI 88.4–96.6). False-positive and false-negative results were observed in 4.4% and 3.2% of cases, respectively.

Findings of HF-US of the vulvar skin and mucosa in the study participants

All patients (n = 159) underwent HF-US examination of the skin using a Digital Ultraschall Bildsystem ultrasound scanner (DUB22-75, Taberna pro Medicum, Germany) with DUB-SkinScan software. The examination began with a 22 MHz transducer, since it provides a large visualization area, allowing evaluation of fine epidermal and dermal structures with the ultrasonic penetration depth of 6-7 mm. During the study, the frequency was increased to a 75 MHz to obtain the clearest image of the skin layer of interest. In this frequency range, the penetrating depth is minimal amounting to 0.13–3 mm, which provides higher image resolution and a detailed analysis of the epidermal and basement membrane structures. The assessment of the lesions included its comparison with the normal skin of the symmetric opposite location or adjacent areas.

At HF-US of the skin and mucosa of the external genitalia in the study group patients (n = 90) with visible GWs and the control group (n = 69) without visible GWs, the main sonographic signs of GW were an irregular external contour of the lesion in the form of waves and ridges, clear demarcation between stroma and the external region, decrease in the ultrasonic density of the lesion and reduced ultrasonic signal in the underlying tissue (Table 2, Fig. 1, 2). When examining patients with GW with HF-US in this area, we identified signs typical for these conditions including homogeneous hypo- or anechoic lesions with an irregular external contour in the form of pyramids, waves or crests, clear demarcation from the underlying tissues and a decrease in ultrasonic density of the underlying tissues. The combinations of these signs were diagnostic for GW.

The results of HF-US showed that in the study group 3 patients had no sonographic signs of GW, while in the control group 7 patients had various combinations of sonographic signs: 5 patients had a combination of 3 sonographic signs listed above, and two women had two sonographic signs of GW.

In three study group patients, the diagnosis of GW was not confirmed by morphologic studies; one of them had molluscum contagiosum, and in the remaining 2 cases, no morphological changes were detected. It turned out that in these patients, small hymen fragments were confused with GW. False-negative results seem to be associated with the subjectivity of the GW evaluation.

Therefore, according to the findings of HF-US of the skin and mucosa, a combination of any three sonographic signs of GW is diagnostic of GW. The sensitivity and specificity of HF-US for the detection of vulvar GW were 100% and 97% (95% CI 93.1–101), respectively. Positive and negative predictive values were 97.8% (95% CI 95–101) and 100%, respectively. The predictive accuracy of the method was 98.8% (95% CI 97–101). False-positive results were observed in 2.9% of patients, and there were no false-negative results. In diagnosing vulvar GW, HF-US of the vulvar skin and mucosa is superior to vulvoscopy.

The results of the morphological assessment of GW biopsy tissue in the study group patients

GW tissue samples were biopsied simultaneously with GW surgical destruction using the DermoPunch skin biopsy tool. The device allows a more complete surgical excision of GW than that performed by a surgical scalpel, as it entirely removes even minor lesions, together with the stroma in the deep submucosal layer, which is not possible with standard surgical excision. All study group participants (n = 90) underwent GW tissue biopsy, followed by a morphological study. Specific morphological signs of GW were found in 87 (96.6%) patients; in two patients no abnormality was found (myrtle nipples - hymen fragments were confused with GW); in one patient a keratinizing papilloma was detected. The morphological study is considered as the most effective method in the diagnosis of GW.

Discussion

Dermatologic ultrasound imaging is a “fast” and comparatively cheap technique for in vivo skin examination. The real-time visualization, a high-resolution view of the skin structures, practically side-effect-free non-ionizing media, reproducibility, and the absence of contraindications are indisputable advantages of cutaneous sonography.

Currently, HF-US is a promising method for assessing and studying the skin in a wide variety of diseases. HF-US for diagnosing hidradenitis has been reported. Analysis of the sonographic image revealed numerous subclinical lesions that could not be reliably discriminated through palpation only [21, 22]. Gupta A.K., Turnbull D.H. et al. [23] described distinct characteristic HF-US signs of psoriasis. Jambusaria-Pahlajani A., Schmults C.D. [24] reported that the sensitivity and specificity of HF-US in evaluating the structure of the skin tumors were 80 and 84%, respectively.

Numerous studies confirm a strong correlation (correlation coefficient 88–96%) between ultrasound findings and histological data regarding the depth of tumor invasion in melanoma and basal cell carcinoma, which is especially important when planning the surgery. Measurement of thickness, invasion depth, and assessment of the borders of skin tumors may increase the number of radical treatments and improve long-term outcomes [25–27]. Our study also demonstrated that with HF-US provides the opportunity to determine the size, contour, structure, and penetration depth of GW lesions and further carry out rational destruction, which helps reduce the GW recurrence rate. Jasaitiene S., Valiukeviciene D. [27] described specific sonographic characteristics of large clusters of apoptotic or para-keratotic cells, as well as cell death detected by HF-US.

A high-frequency ultrasound image of healthy skin consists of three layers: 1. - highly echogenic entry echo; 2. - dermis –a layer rich in dispersed echoes of varying intensities, and 3. - subcutaneous tissue – an anechoic or low echogenicity layer, in which it is sometimes possible to observe the structures with enhanced echogenicity related to the presence of septae.

On the sonographic appearance of normal skin, the echogenicity of each layer depends on its main component that in the epidermis is the keratin; in the dermis, is the collagen; and in subcutaneous tissue, is the fat lobules and connective tissue fibers. The epidermis of healthy skin is presented as a hypoechoic line in the overall body skin, except the palmar and plantar areas, where the epidermis is visualized as a two-layer hyperechoic line. The thickness of the epithelium of unchanged areas is uniform. The epithelium is seen as the external hyperechoic and the underlying hypoechoic layer. The thickness of the entry echo ranged from 45 to 65 microns, the thickness of the underlying layer from 145 to 300 microns. The entire thickness of the epithelium along the scanning front is easy to visualize; the thickness of the epithelium of the unchanged areas is uniform and ranges from 190 to 350 microns. In our study, the sonographic appearance of GW had the following signs: the external contour of the epithelium is significantly deformed and extends outwards. The deformity has the form of “ridges,” pyramid-shaped pointed outgrowths, large waves, which are isolated or form conglomerates and clusters. The changed (in the projection of the lesion) epithelium is presented as significantly thickened compared with the unchanged areas. The height of the lesions ranged from 500 to 2500 microns. It is noteworthy that such “ridges, pyramid-shaped outgrowths were themselves again deformed due to the effect of the transducer on quite tender wart vegetations. We observed a weakening of the ultrasound signal in the deeper sections (light ultrasound shadow). These changes are associated with the dispersion of the echo signal due to the complex relief (vegetation) of the surface of the lesions. The external boundary of the lesion with the underlying connective tissue plate (stroma) may be even or somewhat deformed (“concave”) in the direction of the underlying tissues. Internal parts of the lesion have reduced echogenicity compared with the unchanged epithelium. The internal echo structure of the lesion is homogeneous.

In our study on HF-US of skin and mucosa of the external genitalia in patients of the study group (n = 90) with visible signs of GW and control group (n = 69) without visible GW, the main sonographic signs of GW were an irregular external contour of the lesion in the form of waves and ridges, clear demarcation between stroma and the external region, decrease in the ultrasonic density of the lesion and reduced ultrasonic signal in the underlying tissue. Thus, the identification of a combination of any 3 GW sonographic signs allows reliable verification of GW diagnosis, including the diagnosis of invisible subclinical GWs, which are the source of infection, and timely treatment initiation. HF-US allows a real-time estimation of lesion depth, which implies the possibility of its rational destruction to the required depth and thus reduces the number of recurrences. The sensitivity and specificity of HF-US in our study were higher than those of extended vulvoscopy and were 100 and 97%; the predictive accuracy of the method was 98.8%. The sensitivity and specificity of vulvoscopy were 94.4 and 90%, respectively, with the predictive accuracy of 92.4%. The sensitivity and specificity of the morphological study was 100%.

Conclusion

HF-US improves the effectiveness of non-invasive diagnosis of GWs, providing the opportunity to detect invisible subclinical lesions and measure the depth of visible GWs, which implies the possibility of their rational destruction to the required depth and thus reducing the number of recurrences. The advantages of the skin HF-US include non-invasiveness, safety, real-time visualization, the absence of contraindications, and the possibility of repeat examinations of any part of the skin with an analysis of the tissue microstructure in situ. The method does not require additional preparation or examination of the patient.

The RF patent for invention No. 2607956 dated January 11, 2017, has been granted.