COLPOSCOPIC DIFFERENTIATION OF ABNORMALITIES ARISING IN THE TRANSFORMATION ZONE

The Application of Artificial Intelligence-Assisted Colposcopy in a Tertiary Care Hospital within a Cervical Pathology Diagnostic Unit

The rising global incidence of cervical cancer is estimated to have affected more than 600,000 women, and nearly 350,000 women are predicted to have died from the disease in 2020 alone. Novel advances in cancer prevention, screening, diagnosis and treatment have all but reduced the burden of cervical cancer in developed nations. Unfortunately, cervical cancer is still the number one gynecological cancer globally. A limiting factor in managing cervical cancer globally is access to healthcare systems and trained medical personnel. Any methodology or procedure that may simplify or assist cervical cancer screening is desirable. Herein, we assess the use of artificial intelligence (AI)-assisted colposcopy in a tertiary hospital cervical diagnostic pathology unit. The study group consisted of 48 women (mean age 34) who were referred to the clinic for a routine colposcopy by their gynecologist. Cervical images were taken by an EVA-Visualcheck ^TM colposcope and run through an AI algorithm that gave real-time binary results of the cervical images as being either normal or abnormal. The primary endpoint of the study assessed the AI algorithm’s ability to correctly identify histopathology results of CIN2+ as being abnormal. A secondary endpoint was a comparison between the AI algorithm and the clinical assessment results. Overall, we saw lower sensitivity of AI (66.7%; 12/18) compared with the clinical assessment (100%; 18/18), and histopathology results as the gold standard. The positive predictive value (PPV) was comparable between AI (42.9%; 12/28) and the clinical assessment (41.8%; 18/43). The specificity, however, was higher in the AI algorithm (46.7%; 14/30) compared to the clinical assessment (16.7%; 5/30). Comparing the congruence between the AI algorithm and histopathology results showed agreement 54.2% of the time and disagreement 45.8% of the time. A trained colposcopist was in agreement 47.9% and disagreement 52.1% of the time. Assessing these results, there is currently no added benefit of using the AI algorithm as a tool of speeding up diagnosis. However, given the steady improvements in the AI field, we believe that AI-assisted colposcopy may be of use in the future.

Characteristic findings of high-grade cervical intraepithelial neoplasia or more on magnifying endoscopy with narrow band imaging

BACKGROUND

Colposcopy, which is a standard modality for diagnosing cervical intraepithelial neoplasia (CIN), can have limited accuracy owing to poor visibility. Flexible magnifying endoscopy with narrow band imaging (ME-NBI) has excellent diagnostic accuracy for early gastrointestinal neoplasms and is expected to be highly useful for CIN diagnosis. This study aimed to determine the characteristic findings and evaluate the diagnostic ability of ME-NBI for lesions ≥ CIN 3.

METHODS

A well-designed prospective diagnostic case series conducted at multiple tertiary-care centers. A total of 24 patients who underwent cervical conization with a preoperative diagnosis of high-grade squamous cell intraepithelial lesions (HSILs) or lesions ≥ CIN 3 were enrolled. Prior to conization, still images and video of ME-NBI were captured to investigate the cervical lesions. The images were reviewed based on histological examination of the resected specimens.

RESULTS

The NBI-ME images revealed the following abnormal findings: (1) light white epithelium (l-WE), (2) heavy white epithelium (h-WE), and (3) atypical intra-epithelial papillary capillary loop (IPCL). Pathological examination of the resected specimens confirmed cervical lesions ≥ CIN 3 in 21 patients. The ME-NBI findings were classified into four groups: l-WE, l-WE with atypical IPCL, h-WE, and h-WE with atypical IPCL, at rates of 0, 23.8, 9.5, and 66.7%, respectively. Additionally, all 3 patients with micro-invasive carcinoma showed a strong irregularity of IPCLs.

CONCLUSION

The lesions ≥ CIN 3 demonstrated characteristic ME-NBI findings of h-WE alone, or l-/h-WE with atypical micro-vessels. This study indicates that ME-NBI may have novel value for CIN diagnosis.

Colpocytological abnormalities in HIV infected and uninfected pregnant women: prevalence, persistence and progression

In this retrospective case-control study, we analyse data of 48 HIV-positive pregnant patients, versus a control group of 99 HIV-negative pregnant women, followed as outpatients by our department from 2009 to 2014. The aims of the study were to investigate the prevalence, persistence and progression of cervical squamous intraepithelial lesions (SIL) in each group and to correlate colpo-cytological lesions to the socio-demographic and clinical-laboratory findings in the HIV + pregnant women. In our study we observed that immunosuppression, HPV infection and vaginal coinfections were predictive of cervical lesions. Pap smear and colposcopy should be part of routine care for HIV-infected pregnant women because these lesions behave aggressively in these patients. Success of prevention depends on massive access of patients to screening. HAART reduces viral load and maintains CD4 count and can affect progression of SIL. Multidisciplinary services on the same site appear to be one promising strategy to improve compliance in patients. Impact Statement What is already known on this subject: Our study provided novel information on a highly vulnerable population of young HIV + pregnant women. What the results of this study add: We observed that immunosuppression, HPV infection and vaginal coinfections were predictive of cervical lesions remarkable with colposcopy. We could consider these important risk factors to evaluate to establish an appropriate strategy of management for these patients. What the implications are of these findings for clinical practice and/or further research: Association of the risk between SIL presence and HIV and HPV infection also deserves additional investigation. We believe that Pap smears and colposcopies should be part of the routine care for HIV-infected women because these lesions behave particularly aggressively in these patients.

ASCCP Colposcopy Standards: Role of Colposcopy, Benefits, Potential Harms, and Terminology for Colposcopic Practice

OBJECTIVES

The American Society for Colposcopy and Cervical Pathology Colposcopy Standards address the role of and approach to colposcopy and biopsy for cervical cancer prevention in the United States. Working Group 1 was tasked with defining the role of colposcopy, describing benefits and potential harms, and developing an official terminology.

METHODS

A systematic literature review was performed. A national survey of American Society for Colposcopy and Cervical Pathology members provided input on current terminology use. The 2011 International Federation for Cervical Pathology and Colposcopy terminology was used as a template and modified to fit colposcopic practice in the United States. For areas without data, expert consensus guided the recommendation. Draft recommendations were posted online for public comment and presented at an open session of the 2017 International Federation for Cervical Pathology and Colposcopy World Congress for further comment. All comments were considered for the final version.

RESULTS

Colposcopy is used in the evaluation of abnormal or inconclusive cervical cancer screening tests. Colposcopy aids the identification of cervical precancers that can be treated, and it allows for conservative management of abnormalities unlikely to progress. The potential harms of colposcopy include pain, psychological distress, and adverse effects of the procedure. A comprehensive colposcopy examination should include documentation of cervix visibility, squamocolumnar junction visibility, presence of acetowhitening, presence of a lesion(s), lesion(s) visibility, size and location of lesions, vascular changes, other features of lesion(s), and colposcopic impression. Minimum criteria for reporting include squamocolumnar junction visibility, presence of acetowhitening, presence of a lesion(s), and colposcopic impression.

CONCLUSIONS

A recommended terminology for use in US colposcopic practice was developed, with comprehensive and minimal criteria for reporting.

Accuracy of combinations of visual inspection using acetic acid or lugol iodine to detect cervical precancer: a meta-analysis

BACKGROUND

Visual inspection of the cervix with acetic acid (VIA) or with Lugol's iodine (VILI) have been evaluated for cervical cancer screening in developing countries.

OBJECTIVES

To assess the diagnostic accuracy and clinical utility of visual methods to detect cervical intraepithelial neoplasia grade 2+ (CIN2+) using: (1) VIA alone; (2) VILI alone; (3) co-testing; and (4) VILI as a triage test of a positive VIA result.

SEARCH STRATEGY

PubMed, EMBASE, and the Cochrane Library were searched up to May 2016.

SELECTION CRITERIA

All reports on the accuracy of VIA and VILI, or combinations of VIA/VILI, to detect CIN2+ were identified. Histology and colposcopy when no biopsy was taken were used as the reference standard.

DATA COLLECTION AND ANALYSIS

Selected studies were scored on methodological quality, and sensitivity and specificity were computed. Clinical utility was assessed from the positive predictive value (PPV) and the complement of the negative predictive value (cNPV).

MAIN RESULTS

We included 23 studies comprising 101 273 women. The pooled sensitivity and specificity of VILI was 88 and 86%, respectively. VILI was more sensitive, but not less specific, compared with VIA (relative sensitivity = 1.11; 95% confidence interval, 95% CI, 1.06-1.16; relative specificity = 0.98; 95% CI 0.95-1.01). Co-testing was hardly more sensitive, but significantly less specific, than VILI alone. VILI to triage VIA-POSITIVE women was not less sensitive, but more specific, compared with VIA alone (relative sensitivity = 0.98, 95% CI 0.96-1.01; relative specificity = 1.04, 95% CI 1.02-1.05). The average PPVs were low (range 11-16%), whereas the cNPV varied between 0.3% (VILI, co-testing) and 0.6% (triage).

CONCLUSIONS

Although imperfect, VILI alone appeared to be the most useful visual screening strategy.

TWEETABLE ABSTRACT

VILI alone seems to be the most useful visual screening test for cervical cancer screening.

New technologies and advances in colposcopic assessment

To have a good grasp of clinical colposcopy, it is necessary to understand the histopathologic structure of the normal and dysplastic cervical epithelium. Previous meta-analyses had indicated high overall sensitivity of colposcopy in detecting dysplastic lesions, but recent studies have suggested that the technique has much lower sensitivity in detecting high-grade intraepithelial neoplasia. The best practice in colposcopy relies on accurately taking a biopsy from the correct (i.e. most morphological abnormal) site, and by taking more than one biopsy, the sensitivity for detection of high-grade cervical intraepithelial neoplasia can be increased. Cytological screening programmes of proven and maintained high quality will enhance the predictive colposcopic accuracy for high-grade cervical intraepithelial neoplasia after referral. With the advent of computerised colposcopy and the Internet, digital imaging can be transmitted in real-time for instant viewing, facilitating distant consultation and education. This form of 'telemedicine' will allow family practice and remote areas to have access to colposcopy expertise. Of all the currently available technological adjuncts to colposcopy, spectroscopy devices have demonstrated relatively high sensitivities, and seem to have the best potential to become the technique of choice in future routine clinical practice in developed countries following the human papillomavirus vaccination. Other alternatives may need to be used in parts of the globe with high disease incidence and without organised screening or vaccination programmes. Opportunities remain for global collaboration in research, education and training to promote more effective and affordable cervical screening, and to enhance the skills of colposcopists worldwide.

Clinical study of quantitative diagnosis of early cervical cancer based on the classification of acetowhitening kinetics

A quantitative colposcopic imaging system for the diagnosis of early cervical cancer is evaluated in a clinical study. This imaging technology based on 3-D active stereo vision and motion tracking extracts diagnostic information from the kinetics of acetowhitening process measured from the cervix of human subjects in vivo. Acetowhitening kinetics measured from 137 cervical sites of 57 subjects are analyzed and classified using multivariate statistical algorithms. Cross-validation methods are used to evaluate the performance of the diagnostic algorithms. The results show that an algorithm for screening precancer produced 95% sensitivity (SE) and 96% specificity (SP) for discriminating normal and human papillomavirus (HPV)-infected tissues from cervical intraepithelial neoplasia (CIN) lesions. For a diagnostic algorithm, 91% SE and 90% SP are achieved for discriminating normal tissue, HPV infected tissue, and low-grade CIN lesions from high-grade CIN lesions. The results demonstrate that the quantitative colposcopic imaging system could provide objective screening and diagnostic information for early detection of cervical cancer.

Prise en charge des lésions malpighiennes intra-épithéliales de type CIN2 et CIN3 par vaporisation au laser

OBJECTIVES

This study was carried out over an 8-year period in order to evaluate the long-term effectiveness of laser CO2 vaporization in the treatment of squamous intraepithelial lesion of type CIN2 and CIN3.

MATERIALS AND METHODS

A retrospective study of 52 cases of cervical lesions of type CIN2 and CIN3 treated in first intention by laser CO2 vaporization was carried out at the hospital Jeanne-de-Flandre in CHRU of Lille from 1996 to 2003. This treatment was performed on only high-grade exo-cervical lesions, of small size (<2cm2), after a complete colposcopic examination.

RESULTS

Fifty-two patients were treated by first-intention laser vaporization only. Mean age was 29.4 years and 51.9% were nulliparous. At the first cyto-colposcopic control, there were 17 persistent lesions (32.7%). Among the 35 patients without persistent lesion, 29 achieved cure (absence of recurrence), 4 presented a recurrence and 2 were lost to follow-up.

CONCLUSION

The current data of the literature concerning the treatment by laser CO2 vaporization authorize application of this method for certain high-grade exocervical lesions after a complete colposcopic examination. This type of treatment remains less aggressive than a surgical treatment. The high rate of residual lesions in particular in the event of CIN3 can be due to an incomplete destruction of the lesion. Patients should thus be advised that monitoring is an integral part of the treatment. Laser vaporization could be limited to CIN1 and CIN2 lesions.

Cervical squamous intraepithelial lesions of low-grade in HIV-infected women: recurrence, persistence, and progression, in treated and untreated women

BACKGROUND

Human immunodeficiency virus (HIV)-infected patients are more predisposed than HIV-negative women to develop squamous intraepithelial lesions (SIL) of the uterine cervix, and cervical dysplasia may be of higher grade in HIV-positive women than in HIV-negative subjects, with more extensive and multi-centric involvement of the lower genital tract by human papillomavirus (HPV)-associated lesions. Moreover, recurrence and progression rate of cervical intraepithelial neoplasia (CIN) is particularly higher in immunocompromised women.

DESIGN

Retrospective case-control study of HIV-positive women and HIV-negative controls, all affected by low-grade SIL of the uterine cervix, treated by loop excision or followed-up without treatment. Correlation of progression and recurrence of SIL with HIV status and CD4+ count.

PATIENTS

From September 1990 to October 1997, 75 HIV-positive low-grade-SIL patients, 47 treated and 28 followed-up without treatment, and 75 HIV-negative low-grade-SIL controls, 45 treated and 30 followed-up.

RESULTS

Among treated patients, 17/47 (36.2%) HIV-positive and 5/45 (11.1%) controls had recurrence (P < 0.0101, O.R. = 4.53, 95% CI = 1.5-13.7), progression of untreated lesion was seen in 15/28 (53.6%) HIV-positive and 7/30 (23%) controls (P < 0.05, O.R. = 3.79, 95% CI = 1.23-11.69). The risk of recurrence or progression of low-grade SIL linked to HIV seropositivity is about 4-5 times higher in comparison with seronegative counterpart, matched for age, risk factors and lesion size. More significantly, considering the cut-off of 200 CD4+/mm(3) in HIV-positive women, 13/17 cases of recurrence (P < 0.05, O.R. = 4.88, 95% CI = 1.28-18.58) and 10/15 cases with progression (P < 0.05, O.R. = 6.67, 95% CI = 1.24-35.73) were immunocompromised (<200 CD4+/mm3), with a significant higher risk of recurrence or progression linked to immunodeficiency status. Considering time of progression or recurrence, during follow-up, Kaplan-Meier curves shows that HIV-positive status and immunodeficiency are correlated with more rapid evolution of cervical dysplasia and HPV-related lesions: comparison of recurrence in treated patients report P < 0.005 and progression in untreated P<0.05 (Mantel-Haenszel log-rank test).

CONCLUSIONS

Immunological status seems to be a determinant factor in prognosis of cervical SIL, HIV-positive women affected by this lesion, even if low-grade, need more aggressive management than the immunocompetent counterpart. Strict cytologic and colposcopic screening is recommended and CD4+ count and HPV-DNA testing may be useful risk indicators. Excisional procedures are preferred, while ablative treatments or wait and see policy may expose to some risk this type of population with poor compliance to follow-up.

The role of low-level magnification in visual inspection with acetic acid for the early detection of cervical neoplasia

Several studies have investigated the accuracy of naked eye visual inspection with acetic acid (VIA) in the early detection of cervical neoplasia. It is not clear whether low-level (2-4x) magnification (VIAM) can improve the sensitivity and specificity of VIA. The accuracy of both VIA and VIAM, provided by independent health workers, were evaluated in three cross-sectional studies involving 18,675 women aged 25-65 years in Kolkata and Mumbai in India. All screened women were investigated with colposcopy and biopsies were obtained based on colposcopy findings. The final disease status was based on the reference standard of histology (if biopsies had been taken) or colposcopy. Data from the studies were pooled to calculate the test characteristics for the detection of high-grade squamous intraepithelial lesions (HSIL). 14.1% and 14.2% were positive on testing with VIA and VIAM respectively. Two hundred twenty-nine were diagnosed with HSIL and 68 with invasive cancer. The pooled sensitivity, specificity, positive and negative predictive values for VIA in detecting high-grade squamous intraepithelial lesions (HSIL) were 60.3% (95% CI: 53.6-66.7), 86.8% (95% CI: 86.3-87.3), 5.9% (95% CI: 5.0-7.0), and 99.4% (95% CI: 99.2-99.5), respectively. The values were 64.2% (95% CI: 57.6-70.4), 86.8% (95% CI: 86.2-87.3), 6.3% (95% CI: 5.3-7.3) and 99.4% (95% CI: 99.3-99.6), respectively, for VIAM. Low-level magnification did not improve the test performance of naked eye visualization of acetic acid impregnated uterine cervix.

A clinical study of optical biopsy of the uterine cervix using a multispectral imaging system

OBJECTIVE

To present the clinical application of the multispectral imaging colposcopic system (MIS colposcopy).

METHODS

MIS colposcopy was performed on 123 enrolled women. After a 3% acetic acid application, sequential images were captured, analyzed, and stored automatically. Directed biopsies were taken from distinct marked acetic acid-responsive tissue areas indicated on the monitor, while a real-time assessment of the curves of intensity of the backscattered light (IBSL) vs. time was performed. Blind biopsies were taken from non-acetowhitening areas. Histological findings were correlated with MIS colposcopy results and compared with conventional colposcopy and Pap test results.

RESULTS

Acetic acid-tissue interaction resulted in temporal and spatial alterations to the light scattering properties of the abnormal tissue that was analyzed. The shape of IBSL curve and the "relaxation time" (the time it takes for IBSL to decay to 1/e of its peak value) changed in accordance with the underlying lesion. More severe CIN lesions lead to higher maximum IBSL; longer durations of acetowhitening lead to increasingly delayed exponential decay of IBSL curve. To compare with histological examination, MIS colposcopy had a 1.7% false-diagnostic rate, while PAP test and conventional colposcopy had 24.4% and 22% false-diagnostic rates, respectively. A triple exponential function created a "pseudocolor" image that comprised the grade map of the lesion, and this is frequently representative of the duration/degree of the induced alterations.

CONCLUSION

Improved diagnostic information can be gained by recording the optical information in a narrow spectral range with high spatial resolution. MIS colposcopy can be used in the diagnosis of uterine cervix pathological conditions and in the differentiation between CIN lesions.

Accuracy of visual screening for cervical neoplasia: Results from an IARC multicentre study in India and Africa

Visual inspection-based screening tests, such as visual inspection with 4% acetic acid (VIA) and with Lugol's iodine (VILI), have been proposed as alternatives to cytology in mass screening programs. To date, there is only limited information on the accuracy of these tests in detecting High-grade Squamous Intraepithelial Lesions (HSIL). Eleven cross-sectional studies involving 56,939 women aged 25-65 years were conducted in Burkina Faso, Congo, Guinea, India, Mali and Niger to evaluate the accuracy of VIA and VILI performed by health workers. A common protocol and questionnaire was used. For final diagnosis, all women were investigated with colposcopy and biopsies were taken when necessary. Data from the studies were pooled to calculate sensitivity, specificity and predictive values of the tests for the detection of HSIL. Of the screened women, 16.1% and 16.4% were positive on examination using, respectively, VIA and VILI; 1,063 were diagnosed with HSIL. The pooled sensitivity, specificity, positive and negative predictive values for VIA were 76.8% (95% CI: 74.2-79.4%), 85.5% (95% CI: 85.2-85.8%), 9.4% (95% CI:8.8-10.8%) and 99.5% (95% CI:99.4-99.6%), respectively. The values were 91.7% (95% CI: 89.7-93.4%), 85.4% (95% CI: 85.1-85.7%), 10.9% (95% CI: 10.2-11.6%) and 99.8% (95% CI:99.7-99.9%), respectively for VILI. The range of sensitivity and specificity for VIA was 56.1-93.9% and 74.2-93.8%, respectively, between studies and were 76.0-97.0 % and 73.0-91.3% for VILI. VILI had a significantly higher sensitivity than VIA in detecting HSIL, but specificity was similar. VILI appears to be a more accurate visual test for use in screening and treatment programs in low-resource settings.

When to suspect squamous cancer at colposcopy

Advanced practice nurses are increasingly using colposcopy to investigate abnormal cervical cytology. Specific colposcopic features can lead the clinician to suspect squamous cancer. This article discusses vascular pattern study, examination evaluation, lesion grading, and correlating colposcopy with histology and cytology to aid in identifying squamous cancer.

Colposcopic appearance of cervical intraepithelial neoplasia is age dependent

OBJECTIVE

We investigated to determine whether colposcopic, histologic, and virologic parameters of cervical intraepithelial neoplasia are influenced by a patient's age.

STUDY DESIGN

A cohort of 967 women with a mean age of 37.1 years underwent screening for detection of cervical intraepithelial neoplasia by colposcopy, cytologic examination, and testing for high-risk human papillomaviruses with the Hybrid Capture System (Digene, Silver Springs, Md) and a general primer and type-specific primer polymerase chain reaction system. Cervicography was used for documentation and reproducible evaluation of the colposcopic appearance of the cervix. In 86% of patients with trivial colposcopic changes of doubtful significance (100/116) and 89% of patients with colposcopic changes consistent with cervical intraepithelial neoplasia (89/99), punch biopsy specimens were taken for histologic evaluation.

RESULTS

In patients with trivial colposcopic changes of doubtful significance, histologically confirmed cervical intraepithelial neoplasia was almost as frequent (32%, 37/116) as in patients with colposcopic changes consistent with cervical intraepithelial neoplasia (43%, 43/99, difference not significant). The ratio between colposcopic evidence of cervical intraepithelial neoplasia and trivial colposcopic changes was 1.9 in patients <35 years old with cervical intraepithelial neoplasia, versus 0.5 in patients >/=35 years old with cervical intra-epithelial neoplasia (P =.005). Patients with trivial colposcopic changes of doubtful significance were older (median age 36 years) than were patients with colposcopic changes consistent with cervical intraepithelial neoplasia (median age 29 years, P =. 008). In patients with cervical intraepithelial neoplasia who had no or trivial colposcopic changes, the thickness of neoplastic epithelium was smaller (P =.008) and the number of cellular layers was lower (P =.01) than in patients with cervical intraepithelial neoplasia who had colposcopic changes consistent with cervical intraepithelial neoplasia. In patients <35 years old the rate of positive results for a high-risk human papillomavirus (P <.005) and the viral load (difference not significant) were higher than in women >/=35 years old. The rate of positive results for high-risk human papillomaviruses differed independently of age among patients with normal colposcopic findings, patients with trivial colposcopic changes of doubtful significance, and patients with colposcopic changes consistent with cervical intraepithelial neoplasia (P <.005).

CONCLUSIONS

In women >/=35 years old cervical lesions associated with intraepithelial neoplasia are thinner and thus less colposcopically conspicuous than those in women <35 years old. Patients >/=35 years old with acetowhite cervical lesions consistent with trivial changes of doubtful significance should therefore undergo punch biopsy for histologic evaluation.

Dense acetowhite: a high grade lesion?

To describe the ability of colposcopy in predicting the high grade squamous intraepithelial lesion (SIL) a retrospective, descriptive study was undertaken from January 1, 1993 until November 30, 1995. Agreement between colposcopy and histology results was described in unweighted Kappa. Agreement between colposcopy diagnosis and histology results taken from the dense acetowhitening areas was poor (Kappa 0.5). This agreement will be higher in areas of acetowhite that has an abrupt peeling margin in 72/74 cases. It seems that dense acetowhite areas showed high grade lesions in most of cases observed especially if they had also abrupt peeling margin.