An estimated 16 million women are infected with HIV worldwide, with the majority living in Sub-Saharan Africa.1 Cervical cancer is a preventable disease caused by human papilloma virus (HPV) and is a leading cause of cancer deaths among women in low-income countries.1–4 Despite the large burden of disease, only 20%–40% of HIV-infected women are screened for cervical cancer. Although 10%–30% are found to have lesions requiring treatment, less than 10% receive appropriate therapy.1,2,5–7 The reasons behind the failure to implement effective cervical cancer prevention programs in HIV-infected women are multiple. Prevention of cervical cancer requires widespread screening, accurate diagnosis of precursor lesions, appropriate triaging, and therapy.1,7 Cervical cytology, the screening test most commonly used in developed countries, requires multiple visits by clients, screening at regular intervals because of low sensitivity and excellent laboratory infrastructure, including cytology and pathology.6,8–11 Cervical cancer prevention programs targeting HIV-infected women face the additional challenge of identifying and treating a disease that seems to have a more rapid progression and higher rates of recurrences than in the general population.12–15
Prevalence rates for HIV in Botswana are currently approximately 28% in women in the 15–49 age group.16–19 By extrapolation, a large proportion of Botswana women can be assumed to be at increased risk of precervical cancer lesions and possible invasive cervical cancer, thus making this a significant public health concern in Botswana.20,21 Like in many resource-limited settings, medical and technical resources are strained in Botswana, resulting in challenges in cytology-based screening.1,7
Visual inspection with acetic acid (VIA) has recently emerged as an inexpensive practical alternative to cytology-based screening.5,11,22–27 Application of 4% or 5% acetic acid to the cervix results in dysplastic and neoplastic epithelium transiently appearing white, with normal cervical squamous epithelium assuming a pink colour. VIA is more sensitive but less specific than the Pap smear, with estimated sensitivity of 66%–99% or 55%–90%, and specificity of 64%–98% or 65%–92%.5,6,11,24,25,28 VIA can be performed by midwives, nurses, and other health care workers, which decreases barriers regarding staff shortages.26,29–32
Women with abnormal Pap smears or VIA results generally are referred for colposcopy, which includes magnified visual inspection of the cervix after application of acetic acid. Colposcopy has been extremely successful as a diagnostic modality for cervical precancer lesions and provides increased sensitivity compared with VIA.5,27 However, use of this technology in developing nations is severely limited by the cost of the colposcope.1 To overcome this limitation, lower cost methods to provide easily interpretable magnified images of the cervix were developed. One such method is cervicography in which the cervix is inspected through magnified photographic images.1,27 Cervicography has been shown in multiple studies to be a reliable method of detecting precancerous cervical lesions.25,28,33 Digital cameras have the advantage of easy portability and relative cost effectiveness compared with colposcopy. These cameras have been shown to produce quality high-resolution images, which may be used to perform magnified examination of the cervix and vagina (enhanced digital imaging, EDI).27,30,31,34
To take advantage of the potential benefits of VIA/EDI, we implemented a 2-tier community-based cervical cancer prevention program for HIV-infected women in Botswana. We describe the program and report the outcomes of the first 23 months (March 2009 through January 2011) of its implementation.
DESIGN AND METHODS
The initial phase of this program required training of staff in VIA/EDI and cryotherapy (see and treat).1 As part of on-going capacity, building other health practitioners were also trained. The community-based clinic was located at a primary clinic in Gaborone, the capital city. HIV antiretroviral therapy and minimal access to cytology-based screening was available at this clinic before our pilot. All providers were licensed nurses or nurse-midwives who were trained to perform VIA, EDI, and cryotherapy. Nurses were chosen as the program's primary providers because they are more available than physicians in Botswana and many are familiar with performing pelvic exams and function independently in clinic settings. A 3-day didactic training session took place in Zambia in January 2009, where a similar project has been running since 2006. After the didactic training, nurses spent 2 weeks in clinics gaining hands-on experience in VIA, EDI, cryotherapy, and indications for referral, under the guidance of the Zambian team. The training continued for 6 weeks in Botswana under the guidance of our program gynaecologist. At the end of the practicum, each nurse had successfully performed a minimum of 100 VIA examinations, 100 EDI photographs, and 35 cryotherapies.
The community clinic was set up within the same facility that has an Adult HIV clinic as part of the National HIV program. The HIV clinic acted as a referral centre for 5 other local clinics. Both men and women were referred to this clinic once tested positive for HIV. The services provided at this clinic include assessment for initiation of HIV treatment, follow-up of HIV-infected adults from the catchment area both on treatment and those not needing treatment yet. Any HIV-infected woman enrolled in this clinic who self-referred or was referred by a health care worker, and chose to have cervical cancer screening at our clinic, was cared for in our program.
Procedures at the Community-Based Clinic
Women presenting to the “see and treat” clinic were counseled regarding cervical cancer prevention and consented to allow cervical photography and cryotherapy if needed. During speculum inspection, women were assessed for lesions suspicious of cervical cancer (raised, ulcerative lesions with contact bleeding, bizarre blood vessel patterns); these were referred to the tertiary hospital immediately. Women were also assessed for sexually transmitted infections and appropriate national treatment guidelines followed. VIA using white household vinegar was immediately followed by EDI with a digital camera, and the results were categorized as positive based on the following:
1. Observation of a well-defined opaque aceto-white area close to the squamous-cervical junction
2. Observation of dense aceto-whitening of a cervical growth
VIA-negative women were offered reassurance. Cryotherapy at the community-based clinic was only performed on women diagnosed with low-grade lesions meeting all of the following criteria:
1. An opaque lesion involving less than 3 quadrants of the transformation zone
2. No extension of the lesion into the endocervical canal or onto the vaginal wall
3. The entire lesion could be covered by the cryotherapy probe
4. No clinical evidence of severe lesion or invasive cancer
Cryotherapy was performed using liquid nitrogen and a 15–24 mm ectocervical cryoprobe tip (probe size varied depending on lesion size) with a shallow nipple. A single freeze for 7 minutes was applied. No local anesthesia or analgesics were used before the procedure. Women with abnormal VIA or EDI that did not meet the criteria for immediate treatment were referred to our colposcopy/loop electrosurgical excision procedure (LEEP) clinic at the Princess Marina Hospital for further evaluation.
Colposcopy and Treatment of Precancerous Lesions
Colposcopy was performed in women with high-grade lesions according to VIA/EDI or those whose lesions were inappropriate for cryotherapy. Unlike in traditional colposcopy examination, punch biopsies were not obtained from abnormal areas on the cervix. The “see and treat” approach was followed here as well, with lesions either treated with LEEP or cautery.
Women with extensive lesions reaching the vaginal wall were referred to the gynecology department for cone biopsy or hysterectomy. Excised tissue specimens were processed in the pathology laboratory. Women were given home-care instructions, asked to avoid sexual intercourse for 6 weeks, and instructed to return to the clinic if they had fever for more than 2 days, were passing blood clots, or had severe lower abdominal pain. Women with suspected invasive cancer were referred to the Gynaecology Department at the Princess Marina Hospital for surgery and/or radiotherapy.
Quality Control and Nursing Education
Weekly, nurses participated in an EDI quality-control meeting that involved reviewing all images of cervigrams from the prior week on a large screen using a projector. Discrepancies between the nurse and the gynecologist expert consultant were assessed with the final decision based on the opinion of the gynecologist. Weekly reviews included an analysis of the correlation between the digital image and histology if available as a means of increasing the understanding of the myriad visual manifestations of cervicitis, squamous metaplasia, nabothian gland cysts, endocervical polyps, glandular hyperplasia, cervical hyperplasia, keratosis, cervical intraepithelial neoplasia (CIN) 1/HPV, CIN2, CIN3, microinvasive cancer, and invasive cancer.
Data and Statistical Analysis
Data entry was performed at the treatment site using Access, whereas data analysis was carried out using STATA 11.0. Comparisons were performed using a Mann–Whitney or Student t test according to the distribution of the variables. The sensitivity, specificity, negative and positive predictive values of nurse's assessments were calculated using the gynecologist evaluation as the gold standard. In addition, intersubject and intrasubject variability was determined by asking the nurses to diagnose a selected set of 100 high-quality pictures from our patients. Pictures showing normal and abnormal cervixes were included in the set, and the answers were not provided to the nurses. We used the gynecologist diagnosis (for normal cervix and low-grade lesions requiring cryotherapy but not referral) and pathology (for an abnormal cervix). The test was administered 2 weeks later showing the same set of pictures in a different order. Intrasubject and intersubject variability was calculated using the kappa coefficient.
This was set up as a public health care program to provide service, with a monitoring and evaluation component. The Health Research and Development Committee of the Botswana Ministry of Health approved the program protocol and data collection for programmatic evaluation, including consent for imaging.
During the first 23 months 2175 HIV-infected women were evaluated (Fig. 1). Our population was included predominantly young [mean age 34.9, 95% confidence interval (CI), 34.6 to 35.2] single women from lower socioeconomical strata (as indicated by their average monthly household income; Table 1). The average CD4 T-cell count was 405 cells per square millimeter and the 68.8% of the women were on antiretroviral therapy.
Two hundred and fifty-three of 2175 women screened (11.6%) were found to have low-grade lesions and received same-day cryotherapy. One thousand three hundred forty-seven (61.9%) women were considered to have a normal examination and 575 (27.3%)were referred for further treatment (Fig. 1). Of the 1347 women initially considered to have normal exams, 267 (19.8%) were recalled when the gynecologist was unable to confirm the initial decision of the nurse either because the images were of poor quality (68, 25.6%; Table 2) or because a possible lesion could not be excluded (199, 74.4%; Table 2). Thus, a total of 1333 (1347 + 253 to 267) (61.3%) women received appropriate same-day screening and treatment without the need for recall or referral. Two hundred ten (78.6%) of the 267 recalled, and 499 (86.8%) of the 575 referred women were seen at the referral clinic (overall follow-up 709 of 842, 84.2%). Reasons for recall and referral are shown in Table 2.
Five hundred and six (71.4% of the 709 women; 210 recall plus 499 referred) required treatment. None of the women recalled were found to have high-grade lesions (Tables 2 and 3). Overall, our program identified and treated 264 precancerous lesions (CIN 2 or 3) and identified 6 microinvasive cancers, which were referred for further management (Tables 2 and 3).
Of the 210 women who were recalled and followed-up by the time of this analysis, 164 required no treatment and 25 required cryotherapy/cautery. Tissue was obtained from 46 other recalled women, and results were available for 31 (12 + 9) of them. Twelve of them showed changes compatible with CIN 1 lesions and 9 showed chronic cervicitis (Table 3). Of the 460 women who were referred and followed-up, 39 required no treatment and 32 underwent cryotherapy/cautery for low-grade lesions. Of the 319 remaining ones who were referred and had pathological results by the time of analysis, 49 had CIN1, 62 had CIN2, 202 had CIN 3, and 6 had microinvasive cancer.
We based the calculations of the sensitivity and specificity of the program on the performance of peripheral clinic and its capacity to appropriately identify patients who were normal, patients who could be treated on-site, and patients who needed referral. Using this approach, the 267 patients recalled for evaluation were considered to be “undertreated” by the first tier of the program. Although the majority of those patients were found to have no lesions (164 of the 210 patients who followed up), we believe our approach provides a better understanding of the overall performance of the system. The nurses at the community-based clinic diagnosed 1080 (1347 to 267) women were confirmed to be negative, 237 were declared appropriately diagnosed and treated, 16 were “potentially overtreated (considered “misdiagnosed” for the purpose of this analyses), and 575 were appropriately referred. Thus, the sensitivity and specificity of the nurses' assessment (and, therefore, of the first tier of our program) were 75.3% (95% CI: 72.5% to 77.8%) and 98.5% (95% CI: 97.6% to 99.1%), respectively.
The level of agreement between nurses and the gynecologist in the evaluation of digital pictures (agreement on 83.3% of the observations, kappa = 0.63, 95% CI: 0.48 to 0.77, P < 0.01). Although there were some differences in the intersubject agreement between nurses (agreement on 86.5% of the observations, kappa = 0.69, 95% CI: 0.54 to 0.83, P < 0.01), the intrasubject agreement was substantially higher (agreement on 91.2% of the observations, kappa = 0.72, 95% CI: 0.64 to 0.80, P < 0.001).
Our results demonstrate that community-based “see and treat” programs using VIA/EDI can be valuable additions to screening programs for cervical cancer prevention in HIV-infected women in settings like Botswana. To our knowledge, this is the first community-based VIA/EDI screening program targeting exclusively HIV-infected women. Our approach was efficient and had a high throughput. In addition, visualizing lesions on the digital images by the patients serves as a form of patient education and may lead to improved follow-up rates. The digital images also allowed the implementation of quality control measures that further improve safety.34 Although a significant percentage of women required a second visit to receive definitive treatment, the large majority of women were able to be screened and treated on a single visit.
Consistent with our findings, digital cameras, with their advantageous portability, ease of use, and relative cost-effectiveness, have been shown to produce quality high-resolution images, which may be used to perform magnified examination of the cervix and vagina. Mobile telemedicine phones with high-resolution camera handset technology has the added advantage of not requiring internet connections or electricity and allows images to be transmitted immediately. This “simultaneous telemedicine” technology could provide the opportunity for evaluation by an expert at a distant location although the patient is being evaluated in clinic.34
Our results are comparable to those reported by similar VIA-based prevention programs from resource-limited settings.1,5,29,33 Among the patients screened through our program, only one-third of women had been screened for cervical cancer by Pap smear in the past. Of interest, the large majority of them knew their results (85%). However, despite abnormalities in approximately 20%, follow-up or treatment was never provided, highlighting some of the barriers to care within cytology-based programs. The potential benefit of treating early HPV lesions in HIV-infected women are important advantages of the VIA/EDI approach. Our program provided successful treatment for 85% of the patients who needed it, a reasonable number of them on the same day. Referral for LEEP required a second visit, but multiple visits were minimized by providing “see and LEEP”. Providing VIA/EDI/cryotherapy and LEEP at the same facility can eliminate the need for second visit for those patients not legible for cryotherapy.
Concerns have been raised about the safety of the “see and treat” approach. In particular, concern that lesions may be missed by nurses performing point-of-care evaluation and treatment has limited the uptake of this approach. In our program, during the quality assurance evaluations performed weekly, only 12% (267/2, 175) of the women assessed at the community clinic were recalled for further evaluation by the specialist. The main reason for recall was the poor quality of the picture or the inability to completely rule out the possibility of a lesion by the gynecologist based only on the picture. It is highly reassuring that missed lesions were found only in 3.5% (7/210 who showed up) of the women recalled. Importantly, 164 of 210 (78%) patients recalled did not need any treatment and those that did require treatment had either low-grade lesions (12) or other lesions (9) not related to HPV infection.
Although we detected 264 women with CIN 2 or 3, the magnitude of the impact that such programs will have on cervical cancer can only be established through long-term prospective studies. The 264 premalignant lesions identified and treated during the 23 months of our program represent nearly 70% of all the premalignant lesions treated in the region over the same period. “See and treat” may offer a viable alternative to reach women living in distant and/or underserved regions of the country (eg, rural), whereas cytology-based screening programs may still play a role in areas where the appropriate infrastructure and human resources are available.
This program has shown that screening and treating women using a “see and treat” approach of VIA/EDI followed by cryotherapy and referral for LEEP therapy is feasible, effective, and has a significant impact on the identification and treatment of precancerous and microinvasive cancerous lesions in HIV-infected women in Botswana. In addition, we showed that including telemedicine technology is feasible in these settings and could add to the overall benefit of these programs. In resource-limited settings, “see and treat” approaches are highly likely to reduce the risks of a common and preventable cancer in women, especially if both cryotherapy and LEEP can be provided at the same facility to truly allow same day treatment for all lesions.
The authors thank the staff of the Botswana-UPenn Partnership Women's Health clinic for their admirable devotion to the care of the patients enrolled in this study. We thank the women participating in the program for their invaluable contribution.
1. Cervical cancer screening in developing countries. Report of a WHO Consultation. Geneva, Switzerland: World Health Organization; 2002: ISBN 92 4 154572 0.
3. Hawes SE, Critchlow CW, Faye Niang MA, et al.. Increased risk of high-grade cervical squamous intraepithelial lesions and invasive cervical cancer among African women with human immunodeficiency virus type 1 and 2 infections. J Infect Dis. 2003;188:555–563.
4. Firnhaber C, Evans D, Friedman-Khalili R, et al.. Seroprevalence of HPV vaccine types 6, 11, 16 and 18 in HIV-infected women from South Africa, Brazil and Botswana. J Clin Virol. 2011;52:265–268.
5. Arbyn M, Sankaranarayanan R, Muwonge R, et al.. Pooled analysis of the accuracy of five cervical cancer screening tests assessed in eleven studies in Africa and India. Int J Cancer. 2008;123:153–160.
6. Denny L, Quinn M, Sankaranarayanan R. Chapter 8: screening for cervical cancer in developing countries. Vaccine. 2006;24:(suppl 3):S3/71–S3/77.
7. Finn A. HPV prevention: waiting for screening services. Lancet. 2010;375:1694.
8. Danso D, Lyons F, Bradbeer C. Cervical screening and management of cervical intraepithelial neoplasia in HIV-positive women. Int J STD AIDS. 2006;17:579–584; quiz: 585–577.
9. Kim JJ, Brisson M, Edmunds WJ, et al.. Modeling cervical cancer prevention in developed countries. Vaccine. 2008;26:(suppl 10):K76–K86.
10. Leinonen M, Nieminen P, Kotaniemi-Talonen L, et al.. Age-specific evaluation of primary human papillomavirus screening vs conventional cytology in a randomized setting. J Natl Cancer Inst. 2009;101:1612–1623.
11. Sankaranarayanan R, Budukh AM, Rajkumar R. Effective screening programmes for cervical cancer in low- and middle-income developing countries. Bull World Health Organ. 2001;79:954–962.
12. De Vuyst H, Franceschi S. Human papillomavirus vaccines in HIV-positive men and women. Curr Opin Oncol. 2007;19:470–475.
13. De Vuyst H, Lillo F, Broutet N, et al.. HIV, human papillomavirus, and cervical neoplasia and cancer in the era of highly active antiretroviral therapy. Eur J Cancer Prev. 2008;17:545–554.
14. Strickler HD, Burk RD, Fazzari M, et al.. Natural history and possible reactivation of human papillomavirus in human immunodeficiency virus-positive women. J Natl Cancer Inst. 2005;97:577–586.
15. van der Burg SH, Palefsky JM. Human immunodeficiency virus and human papilloma virus—why HPV-induced lesions do not spontaneously resolve and why therapeutic vaccination can be successful. J Transl Med. 2009;7:108.
16. Central Statistics Office, Republic of Botswana. Stats Brief: Preliminary Results of the AIDS Impact Survey III. Gaborone: Botswana; 2008.
17. International Group on Analysis of Trends in HIV Prevalence and Behaviours in Young People in Countries most Affected by HIV. Trends in HIV prevalence and sexual behaviour among young people aged 15-24 years in countries most affected by HIV. Sex Transm Infect. 2010;86:(suppl 2):ii72–ii83.
18. Bussmann H, de la Hoz Gomez F, Roels TH, et al.. Prevalence of transmitted HIV drug resistance in Botswana: lessons learned from the HIVDR-threshold survey conducted among women presenting for routine antenatal care as part of the 2007 national sentinel survey. AIDS Res Hum Retroviruses. 2011;27:365–372.
19. Riviello ED, Sterling TR, Shepherd B, et al.. HIV in the workplace in Botswana: incidence, prevalence, and disease severity. AIDS Res Hum Retroviruses. 2007;23:1453–1460.
20. Macleod IJ, O'Donnell B, Moyo S, et al.. Prevalence of human papillomavirus genotypes and associated cervical squamous intraepithelial lesions in HIV-infected women in Botswana. J Med Virol. 2011;83:1689–1695.
21. Ramogola-Masire D, McGrath CM, Barnhart KT, et al.. Subtype distribution of human papillomavirus in HIV-infected women with cervical intraepithelial neoplasia stages 2 and 3 in Botswana. Int J Gynecol Pathol. 2011;30:591–596.
22. Sankaranarayanan R, Esmy PO, Rajkumar R, et al.. Effect of visual screening on cervical cancer incidence and mortality in Tamil Nadu, India: a cluster-randomised trial. Lancet. 2007;370:398–406.
23. Sankaranarayanan R, Mathew B, Jacob BJ, et al.. Early findings from a community-based, cluster-randomized, controlled oral cancer screening trial in Kerala, India. The Trivandrum Oral Cancer Screening Study Group. Cancer. 2000;88:664–673.
24. Sankaranarayanan R, Nene BM, Shastri SS, et al.. HPV screening for cervical cancer in rural India. N Engl J Med. 2009;360:1385–1394.
25. Sankaranarayanan R, Shyamalakumary B, Wesley R, et al.. Visual inspection with acetic acid in the early detection of cervical cancer and precursors. Int J Cancer. 1999;80:161–163.
26. Nene BM, Deshpande S, Jayant K, et al.. Early detection of cervical cancer by visual inspection: a population-based study in rural India. Int J Cancer. 1996;68:770–773.
27. Berkhof J, de Bruijne MC, Zielinski GD, et al.. Evaluation of cervical screening strategies with adjunct high-risk human papillomavirus testing for women with borderline or mild dyskaryosis. Int J Cancer. 2006;118:1759–1768.
28. Sankaranarayanan R, Nene BM, Dinshaw K, et al.. Early detection of cervical cancer with visual inspection methods: a summary of completed and on-going studies in India. Salud Publica Mex. 2003;45:(suppl 3):S399–S407.
29. Nene BM, Hiremath PS, Kane S, et al.. Effectiveness, safety, and acceptability of cryotherapy by midwives for cervical intraepithelial neoplasia in Maharashtra, India. Int J Gynaecol Obstet. 2008;103:232–236.
30. Gage JC, Rodriguez AC, Schiffman M, et al.. An evaluation by midwives and gynecologists of treatability of cervical lesions by cryotherapy among human papillomavirus-positive women. Int J Gynecol Cancer. 2009;19:728–733.
31. Gage JC, Rodriguez AC, Schiffman M, et al.. Treatability by cryotherapy in a screen-and-treat strategy. J Low Genit Tract Dis. 2009;13:174–181.
32. Blumenthal PD, Lauterbach M, Sellors JW, et al.. Training for cervical cancer prevention programs in low-resource settings: focus on visual inspection with acetic acid and cryotherapy. Int J Gynaecol Obstet. 2005;89:(suppl 2):S30–S37.
33. Cuzick J, Arbyn M, Sankaranarayanan R, et al.. Overview of human papillomavirus-based and other novel options for cervical cancer screening in developed and developing countries. Vaccine. 2008;26:(suppl 10):K29–K41.
34. Quinley KE, Gormley RH, Ratcliffe SJ, et al.. Use of mobile telemedicine for cervical cancer screening. J Telemed Telecare. 2011;17:203–209.
© 2012 Lippincott Williams & Wilkins, Inc.