Evaluation of direct visual inspection of the cervix in detecting cytology diagnosed squamous intraepithelial lesion in women of known HIV status. A randomized trial (CANHIV study).
A two-arm, open label, randomized study, evaluated the test characteristics of visual inspection of cervix with Acetic acid (VIA) and Lugol's Iodine (VILI) in detecting cytology diagnosed squamous intraepitheliallesion (SIL) in 1160 women of known HIV status in southwestern Nigerian. Using SIL as reference standard and the HIV status masked, VIA and VILI had similar test characteristics except for the positive predictive value in which VIA value of 91.5% was significantly higher than 77.7% for VILI ( p=0.01). Among HIV positive women, VILI performed poorly across all the 4 test characteristics compared to VIA. Among severely immuno-compromised HIV positive participants VILI performance was consistently below 80% across all test characteristics (sensitivity-70.0%; specificity-66.9%; positive predictive value-46.7%; negative predictive value (NPV) -50.0%) compared to VIA (Senstivity-71.3%; specificity-88.2%; positive predictive value-83.3%; negative predictive value-88.2). Our study shows that VILI is insufficiently sensitive and specific in the presence of HIV infection especially in those with severe immunosuppression. Based on VIA's acceptable sensitivity and NPV in all situations, it is recommended for cervical cancer screening in HIV positive women and in settings of high HIV burden. (Afr J Reprod Health 2016; 20[4}: 77-88).
Keywords: HIV, Cervical cancer, squamous intraepitheliallesion, visual inspection with Acetic acid and Lugol's Iodine
Une etude randomisee a deux bras, ouverte, a evalue les caracteristiques de l'exam en visuel du col avec de l'acide acetique (IVA) et de l'iode de Lugol (IVLI) dans la detection d'une lesion intraepitheliale squameuse (LIS) diagnostiquee par a l'aide de la cytologie chez 1160 femmes seropositives connues dans le sud-ouest du Nigeria. En utilisant la LIS comme norme de reference et on masquant la seropositivite, l'IVA et l'IVLI presentaient des caracteristiques de test similaires sauf pour la valeur predictive positive dans laquelle la valeur IVA de 91,5% etait significativement superieure a 77,7% pour IVLI (p = 0,01). Parmi les femmes seropositives, IVLI a presente un mauvais rendement dans toutes les 4 caracteristiques du test par rapport a l'IV A. Chez les participants seropositifs severement immunodeprimes, la performance de l'IVLI etait constamment inferieure a 80% pour toutes les caracteristiques du test (sensibilite-70,0%, specificite-66,9%, valeur predictive positive-46,7%, valeur predictive negative (VPN) -50,0% (Sensibilite-71,3%, specificite-88,2%, valeur predictive positive-83,3%, valeur predictive negative -88,2). Notre etude montre que l'IVLI est insuffisamment sensible et specifique en presence d'infection par le VIH, en particulier chez ceux qui ont une immunosuppression severe. En se fondant sur la sensibilite acceptable de l'IV A dans toutes les situations, il est recommande pour le depistage du cancer du col de l'uterus chez les femmes seropositives et dans les contextes du fardeau eleve du VIH. (Afr J Reprod Health 2016; 20[4}: 77-88).
Mots-des: VIH, cancer du col uterin, lesion intraepitheliale squameuse, inspection visuelle avec de l'acide acetique et de l'iode de Lugol
It is estimated that globally about 2.3 million new HIV infections (1) and about half a million cases of cervical cancer (2) were diagnosed in 2012. The two diseases have been reported to be closely intertwined, and the number of patients with comorbidities continues to grow rapidly (3). Cervical cancer IS one of the common opportunistic cancers and cause of cancer related death in HIV positive women in low income countries (4-6). Notwithstanding the proven effectiveness of cervical cancer prevention and control, in the presence of HIV infection, it is associated with substantially increased case fatality (6).
Cervical cytology-based screening programs continue to be the mainstay of cervical cancer prevention globally. It have demonstrated reduction in the cervical cancer incidence and mortality, particularly in high income countries, which focus on good-quality screening including optimal frequency and adequate coverage (7-10). However, the sensitivity of cytology to detect cervical cancer precursors is ranged from 50 - 80% (11, 12). The sensitivity was however slightly lower for mild and moderate dysplasia (78.1 %) and slightly higher for carcinoma in situ and severe dysplasia (81.4%) and 82.3% for invasive carcinoma (12). In addition, it can only be effectively implemented if infrastructure and laboratory quality assurance requirements are consistently met. As a result of these requirements, cytology based screening programmes cannot be effectively implemented in most low income countries, prompting the recommendation of visual inspection of the cervix with either Acetic acid (VIA) or Lugol's Iodine (VILI) as an alternative cervical cancer screen strategy to cytology based screening programme in low income countries (7-10, 13).
Most countries including Nigeria have recommended the integration of cervical cancer prevention and control services into HIV programmes as a strategy to reduce the incidence, high morbidity and mortality associated with cervical cancer in HIV positive women (13-16). While this strategy seems appropriate and ideal for prevention and control of cervical cancer in HIV positive women, there are scanty data on the test performance of direct visual inspection (DVI), a key tool of the new strategy in the context of HIV infection (17).
Visual inspection with either acetic acid or Lugol's iodine has been extensively evaluated in low-income countries with inconsistent finding in their test characteristics (17-20). While some studies reported VIA to be superior to VILI in their performance (17, 18, 20), others reported that VILI
appears to be a more accurate test (9, 21, 22). In multicenter studies involving some sub Saharan African countries and India to evaluate the test characteristics of VIA and VILI, the researchers concluded that VILI is a more accurate visual test for use in screening and treatment programs in low-resource settings. The pooled sensitivity and negative predictive values for VIA were 76.8% and 99.5% respectively. The values were 91.7% and 99.8%, respectively for VILI (20). Notably these evaluations were in population of unknown HIV status (17-21). The few studies evaluating this test among HIV positives did not analyse for the effect of severe immunosuppression on the sensitivity and specificity of the test, mainly cross sectional and non-randomized studies (17, 23). Utilizing such a tool for screening in the context of HIV infection may be inappropriate, as experiences from the tuberculosis field has shown the failure of the existing TB diagnostic tools like x-ray and sputum smear microscopy in detecting all forms of tuberculosis in severely immune-compromised HIV positive persons (24). In addition, there are scanty data on the performance of DVI in terms of sensitivity and specificity when used under prevailing programmatic conditions m sub-Saharan African countries (25).
In this study, we used a randomized open label trial to investigate the test characteristics of visual inspection of the cervix after application of 5% acetic acid (VIA) and visual inspection after the application of Lugol's iodine (VILI) in detecting cervical squamous intraepithelial lesion (SIL) in women of known HIV status, in the context of actual medical practice in programmatic settings in a high HIV/cervical cancer burden African country; namely Nigeria.
Study design and participants
This study was a multi-site, two-arm, open label randomized trial which evaluated the test characteristics of visual inspection with Acetic acid (VIA arm) versus Visual inspection with Lugol's Iodine (VILI arm) in detecting cervical squamous intraepithelial lesions diagnosed by cytology. A total of 1289 women were approached to participate in the study; of which 1160 eligible women agreed to be part of the study and were randomized into the two study arms.
Inclusion criteria were: adults (> 18 years) with known HIV status or willingness to have an HIV test and written informed consent. Exclusion criteria were: Overt cervical cancer, known reaction to Lugol's iodine, psychiatric illness and alcohol or drug abuse.
The study was conducted in two settings; at the cervical cancer screening Centre, Nigerian Institute of Medical Research (NIMR), Lagos and at the community cervical cancer screening outreach programs in Lagos and Ogun states of Nigeria.
Community cervical cancer screening outreach programme: In June 2011, the Nigerian Institute of Medical Research (NIMR) initiated a community based outreach cervical cancer-screening programme as a corporate social responsibility in two contiguous south western Nigeria states of Lagos and Ogun with a population of 13 million (26, 27). The study was conducted in four urban and six rural communities.
Nigerian Institute of Medical Research, Yaba Lagos currently provides comprehensive HIV care, treatment and support for over 23,000 patients with 65% of the patients coming from Lagos and the remaining from the other neighboring states. The study was initially planned to end in June 20 12 but was extended to December 2012 in order to achieve the required study sample size.
Before the screening, the participants were educated on cervical cancer screening, its' importance, the required follow-up appointment and also on all study related procedures. They were then asked to sign a written informed consent document.
After signing the informed consent form, information on socio-demographic characteristics, sexual and reproductive history was collected using a study case record form prepared by the PI (OE). All participants were subjected to a thorough pelvic examination, in a sequence comprising of; collection of the Pap smear, collection of sample for microbiological examination (when indicated) and Direct Visual Inspection (DVI) using either Acetic Acid (VIA) or Lugol's Iodine (VILI). The clinical examinations and sample collection for cervical Pap tests were performed by physicians and midwives who received a competency based training preparatory to the study.
The women were placed in the modified lithotomy position and cervix was exposed with the help of a disposable Cusco's bivalve speculum to facilitate the cervical examination. Cervical cells scraping was obtained by the use of an Ayres spatula and the smear prepared by spreading the specimen uniformly across a pre-labeled glass slide. This cytology smear was immediately fixed using a commercial fixator containing 95% ethyl alcohol. The slides where then hatched and transported for analysis. After collecting the cervical smear, the same examiner performed VIA or VILI depending on a predetermined group allocation.
VIA procedure and interpretation: After collection of the samples for the Pap test and microbiological test, VIA was performed by generously applying freshly prepared 5% acetic acid on the entire cervix with a cotton swab. After one minute, the cervix was illuminated with a bright lamp and visually examined ('naked eye' examination). The findings of VIA were recorded using the following criteria
1. No Acetowhite lesions
2. Acetowhitening on endocervical polyps, nabothian cysts
3. Prominent white line like acetowhitening of the squamous junction
4. Faint, translucent, ill defined, irregular acetowhite lesions on the cervix
5. Definite, angular, geographic, acetowhite lesions far away from the squamocolumnar junction
1. Opaque, dense, dull, definite, well defined acetowhite lesions touching the squamocolumnar junction or close to external os
2. Large, circumferential, well defined, thick, dense acetowhite lesions.
3. Acetowhite lesions on clinically visible ulceroproliferative growth of the cervix
VILI procedure and interpretation: After collection of the samples for the Pap and microbiological test, VILI was performed by generously applying Lugol's iodine on the entire cervix with cotton swab. The cervix was illuminated with a bright lamp and visually examined. The findings of VILI were recorded using the following criteria:
1. Homogeneous staining of the cervix mahogany brown or black and the columnar epithelium does not change colour and remains pale.
2. Patchy, indistinct, ill defined, colourless or partially brown areas in the transformation zone
3. Scattered, irregular, ill-defined non iodine uptake areas on the cervix, with or without extension to the vagina
4. Thin, yellow, non-iodine uptake areas with angular, or digitating margins, resembling geographical areas, located far away from squamocolumnar junction
1. Well defined, dense, thick, bright, mustard-yellow or saffron--yellow, iodine non uptake areas touching the squamocolumnar junction.
2. Circumferential, well defined, thick, dense, yellow lesion, occupying large portion of the cerv1x
3. Ulceroproliferative growth of the cervix turns yellow
Cytology sample analysis
Cytology sample analysis was at the Department of Pathology, University College Hospital Ibadan, Nigeria and interpretation were according to Bethesda system. The cytopathologists who performed the cytological analyses were blinded to the participants HIV status. A senior pathologist read all tests originally classified as abnormal and 15% of those classified as normal. All slides were pre-coded with the study number before samples were taken. In the event of disagreement between the cytopathologist and senior pathologist report, the slides were sent to another senior pathologist for an independent review. For all such cases, that review constituted the final diagnosis.
HIV test was conducted according to Nigerian National HIV testing and counseling guidelines in all women before enrolment into the study. Diagnosis was based on positive test on double ELISA based algorithm.
CD4 cell count Tests were conducted at the Human Virology Laboratory. Whole blood of the HIV positive women were used to perform CD4 assay using the Cyflow Counter and Kits (Partee, Germany) according to the Manufacturer's instructions.
The outcome measures was cervical squamous intraepithelial lesion or its equivalent of VIA or VILI positivity.
Sample size for this study was calculated to demonstrate VILis' assumed superiority to VIA in terms of sensitivity, specificity, PPV and NPV in diagnosing cervical intraepithelial lesion diagnosed by histology in women of known HIV status. In a multicenter study by Sankaranarayanan R and colleagues in Africa and India, VILI (91.7%) was found to be more sensitivity than VIA(76.8%) in detecting precancerous lesion of
the cervix (20). To this effect a minimum of 530 participants for each arm was sufficient to achieve 80% power at a 5% significance level (i.e. one-sided) with the assumption that the proportion of case correctly detected by VILI will be 90%and assuming maximum dropout rate in the study will be 5%. Thus, 580 participants were randomly allocated to each arm.
Participants were allocated to screening arms according to a computer-generated randomization list prepared and held by Principal Investigator using a free online random number generator by Intemondino group (http://randomnumbergenerator.intemodino.com/en/). On each day of recruitment the team is provided with a list of randomization sequence. This sequence list is read by counselors who were not investigators in the trial. The counselors kept group allocation logs which were not available to the physicians and midwives conducting the examination until the conclusion of enrollment.
Investigator competency was maintained by support supervision in the field and by periodic training and retraining and performance monitoring, along with rates of positive results on screening, in comparison with the supervisors' results.
The reference standard diagnosis for this study was based on cytology findings. Participants with squamous intraepitheliallesion or invasive cancers were considered as true positive cases for the estimation of test accuracy. The estimates for sensitivity, specificity and predictive values and their 95% confidence intervals were calculated using standard formulae for these test characteristics (25). Since all the participants that completed the study were evaluated with the reference investigation (Cytology), the calculations were made directly using a 2 X 2 contingency table, without verification bias.
During data analysis, two reference result threshold were used; squamous intraepithelial lesion and high grade squamous intraepithelial lesion. To determine the effect of immunosuppression among HIV positive participants on the DVI test performance, further stratified analysis was performed using CD4 cell count cut off of 200 cells/mm3. Differences in the sensitivity, specificity, predictive values of VIA and VILI's were also determined. P <0.05 was accepted as level of significance.
A CONSORT chart was developed to show the number of women at each stage of the trial by study arm. This includes the numbers assessed for eligibility, screened, randomized and eventually screened for precancerous lesion of the cervix. Baseline characteristics were stratified by trial arms and summarized to assess the degree of balance between VIA and VILI arms. Median and interquartile ranges were reported for age, parity, life time partner, age at first intercourse, CD4 count and viral load; for the other variables the percentages in each category were summarized between arms.
This trial was approved by the Nigerian Institute of Medical Research Institutional Review Board and was registered with current controlled trials (ISRCTN90623294)
Role of funding source: This study was funded by the OE research budget at Nigerian Institute of Medical Research, Y aba Lagos.
Recruitment and eligibility
A total of 1289 women were approached to participate in the study; of which 98 were found ineligible to participate in the study because they declined to be part of the study (35; 35.7%), to take HIV test (21 ;21.4%) and continue with the cervical cancer screening (42; 42.6%). Of the remaining 1191 women who accepted to be part of the study, 31 women were further excluded because they were aged less than 18 years (19; 61.3%), requires husband's permission (9;29.0%) and menstruating (3;7.7%). The remaining 1160 women were randomized into two arms; 580 each in VIA and VILI arms. Twenty (1.7%) women were excluded from analysis because cytology smear was not taken or cytology result was not available; 8 and 12 respectively in VIA and VILI arms (Figure 1).
The characteristics of the participants in the study by study arms are shown in Table 1. The women in the two test arms were comparable in their baseline characteristics and the prevalence of cervical cell abnormality. A majority of the study participants were less than 40 years of age (55.2%) and HIV infected (53.1%), with CD4 count above 200 cells (91.5%).
Cytology results by DVI test outcome
The distribution of cytology results by VIA and VILI test outcomes are shown in Table 2. Of the 572 participants that were randomized to VIA and had complete result, cytology diagnosed 51(8.9%) to have SIL abnormality and 471(82.3%) as normal. Atypical squamous cell abnormality of unknown status (ASCUS) was the finding in 49 women (8.6%). VIA screening correctly identified 43 (84.3%) of the 51 cytological confirmed SIL as positive for precancerous lesion and 467 (99.2%) of the 471 normal cytology cases as negative for precancerous lesion. Among the 568 women randomized to the VILI arm and who have complete result, VILI correctly identified 36 (80.0%) of the 45 cytological diagnosed SIL cases as positive for precancerous lesion and 398 (88.6%) of the 449 cytological normal cases as negative for precancerous lesion
Safety related issues were only reported in 9 participants; 7 in the VILI arm and 2 in the VIA arms. All reported painful sensation following screening. In both testing arms the occurrence of the events were immediately after the screening and resolved within 48 hours, except in one participant that resulted in ulceration and required antibiotics therapy for 5 days. However, there was no significant difference in safety profile of the two testing arms (p =0.10; odd ratio: 0.28; 95% CI: 0.04-1.46).
Test characteristics of VIA and VILI
The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) and the 95% confidence interval for VIA and VILI in detecting cytology diagnosed SIL or worse lesions are shown in Table 3. The result is presented using two reference threshold of: low threshold of cytology diagnosed squamous intraepithelial lesion or high threshold of cytology diagnosed high grade intraepithelial lesion or worse lesion.
Using a low threshold reference standard of cytology diagnosed squamous intraepithelial lesion and the HIV status of the women masked, VIA and VILI had similar test characteristics except for the PPV in which VIA value of 91.5% was significantly higher than 77.7% for VILI (p=0.01). When the HIV status of the women was unmasked and the data re-analysed; VILI performance was significantly lower across all the 4 test characteristics compared to VIA in HIV positive women. VILI performance was consistently below 80.0% in all test characteristics except for NPV where it was 88.4%. Among HIV negative women, VIA and VILI performance was similar except for PPV in which VIA score of 89.3% was statistically higher than 63.4% for VILI (p=0.0001).
Comparison of VIA and VILI using a high threshold of HSIL reference standard showed similar performance in all test characteristics except for PPV when HIV status of the women was unmasked. PPV of VIA (77.8%) was significantly higher than 22.0% for VILI (p=0.000). Among the HIV positive women, VIA performed better in sensitivity (90.3% vs. 82.4%), specificity (93.1% vs. 71.3%) and PPV (75.3% vs. 19.9%) even though the PPV of both test were below 80.0%. For NPV scores, VILI performed better at 97.8% compared to 94.5% for VIA, however the difference was not statistically significant (P=0.68). Similar findings were noted in HIV negative women except for PPV, were VIA performance was 85.9% vs. the significantly lower value of 61.5% for VILI.
Further, stratification of the women by CD4 cell count cut off of 200 cells/mm3 showed that among the women with CD4 below 200 cells, VIA showed better performance across the 4 test characteristics than VILI. VILI performance was below 80% across all test characteristics (sensitivity: 70.0%; specificity: 66.9%; PPV: 46.7%; NPV: 50.0%). However, among women with CD4 cells count above 200 cells/mm3, VILI performance was below 80.0% in only specificity (73.4%) and PPV (57.5%). VIA performance was above 93.0% across board.
The performance of visual inspection with acetic acid and Lugol's iodine in detecting cytology diagnosed cervical cancer precursors were compared m terms of sensitivity, specificity, positive and negative predictive value. While a multicenter study in India and southern African countries reported the superiority of VILI I over VIA in detecting cervical cell abnormalities (19, 21, 22), more recent studies (17, 18, 20) comparing the test characteristics of VILI and VIA reported the superiority of VIA over VILI, even though both meet the minimum benchmark for a screening test (10, 13,-16). The finding of this study confirms the more recent findings of the superiority of VIA over VILI. In addition, we also found VILI to be an inadequate screening tool for precancerous cervical lesion in HIV positive women. The performance of VILI deteriorated as the CD4 cell count of the HIV positive women reduced to below 200 cells. This finding is similar to what had been previously reported in TB/HIV coinfected infection in which TB diagnostic tools have been found to be less sensitive (24). The explanation for this is not immediately obvious but it may be related to the poor expressions of some protein that interact with Lugol's iodine affecting its reaction with cervical cells. In addition, the large number of HIV positive women in this study compared to previous studies may have exposed the inadequacy of VILI in the presence of HIV infection (17 - 22). Thus, in low resource countries, VIA can be utilized for screening both HIV positive and women of unknown HIV status. However the high default rate in screening programmes based on visual inspection methods should be addressed (26, 27). Complementing the test with colposcopy and thereafter treated in one single visit will resolve this challenge associated with VIA screening (16, 26, 28-30).
Another important finding in this study is the near excellent negative predictive value of VIA in this study ranging from 88.2 - 99.5%. Similar results have been previously reported (17, 18, 23), confirming that if VIA test result is negative, women irrespective of their HIV status can be reassured and safely sent home. Though the NPV of VILI is generally above 70% in most situations, its 50% NPV score among HIV positive women with CD4 cell count below 200 cells makes it an unreliable test in low income, high HIV burden countries, as health workers cannot confidently reassure this category of women that the result is truly negative for precancerous lesion of the cervix.
The positive predictive value of VIA was also good and ranged from 83.3% to 93.4%, showing that when it is used in low income, high HIV positive burden under a see and treat programme. The economic cost and morbidity associated with over treatment is acceptable and health workers can confidently inform women with positive result that they are very likely to have precancerous lesion of the cervix and go ahead and treat. The VILI performance was abysmal as it is only in one situation (HIV status masked) that it performed above 65%, making it an unreliable test that will be associated with a lot of over treatment with its attendant economic cost and morbidity.
This study was limited by using cytology diagnosed lesion as the "reference standard" instead of histology diagnosed lesion. This study was comparing two approved tool for screening of precancerous lesions of the cervix and not evaluating new tool, thus the effect is likely to be minimal and may have no significant impact on the test characteristics observed. In addition we initially planned to use histology as gold standard, but the approving ethics committee declined approval. In their opinion, "the proposed plan to use histology diagnosis as gold standard instead of a less invasive standard like cytology for a study that aimed to compare the test performance of tools that are already approved, in use, involving HIV positives and partly conducted in community outreach programme, may not add any extra value rather may pose a risk to participants and health workers".
The main strength of this study is that it was done in programme setting using randomized controlled trial including over 1000 participants. To our knowledge, this is the only study in the West Africa sub region performed within programme setting with over half of the participants being HIV infected which gave us the statistical power to make definite conclusions about the test performance in HIV situations.
This study shows that visual inspection with Lugol's iodine is insufficiently sensitive and specific in presence of HIV infection especially in those with severe immunosuppression and should not be used to screen for precancerous lesion of cervix in HIV positive women. In HIV positives, those of unknown status and high HIV burden setting, VIA should be the preferred screening test.
We sincerely thank Eva Amadi, Rita Nwosu, Yetunde and her team for their assistance with data and sample collection and Oba Rasheed for coordinating data entry. We are also grateful to staff of Clinical Sciences Division, HIV testing and counseling centre and Human virology laboratory, Nigerian Institute of Medical Research, Lagos.
Trial Registration: Current Controlled Trials ISRCTN90623294
All the authors participated in the planning and design of the study, and all read and approved the final manuscript. OCE conceived the study, participated in the recruitment of study participants, performed the statistical analysis and produced the first draft. TAG, IEI and CVO recruited the participants, conducted all examinations. In collaboration with OCE and IAOU, CAO defined the protocol for cytology analysis and performed the cytological studies.
KOP and POO reviewed the study design, data analytic plan, the statistical output for accuracy and appropriateness and reviewed all the draft manuscripts for important intellectual content. All Authors read and approved the final manuscript.
(1.) UNAIDS. Global report: UNAIS report on global AIDS epidemic 2013. "UNAIDS I JC2502/ l/E"- Revised and reissued, November 2013. ISBN 978-92-9253-032-7 .http://www.unaids.org/en/media/unaids/contentassets/documents/epidemiology/2013/gr2013/UNAIDS_Gl.
(2.) Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM, Forman D, Bray, F (2013). GLOBOCAN 2012 vl.0, Cancer Incidence and Mortality Worldwide: IARC Cancer Base No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer. Available from http://globocan.iarc.fr.
(3.) Kreitchmann R, Bajotto H, da Silva DA, Fuchs SC. Squamous intraepithelial lesions in HIV -infected women: prevalence, incidence, progression and regression. Arch Gynecol Obstet. 2013 ;288(5): 1107-13.
(4.) Purtilo, D. T. "Opportunistic cancers in patients with immunodeficiency syndromes." Archives of pathology & laboratory medicine 1987; 111(12): 1123-1129.
(5.) Lewden C,Salmon D,Morlat P,Bevilacqua S,Jougla E,Bonnet F,Heripret L,Costagliola D, May T,Chene G. Causes of death among human immunodeficiency virus (HIV)-infected adults in the era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS International Journal of Epidemiology 2005;34:121-130.
(6.) Sasco AJ, Jaquet A, Boidin E, Ekouevi DK, Thouillot F, LeMabec T, Forstin MA, Renaudier P, N'Dom P, Malvy D, Dabis F. The challenge of AIDS-related malignancies in sub-Saharan Africa. PLoS One. 2010 Jan 11;5(l):e8621. doi: 10.1371/journal.pone.
(7.) Mishra GA, Pimple SA, Shastri SS. An overview of prevention and early detection of cervical cancers. Indian J Med Paediatr Oncol. 2011 ; 32(3): 125-132. doi: 10.4103/0971-5851.92808.
(8.) Adewole IF, Benedet JL, Crain BT, Follen M. Evolving a strategic approach to cervical cancer control in Africa. Gynecologic oncology. 2005 Dec 31 ;99(3):S209-12.
(9.) Parkin DM. The global health burden of infection-associated cancers in the year 2002. Int J Cancer 2006;1118: 3030-3044.
(10.) Anorlu RI. Cervical cancer: the sub-Saharan African perspective. Reproductive Health Matters 2008; 16(32):41-49.
(11.) Krishnakumar Duraisamy, K.S. Jaganathan and Jagathesh Chandra Bose. Methods of Detecting Cervical Cancer. Advances in Biological Research 2011; 5 (4): 226-232, 2011.
(12.) Soost, H.J., Lange, H.J., Lehmacher, W. and Ruffing-Kullmann, B. The validation of cervical cytology. Sensitivity, specificity and predictive values. Acta cytologica 1990; 35(1):8-14.
(13.) Holschneider CH, Ghosh K, Montz FJ. See-and-treat in the management of high-grade squamous intraepithelial lesions of the cervix: a resource utilization analysis. Obstet Gynecol 1999;94:377-85. doi: http://dx.doi.org/10.1016/S0029-7844(99) 00337-3 PMID: 10472863.
(14.) Cronje HS, Parham GP, Cooreman BF, et al. A comparison of four screening methods for cervical neoplasia in a developing country. Am J Obstet Gynecol 2003; 188:395-400.
(15.) Federal Ministry of Health, Nigeria (FMoH): National guidelines for HIV and AIDs treatment and care in adolescents and adults. Abuja Nigeria: FMoH; 2010.
(16.) Ezechi OC, Gab-Okafor CV, Ostergren PO, Odberg Pettersson K. Willingness and acceptability of cervical cancer screening among HIV positive Nigerian women. BMC Public Health 2013, 13:46 doi: 10.1186/1471-2458-13-46.
(17.) Huchko MJ, Sneden J, Leslie HH, Abdulrahim N, Maloba M, Bukusi E, Cohen CR. A comparison of two visual inspection methods for cervical cancer screening among HIV-infected women in Kenya. Bull World Health Organ 2014;92: 195-1203 doi: http://dx.doi.org/10.2471/BLT.l3.122051.
(18.) Sankaranarayanan R, Wesley L, Thara S, Dhakad N, Chandralekha B,Sebastian P, Chithrathara K, Parkin DM, Nair MK. Test characteristics of visual inspection with 4% acetic acid (VIA) and Lugol's iodine (VILI) in cervical cancer screening in Kerala, India Int. J Cancer 2003; 106:404-408.
(19.) Parkin DM. Screening for cervix cancer in developing countries. In: Miller AB, Chamberlain J, Day NE, Hakama M, Prorok PC, editors. Cancer screening. Cambridge: Cambridge University Press; 1991; 184-98.
(20.) Prevention and Control of Cervical Cancer in the East and Southern Africa Region. Meeting Report. 29 March to 1 April 1998, Nairobi, Kenya. In: Abwao S, Greene P, Sanghvi H, Tsu V, Winkler J, editors.
(21 .) Adefuye PO, Broutet NJ, de San jose S, Denny LA, Trials and projects on cervical cancer and human papillomavirus prevention in sub-Saharan Africa. Vaccine. 2013 Dec 29;31 Suppl 5:F53-9. doi: 10.1016/j.vaccine.2012.06.070.
(22.) Sankaranarayanan R, Basu P, Wesley RS, Mahe C, Keita N, Mbalawa CC, Sharma R, Dolo A, Shastri SS, Nacoulma M, Nayama M, Somanathan T, Lucas E, Muwonge R, Frappart L, Parkin DM. Accuracy of visual screening for cervical neoplasia: Results from an IARC multicentre study in India and Afr ica. Int J Cancer. 2004 Jul20;110(6):907-13.
(23.) Akinwuntan AL, Adesina OA, Okolo CA, Oluwasola OA, Oladokun A, Ifemeje AA, Adewole IF. Correlation of cervical cytology and visual inspection with acetic acid in HIV-positive women. J Obstet Gynaecol. 2008;28:638-641.
(24.) Perkins MD, Cunningham J. Facing the Crisis: Improving the Diagnosis of Tuberculosis in the HIV Era. The Journal of Infectious Diseases 2007; 196:S15-27.
(25.) Federal Ministry of Health. National policy on cervical cancer prevention and control. Federal Ministry of Health in Nigeria, Abuja Nigeria. 2011.
(26.) Ezechi OC, Odberg Pettersson K, Gbajabiamila T A, Idigbe IE, Kuyoro O, Ujah lAO, Ostergren PO. Predictors of default from follow-up care in a cervical cancer screening program using direct visual inspection in south-western Nigeria. BMC Health Services Research 2014, 14:143.
(27.) National population commission, Nigeria. 2006 Population & Housing Census. http://www.population.gov.ng/.
(28.) Wong HB, LIM GH. Measures of Diagnostic Accuracy: Sensitivity, Specificity, PPV and NPV. Proceedings of Singapore Healthcare 2011 ;20(4):316-318.
(29.) Glick SB, Clarke AR, Blanchard A, Whitaker AK. Cervical Cancer Screening, Diagnosis and Treatment Interventions for Racial and Ethnic Minorities: A Systematic Review J Gen Intern Med. 2012 ; 27(8): 1016-1032.
(30.) Mwanahamuntu MH, Sahasrabuddhe VS, Blevins M, Kapambwe S, Shepherd BE, Chibwesha C, Pfaendler KS, Mkumba G, Vwalika B, Hicks ML, Vermund SH, Stringer JSA, Parham GP. Utilization of Cervical Cancer Screening Services and Trends in Screening Positivity Rates in a 'Screen-And-Treat' Program Integrated with HIV/AIDS Care in Zambia. PLoS One. 2013; 8(9): e74607. 2013. doi: 10.1371/joumal.pone.0074607.
Oliver Chukwujekwu Ezechi (*1, 2), Karen Odberg Petterson (2), Titilola A Gbajabiamila1, Ifeoma Eugenia Idigbe (1), Chidinma Vivian Gab-Okafor (1), Clement Abu Okolo (3), Innocent Achaya Otobo Ujah (1) and Per Olaf Ostergren (2)
Clinical Sciences Division, Nigerian Institute of Medical Research, Yaba Lagos, Nigeria (1); Social Medicine and Global Health, Faculty of Medicine, Lund University, Malmo Sweden (2); Department of Pathology, College of Medicine, University of Ibadan, Ibadan Nigeria (3)
(*) For Correspondence: E-mai/:email@example.com; Phone: +2348033065683
Table 1: Baseline Characteristics of the 1133 Participants that Completed the Study Variables All VIA ARM Number of participants 1133[100.0] 572[50.5] Age(years) Median[IQR] 37[31-45] 38[31-45] 18 -24 87[7.7] 45[7.9] 25 - 39 538[47.5] 272[47.6] 40 and above 508[44.8] 255[44.5] Parity Median[IQR] 2[0-4] 2[0-4] 0 193[17.0] 99[17.3] 1 - 4 622[54.9] 321[56.1] 5 and above 318[28.1] 152[26.6] Marital Status Unmarried 38[33.5] 203[35.5] Married 75[66.5] 369[64.5] Educational status Less than secondary Education 25[22.8] 118[20.6] Secondary education and above 875[77.2] 454[79.4] Work Status Employed 97[85.7] 485[84.8] Unemployed 162[4.3] 87[15.2] Tribal Group Major Southern Tribes 68[60.1] 344[60.1] Northern Tribes 231[20.4] 120[21.0 Southern Minority Tribes 221[19.5] 108[18.9] Area of Residence Urban 422[37.2] 208[36.4] Rural 711[62.8] 364[63 .6] Life time sexual partners Median[IQR] 2[1-4] [1-4] 1 375[33.1] 192[33.6] 2 and above 75[66.9] 380[66.4] Age at first intercourse Median[IQR] 20[18-22] 20[17-22] Less than 10 87[7.7] 42[7.3] 10 - 15 86[7.6] 49[8.6] 16 and above 960[84.7] 481[84.1] HIV status Positive 531[46.9] 266[46.5] Negative 602[53.1] 306[53.5] CD4 cell count Median [IQR] (*) 497[33-684] 521[333-720] Less than 200 45[8.5] 22[8.3] 200 and above 486[91.5] 244[91.7] Viral load Median [IQR] (*) 200[200-3720] 200[200-4302] Less than 1000 386[72.7] 197[74.1] 1000 and above 145[27.3] 69[25.9] HIV Treatment status (*) Not on treatment 108[20.3] 54[20.3] On treatment 423[79.7] 212[79.7] Cervical cell abnormality 220[19.4] 101 [17.7] Variables VILIARM P value Number of participants 561[49.5] Age(years) Median[IQR] 36[31-44] 0.96 18 -24 42[7.5] 25 - 39 266[47.4] 40 and above 253[45.1] Parity Median[IQR] 2[0-4] 0 94[16.8] 1 - 4 301[53.7] 0.53 5 and above 166[29.5] Marital Status Unmarried 177[31.6] 0.18 Married 384[68.5] Educational status Less than secondary Education 140[25.0] 0.10 Secondary education and above 421[75 .0] Work Status Employed 486[86.6] 0.42 Unemployed 75[13.4] Tribal Group Major Southern Tribes 337[60.1] 0.81 Northern Tribes 111[19.8] Southern Minority Tribes 113[20.1] Area of Residence Urban 214[38.1] Rural 347[61.9] 0.58 Life time sexual partners Median[IQR] 2[1-4] 1 183[32.6] 0.78 2 and above 378[67.4] Age at first intercourse Median[IQR] 19[18-21] Less than 10 45[8.0] 0.43 10 - 15 37[6.6] 16 and above 479[85.4] HIV status Positive 265[47.2] 0.85 Negative 296[52.8] CD4 cell count Median [IQR] (*) 494[354-667] Less than 200 23[8.7] 0.98 200 and above 242[91.3] Viral load Median [IQR] (*) 200[200-5534] Less than 1000 189[71.3] 0.54 1000 and above 76[28.7] HIV Treatment status (*) Not on treatment 54[20.3] 0.93 On treatment 211[79 .6] Cervical cell abnormality 119[21.0] 0.15 (*) Data only for HIV positive participants Table 2: Comparison of VISUAL Inspection Findings with Cytology Results Visual Inspection with acetic acid Cytology Report findings N=572 Normal(%) Positive(%) N=489 N=83 Normal (471) 467(99.2) 4(0.8) ASCUS(49) 13(26.5) 36(73.5) LSIL(36) 7(19.4) 29(80.6) HSIL/ Invasive 1(6.7) 14(93.3) carcinoma(15) Atypical 1(100.0) 0(0.0) Glandular(1) Visual Inspection with acetic acid Cytology Report findings N=568 Normal(%) Positive(%) N=433 N=135 Normal(449) 398(88.6) 49(11.4) ASCUS(74) 25(33.8) 50(66.2) LSIL(31) 9(29.0) 23(71.0) HSIL/ Invasive 1(7.1) 13(92.9) carcinoma(l4) Atypical 0(0.0) 0(0.0) Glandular(O) Table 3: Test performance of Visual Inspection with Acetic Acid (VIA) and with Lugol's Iodine (VILI) in Detecting Cervical Cytology Diagnosed Squamous Intraepithelial Lesion Test performance Sensitivity (%:95% Cl) Squamous Intraepithelial Lesion cut off point @ ALL ** VIA 82.7[80.9-89.3] ** VILI 87.5[70.3-89.9] @ HIV Positive s only ** VIA 81.9[79.3-96.1] ** VILI 77.2[64.9-84.2] @ HIV negatives only ** VIA 90.0[86.7 -98.9] ** VILI 86.7[84.3-94.1] High Grade Squamous lntraepithelial Lesion cut off point @ ALL ** VIA 93.3[89.2-98.9] ** VILI 92.9[83.4-98.3] @ HIV Positive s only ** VIA 90.3[83.2-99.3] ** VILI 82.4[71.9-93.2] @ HIV negatives only ** VIA 96.0[95.8-1 00.0] ** VILI 80.0[72.5-92.3] @ CD4 cell count <200 (*) ** VIA 71.3[61.2-85.1] ** VILI 70.0[68.8-88.3] @ CD4 cells count [greater than or equal to]200 (*) ** VIA 93.8[84.5-99.1] ** VILI 87.5[75.9-94.5] Test performance Specificity(%:95% Cl) Squamous Intraepithelial Lesion cut off point @ ALL ** VIA 99.2[93.1-100.0] ** VILI 99.2[89.1-99.4] @ HIV Positive s only ** VIA 93.1[90.3-100.0] ** VILI 71.2[63.9-87.4] @ HIV negatives only ** VIA 97.2[86.7-100] ** VILI 94.0[81.4-98.6] High Grade Squamous lntraepithelial Lesion cut off point @ ALL ** VIA 99.2[89.8-100.0] ** VILI 89.6[81.4-91.9] @ HIV Positive s only ** VIA 93.1[87.1-100] ** VILI 71.3[65.3-79.9] @ HIV negatives only ** VIA 99.7[87.5-100.0] ** VILI 80.0[67.8-94.9] @ CD4 cell count <200 (*) ** VIA 88.2[76.3-93.1] ** VILI 66.9[57.8-74.9] @ CD4 cells count [greater than or equal to]200 (*) ** VIA 98.5[86.9-100.0] ** VILI 73.4[65.1-83.9] Test performance Positive Predictive Value(%: 95% Cl) Squamous Intraepithelial Lesion cut off point @ ALL ** VIA 91.5[86.3-97.2] ** VILI 77.7[72.1-85.8] @ HIV Positive s only ** VIA 86.4[79.5-91.3] ** VILI 52.4[48.9-58.3] @ HIV negatives only ** VIA 89.3[82.3-96.3] ** VILI 63.4[55.9-76.1] High Grade Squamous lntraepithelial Lesion cut off point @ ALL ** VIA 87.8[73.4-93.7] ** VILI 22.0[20.9-23.4] @ HIV Positive s only ** VIA 85.3[73.6-87.3] ** VILI 19.9[17.7-22.1] @ HIV negatives only ** VIA 85.7[81.7-90.7] ** VILI 61.5[55.3-67.8] @ CD4 cell count <200 (*) ** VIA 83.3[79.3-90.1] ** VILI 46.7[41.9-50.3] @ CD4 cells count [greater than or equal to]200 (*) ** VIA 93.4[89.5-98.7] ** VILI 57.5[53.4-61.9] Test performance Negative Predictive Value((%: 95% Cl) Squamous Intraepithelial Lesion cut off point @ ALL ** VIA 98.1[93.2-100.0] ** VILI 99.6[93.5-100.0] @ HIV Positive s only ** VIA 97.1[91.9-100.0] ** VILI 88.4[81.2-95.5] @ HIV negatives only ** VIA 99.6[95.3-100.0] ** VILI 98.3[95.1-100.0] High Grade Squamous lntraepithelial Lesion cut off point @ ALL ** VIA 97.8[90.9-100.0] ** VILI 99.7[96.3-100.0] @ HIV Positive s only ** VIA 94.5[91.6-97.6] ** VILI 97.8[93.5-99.8] @ HIV negatives only ** VIA 98.9[95.2-100.0] ** VILI 99.1[95.6-100.0] @ CD4 cell count <200 (*) ** VIA 88.2[82.4-95.1] ** VILI 50.0[44.8-54.6] @ CD4 cells count [greater than or equal to]200 (*) ** VIA 99.5[96.3-100.0] ** VILI 98.9[94.9-100.0]
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ORIGINAL RESEARCH ARTICLE|
|Author:||Ezechi, Oliver Chukwujekwu; Petterson, Karen Odberg; Gbajabiamila, Titilola A.; Idigbe, Ifeoma Eugen|
|Publication:||African Journal of Reproductive Health|
|Date:||Dec 1, 2016|
|Previous Article:||Gender-Based Violence following social media acquaintance in Nigeria.|
|Next Article:||Exploring the barriers to breast and cervical cancer screening in Nigeria: A narrative review.|