Tropical Medicine and International Health doi:10.1111/j.1365-3156.2007.01888.

volume 12 no 9 pp 10181025 september 2007

Comparison of visual inspection and Papanicolau (PAP) smears

for cervical cancer screening in Honduras: should PAP smears
be abandoned?
R. B. Perkins, S. M. Langrish, L. J. Stern, J. Figueroa and C. J. Simon

Department of Obstetrics and Gynecology, Boston University Medical Center, Boston, MA, USA

Summary objective To compare visual inspection with acetic acid (VIA) to Papanicolau (PAP) smears in a
community setting in a developing nation.
methods Women undergoing cervical cancer screening in Honduras received either VIA and PAP
smears (VIA PAP group) or PAP smears alone (PAP-only group). Local healthcare providers performed
PAP screening. A VIA-trained nurse performed VIA exams. All PAP smears were processed in Honduras.
PAP smears from the VIA PAP group were reviewed in the United States. Women with positive VIA
or PAP tests were offered colposcopy. We compared the relative accuracy of PAP smears and VIA and
the proportions of women completing follow-up colposcopy after positive screening tests.
results In total, 1709 PAP smears were performed including women from both the VIA PAP and PAP-
only groups. Nine PAP smears were positive (0.5%). Three women completed colposcopy (33%). All
three had biopsy-confirmed dysplasia. In the VIA PAP group (n = 339), 49 VIA exams were abnormal
(14%) and two PAP smears were abnormal when read in Honduras (0.6%). When reviewed in the
United States, 14 of the 339 PAP smears were abnormal (4%). Forty women (83%) completed follow-up
colposcopy after a positive VIA exam. Twenty-three had biopsy-proven dysplasia. All 23 dysplasia cases
had negative PAP smear readings in Honduras; four PAP smears were reclassified as positive in the
United States.
conclusions Although few developing countries can maintain high-quality PAP smear programmes,
many governments and charitable organizations support cervical cancer screening programmes that rely
on PAP smears. This study underscores the need to promote alternative technologies for cervical cancer
screening in low-resource settings.

keywords cervical cancer prevention, visual inspection with acetic acid, Papanicolau smear, cervical
cytology, developing countries

coverage in Latin America range from a low of 1520% in

Introduction
In Latin America, 76 000 women develop cervical cancer
and 30 000 succumb to the disease each year (Arrossi et al.
2003). Despite attempts to increase screening over the past
several decades, cancer incidence and mortality rates have
shown almost no improvement between 1960 and 1993
(Robles et al. 1996; Hernandez-Avila et al. 1998; Lazcano-
Ponce et al. 1999). The Papanicolau (PAP) smear is known
to be effective in reducing the population-wide incidence of
invasive cervical cancer when at least 70% of the popu-
lation is screened with good-quality PAP smears on a
regular basis (Laara et al. 1987; Arrossi et al. 2003).
However, screening programs in Latin America suffer both
from inadequate population coverage and poor quality of
existing PAP smear services. Estimates of population
Nicaragua to a high of 6580% in Costa Rica and El
Salvador (Agurto et al. 2005). These numbers may not give
an accurate picture of screening coverage because many
screening programs over-screen younger, more affluent,
low-risk women and fail to screen older, rural, high-risk
populations (Lazcano-Ponce et al. 1999).
Poor PAP smear quality also contributes to screening
program failure. An effective PAP smear program requires
consistent access to supplies, trained providers, reliable
transportation of specimens, a high quality laboratory with
adequate materials, trained cytopathologists, and a quality
control system (PATH 2002). Because these requirements
are not met in many countries, the PAP smears are often of
low quality (Bishop et al. 1996) and can fail to detect up to
90% of disease in some settings (Fahey et al. 1995; Nanda

1018 2007 Blackwell Publishing Ltd

Tropical Medicine and International Health
R. B. Perkins et al. Visual inspection outperforms PAP smears
et al. 2000). Thus even women who have participated in
screening may not be adequately protected from develop-
ing invasive cervical cancer.
Because few developing nations can maintain accessible,
high-quality PAP smear services nationwide, most experts
do not recommend the use of PAP smears in low-resource
settings (Denny 2005; Sankaranarayanan et al. 2004; Gol-
die et al. 2005, 2001). The use of acetic acid during visual
examination of the cervix, termed visual inspection with
acetic acid (VIA), has been advocated as an alternative
screening method to PAP smears in developing countries.
The attractive features of VIA include low cost, simple
administration, immediate availability of results and accu-
racy comparable to good quality PAP smears (Cecchini et al.
1993; Van Le et al. 1993; Megevand et al. 1996; Sankara-
narayanan et al. 1998, 1999; University of Zimbwabwe
JHPIEGO Cervical Cancer Project 1999; Denny 2005).
Despite the recommendation of VIA-based screening
systems over PAP smear-based screening systems in devel-
oping countries (Goldie et al. 2001, 2005; Sankara-
narayanan et al. 2004; Denny 2005), many countries are
reluctant to abandon PAP smear-based systems. Even
countries that have failed to show a reduction in mortality
after two decades of nationalized PAP screening programs
have not considered changing screening modalities at a
national level (Lazcano-Ponce et al. 1999). Some of the
reluctance may be due to the long history of success using the
PAP smear in developed countries. Hesitation may also arise
because VIA has primarily been tested in the setting of large,
well-designed research studies, the results of which may not
be generalizable to community settings. This study compares
PAP smears and VIA in a community setting in Honduras.
The following hypotheses were tested: (1) VIA detects more
cervical dysplasia than PAP smears; (2) Patients screened
with VIA are more likely to complete follow-up colposcopy
than those screened with PAP smears.
Materials and methods
Project setting
This project took place in Honduras, where the estimated
cervical cancer incidence is 39.1 100 000, four times the
rate seen in the United States, and the mortality is
16.8 100 000, nearly six times higher than in the United
States (U.S. Department of Health and Human Services
2000, Arrossi et al. 2003). The communities of Yuscaran,
Moroceli and Jicarito in Honduras were chosen because
they represent a rural area where healthcare services,
including PAP smears, are routinely available. Honduras
has no nationwide screening program, but women may
obtain PAP smears in their local health centres, at private
2007 Blackwell Publishing Ltd
volume 12 no 9 pp 10181025 september 2007
clinics, or at special screening days organized by a variety
of private organizations. The total population of these
communities is approximately 30 000, of whom 8000 are
women of reproductive age. Two-thirds of the population
live below the poverty line and half lacks basic sanitation
and potable water. The average person has had 23 years
of schooling; 30% of the population is illiterate (Honduran
State Department 2004).
The data used in this paper were collected during a
cervical cancer outreach and screening project run by a
local non-governmental organization (NGO) in collabor-
ation with the Honduran Ministry of Health and the
Emma Romero de Callejas Cancer Center, a prominent
referral hospital in the capital city of Tegucigalpa. The
primary goal of the project was to increase access to
existing PAP smear services. VIA and secondary slide
review were performed to assess the effectiveness of
existing PAP smear services in detecting cervical dysplasia.
Subjects
Women receiving cervical cancer screening through the
local health centres in Yuscaran, Jicarito and Moroceli
between 1 June 2003 and 31 August 2004 were eligible for
inclusion in this study. Women could receive screening as
part of their routine health care at their local health centres
or through special cervical cancer screening days. Routine
screening was offered for a small fee per protocol of each
local health centre, and was not advertized. Screening days
were advertized by word-of-mouth, fliers and announce-
ments on local radio stations, and services were provided
free-of-charge. All women who requested PAP smears or
VIA received screening, but those who were pregnant or
who had undergone hysterectomy were excluded from the
final analysis.
Screening protocol
All women were screened with PAP smears, the current
standard of care. All women presenting for screening
between 1 June, 2003 and 31 August, 2004 were offered
screening with VIA in addition to PAP smears (VIA PAP
group). VIA was presented to patients as an optional test
for cervical cancer that was offered in addition to their
routine PAP smear. Because this project was designed for
the purpose of community service, not for research, written
informed consent was not obtained. All patients with an
abnormal result on PAP smear or VIA were offered
colposcopy with biopsy, the gold standard for diagnosing
cervical dysplasia.
Papanicolau smears were performed by local nurses,
doctors, or trained paraprofessional staff. All VIA was
1019
Tropical Medicine and International Health
R. B. Perkins et al. Visual inspection outperforms PAP smears
performed by one nurse who received VIA training
according to the standards developed by the International
Agency for Research of Cancer (IARC) (Sellors & San-
karanarayanan 2003). All colposcopies were performed by
physicians from the Emma Romero de Callejas Cancer
Center.
All PAP smears were obtained using Ayre spatulas and
spray fixative, and all were processed through central
screening laboratories in the capital city. Tests were read
by cytotechnologists in accordance with the protocol of the
processing centre. PAP results were reported according to
the 2001 Bethesda system, and considered positive if
atypical squamous cells of undetermined significance
(ASCUS) or any squamous intraepithelial lesion (SIL) were
reported. In accordance with the protocol of the Honduran
cytology laboratory, all positive PAP smear samples and
10% of the negative samples were reviewed by a patho-
logist affiliated with the Honduran Ministry of Health. All
biopsy samples were processed by the pathology laboratory
used by the Ministry of Health. All PAP smears performed
in the VIA PAP group were reviewed at Brigham and
Womens Hospital, a Harvard-affiliated teaching hospital
in the United States, and the results were compared with
those obtained in Honduras.
Data analysis
This study was a retrospective analysis of data that had
been collected during the cervical cancer outreach and
screening project. During the project, medical record data
including demographic information, prior screening his-
tory, PAP smear, VIA, colposcopy and biopsy results were
collected prospectively and kept in a Microsoft Excel
spreadsheet for the use of local health providers. The use of
these anonymized medical record data for research pur-
poses was approved by the Institutional Review Board at
Boston Medical Center. Descriptive statistics were used to
analyse the demographic data and to describe the results of
each screening test. McNemars statistic was used to
compare the relative accuracy of PAP and VIA. Chi-
squared tests were used to compare the proportions of
women completing diagnostic and treatment recommen-
dations in each group.
Results
Demographics
Between 1 June 2003 and 31 August 2004, a total of 1709
women underwent screening with either PAP smear plus
VIA or PAP smear alone. During the first year of the
project, 1370 women were screened with PAP smears only
1020
volume 12 no 9 pp 10181025 september 2007
(PAP-only group). During the last 3 months of the project,
339 women were screened with both VIA and PAP smears
(VIA PAP group). All women who were offered VIA in
addition to routine PAP smear screening agreed to undergo
both tests. Table 1 compares the demographic character-
istics of these groups. Women did not differ on average
age, parity, number of lifetime sexual partners, use of birth
control, or cigarette smoke exposure. Women in the
VIA PAP group initiated sexual intercourse slightly
younger, were more likely to be living with a partner and
unmarried, and were more likely to be exposed to wood
smoke than women in the PAP-only group. Ninety per cent
of women in the VIA PAP group and 89% of women in the
PAP-only group reported having had at least one PAP
smear in the past, and 76% of the VIA PAP group and
73% of the PAP-only group had been screened within the
past year.
Comparison of screening tests
Visual inspection with acetic acid greatly increased the
detection of cervical dysplasia compared with PAP smears
(Table 2). Of the 339 women screened with both VIA and
PAP smears (VIA PAP group), 49 (14%) had positive VIA
tests and 2 (0.6%) had positive PAP smears. Forty women
with positive VIA tests completed follow-up colposcopy,
and 23 (58%) of them had biopsy-confirmed dysplasia.
None of those 23 women had a positive PAP smear as read
in Honduras, indicating that at least 23 cases of dysplasia
were missed by PAP smear. One of the two patients with a
positive PAP smear had confirmed dysplasia on biopsy but
a negative VIA exam. The sensitivity and specificity cannot
be calculated directly because colposcopy was not per-
formed on women with negative tests. However, the
relative sensitivity and specificity can be compared using
McNemars statistic, which compares only the discordant
cells within each of the diseased and non-diseased groups.
Using McNemars statistic, VIA was significantly more
accurate for the detection of cervical dysplasia than PAP
(P < 0.0001).
When the 339 PAP smear samples from the VIA PAP
group were reviewed at Brigham and Womens Hospital in
the United States, 14 were positive, including one of the
two that was read as positive in Honduras. The screening
project had been completed by the time these results were
obtained, so colposcopic follow-up was not available for
all patients. Six of the 14 patients had previously under-
gone colposcopy due to an abnormal result on either VIA
or the Honduran PAP smear reading, and four had
dysplasia confirmed on biopsy.
Of the remaining 1370 women screened only with PAP
smears (PAP-only group), seven had positive tests (0.5%).
2007 Blackwell Publishing Ltd
Tropical Medicine and International Health volume 12 no 9 pp 10181025 september 2007

R. B. Perkins et al. Visual inspection outperforms PAP smears

Table 1 Demographic characteristics

PAP VIA PAP-only P-value
(n = 339), (n = 1004*), (students
Continuous Variables Mean SD Mean SD t-test)

Age 36.8 13.3 37.3 12.6 0.53

Age at first sexual intercourse 17.7 3.0 18.1 3.5 0.06
Parity 4.5 3.0 4.3 2.8 0.35
Number of lifetime sexual partners 1.5 0.7 1.6 1.0 0.64
Categorical variables Percent (n) Percent (n) Chi-squared test
Birth control
None 23% (79) 26% (251) 0.13
Tubal Ligation 20% (67) 24% (231)
Depoprovera 16% (54) 16% (159)
Oral contraceptive pills 11% (37) 7% (66)
Intrauterine device 13% (44) 10% (96)
Menopause 13% (45) 15% (147)
Condoms 2% (6) 2% (18)
Other 2% (7) 2% (15)
Marital status
Married 26% (89) 32% (309) 0.002
Single 15% (49) 20% (197)
Living with partner 59% (200) 48% (473)
Cigarette exposure 33% (111) 30% (204) 0.4
Wood smoke exposure 96% (321) 79% (536) <0.0001
Prior PAP smear
No 10% (34) 11% (108) 0.6
Yes 90% (302) 89% (861)
Date of most recent prior PAP smear
1 year ago 76% (231) 73% (625) 0.12
23 years ago 15% (44) 13% (114)
3 years ago 9% (27) 14% (122)
Result of most recent prior PAP smear
Abnormal 1% (5) 1% (9) <0.0001
Did not receive unsure of result 14% (42) 18% (155)
Infection 1% (3) 9% (75)
Inflammation 26% (80) 21% (186)
Negative 57% (174) 52% (452)

*Includes data for 1004 women with available demographic information. Women for
whom only PAP results were available (n = 366) were not included in the demographic
analysis.
Other birth control methods included abstinence, rhythm method, withdrawal, lacta-
tional amenorrhea and vasectomy.
Five women total in both groups were smokers; all others were exposed to secondhand
smoke.
Wood smoke exposure from cooking in unventilated kitchens increases cervical dysplasia
risk similarly to cigarette smoke (Ferrera et al. 2000).

Two women completed follow-up, and both had biopsy-
confirmed dysplasia.
Completion of follow-up colposcopy
The proportion of women who completed follow-up col-
poscopy after receiving an abnormal PAP smear result was
compared with the proportion that completed follow-up
after an abnormal VIA examination. Forty-eight women
were positive on VIA only and 40 completed follow-up, a
compliance rate of 83%. Nine of the 1709 PAP smears
processed in Honduras were screen positive, and three of
these women completed follow-up. One of the nine women
was positive on both VIA and PAP smear. She was excluded
from the compliance analysis as her follow-up could indicate
response to either a positive PAP or a positive VIA. Of the
remaining eight women, three completed follow-up, giving a
compliance rate of 38% (Table 2).

2007 Blackwell Publishing Ltd 1021

Tropical Medicine and International Health volume 12 no 9 pp 10181025 september 2007

R. B. Perkins et al. Visual inspection outperforms PAP smears

Table 2 Comparison of Papanicolau (PAP) smear and visual inspection with acetic acid (VIA): accuracy and loss to follow-up

Total Screen Completed follow-up Confirmed dysplasia

Screening group population (n) Screening test positive* (n) colposcopy (n) on biopsy (n)

VIA PAP group 339 VIA 49 40 23

PAP: results from Honduras 2 1 1
PAP: review in United States 14 NA 4
PAP-only group 1370 PAP: results from Honduras 7 2 2

*The two PAP smears in the VIA PAP group that were read as positive in Honduras were low-grade squamous intraepithelial lesions
(LSIL). Of the 14 PAP smears in the VIA PAP group called positive on review in the United States, 10 were atypical squamous cells of
undetermined significance (ASCUS), two were atypical squamous cells-probable high grade (ASC-H), and two were squamous intraepi-
thelial lesion-unable to classify (SIL-UNC). Of the seven positive PAP smears in the PAP-only group, five were ASCUS, one was LSIL, and
one was high-grade squamous intraepithelial lesion (HSIL).
Of the three women with biopsy-confirmed dysplasia diagnosed by PAP smear read in Honduras, two had cervical intraepithelial
neoplasia (CIN) grade I, and 1 had CINIII. Of the 23 women with biopsy-confirmed dysplasia diagnosed by VIA, 21 had CINI and two had
CINIII.

facilities, so all cytopathologists working in Honduras

Discussion
The data from this community-based study indicate that
PAP smears are ineffective in Honduras. When VIA was
used to screen a population of 339 women, 23 cases of
biopsy-confirmed dysplasia were discovered, all of which
were missed when the PAP smears were interpreted in
Honduras. The poor PAP smear quality is especially
unfortunate in these communities because women are
committed to screening. Ninety per cent of women
participating in this screening project had been screened at
least once in the past, and a previous survey of women in
these communities revealed that 83% were participating in
screening (data not shown). With more than 80% of
women participating in screening, cervical cancer rates
should decrease dramatically in these communities (Laara
et al. 1987). But because the PAP smears are not detecting
pre-cancerous lesions, these women are not protected
against the development of invasive cervical cancer.
The poor performance of PAP smears in Honduras is
likely due to a combination of material and personnel
factors. In Honduras, the Ayre spatula alone is commonly
used for specimen collection, and recent literature indicates
that sampling using an extended tip spatula or Ayre spatula
plus cytobrush is more effective (Martin-Hirsch et al.
1999). However, many steps of the cervical cancer
screening process in Honduras PAP collection, slide
fixation, slide preparation and slide review suffer from
lack of supplies and lack of adequately trained personnel.
Due to limited resources, cytobrushes are rarely used in
slide collection, and when Ayre spatulas are unavailable,
wooden tongue depressors may be used. Many of the stains
are diluted and or recycled to extend their use, making
slides more difficult to interpret. In addition, Honduras
does not currently have any cytopathologist training
during the study period were trained in neighbouring
countries and did not have regular access to continuing
education. When cytopathologists in the United States
reviewed the Honduran PAP smears, they commented on
the poor quality of cell fixation and staining, but were able
to identify several positive PAP tests that had been read as
negative in Honduras. Thus both slide preparation and
slide interpretation appear to contribute to the high false-
negative rate.
Because both PAP smears and VIA are screening tests,
they are only successful in preventing the development of
invasive cancer when the appropriate diagnostic and treat-
ment procedures are completed following a positive test. In
this study, women were more than twice as likely to comply
with follow-up when screened with VIA compared with
PAP smears (83% vs. 38%). The difference in follow-up
may be due to the immediate communication of results with
VIA. Because the results of the PAP smear are not imme-
diately available, each woman must return to the health
centre for a second visit to receive her results. The
immediate availability of results with VIA not only elimin-
ates the inconvenience of an additional clinic visit, but it
ensures that every woman knows whether her examination
was positive or negative and allows frank, in-the-moment
discussion between the woman and her healthcare provider
about the importance of completing timely follow-up.
When the women presenting for screening in this study were
asked about the results of their prior PAP smears, between
14% and 18% were unaware of the results, and were
therefore unsure if follow-up had been indicated.
Due to its favourable follow-up rate, low cost, high
sensitivity and ability to be combined with immediate
treatment (Mandelblatt et al. 2002; Denny 2005), VIA is
favoured over PAP smear screening in developing nations

1022 2007 Blackwell Publishing Ltd

Tropical Medicine and International Health
R. B. Perkins et al. Visual inspection outperforms PAP smears
even when the detection of dysplasia is equivalent between
the two methods (Sankaranarayanan et al. 2004; De Vuyst
et al. 2005; Denny 2005). VIA does have a higher false-
positive rate than PAP smear, however, which results in
more referrals to colposcopy and can result in overtreat-
ment (Clayes et al. 2003). In addition, performing VIA is
more difficult than performing a PAP smear, so more
provider training and supervision is needed for a pro-
gramme to be successful. Introducing VIA at a national
level would require a significant investment in the training
and ongoing education of community nurses.
Improving the quality of PAP smears by using higher
quality materials and providing training for cytopatholo-
gists is also an option, and these efforts may be successful
in the short-term. Over time, however, VIA may prove to
be more cost-effective than PAP smears because the cost
per VIA screen is considerably less than the cost per PAP
smear. Compared with the PAP smear, VIA is considered a
very cost-effective intervention (Goldie et al. 2001, 2005;
Legood et al. 2005), comparable with vaccinating against
Hepatitis B in India or using bed nets for malaria
prevention in Kenya (Goldie et al. 2005). A full cost-
effectiveness analysis was not performed in this case.
However, the cost per PAP smear, based on data from the
Emma Romero de Callejas Cancer Center, was $3 (US),
while the cost per VIA exam, estimated using the direct
purchase price of the materials and the average hourly
wage of a Ministry of Health nurse, was $0.22 (US).
Because this study relied on medical record data gath-
ered during a community outreach project, it has a number
of limitations. The true prevalence of disease and the
sensitivity and specificity of PAP and VIA in rural
Honduras remain unknown as colposcopy was not
performed on all patients. In addition, efforts were made
to facilitate dissemination of results and to provide
transportation to diagnostic and treatment services that
were provided free-of-charge. This may have artificially
increased the numbers of women complying with follow-
up over what would be observed without these services.
We do not know the exact percentage of women who
participate in regular screening because data from health
centres and private organizations that were not participa-
ting in this project were unavailable. However, previous
work indicates that the majority of women do participate
in screening. Furthermore, because all PAP smears are
processed through the same laboratories, we would expect
to find similar results if these additional data were
included. The final limitation is that one provider was
trained during the course of the project and she performed
all VIA examinations. As VIA is provider-dependent, the
results must be replicated with other providers. However,
the positive predictive value of VIA obtained in this study
2007 Blackwell Publishing Ltd
volume 12 no 9 pp 10181025 september 2007
was consistent with other published studies in Latin
America (Claeys et al. 2003; Jeronimo et al. 2005).
Conclusions
Numerous research studies support the use of VIA over
PAP smears for cervical cancer screening in developing
countries. However, most countries continue to use PAP
smears for the majority of cervical cancer screening, and
many international aid organizations working in devel-
oping countries continue to promote the use of PAP
smears. This study found that VIA was considerably more
accurate than PAP smears and resulted in greater
compliance with follow-up in a community setting in a
developing nation. Ministries of Health and international
charitable organizations should increase support for
alternative technologies for cervical cancer screening in
developing countries.
Acknowledgements
We thank the staff of the Emma Romero de Callejas
Cancer Center and of the Honduran Ministry of Health
clinics in Jicarito, Moroceli and Yuscaran. Funding for the
cervical cancer prevention programme was provided by the
Alliance for Cervical Cancer Prevention Small Grants
Program. Funding for research was provided by the
Building Interdisciplinary Careers in Womens Health
(BIRCWH) Scholar Program through the United States
National Institutes of Health.
References
University of Zimbabwe JHPIEGO Cervical Cancer Project
(1999) Visual inspection with acetic acid for cervical-cancer
screening: test qualities in a primary-care setting. Lancet 353,
869873.
U.S. Department of Health and Human Services (2000) Healthy
People 2010, Vol 1 Cancer. http://www.healthypeople.gov/
document/HTML/Volume1/03Cancer.htm (accessed July 23,
2007).
Agurto I, Arrossi S, White S et al. (2005) Involving the community
in cervical cancer prevention programs. International Journal of
Gynaecology and Obstetrics 89(Suppl. 2), S38S45.
Arrossi S, Sankaranarayanan R & Parkin DM (2003) Incidence
and mortality of cervical cancer in Latin America. Salud Publica
de Mexico 45(Suppl. 3), S306S314.
Bishop A, Sherris J, Tsu VD & Kilbourne-Brook M. (1996) Cer-
vical dysplasia treatment: key issues for developing countries.
Bulletin of the Pan American Health Organization 30, 378386.
Cecchini S, Bonardi R, Mazzotta A, Grazzini G, Iossa A & Ciatto
S. (1993) Testing cervicography and cervicoscopy as screening
tests for cervical cancer. Tumori 79, 2225.
1023
Tropical Medicine and International Health
R. B. Perkins et al. Visual inspection outperforms PAP smears
Claeys P, De Vuyst H, Gonzalez C, Garcia A, Bello RE &
Temmerman M (2003) Performance of the acetic acid test
when used in field conditions as a screening test for cervical
cancer. Tropical Medicine and International Health 8, 704
709.
De Vuyst H, Claeys P, Njiru S et al. (2005) Comparison of pap
smear, visual inspection with acetic acid, human papillomavirus
DNA-PCR testing and cervicography. International Journal of
Gynaecology and Obstetrics 89, 120126.
Denny L. (2005) The prevention of cervical cancer in developing
countries. BJOG 112, 12041212.
Fahey MT, Irwig L & Macaskill P. (1995) Meta-analysis of Pap
test accuracy. American Journal of Epidemiology 141, 680689.
Ferrera A, Velema JP, Figueroa M et al. (2000) Co-factors related
to the causal relationship between human papillomavirus and
invasive cervical cancer in Honduras. International Journal of
Epidemiology 29, 817825.
Goldie SJ, Kuhn L, Denny L, Pollack A & Wright TC. (2001)
Policy analysis of cervical cancer screening strategies in low-
resource settings: clinical benefits and cost-effectiveness. JAMA
285, 31073115.
Goldie SJ, Gaffikin L, Goldhaber-Fiebert JD et al. (2005) Cost-
effectiveness of cervical-cancer screening in five developing
countries. New England Journal of Medicine 353, 21582168.
Hernandez-Avila M, Lazcano-Ponce EC, de Ruiz PA & Romieu I
(1998) Evaluation of the cervical cancer screening programme in
Mexico: a population-based casecontrol study. International
Journal of Epidemiology 27, 370376.
Honduran State Department (2004) Presidential report by the
Secretary of State of Honduras: Appendix 4Information by
department and municipality. Honduran State Department,
Tegucigalpa, Honduras, 137.
Jeronimo J, Morales O, Horna J et al. (2005) Visual inspection
with acetic acid for cervical cancer screening outside of low-
resource settings. Revista Panamericana de Salud Publica 17,
15.
Laara E, Day NE & Hakama M (1987) Trends in mortality from
cervical cancer in the Nordic countries: association with
organised screening programmes. Lancet 1, 12471249.
Lazcano-Ponce EC, Moss S, Alonso de Ruiz P, Salmeron Castro J
& Hernandez Avila M. (1999) Cervical cancer screening in
developing countries: why is it ineffective? The case of Mexico
Archives of Medical Research 30, 240250.
Corresponding Author Rebecca Perkins, Department of Obstetrics and Gynecology, Boston University Medical Center, 85 E.
Concord St. 6th Floor, Boston, MA 02118, USA. Tel.: +1 617 710 7446; Fax +1 617 277 2370; E-mail rebecca.perkins@bmc.org.
1024
volume 12 no 9 pp 10181025 september 2007
Legood R, Gray AM, Mahe C et al. (2005) Screening for cervical
cancer in India: How much will it cost?. A trial based analysis of
the cost per case detected International Journal of Cancer 117,
981987.
Mandelblatt JS, Lawrence WF, Gaffikin L et al. (2002) Costs and
benefits of different strategies to screen for cervical cancer in
less-developed countries. Journal of the National Cancer Insti-
tute 94, 14691483.
Martin-Hirsch P, Lilford R, Jarvis G et al. (1999) Efficacy of
cervical-smear collection devices: a systematic review and meta-
analysis. Lancet 354, 17631770.
Megevand E, Denny L, Dehaeck K, Soeters R & Bloch B. (1996)
Acetic acid visualization of the cervix: an alternative to cytologic
screening. Obstetrics and Gynecology 88, 383386.
Nanda K, McCrory DC, Myers ER et al. (2000) Accuracy of the
Papanicolaou test in screening for and follow-up of cervical
cytologic abnormalities: a systematic review. Annals of Internal
Medicine 132, 810819.
PATH (2002) Alliance for Cervical Cancer Prevention Fact Sheet:
Pap smears, an Important but Imperfect Method. http://www.
path.org/files/RH_pap_smears.pdf (accessed July 23, 2007).
Robles SC, White F & Peruga A. (1996) Trends in cervical cancer
mortality in the Americas. Bulletin of the Pan American Health
Organization 30, 290301.
Sankaranarayanan R, Wesley R, Somanathan T et al. (1998) Vis-
ual inspection of the uterine cervix after the application of acetic
acid in the detection of cervical carcinoma and its precursors.
Cancer 83, 21502156.
Sankaranarayanan R, Shyamalakumary B, Wesley R, Sreedevi
Amma N, Parkin DM & Nair MK (1999) Visual inspection with
acetic acid in the early detection of cervical cancer and precur-
sors. International Journal of Cancer 80, 161163.
Sankaranarayanan R, Basu P, Wesley RS et al. (2004) Accuracy of
visual screening for cervical neoplasia: Results from an IARC
multicentre study in India and Africa. International Journal of
Cancer 110, 907913.
Sellors JW & Sankaranarayanan R. (2003) Colposcopy and
Treatment of Cervical Intraepithelial Neoplasia. A Beginners
Manual. International Agency for Research of Cancer, Lyon
Cedex, France.
Van Le L, Broekhuizen FF, Janzer-Steele R, Behar M & Samter T
(1993) Acetic acid visualization of the cervix to detect cervical
dysplasia. Obstetrics and Gynecology 81, 293295.
2007 Blackwell Publishing Ltd
Tropical Medicine and International Health volume 12 no 9 pp 10181025 september 2007

R. B. Perkins et al. Visual inspection outperforms PAP smears

Comparaison de linspection visuelle et du frottis de PAP pour le criblage du cancer cervical au Honduras: les frottis de PAP devraient-ils etre abandonnes ?

objectif Comparer linspection visuelle basee sur lutilisation de lacide acetique (VIA) au frottis de PAP dans une communaute dun pays en voie de
developpement.
methodes Les femmes inclues dans le criblage pour le cancer cervical au Honduras ont subi soit le test VIA et le frottis de PAP (groupe VIA PAP) ou le
frottis de PAP seul (groupe PAP seul). Les fournisseurs locaux de soins de sante ont effectue le criblage par le frottis de PAP. Une infirmiere formee a
effectue les examens VIA. Tous les frottis de PAP ont ete realises au Honduras. Les frottis de PAP du groupe VIA PAP ont ete re-analyses aux Etats-
Unis. Les femmes avec un resultat positif par le test VIA ou le test PAP ont recu une colposcopie. Nous avons compare la precision relative des frottis de
PAP et du VIA et les proportions de femmes qui ont complete le suivi colposcopique a la suite des tests de criblage positifs.
resultats 1709 frottis de PAP ont ete realises comprenant des femmes issues du groupe VIA PAP et du groupe PAP seul. Neuf frottis de PAP etaient
positifs (0,5%); trois femmes ont suivi la colposcopie (33%). Toutes les 3 avaient une biopsie confirmee de dysplasie. Dans le groupe VIA PAP
(n = 339), 49 examens VIA (14%) et 2 (0,6%) frottis de PAP etaient anormaux selon les resultats de lecture au Honduras. Lorsque reexamines aux
Etats-Unis, 14 des 339 frottis de PAP etaient anormaux (4%). Quarante femmes (83%) ont complete le suivi colposcopique suite a un examen VIA
positif. Vingt-trois dentre elles avaient une biopsie confirmee de dysplasie. Tous les 23 cas de dysplasie avaient des resultats negatifs pour le frottis de
PAP au Honduras; 4 frottis de PAP ont ete reclasses comme etant positifs aux Etats-Unis.
conclusions Bien que peu de pays en voie de developpement puissent maintenir des programmes de frottis de PAP de haute qualite, beaucoup de
gouvernements et organisations de charite soutiennent les programmes de criblage du cancer cervical bases sur le frottis de PAP. Cette etude souligne la
necessite de favoriser des technologies alternatives pour le criblage du cancer cervical dans des regions a faibles ressources.

mots cles prevention du cancer cervical, inspection visuelle a lacide acetique (VIA), frottis de PAP, cytologie cervicale, pays en voie de developpement

Comparacion de la inspeccion visual y el frotis de PAP para el cribado de cancer de cerviz en Honduras: deberan abandonarse los frotis de PAP?

objetivo Comparar la inspeccion visual con acido acetico (IVA) con los frotis de PAP en una comunidad de un pas en vas de desarrollo.
metodos Las mujeres que estaban pasando un cribado para cancer de cerviz recibieron IVA o frotis de PAP (grupo VIA PAP), o solo frotis de PAP
(grupo solo-PAP). Los trabajadores sanitarios locales realizaron las pruebas de PAP. Una enfermera entrenada en VIA realizo estas pruebas. Todos los
frotis PAP fueron procesados en Honduras. Los frotis del grupo VIA PAP fueron posteriormente revisados en los Estados Unidos. A las mujeres con una
prueba VIA o PAP positivas se les ofrecio una colposcopa. Se comparo la precision relativa de las pruebas de PAP y VIA, as como la proporcion de
mujeres que completaban el seguimiento con colposcopa despues de una prueba de cribaje positiva.
resultados Se realizaron 1709 frotis de PAP incluyendo a mujeres de ambos grupos (VIA PAP y solo-PAP). Nueve frotis de PAP fueron positivos
(0.5%). Tres mujeres completaron la colposcopa (33%) y las 3 tenan displasia confirmada por biopsia. En el grupo VIA PAP (n = 339), 49 pruebas
VIA eran anormales(14%) y 2 frotis de PAP eran anormales cuando su lectura se realizo en Honduras (0.6%). Al ser revisados en los Estados Unidos, 14
de los 339 frotis de PAP eran anormales (4%). Cuarenta mujeres (83%) completaron el seguimiento con colposcopa despues de una prueba VIA
positiva. Veintitres tenan displasia probada mediante biopsia. Los 23 casos de displasia tenan un resultado negativo en los frotis de PAP segun la
lectura realizada en Honduras; 4 de los frotis de PAP fueron reclasificados como positivos en los Estados Unidos.
conclusiones Aunque son pocos los pases en vas de desarrollo que pueden mantener un programa de frotis de PAP de buena calidad, muchos
gobiernos y organizaciones caritativas apoyan los programas de cribaje de cancer de cerviz que se basan en la prueba de PAP. Este estudio apoya la
necesidad de promover tecnicas alternativas para el cribaje de cancer de cerviz en lugares con pocos recursos.

palabras clave prevencion del cancer de cerviz, inspeccion visual con acido acetico (IVA), frotis de PAP, citologa de cerviz, pases en vas de desarrollo

2007 Blackwell Publishing Ltd 1025