The European Journal of Contraception & Reproductive

Health Care

ISSN: 1362-5187 (Print) 1473-0782 (Online) Journal homepage: [Link]

The risk of cervical atypia in oral contraceptive

users

Indira Adhikari, Tiina Eriksson, Tapio Luostarinen, Matti Lehtinen & Dan
Apter

To cite this article: Indira Adhikari, Tiina Eriksson, Tapio Luostarinen, Matti Lehtinen & Dan
Apter (2018): The risk of cervical atypia in oral contraceptive users, The European Journal of
Contraception & Reproductive Health Care, DOI: 10.1080/13625187.2018.1431214

To link to this article: [Link]

Published online: 07 Feb 2018.

Submit your article to this journal

Article views: 11

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

[Link]
THE EUROPEAN JOURNAL OF CONTRACEPTION & REPRODUCTIVE HEALTH CARE, 2018
[Link]

EPIDEMIOLOGICAL STUDY

The risk of cervical atypia in oral contraceptive users

Indira Adhikaria, Tiina Erikssona, Tapio Luostarinenb, Matti Lehtinenb and Dan Apterc
a
Department of Health Sciences, University of Tampere, Tampere, Finland; bKarolinska Institute, Huddinge, Sweden; cVL-Medi, Helsinki,
Finland

ABSTRACT ARTICLE HISTORY

Background: The interactions of oral contraceptive (OC) use, risk of human papillomavirus (HPV) Received 17 November 2017
infection and associated cellular atypia are complex. We investigated the association between his- Revised 10 January 2018
tory of OC-use, and cytological or histopathological abnormalities in a cohort of non-HPV vacci- Accepted 16 January 2018
nated originally 16–17-year-old women participating the PATRICIA trial for 4 years. Published online 8 February
2018
Methods: The total number of hepatitis A-virus (control) vaccine recipients participating in the clin-
ical PATRICIA trial in Finland was 2399. Nine-hundred and ninety-nine women returned question- KEYWORDS
naires on living conditions-life habits and sexual health after completing the study. Mean age at ASCUS; cervical neoplasia;
answering the questionnaire at the end of the clinical trial was 22 years. Age at sexual debut varied contraceptives; HPV;
between 12 and 16 years for majority of the women. Cervical cytological samples were obtained squamous intraepithelial
every 6 months throughout the PATRICIA trial. The relative risk of cervical atypia associated with lesion
time since start of oral contraceptives use was calculated as odds ratio (OR) with 95% confidence
interval (CI) using logistic regression.
Results: Compared to never-users, the smoking and age-at-sexual-debut adjusted relative risk of
cervical intraepithelial neoplasia grade 1 (CIN1) in women who had started the use of oral contra-
ceptives for more than 1 year was low (OR 0.2, 95% CI: 0.1–0.7). Risk of cytological atypia was also
reduced (OR 0.6) albeit not significantly (95% CI: 0.3–1.3).
Conclusions: Use of oral contraceptives does not increase the risk of cervical atypia but when
established might instead be protective.

Introduction We evaluated the association between history of OC-use,

and development of incident cytological or histopatho-
Cervical cancer is the third leading cancer among women
logical abnormalities in a cohort of young women attend-
globally with estimated age-standardised incidence and
ing the control arm of a clinical vaccination trial for 4 years
mortality rates of 14.0 and 6.8 per 100,000 person-years [1].
and answering a health survey questionnaire at the end of
IARC (International Agency for Research on Cancer) has
defined cervical intraepithelial neoplasia (CIN) as the spec- the trial [16–20].
trum of pre-invasive diseases progressing through various
degrees of dysplasia to invasive cervical cancer (ICC) [2].
CIN is classified into three grades: CIN1, CIN2 and CIN3 Materials and methods
equivalent to mild, moderate and severe dysplasia as well Our study material stems from the control arm of a double-
as carcinoma in situ, respectively [3]. The corresponding blinded, multi-national randomised controlled PATRICIA trial
epithelial cellular abnormalities, most notably: ASCUS (atyp-
registration number: (NCT00122681) [16]. The primary aim
ical squamous cells of undetermined significance), LSIL (low
of the trial was to evaluate the HPV16/18-related vaccine
grade squamous intraepithelial lesions), HSIL (high-grade
efficacy (VE) against CIN2þ. Altogether 18,644 women from
squamous intraepithelial lesions), ASC-H (atypical squamous
14 countries were randomly assigned to receive HPV16/18
cells cannot exclude HSIL) and AGC (atypical glandular
vaccine (intervention arm) or at least one dose of hepatitis
cells) are reported according to the Bethesda system [4].
A virus (HAV) vaccine in the control arm. In Finland, 16–17-
The treatment and management of women with cervical
abnormalities currently rely on this classification [4,5]. year-old women were enrolled in a population-based fash-
The major risk factor for cervical neoplasia is persistent ion [16]. All 2409 and 2399 participants received at least
infection with oncogenic human papillomavirus (HPV) one dose of HPV16/18 vaccine or HAV vaccine, respectively,
which drives the progression of cervical lesions through the in the two trial arms. Full descriptions of the trial and final
above-mentioned two (CIN1 and CIN3) or three (CIN1, CIN2 results on the primary endpoints, VE estimated persistent
and CIN3) different grades of intraepithelial neoplasia into HPV infection and various CIN grades, have been reported
ICC [6,7]. Other suggested risk factors for CIN are smoking, earlier [17,18]. In Finland, the trial was approved by the
Chlamydia trachomatis, parity and long-term use of oral Finnish national ethics committee (TUKIJA 1174/04) in 2004.
contraceptives (OC) [8–13]. The risk of CIN associated with Written informed consent was obtained from all the
OC-use has, however, been questioned [14,15]. participants.

CONTACT Indira Adhikari [Link]@[Link], [Link].x@[Link] Department of Health Sciences, University of Tampere,
Tampere, Finland
ß 2018 The European Society of Contraception and Reproductive Health
2 I. ADHIKARI ET AL.

In short, ours is a longitudinal study planned a priori before or after first sexual intercourse. The time of start of
and established when the clinical phase III PATRICIA trial OC-use was quantified in years. The time since the start
was started in Finland [16]. As planned, the questionnaire of OC-use was further categorised into 0–1 year, 2–3
was implemented to both randomised arms (vaccine and years, 4–5 years, 6–7 years, 8 years or more. Those having
control arms) at the end of the clinical trial when the study a year or less from the start of OC-use were considered
subjects were approximately 22 years of age and had a separately from those having more than a year from the
possibility to report start of their OC-use over time. Only start of OC-use. Random breaks in OC-use were not
women in the HAV vaccine control arm who had answered recorded.
the questionnaire after exiting the trial were included in Covariables were smoking (‘never-smokers’, ‘past smok-
the study (Figure 1) [19]. Information about the history of ers’ and ‘present smokers’), alcohol use (‘never-users’, ‘past
OC-use, smoking, and sexual habits were derived from the users’ and ‘present users’), condom use (‘never-users’,
questionnaire. ‘users’ and ‘do not know’), age at sexual debut (‘12–16’ and
Cervical cytological samples were obtained in conjunc- ‘17–22’), number of life-time sexual partners (‘none’, ‘1’,
tion of pelvic examination every 6 months during the ‘2–4’, ‘5–9’ and ‘more than 10’) and age at answering to
4-year trial. The first findings of atypical squamous cells of the questionnaire. Information on Chlamydia trachomatis
undetermined significance (ASCUS), low-grade squamous testing at the study visits was also available for the statis-
intraepithelial lesions (LSIL) and high-grade squamous intra- tical analysis.
epithelial lesions (HSIL) were registered as index incident Incidence rates of cytological and histopathological
cases for the statistical analysis. abnormalities were calculated with 95% confidence interval
The histopathologically confirmed cervical intraepithelial (95% CI) and expressed per 100 person years using the
lesions (CIN) grades 1, 2 and 3 were diagnosed in colpos- concept of survival analysis. Odds ratios (OR) with 95% CI
copy-directed biopsy samples obtained during the trial. The were calculated using logistic regression in Stata version
first CIN1, CIN2 and CIN3 diagnoses were listed as the index 14.0 (Stata Corp LP, Statistical Software: Release 14. College
cases for the statistical analysis in this study. Station, TX).
The use of OC was the main variable of interest. It was
used as a categorical variable with three categories:
Results
‘never-users’, ‘past users’ and ‘present users’. The length of
OC-use was not asked explicitly. It was inferred from the The total number of HAV vaccine recipients, who returned
start of OC-use and age at first sexual intercourse; either the questionnaire after the last study visit was 999. A total

Figure 1. Flow diagram of study subjects selection.

 THE EUROPEAN JOURNAL OF CONTRACEPTION & REPRODUCTIVE HEALTH CARE 3

Table 1. Characteristics of the study material by oral contraceptive (OC)

Incidence rate

(4.2–11.4)

(6.4–14.9)
Never-users

(2.1–7.7)

(0.1–3.0)
use.

(95% CI)
(N ¼ 62)
Short-term Long-term
Never-users OC users OC users

4.0
6.9
0.4
9.8
(N ¼ 62) (N ¼ 11) (N ¼ 810)
Category n % n % n %

15

22
9

1
n
Age
22 30 48.4 5 45.5 371 45.8
23 32 51.6 6 54.5 437 54.0

OC-use >1 year

24 0 0 0 0 2 0.20

Incidence rate

(4.0–5.6)
(4.8–6.5)
(0.0–0.3)
(7.4–9.5)
(N ¼ 810)
Total 62 100 11 100 810 100

(95% CI)
Smoking
Never 44 71.0 11 100 449 55.7

4.7
5.6
0.1
8.4
Past smoker 3 4.80 0 0 90 11.2
Present smoker 15 24.2 0 0 267 33.1
Total 62 100 11 100 806 100
Condom use

4
141
164

252
n
Never-users 25 43.1 3 27.3 369 45.8
User 16 27.6 7 63.6 379 47.1
Do not know 17 29.3 1 9.10 57 7.10

OC-use 8 years
Total 58 100 11 100 805 100

Incidence rate

8.8 (6.8–11.5)
5.6 (4.0–7.8)
5.7 (4.1–8.0)
Table 2. Incidence rate (per 100 women years) of abnormal cytological findings by time since the start of oral contraceptive (OC) use during a clinical follow-up of 4 years.
Age at sexual debut

(N ¼ 167)

(95% CI)
12–16 19 39.6 0 0 577 71.2

NA
17–22 29 60.4 11 100 233 28.8
Total 48 100 11 100 810 100
Number of sexual partners
0 0 0 0 0 1 0.10
1 16 32.7 7 63.6 107 13.2

34
35

54
0
2–4 15 30.6 3 27.3 261 32.2

n
5–9 7 14.3 1 9.10 225 27.8
>10 11 22.4 0 0 216 26.7

OC-use 6–7 years

Total 49 100 11 100 810 100

Incidence rate

(7.2–10.2)
(3.6–5.8)
(4.3–6.7)
(0.1–0.7)
(N ¼ 399)

(95% CI)
4.6
5.4
0.3
8.5
of 914, who met the inclusion criteria, were eligible for this
study (Figure 1). Mean age at answering the questionnaire
was 22.5 years. The mean age at sexual debut was 15.9
67

4
78

125
(12–22) years. Most of the women reported of having had
n

2 to 4 life-time sexual partners and reported having never

smoked (Table 1). The total numbers of OC users and
OC-use 4–5 years

Incidence rate

7.9 (6.1–10.2)
never-users were 843 and 62, respectively. Some data on
4.0 (2.8–5.8)
5.7 (4.2–7.8)
(N ¼ 192)

(95% CI)

the time since start of OC-use was missing which decreased

NA

the number of OC-users to 821 while categorising them

further by time of OC-use.
The total number of ASCUS, LSIL and HSIL findings were
151, 182 and 5, respectively. The incidence rates of LSIL
29

0
40

57
n

were somewhat higher among never users 6.9 (95% CI

4.2–11.4) or the women who had just started the use of OC
OC-use 2–3 years

(0 to 1 year ago), 7.7 (95% CI 2.5–23.9) as compared to

Incidence rate

5.6 (3.1–10.1)
5.8 (3.2–10.5)

8.2 (5.0–13.3)
(95% CI)
(N ¼ 52)

those who had started more than a year ago 5.6 (95% CI:
NA

4.8–6.5) (Table 2). This was not the case for ASCUS findings
(Table 2).
The total numbers of CIN1, CIN2 and CIN3 cases were
40, 20 and 8, respectively. The incidence rates of CIN1 per
11

0
11

16
n

100 person-years were higher among never users 2.6 (95%

Any abnormality means either ASCUS or LSIL or HSIL.

CI: 1.1–5.7) or women who had just started the use of OC

OC-use 0–1 years

(2.3, 95% CI: 0.3–16.3) compared to women who had

Incidence rate

2.4 (0.3–17.1)
7.7 (2.5–23.9)

9.6 (3.6–25.6)

started to use OC for more than a year (1.0, 95% CI:

(95% CI)
(N ¼ 11)

NA

0.7–1.5) (Table 3).

There was no significantly increased risk of ASCUS and
LSIL associated with OC use (Table 4). Compared to
the never-users, there appeared to be a slightly increased
relative risk of ASCUS associated with more than a year of
n
1
3
0
4

OC-use, but it vanished after adjusting for smoking and

age-at-sexual-debut (OR 1.0, 95% CI: 0.4–2.3) (Table 4).
Any abnormalitya

Most notably the adjusted relative risk of CIN1 in

women who had started the use of OC for more than a
Category
ASCUS

year ago was significantly reduced (OR 0.2, 95% CI: 0.1–0.7)
HSIL
LSIL

as compared to never users (Table 4).

a
4 I. ADHIKARI ET AL.

Discussion

Incidence rate

2.6 (1.1–5.7)
0.8 (0.2–3.4)

3.0 (1.4–6.3)
Never-users

(95% CI)
(N ¼ 62)

NA
Findings and interpretations
We found decreased risk of CIN1 and cytological atypia
among women who had started OC use 2–8 (or more)
years ago. The established use of OC may have a protective

n
6
2
0
7
OC-use >1 year effect against CIN1.

Incidence rate

(0.7–1.5)
(0.4–0.9)
(0.1–0.5)
(1.1–2.0)
(N ¼ 810)

(95% CI) Strengths and weaknesses of the study

The strength of our study is the large cohort data
1.0
0.6
0.2
1.5
(PATRICIA) with meticulous clinical follow-up over four
years, a period during which they received sexual health
counselling and different free modes of contraception.
8
33
18

47
n

Cervical cytological sampling was done every 6 months

Table 3. Incidence rate (per 100 women years) of abnormal histopathological findings by time since the start of oral contraceptive (OC) use during a clinical follow-up of 4 years.

irrespective of OC-use. The study was accomplished among

OC- use 8 years

women less than 25 years of age by what time OC-use had

Incidence rate

1.5 (0.8–2.8)
0.5 (0.1–1.4)

1.7 (0.9–3.0)
(N ¼ 167)

(95% CI)

been well established for up to 10 years. Thus, the results

NA

may be generalisable. The limitation of our study is that

the questionnaires were filled at the end of the clinical trial,
and random breaks in OC-use were not registered. The
high numbers of condom users among both the short-term
and long-term OC users might have reduced the possibil-
3
0
10

11
n

ities to disclose the effect of OC use in our study. The

observed statistically significant decreased risk of CIN1 asso-
OC-use 6–7 years

ciated with established OC use, however, indicates that this

Incidence rate

(0.6–1.6)
(0.2–0.9)
(0.1–0.8)
(0.9–2.1)
(N ¼ 399)

(95% CI)

was not the case. Lack of statistically significant differences

in Condom use between or within the different groups of
1.0
0.4
0.3
1.3

OC users is reassuring.

Difference and similarities in relation to other studies

7
5
15

21
n

Our findings on the protective association are in line with

previous studies which have found a negative association
OC-use 4–5 years

Incidence rate

between OC-use and cervical intraepithelial neoplasia grade

(0.4–1.8)
(0.3–1.6)
(0.1–1.1)
(0.9–2.8)
(N ¼ 192)

(95% CI)

1 [21,22]. At the same time, a positive association between

OC-use and the risk of more severe CIN grades has also
0.8
0.7
0.3
1.6

been reported [23–26]. Our findings lost statistical signifi-

cance when CIN2 and CIN3 lesions were also considered.
On the other hand, we found no significant risk between
6
5
2
12
n

OC-use and abnormal cytological findings. This is in line

with most of the earlier studies [27–29] and suggests that
misclassification of cytological diagnoses may be an issue.
OC-use 2–3 years

Incidence rate

(0.2–3.9)
(0.5–4.6)
(0.1–3.4)
(0.5–4.5)

Kruger-Kjaer et al. also found that long-term OC-use is asso-

(95% CI)
(N ¼ 52)

ciated with the development of HSIL, which corresponds to

CIN2 and CIN3, but not with the LSIL or ASCUS [30].
1.0
1.5
0.5
1.4

The absence of significant positive or negative associa-

tions between the start of OC-use and abnormal cytological
findings is probably due to low numbers. In addition, the
n
2
3
1
3

low number of more severe findings could be attributable

to the age of women during the follow-up. The trial partici-
OC-use 0–1 years

pants were 16 to 22 years old, which, is relatively short

Incidence rate

2.3 (0.3–16.3)

2.3 (0.3–16.3)
CIN1 þ includes CIN1, CIN2 and CIN3.
(95% CI)
(N ¼ 11)

time period to develop high-grade cervical lesions [31].

NA
NA

On the other hand, the decreased relative risk of CIN1

associated with established OC-use (more than a year since
the start) against CIN1 is noteworthy. CIN1 is the clinical
manifestation of HPV infection which develops relatively
rapidly. HPV infections cluster close to start of OC-use or
n
1
0
0
1

with early sexual debut [32–34]. Our finding suggests that

established OC-use, a probable surrogate of established
Category

1þa

sexual relationships may have a protective effect against

1
2
3
CIN
CIN
CIN
CIN

cervical neoplasia in young adult Finnish women.

a
 THE EUROPEAN JOURNAL OF CONTRACEPTION & REPRODUCTIVE HEALTH CARE 5

Table 4. Adjusted relative risk (odds ratio, OR with 95% confidence interval, 95% CI) of abnormal cytological and histopathological findings associated with
time since the start of oral contraceptive (OC) use.
Never users
OC-use 0–1 years (N ¼ 11) OC-use >1 year (N ¼ 810) (N¼ 62)
Category n Crude OR Adj. OR (95% CI) n Crude OR Adj. OR (95% CI) n OR
ASCUS 1 0.5 0.5 (0.1–4.6) 141 1.2 1.0 (0.4–2.3) 9 1.0
LSIL 3 1.1 1.0 (0.2–4.6) 164 0.7 0.6 (0.3–1.3) 15 1.0
CIN1 1 0.9 0.7 (0.1–7.3) 33 0.3 0.2 (0.1–0.7) 6 1.0
CIN1þ 1 0.7 0.7 (0.1–6.9) 47 0.4 0.4 (0.1–1.1) 7 1.0

Open questions and future research [11] Luostarinen T, Namujju PB, Merikukka M, et al. Order of HPV/
Chlamydia infections and cervical high-grade precancer risk: a
In conclusion, established oral contraceptive use seems to case-cohort study. Int J Cancer. 2013;133:1756–1759.
be associated with a protective effect against mild cervical [12] Muwonge R, Ngo Mbus L, Ngoma T, et al. Socio-demographic
and reproductive determinants of cervical neoplasia in seven
abnormality and CIN1, that is, clinical manifestations of HPV
sub-Sahara African countries. Cancer Causes Control.
infection. Larger prospective studies aiming to determine 2016;27:1437–1446.
the association between cervical lesions and oral contracep- [13] Castellsague X, Munoz N. Chapter 3: cofactors in human
tives are required. papillomavirus carcinogenesis–role of parity, oral contracep-
tives, and tobacco smoking. J Natl Cancer Inst Monogr.
2003;20–28.
[14] Ajah LO, Chigbu CO, Ozumba BC, et al. Is there any association
between hormonal contraceptives and cervical neoplasia in a
Disclosure statement poor Nigerian setting? Onco Targets Ther. 2015;8:1887–1892.
ML and DA have received grants for HPV vaccination studies from [15] Syrj€anen K, Shabalova I, Petrovichev N, et al. Oral contracep-
Merck & Co. Inc. and GSK vaccines through their employers, University tives are not an independent risk factor for cervical intraepithe-
of Tampere (ML) and Family Federation Finland (DA). lial neoplasia or high-risk human papillomavirus infections.
Anticancer Res. 2006;26:4729–4740.
[16] Lehtinen M, Apter D, Dubin G, et al. Enrolment of 22,000 ado-
Funding lescent women to cancer registry follow-up for long-term
human papillomavirus vaccine efficacy: guarding against
The primary PATRICIA study was funded by GSK Biologicals. The cur- guessing. Int J STD Aids. 2006;17:517–521.
rent manuscript reports on an analysis performed on the data from [17] Paavonen J, Jenkins D, Bosch XF, et al. Efficacy of a prophy-
the Finnish study sites under the accountability of the manuscript lactic adjuvanted L1 VLP vaccine against infection with
authors, who received additional funding from the Finnish Cancer HPV16/18 in young women: an interim analysis of a phase III
Organization and Academy of Finland. double-blind, randomized controlled trial. Lancet. 2007;369:
2161–2170.
[18] Lehtinen M, Paavonen J, Wheeler C, et al. Overall efficacy of
References HPV-16/18 vaccine against the most stringent cervical pre-can-
cer end-points: end-of study report of a double blind, random-
[1] International Agency for Research on Cancer (IARC). ized trial. Lancet Oncol. 2012;13:89–99.
GLOBOCAN. Estimated Cancer Incidence, Mortality and [19] Eriksson T, Torvinen S, Woodhall SC, et al. Impact of HPV16/18
Prevalence worldwide [Internet]. 2012; [cited 2017 April 12]. vaccination on quality of life: a pilot study. Eur J Contracept
Available from: [Link] Reprod Health Care. 2013;18:364–371.
aspx [20] Woodhall SC, Eriksson T, Nykanen AM, et al. Impact of HPV
[2] Sellors JW, Sankaranarayanan R, Colposcopy and treatment of vaccination on young women's quality of life - a five year fol-
cervical intraepithelial neoplasia: a beginner’s manual [Internet]. low-up study. Eur J Contracept Reprod Health Care.
Lyon, France: IARC; 2017. Available from: [Link] 2011;16:3–8.
[Link]?lang¼1&chap¼2 [21] Chih HJ, Lee AH, Colville L, et al. Condom and oral contracep-
[3] Buckley CH, Butler BE, Fox H. Cervical intraepithelial neoplasia. J tive use and risk of cervical intraepithelial neoplasia in
Clin Pathol. 1982;35:1–13. Australian women. J Gynecol Oncol. 2014;25:183–187.
[4] Bergeron C. The 2001 Bethesda system. Salud P
ublica M
ex. [22] Longatto-Filho A, Hammes LS, Sarian LO, et al. Hormonal con-
traceptives and the length of their use are not independent
2003;45:(3):340–344.
risk factors for high-risk HPV infections or high-grade CIN.
[5] Wright TC, Jr., Massad LS, Dunton CJ, et al. 2006 Consensus
Gynecol Obstet Invest. 2011;71:93–103.
Guidelines for the management of women with abnormal cer-
[23] Smith JS, Green J, Berrington de Gonzalez A, et al. Cervical can-
vical screening tests. J Low Genit Tract Dis. 2007;11:201–222.
cer and use of hormonal contraceptives: a systematic review.
[6] Walboomers JM, Jacobs MV, Manos MM, et al. Human papillo-
Lancet. 2003;361:1159–1167.
mavirus is a necessary cause of invasive cervical cancer world-
[24] Moreno V, Bosch FX, Munoz N, et al. Effect of oral contracep-
wide. J Pathol. 1999;189:12–19.
tives on risk of cervical cancer in women with human papillo-
[7] Zur Hausen H. Papillomaviruses in the causation of human can-
mavirus infection: the IARC multicentric case-control study.
cers – a brief historical account . Virology. 2009;384:260–265. Lancet. 2002;359:1085–1092.
[8] Kjellberg L, Hallmans G, Ahren AM, et al. Smoking, diet, preg- [25] Appleby P, Beral V, Berrington de Gonzalez A, et al. Cervical
nancy and oral contraceptive use as risk factors for cervical cancer and hormonal contraceptives: collaborative reanalysis of
intra-epithelial neoplasia in relation to human papillomavirus individual data for 16,573 women with cervical cancer and
infection. Br J Cancer. 2000; 82:1332–1338. 35,509 women without cervical cancer from 24 epidemiological
[9] Simen-Kapeu A, La Ruche G, Kataja V, et al. Tobacco smoking studies. Lancet. 2007;370:1609–1621.
and chewing as risk factors for multiple human papillomavirus [26] Zondervan KT, Carpenter LM, Painter R, et al. Oral contracep-
infections and cervical squamous intraepithelial lesions in two tives and cervical cancer–further findings from the Oxford fam-
countries (Cote d’Ivore and Finland) with different tobacco ily planning association contraceptive study. Br J Cancer.
exposure. Cancer Causes Control. 2009;20:163–170. 1996;73:1291–1297.
[10] Lehtinen M, Ault KA, Lyytikainen E, et al. Chlamydia trachomatis [27] Binesh F, Akhavan A, Pirdehghan A, et al. Does oral contracep-
infection and risk of cervical intraepithelial neoplasia. Sex tive pill increase the risk of abnormal Pap smear? Iran J Reprod
Transm Infect. 2011;87:372–376. Med. 2013;11:761–766.
6 I. ADHIKARI ET AL.

[28] Kazerooni T, Mosalaee A. Does contraceptive method change [32] Nygard JF, Sauer T, Skjeldestad FE, et al. CIN 2/3 and cervical
the Pap smear finding? Contraception. 2002;66:243–246. cancer after an ASCUS pap smear. A 7-year, prospective study
[29] Giuliano AR, Papenfuss M, De Galaz EM, et al. Risk factors for of the Norwegian population-based, coordinated screening pro-
squamous intraepithelial lesions (SIL) of the cervix among gram. Acta Cytol. 2003;47:991–1000.
women residing at the US-Mexico border. Int J Cancer. [33] Nygard JF, Sauer T, Nygard M, et al. CIN 2/3 and cervical cancer
2004;109:112–118. in an organised screening programme after an unsatisfactory or
[30] kruger-Kjaer S, van den B, Adriaan JC, et al. Different risk factor a normal Pap smear: a seven-year prospective study of the
patterns for high-grade and low-grade intraepithelial lesions on Norwegian population-based screening programme. J Med
the cervix among HPV-positive and HPV-negative young Screen. 2004;11:70–76.
women. Int J Cancer. 1998;76:613–619. [34] Sun LL, Chen W, Fan YY, et al. The presence of advanced
[31] Watson RA. Human papillomavirus: confronting the lesions and associating risk factors for advanced cervical carcin-
epidemic-a Urologist's Perspective. Rev Urol. 2005;7: oma in patients with atypical sguamous cells of undetermined
135–144. significance. Eur J Gynaecol Oncol. 2015;36:585–589.