cme article journal watch paediatrics...

ISSN 1015-4361(PHILIPPINES)

PhilippinesJAN/FEB 2012 Vol. 38 No. 1

JOURNAL OF PAEDIATRICS, OBSTETRICS & GYNAECOLOGY

Your partner in paediatric and O&G practice

JUL/AUG 2013 Vol. 39 No. 4 Your partner in paediatric and O&G practice

www.jpog.com

CME ARTICLE

Hormonal Contraception and Cancers

JOURNAL WATCH

PAEDIATRICS

Management of Hearing Loss in Children

Constipation in Infants and Children

GYNAECOLOGY

Ovarian Cancer: Current Management and Future Directions

OBSTETRICS

Management of Early Pregnancy

Complications

JPOG_JulAug13_CVR_FINAL.indd 1 8/5/13 2:00 PM

JUL/AUG 2013

Vol. 39 No. 4


Journal Watch

133 • Calcium intake and mortality in Swedish women

• Outcome of extreme preterm birth in England, 1995–2006

134 • New anti-interleukin therapies for systemic JIA

• CPAP vs surfactant and higher vs lower oxygen saturation for extremely preterm infants: Outcomes at 18–22 months

135 • Paromomycin for cutaneous leishmaniasis

• HMPV infection in young children in the US

136 • Risk factors for stillbirth • Fetal macrosomia in developing countries

Board Director, Paediatrics

Professor Pik-To CheungAssociate ProfessorDepartment of Paediatrics and Adolescent MedicineThe University of Hong Kong

Board Director, Obstetrics and Gynaecology

Professor Pak-Chung HoHead, Department of Obstetrics and GynaecologyThe University of Hong Kong

Editorial Board Professor Biran AffandiUniversity of Indonesia

Dr Karen Kar-Loen ChanThe University of Hong Kong

Professor Oh Moh ChayKK Women’s and Children’s Hospital, Singapore

Associate Professor Anette JacobsenKK Women’s and Children’s Hospital, Singapore

Professor Rahman JamalUniversiti Kebangsaan Malaysia

Dato’ Dr Ravindran JegasothyHospital Kuala Lumpur, Malaysia

Associate Professor Kenneth KwekKK Women’s and Children’s Hospital, Singapore

Dr Siu-Keung LamPrestige Medical Centre, Hong Kong

Professor Terence LaoChinese University of Hong Kong

Dr Kwok-Yin LeungThe University of Hong Kong

Dr Tak-Yeung LeungChinese University of Hong Kong

Professor Tzou-Yien LinChang Gung University, Taiwan

Professor Somsak LolekhaRamathibodi Hospital, Thailand

Professor Lucy Chai-See LumUniversity of Malaya, Malaysia

Professor SC NgNational University of Singapore

Professor Hextan Yuen-Sheung NganThe University of Hong Kong

Professor Carmencita D PadillaUniversity of the Philippines Manila

Professor Seng-Hock QuakNational University of Singapore

Dr Tatang Kustiman SamsiUniversity of Tarumanagara, Indonesia

Professor Alex SiaKK Women’s and Children’s Hospital, Singapore

Dr Raman SubramaniamFetal Medicine and Gynaecology Centre, Malaysia

Professor Walfrido W Sumpaico MCU-FDT Medical Foundation, Philippines

Professor Cheng Lim TanKK Women’s and Children’s Hospital, Singapore

Professor Kok Hian TanKK Women’s and Children’s Hospital, Singapore

Professor Surasak TaneepanichskulChulalongkorn University, Thailand

Professor Eng-Hseon TayThomson Women Cancer Centre, Singapore

Professor PC WongNational University of Singapore

Adjunct Professor George SH YeoKK Women’s and Children’s Hospital, Singapore

Professor Hui-Kim YapNational University of Singapore

Professor Tsu-Fuh YehChina Medical University, Taiwan

133

136

JPOG JUL/AUG 2013 • i

JPOG_JulAug_2013_TOC_Final.indd 1 8/5/13 1:52 PM

JUL/AUG 2013

Vol. 39 No. 4


Enquiries and Correspondence

137

JPOG JUL/AUG 2013 • ii

Publisher Ben YeoPublication Manager Marisa LamManaging EditorGreg TownAssociate Editor Grace LingDesignersAgnes Chieng, Sam ShumProduction Edwin Yu, Ho Wai Hung, Steven Cheung Circulation Christine ChokAccounting Manager Minty KwanAdvertising CoordinatorRachael Tan

Published by: MIMS Asia Pacific27th Floor, OTB Building160 Gloucester Road, Wan Chai, Hong KongTel: (852) 2559 5888 Email: [email protected]

PUBLISHER: Journal of Paediatrics, Obstetric & Gynaecology (JPOG) is published 6 times a year by MIMS Pte Ltd. CIRCULATION: JPOG is a controlled circulation for medical practitioners in South East Asia. It is also available on subscription to members of allied professions. SUBSCRIPTION: The price per annum is US$42 (surface mail, students US$21) and US$48 (overseas airmail, students US$24); back issues US$8 per copy. EDITORIAL MATTER published herein has been prepared by professional editorial staff. Views expressed are not necessarily those of MIMS Pte Ltd. Although great care has been taken in compiling and checking the information given in this publication to ensure that it is accurate, the authors, the publisher and their servants or agents shall not be responsible or in any way liable for the continued currency of the information or for any errors, omissions or inaccuracies in this publication whether arising from negligence or otherwise howsoever, or for any consequences arising therefrom. The inclusion or exclusion of any product does not mean that the publisher advocates or rejects its use either generally or in any particular field or fields. COPYRIGHT: © 2013 MIMS Pte Ltd. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, in any language, without written consent of copyright owner. Permission to reprint must be obtained from the publisher. ADVERTISEMENTS are subject to editorial acceptance and have no influence on editorial content or presentation. MIMS Pte Ltd does not guarantee, directly or indirectly, the quality or efficacy of any product or service described in the advertisements or other material which is commercial in nature. Philippine edition: Entered as second-class mail at the Makati Central Post Office under Permit No. PS-326-01 NCR, dated 9 Feb 2001. Printed by Fortune Printing International Ltd, 3rd Floor, Chung On Industrial Building, 28 Lee Chung Street, Chai Wan, Hong Kong.

ChinaYang XuanTel: (86 21) 6157 3888Email: [email protected]

Hong KongKristina Lo-Kurtz, Jacqueline Cheung, Marisa Lam, Miranda WongTel: (852) 2559 5888Email: [email protected]

IndiaMonica BhatiaTel: (91 80) 2349 4644 Email: [email protected]

KoreaChoe Eun YoungTel: (82 2) 3019 9350Email: [email protected]

IndonesiaHafta Hasibuan, Sri Damayanti, Ritta PamolangoTel: (62 21) 729 2662Email: [email protected]

MalaysiaMeera Jassal, Lee Pek Lian, Sheena Subash, Grace YeohTel: (60 3) 7954 2910Email: [email protected]

PhilippinesMarian Chua, Kims Pagsuyuin, Rowena Belgica, Philip KatipunanTel: (63 2) 886 0333Email: [email protected]

SingaporeJason Bernstein, Carrie Ong, Josephine Cheong, Melanie NyamTel: (65) 6290 7400Email: [email protected]

ThailandWipa SriwijitchokTel: (66 2) 741 5354Email: [email protected]

VietnamNguyen Thi Lan Huong, Nguyen Thi My DungTel: (84 8) 3829 7923Email: [email protected]

Europe/USAKristina Lo-KurtzTel: (852) 2116 4352Email: [email protected]

Review ArticlePaediatrics

137 The Management of Hearing Loss in Children

Universal neonatal hearing screening aims to detect the 1 in 1,000 babies born in the UK with a permanent hearing loss detectable at birth. However, children may present later to the paediatrician with hearing difficulties. This article aims to discuss the clinical assessment of hearing and provides an overview of the management options available in the treatment of hearing loss.

Marianne D Elloy, Andrew H Marshall

Review ArticleObstetrics

146 Management of Early Pregnancy Complications

Complications of early pregnancy are common, including pregnancy loss, threatened miscarriage, ectopic pregnancy, molar pregnancy and hyperemesis. This review discusses the different presentations, diagnoses and management of the common problems complicating early pregnancy.

Harriet Pugsley, Judith Moore

146



JPOG JUL/AUG 2013 • iii

Review ArticlesComprehensive reviews providing the latest clinical information on all aspects of the management of medical conditions affecting children and women.

Case StudiesInteresting cases seen in general practice and their management.

Pictorial MedicineVignettes of illustrated cases with clinical photographs.

For more information, please refer to the Instructions for Authors on our website www.jpog.com, or contact:The EditorMIMS Pte Ltd, 6 Shenton Way, #15-08 Tower 2, Singapore 068809Tel: (65) 6290 7400 Fax: (65) 6290 7401 E-mail: [email protected]

Lisa Low, Illustrator

JUL/AUG 2013

Vol. 39 No. 4

The Cover:Early Pregnancy Complications

© 2013 MIMS Pte Ltd

Review ArticleGynaecology

155 Ovarian Cancer: Current Management and Future Directions

Ovarian cancer has the highest mortality of all the gynaecological malignancies. Treatment of advanced epithelial ovarian cancer usually involves debulking surgery and chemotherapy. Treatment may prolong life and palliate symptoms but it is rarely curative. New treatments are constantly being developed and offer the hope of improved outcomes.

Siân E Taylor, John M Kirwan

Review ArticlePaediatrics

164 Constipation in Infants and Children

Constipation is a common problem in children and is usually functional, related to stool-withholding. Successful management requires parent education, behavioural strategies, laxative agents (often long term) and ongoing review.

Taya Dowling, Scott Nightingale

Continuing Medical Education

169 Hormonal Contraception and Cancers

This article reviews the evidence regarding the relationship between hormonal contraceptive use and the development of cancer, with the discussion focusing mainly on carcinoma of the breast and female genital tract.

Wong Yuen Kwan Alice

155

169

Philippines



JPOG JUL/AUG 2013 • 169

Hormonal Contraception and Cancers Wong Yuen Kwan Alice, MBBS, FRCOG, FHKAM(O&G), FHKCOG, Cert HKCOG(Reprod Med)

Oral contraceptive use puts women at risk for certain carcinomas.

INTRODUCTION

The use of hormones has provided great

convenience to a woman’s life, be it for

therapeutic use or as a lifestyle drug, ie,

contraceptives. However, the duration of

hormone use is frequently long, in terms

of years. Concerns have been raised about

the possibility of a relationship between

cancer development and long-term hormo-

nal influence. This article reviews the evi-

dence regarding the relationship between

hormonal contraceptive use and the devel-

opment of cancer, with the discussion fo-

cusing mainly on carcinoma of the breast

and female genital tract.

HISTORICAL EVIDENCE

The first report dated back to 1972 when

combined oral contraceptive pills (COC)

containing mestranol and norethynodrel

appeared to cause a case of metastatic

breast cancer in a female rhesus monkey.1

Soon after, there were further similar re-

ports of development of breast cancer in

beagles and rodents after exposure to hor-

mones contained in today’s COC.2–4

In June 2005, the International Agen-

cy for Research on Cancer (IARC) Working

Group of the World Health Organization

(WHO) met in Lyon, France, and classified

combined oral contraceptives and com-

bined oestrogen-progestogen hormone

therapy as ‘carcinogenic’ to humans.

THEORY OF ‘CARCINOGENESIS’

Carcinogenesis involves two steps, name-

ly, initiation and promotion. Most of the

studies on the relationship between hor-

mones and cancer development involved

the latter step. In 1989, Anderson et al

reported that nulliparous women who

took COC had a significantly higher rate of

breast cell division.5 It was also found that

COC caused a rise in epithelial cell pro-

liferation of the glandular breast,6 leading

to an increase in accumulation of random

genetic errors.7

However, whether these proliferat-

ing effects on normal epithelia, as a result

of replication error, may cause malignant

transformation has not yet been proven,

although DNA repair is hampered by ac-

tivated proliferation.8 With the end point

being ‘chromosomal mutation’, numer-

ous chromosomal aberrations have been

JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 169 8/5/13 2:13 PM


observed with natural and synthetic oes-

trogens and progestogens. There was no

proof of sufficient strength to show that

these proliferative effects could induce

tumours. Hormonal tumour promotion also

cannot be discriminated from a causal re-

lation with breast tumour induction.

Moreover, toxicity studies used

animal models. Species-specific effects

might not allow for extrapolation of the ef-

fects to humans. Supraphysiological doses

had been used in animal studies and this

might not reflect actual clinical use. Pos-

sibility of genetic predisposition and other

environmental factors were not taken into

account.

CARCINOMA OF THE BREAST

In 1981, Pike et al reported that women

who took COC for 4 years or more prior to

their first full-term pregnancy experienced

a 125% increased risk of developing carci-

noma of the breast and a 250% increase

in risk with 8 years or more of COC use.9

Similarly, in 1989, Chilvers et al reported

that women under the age of 36 who used

COC for at least 4 years before their first

full-term pregnancy had at least 44% in-

creased risk of breast cancer.10

However, the Cancer and Steroid Hor-

mone Study (CASH), which was one of the

largest case-control studies in the 1980s,

Combined oral contraceptives play a role in the promotion of carcinogenesis.

found no association between breast can-

cer and COC use for women up to the age

of 54. Risk was found only among a sub-

group of women who underwent menarche

before age 13 and used COC for more than

10 years before their first birth.11

In the 1990s, the Collaborative Group

on Hormonal Factors in Breast Cancer in

Oxford, UK, analysed individual data of

53,297 women with breast cancer and

100,239 women without breast cancer

from 54 studies conducted in 25 countries.

The results provided two strong conclu-

sions. First, while women are taking COC

and in a period of 10 years after stopping,

there is a small increase in the relative

risk (RR) of having breast cancer: RR of

1.24 (95% CI, 1.15–1.33) for current users;

RR of 1.16 (95% CI, 1.08–1.23) 1–4 years

after stopping; and RR of 1.07 (95% CI,

1.02–1.13) 5–9 years after stopping. Sec-

ond, there is no significant excess risk of

having breast cancer diagnosed 10 or more

years after stopping use (RR, 1.01 [95%

CI, 0.96–1.05]). The cancers diagnosed in

women who had used COC were less ad-

vanced clinically than those diagnosed in

never-users; the RR for tumours that had

spread beyond the breast compared with

localized tumours was 0.88 (95% CI, 0.81–

0.95). There was no pronounced variation

in the results for recency of use between

women with different background risks of

breast cancer, including women from dif-

ferent countries and ethnic groups, women

with different reproductive histories, and

those with or without a family history of

breast cancer. Other features of hormonal

use, such as duration of use, age at first

use, and the dose and type of hormone




within the contraceptives, had little ad-

ditional effect on breast cancer risk, once

recency of use had been taken into ac-

count.12

Other important studies worth men-

tioning include the Nurses’ Health Study

(1997),13 Women’s Lifestyle and Health

Cohort Study (2002),14 Oxford-Family Plan-

ning Association study (1981),15 Mayo

Clinic Meta-analysis (2006),16 and Royal

College of General Practitioners study

(2007).17 There were some inconsistencies

among their findings, and the increase in

RR shown by some of these studies was

modest. Possible confounders that may

be related to the use of high-dose COC

include the clinical practice during the

time of these studies and the possibil-

ity of the nature of recall bias. The study

by the Collaborative Group on Hormonal

Factors in Breast Cancer12 also shared a

similar problem. Another weakness of this

study was that it analysed pooled data

from studies which examined women with

breast cancer from as far back as the early

1970s. Taking data from studies which in-

terviewed women before the 1980s might

underestimate the risk of breast cancer

development because the latent period

for cancer development was too short and

few women had used COC for significant

periods of time prior to their first full-term

pregnancy in the late 1960s and early

1970s as compared with women of the

late 1970s and 1980s.

In 2005, the IARC classified COC and

combined oestrogen-progestogen hor-

mone therapy as carcinogenic to humans.

The Working Group mentioned a ‘slightly

increased risk of breast cancer in current

and recent users of hormonal contracep-

tives’. This risk disappears 10 years after

cessation of COC use and will be similar to

that in never-users.18 The Working Group

also acknowledged that their statement

does not meet the overall net public health

outcome, be this of a beneficial or ad-

verse effect other than cancer, and there

is no reason to change the current clini-

cal practice, particularly when the risks of

unwanted pregnancy are taken into con-

sideration.18

For BRCA mutation carriers, who al-

ready have a 50–80% increase in risk of

breast cancer, the use of COC will be of

concern. Among BRCA1 mutation carriers,

those who first used COC before 1975,

who used them before age 30, or who

used for 5 years or more might have an

increased risk of breast cancer. COC do not

appear to be associated with risk of breast

cancer in BRCA2 carriers; however, data to

support this are limited.19

On the use of depot medroxyproges-

terone acetate (MPA), pooled analysis of

two major case-control studies (one in

New Zealand20 and the other under the

auspices of the WHO21) found no increase

in risk for breast cancer. A currently unex-

plained pattern of increased risk in recent

users mimics that seen with COC.22

In a study involving completed ques-

Hormonal contraceptive use increases the relative risk of having breast cancer.



tionnaires from 17,360 levonorgestrel-

releasing intrauterine system (LNG-IUS)

users, there was no apparent associa-

tion between the length of time elapsed

from the LNG-IUS insertion up to 10 years

and yearly incidence of breast cancer in

the Finnish female population (data from

the Finnish Cancer Registry). A causal

relationship between LNG-IUS use and

occurrence of breast cancer was not sup-

ported.23

CARCINOMA OF THE UTERINE CORPUS

A meta-analysis of 10 case-control stud-

ies (published up to 1996; 1,728 cases and

6,243 controls) and another cohort study

(440,000 woman-years of observation)

both showed statistically reduced RR for

carcinoma of endometrium in COC users.

This RR was negatively associated with

the duration of COC use; the risk reduction

was 56% with 4 years’ use, 67% with 8

years’ use, and 72% with 12 years’ use.24

The Oxford-Family Planning Asso-

ciation (Oxford-FPA) contraceptive study,

which took place in 1968–2004, involved

540,000 woman-years of observation.

There were 50 women with carcinoma of

the uterine corpus in the control group and

27 women in the COC user group. The RR

for ever-users versus never-users was 0.3

(95% CI, 0.2–0.6). The risk was further

found to be negatively related to the du-

ration of COC use, ie, a RR of 0.6 (95%

CI, 0.3–1.1) for up to 48 months’ use, 0.4

(95% CI, 0.2–0.5) for 49–96 months’ use,

and 0.1 (95% CI, 0.0–0.4) for > 97 months’

use.25 The Royal College of General Prac-

titioners oral contraceptive study also

showed similar findings.17

Another meta-analysis of 11 epide-

miological studies found that the more

recent the use of COC, the lower the risk

for carcinoma of the uterine corpus. Such

protective effect from the previous use of

COC would attenuate with time after dis-

continuation. The RR was 0.33, 0.41 and

0.51 for 5, 10, and 20 years of ceasing, re-

spectively. Even after more than 20 years

of cessation of use, the protective effect

is still significant, ie, at 50% less than

non-users.24

Prolonged and unremitting mitotic

activity of the endometrium due to unop-

posed oestrogenic stimulation has been

proposed to be the cause of development

of the majority of cases of endometrial ad-

enocarcinoma. COC suppress endometrial

mitotic activity, leading to apoptosis, thus

reducing the risk of endometrial cancer.

The use of depot MPA is associated

with an 80% risk reduction of endometrial

adenocarcinoma, a level of protection even

greater than that observed with COC. The

effect was also found to be long-term.22

CARCINOMA OF THE OVARY

The Collaborative Group on Epidemiologi-

cal Studies of Ovarian Cancer published a

collaborative reanalysis of data from 45

epidemiological cohort and case-control

The protective effect of combined oral contraceptives against ovarian carcinoma is long-lasting.




studies, which included 23,257 women

with ovarian cancer and 87,303 controls

from 21 countries. It was found that 7,308

(31%) of the women with ovarian cancer

and 32,717 (37%) of the controls had used

COC, and the average duration of use was

4.4 and 5.0 years, respectively. The over-

all RR for ever-users versus never-users

was 0.73 (95% CI, 0.70–0.76). The longer

the duration of use, the lower the risk for

ovarian cancer development. The overall

RR decreased by 20% for each 5 years of

use. For women who had used COC for 15

years, the risk was almost halved. The pro-

tective effect started after at least 1 year

of COC use. The RR for < 1 year, 1–4 years,

5–9 years, 10–14 years, and ≥ 15 years of

use were 1.0, 0.78, 0.64, 0.56, and 0.42,

respectively.26

This reanalysis also found that the

more recent the use of COC, the lower the

RR of ovarian cancer. The proportional de-

cline in RR per 5 years of COC use was

29% for < 10 years of cessation of use,

19% for 10–19 years, and 15% for 20–29

years. The longer the duration of use, the

higher the protective effect irrespective

of the time elapsed from ceasing. The

start age of COC use and age of last use

seemed to have no effect on the protec-

tion. Low-dose pill use was found to have

an identical RR compared with high-dose

pill use.26

The Oxford-FPA contraceptive study,

which included 17,032 women aged 25–39

recruited at 17 family planning clinics in

England and Scotland between 1968–1974

with a long follow-up till 2004, analyzed

a total of 540,000 woman-years of ob-

servation and 58 ovarian cancer cases in

the control group and 48 cases in the COC

group.25 The overall ovarian cancer RR for

ever-users versus never-users was 0.5

(95% CI, 0.3–0.7). The risk of ovarian can-

cer was significantly lower in women on

oral contraceptives for more than 4 years.

The Royal College of General Prac-

titioners oral contraceptive study, which

started in 1968, collected data from

23,377 COC users and 23,796 never-users

over a period of 14 months, with 339,000

woman-years of observation for never-

users and 744,000 woman-years for ever-

users.17 It was found that the RR was 0.51

for ever-users as compared with never-

users. Similarly, a statistically significant

gradual decrease in risk with increasing

duration of COC use was observed. The

protective effect was found to last for at

least 15 years after stopping COC.

In the early 1970s, it was proposed

that defective cellular repair after ovula-

tion represents the major risk factor for

ovarian cancer development.27 More re-

cent theories assume that ovarian can-

cer development is attributed to either

activated proto-oncogenes or inactivated

tumour-suppressor genes,27,28 which seem

to point to abnormalities of genomic DNA

quantity and quality, with the resulting de-

fects in post-ovulatory ovarian cellular re-

pair being the causative factor for ovarian

cancer. Thus, the protective effect of COC

against ovarian cancer may be attributed

to the resulting anovulation during their

use, which prevents genetic predisposing

cellular repair defects to be expressed.

Although depot MPA also suppresses

ovulation and would theoretically lower

the risk of ovarian cancer, a hospital-based

WHO case-control study failed to uncover

such a protective effect.22

CARCINOMA OF THE CERVIX

There is evidence suggesting that long-

term use of COC for 5 years or more may

be associated with an increased risk of

cervical cancer.29 A meta-analysis of 28

studies, involving 12,531 women with

cervical cancer, suggested that the risk of

cervical cancer may decrease after stop-

ping the use of COC.30 Another IARC analy-

sis which included eight studies found a

fourfold increase in risk among women

with over 5 years of COC use. The risk was

also increased in women who started us-

ing COC before the age of 20 and in those

who had used COC within the previous 5

years.31

The mechanism for increased risk of

cervical cancer in COC users is uncertain.

Human papillomavirus (HPV) has been

recognized to be the major cause of carci-

noma of the cervix. Steroid contraception

has been postulated to be able to bind to

specific DNA sequences within transcrip-

tional regulatory regions on the HPV DNA,

either to increase or suppress the tran-

The protective effect of

COC against ovarian cancer

may be attributed to the

resulting anovulation

during their use



scription of various genes. It was suggest-

ed that the regulatory region of HPV type

16 viral genome indicates transcriptional

control of the HPV genome and might con-

tain enhancer elements that are activated

by steroid hormones.32

In COC users, the cervical mucus be-

comes scanty, thick, and highly viscous. It

has been hypothesized that such mucus

may modulate and prolong the effect of

carcinogenic agents and pathogens (in-

cluding HPV), which might have been car-

ried by coitus, on the cervical squamoco-

lumnar junction, causing them to become

difficult to be removed.33

However, the majority of studies did

not analyse the HPV status of COC users

and controls. Moreover, early use of COC

might be related to early onset of sexual

activity, which is itself a significant risk

factor for HPV infection and development

of cervical cancer. The lower use of the

barrier method of contraception in COC

users might be another accountable fac-

tor for the increase in risk of HPV infec-

tion. However, with the development of

HPV vaccines, the observed association

of increased risk of cervical cancer in COC

users might be changed, and fear of cervi-

cal cancer should not be a reason to avoid

COC use.

A large, population-based, case-con-

trol study in Costa Rica, a hospital-based

WHO case-control study in Thailand, Mex-

ico and Kenya, and a study in New Zealand

found that the risk of cervical cancer did

not appear to be affected by depot MPA

use.22

CONCLUSION

The majority of studies on the relationship

between hormonal contraceptive use and

development of cancer have focused on

COC and breast cancer, albeit with con-

flicting results. From the cumulative expe-

rience and meta-analyses of large epide-

miological studies with a long follow-up

duration, the present evidence suggests

an increase in risk of breast cancer devel-

opment mainly in current COC users, with

Figure 1. Risks of cancer development with duration of oral contraceptive pill (OCP) use

10

1.0

0.1

Rela

tive

risk

for c

ance

r dev

elop

men

t

Duration of OCP use (mo)0–48 49–96 > 97

Increased risk (?) of cervical cancer(Decreased risk on stopping OCP)

Increased risk of breast cancer(Decreased risk on stopping OCP)

Decreased risk of ovary cancer

Decreased risk of endometrial cancer




REFERENCES

the risk decreased on stopping therapy.

By 10 years of cessation of use, the risk

is similar to that in never-users. Similar-

ly, the risk of carcinoma of the cervix is

found to be increased in current COC us-

ers. However, COC use confers a strong

and prolonged protective effect against

carcinoma of the endometrium and ovary,

which will last even after over 20 years of

cessation of therapy (Figure 1).

Depot MPA was observed to cause

a plausible increased risk of breast can-

cer in current or recent users, similar to

that observed in COC users. It does not

seem to affect the overall breast cancer

risk. However, there is strong evidence of

prolonged decreased risk for carcinoma of

the uterine corpus. No strong association

has been noted for carcinoma of the cervix

and ovary.

When counselling women regarding

hormonal contraception, the issue of po-

tential carcinogenic effect from its use is

to be included. However, its prescription

should be based on an individual risk-

benefit assessment, provided contraindi-

cations are taken into account and regular

visits to doctors or health-care profession-

als are made.

1. Kirschstein RL, Rabson AS, Rusten GW. Infil-trating duct carcinoma of the mammary gland of a rhesus monkey after administration of an oral contraceptive: a preliminary report. J Natl Cancer Inst 1972;48:551–553.

2. Welsch CW, Adams C, Lambrecht LK, Hassett CC, Brooks CL. 17β-oestradiol and Enovid mam-mary tumorigenesis in C3H/HeJ female mice: counteraction by concurrent 2-bromo-alpha-ergocryptine. Br J Cancer 1977;35:322–328.

3. Geil RG, Lamar JK. FDA studies of estrogen, progestogens and estrogen/progesterone combi-nations in the dog and monkey. J Toxicol Environ Health 1977;3:179–193.

4. Shubik P. Oral contraceptives and breast cancer: laboratory evidence. IARC Sci Publ 1985;(65):33–35.

5. Anderson TJ, Battersby S, King RJ, McPher-son K, Going JJ. Oral contraceptive use influ-ences resting breast proliferation. Hum Pathol 1989;20:1139–1144.

6. Cogliano V, Grosse Y, Baan R, et al; WHO Inter-national Agency for Research on Cancer. Carci-nogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lan-cet Oncol 2005;6:552–553.

7. Preston-Martin S, Pike MC, Ross RK, Hender-son BE. Epidemiologic evidence for the increased cell proliferation model of carcinogenesis. Prog Clin Biol Res 1991;369:21–34.

8. Henderson BE, Feigelson HS. Hormonal car-cinogenesis. Carcinogenesis 2000;21:427–433.

9. Pike MC, Henderson BE, Casagrande JT, Rosa-rio I, Gray GE. Oral contraceptive use and early

abortion as risk factors for breast cancer in young women. Br J Cancer 1981;43:72–76.

10. Chilvers C, McPherson K, Peto J, et al; UK Na-tional Case-Control Study Group. Oral contracep-tive use and breast cancer risk in young women. Lancet 1989;333:974–982.

11. Cancer and Steroid Hormone Study of the Centers for Disease Control and the National In-stitute of Child Health and Human Development. Oral contraceptive use and the risk of breast can-cer. N Engl J Med 1986;315:405–411.

12. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal con-traceptives: collaborative reanalysis of individual data on 53297 women with breast cancer and 100239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–1727.

13. Colditz GA, Manson JE, Hankinson SE. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J Wom-ens Health 1997;6:49–62.

14. Kumle M, Weiderpass E, Braaten T, Persson I, Adami HO, Lund E. Use of oral contraceptives and breast cancer risk: The Norwegian-Swedish Women’s Lifestyle and Health Cohort Study. Cancer Epidemiol Biomarkers Prev 2002;11:1375–1381.

15. Vessey MP, McPherson K, Doll R. Breast can-cer and oral contraceptives: findings in Oxford-Family Planning Association contraceptive study. Br Med J (Clin Res Ed) 1981;282:2093–2094.

16. Kahlenborn C, Modugno F, Potter DM, Severs WB. Oral contraceptive use as a risk factor for premenopausal breast cancer: a meta-analysis.

Mayo Clin Proc 2006;81:1290–1302.

17. Hannaford PC, Selvaraj S, Elliott AM. Cancer risk among users of oral contraceptives: cohort data from the Royal College of General Practition-er’s oral contraception study. BMJ 2007;335:651.

18. Schneider HPG, Mueck AO, Kuhl H. IARC monographs program on carcinogenicity of com-bined hormonal contraceptives and menopausal therapy. Climacteric 2005;8:311–316.

19. Narod SA, Dubé MP, Klijn J, et al. Oral contra-ceptives and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers. J Natl Cancer Inst 2002;94:1773–1779.

20. Paul C, Skegg DCG, Spears GFS. Depot me-droxyprogesterone (Depo-Provera) and risk of breast cancer. BMJ 1989;299:759–762.

21. Breast cancer and depot-medroxyprogester-one acetate: a multinational study. WHO Collabo-rative Study of Neoplasia and Steroid Contracep-tives. Lancet 1991;338:833–838.

22. Kaunitz AM. Depot medroxyprogesterone acetate contraception and the risk of breast and gynecologic cancer. J Reprod Med 1996;41(5 Suppl):419–427.

23. Backman T, Rauramo I, Jaakkola K, et al. Use of the levonorgestrel-releasing intrauter-ine system and breast cancer. Obstet Gynecol 2005;106:813–817.

24. Schlesselman JJ. Risk of endometrial cancer in relation to use of combined oral contraceptives: a practitioner’s guide to meta-analysis. Hum Re-prod 1997;12:1851–1863.

25. Vessey M, Painter R. Oral contraceptive use and cancer: findings in a large cohort study,

1968–2004. Br J Cancer 2006;95:385–389.

26. Collaborative Group on Epidemiological Stud-ies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371:303–314.

27. Fathalla MF. Incessant ovulation—a factor in ovarian neoplasia? Lancet 1971;298:163.

28. Casagrande JT, Louie EW, Pike MC, Roy S, Ross RK, Henderson BE. “Incessant ovulation” and ovarian cancer. Lancet 1979;314:170–173.

29. Franceschi S. The IARC commitment to can-cer prevention: the example of papillomavirus and cervical cancer. Recent Results Cancer Res 2005;166:277–297.

30. Smith JS, Green J, Berrington de Gonzalez A, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–1167.

31. Moreno V, Bosch FX, Munoz N, et al . Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359:1085–1092.

32. Moodley M, Moodley J, Chetty R, Herrington CS. The role of steroid contraceptive hormones in the pathogenesis of invasive cervical cancer: a review. Int J Gynecol Cancer 2003;13:103–110.

33. Guven S, Kart C, Guvendag Guven ES, Gu-nalp GS. The underlying cause of cervical can-cer in oral contraceptive users may be related to cervical mucus changes. Med Hypotheses 2007;69:550–552.

About the Author

Dr Wong is Consultant in the Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Hong Kong.


This continuing medical education service is brought to you by MIMS. Read the article ‘Hormonal Contraception and Cancers’ and answer the following questions. Answers are shown at the bottom of this page. We hope you enjoy learning with JPOG.

CME Answers for JPOG Jan/Feb 2012

HKCOG CME Article: Management of Pregnancies With Previous Caesarean Section

Answers

1 2 3 4 5 6 7 8 9 10

T F F T T T T T T F

CME Article


Answer True or False to the questions below.

1. Combined oral contraceptives (COC) are classified by IARC as ‘carcinogenic’ to humans.

2. There is strong evidence for COC to cause a rise in epithelial cell proliferation of the glandular breast, leading to increase in genetic error and resulting in malignant transformation and breast cancer development.

3. The results of COC toxicity studies using animal models can accurately reflect the effects in humans.

4. The Collaborative Group on Hormonal Factors in Breast Cancer concluded that there was a small increase in the relative risk of having breast cancer in current and recent COC users.

5. Depot medroxyprogesterone acetate (MPA) users have decreased risk of carcinoma of the breast.

6. The protective effect of COC against carcinoma of the uterine corpus disappears after 10 years of stopping.

7. Low-dose COC has been shown to have a similar protective effect against carcinoma of the ovary compared with high-dose COC.

8. There is good evidence for the protective effect of depot MPA against carcinoma of the ovary.

9. There is evidence suggesting that long-term use of COC increases the risk of cervical cancer.

10. The carcinogenic potential of hormonal therapy should not be discussed during contraceptive counselling.

True False


CME Questions CME QuestionsAnswers

1 2 3 4 5 6 7 8 9 10

T F F T F F T F T F


Journal Watch


Peer reviewed

GYNAECOLOGY

Calcium intake and mortality in Swedish women

Meta-analyses of randomized studies have shown

that taking calcium supplements is associated with

increased risk of coronary disease and stroke. A

Swedish cohort study has confirmed the increased

risk for cardiovascular disease in general but not

for stroke.

The Swedish mammography cohort was set

up in 1987 and included 61,433 women born be-

tween 1914 and 1948. National registries provided

data about all-cause and cardiovascular mortality

over a mean follow-up of 19 years. Food frequency

questionnaires in 1987 and 1997 provided data

about dietary intake and use of calcium supple-

ments for 38,984 women. The relationship between

calcium intake and all-cause mortality took the

form of a ‘J-shaped curve’, with higher mortality

at both extremes of intake. An intake of 1,400 mg

a day of calcium was associated with significant

increases of 40% in all-cause mortality, 49% in

cardiovascular mortality, and 114% in coronary

disease mortality, and no significant change in

stroke mortality, compared with a calcium intake of

600–1,000 mg a day. After further statistical analy-

sis, low intakes of calcium (< 600 mg/day) were

no longer significantly associated with increased

mortality. Among people taking calcium tablets and

with a dietary calcium intake of > 1,400 mg/day,

all-cause mortality was increased 2.6-fold.

High-calcium intake is associated with in-

creased all-cause and cardiovascular mortality.

Michaëlson K et al. Long-term calcium intake and rate of all cause and cardiovascular mortality: community-based prospective longitudinal cohort study. BMJ 2013; 346: 14 (f228).

Outcome of extreme preterm birth in England, 1995–2006

Two successive papers in the BMJ have examined

short-term and long-term outcomes for extremely

preterm births in England in 1995 and 2006.

The prospective national cohort studies pro-

vided short-term data about 666 babies born at 22–

25 weeks’ gestation in England in March to Decem-

ber 1995 and all 3,133 babies born at 22–26 weeks’

gestation in 2006. In 2006, 56% of infants born at

22 weeks and 98% of those born at 26 weeks were

born alive. Active care at birth was withheld from

73% of infants born at 22 weeks, 16% at 23 weeks,

and < 2% at 24 weeks or later. Survival rates for

live-born infants were 2% at 22 weeks, 19% at 23

weeks, 40% at 24 weeks, 66% at 25 weeks, and

77% at 26 weeks. More than two-thirds (68%) of

survivors had bronchopulmonary dysplasia, 16%

were treated (laser treatment) for retinopathy of

prematurity, and 13% had a serious abnormality on

cerebral ultrasound. In 2006, compared with 1995,

44% more infants born alive at 22–25 weeks were

admitted to neonatal intensive care, and survival

of infants born at 23, 24, and 25 weeks increased

by 9.5%, 12%, and 16%, respectively. Overall, the

proportion treated for retinopathy increased from

12% in 1995 to 22% in 2006.

Neurodevelopmental outcomes at ages 2–3

years were assessed for 1,031 survivors in the 2006

cohort. The prevalence of moderate or severe im-

pairment was 45% among survivors born at 22–23

weeks, 30% at 24 weeks, 25% at 25 weeks, and

20% at 26 weeks. Overall, one in seven (14%) had

cerebral palsy, usually mild or moderate. Mean

predicted adjusted mental development index quo-

tients (Bayley scales) were 80 (22–23 weeks), 87

(24 weeks), 88 (25 weeks), and 91 (26 weeks). In

the 2006 cohort, a greater absolute number of chil-

dren than in 1995 will need lifelong special care. It

is calculated that of every 100 infants born at 24

weeks, 60 will die despite intensive care and 12

of the 40 survivors will have serious impairments.

Further follow-up of the 2006 cohort is planned.

Between 1995 and 2006 the survival of ex-

tremely preterm infants in England improved, but

PAEDIATRICS

JPOG_JulAug_2013_COMBINE_Final.indd 133 8/5/13 2:18 PM


more survived with disability. The writers of an

editorial point out that in the Netherlands, infants

born before 24 completed weeks are not routinely

offered neonatal intensive care.

Costeloe KL et al. Short term outcomes after extreme preterm birth in England: comparison of two birth cohorts in 1995 and 2006 (the EPICure studies). BMJ 2012; 345: 14 (e7976); Moore T. Neurological and developmental outcome in extremely preterm children born in England in 2006 and 1995: the EPICure studies. Ibid: 15 (e7961), Groenendaal F, Uiterwaal C. Long-term follow-up of extremely preterm neonates. Ibid:10 (e8252) (editorial).

New anti-interleukin therapies for systemic JIA

Systemic juvenile idiopathic arthritis (systemic JIA)

frequently leads to joint damage and disability and

is accompanied by systemic features such as fever,

rash, hepatosplenomegaly, and serositis. High-

dose steroid treatment may lead to severe toxicity,

and treatments such as methotrexate and tumour

necrosis factor inhibitors may be ineffective. The

effectiveness of antibodies to the interleukin-6

receptor and to interleukin-1β has been shown in

successive papers in the New England Journal of

Medicine.

The humanized, antihuman interleukin-6-re-

ceptor monoclonal antibody, tocilizumab, was as-

sessed in a placebo-controlled trial at 43 centres

in Europe North America, and South America. The

trial included 112 children aged 2–17 years with

active, treatment-resistant systemic JIA. Random-

ization (2:1) was to intravenous tocilizumab or

placebo every 2 weeks for 12 weeks. At 12 weeks,

an improvement of at least 30% on the American

College of Rheumatology JIA score (JIA ACR 30)

was achieved by 85% (tocilizumab) vs 24% (pla-

cebo), a highly significant difference. At week 52

a JIA ACR 70 response (at least 70% improvement)

was achieved by 80% in the tocilizumab group and

a JIA ACR 90 response by 59%. Steroid therapy had

been stopped by 52% in this group, and 48% had no

active arthritis. Common adverse events with tocili-

zumab included infections, neutropenia, and raised

aminotransferase levels.

The fully human, anti-interleukin-1β mono-

clonal antibody, canakinumab, was assessed in two

international trials reported together, including 84

and 100 patients. In the first trial, randomization

was to subcutaneous canakinumab or placebo, and

an adapted JIA ACR 30 response was achieved by

84% (canakinumab) vs 10% (placebo). In the sec-

ond trial, 100 patients who had responded to 32

weeks of canakinumab were randomized to con-

tinued canakinumab or to placebo. A disease flare

occurred in 74% (canakinumab) vs 25% (placebo).

The median time to disease flare was incalculable

in the canakinumab group and 236 days in the pla-

cebo group. The disease became inactive in 62%

vs 34%. One in three patients on canakinumab was

able to discontinue steroid therapy. Infections were

frequent with canakinumab and five patients (ver-

sus two in the placebo group) developed the mac-

rophage activation syndrome.

Both tocilizumab and canakinumab were ef-

fective treatment for systemic JIA but more data

are needed about toxicity.

De Benedetti F et al. Randomized trial of tocilizumab in systemic juvenile idiopathic arthritis. NEJM 2012; 367: 2385–95; Ruperto N et al. Two randomized trials of canakinumab in systemic juvenile idiopathic arthritis. Ibid: 2396–406; Sandborg C, Mellins ED. A new era in the treatment of systemic juvenile idiopathic arthritis. Ibid: 2439–40 (editorial).

CPAP vs surfactant and higher vs lower oxygen saturation for extremely preterm infants: Outcomes at 18–22 months

The Surfactant, Positive Pressure, and Pulse Oxim-

etry randomized trial was a multicentre, random-

ized, controlled trial with a 2 × 2 multifactorial

design in which 1,316 extremely preterm infants

(born at 24 weeks 0 days to 27 weeks 6 days) were

randomized at 20 US centres to early continuous

positive airway pressure (CPAP), or early surfactant

via an endotracheal tube and to a target oxygen

saturation of 85–89% or of 91–95%. Early assess-

ment (at 36 weeks’ postmenstrual age) showed

similar rates of death or bronchopulmonary dyspla-

sia with either CPAP or surfactant, and the lower

target range for oxygen saturation was associ-

ated with less retinopathy of prematurity but more


Journal Watch


Peer reviewedPeer reviewed

deaths. Now, surviving infants have been assessed

at 18–22 months.

Neurodevelopmental status was assessed

at 18–22 months in 990 of 1,058 surviving infants

(94%). Death or neurodevelopmental impairment

occurred in 27.9% (CPAP) vs 29.9% (surfactant),

a non-significant difference, and in 30.2% (lower

oxygen saturation target) vs 27.5% (higher oxygen

saturation target), also a non-significant differ-

ence. There was a significant increase in mortality

with the lower oxygen saturation target (22.1% vs

18.2%).

These researchers conclude that outcomes

are similar with early CPAP and with early surfac-

tant, but the lower oxygen saturation target should

not be used in the care of extremely premature

babies.

Vaucher YE et al. Neurodevelopmental outcomes in the early CPAP and pulse oximetry trial. NEJM 2012; 367: 2495–504.

Paromomycin for cutaneous leishmaniasis

York. It included inpatients and outpatients in No-

vember to May each year between 2003 and 2009

with children presenting with an acute respiratory

illness or fever. The rate of HMPV detection (using

reverse transcriptase–polymerase chain reaction

assay on nose and throat swabs) was 200/3,490

(6%) among children in hospital, 222/3,257 (7%)

among children in outpatient clinics, 224/3,001

(7%) among children in the emergency department,

and 10/770 (1%) among healthy children in well-

child primary care clinics. Rates of hospital admis-

sion with HMPV infection among children aged < 5

years were 1 in 1,000 (3 in 1,000 at age < 6 months,

and 2 in 1,000 at age 6–11 months). Among chil-

dren admitted to hospital with an acute respiratory

illness or fever, those with HMPV infection were

older, more likely to be diagnosed as pneumonia or

asthma, to need supplemental oxygen, and to stay

longer in intensive care, compared with children

testing negative for HMPV. It was estimated that

among 1,000 children of this age, HMPV would,

each year, cause 55 clinic visits and 13 visits to the

emergency department. Among children admitted

to hospital, coexisting high-risk conditions (pre-

Leishmaniasis is prevalent in Eurasia, Africa, and

the Americas, and although cutaneous leishmani-

asis eventually resolves without treatment it is the

cause of much morbidity. Cutaneous leishmaniasis

due to Leishmania major is prevalent in Tunis, and

a trial there has shown topical paromomycin to be

effective treatment.

A total of 375 patients aged 5–65 years (half

of them children < 17 years old) were randomized

to three groups: 15% paromomycin, 15% paromo-

mycin plus 0.5% gentamicin, or vehicle alone (pla-

cebo), all applied as topical creams for 20 days to

all ulcerated skin lesions including an index lesion

(1–5 cm diameter with leishmania demonstrated).

Cure of the index lesion was achieved in 82% (par-

omomycin), 81% (paromomycin/gentamicin), and

58% (placebo). Only seven patients (five in the pla-

cebo group) had any persisting lesions after cure of

the index lesion. Mild to moderate local reactions

occurred with paromomycin.

Topical treatment with paromomycin cream,

with or without gentamicin, was effective treat-

ment for cutaneous leishmaniasis due to L major.

Salah AB et al. Topical paromomycin with or without gentamicin for cutaneous leishmaniasis. NEJM 2012; 368: 524–32.

HMPV infection in young children in the US

Human metapneumovirus (HMPV) is a paramyxovi-

rus discovered in 2001 as a cause of acute respi-

ratory illness in infants and young children world-

wide. It also affects old people and people with

debilitating illnesses. A study at three US sites

has provided more data about the epidemiology of

HMPV in children under the age of 5 years.

The survey by the Centers for Disease Con-

trol and Prevention (CDC) New Vaccine Surveillance

network took place at three sites, in Cincinnati,

Ohio; Nashville, Tennessee; and Rochester, New



mature birth, immunodeficiency, chronic disease of

lungs, heart, or kidneys, cancer, or sickle-cell dis-

ease) were present in 40% (HMPV-positive) vs 30%

(HMPV-negative).

In the US, HMPV is an important cause of

acute respiratory illness and fever among young

children.

Edwards MK et al. Burden of human metapneumovirus infection in young children. NEJM 2013; 368: 633–43.

Risk factors for stillbirth

There has been little improvement in stillbirth rates

in recent years, and the importance of risk factors

has been uncertain. A study in the West Midlands

region of England has included 91,829 live births

and 389 stillbirths.

Risk factors for stillbirth included parity

(para 0 and 3+), ethnicity (African, Afro-Caribbean,

using country-specific data) for infants born in

hospital in 23 developing countries in Africa and

Latin America (2004–2005) and Asia (2007–2008),

including 276,436 singleton live births or fresh still-

births. The 90th percentile for birth weight varied

from 3,250 g in India to 4,050 g in Algeria, and the

prevalence of a birth weight of 4,000 g or greater

was 0.5% in India and 14.9% in Algeria. Factors

found to be significantly associated with macroso-

mia included higher maternal age (20–34 years),

higher maternal height, higher parity, higher mater-

nal BMI, maternal diabetes, post-term pregnancy,

and male fetus. Macrosomia was associated with

increased risk of caesarean section and of adverse

maternal outcomes. The risk of adverse perinatal

outcome was increased in Asia.

The increase in obesity and diabetes in

women of reproductive age might have led to an

increase in fetal macrosomia worldwide. Research

into ways of controlling these factors is needed.

Koyanagi A et al. Macrosomia in 23 developing countries: an analysis of a multicountry, facility-based, cross-sectional survey. Lancet 2013; 381: 476–83; Dennedy M, Dunne F. Macrosomia: defining the problem worldwide. Ibid: 435–6 (comment).

Indian, or Pakistani), maternal obesity (BMI 30 or

greater), smoking, pre-existing diabetes, history of

mental health problems, and pregnancy complica-

tions (antepartum haemorrhage, fetal growth re-

striction). The greatest risk factor was fetal growth

restriction which increased the risk of stillbirth by a

factor of 7.8 in non-smoking mothers, 5.7 in smok-

ing mothers, and 10.0 in mothers only exposed to

passive smoking. The population attributable risk

from fetal growth restriction was 6.2% if detected

antenatally and 32.0% when not detected antena-

tally. Antenatal detection of fetal growth restric-

tion was associated with delivery 10 days earlier

on average. The stillbirth rate (per 1,000 births)

was 4.2 overall, 2.4 in pregnancies with no fetal

growth restriction, 9.7 when fetal growth restric-

tion was detected antenatally, and 19.8 when fetal

growth restriction was not detected antenatally.

This study identifies fetal growth restriction

as the main risk factor for stillbirth. Antenatal de-

tection of fetal growth restriction and appropriate

early delivery might prevent 600 stillbirths each

year in the UK.

Gardosi J et al. Maternal and fetal risk factors for stillbirth: population based study. BMJ 2013; 346: 15 (f108); McCowan LME, Groom KM. Identifying risk factors for stillbirth. Ibid: 7 (f416) (editorial).

Fetal macrosomia in developing countries

Fetal macrosomia may lead to perinatal death,

perinatal asphyxia, shoulder dystocia, caesarean

section, maternal haemorrhage, prolonged labour,

and perinatal trauma. In the developed world, the

prevalence of macrosomia has increased along

with maternal obesity and diabetes. Little is

known, however, about fetal macrosomia in devel-

oping countries. Now, the WHO Global Survey on

Maternal and Perinatal Health has provided data

about macrosomia (birth weight > 90th percentile

OBSTETRICS


Imaging Paediatric Brain Tumours

Tang Phua Hwee, MBBS, FRCR, MMed Diagnostic Radiology

PAEDIATRICS i Peer reviewed


INTRODUCTION

Prompt identification and management of hearing loss in childhood is essential to en-

sure optimal speech and language development in the early years of life and optimal

school performance for older children. The implementation of universal neonatal hear-

ing screening has facilitated earlier identification of congenital losses, with 1 in 1,000

babies being born in the UK with a permanent hearing loss detectable at birth. However,

for more common acquired otological conditions of childhood such as otitis media with

effusion (OME) or hearing loss in the presence of complex medical needs, the paediatri-

cian may be the first port of call for parents or concerned allied health professionals,

and an understanding of the presentation, assessment and management options can

ensure optimal outcomes for this group of children. The aim of this review is to discuss

the clinical assessment of hearing loss and provide an overview of the management

options available.

Types of Hearing LossHearing loss is classified by both type and severity. The types of hearing loss include

conductive, sensorineural, or mixed hearing loss, which can be subdivided into congeni-

tal or acquired. A differential diagnosis is included in Table 1. The severity of hearing

loss in decibels hearing level (dBHL) is based on a ‘pure tone average’, which is the

mean hearing threshold at four different sound frequencies (0.5, 1, 2, and 4 kHz). This

allows classification of the patients hearing as normal (< 25 dBHL) or determines the

extent of the loss; mild (25–50 dBHL), moderate (51–70 dBHL), severe (71–90 dBHL),

profound (91–110 dBHL), or total (> 110 dBHL). Management of the hearing loss thus

depends on its aetiology and the extent of the loss.


The Management of

Hearing Loss in Children

Marianne D Elloy, MBBS, MRCS, DOH-NS; Andrew H Marshall, FRCS(ORL-HNS)



PAEDIATRICSPAEDIATRICS I Peer revIewed

Table 1. Differential diagnosis of paediatric hearing loss

Inheritance/aetiology Sensorineural HL Conductive HL

Congenital Non-syndromic

Autosomal dominantAutosomal recessiveX-linked

DFNA chromosome loci: approx 40 gene loci identified. Most common is GJB2 (connexin 26)

DFNB chromosome loci: approx 35 gene loci identified. Most common is GJB2 (connexin 26)

DFN chromosome loci: approx 5 gene loci identified

Congenital ossicular anomalies/fixation

Syndromic Autosomal Dominant Branchio-oto-renal syndromeCraniosynostosis syndromes (Crouzon, Apert,

Muenke, Pfeiffer)DiGeorge syndromeNoonan syndrome

Osteogenesis imperfectaMarshall syndromeNeurofibromatosis type 2Saethre-Chotzen syndrome Stickler syndromeWaardenburg syndrome

Treacher Collins syndrome

Autosomal recessive Albers-Schonberg diseasePendred syndromeEnlarged vestibular aqueduct

syndromeUsher syndrome

X-linked Alport syndromeMohr-Tranebjaerg syndrome

Trisomy Sporadic, autosomal dominant or autosomal recessive

Down syndromeCHARGE syndrome

Goldenhar syndrome

Alport syndrome

Acquired Prenatal Intrauterine infectionsIntrauterine exposure to

ototoxic drugsPostnatal Infectious AOM and complications of

AOM MeningitisViral: measles, mumps, CMV

Inflammatory Chronic suppurative otitis media

OME

Trauma Tympanic membrane perforation

Temporal bone fractureAOM = acute otitis media; CMV = cytomegalovirus; OME = otitis media with effusion.




CLINICAL ASSESSMENT

detectionThe detection of hearing loss either occurs as a re-

sult of a universal neonatal hearing screening pro-

gramme or as a result of parental or professional

concern.

Universal neonatal hearing screening is now

well established in the United Kingdom, and its aim

is for early identification of hearing loss as early

intervention results in better speech and language

outcomes.

The programme aims for babies to be screened

by the gestational age of 44 weeks but preferably

before discharge from hospital. Premature babies

should not be screened before the gestational age

of 34 weeks. Two screening pathways exist: the

well baby protocol and the neonatal intensive care

unit/special care baby unit (NICU/SCBU) protocol

(for those babies admitted to NICU/SCBU for more

than 48 hours). The exclusions for hearing screen-

ing are ear atresia, microtia, or meningitis, with a

recommendation of direct referral for audiological

assessment in this high-risk group.

A dedicated hearing screening team under-

takes the assessments. The first step of the well

baby protocol comprises otoacoustic emissions

(OAEs) testing. If the baby passes this test and

there is no history of risk factors, then the baby is

discharged from the programme. If the baby fails

the test, it can be repeated on a second occasion;

however, if the test is failed again, an automated

auditory brainstem response (AABR) test is per-

formed. If this is passed and there are no risk fac-

tors, the baby can be discharged; however, if the

AABR is missed or incomplete, or if there are no

clear responses in one or both ears, or risk factors

are identified, further audiological assessment will

be arranged. The NICU/SCBU protocol differs with

all babies undergoing an OAE and AABR test. The

audiological referral criteria are the same as for

the well baby protocol; in addition, if either test

is missed or the tests provide inconsistent results,

then audiological review is recommended. A poten-

tial pitfall using OAEs as a screening tool is the po-

tential failure to identify a condition called auditory

neuropathy. In these children, OAEs are present but

their auditory brainstem response and functional

hearing can be very poor. This condition is more

prevalent in babies that would be identified by the

NICU/SCBU protocol in particular due to prematu-

rity or neonatal infection.

HistoryThe history in neonates should explore the risk fac-

tors identified in the neonatal hearing screening

pathway which includes parental or professional

concern regarding the infants’ hearing or develop-

ment of auditory or vocal behaviour, high risk of

chronic middle ear problems, for example, Down

syndrome or cleft palate, craniofacial anomalies,

family history of permanent sensorineural hearing

loss from early childhood (in parents or siblings),

intermittent positive pressure ventilation, on NICU

or SCBU for more than 5 days, jaundice or hyperbili-

rubinaemia requiring exchange transfusion, proven

or possible congenital infections, TORCH (Toxoplas-

mosis Other: syphilis Rubella CMV Herpes), neuro-

degenerative or neurodevelopmental disorders, and

exposure to ototoxic drugs with monitored levels

outside the therapeutic range. However, it is im-

portant to be aware that 50% of newborns born

with permanent bilateral congenital hearing loss

do not have any known risk factors. Other risk fac-

tors which have been identified include bacterial

meningitis, whereby 10% of children will develop

a subsequent sensorineural hearing loss, Apgar

scores (0–4 at 1 minute and 0–6 at 5 minute), birth

weight less than 1,500 g, and consanguinity.

The history in older children should focus on




the impact of the hearing loss on the child’s speech

and language development, and school perfor-

mance including social interaction with their peers.

The aforementioned risk factors should be explored

and, in addition, enquiry about previous head injury

and temporal bone fracture.

examinationIn neonates with hearing loss, a full clinical exami-

nation should be undertaken as syndromic hearing

loss can occur in up to 30% of children with bilat-

eral permanent hearing loss and other associated

features may be noted.

Otoscopy may be normal. Examination for

craniofacial anomalies, preauricular pits, sinuses,

and branchial pits should be undertaken. Some syn-

dromes with hearing loss can be associated with

ophthalmological anomalies and all children with

a moderate or greater sensorineural hearing loss

should be reviewed by an ophthalmologist.

In older children, the most common cause of

childhood hearing loss is OME, and a number of dif-

ferent characteristic otoscopic appearances have

been identified including a dull tympanic membrane,

loss of the light reflex, flattening of the handle of

the malleus, a golden or blue hue of the tympanic

membrane, and indeed sometimes an air fluid me-

niscus or retrotympanic bubbles. Otosocopy should

also assess for the presence impacted cerumen, a

foreign body, infection, congenital cholesteatoma,

tympanosclerosis (the presence of calcification in

the tympanic membrane which can sometimes also

affect the ossicular chain), perforation (including

the site, extent, and status of the middle ear mu-

cosa), or an attic defect with the possibility of cho-

lesteatoma.

The facial nerve should also be assessed par-

ticularly in the presence of pathology.

investigations

Audiological Assessments

The objective audiological assessments can be per-

formed for children of any age as no contribution to

the testing process is required by the patient.

Otoacoustic emissions: the principle of

OAEs is that objective sounds are emitted from the

outer hair cells of a normally functioning cochlea.

OAEs can be spontaneous or occur in response to

acoustic stimulus. Transient-evoked OAEs are used

in neonatal hearing screening, whereby broadband

clicks are delivered to the ear by a handheld probe

which also contains a microphone to detect the

emissions. The presence of OAEs indicates a hear-

ing threshold of 20–40 dBHL. The test is quick and

easy to administer, is not affected by sleep, and has

a high sensitivity (97%), making it a useful screen-

Syndromic hearing

loss can occur in up

to 30% of children

with bilateral

permanent hearing

loss and other

associated features

may be noted




ing tool. The limitations include inability to esti-

mate hearing thresholds or assess specific frequen-

cies. Specificity is low and patients can fail if they

have impacted wax or OME. In contrast, a patient

can pass this test but still have hearing problems

due to an auditory neuropathy.

Auditory brainstem response: Auditory

brainstem response can be undertaken as an auto-

mated test or a manually interpreted test to gain in-

formation about frequency-specific hearing thresh-

olds including bone conduction thresholds and

aided thresholds, which is useful in the presence

of ear canal atresia and microtia. The test takes

longer to administer and may require sedation or

general anaesthesia. The test records the activity

of the eighth cranial nerve and auditory pathways in

response to acoustic stimulus (via headphones) by

adhesive scalp electrodes. It has a high specificity

and sensitivity (> 90%).

Tympanometry: this test does not assess

hearing but is used to assess the compliance of

the tympanic membrane and is particularly useful

in the assessment of OME. The shape of the graph

produced gives information about middle ear com-

pliance, and a normal peak (type A) suggests nor-

mal middle ear function, a flattened peak (type B)

is suggestive of OME (or perforation in the presence

of a high ear canal volume), and a peak shifted to a

more negative pressure (type C) suggests Eustachi-

an tube dysfunction.

Behavioural Testing

Visual reinforcement audiometry: this can be

used for children age 6 months to 3 years. It is un-

dertaken in a specially adapted audiology room us-

ing either sound field speakers or ear inserts. The

child must be able to sit on a parent’s lap and be

able to turn their head to the sound. A distracter en-

tertains the child with toys and sounds are played

via the speakers; when the child turns correctly in

response to the sound, a visual reward is triggered

by the audiologist conducting the test, for example,

an illuminated moving toy. The advantages are that

hearing thresholds can be determined; however,

children can tire and lose interest which reduces

accuracy.

Conditioning/play audiometry: this can be

used for children aged above 2–3 years who can

obey simple commands. The child is instructed to

perform a task each time they hear a sound pre-

sented by headphones or sound field, for example,

put a wooden man into a wooden boat. This is easy

to perform and can establish hearing thresholds but

is dependent on the compliance of the child.

Pure tone audiometry: this is the same test

as used for testing the hearing in adults, requiring

the patient to press a button each time they hear

a sound. This can be used in children aged 4 years

and above, depending on their compliance. Parents

and siblings should be investigated with pure tone

audiometry.

Imaging: imaging is indicated in patients with

a bilateral severe to profound hearing loss, severe

to profound unilateral hearing loss, or a progres-

sive loss. Local protocols vary and a combination of

computed tomography or magnetic resonance imag-

ing (MRI) scanning can be utilized. The anatomical

features or variations of the cochlear nerves and

structure of the cochlea are well demonstrated by

MRI. However, the presence of ossification within

the cochlea, in particular, following meningitis is

optimally imaged using computed tomography.

Other investigationsElectrocardiogram (ECG): all children identified

to have a bilateral permanent hearing loss should

undergo an ECG to assess for a prolonged QT inter-

val which is typically found in Jervell and Lange-

Neilsen syndrome, and if anomalies are identified

referral to a cardiologist should be undertaken.




Blood tests and urinalysis: screening for

congenital infections can be undertaken by test-

ing for antibodies to toxoplasmosis, syphilis, ru-

bella, cytomegalovirus, and herpesvirus hominis. A

blood test for connexin 26, the most common non-

syndromic genetic cause of hearing loss, can be

undertaken. Urea and electrolytes accompanied by

urinalysis for haematuria should be undertaken as

part of a screen for Alport syndrome and branchio-

oto-renal syndrome. Thyroid function tests may be

normal in the early stages of Pendred syndrome,

and the classical investigation described is the per-

chlorate discharge test.

MANAGEMENT

For optimal outcomes, the management of paedi-

atric hearing loss requires a multidisciplinary ap-

proach. This can involve a host of medical speciali-

ties, in addition to ear, nose and throat surgeons,

the general practitioner, paediatricians, ophthal-

mologists, to manage ocular anomalies and the

geneticists to investigate aetiology of hearing loss

and undertake genetic counselling with the child’s

family.

The child’s family will require a significant

amount of support and input from the allied health

professionals. The audiologists not only assess

the child’s hearing thresholds but also are the key

professional group in the provision of adjuncts to

hearing rehabilitation including the management

of hearing aids, bone anchored hearing aids, and

cochlear implant programming. The teachers of

the deaf provide support in the learning environ-

ment before children even start school. They spend

time in the classroom to assess what adjuncts may

be required and train the teachers in mainstream

schools on how to support children with hearing im-

pairment. The speech and language therapists’ in-

put is essential to optimize language development,

and they often work both in the community and the

school to provide input. At the initial consultation,

simple measures to enhance a child’s hearing abili-

ties should be discussed with the parents, including

classroom placement strategies.

Conservative ManagementChildren who are making satisfactory progress de-

spite a mild to moderate hearing loss may be man-

aged with supportive measures. The majority of

children with OME are managed conservatively as a

significant proportion resolve spontaneously.

In those children requiring intervention, the

use of hearing aids for amplification is the main-

stay of auditory rehabilitation. Most children will

require behind-the-ear digital hearing aids, with

multiple brands and models available. These can be

programmed to the child’s hearing loss and a mould

can be made to fit specifically in their ear, which

can be personalized with the logo of their favourite

football team being popular for boys and glitter be-

ing a hit with girls! For children with ear canal atre-

Figure 1. Soft band hearing aid.




sia, microtia and other conductive deficits, a soft

band hearing aid – a bone conductor hearing aid

attached to a soft head band is particularly useful

(Figure 1). Adjunctive devices such as FM systems

for the classroom can also be valuable in hearing

rehabilitation.

Surgical ManagementGrommets (ventilation tubes): grommets are

small (plastic or titanium) tubes that are inserted

into the tympanic membrane to aerate the middle

ear. Multiple types of grommets are available with

the design impacting on the duration that the grom-

met is maintained in position. The main indication

for grommet insertion is persistent OME with hear-

ing loss. The National Institute for Clinical Excel-

lence (NICE) guidance (2008) advocates a 3-month

period of active observation as during this period

the effusion will resolve in 60% of children. For

persistent bilateral OME with hearing in the better

ear of 25–30 dBHL or worse, grommets can be of-

fered, or if the hearing is less than 25–30 dBHL but

is significantly impacting the child’s development

or education. Children with Down syndrome often

have persistent problems with OME, and grommet

placement can be technically difficult owing to nar-

row ear canals and children often require multiple

sets of grommets. As such, a separate pathway is

provided for children with Down syndrome, with

more emphasis on utilization of hearing aids than

grommets. A third pathway for children with cleft

palate advocates the use of grommets in the pres-

ence of persistent OME and hearing loss.

The advantages of grommet placement are

instantaneous improvement in hearing. This quick

procedure is carried out under general anaesthetic

in children and often preferred by parents to the

alternative; a hearing aid as compliance is not an

issue. The risks of surgery include bleeding, infec-

tion, and perforation. In the event of infection, a

1-week course of topical antibiotic ear drops is ad-

vocated as the first-line treatment, on rare occa-

sions, for refractory infections grommets have to be

surgically removed.

Bone-anchored hearing aid (BAHA): the

indications for BAHA in children include congeni-

tal aural atresia and microtia, chronic suppurative

otitis media, persistent OME, chronic otitis externa,

unilateral profound hearing loss, failure with con-

ventional aids, and trauma to the external ear ca-

nal. A BAHA is a titanium screw inserted into the

Figure 2. Cochlear implant.

How a cochlear implant works1. The sound processor (A) captures sound and converts it

into digital code.2. The sound processor transmits the digitally coded sound

through the coil (B) to the implant (C) just under the skin.3. The implant converts the digitally coded sound to electrical

signals and sends them along the electrical signals and sends them along the electrode array, which is positioned in the cochlea.

4. The implant’s electrodes stimulate the cochlea’s hearing nerve fibres, which relay the sound signals to the brain to produce hearing sensations.




calvarial bone behind the ear. Osseointegration

occurs, resulting in bony growth into the titanium

screw. Once this has occurred, typically after 3–6

months, a bone conduction hearing aid system can

be attached, providing the child with amplification.

Typically, children cannot undergo implantation un-

til the age of 4 years; until a child is old enough to

undergo implantation, a soft band hearing aid can

be used.

The risks of surgery include failure of osse-

ointegration, soft tissue complications, trauma,

fixture displacement, and extradural haematoma.

Outcomes measures include favourable audiologi-

cal outcomes in comparison to conventional bone

conductor aids and good patient usage and satis-

faction after up to 10 years after fitting.

Cochlear implant: A cochlear implant is

a surgically implantable device which bypasses

damaged hair cells in the cochlea and directly

stimulates the cochlear nerve (Figures 2 and 3). It

comprises a microphone (which captures speech

and sound), processor (which converts the sound

into an electronic signal), transmitting coil (which

transmits the electronic signal), internal receiver/

stimulator (which receives the electronic signal and

converts it to electrical impulses), and an electrode

array (which delivers the electrical impulses to the

cochlear nerve). Modern electrodes are multichan-

nel with the common brands having 16 and 22 chan-

nels. Each channel stimulates a specific frequency.

Insertion of an electrode into the cochlear risks

destroying any residual hearing and makes the ear

reliant on the cochlear implant. However, hybrid

cochlear implants have been recently developed

for use in patients with serviceable low-frequency

hearing but a profound high-frequency loss, where-

by a custom made shortened electrode is inserted

with the aim of preserving existing hearing in the

low frequencies.

NICE guidance introduced in 2009 recom-

mended that simultaneous bilateral cochlear im-

plantation was indicated in children with severe

to profound hearing loss who did not receive ad-

equate benefit from conventional hearing aids. The

guidelines defined the severe to profound hearing

loss as ‘only hearing sounds louder than 90 dBHL

at the frequencies of 2 and 4 kHz without hearing

aids’ and recommended the utilization of speech,

language, and listening skills appropriate for age,

Figure 3. Cochlear implant.




combined with development stage and cognitive

ability as factors to consider when assessing the

benefit of hearing aids. The main contraindications

to cochlear implantation include lack of a cochlear

nerve and complete ossification of the cochlea.

Patients who are candidates for cochlear im-

plantation will be assessed by a specialist multi-

disciplinary cochlear implant team including au-

diological investigations, imaging, trial of hearing

aids, review by the speech and language therapists,

and teachers of the deaf to assess for suitability.

Confirmation that children are vaccinated against

pneumococcus will be sought preoperatively from

the general practitioner.

The risks of surgery include altered taste, in-

fection, meningitis, device failure, and facial nerve

injury. It is also important to note that children with

cochlear implants cannot undergo MRI.

Outcome measures have been used to assess

the efficacy of cochlear implantation, and there is

a wealth of literature supporting the success of

hearing rehabilitation and these include ‘sound lo-

calization’ and ‘speech recognition in noise’ meas-

ures and quality of life. Reports have revealed that

many children who received cochlear implants at an

early age are achieving age-appropriate academic

performance.

FURTHER READING

Browning G. Clinical otology & audiology. London: Arnold, 2001.Gerber S. The handbook of pediatric audiology. Washington: Gallau-

det, 1996.Graham J, Scadding G, Bull P. Pediatric ENT. Heidelberg: Springer,

2007.Johnston J, Durieux-Smith A, O’Connor A, Fitzpatrick E. Bilateral

cochlear implants: a critical review. Int J Audiol 2009;48:601–617.Kral A, O’Donoghue G. Profound deafness in childhood. N Engl J Med

2010;363:1438–1450.Kunst D, Kremer H, Cremers C. Genetics for ENT specialists. London:

Remedica, 2005.McDermott A-L, Sheehan P. Bone anchored hearing aids in children.

Curr Opin Otolaryngol Head Neck Surg 2009;17:488–493.NHS. Newborn Hearing Screening Programme. Available online at

http://hearing.screening.nhs.uk/; 2011 (accessed 5 Mar 2011).NICE. Cochlear implants for children and adults with severe to

profound deafness: NICE technology appraisal guidance 166. Avail-able online at http://guidance.nice.org.uk/TA166; 2009 (accessed 5 Mar 2011).

NICE. Surgical management of otitis media with effusion in children: NICE clinical guideline 60. Available online at http://guidance.nice.org.uk/CG60; 2008 (accessed 5 Mar 2011).

Papsin B, Gordon K. Bilateral cochlear implants should be the stan-dard for children with bilateral sensorineural deafness. Curr Opin Otolaryngol Head Neck Surg 2008;16:69–74.

Phillips J, Yung M, Burton M, Swan I. Use of aminoglycoside contain-ing ear drops in the presence of a perforation evidence review and ENT UK consensus statement. Available online at http://www.entuk.org/news/news/attachments/eardrops; 2007 (accessed 5 Mar 2011).

US Preventative Services Task Force. Universal screening for hearing loss in newborns: US Preventative Services Task Force recommen-dation statement. Pediatrics 2008;122:143–148.

© 2011 Elsevier Ltd. Initially published in Paediatrics and Child Health 2011;22(1):13–18.

About the AuthorsMarianne D Elloy is Specialist Registrar in Otolaryngology at the Royal Derby Hospital, Derby, UK. Andrew H Marshall is Consul-tant in Otolaryngology at the Queens Medical Centre, Notting-ham, UK.

Practice points

• Universal neonatal hearing screening identifies 1 in 1,000 babies with a permanent hearing impairment at birth.

• Management of hearing loss is a multidisciplinary approach.• Hearing loss can be associated with a syndrome, and associated

features should be identified and investigated.• Early identification and intervention of hearing loss improves

speech and language outcomes.• Conventional digital hearing aids or soft band bone conductor

hearing aids can be used for children.• If hearing aids provide inadequate response to sound, patients

can be referred for assessment for either bone-anchored hearing aid or cochlear implants, depending on their hearing thresholds and pathology.


Professor Nimish Vakil talks about management of patients with refractory GERD.“Successful treatment of refractory GERD requires thorough investigation of the patient situation.”

Professor David Liebermanshares his perspective on the present and future of colorectal cancer screening.“There is a lot of potential to prevent many cancers if we can improve the rate of CRC screening.”

Dr Markus Cornbergdiscusses the management of chronic hepatitis B.“The aim of therapy should be the cure or control of HBV infection without the need for life-long treatment.”

In this Series, find out what these medical experts have to say about latest updates in the management of refractory GERD, the management of chronic hepatitis B and the present & future of colorectal cancer screening.

Current Opinion in Gastroenterology

SCAN TO WATCH VIDEO

Brought to you by MIMS

MIMS Video Series features interviews with leading experts.

Got a spare 5 minutes? Go to www.mims.asia/video_series

MIMS Video Series

By Doctors • For DoctorsBY DOCTORS

FOR DOCTORS

MIMSVideo2013_GERD_ad_206x276.indd 1 1/8/13 12:42 PM



PAEDIATRICSOBSTETRICS i Peer reviewed


INTRODUCTION

Most women presenting with complications in early pregnancy are assessed, diagnosed

and managed at early pregnancy assessment units (EPAUs). These units aim to provide

thorough assessments, access to specialist investigations (scan, human chorionic gon-

adotrophin [hCG]), a rapid turnaround of results, and co-ordination of further manage-

ment.

The EPAU enables continuity of care, fewer admissions, and planned follow-up. It

is beneficial in the provision of open access for GPs, and ideally patients particularly

following a previous pregnancy loss. By streamlining investigations and treatment, this

system is also more cost-effective.

For women who have had previous pregnancy complication, a familiar setting and

ongoing support in a future pregnancy is a valued service.

A downside of the EPAU system is that it is often only available at limited times,

thus for complications occurring outside of these hours, patients require ward contact

numbers and more frequent inpatient based care.

HYPEREMESIS GRAVIDARUM

Over 50% of women suffer from nausea in pregnancy. Hyperemesis gravidarum is the

inability to maintain hydration, resulting in dehydration and ketonuria as a result of

nausea and vomiting in pregnancy. It affects between 0.1% and 1% of women. Patients

become dehydrated and ketonuric, develop an electrolyte imbalance (hyponatraemia and

hypokalaemia), and in severe untreated cases a nutritional (thiamine) deficiency culmi-

nates in Wernicke’s encephalopathy.

Management of Early Pregnancy Complications

Harriet Pugsley, MB ChB, MRCOG; Judith Moore, MRCOG

OBSTETRICS i Peer reviewed



OBSTETRICS I Peer revIewed

Symptoms develop around 6–8 weeks of gesta-

tion and are directly related to levels of hCG, peak-

ing towards the end of the first trimester before

settling in the second trimester. Women present

with nausea, weakness, vomiting, and occasionally

ptyalism (inability to swallow saliva). On examina-

tion, there are signs of dehydration and tachycardia

with or without hypotension. Those who have suf-

fered with hyperemesis in a previous pregnancy are

more likely to develop similar symptoms in subse-

quent pregnancies. Nausea and vomiting present-

ing after 12 weeks of gestation is not hyperemesis

gravidarum.

The proposed pathophysiology behind hy-

peremesis is related to the hCG. The hCG molecule

has common alpha subunit with thyroid-stimulating

hormone and is thought to exert its effect via a

temporary physiological thyrotoxicosis; there may

be evidence of a raised free thyroxine and a low

thyroid-stimulating hormone. This accounts for the

timing of the onset and settling of symptoms in cor-

relation with levels of hCG. There may also be psy-

chological and cultural factors.

In all presentations, a multiple or molar preg-

nancy should be excluded as these conditions also

result in an increased hCG level.

Many women with mild symptoms are managed

in the community; once women are ketotic and un-

able to maintain hydration, admission is necessary.

For the majority of women, intravenous rehydration

with antiemetic medication, on a day case basis

without admission, is sufficient to break the cycle

and allow the patient to re-establish oral intake.

Many units have managed to expand the role of the

EPAU to incorporate this service and release capac-

ity on the admitting wards. However, more severe

cases do require an inpatient admission, further

rehydration, antiemetics, thiamine supplements,

daily electrolyte analysis, and a gradual resumption

of oral intake and monitoring of ketonuria. Intrave-

nous rehydration with normal saline or Hartmann’s

with potassium supplementation and monitoring of

electrolytes and ketonuria should be employed. In-

fusions containing dextrose should be avoided as

they may precipitate Wernicke’s encephalopathy.

Women requiring admission are intravascularly de-

hydrated posing an increased thrombotic risk, thus

thromboprophylaxis should be considered. Rarely,

intractable cases may require treatment with ster-

oids to relieve symptoms.

Hyperemesis is not uncommon; the majority of

cases can be successfully treated as a day admis-

sion to an EPAU. The outcome for the both preg-

nancy and patient are excellent.

MISCARRIAGE

Fifteen percent to 20% of pregnancies end in mis-

carriage.

Miscarriage was traditionally classified as

threatened, inevitable, complete, incomplete, or

missed. Most women present with pain and bleed-

ing in early pregnancy. Alternatively, miscarriage

may be diagnosed at the early dating scan with

no prior warning symptoms, this is classified as a

missed miscarriage.

Practice points I

Hyperemesis

• 50% of women suffer from nausea in pregnancy

• 0.1–1% of women suffer from hyperemesis gravidarum

• Symptoms peak at the end of the first trimester

• In severe cases, there is a risk of Wernicke’s encephalopathy re-

sulting from a nutritional imbalance of reduced thiamine

• Treatment includes intravenous rehydration, electrolyte monitor-

ing and restitution, antiemetics, thiamine supplementation and,

in non-resolving cases, steroid therapy

• Molar and multiple pregnancies should be excluded




An episode of pain or bleeding in an early

pregnancy subsequently demonstrated to be a vi-

able pregnancy is termed a threatened miscarriage.

Traditionally, classifications of complete, in-

complete and inevitable miscarriage have been

used. Diagnosis and management are based on

clinical findings: (1) examination of vaginal loss

for products of conception, and (2) examination

of the cervical, to determine whether it is open or

closed. More recently, there has been a tendency

to be increasingly conservative in management of

miscarriage with routine use of ultrasound to aid

management. Once an intrauterine pregnancy is

demonstrated by ultrasound, with uncertain viabil-

ity, a follow-up scan will be arranged 10–14 days

later to determine whether the pregnancy is ongo-

ing or not. If initial symptoms are of heavy bleeding,

or if they worsen between appointments, then med-

ical intervention in terms of a surgical evacuation

may be necessary at an earlier stage. Patients with

a pregnancy of unknown viability, who are waiting

for follow-up, must be informed of the risk of pain

and bleeding between appointments and be given

appropriate contact numbers and advised to return

if their bleeding becomes heavy.

If there are symptoms (or with minimal symp-

toms), treatment of miscarriage is based on three

alternatives: expectant, medical, or surgical. Ex-

pectant management employs awaiting the natural

course of events, for the products to pass sponta-

neously. Medical management involves combina-

tions of oral or vaginal prostaglandins to induce the

completion of miscarriage. Surgical management

involves an operation, usually vacuum aspiration,

to remove any remaining products of pregnancy.

There are risks and benefits associated with

all approaches:

1. Expectant management requires a longer fol-

low-up, often multiple visits, and is associated

with more prolonged and heavier bleeding.

Owing to the unpredictable of length of follow-

up and number of visits associated with expect-

ant management, many women veer away from

this option. There is a higher risk of treatment

failure and an increased need for later surgery.

This method of treatment may be more effec-

tive for those who are symptomatic of pregnan-

cy loss (ie, those presenting with an incomplete

miscarriage) because the process of spontane-

ous loss will have already begun. The timescale

for completion of expectant management can

extend over 2–6 weeks with the emphasis on

planned follow-up at specified intervals for

assessments usually with ultrasound scans

to ensure all products have passed. Although

figures differ depending on the length and type

of follow-up that patients have (clinical versus

ultrasound), around 80% complete without

intervention. When counselling women, the

possibility of a period of prolonged follow-up

and of the risk of incomplete loss with the

subsequent need for surgery should be stressed.

2. Medical management also aims to avoid

Practice points II

Miscarriage

• Affects up to 1 in 5 pregnancies

• Treatment options are expectant, medical, and surgical

• Expectant management aims to avoid surgery, may result in

prolonged follow-up with a risk of heavier bleeding and failed

treatment

• Medical management aims to avoid surgery, may be uncomfort-

able with heavier bleeding and risk of later surgery

• Surgical management allows early completion of treatment with

the risk of surgical and anaesthetic complications

• There is no difference in post-treatment conception rates regard-

less of method of management




surgery and is an accelerated method of

conservative management with a more predict-

able timescale of completion. It can be carried

out as an inpatient or outpatient procedure,

provided support is available for women at

home if needed. If managed in the community,

follow-up is necessary once again, to ensure

completion and that ongoing complications are

recognized and treated.

3. The benefit of surgical management is to

limit the need for prolonged follow-up,

reduce the number of symptomatic days, and

for early closure of treatment. Complications,

although infrequent, include uterine perforation,

cervical trauma, incomplete evacuation, the

risk of the anaesthetic, and a slightly higher

infection rate when compared with expectant

management.

The miscarriage treatment trial and subse-

quent Cochrane reviews (2006, 2010) have con-

cluded that there is no superior method of man-

agement and have recommended that the woman’s

preferences are taken into account when planning

care; treatment should therefore be patient-guided,

based on an informed decisions.

Importantly, the long-term follow-up of the

miscarriage treatment trial concluded that there

is no difference in later conception rates follow-

ing the different approaches to management. It has

also been suggested that empowering patients by

choice in their management reduces subsequent

anxiety and depression rates. Women who have

been involved in the decision-making process have

subsequently scored more favourably on quality-of-

life outcome questionnaires.

recurrent MiscarriageRecurrent miscarriage is defined as three or more

consecutive miscarriages and affects 1% of couples

trying to conceive.

It is recommended that investigations into

why women have miscarried should be implement-

ed following the third consecutive miscarriage.

For the majority of women, the results are normal.

Prognosis for the future worsens with increasing

numbers of pregnancy losses, advancing maternal

age, normal histopathology, and a normal parental

karyotype. Recurrent miscarriage is particularly

devastating for couples as they may be able to

conceive relatively easily but not carry an ongoing

pregnancy; they may thus need increased support

and counselling. EPAUs arrange specialist follow-

up and offer a patient educational role with the

provision of information combined with access to

support groups and counselling.

Following recurrent miscarriage, investiga-

tions may include screening for antiphospholipid

syndrome, karyotyping the products of conception,

diagnostic imaging for structural abnormalities, and

parental karyotyping looking for balanced translo-

cations. The incidence of controlled diabetes and

thyroid disease are no different in this population

when compared with that of the general public.

As stated, for the majority of patients, inves-

tigations will be normal and no cause is identified.

In a minority, there may be a treatable haemato-

logical factor such as antiphospholipid syndrome.

In the case of recurrent miscarriage, positive serum

Practice points III

Recurrent miscarriage

• Defined as three or more consecutive miscarriages

• Affects 1% of couples

• For the majority of patients, all investigations are normal

• A minority of patients will have antiphospholipid syndrome,

which can be treated with heparin and aspirin to improve the

chance of an ongoing pregnancy




blood results on two separate occasions 12 weeks

apart with a history of three consecutive pregnancy

losses before 10 weeks are necessary to confirm

antiphospholipid syndrome. (Positive bloods in as-

sociation with a single pregnancy loss above 10

weeks or a poor obstetric outcome before 34 weeks

secondary to placental disease also fulfil the crite-

ria for antiphospholipid syndrome.) For these wom-

en, treatment with aspirin and heparin will improve

the chance of an ongoing pregnancy. Without treat-

ment, the likelihood of a term pregnancy may be as

low as 10%; with treatment, the miscarriage rate

may be reduced by up to 54%.

MOLAR PREGNANCY

Molar pregnancies are rare, affecting roughly 1 in

700 conceptions. They are the result of an abnor-

mal conception and can be either complete moles

or partial moles. They are more common at the ex-

tremes of reproductive ability and among people

living in Asia. The risk of recurrence is low: 1–2%.

Complete moles are diploid, contain only pa-

ternal chromosomes, and are either the result fer-

tilization of an empty ovum with duplication of a

single sperm or dispermic fertilization. They contain

no fetal tissue.

Partial moles are triploid in nature, may con-

tain fetal parts, and are the result of dispermic fer-

tilization.

The concern with molar conceptions is the

risk of progression to neoplasia if left untreated.

Molar pregnancies can present as a miscarriage

or, more rarely, after a normal pregnancy. They are

associated with characteristic appearances on an

ultrasound scan (snow storm/placental vacuoles).

Molar pregnancies are usually detected in the

first trimester because women either present with

bleeding, resulting in an early ultrasound scan, or

are suspected on the dating ultrasound scan. They

are also associated with a raised hCG level, and the

condition should be excluded in women presenting

with hyperemesis.

Diagnosis is sometimes retrospective follow-

ing histological examination of products of concep-

tion removed following a surgical evacuation. In

cases of miscarriage, products should be screened

for gestational trophoblastic disease.

Once suspected, the majority are easily treat-

ed by surgical evacuation alone, with histological

confirmation. However, owing to the low but poten-

tially life-threatening risk of ongoing trophoblastic

tissue and neoplasia, registration and referral to

specialist tertiary care is crucial.

Due to the nature of hydatidiform gestational

trophoblastic disease/neoplasia (GTN) with the po-

tential for progressive disease, long-term follow-up

is necessary. This involves prolonged measurement

of urine hCG levels to ensure disease-free status,

avoidance of future pregnancy until hCG levels have

been normal for 6 months, and further monitoring

after all future pregnancies.

Treatment with surgical evacuation and pro-

longed monitoring of hCG levels are usually all that

is required, with repeat monitoring of levels in sub-

sequent pregnancies. All UK cases of gestational

Practice points IV

Molar pregnancy

• Affects 1 in 700 conceptions

• The concern is the risk of progression to neoplasia if unrecog-

nized or untreated

• All cases must be registered at a tertiary referral centre and re-

quire long-term follow-up

• The risk of recurrence is low (1–2%)

• Cases which have progressed to neoplasia are responsive to

treatment; 5–6% require chemotherapy with a 98% cure rate




trophoblastic disease are monitored via a national

registration programme consisting of three centres

based in Sheffield, Dundee, and Charring Cross.

hCG is a marker of ongoing trophoblastic dis-

ease. If levels are persistently elevated, further

treatment is required; this is usually medical with

methotrexate and, in rare cases, chemotherapy; a

repeat evacuation is not usually helpful. GTN is

extremely responsive to treatment. Chemotherapy

is necessary in only 5–6% of cases with a greater

than 98% cure rate and is more commonly required

following complete rather than partial molar preg-

nancies.

Following diagnosis and treatment, women

are advised to delay further conception until 6

months following normalization of the hCG levels

and a year after completion of chemotherapy. Bar-

rier contraception is advised; hormonal contracep-

tion is an option once hCG levels have normalized.

If contraception has been commenced before diag-

nosis, there may be a slight increase in the risk of

gestational trophoblastic disease progressing to

neoplasia with combined preparations, but there

is no evidence of detriment from single agent pro-

gestogens.

Thankfully, the recurrence risk is low, so wom-

en who continue to have a future pregnancy can be

reassured that 98 out of 100 conceptions will not be

a further molar pregnancy. In the women in whom a

second molar pregnancy does occur, it is usually of

the same histological type as the first.

ECTOPIC PREGNANCY

Ectopic pregnancy complicates approximately 1 in

50 pregnancies. Risk factors include previous ec-

topic pregnancy, a history of pelvic infection, or

past pelvic surgery. If not recognized or managed

inappropriately, it can become a life-threatening

condition as a result of invasion and rupture of a

blood vessel, resulting in massive pelvic haemor-

rhage; this can happen with or without rupture of

the Fallopian tube.

Presentation, as with most early pregnancy

complications, is with pain with or without vaginal

bleeding, or as collapse with associated concern-

ing symptoms of dizziness and fainting. Diagno-

sis is based on history, clinical examination, and

investigations, including transvaginal ultrasound

scan, serial hCG measurements, and laparoscopy.

In a normal pregnancy, the serum hCG level should

rise by at least 60% in 48 hours, although anything

above 30% can be consistent with normal early

pregnancy. The hCG levels in an ectopic pregnancy

may initially rise at a normal rate, but by the time

of presentation the rise is usually suboptimal. Once

suspected, ectopic pregnancies might be suggested

by ultrasound evidence of gestational tissue out-

side of the body of the uterus with an empty uter-

ine cavity. However, laparoscopy remains the gold

standard for confirmation of diagnosis and allows

concurrent treatment.

Although diagnosis can be suggested by pel-

vic ultrasound and hCG, confirmation can often

take several days because of presenting features

in common with other early pregnancy complica-

tions, uncertainty on ultrasound scan, and the need

Practice points V

Ectopic pregnancy

• Affects 1 in 50 pregnancies

• The main concern is the risk of life-threatening bleeding

• Management options depend largely on presentation and include

mainly surgery or methotrexate (rarely, expectant management

may be employed in the case of resolving trophoblast)

• Salpingotomy is only recommended during surgical treatment if

there is a concern that the contralateral tube is non-functional




for serial blood levels. In the stable patient in this

situation, the 48-hour hCG trend and symptoms are

crucial factors in determining timing and options

for treatment. In cases of haemodynamic insta-

bility secondary to a suspected bleeding ectopic

pregnancy, emergency surgery is necessary either

by laparotomy or laparoscopy, in order to stem the

bleeding and remove the ectopic as rapidly and

safely as possible.

Once diagnosed, there are several options for

the management of an ectopic pregnancy, depend-

ing mainly on presentation and individual circum-

stances. For example, in the acute emergency due

to a bleeding ectopic, urgent surgery is required

as a lifesaving procedure. However, the majority

of ectopic pregnancies are treated by scheduled

laparoscopic surgery or by medical management

with methotrexate. Conservative management is

also an option with resolving trophoblast (the trend

is for the hCG level to fall). Laparoscopic surgery

may be initially diagnostic to confirm the ectopic

pregnancy as the cause of pain and then therapeu-

tic by salpingectomy or salpingotomy as definitive

treatment.

Any surgery performed will have implications

for future fertility potential. At the time of surgery,

both Fallopian tubes should be assessed as the

health of the unaffected Fallopian tube is also rele-

vant. The surgical recommendation is for salpingec-

tomy to be performed if the contralateral tube is

deemed healthy and that salpingotomy should only

be performed if there is doubt about fertility with

the remaining tube. This is because preservation of

the Fallopian tube following an ectopic pregnancy

retains the risk of a subsequent ectopic pregnancy

in the same tube, whereas a salpingectomy removes

this risk. The fertility rate from a single functional

tube should be sufficient to allow conception.

If there is a diseased contralateral Fallopian

tube (such as a hydrosalpinx) and a salpingectomy

is performed, future options for conception are like-

ly to depend on in vitro fertilization. The limitation

of visual assessment of the tube should be remem-

bered as it will give an indication of gross disease

but not assess tubal patency for which a dye test is

more accurate.

Owing to the risk of life-threatening bleeding

from an ectopic pregnancy and the need to be ab-

solutely certain that a viable intrauterine pregnancy

has been excluded, there are rigid criteria to be met

before medical management becomes an option.

It is recommended that ultrasound findings

should show a suspected ectopic mass less than 3

cm in size with minimal free fluid and no visible

fetal heart. There should be minimal pain on ex-

amination and a suboptimal rise in hCG which is

initially less than 3,000 IU/L. (The Royal College of

Obstetricians and Gynaecologists disseminated na-

tional recommendation, from which individual units

For the majority of patients with recurrent miscarriage, investigations are normal and no cause is identified.




set their own treatment pathways). These criteria

are employed to enable patient selection so that

medical management is an option for appropriate

patients who are unlikely to have spontaneous

bleeding whilst undergoing treatment. Success of

treatment is assessed by subsequent hCG estima-

tions.

The main benefit of methotrexate is the avoid-

ance of surgery. The downside includes the low risk

of failure of treatment in cases in which ongoing

trophoblast results in bleeding from the ectopic

site and the need for surgical intervention.

Methotrexate has antifolate properties and is

teratogenic, so patients must be suitably advised

to delay conception following treatment, and care

must be taken to encourage folic acid prior to and

during subsequent conceptions.

Expectant management can be employed for

cases of resolving trophoblast, which may either be

a resolving ectopic pregnancy or an early pregnan-

cy miscarriage. For safety, levels of hCG must be

decreasing and women must be able to seek appro-

priate help and attend hospital easily should they

need to. Follow-up needs to be rigorous to ensure

that the hCG titres have returned to non-pregnant

levels and complications are not missed.

For women who have experienced an ectopic

pregnancy, fears for the future include the risk of

a recurrent ectopic pregnancy, concerns regarding

the ability to conceive again, and fear relating to

the health risks of a recurrent ectopic pregnancy.

It may be particularly worrying for women to plan a

further pregnancy if they have already experienced

life-threatening emergency surgery from a ruptured

ectopic or a previous pregnancy complication. If one

Fallopian tube is diseased, it is likely that the other

tube is also affected, and spontaneous conception

may be delayed or impossible. Therefore fertility

may well depend on in vitro fertilization which may

not be an option for many women (limited National

Health Service funding, personal finances, and

emotional reasons).

CONCLUSION

For the majority of women, conception, pregnancy,

and birth are straightforward, resulting in an uncom-

plicated confinement, delivery, and healthy baby.

For women who do experience complications,

explanation, advice, follow-up, and necessary in-

vestigations will help when planning for the future

and deciding whether to try for a further pregnancy.

Women who are informed of the long-term effects of

treatment, for example, that surgical/medical man-

agement following a miscarriage does not convey

differences in conception rates, and who are given

an informed choice regarding their management,

may find it easier to come to terms with their loss

and have less anxiety and depression long-term. It

must be remembered that each case is individual,

all circumstances are different and women need

Women who are informed

of the long-term effects

of treatment... and who are

given an informed choice

regarding their management,

may find it easier to come

to terms with their loss

and have less anxiety and

depression long-term.




FURTHER READING

Royal College of Obstetricians and Gynaecologists. The manage-ment of early pregnancy loss. Green-top Guideline No. 25. London: RCOG, 2006.

Trinder J, Brocklehurst P, Porter R, Read M, Vyas S, Smith L. Management of miscarriage: expectant, medical or surgical? Results of randomized controlled trial (miscarriage treatment (MIST) trial). BMJ 2006;332:1235.

Smith LFP, Ewings PD, Quinain C. Incidence of pregnancy after expectant, medical or surgical management of spontaneous first trimester miscarriage. Long term follow-up of miscar-riage treatment (MIST) randomized controlled trial. BMJ 2009;339:b3827.

Nanda K, Peloggia A, Grimes DA, Lopez LM, Nanda G. Expect-ant care versus surgical treatment for miscarriage. Cochrane Database Syst Rev 2006. Issue 2. Art. No.: CD003518. pub. 2.

Neilson JP, Gyte GML, Hickey M, Vaquez JL, Don L. Medical treatments for incomplete miscarriage (less than 24 weeks). Cochrane Database Syst Rev 2010. Issue 1. Art. No.: CD007223. pub. 2.

Neilson JP, Hiskey M, Vaquez JL. Medical treatment for early fetal death (less than 24 weeks). Cochrane Database Syst Rev 2006. Issue 3. Art. No.: CD002253. pub. 3.

Rai RS, Clifford K, Cohen H, Regan L. High prospective fetal loss rate in untreated pregnancies of women with recurrent miscarriage and antiphospholipid antibodies. Hum Reprod 1995;10:3301–3304.

Royal College of Obstetricians and Gynaecologists. The manage-ment of recurrrent miscarriage. Green-top Guideline No. 17. London: RCOG, 2011.

Royal College of Obstetricians and Gynaecologists. The manage-ment of gestational trophoblastic neoplasia. Green-top Guide-line No. 38. London: RCOG, 2010.

Royal College of Obstetricians and Gynaecologists. The manage-ment of tubal pregnancy. Green-top Guideline No. 21. London: RCOG, 2004.

© 2012 Elsevier Ltd. Initially published in Obstetrics, Gynaecol-ogy and Reproductive Medicine 2012;22(3):76–80.

About the AuthorsHarriet Pugsley is Specialist Registrar in Obstetrics and Gynae-cology. Judith Moore is Consultant Obstetrician and Gynaecolo-gist at Nottingham United Hospitals NHS Trust, Nottingham City Hospital, Nottingham, UK.

to be informed of the risks and benefits of all the

available treatment options.

The role of the EPAU, with open access and

support in future pregnancies in terms of regular

scanning and as a point of contact for the individual,

has been shown to improve the successful outcome

of future pregnancies following miscarriage. Open

access and choice in care options have also been

advocated as helping women to feel more in con-

trol of their treatment and reducing post-treatment

rates of anxiety and depression.

Besides the feeling of loss, it is normal for

women to feel guilt, blame, and anxiety. Women

will have concern for their own health, particularly

if they have required emergency treatment. Part-

ners may also harbour feelings of guilt, responsi-

bility, and helplessness for the situation, and may

have conflicting emotions with concerns regarding

the risk to their loved ones’ health in a future preg-

nancy, balanced with their desire to increase their

family.

It is important for patients who have had an

ectopic pregnancy to understand that they require

an early ultrasound in subsequent pregnancies to

exclude a recurrence and to seek a medical assess-

ment early in pregnancy.

Previously, in future conceptions, early open

access and regular specialist follow-up have been

available, enabling reassurance, investigations,

early detection of further complications, along with

support throughout subsequent pregnancies. Unfor-

tunately as healthcare services are put under ever

increasing pressure to reduce costs, open access

may no longer be deemed a financially viable ser-

vice.

It is important that appropriate investigations

are arranged and results collated and explained. It

is also necessary that patients are given appropri-

ate counselling, information, and links to support

groups should they require additional care.


Autism Spectrum Disorders Patricia Howlin, BA MSc PhD FBSP, Professor of Clinical Child Psychology

GYNAECOLOGY I Peer revIewed


GYNAECOLOGY i Peer reviewed

INTRODUCTION

The lifetime prevalence of ovarian cancer in the developed world is 1–2%. It is often

described as a silent killer; however, the majority of women frequently experience symp-

toms in the months leading to diagnosis. The majority of ovarian cancers are diagnosed

at an advanced stage. In England and Wales, ovarian cancer kills more women than all

of the other gynaecological malignancies combined.

Pregnancy, breastfeeding, and use of the oral contraceptive pill all appear to pro-

tect against the development of epithelial ovarian cancer; the lower incidence seen in

less developed countries may be related to a higher birth rate.

Patients with ovarian cancer are best managed by multidisciplinary teams. These

usually include nurse specialists, medical oncologists, histopathologists, radiologists,

palliative care specialists, and gynaecological oncologists, in collaboration with the

patients and their families.

TYPES OF OVARIAN CANCER

Primary ovarian tumour types include epithelial, sex cord–stromal and germ cell tu-

mours. Tumours not specific to the ovaries also occur, such as sarcomas and lymphomas.

Metastatic tumours from breast, stomach, and endometrial primaries are not uncommon.

epithelial TumoursMore than 80% of ovarian cancer is epithelial in origin. The most common subtype

is serous, accounting for about 50%, followed by endometrioid, mucinous, clear cell,

transitional (Brenner), mixed, and undifferentiated tumours. These have previously been

Ovarian Cancer: Current Management and Future Directions

Siân E Taylor, BSc, MB ChB, MRCOG, MD; John M Kirwan, MB ChB, MRCOG



GYNAECOLOGYGYNAECOLOGY I Peer revIewed

treated as essentially the same disease with dif-

fering histology; however, it is becoming increas-

ingly evident that they behave as distinct entities.

For example, endometrioid and clear cell cancers

are strongly linked with endometriosis, whilst many

mucinous cancers originate in the appendix and evi-

dence increasingly points to serous tumours arising

from dysplastic endometrium in the distal fallopian

tube. Response to chemotherapy varies; serous tu-

mours tend to be highly chemosensitive, but clear

cell and mucinous tumours are more resistant to

conventional chemotherapy.

Primary peritoneal cancer is histologically

indistinguishable from metastatic serous ovarian

cancer. It is diagnosed in the absence of any clear

ovarian primary. Treatment is the same as for ovar-

ian cancer, although, as there is often no mass to

debulk, chemotherapy is more often used as the

primary treatment.

Borderline ovarian tumours are not truly can-

cers but are termed ‘borderline’ because they show

histological features that are intermediate between

benign and malignant tumours. They are staged in

exactly the same way and sometimes spread be-

yond the ovary to produce non-invasive implants in

the omentum and the peritoneum. They can recur

after long periods; cases have been documented

with disease returning over 30 years after the initial

presentation. They are typically found in a younger

population compared with epithelial cancers, one-

third occur in women under the age of 40. The main

treatment is surgical excision, and opinions differ

in the extent of surgery required. It is probable that

they represent premalignant disease for low-grade

ovarian carcinomas.

Sex Cord–Stromal TumoursSex cord–stromal tumours account for approximate-

ly 7% of all malignant ovarian tumours. They arise

from a combination of the hormone-producing cells

of the ovary and stromal fibroblasts. Seventy per-

cent of all malignant sex cord–stromal tumours are

granulosa cell tumours. The majority occur in wom-

en in their sixth decade, although a small proportion

arises in young women and prepubertal girls. Gran-

ulosa cell tumours may secrete sex hormones; most

secrete oestrogen (although androgen-secreting va-

rieties do occur) potentially leading to endometrial

hyperplasia and carcinoma. Presenting symptoms

include abdominal distension, acute abdominal

pain, and abnormal vaginal bleeding. As most pre-

sent at an early stage, the prognosis is good. Treat-

The histological subtypes of epithelial ovarian cancer are distinct entities, requiring different treatment strategies.




ment is principally surgical with platinum-based

chemotherapy for advanced disease. Surgical treat-

ment is as for epithelial ovarian cancer, although in

young women with early disease, fertility preserva-

tion is an option. Other stromal tumours are rare

and include thecomas, fibromas, Sertoli-Leydig cell

tumours, and gynandroblastomas.

Malignant Germ Cell TumoursMalignant germ cell tumours occur chiefly in girls

and young women. The most common variety is the

dysgerminoma, the counterpart to the seminoma in

the male. Other types include the yolk sack tumour,

embryonal carcinoma, polyembryoma, non-gesta-

tional choriocarcinoma, and teratoma. They usually

present with abdominal pain, which is sometimes

acute, and a palpable pelvic mass. Treatment for

early-stage disease is surgical. As over 60% are

confined to one ovary at diagnosis, fertility-sparing

surgery is usual with unilateral salpingo-oophorec-

tomy or even ovarian cystectomy in selected cases

with otherwise normal ovaries. Dysgerminomas are

highly radiosensitive, but platinum-based chemo-

therapy is currently the preferred option as radio-

therapy usually results in premature ovarian failure.

Non-dysgerminoma tumours are treated with the

chemotherapy combination of bleomycin, etopo-

side, and cisplatin. Response rates are excellent

with cure rates approaching 100% in early-stage

disease and up to 75% in advanced disease.

FAMILIAL OVARIAN CANCER

Approximately 5–10% of all ovarian cancer is as-

sociated with a genetic predisposition. Individuals

carrying these gene defects have a significantly

higher risk of developing ovarian cancer than the

general population (Table 1).

A strong family history of breast and/or ovar-

ian cancer, especially at a relatively young age,

may indicate the presence of one of several pos-

sible BRCA1 or BRCA2 gene mutations. The gene

products are involved in DNA repair. These gene

mutations are particularly common amongst the

Ashkenazi Jewish population. Men carrying these

genes are at increased risk of pancreatic cancer

as well as male breast cancer. Endometrial, colon,

ovarian, and other cancers cluster in families with

the Lynch II or hereditary non-polyposis colorectal

cancer (HNPCC) syndrome. This is caused by a vari-

ety of defects in the DNA mismatch repair system.

BRCA1/2 and HNPCC are inherited in an autosomal

dominant fashion; further mutation or epigenetic

silencing of the second allele results in malignancy.

Prophylactic risk-reducing surgery is recom-

mended for BRCA1 and BRCA2 carriers when they

reach the age of 35 or have completed their fami-

lies. This usually consists of a laparoscopic bilateral

salpingo-oophorectomy. The risk of ovarian cancer

under 35 years in BRCA patients is low, and remov-

al of the ovaries has the dual advantage of reducing

the risks of both ovarian and breast cancer. HNPCC

carriers are offered hysterectomy and bilateral sal-

pingo-oophorectomy when their family is complete.

Until patients are ready to undergo surgery, annual

screening is commonly offered with cancer antigen

125 (CA 125) and transvaginal ultrasound; however,

Table 1. Approximate lifetime percentage risk of developing ovarian cancer

General populationOne first-degree relative affected under 55 yearsOne first-degree relative affected over 55 yearsTwo first-degree relatives affectedBRCA1 carrierBRCA2 carrierHNPCC carrier

1.6%5.2%3.4% 7%

28–44%27%12%

HNPCC = hereditary non-polyposis colorectal cancer.




there is no evidence that this is effective. Even if

the ovaries are removed, there is a small continuing

risk of developing primary peritoneal cancer.

The lack of one of these specific conditions

does not exclude an increased cancer risk, as not all

inherited ovarian cancer syndromes have had their

genetic origin fully characterized.

BIOMARKERS AND SCREENING

The most commonly used marker for ovarian cancer

is CA 125. This is a glycoprotein that is released

into the bloodstream by any condition that disturbs

the peritoneum, including any peritoneal cancer,

cirrhosis, congestive cardiac failure, endometrio-

sis, and pelvic inflammatory disease. Pregnancy

also causes a variable increase in serum levels. It

is therefore notoriously non-specific at low levels.

However, serous ovarian cancer can cause dramatic

increases in CA 125. This is only likely to occur

when the disease has already spread beyond the

ovary. Other tumour subtypes, especially mucinous,

often produce a more modest elevation.

Other markers are also used. Germ cell ovarian

cancers often secrete highly specific tumour mark-

ers which are useful in diagnosis and monitoring,

eg, α-fetoprotein, β-hCG, and lactate dehydroge-

nase. These should be tested, in addition to CA

125, in women under the age of 40 years with a

suspicious pelvic mass. In addition, inhibin may be

of use as a marker for mucinous and granulosa cell

tumours.

Novel markers are currently under development

both as panels and individually. Human epididymis

protein 4 has been suggested as a biomarker with

equal or greater sensitivity and specificity than CA

125, however clinical trials are awaited.

The PLCO (prostate, lung, colon, ovary) trial,

a huge US-based randomized controlled trial of

cancer screening, has recently reported. It con-

Screening the anxious low-risk patient for ovarian cancer may not be beneficial because of the risk of false-positive result.




cluded that screening the general population with

annual CA 125 and transvaginal ultrasound does

not reduce ovarian cancer mortality. Indeed, a sig-

nificant number of women had major complications

from surgery performed because of a false-positive

screening test.

There are also two large UK-based multicentre

trials investigating ovarian cancer screening. The

first is in postmenopausal women without a signifi-

cant family history of ovarian cancer and also uses a

combination of transvaginal ultrasound and CA 125

(UK Collaborative Trial of Ovarian Cancer Screen-

ing [UKCTOCS]). The second is in women with a

significant family history and is testing a panel of

biomarkers in addition to CA 125 and transvaginal

ultrasound (UK Familial Ovarian Cancer Screening

Study). Both have now finished recruitment and are

due to report in the next few years.

Screening with a combination of CA 125, ul-

trasound, and pelvic examination is commonly per-

formed for anxious patients who desire screening

for ovarian cancer. If the findings of the PLCO trial

are duplicated in the UKCTOCS trial, this practice is

likely to become hard to defend in low-risk women.

INVESTIGATION

The first symptoms of ovarian cancer usually

emerge some time before diagnosis. These com-

monly include early satiety, changes in bowel

habit, bloating, urinary frequency, and pelvic and

abdominal pain. Patients with advanced cancer of-

ten complain of abdominal swelling and discomfort

due to ascites with or without a large abdomin-

opelvic mass. Eating is often difficult, and patients

may notice weight loss, apart from the distended

abdomen. It is not uncommon for patients to pre-

sent with a swollen leg secondary to a deep vein

thrombosis. However, most small ovarian cancers

are asymptomatic when confined to the ovaries and

therefore difficult to detect.

Recent National Institute for Clinical Excel-

lence (NICE) guideline recommend that women,

especially those over 50 years old, who experience

symptoms persistent or frequent symptoms as de-

scribed above, or new-onset symptoms suggestive

of irritable bowel syndrome, should have CA 125

Table 2. FIGO staging of ovarian cancer

Stage Description

I Confined to ovaries

Ia One ovary, no ascites present containing malignant cells, no tumour on external surface, capsule intact

Ib Both ovaries, no ascites present containing malignant cells, no tumour on external surfaces, capsule intact

Ic Tumour limited to one or both ovaries with any of the following: tumour on the surface on one or both ovaries, capsule ruptured, ascites present with malig-nant cells or positive peritoneal washings

II Growth involving one or both ovaries with pelvic exten-sion

IIa Extension and/or metastases to uterus and/or fallopian tubes

IIb Extension to other pelvic tissues

IIc Tumour stage IIa or IIb but with tumour on surface of one or both ovaries, capsule ruptured, ascites pres-ent containing malignant cells or positive peritoneal washings

III Tumour involving one or both ovaries with microscopi-cally confirmed peritoneal implants outside the pelvis and/or regional lymph node metastasis

IIIa Microscopic peritoneal metastasis beyond the pelvis

IIIb Macroscopic peritoneal metastasis beyond the pelvis, 2 cm or less in greatest dimension

IIIc Abdominal implants greater than 2 cm in diameter and/or regional lymph nodes metastasis

IV Distant metastasis beyond the peritoneal cavity. Includes liver parenchymal metastasis and/or pleural effusion with positive cytology

FIGO = International Federation of Gynecology and Obstetrics.




measured. If the level is > 35, an ultrasound of the

abdomen and pelvis should be performed.

The ultrasound and CA 125 together are used

to calculate the risk of malignancy index (RMI): RMI

= U × M × CA 125, where U = ultrasound score (1

point for each of multilocular cysts, solid areas,

ascites, bilateral lesions, metastases. U = 0 for 0

points, U = 1 for 1 point, U = 3 for 2–5 points); M =

menopausal status (premenopausal = 1, postmeno-

pausal = 3); CA 125 = serum CA 125 level.

A score of 250 has been chosen by NICE to

guide triage to either surgery in a cancer centre un-

der the care of a specialist multidisciplinary team,

including subspecialist gynaecological oncologists

(≥ 250), or to care under a general gynaecologist

with an interest in gynaecological oncology in a

cancer unit (< 250).

If ovarian cancer is suspected, a computed to-

mography of the abdomen and pelvis (and thorax if

clinically indicated) should be performed, prior to

surgery, to assess the extent of the disease.

STAGING

Staging is performed to guide treatment and to

provide information on prognosis. Traditionally, this

has been achieved by performing a staging lapa-

rotomy. Information can also be gleaned from ra-

diological investigations, guided biopsy, and cytol-

ogy of ascitic or pleural fluid. The current staging

system was devised by International Federation of

Gynecology and Obstetrics (FIGO) (see Table 2).

TREATMENT

Treatment in ovarian cancer depends upon the stage

at presentation and the histological subtype. Non-

epithelial cancers are discussed earlier. In general,

epithelial ovarian cancers are treated with a combi-

nation of surgery and chemotherapy. Except in very

early disease, treatment is rarely curative but it can

provide symptom relief and prolong life.

Traditionally, ovarian cancer is treated with a

staging and debulking laparotomy followed by six

cycles of chemotherapy. Second-look laparotomy,

where a second surgical debulking procedure is

performed after completion of chemotherapy, is not

beneficial. Data from the EORTC 55971 trial, which

compared interval debulking surgery performed

midway between six cycles of chemotherapy with

the traditional laparotomy and six cycles of post-

operative chemotherapy, suggested that neoadju-

vant chemotherapy in bulky stage IIIc/IV disease

did not adversely affect prognosis and that interval

debulking is associated with a lower post-operative

morbidity and mortality. The CHORUS (CHemother-

apy OR Upfront Surgery) trial also addresses this

question and its publication is awaited.

In light of this, interval debulking surgery, per-

Advanced epithelial ovarian cancer is difficult to cure.




formed midway through chemotherapy, is gaining

popularity.

SURGERY

Surgical treatment, whether primary or as an inter-

val debulking procedure, involves a midline laparot-

omy, sampling of ascitic fluid or peritoneal wash-

ings for cytology, full assessment by palpation of

all peritoneal surfaces and biopsy of any suspicious

areas, removal intact of any encapsulated masses

or debulking of tumour, sampling of suspicious

pelvic and para-aortic lymph nodes, omentectomy,

total abdominal hysterectomy, and bilateral salpin-

go-oophorectomy. The aim is to completely remove

all visible disease. This is thought to promote an

optimum response to chemotherapy. Surgery also

provides ample material for diagnostic histological

assessment. The evidence for debulking improving

chemosensitivity is not absolute. Many argue that

the ability to perform optimal cytoreduction is a

more a reflection of favourable tumour biology with

an intrinsically better prognosis than the surgery

itself influencing outcome. However, it has been

demonstrated that survival in stage III disease is

improved by the primary surgery being performed by

a specialist gynaecological oncologist, rather than

an obstetrician/gynaecologist or a general surgeon.

Surgical aggressiveness varies considerably

between continents, countries, and individual units.

Some studies have shown improvements in survival

with radical surgery, including liver resection, bowel

resection, and splenectomy. Published case series

show small or no increases in mortality and morbid-

ity. However, these studies usually involve highly se-

lected cases from single institutions and may not be

directly applicable to wider practice. Some disease

remains unresectable to even the most adventurous

surgeon; this includes small bowel mesenteric dis-

ease and disease of the portal triad.

Recent NICE guidance recommends that lym-

phadenectomy is restricted to removal of clinically

involved nodes. Full pelvic and para-aortic node dis-

section in suspected stage I disease is not recom-

mended.

Less aggressive surgery is preferred in early

epithelial ovarian cancer in young women who wish

to preserve their fertility. About 8% of stage I epi-

thelial ovarian cancers occur in women under the

age of 35. A proportion of these will not have com-

pleted childbearing and may wish to consider fertil-

ity-sparing surgery. Suitable patients include those

with stage IA, grade 1 or possibly grade 2 disease.

Such conservative surgery would typically consist

of peritoneal fluid cytology, unilateral salpingo-

oophorectomy, omental biopsy, and careful inspec-

tion of the contralateral ovary and nodal chain. One

case series describes 282 women treated conserva-

tively for epithelial ovarian cancer. Just over 30%

The decision to adopt palliation in patients with ovarian cancer can be challenging.

The decision to adopt palliation in patients with ovarian cancer can be challenging.




subsequently went on to have term deliveries. Four

percent died of conditions related to their disease.

Surgery also plays a role in palliative care of

patients with ovarian cancer. Bowel obstruction is

common in the end stages of the disease; post-mor-

tem studies of women dying with ovarian cancer

revealed bowel obstruction in almost 50%. Before

contemplating such surgery, consideration must be

given as to whether it is appropriate. Surgery should

only be performed if it has a reasonable chance of

success, and risks need to be carefully balanced

against potential symptom relief. Contraindications

to surgery include patient refusal, rapidly accumu-

lating acsites, high obstruction, multiple levels of

obstruction, and poor nutritional status. Decisions

involving palliative surgery should involve the mul-

tidisciplinary team and careful discussion with the

patient and her relatives.

CHEMOTHERAPY

Patients should have histological confirmation of

their ovarian cancer prior to starting chemotherapy.

This may be obtained through image-guided percu-

taneous biopsy, or where this is not possible or the

results are inadequate, by laparoscopic biopsy.

The current standard first-line chemotherapy

regimen for ovarian cancer involves intravenous ad-

ministration of a platinum-based drug with a taxa-

ne, usually paclitaxel, given 3 weekly for six cycles.

Most units prefer carboplatin to cisplatin as it has

a less toxic side effect profile. Evidence for the use

of paclitaxel is drawn from its efficacy in relapsed

ovarian cancer. However, paclitaxel significantly in-

creases the risk of neuropathy when compared with

carboplatin alone, and some patients have anaphy-

lactic reactions to taxanes. For these reasons, it

is not universally used. Chemotherapy can also be

given via the intraperitoneal route; however, cur-

rently in the UK its use is not recommended outside

of clinical trials.

Epithelial ovarian cancer is one of the more

chemosensitive solid tumours, and complete clini-

cal and radiological response occurs in up to 50%

with the above regimen. Conversely, 20–30% will

show no evidence of response. Unfortunately, the

majority of patients with advanced ovarian cancer

will relapse. Chemotherapy for recurrent disease is

determined in part by the length of time before re-

lapse occurs. If it is more than 6 months afterwards,

it is potentially platinum-sensitive and, unless con-

traindicated, a regimen containing platinum will be

used again. A taxane is likely to be included, espe-

cially if not used initially. Paclitaxel is sometimes

used at a lower dose at more frequent intervals;

this appears to reduce the adverse effects experi-

enced. Response rates are in the order of 30%. If

relapse occurs within 6 months, second-line drugs,

such as liposomal doxorubicin and topotecan, may

be considered. As response rates are low, approxi-

mately 10–20%, the choice of drug is made bearing

in mind side effect profiles and ease of administra-

tion.

Novel chemotherapeutic agents are constantly

being developed, some of the most promising effect

the functioning of vascular endothelial growth fac-

Practice points

• Ovarian cancer is best managed within a cancer centre by a multidisciplinary team

• Women with symptoms suggestive of ovarian cancer should have a CA 125 performed as an initial screen, followed by ultrasound if abnormal

• Ovarian cancer screening with ultrasound and CA 125 does not appear to be helpful in the low-risk population

• Treatment involves a combination of surgery and platinum-based chemotherapy

• Fertility-sparing surgery is possible in women with early-stage disease




tor. Interim analysis of the ICON 7 trial of standard

therapy with or without bevacizumab (a monoclo-

nal antibody to vascular endothelial growth factor)

shows a sustained improvement in progression-free

survival in a subgroup of women with advanced dis-

ease and suboptimal surgical debulking.

Hormonal therapies are occasionally used;

these probably act by reducing oestrogen activity

and include tamoxifen, aromatase inhibitors, and

gonadotropin-releasing hormone analogues. Re-

sponse rates of 10–15% have been achieved in re-

lapsed disease. Their main advantage is their mini-

mal side effects when compared with conventional

chemotherapy.

PALLIATIVE CARE

Palliation is an integral part of the care of pa-

tients with ovarian cancer. Maintaining a balance

between optimism and pragmatism, together with

knowing when the emphasis of care should tilt to-

wards palliation, are some of the most challenging

aspects of caring for this group of patients.

Almost all patients with advanced ovarian

cancer, whether in the terminal phase or not, will

have distressing symptoms that require treatment.

These symptoms may be due to the disease itself

or secondary to their treatment. Common symptoms

include nausea, pain, loss of appetite, constipation,

and abdominal distension. This is too small a space

to describe all the possible treatment strategies

involved, but it is worth emphasizing a few princi-

ples. It is important to take a history, examine the

patient, and review the drug chart and case notes.

Then consider what investigations may be helpful

and what drug or intervention will best treat the

most likely cause of her symptoms. These patients

are usually complex with multiple problems; how-

ever, a logical and systematic approach will help to

identify the best treatment. Subcutaneous infusions

delivered by syringe driver are useful as they permit

a steady concentration of drug and oral medications

may be poorly absorbed. Several hospices publish

their guidelines for managing symptoms in pallia-

tive care on the internet, which can be very helpful.

Finally, the physical needs of a seriously ill pa-

tient are only one facet of their care. Social, spiritu-

al, and emotional needs also need to be addressed,

both for the patient and their relatives. Specialist

oncology nurses and palliative care input is essen-

tial, as are discussions on resuscitation status and

preferred place of death. These aspects of care are

easily overlooked but can make the difference be-

tween a peaceful and a difficult death.

FURTHER READING

Agarwal R, Linch M, Kaye SB. Novel therapeutic agents in ovarian cancer. EJSO 2006;32:875–886.

Aletti GD, Gallenberg MM, Cliby WA, Jatoi AJ, Hartmann LC. Current management strategies for ovarian cancer. Mayo Clin Proc 2007;82:751–770.

Blagden S, Gabra H. Future directions in the management of epithelial ovarian cancer. Future Oncol 2008;4:403–411.

Cannistra SA. Cancer of the ovary. N Engl J Med 2004;351:2519–2529.Colombo N, Parma G, Zanagnolo V, Insinga A. Management of ovarian

stromal cell tumours. J Clin Oncol 2007;25:2944–2951.Gershenson DM. Management of ovarian germ cell tumours. J Clin

Oncol 2007;25:2938–2943.Goff BA, Mandel L, Muntz HG, Melancon CH. Ovarian carcinoma

diagnosis, results of a national ovarian cancer survey. Cancer 2000;89:2068–2075.

Markman M. New, expanded, and modified use of approved antineo-plastic agents in ovarian cancer. The Oncologist 2007;12:186–190.

NICE clinical guideline 122, Ovarian Cancer.Nickles Fader A, Rose PG. Role of surgery in ovarian carcinoma. J Clin

Oncol 2007;25:2873–2883.Rao G, Crispens M, Rothenberg ML. Intraperitoneal chemotherapy

for ovarian cancer: overview and perspective. J Clin Oncol 2007;25:2867–2872.

© 2011 Elsevier Ltd. Initially published in Obstetrics, Gynaecology

and Reproductive Medicine 2011;22(2):34–37.

About the AuthorsSiân E Taylor is Subspecialty Trainee in the Department of Gynae-cologial Oncology at Liverpool Women’s Hospital, Liverpool, UK. John M Kirwan is Consultant Gynaecological Oncologist at Liverpool Women’s Hospital, Liverpool, UK.


MIMS.com now comes with a brand new look and mobile app!Along with a redesign that helps enhance ease of access to crucial clinical information, MIMS.com can now be accessed through a FREE mobile application, available on the iTunes store.

With MIMS available across multiple platforms and devices, you can now easily and conveniently find the most up-to-date and relevant drug information anytime, anywhere.

CONNECT WITH MIMS:

www.mims.com MIMS mobile/tablet app facebook.com/mimscom

Journal of Paediatrics, Obstetrics & Gynaecology

MIMScomNew206x276_0613.indd 1 6/4/13 3:45 PM





Constipation in Infants and Children

Taya Dowling, BMedSci, MB BS; Scott Nightingale, BMed(Hons), MClinEpid, FRACP

REMEMBER

• Constipation is a common paediatric presentation.

• Constipation is defined based on the frequency of stooling (which varies widely

depending on the age of the child), and more importantly the consistency, size and

difficulty with which stools are passed (see the box on page 165).1

• Constipation arising beyond the neonatal period is most often functional (ie, does not

result from any identifiable organic pathology), and may be perpetuated by voluntary

withholding of stool to avoid painful defaecation.

• There are several ‘red flags’ that should prompt further investigation for a contrib-

uting medical or surgical condition, but these are uncommon (see the box on page

165).2,3

• Faecal incontinence is thought to be related to over distension of the rectum with

stool, shortening of the anal canal and resulting impairment of normal continence

mechanisms. Incontinence is not deliberate or caused by laziness on the part of the

child.

• Untreated constipation and faecal incontinence (encopresis) can have a significant

psychosocial impact on a child.

• Management of constipation is often a long- term process that requires the com-

plementary approaches of careful education of the child and parents, behavioural

techniques, laxative agents and review.4

ASSESSMENT

• The focus should be on identifying the rare child with an organic cause for consti-





pation, and determining whether the child has

faecal impaction.

• The history should include a detailed descrip-

tion of the child’s stool, stool frequency, incon-

tinence, withholding behaviour and any symp-

toms associated with defaecation, such as pain,

bleeding and straining.

• Important aspects of the history include age at

onset, growth trends, diet history and the pres-

ence of ‘red flags’ (see the box on this page).

• A thorough physical examination should be per-

formed, particularly focusing on growth param-

eters, palpable abdominal faecal masses, in-

spection of the perianal and lumbosacral regions

and lower limb neurological examination. Poor

growth may occur with Hirsch sprung disease,

hypothyroidism and coeliac disease.

• Digital rectal examination should be avoided in

primary health care as it rarely contributes to the

clinical assessment and can be particularly dis-

tressing for the child.

• Impaction is suggested by faecal incontinence or

a palpable faecal mass (preferably determined

via abdominal palpation).

• If the likely diagnosis is functional constipation

then no further invest igation is needed. Abdomi-

nal X- rays are not needed to diagnose constipa-

tion or to determine response to therapy.5

• If a pathological cause for constipa tion is sus-

pected then appropriate investigations should be

performed in consultation with a paediatrician or

paediatric surgeon.

MANAGEMENT

• A combination of management approaches that

complement each other are almost always re-

quired in the management of childhood consti-

pation. Individual elements in isolation (eg, di-

simpaction without maintenance laxatives) are

unlikely to be unsuccessful.

• Education of parents and caregivers about the

relationship between behavioural aspects (eg,

fear of pain and withholding) and functional

‘Red flag’ feature2,3

History• Constipation from the neonatal period• Failure to pass meconium by the age of 48 hours• ‘Ribbon stools’ suggesting anorectal stricture or stenosis• Abdominal distension and vomiting• Poor weight gain or weight loss• Leg weakness or delayed gross motor development

Examination• Gross abdominal distension• Abnormal appearance, position or patency of anus – fistulae, bruis-

ing, multiple fissures or fissures away from the midline, tight or patulous anus, anteriorly placed anus, absent anal wink

• Lumbosacral abnormalities – evidence of sacral agenesis, discol-oured skin, naevi, sinus, hairy patch, lipoma, central pit, scoliosis

• Gluteal asymmetry or wasting• Absent cremasteric reflex• Abnormal results on lower limb neurological examination – de-

formity such as talipes, abnormal reflexes

Rome III diagnostic criteria for functional constipation in children1

At least two of the following features have been present for at least 2 months in a child aged 4 years or older (developmental age):• two or fewer defaecations in the toilet per week• at least one episode of faecal incontinence per week• history of retentive posturing or excessive volitional stool retention• history of painful or hard bowel movements• presence of a large faecal mass in the rectum• history of large diameter stools that obstruct the toilet




constipation is vital. They should be informed

that this is usually a chronic problem, requiring

long-term management. The rationale behind the

various aspects of management should be made

clear, and education should be reviewed on sub-

sequent visits.

• Faecal disimpaction, if necessary, should be

achieved using laxatives (see the Table). Except

in infants, oral therapy should generally be tried

first. In children, rectal therapy should usually

be reserved for those with more severe or un-

responsive constipation, leading to persistent

rectal discomfort or unproductive straining. Oral

macrogol 3350 (polyethylene glycol) was found

to be equally effective to enema therapy for dis-

impaction in a randomized study of 90 children.6

If multiple enemas or nasogastric lavage are re-

quired for disimpaction then a paediatrician or

paediatric surgeon should be involved.

• Maintenance laxative therapy should be started

after disimpaction and often needs to continue

for many months after normalization of stool-

ing. A plan should be made to restart laxatives

promptly on signs of relapse.

• The objectives of maintenance therapy are that:

– the child passes regular soft stools (eg, 1 to

2 per day) without discomfort or excessive ef-

fort

– the rectum remains empty to prevent re-

impaction.

• There are many options available for mainte-

nance laxative therapy (see the Table), with

choice influenced by the age of the child, pre-

vious experience, ease of administration and

palatability to the child:

– Liquid paraffin should be avoided in infants

and those at risk of aspiration because of the

risk of lipoid pneumonia, and should not be

given within 2 hours of sleep.

– In children, the osmotic laxative macrogol

Parents and caregivers are to be educated to encourage the child with faecal incontinence develop a healthy stooling habit.




was found to result in more stools per week

and less need for additional therapy when

compared with lactulose in a meta -analysis of

four studies.7 Macrogol is not currently recom-

mended for children younger than 2 years in

Australia because of lack of safety data; how-

ever, these data are accumulating.

– Stimulant laxatives such as bisacodyl, sen-

nosides and sodium picosulfate are effective

adjuncts to osmotic laxatives when necessary.

• There are few data to support widely held con-

cerns regarding long- term use of laxatives, in

particular stimulant laxatives. Clinical studies

show that long- term use of osmotic laxatives is

safe and well tolerated.7 Caregivers should be

educated about this safety and the need for long-

term maintenance therapy.

• Behavioural measures combined with laxative

therapy are superior to either therapy alone in

children with faecal incontinence.8 It is impor-

tant that behavioural measures are applied con-

sistently with the aim of developing a healthy

stooling habit. The child should be encouraged to

sit on the toilet for 5 minutes, two or three times

a day, within 30 minutes after meals to take ad-

vantage of the gastrocolic reflex. Positive rein-

forcement for sitting (eg, using reward charts)

should be encouraged. Children should never be

punished for being constipated or incontinent.

• A healthy diet should be encouraged rather than

a high- fibre diet, as there is little evidence that

increasing dietary fibre is an effective treat-

ment for childhood constipation.2 There is lim-

ited evidence that avoiding cow milk may result

in improvement in some children with chronic

constipation, particularly those with atopic ten-

dencies.9,10 Any trial of such an elimination diet

should be limited to 2 to 4 weeks, and the child

should be re- challenged to confirm any effect.

Prolonged elimination diets require supervision

by a qualified dietitian to prevent any nutritional

deficiencies and there should be repeated at-

tempts to normalize the diet.

• Aside from situations of clinical dehydration,

there is no evidence that increasing water intake

is beneficial in constipation. However, ensuring

adequate fluid intake is important when using

osmotic laxatives to avoid dehydration.

• Regular review is required to monitor response

to therapy, adjust laxative dose, reinforce educa-

Table. Initial laxative options for infants and children

Infants (< 12 months) Children

Disimpaction • Glycerol suppository 700 mg • Macrogol 3350 1 to 1.5 g/kg/day (max 52 g for children aged 2 to 5 years and 78 g for children aged 6 years and over) until disimpacted4*

Maintenance† • Sorbitol-containing fruit juices, such as prune, pear, apple (50 to 100 mL/day) or

• Lactulose 5 mL daily

• Macrogol 3350 0.4 to 0.8 g/kg daily up to 17 g* or• Paraffin oil 1 to 3 mL/kg daily (start 10 to 20 mL

daily)‡ or• Lactulose 10 to 15 mL daily

* Macrogol 3350 (polyethylene glycol) is available in Australia in a variety of formulations, in scoop packs or sachets. † Doses are a guide only and should be titrated to effect every 2 to 3 days as required, with consideration of the maximum recommended dose.‡ Not recommended in infants, or in children with gastro-oesophageal reflux or risk of aspiration.




REFERENCES

tion and support the family through what is often

a long and frustrating period.

• Lack of response to management should prompt

review of all aspects of the management plan.

Persisting failure to respond may prompt recon-

sideration of pathological causes and investiga-

tion for these.

CONCLUSION

• Constipation is a common paediatric presenta-

tion and is usually functional.

• Further investigation or referral is guided by the

presence of ‘red flags’.

• Untreated constipation and faecal incontinence

(encopresis) can have a significant psychosocial

impact on a child.

• Management of constipation is often a long-

term process that requires the complementary

approaches of careful education of the child and

parents, behavioural techniques, laxative agents

and review.

© 2013 Medicine Today Pty Ltd. Initially published in Medicine Today July 2013;14(7):71–73. Reprinted with permission.

About the AuthorsDr Dowling is a Paediatric Registrar, Hunter New England Local Health

District, Newcastle. Dr Nightingale is a Paediatric Gastroenterolo-

gist, John Hunter Children’s Hospital; and Conjoint Lecturer, School

of Medicine and Public Health, University of Newcastle, Newcastle,

NSW, Australia.

Series Editor: Associate Professor Simone Strasser, MD, FRACP, Clini-

cal Associate Professor, Central Clinical School (Medicine), University

of Sydney; and Senior Staff Specialist, AW Morrow Gastroenterol-

ogy and Liver Centre, Royal Prince Alfred Hospital, Sydney, NSW,

Australia.

1. Drossman DA. Rome III: the functional gastro-intestinal disorders. 3rd ed. McLean, VA: Degnon Associates; 2006.

2. National Institute for Health and Care Excellence (NICE). Constipation in children and young people: diagnosis and management of idiopathic childhood constipation in primary and secondary care: quick reference guide. London: NICE; 2010.

3. Constipation Guideline Committee of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. Evaluation and treatment of constipation in infants and children: recommen-

dations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pedi-atr Gastroenterol Nutr 2006;43:e1-e13.

4. North American Society for Pediatric Gastroen-terology, Hepatology and Nutrition. Evaluation and treatment of constipation in children: summary of updated recommendations of the North American Society for Pediatric Gastroenterology, Hepatol-ogy and Nutrition. J Pediatr Gastroenterol Nutr 2006;43:405-407.

5. Berger MY, Tabbers MM, Kurver MJ, Boluyt N, Benninga MA. Value of abdominal radiography, co-

lonic transit time, and rectal ultrasound scanning in the diagnosis of idiopathic constipation in children: a systematic review. J Pediatr 2012;161:44-50, e41-42.

6. Bekkali NL, van den Berg MM , Dijkgraaf MG, et al. Rectal fecal impaction treatment in childhood constipation: enemas versus high doses oral PEG. Pediatrics 2009;124:e1108-E1115.

7. Gordon M, Naidoo K, Akobeng AK, Thomas AG. Osmotic and stimulant laxatives for the manage-ment of childhood constipation. Cochrane Database Syst Rev 2012;7:CD009118.

8. Brazzelli M, Griffiths PV, Cody JD, Tappin D. Be-havioural and cognitive interventions with or with-out other treatments for the management of faecal incontinence in children. Cochrane Database Syst Rev 2011;12:CD002240.

9. Iacono G, Cavataio F, Montalto G, et al. Intoler-ance of cow’s milk and chronic constipation in chil-dren. N Engl J Med 1998;339:1100-1104.

10. Irastorza I, Ibanez B, Delgado-Sanzonetti L, Maruri N, Vitoria JC. Cow’s milk–free diet as a therapeutic option in childhood chronic constipa-tion. J Pediatr Gastroenterol Nutr 2010;51:171-176.




Hormonal Contraception and Cancers Wong Yuen Kwan Alice, MBBS, FRCOG, FHKAM(O&G), FHKCOG, Cert HKCOG(Reprod Med)

Oral contraceptive use puts women at risk for certain carcinomas.

INTRODUCTION

The use of hormones has provided great

convenience to a woman’s life, be it for

therapeutic use or as a lifestyle drug, ie,

contraceptives. However, the duration of

hormone use is frequently long, in terms

of years. Concerns have been raised about

the possibility of a relationship between

cancer development and long-term hormo-

nal influence. This article reviews the evi-

dence regarding the relationship between

hormonal contraceptive use and the devel-

opment of cancer, with the discussion fo-

cusing mainly on carcinoma of the breast

and female genital tract.

HISTORICAL EVIDENCE

The first report dated back to 1972 when

combined oral contraceptive pills (COC)

containing mestranol and norethynodrel

appeared to cause a case of metastatic

breast cancer in a female rhesus monkey.1

Soon after, there were further similar re-

ports of development of breast cancer in

beagles and rodents after exposure to hor-

mones contained in today’s COC.2–4

In June 2005, the International Agen-

cy for Research on Cancer (IARC) Working

Group of the World Health Organization

(WHO) met in Lyon, France, and classified

combined oral contraceptives and com-

bined oestrogen-progestogen hormone

therapy as ‘carcinogenic’ to humans.

THEORY OF ‘CARCINOGENESIS’

Carcinogenesis involves two steps, name-

ly, initiation and promotion. Most of the

studies on the relationship between hor-

mones and cancer development involved

the latter step. In 1989, Anderson et al

reported that nulliparous women who

took COC had a significantly higher rate of

breast cell division.5 It was also found that

COC caused a rise in epithelial cell pro-

liferation of the glandular breast,6 leading

to an increase in accumulation of random

genetic errors.7

However, whether these proliferat-

ing effects on normal epithelia, as a result

of replication error, may cause malignant

transformation has not yet been proven,

although DNA repair is hampered by ac-

tivated proliferation.8 With the end point

being ‘chromosomal mutation’, numer-

ous chromosomal aberrations have been



observed with natural and synthetic oes-

trogens and progestogens. There was no

proof of sufficient strength to show that

these proliferative effects could induce

tumours. Hormonal tumour promotion also

cannot be discriminated from a causal re-

lation with breast tumour induction.

Moreover, toxicity studies used

animal models. Species-specific effects

might not allow for extrapolation of the ef-

fects to humans. Supraphysiological doses

had been used in animal studies and this

might not reflect actual clinical use. Pos-

sibility of genetic predisposition and other

environmental factors were not taken into

account.

CARCINOMA OF THE BREAST

In 1981, Pike et al reported that women

who took COC for 4 years or more prior to

their first full-term pregnancy experienced

a 125% increased risk of developing carci-

noma of the breast and a 250% increase

in risk with 8 years or more of COC use.9

Similarly, in 1989, Chilvers et al reported

that women under the age of 36 who used

COC for at least 4 years before their first

full-term pregnancy had at least 44% in-

creased risk of breast cancer.10

However, the Cancer and Steroid Hor-

mone Study (CASH), which was one of the

largest case-control studies in the 1980s,

Combined oral contraceptives play a role in the promotion of carcinogenesis.

found no association between breast can-

cer and COC use for women up to the age

of 54. Risk was found only among a sub-

group of women who underwent menarche

before age 13 and used COC for more than

10 years before their first birth.11

In the 1990s, the Collaborative Group

on Hormonal Factors in Breast Cancer in

Oxford, UK, analysed individual data of

53,297 women with breast cancer and

100,239 women without breast cancer

from 54 studies conducted in 25 countries.

The results provided two strong conclu-

sions. First, while women are taking COC

and in a period of 10 years after stopping,

there is a small increase in the relative

risk (RR) of having breast cancer: RR of

1.24 (95% CI, 1.15–1.33) for current users;

RR of 1.16 (95% CI, 1.08–1.23) 1–4 years

after stopping; and RR of 1.07 (95% CI,

1.02–1.13) 5–9 years after stopping. Sec-

ond, there is no significant excess risk of

having breast cancer diagnosed 10 or more

years after stopping use (RR, 1.01 [95%

CI, 0.96–1.05]). The cancers diagnosed in

women who had used COC were less ad-

vanced clinically than those diagnosed in

never-users; the RR for tumours that had

spread beyond the breast compared with

localized tumours was 0.88 (95% CI, 0.81–

0.95). There was no pronounced variation

in the results for recency of use between

women with different background risks of

breast cancer, including women from dif-

ferent countries and ethnic groups, women

with different reproductive histories, and

those with or without a family history of

breast cancer. Other features of hormonal

use, such as duration of use, age at first

use, and the dose and type of hormone




within the contraceptives, had little ad-

ditional effect on breast cancer risk, once

recency of use had been taken into ac-

count.12

Other important studies worth men-

tioning include the Nurses’ Health Study

(1997),13 Women’s Lifestyle and Health

Cohort Study (2002),14 Oxford-Family Plan-

ning Association study (1981),15 Mayo

Clinic Meta-analysis (2006),16 and Royal

College of General Practitioners study

(2007).17 There were some inconsistencies

among their findings, and the increase in

RR shown by some of these studies was

modest. Possible confounders that may

be related to the use of high-dose COC

include the clinical practice during the

time of these studies and the possibil-

ity of the nature of recall bias. The study

by the Collaborative Group on Hormonal

Factors in Breast Cancer12 also shared a

similar problem. Another weakness of this

study was that it analysed pooled data

from studies which examined women with

breast cancer from as far back as the early

1970s. Taking data from studies which in-

terviewed women before the 1980s might

underestimate the risk of breast cancer

development because the latent period

for cancer development was too short and

few women had used COC for significant

periods of time prior to their first full-term

pregnancy in the late 1960s and early

1970s as compared with women of the

late 1970s and 1980s.

In 2005, the IARC classified COC and

combined oestrogen-progestogen hor-

mone therapy as carcinogenic to humans.

The Working Group mentioned a ‘slightly

increased risk of breast cancer in current

and recent users of hormonal contracep-

tives’. This risk disappears 10 years after

cessation of COC use and will be similar to

that in never-users.18 The Working Group

also acknowledged that their statement

does not meet the overall net public health

outcome, be this of a beneficial or ad-

verse effect other than cancer, and there

is no reason to change the current clini-

cal practice, particularly when the risks of

unwanted pregnancy are taken into con-

sideration.18

For BRCA mutation carriers, who al-

ready have a 50–80% increase in risk of

breast cancer, the use of COC will be of

concern. Among BRCA1 mutation carriers,

those who first used COC before 1975,

who used them before age 30, or who

used for 5 years or more might have an

increased risk of breast cancer. COC do not

appear to be associated with risk of breast

cancer in BRCA2 carriers; however, data to

support this are limited.19

On the use of depot medroxyproges-

terone acetate (MPA), pooled analysis of

two major case-control studies (one in

New Zealand20 and the other under the

auspices of the WHO21) found no increase

in risk for breast cancer. A currently unex-

plained pattern of increased risk in recent

users mimics that seen with COC.22

In a study involving completed ques-

Hormonal contraceptive use increases the relative risk of having breast cancer.



tionnaires from 17,360 levonorgestrel-

releasing intrauterine system (LNG-IUS)

users, there was no apparent associa-

tion between the length of time elapsed

from the LNG-IUS insertion up to 10 years

and yearly incidence of breast cancer in

the Finnish female population (data from

the Finnish Cancer Registry). A causal

relationship between LNG-IUS use and

occurrence of breast cancer was not sup-

ported.23

CARCINOMA OF THE UTERINE CORPUS

A meta-analysis of 10 case-control stud-

ies (published up to 1996; 1,728 cases and

6,243 controls) and another cohort study

(440,000 woman-years of observation)

both showed statistically reduced RR for

carcinoma of endometrium in COC users.

This RR was negatively associated with

the duration of COC use; the risk reduction

was 56% with 4 years’ use, 67% with 8

years’ use, and 72% with 12 years’ use.24

The Oxford-Family Planning Asso-

ciation (Oxford-FPA) contraceptive study,

which took place in 1968–2004, involved

540,000 woman-years of observation.

There were 50 women with carcinoma of

the uterine corpus in the control group and

27 women in the COC user group. The RR

for ever-users versus never-users was 0.3

(95% CI, 0.2–0.6). The risk was further

found to be negatively related to the du-

ration of COC use, ie, a RR of 0.6 (95%

CI, 0.3–1.1) for up to 48 months’ use, 0.4

(95% CI, 0.2–0.5) for 49–96 months’ use,

and 0.1 (95% CI, 0.0–0.4) for > 97 months’

use.25 The Royal College of General Prac-

titioners oral contraceptive study also

showed similar findings.17

Another meta-analysis of 11 epide-

miological studies found that the more

recent the use of COC, the lower the risk

for carcinoma of the uterine corpus. Such

protective effect from the previous use of

COC would attenuate with time after dis-

continuation. The RR was 0.33, 0.41 and

0.51 for 5, 10, and 20 years of ceasing, re-

spectively. Even after more than 20 years

of cessation of use, the protective effect

is still significant, ie, at 50% less than

non-users.24

Prolonged and unremitting mitotic

activity of the endometrium due to unop-

posed oestrogenic stimulation has been

proposed to be the cause of development

of the majority of cases of endometrial ad-

enocarcinoma. COC suppress endometrial

mitotic activity, leading to apoptosis, thus

reducing the risk of endometrial cancer.

The use of depot MPA is associated

with an 80% risk reduction of endometrial

adenocarcinoma, a level of protection even

greater than that observed with COC. The

effect was also found to be long-term.22

CARCINOMA OF THE OVARY

The Collaborative Group on Epidemiologi-

cal Studies of Ovarian Cancer published a

collaborative reanalysis of data from 45

epidemiological cohort and case-control

The protective effect of combined oral contraceptives against ovarian carcinoma is long-lasting.




studies, which included 23,257 women

with ovarian cancer and 87,303 controls

from 21 countries. It was found that 7,308

(31%) of the women with ovarian cancer

and 32,717 (37%) of the controls had used

COC, and the average duration of use was

4.4 and 5.0 years, respectively. The over-

all RR for ever-users versus never-users

was 0.73 (95% CI, 0.70–0.76). The longer

the duration of use, the lower the risk for

ovarian cancer development. The overall

RR decreased by 20% for each 5 years of

use. For women who had used COC for 15

years, the risk was almost halved. The pro-

tective effect started after at least 1 year

of COC use. The RR for < 1 year, 1–4 years,

5–9 years, 10–14 years, and ≥ 15 years of

use were 1.0, 0.78, 0.64, 0.56, and 0.42,

respectively.26

This reanalysis also found that the

more recent the use of COC, the lower the

RR of ovarian cancer. The proportional de-

cline in RR per 5 years of COC use was

29% for < 10 years of cessation of use,

19% for 10–19 years, and 15% for 20–29

years. The longer the duration of use, the

higher the protective effect irrespective

of the time elapsed from ceasing. The

start age of COC use and age of last use

seemed to have no effect on the protec-

tion. Low-dose pill use was found to have

an identical RR compared with high-dose

pill use.26

The Oxford-FPA contraceptive study,

which included 17,032 women aged 25–39

recruited at 17 family planning clinics in

England and Scotland between 1968–1974

with a long follow-up till 2004, analyzed

a total of 540,000 woman-years of ob-

servation and 58 ovarian cancer cases in

the control group and 48 cases in the COC

group.25 The overall ovarian cancer RR for

ever-users versus never-users was 0.5

(95% CI, 0.3–0.7). The risk of ovarian can-

cer was significantly lower in women on

oral contraceptives for more than 4 years.

The Royal College of General Prac-

titioners oral contraceptive study, which

started in 1968, collected data from

23,377 COC users and 23,796 never-users

over a period of 14 months, with 339,000

woman-years of observation for never-

users and 744,000 woman-years for ever-

users.17 It was found that the RR was 0.51

for ever-users as compared with never-

users. Similarly, a statistically significant

gradual decrease in risk with increasing

duration of COC use was observed. The

protective effect was found to last for at

least 15 years after stopping COC.

In the early 1970s, it was proposed

that defective cellular repair after ovula-

tion represents the major risk factor for

ovarian cancer development.27 More re-

cent theories assume that ovarian can-

cer development is attributed to either

activated proto-oncogenes or inactivated

tumour-suppressor genes,27,28 which seem

to point to abnormalities of genomic DNA

quantity and quality, with the resulting de-

fects in post-ovulatory ovarian cellular re-

pair being the causative factor for ovarian

cancer. Thus, the protective effect of COC

against ovarian cancer may be attributed

to the resulting anovulation during their

use, which prevents genetic predisposing

cellular repair defects to be expressed.

Although depot MPA also suppresses

ovulation and would theoretically lower

the risk of ovarian cancer, a hospital-based

WHO case-control study failed to uncover

such a protective effect.22

CARCINOMA OF THE CERVIX

There is evidence suggesting that long-

term use of COC for 5 years or more may

be associated with an increased risk of

cervical cancer.29 A meta-analysis of 28

studies, involving 12,531 women with

cervical cancer, suggested that the risk of

cervical cancer may decrease after stop-

ping the use of COC.30 Another IARC analy-

sis which included eight studies found a

fourfold increase in risk among women

with over 5 years of COC use. The risk was

also increased in women who started us-

ing COC before the age of 20 and in those

who had used COC within the previous 5

years.31

The mechanism for increased risk of

cervical cancer in COC users is uncertain.

Human papillomavirus (HPV) has been

recognized to be the major cause of carci-

noma of the cervix. Steroid contraception

has been postulated to be able to bind to

specific DNA sequences within transcrip-

tional regulatory regions on the HPV DNA,

either to increase or suppress the tran-

The protective effect of

COC against ovarian cancer

may be attributed to the

resulting anovulation

during their use



scription of various genes. It was suggest-

ed that the regulatory region of HPV type

16 viral genome indicates transcriptional

control of the HPV genome and might con-

tain enhancer elements that are activated

by steroid hormones.32

In COC users, the cervical mucus be-

comes scanty, thick, and highly viscous. It

has been hypothesized that such mucus

may modulate and prolong the effect of

carcinogenic agents and pathogens (in-

cluding HPV), which might have been car-

ried by coitus, on the cervical squamoco-

lumnar junction, causing them to become

difficult to be removed.33

However, the majority of studies did

not analyse the HPV status of COC users

and controls. Moreover, early use of COC

might be related to early onset of sexual

activity, which is itself a significant risk

factor for HPV infection and development

of cervical cancer. The lower use of the

barrier method of contraception in COC

users might be another accountable fac-

tor for the increase in risk of HPV infec-

tion. However, with the development of

HPV vaccines, the observed association

of increased risk of cervical cancer in COC

users might be changed, and fear of cervi-

cal cancer should not be a reason to avoid

COC use.

A large, population-based, case-con-

trol study in Costa Rica, a hospital-based

WHO case-control study in Thailand, Mex-

ico and Kenya, and a study in New Zealand

found that the risk of cervical cancer did

not appear to be affected by depot MPA

use.22

CONCLUSION

The majority of studies on the relationship

between hormonal contraceptive use and

development of cancer have focused on

COC and breast cancer, albeit with con-

flicting results. From the cumulative expe-

rience and meta-analyses of large epide-

miological studies with a long follow-up

duration, the present evidence suggests

an increase in risk of breast cancer devel-

opment mainly in current COC users, with

Figure 1. Risks of cancer development with duration of oral contraceptive pill (OCP) use

10

1.0

0.1

Rela

tive

risk

for c

ance

r dev

elop

men

t

Duration of OCP use (mo)0–48 49–96 > 97

Increased risk (?) of cervical cancer(Decreased risk on stopping OCP)

Increased risk of breast cancer(Decreased risk on stopping OCP)

Decreased risk of ovary cancer

Decreased risk of endometrial cancer




REFERENCES

the risk decreased on stopping therapy.

By 10 years of cessation of use, the risk

is similar to that in never-users. Similar-

ly, the risk of carcinoma of the cervix is

found to be increased in current COC us-

ers. However, COC use confers a strong

and prolonged protective effect against

carcinoma of the endometrium and ovary,

which will last even after over 20 years of

cessation of therapy (Figure 1).

Depot MPA was observed to cause

a plausible increased risk of breast can-

cer in current or recent users, similar to

that observed in COC users. It does not

seem to affect the overall breast cancer

risk. However, there is strong evidence of

prolonged decreased risk for carcinoma of

the uterine corpus. No strong association

has been noted for carcinoma of the cervix

and ovary.

When counselling women regarding

hormonal contraception, the issue of po-

tential carcinogenic effect from its use is

to be included. However, its prescription

should be based on an individual risk-

benefit assessment, provided contraindi-

cations are taken into account and regular

visits to doctors or health-care profession-

als are made.

1. Kirschstein RL, Rabson AS, Rusten GW. Infil-trating duct carcinoma of the mammary gland of a rhesus monkey after administration of an oral contraceptive: a preliminary report. J Natl Cancer Inst 1972;48:551–553.

2. Welsch CW, Adams C, Lambrecht LK, Hassett CC, Brooks CL. 17β-oestradiol and Enovid mam-mary tumorigenesis in C3H/HeJ female mice: counteraction by concurrent 2-bromo-alpha-ergocryptine. Br J Cancer 1977;35:322–328.

3. Geil RG, Lamar JK. FDA studies of estrogen, progestogens and estrogen/progesterone combi-nations in the dog and monkey. J Toxicol Environ Health 1977;3:179–193.

4. Shubik P. Oral contraceptives and breast cancer: laboratory evidence. IARC Sci Publ 1985;(65):33–35.

5. Anderson TJ, Battersby S, King RJ, McPher-son K, Going JJ. Oral contraceptive use influ-ences resting breast proliferation. Hum Pathol 1989;20:1139–1144.

6. Cogliano V, Grosse Y, Baan R, et al; WHO Inter-national Agency for Research on Cancer. Carci-nogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lan-cet Oncol 2005;6:552–553.

7. Preston-Martin S, Pike MC, Ross RK, Hender-son BE. Epidemiologic evidence for the increased cell proliferation model of carcinogenesis. Prog Clin Biol Res 1991;369:21–34.

8. Henderson BE, Feigelson HS. Hormonal car-cinogenesis. Carcinogenesis 2000;21:427–433.

9. Pike MC, Henderson BE, Casagrande JT, Rosa-rio I, Gray GE. Oral contraceptive use and early

abortion as risk factors for breast cancer in young women. Br J Cancer 1981;43:72–76.

10. Chilvers C, McPherson K, Peto J, et al; UK Na-tional Case-Control Study Group. Oral contracep-tive use and breast cancer risk in young women. Lancet 1989;333:974–982.

11. Cancer and Steroid Hormone Study of the Centers for Disease Control and the National In-stitute of Child Health and Human Development. Oral contraceptive use and the risk of breast can-cer. N Engl J Med 1986;315:405–411.

12. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal con-traceptives: collaborative reanalysis of individual data on 53297 women with breast cancer and 100239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–1727.

13. Colditz GA, Manson JE, Hankinson SE. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J Wom-ens Health 1997;6:49–62.

14. Kumle M, Weiderpass E, Braaten T, Persson I, Adami HO, Lund E. Use of oral contraceptives and breast cancer risk: The Norwegian-Swedish Women’s Lifestyle and Health Cohort Study. Cancer Epidemiol Biomarkers Prev 2002;11:1375–1381.

15. Vessey MP, McPherson K, Doll R. Breast can-cer and oral contraceptives: findings in Oxford-Family Planning Association contraceptive study. Br Med J (Clin Res Ed) 1981;282:2093–2094.

16. Kahlenborn C, Modugno F, Potter DM, Severs WB. Oral contraceptive use as a risk factor for premenopausal breast cancer: a meta-analysis.

Mayo Clin Proc 2006;81:1290–1302.

17. Hannaford PC, Selvaraj S, Elliott AM. Cancer risk among users of oral contraceptives: cohort data from the Royal College of General Practition-er’s oral contraception study. BMJ 2007;335:651.

18. Schneider HPG, Mueck AO, Kuhl H. IARC monographs program on carcinogenicity of com-bined hormonal contraceptives and menopausal therapy. Climacteric 2005;8:311–316.

19. Narod SA, Dubé MP, Klijn J, et al. Oral contra-ceptives and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers. J Natl Cancer Inst 2002;94:1773–1779.

20. Paul C, Skegg DCG, Spears GFS. Depot me-droxyprogesterone (Depo-Provera) and risk of breast cancer. BMJ 1989;299:759–762.

21. Breast cancer and depot-medroxyprogester-one acetate: a multinational study. WHO Collabo-rative Study of Neoplasia and Steroid Contracep-tives. Lancet 1991;338:833–838.

22. Kaunitz AM. Depot medroxyprogesterone acetate contraception and the risk of breast and gynecologic cancer. J Reprod Med 1996;41(5 Suppl):419–427.

23. Backman T, Rauramo I, Jaakkola K, et al. Use of the levonorgestrel-releasing intrauter-ine system and breast cancer. Obstet Gynecol 2005;106:813–817.

24. Schlesselman JJ. Risk of endometrial cancer in relation to use of combined oral contraceptives: a practitioner’s guide to meta-analysis. Hum Re-prod 1997;12:1851–1863.

25. Vessey M, Painter R. Oral contraceptive use and cancer: findings in a large cohort study,

1968–2004. Br J Cancer 2006;95:385–389.

26. Collaborative Group on Epidemiological Stud-ies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371:303–314.

27. Fathalla MF. Incessant ovulation—a factor in ovarian neoplasia? Lancet 1971;298:163.

28. Casagrande JT, Louie EW, Pike MC, Roy S, Ross RK, Henderson BE. “Incessant ovulation” and ovarian cancer. Lancet 1979;314:170–173.

29. Franceschi S. The IARC commitment to can-cer prevention: the example of papillomavirus and cervical cancer. Recent Results Cancer Res 2005;166:277–297.

30. Smith JS, Green J, Berrington de Gonzalez A, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–1167.

31. Moreno V, Bosch FX, Munoz N, et al . Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359:1085–1092.

32. Moodley M, Moodley J, Chetty R, Herrington CS. The role of steroid contraceptive hormones in the pathogenesis of invasive cervical cancer: a review. Int J Gynecol Cancer 2003;13:103–110.

33. Guven S, Kart C, Guvendag Guven ES, Gu-nalp GS. The underlying cause of cervical can-cer in oral contraceptive users may be related to cervical mucus changes. Med Hypotheses 2007;69:550–552.

About the Author

Dr Wong is Consultant in the Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Hong Kong.


This continuing medical education service is brought to you by MIMS. Read the article ‘Hormonal Contraception and Cancers’ and answer the following questions. Answers are shown at the bottom of this page. We hope you enjoy learning with JPOG.

CME Answers for JPOG Jan/Feb 2012

HKCOG CME Article: Management of Pregnancies With Previous Caesarean Section

Answers

1 2 3 4 5 6 7 8 9 10

T F F T T T T T T F

CME Article


Answer True or False to the questions below.

1. Combined oral contraceptives (COC) are classified by IARC as ‘carcinogenic’ to humans.

2. There is strong evidence for COC to cause a rise in epithelial cell proliferation of the glandular breast, leading to increase in genetic error and resulting in malignant transformation and breast cancer development.

3. The results of COC toxicity studies using animal models can accurately reflect the effects in humans.

4. The Collaborative Group on Hormonal Factors in Breast Cancer concluded that there was a small increase in the relative risk of having breast cancer in current and recent COC users.

5. Depot medroxyprogesterone acetate (MPA) users have decreased risk of carcinoma of the breast.

6. The protective effect of COC against carcinoma of the uterine corpus disappears after 10 years of stopping.

7. Low-dose COC has been shown to have a similar protective effect against carcinoma of the ovary compared with high-dose COC.

8. There is good evidence for the protective effect of depot MPA against carcinoma of the ovary.

9. There is evidence suggesting that long-term use of COC increases the risk of cervical cancer.

10. The carcinogenic potential of hormonal therapy should not be discussed during contraceptive counselling.

True False


CME Questions CME QuestionsAnswers

1 2 3 4 5 6 7 8 9 10

T F F T F F T F T F


cme article journal watch paediatrics...

Documents