Content uploaded by Alison Mcfadden
Author content
All content in this area was uploaded by Alison Mcfadden on Jun 25, 2014
Content may be subject to copyright.
Public Health Nutrition: 11(11), 1188–1199 doi:10.1017/S1368980008001791
Cleaning and sterilisation of infant feeding equipment:
a systematic review
Mary J Renfrew
1,
*, Marie McLoughlin
2
and Alison McFadden
1
1
Mother and Infant Research Unit, Department of Health Sciences, Area 4, Seebohm Rowntree Building,
University of York, Heslington, York YO10 5DD, UK:
2
Oxfordshire Primary Care Trust, Jubilee House,
5510 John Smith Drive, Oxford Business Park South, Cowley, Oxford OX4 2LH, UK
Submitted 3 August 2007: Accepted 14 December 2007: First published online 26 February 2008
Abstract
Aim: To assess the clinical and cost-effectiveness of different methods of cleaning
and sterilisation of infant feeding equipment used in the home.
Design: Systematic review of studies from developed countries on the effective-
ness of methods of cleaning and sterilisation of infant feeding equipment used in
the home. A brief telephone survey of UK-based manufacturers of infant feeding
equipment and formula to ascertain the evidence base used for their recom-
mendations, and a comparison of current relevant guidelines in developed
countries, informed the work.
Results: National guidelines from six countries demonstrated variation and lack of
evidence to support current guidance. Manufacturers did not report evidence of
effectiveness to support their recommendations. Nine studies were identified;
eight conducted between 1962 and 1985 and one in 1997. All had methodological
weaknesses. Hand-washing was identified as fundamentally important. Health
professionals were reported as not providing appropriate education on the
importance and methods of cleaning and sterilisation. Mothers of subsequent
babies and women from lower socio-economic groups were less likely to follow
recommended procedures.
Conclusion: There is a lack of good-quality evidence on effective ways of
cleaning and sterilising infant feeding equipment in the home. The evidence base
does not answer the question about which of the methods in common use is most
effective or most likely to be used by parents. Hand-washing before handling
feeding equipment remains important. Further research on the range of methods
used in the home environment, including assessment of the views of parents and
carers, is required.
Keywords
Bottle feeding
Infant nutrition
Infant formula
Infection
Sterilisation
Systematic review
In the UK and many other Western countries, breast-
feeding rates fell throughout the first half of the 20th
century and remain at low levels. Over the past decades,
work conducted to raise the rates of breast-feeding has
met with some success
(1,2)
, but serious problems remain
internationally as highlighted in the second Innocenti
Declaration on the protection, promotion and support of
breast-feeding
(3)
. In the UK around 24 % of babies will be
formula-fed from birth, and by 6 weeks following birth
79 % of babies will be fed exclusively or partially on infant
formula
(2)
, with the highest rates of formula feeding
among women of lower socio-economic status. The
situation is similar in other Western countries including
the USA
(4)
, while in other countries, including China,
previously high breast-feeding rates are declining
rapidly
(5)
. Worldwide, many millions of babies will be fed
breast milk substitutes, usually by use of a plastic bottle
and teat. The health consequences of this are enormous,
and result in increased mortality and morbidity in both
developed and developing countries
(6,7)
.
Risks to the baby from formula feeding not only
include the intrinsic nutritional and immunological defi-
ciencies of formula compared with breast milk. They also
include risks that could potentially be reduced, such as
errors in the manufacturing process, contamination dur-
ing storage and transport
(8)
, errors in reconstitution of
infant formula in the home
(9)
, and ensuring effective
cleaning and sterilisation of the equipment used.
Article 1 of the WHO Code on the Marketing of
Breastmilk Substitutes endorses ‘the proper use of
breastmilk substitutes, when these are necessary, on the
basis of adequate information’
(10)
. What are the evidence-
based ways of ensuring the ‘proper use’ of such sub-
stitutes? Unlike the evidence base for the promotion and
*Corresponding author: Email mjr505@york.ac.uk r The Authors 2008
support of breast-feeding, which has been strengthened
considerably in recent decades
(11–13)
, the evidence base
for reducing the risks of formula feeding seems to be
scanty. It is not clear from the epidemiological literature
what proportion of common diseases such as gastro-
enteritis is an inevitable consequence of the use of a
product of lower quality than breast milk and what pro-
portion could be prevented by improved practice. One
recent study
(7)
has suggested that, in formula-fed infants,
there was significantly more diarrhoeal disease in those
infants whose carer did not sterilise bottles and teats with
steam or chemicals, particularly in infants under 6 months
of age (adjusted OR 5 9?13; 95 % CI 1?17, 71?39;
P 5 0.012). No information was available, however, on
the relative effectiveness of the different methods being
used.
Milk is the perfect medium for growth of bacteria, and
therefore poorly cleaned feeding equipment can be a
potent source of infection for babies
(14)
. The organisms
of most concern are reported to be Salmonella
(8)
and
Enterobacter sakazakii
(8,15)
. To reduce contamination
with these organisms, the Food Standards Agency and
Department of Health in England have recently recom-
mended that formula feeds are made up using boiled
water that is .708C (water that has been boiled and left to
cool for no more than 30 min) and that formula is made
up fresh for each feed
(16)
. A report by the European Food
Safety Authority (EFSA)
(8)
concluded that cleaning and
sterilisation of equipment in the home is a critical part of
the avoidance of infection; recommendations include the
use of ‘sterile bottles, achieved by heating and chemical
methods’, although no evidence is provided on the rela-
tive effectiveness of these methods. The recent WHO
guidelines published in 2006 and updated in 2007
(17)
are
consistent with the EFSA recommendations and suggest
that manufacturer’s instructions should be followed for
chemical or steam sterilisation procedures. These guide-
lines are stated (p. 2) to be ‘largely based on the findings
of a quantitative microbiological risk assessment for
Enterobacter sakazakii ’. Effective cleaning and sterilisa-
tion of infant feeding equipment offers the opportunity to
minimise risks to the baby and could result in significant
clinical and cost benefits
(18)
.
Various methods of cleaning and treating infant feeding
equipment are used internationally (Table 1). Potential
problems in using these methods routinely include cost,
the time-consuming and complex nature of some meth-
ods, confusion over the length of time equipment should
be boiled, left to soak or left in the microwave, whether
or not equipment left to soak in hypochlorite should be
rinsed, and how equipment immersed in boiling water or
hypochlorite should be removed and dried. Such confu-
sion, expense and the time needed may result in a lack of
compliance. It is also not clear if basic hygiene measures
such as hand-washing are seen by parents and carers as
important in the face of more complicated approaches.
Furthermore, use of dishwashers has been implicated in
the release of plasticisers following a relatively small
number of washes
(19,20)
. It is not surprising, therefore,
that there is variation in the information and advice given
by health professionals. One survey conducted in Scotland
found that, before the birth of their baby, only 40 % (n 25)
of women considering bottle feeding had been given any
information on sterilising equipment
(21)
. It is essential to
maximise the opportunities for women to breast-feed,
and it is also very important to offer opportunities for
parents to learn about minimising the risks of formula
feeding.
Preliminary work
Two brief investigations were undertaken to inform this
review and to search for unpublished studies. First, UK-
based manufacturers of infant feeding equipment and
formula who made recommendations for cleaning and
sterilising techniques in the literature accompanying their
product were identified. The health advisor for each
company was contacted by telephone by one of us (M.M.);
all agreed to be interviewed. They were asked for infor-
mation about the evidence base that informed their
published recommendations. Responses are summarised
in Table 2. Although two infant formula companies and
two manufacturers of feeding equipment reported that
they based their instructions on Department of Health
guidelines or policy, at the time of the interviews (2003)
no such guidelines existed on the cleaning and sterilisa-
tion of infant feeding equipment. No relevant studies
were identified.
Second, we contacted the UNICEF Baby Friendly
Initiative Co-ordinator in similar developed countries and
requested copies of relevant national guidelines and
information about the evidence base used. The results
for these six countries are summarised in Table 3 and
Table 1 Methods of cleaning or sterilising infant feeding equipment
Method Description
Boiling Placing equipment in a saucepan of boiling water
Chemical
sterilisation
Use of a hypochlorite solution into which the
feeding equipment is placed
Electric steam
steriliser
This is a container with a hotplate set into the
base. The equipment is stacked into a
designed plastic basket; a small amount of
water is added to the base and the unit closed
by a lid. The directly heated unit then boils the
water until all of it has vaporised, then switches
itself off. The items to be sterilised are
surrounded by steam, which kills the bacteria
only if the equipment has the closed end facing
upwards
Microwave
steriliser
Feeding equipment is placed in a basket with a
lid. Water is added to the base of the unit and
placed in the microwave; the water boils by the
action of the microwave energy
Dishwasher Feeding equipment is placed in the rack of the
dishwasher and the cycle run as normal
Cleaning and sterilisation of infant feeding equipment 1189
demonstrate variation in whether sterilisation was
recommended for all babies (UK, Ireland)
(22–24)
, only for
those under 3 months of age (New Zealand
(25)
,Norway
(26)
,
USA
(27)
), or only on the recommendation of a health
professional (Canada)
(28)
. In the USA and Norway the use
of a dishwasher was recommended as an alternative to
sterilisation. There were also variations in the detailed
instructions for cleaning teats (i.e. use of salt or not), and
in whether bottles and teats sterilised by the chemical
method should be rinsed before use. No international
respondent was aware of any evidence of effectiveness
informing the guidelines from their country.
The aim of the systematic review described in the
present paper was to assess the clinical and cost-effec-
tiveness of different methods of cleaning and sterilisation
of infant feeding equipment used in the home.
Methods
Literature search
The first search was conducted in 2003 on the following
databases: Medline, Embase, CINAHL, Psycinfo, British
Nursing Index (BNI), Allied and Complementary Medicine,
Premedline, Health Management Information Consortium
(HMIC), EBM reviews, SIGLE and the Cochrane library
database (which included CDSR, ACP Journal Club, CCTR
and DARE). Electronic database searches were supple-
mented with hand searches of the references of selected
papers, relevant policy documents and consultation with
the key professionals and companies in this field. Grey
literature and unpublished studies not recorded on SIGLE
were identified by searching the National Research Reg-
ister and the NHS Centre for Reviews and Dissemination
database. A broad search strategy was used to identify all
relevant literature, using the following keywords and
Mesh terms: bottle$, artificial feed$, formula feed$, infant
feed$, teat$, sterili?$, clean$, wash$, prepar$, disinfect$,
saniti$, hygien$, breastfeeding, breast pump, germ free,
spotless, infection, uncontaminated. A second, structured
search was run on Medline in 2006 which identified only
one additional paper; the full search strategy is shown in
Table 4.
Inclusion/exclusion criteria
No date limits were set on either search. The review
included research studies from developed countries that
examined methods of cleaning and/or sterilisation of
infant feeding equipment, either in the home or applic-
able to home conditions, regardless of the research
design used. Studies from developing countries were
excluded, as the sterilisation and infection issues are
different in such dissimilar settings.
Outcomes
Outcomes of interest included clinical outcomes in
infants; results of microbiological tests of bottles and
teats; behavioural outcomes for carers; and costs.
Data extraction
Data were systematically extracted by one reviewer (M.M.
or A.M.) using pre-designed data extraction forms, and
were checked by a second reviewer (M.J.R.). Included
studies were assessed for quality using criteria published
by the Centre for Reviews and Dissemination
(29)
.
Results
Published literature
Of the 1520 references identified, only nine studies pub-
lished in eight papers
(18,30–36)
met the inclusion criteria; all
but one
(18)
were identified in the early search. The majority
of the other citations were not research studies. No
systematic reviews were identified. Details of the nine
included studies are given in Table 5 and the quality
of the included studies is summarised in Table 6. Eight
were cond ucted between 1962 and 1987
(30–36)
and one
in 1998
(18)
. Three were conducted in the USA
(33,34,36)
and
six in the UK
(18,30–32,35)
. One was a randomised con-
trolled trial
(34)
, four were non-randomised controlled
trials
(18,32,33,36)
, and four were surveys
(30–32,35)
. The number
of participants ranged from twenty-six to 758 (median,
sixty-three). No studies examined cost-effectiveness. All
studies had serious method ological weaknesses.
Two studies examined infant morbidity related to the
cleaning method used
(32,33)
. No significant difference in
the incidence of infections or illness was identified,
although no studies were of appropriate design or
sufficient quality or size to answer these important
questions.
The largest study
(30)
, conducted in 1970, took place
across four different geographical areas in the UK,
each selected because of proximity to a public health
Table 2 Responses to the survey of manufacturers of infant
feeding formula and infant feeding equipment
Type of company
Evidence used to inform their
recommendations
No. of
companies
Manufacturers of
infant formula (n 3)
Department of Health
guidelines
1
Manufacturers of the
sterilisation equipment
1
UK policy and expert advisor 1
Manufacturers of
infant feeding
equipment (n 7)
Department of Health
guidelines
2
British Standards 1
Expert advisor 1
Company carries out
independent tests on the
equipment to test the
efficacy of sterilisation
3
1190 MJ Renfrew et al.
Table 3 Comparison of national guidelines for cleaning and sterilising infant feeding equipment
Country
Guidelines
issued by Guidelines for Cleaning
Sterilisation
recommended
Recommended methods of sterilisation
and specific instructions Additional recommendations
UK Department of
Health
Public Clean bottles and teats in
hot soapy water using a
clean bottle brush.
Squirting water through
the teat will help remove
every trace of milk
Yes Chemical: follow the manufacturer’s
instructions. Change the sterilising
solution every 24 h, leave equipment in the
sterilising solution for at least 30 min, make
sure there is no air trapped in the bottles or
teats, keep all the equipment under the
solution with a floating cover, before
making up the feed, shake off any excess
solution from the bottle and teat or rinse
with cooled boiled water from the kettle
Steam steriliser: follow the manufacturer’s
instructions, make sure the openings of the
bottles and teats are face down in the
steriliser, any equipment not used straight
away should be re-sterilised before use
USA Department of
Agriculture
(USDA)
Health
professionals
Equipment should be
thoroughly cleaned
using soap, hot water
and bottle and nipple
brushes
Sterilisation or
dishwasher
for babies
,3 months old
Boiling: boiling water for 5 min After 3 months, unless otherwise
indicated by a health-care provider,
bottles should be thoroughly
washed using soap, hot water and
bottle and nipple brushes or
cleaned in a dishwasher
Dishwasher: wash in a properly functioning
dishwasher machine
Canada Canadian Medical
Association
Public All feeding-related items
should be washed in hot
soapy water and
thoroughly rinsed
All feeding equipment
should be sterilised
before first use, then
check with a doctor
Boiling: items should be fully submerged,
for five consecutive minutes, in water
that is at a rolling boil
Use dishwasher-proof nipples
(teats), when nipples are to be
cleaned in a dishwasher
New
Zealand
National
Government
Health
Department
Public and health
professionals
Clean bottles with a bottle
brush in warm soapy
water. Clean teats in warm
soapy water and rub salt
inside the teat and through
holes to remove milk
residue. Rinse bottles and
teats well with running cold
water, ensuring this is
squirted through holes
in teats
Yes, for babies
,3 months
Boiling: boil the feeding and mixing utensils
in a pot of water big enough to cover
everything for 5 min. Use clean tongs to
lift everything out and drain well and
store in a clean dry place until required
When the baby is older, thorough
washing is enough
Chemical: completely immerse all items in a
large plastic container with a lid. Make the
sterilising solution according to the
directions. Equipment must be fully
immersed for 1 h. Drain bottles and teats
well but do not rinse before use. Sterilising
solution must be renewed every 24 h
Follow the manufacturer’s instructions if
you use any of the following methods:
Microwave – with a special microwave
sterilising unit
Steam steriliser
Norway National
Government
Health
Department
Public Rinse and fill bottles/cups
with cold water. Wash
well, with a brush used
only for bottles, in hot
water with detergent.
Rinse with as hot water
as possible. Teats should
be carefully cleaned
Yes, ‘in first weeks’ Boiling: boil bottles and teats. Let the bottles/
cups air-dry. Teats should also be boiled
Dishwasher: use dishwasher if you have one
Cleaning and sterilisation of infant feeding equipment 1191
laboratory. Methods used in the home differed across the
four areas; these included the hypochlorite method alone
(48 %), boiling (30 %), hypochlorite and boiling together
(11 %), and the remainder either used another method
(not stated) or nothing (12 %). Bottles and teats sterilised
using the hypochlorite method had lower colony counts
(bacterial count #5: 63 % (n 475) of bottles; 52 % (n 395)
of teats) when samples were tested in the laboratory.
The majority of mothers who used the hypochlorite
method lived in rural areas and also spent more time
washing and sterilising the equipment. The majority did
not carry out sterilisation procedures as recommended, in
spite of the fact they had received health education in this
field. Clegg et al.’s study in England in 1977
(31)
found
similar results.
The most recent study
(18)
examined the effectiveness of
commonly used cleaning and disinfecting procedures on
the removal of enterotoxigenic Bacillus cereus from
feeding bottles. This non-randomised experimental study
was conducted in the laboratory, but included subjecting
bottles to storage conditions which may occur in the
home. The results showed that thorough cleaning
reduced, but did not completely eliminate, all microbes.
All of the three disinfection procedures tested (one
chemical and two thermal) eliminated the organism at
low levels of contamination (,10
5
organisms/ml) but the
chemical method failed to eliminate B. cer eus at poten-
tially hazardous levels ($10
5
organisms/ml) which may
occur with improper use in the home.
Table 3 Continued
Country
Guidelines
issued by Guidelines for Cleaning
Sterilisation
recommended
Recommended methods of sterilisation
and specific instructions Additional recommendations
Ireland UNICEF UK
guidelines
Public (widely
used by health
professionals)
Wash all bottles and other
equipment thoroughly in
hot soapy water using a
bottle brush. Scrub inside
and outside the bottle. Pay
particular attention to the
rim. Use small teat brush
to clean inside teat or turn
inside out and wash in hot
soapy water. If salt used,
always rinse off
completely. Rinse all
washed equipment before
sterilising
Yes, up to 1 year Boiling: put equipment into pan filled with
water. Make sure no air is trapped. Cover
with lid and bring to boil. Boil for at least
10 min. Keep pan covered until equipment
is needed
Chemical: make up solution following
manufacturer’s instructions. Submerge
equipment in the solution, making sure no
air is trapped. Use plunger to keep
equipment under water. Leave in solution
for at least 30 min. If you rinse before
using, use water that has been boiled and
allowed to cool. Always wash hands before
removing equipment from steriliser
Steam or microwave: always follow the
manufacturer’s instructions
Department of
Health and
Children, Health
Promotion Unit
Public Bottles and teats should be
thoroughly washed after
use
Yes, up to 1 year Sterilise up to 1 year
Table 4 Structured search strategy, 2006
1 bottle feeding/
2 (bottle$ adj8 feed$).ti.
3 (bottle$ adj8 feed$).ab.
4 (nursing adj2 bottle$).ti.
5 (milk adj2 bottle$).ti.
6 (teat or teats).ti,ab.
7 animal/
8 6 not 7
9 nipples.ti,ab.
10 (bottle$ adj4 nipple$).ti,ab.
11 10 not 7
12 bottle$.ti,ab.
13 exp infant/ or infant, newborn/
14 exp infant/
15 12 and 14
16 feeding equipment.ti,ab.
17 1 or 2 or 3 or 4 or 5 or 8 or 11 or 16
18 (editorial or letter or comment).pt.
19 17 not 18
20 *dental caries/
21 19 not 20
22 disinfection/
24 microwaves/
25 sterilization/
26 clean$.ti,ab.
27 (sterilis$ or steriliz$).ti,ab.
28 hygiene.ti,ab.
29 disinfect$.ti,ab.
30 22 or 24 or 25 or 26 or 27 or 28 or 29
31 21 and 30
33 from 31 keep 1–38
1192 MJ Renfrew et al.
Table 5 Summary of studies included in the present review
Study Setting Participants
Methods of preparation of infant
feeding equipment Data collection methods Main findings
Hargrove et al.
(1974)
(33)
: non-
randomised
controlled trial
USA 26 babies selected from a
hospital nursery representing
a wide range of socio-
economic levels
Intervention group (n 13): babies were fed
formula prepared by the clean method –
bottles and nipples (teats) were washed
with hot soapy water and rinsed with hot
running water. Bottles were filled
immediately before each feed
Mothers recorded signs of illness
in their baby; home visits to
assess infant well-being,
cleanliness of the home and
preparation time; lab analysis
of samples of unused formula
Clinical: no differences occurred in frequency
of illness in the 2 groups
Control group (n 13): babies were fed
formula prepared by terminal
(not defined) sterilisation
Microbiology: 70 % of each set of formula
samples grew aerobic bacteria, whether
prepared by the clean or sterilisation
technique; 6 samples from the clean group
and 5 from the sterilised group showed
growth of coliform organisms
High incidence of positive cultures occurred
regardless of socio-economic background,
housekeeping habits or method of
preparation used
Behaviour: women from the clean method
group spent on average 10 min a day on
preparation v. 30–45 min spent by the
control group
Hughes et al.
(1987)
(34)
:
randomised
controlled trial
USA 284 bottle-fed infants and their
mothers who delivered in a
regional hospital; babies
were full term
Group 1: mothers were individually taught
the terminal method of formula preparation
according to written guidelines (no further
details given; the discussion indicates this
is similar to ‘traditional sterilisation’)
Accident and emergency
records reviewed for
cases of gastroenteritis
Clinical: 11 (3?9 %) cases of gastroenteritis
were documented, 5 in the ‘terminal’ group
and 6 in the ‘clean’ group
Group 2: mothers were individually taught the
clean method of formula preparation
according to written guidelines (no further
details given: the discussion indicates is based
on washing with hot water and soap)
Number of participants assigned to each
group not reported
No significant difference found between the
method of formula preparation taught to
mothers and incidence of gastroenteritis in
their infants
Jacob (1985)
(35)
:
survey
UK 30 mothers with babies aged
5 months; 6 mothers were
randomly selected from
each of 5 child health
clinics in catchment area
28 mothers sterilised bottles and teats
using cold method (‘soaking in a
hypochlorite solution’); 1 used boiling
technique; 1 did not use any method
Interviews with mothers Behaviour: 14 (46?6 %) mothers were
sterilising ‘correctly’ and 16 (53?3 %) were
not
Of those not sterilising ‘correctly’, 14 did not
use salt to clean the teats, 5 did not use
detergent, 3 did not sterilise if another fluid
was used instead of milk, 1 did not sterilise
the teats, 1 did not sterilise at all and 1 did
not soak the equipment for minimum
recommended time
81 % of mothers who were not sterilising
‘correctly’ were from social classes 4 & 5 v.
17 % from social classes 1 & 2
Majority of the mothers who were not
sterilising correctly were multiparous
(P , 0?02)
Women reported learning how to sterilise
equipment from manufacturers’
instructions ( n 24, 80 %) and from
instructions during their postnatal stay in
hospital (n 15, 50 %) as well as family and
friends (n 11, 36 %)
Cleaning and sterilisation of infant feeding equipment 1193
Table 5 Continued
Study Setting Participants
Methods of preparation of infant
feeding equipment Data collection methods Main findings
Vaughan et al.
(1962)
(36)
: non-
randomised
controlled trial
USA 63 mothers selected ‘essentially
at random’ from a well baby
clinic with infants aged 6–16
weeks; alternate mothers
allocated to intervention group
Intervention group (n 29): mothers performed
the bottle-at-a-time method (at each feed:
wash hands, wash bottle and teat with
bottle brush, rinse thoroughly; rinse after
feeding and leave to dry until next feed)
Initial interview; home visits to
evaluate home conditions;
bacteriological cultures form
homes; mothers’ reports of
illness of the infant
No results presented by intervention/control
groups
Control group (n 29): mothers continued
using their ‘usual technique’ (no details)
Microbiology: 20 % (n 45) of samples from
homes designated ‘sanitary’ showed heavy
growth of organisms v.36%(n 26) from
homes designated ‘unsanitary’
Largest percentage of heavy grow th (19/23)
and most regular isolation of Escherichia coli
(8/28) were from containers in which milk
wasstoredinbulk
Gatherer (1978)
(32)
:
study was in two
parts, (1) survey
and (2) non-
randomised trial of
two products
UK Part 1: 94 mothers and their
babies aged 2–6 months
Cold chemical (hypochlorite solution) Interview; sample of sterilising
fluid from home taken for lab
analysis of available chlorine
Bacteriological: results showed nil growth in
91 % (n 86) of bottles and 75 % (n 71) of
teats. Nearly 30 % (n 28) of mothers were
using a sterilising fluid which gave a low
estimation of residual chlorine, 11 % (n 10)
of these had poor results with either
bottles or teats
Behaviour: nearly all mothers (no. not
reported) were ‘correct’ in their account of
bottle hygiene but not of teats. Many
mothers (no. not reported) thought rinsing
or using a brush was not necessary, 14
mothers did not sterilise dummies, small
feeders and teething rings
Of the 36 % (n 34) of mothers who did not
attend antenatal classes, 5 had poor
results with bottles and 17 with teats
UK Part 2: 51 mothers divided
into 2 groups
Group 1 (n 27): fluid hypochlorite solution Bacteriological assessment of
bottles and teats
Microbiology: on bacteriological assessment,
all of the feeding bottles gave satisfactory
results, as did all but 2 of the teats (no
further information given)
Group 2 (n 24): unit dose crystals product
Both groups of mothers were given additional
instructions and encouragement by
midwives and health visitors
Anderson and
Gatherer
(1970)
(30)
: survey
UK 1000 homes, with babies aged
2–4 months, were selected
in 4 geographical areas which
contained a cross-section of
socio-economic groups; 758
completed the questionnaire
and gave bottles and teats
to be tested
94 % of mothers reported they sterilised
feeding bottles and teats; 48 % (n 362)
used the hypochlorite method alone; 30 %
(n 230) relied solely on boiling; 11 % used
both methods together; the remainder used
another method or did nothing
Questionnaire; bacteriological
assessment of bottles
and teats
Microbiology: 63 % (n 475) of bottles and
52 % (n 395) teats had a bacterial count of
#5 colonies. Area b was significantly
worse than the other areas, both for bottles
and teats (P , 0?001)
Health visitors arrived without prior warning and
collected a bottle and teat which were,
according to the mother, clean and ready
to be filled for the morning feed
78 % (n 281) of bottles and 70 % (n 253) of teats
sterilised by hypochlorite had #5coloniesv.
46 % (n 106) of bottles and 34 % (n 77) of
teats sterilised by boiling method
Behaviour: more mothers using hypochlorite
method used a more thorough cleansing
routine – more of them used bottle brushes
and liquid detergent and more rubbed the
teats with salt and turned them inside out
(no figures given)
1194 MJ Renfrew et al.
Table 5 Continued
Study Setting Participants
Methods of preparation of infant
feeding equipment Data collection methods Main findings
Clegg et al.
(1977)
(31)
:
survey
UK 63 mothers, all having babies
aged 2–5 months who were
being bottle-fed, from wide
range of socio-economic
groups
3 used the boiling method; 60 used some
form of hypochlorite solution (not defined)
Bacteriological assessment of
bottles and teats
Microbiology: 98?1 % of bottles and 90?6% of
teats had a residual count of less than 5
colonies/ml. As only 3 mothers were using
the boiling method it was impossible to
make any significant notes in comparison
to the other technique
Mothers were provided with a commercial
sample of a stabilised solution of 1 %
sodium hypochlorite to use for treatment
of their babies’ feeding utensils; each
mother was visited twice by a
microbiologist who collected a bottle
and a teat which had been immersed
for minimum of 30 min in a sterilising
unit using the aseptic technique
Rowan and Anderson
(1998)
(18)
:
non-randomised
lab-based trial
of 3 procedures
UK lab 20 infant feeding bottles Bottles contaminated with different levels
of enterotoxigenic Bacillus cereus were
subjected to commonly used cleaning
and disinfecting procedures. Uncleaned,
partially cleaned and thoroughly cleaned
bottles were subjected to the following:
Bacteriological assessment of
bottles and teats
Microbiology: the greater the level of infant
bottle cleaning, the larger the reduction in
microbial numbers. Effectiveness of each
cleaning stage at removing B. cereus was
significant (P , 0?05). Thorough cleaning
did not remove all spores and potentially
hazardous levels remained in more
contaminated bottles
Steam sterilisation: bottles automatically
steamed at 1008C for 15 min
All methods of disinfection successfully
reduced B. cereus to non-detectable level
when initial contamination was #10
5
colonies/ml. B cereus emerged earlier
(after 14 h) in uncleaned bottles subjected
to chemical disinfection. Both thermal
disinfection methods did not totally
eliminate B. cereus after 18 h. Level of
contamination and degree of bottle
cleaning affected length of time that B.
cereus remained at undetectable levels
(P , 0?05). Chemical method failed to
disinfect uncleaned feeding bottles
contaminated with 10
5
colonies/ml.
Potentially hazardous levels detected after
14 h storage following thermal disinfection.
Both steam disinfection methods equally
efficient at removing B. cereus from bottles
contaminated with #10
5
colonies/ml
(P , 0?05) and both methods significantly
better than chemical method (P , 0?05)
Microwave steam sterilisation: bottles placed
in a sterilising unit and steamed at 1008C
in a microwave oven for 9 min
Chemical sterilisation: bottles immersed in
sodium hypochlorite solution for 90 min
Cleaning and sterilisation of infant feeding equipment 1195
Table 6 Design and quality assessment* of the experimental studies included in the present review
Study Design Allocation concealment
Comparability of
groups Eligibility Blinding Analysis
Hargrove et al.
(1974)
(33)
Non-randomised
controlled trial
Group allocation
chosen by mother’s
preference
No information Healthy full-term
newborns, bottle
feeding, from one
hospital nursery
Blinding of participants not
possible: no evidence of
blinding of care providers or
outcome assessors
Point estimates/variability
not reported; ITT
analysis not stated
Hughes et al. (1987)
(34)
Randomised
controlled trial
No information on
allocation;
concealment not
possible
No information Term babies with stable
temperature and blood
sugar
Blinding of participants not
possible: no evidence of
blinding of care providers or
outcome assessors
Point estimates/variability
not reported; ITT
analysis not stated
Jacob (1985)
(35)
Survey Not an experimental design – background information only
Vaughan et al.
(1962)
(36)
Non-randomised
controlled trial
Alternate allocation,
method not
described;
concealment not
possible
No information Mothers who attended
well baby clinics
selected ‘at random’
Blinding of participants not
possible: no evidence of
blinding of care providers or
outcome assessors
No results presented by
groups as allocated
Gatherer (1978)
(32)
Survey Not an experimental design – background information only
Non-randomised
trial of two
products
No information on
allocation;
concealment not
possible
No information Informed volunteers,
wished to bottle
feed
Blinding of participants not
possible: no evidence of
blinding of care providers or
outcome assessors
Point estimates/variability
not reported; ITT
analysis not stated
Anderson and Gatherer
(1970)
(30)
Survey Not an experimental design – background information only
Clegg et al. (1977)
(31)
Survey Not an experimental design – background information only
Rowan and Anderson
(1998)
(18)
Non-randomised
lab-based trial of
three procedures
All bottles subjected
to three procedures
in turn
All bottles
subjected
to three
procedures
in turn
Bottles prepared to a
protocol in lab
conditions
Blinding of outcome assessors not
described
All results presented for
all bottles in groups
as treated
ITT, intention to treat.
*Using criteria in Khan et al.
(29)
.
1196 MJ Renfrew et al.
Authors of several of the included studies
(33,34)
sug-
gested that the ‘clean’ (washed with hot soapy water and
rinsed with hot running water) method is a safe alter-
native to traditional ‘sterilisation’ techniques, provided
the safety of the water is assured. However, three stu-
dies
(30,32,35)
found higher numbers of organisms on teats,
suggesting that they are more difficult to clean effectively
than bottles. Gatherer
(32)
indicated that bacteriology
results were excellent using either thorough cleaning or
sterilisation; this was attributed to the education provided
for mothers. Other studies
(30–32,35)
identified a link
between the failure to correctly sterilise and prepare
formula feeds and a lack of ante- and postnatal education
from health professionals, multiparity (mothers with two
or more children) and low socio-economic status. Several
authors
(18,30,31,37,38)
have suggested that improved teach-
ing for mothers and consistent advice from health pro-
fessionals
(38)
are key factors in improving sterilisation and
cleaning of infant feeding equipment.
Discussion
The striking finding from the present study is the lack of
good-quality information on clinical and cost-effective
ways of cleaning and sterilising infant feeding equipment
in the home, especially under conditions relevant to
families in developed countries in the 21st century. Only
two studies examined clinical outcomes for the babies.
Design flaws identified included lack of randomisation,
inadequate sample size and selection bias. This lack of
high-quality evidence probably explains the variation in
international guidelines that we identified.
The majority of studies identified were conducted in
the 1950s to the 1980s, before the introduction of
microwaves and the widespread use of dishwashers in
the home. It was during the 1950s that there was a steady
improvement in infant mortality rate due to the reduction
in deaths from childhood infectious diseases, including
gastroenteritis
(30,39,40)
. This was in part due to the
recognition of the dangers of contaminated water and the
introduction of chemical sterilisation methods for infant
feeding equipment
(41)
, and forms of sterilisation used
in the 1950s are still recommended. Only one study
(18)
tested more recent approaches (electric and microwave
steam sterilisation) and found both to be more effective
than chemical sterilisation. The contribution of micro-
waves, with the potential for overheating, and dish-
washers, with the possible hazards of salt and detergent
residues and the release of plasticisers, remains to be
examined.
Our preliminary work found that the evidence base for
the current advice given by some infant formula or
feeding equipment manufacturers in the UK is unclear.
Several respondents stated that they based their advice on
Department of Health guidelines, but we were unable to
identify the existence of such guidelines at that time.
Since we conducted this preliminary work, limited
guidelines have been issued by the Food Standards
Agency in the UK
(16)
and by EFSA
(8)
in Europe, in
response primarily to recent concern about E. sakazakii
and Salmonella. The recommendation in those guidelines
to make up each feed as required is likely to have
implications for compliance with complex sterilisation
procedures.
In the absence of a secure evidence base, it is not clear
what advice health professionals should be giving. Such
information as we have from the included papers in this
review, however, suggests that input from health profes-
sionals does have the potential to make a difference,
probably in promoting compliance with whichever
method is used.
We found no studies that examined the views and
experiences of parents and carers about the problems of
cleaning and sterilising in the home environment, or ways
in which the process might be simplified and made more
efficient and effective.
Unanswered questions remain: what are parents and
carers actually doing? Which methods are easier and
achieve best compliance? Which methods achieve the
best clinical outcomes for the babies? Are basic hygiene
measures such as thorough hand-washing, cleaning of
equipment and a clean water supply sufficient?
Conclusion
Bottle feeding carries inherent risks for the baby and is a
potential source of bacterial contamination. There is
scanty of evidence, however, on clinical and cost-effec-
tive ways of cleaning and sterilising infant feeding
equipment, including both bottles and teats, in the home,
and there is a risk that some parents may reject current
methods as cumbersome, time-consuming and expen-
sive. Further, the focus on ‘sterilisation’ procedures could
have a paradoxical effect of reducing basic hygiene pro-
cedures such as hand-washing and thorough washing
and drying of equipment, as it could induce complacency
about the possibility of infection. The current evidence
base provides little information about effectiveness of the
range of old and new methods used, and there is no
evidence that either manufacturers or health policy
makers have identified the problems parents might face
in their own homes. Studies are old, of poor quality and
largely irrelevant to circumstances in the home in the
early 21st century.
Further r esearch
Research is needed to determine which methods of
cleaning or sterilisation of infant feeding equipment are
Cleaning and sterilisation of infant feeding equipment 1197
most effective and efficient, for both bottles and teats.
Epidemiological studies could contribute to a more
comprehensive assessment of the risks of poor practice
and the likely effects of improved procedures in this
important public health issue. Further surveys and qua-
litative studies could provide updated information on
what parents and carers actually do in the home envir-
onment and what the barriers to good practice are. Such
information would inform the design of a randomised
controlled trial, adequately powered, of the range of
methods in common use, measuring clinical and cost
outcomes, as well as the views of parents and carers.
Acknowledgements
None of the authors has any competing interests.
We are grateful for the input of international colleagues
in providing and translating their national guidelines:
Cindy Turner-Maffei, National Coordinator, Baby
Friendly, USA; Julie Stufkens, Executive Officer, New
Zealand Breastfeeding Authority, New Zealand; Pat
Martz, Project Manager, Population Health Strategies
Branch, Alberta Health and Wellness, Canada; Rachel Myr,
Staff breastfeeding specialist midwife, Sørlandet Sykehus,
Kristiansand, Norway; Elisabeth Tufte, Norwegian
Resource Centre for Breastfeeding, Norway; Genevieve
Becker, National Coordinator, Baby Friendly, Ireland.
We also thank the anonymous respondents from manu-
facturing companies who provided information.
References
1. Hamlyn B, Brooker S, Oleinikova K & Wands S (2002)
Infant Feeding 2000. London: The Stationery Office.
2. Bolling K, Grant C, Hamlyn B & Thornton A (2007) Infant
Feeding Survey 2005. London: The Information Centre,
Government Statistical Service, UK Health Departments.
3. UNICEF/World Health Organization (1990). Innocenti
Declaration On the Protection, Support and Promotion of
Breastfeeding. http://www.unicef.org/programme/breast
feeding/innocenti.htm
4. Centers for Disease Control and Prevention (2006) Breast-
feeding Practices – Results from the 2004 National
Immunization Survey. http://www.cdc.gov/breastfeeding/
data/NIS_data/data_2004.htm (accessed October 2006).
5. Xu F, Binns C, Nazi G, Shi L, Zhao Y & Lee A (2006)
A comparison of breastfeeding among Han, Uygur and
other ethnic groups in PR China. BMC Public Health
6, 196.
6. Edmond KM, Zandoh C, Quigley MA, Amenga-Etego S,
Owusu-Agyei S & Kirkwood BR (2006) Delayed breast-
feeding initiation increases risk of neonatal mortality.
Pediatrics 117, 380–386.
7. Quigley MA, Cumberland P, Cowden JM & Rodrigues LC
(2006) How protective is breastfeeding against diarrhoeal
disease in infants in 1990 England? A case–control study.
Arch Dis Child 91, 245–250.
8. European Food Safety Authority (2004) Opinion of the
Scientific Panel on Biological Hazards on a request from
the Commission related to the microbiological risks in
infant formulae and follow-on formulae. EFSA Journal
113, 1–35.
9. Renfrew MJ, Ansell P & Macleod KL (2003) Formula feed
preparation: helping reduce the risks; a systematic review.
Arch Dis Child 88, 855–858.
10. World Health Organization (1981) International Code of
Marketing of Breast Milk Substitutes. Resolution passed by
the World Health Assembly WHA 34.32. Geneva: WHO.
11. Fairbank L, O’Meara S, Renfrew MJ, Woolridge M, Sowden AJ
& Lister-Sharp D (2000) A systematic review to evaluate the
effectiveness of interventions to promote the initiation of
breastfeeding. Health Technol Assess 25, 1–171.
12. Tedstone A, Dunce N, Aviles M, Shetty P & Daniels L (1998)
Effectiveness of Interventions to Promote Healthy Feeding
in Infants under One Year of Age: A Review. Health
Promotion Effectiveness Reviews. London: Health Education
Authority.
13. Renfrew MJ, Dyson L, Wallace L, D’Souza L, McCormick FM
& Spiby H (2005) The Effectiveness of Health Interventions
to Promote the Duration of Breastfeeding: Systematic
Review, 1st ed. London: National Institute for Health and
Clinical Excellence.
14. Thompson J (2002) Benefits of breastfeeding and current
controversies: part two. Community Pract 75, 106–107.
15. Forsythe SJ (2005) Enterobacter sakazakii and other
bacteria in powdered infant milk formula. Matern Child
Nutr 1, 44–50.
16. Food Standards Agency (2006) Guidance on preparing infant
formula February 2006. http://www.food.gov.uk/news/
newsarchive/2005/nov/infantformulastatementnov05
17. World Health Organization/Food and Agriculture Organi-
zation of the United Nations (2007) Safe preparation,
storage and handling of powdered infant formula: Guide-
lines. http://www.who.int/foodsafety/publications/micro/
pif_guidelines.pdf (accessed October 2007).
18. Rowan N & Anderson J (1998) Effectiveness of cleaning
and disinfection procedures on the removal of enterotoxi-
genic Bacillus cereus from infant feeding bottles. J Food
Prot 61, 196–200.
19. Mountfort KA, Kelly J, Jickells SM & Castle L
(1997) Investigation into the potential degradation of
polycarbonate baby bottles during sterilisation with
consequent release of bisphenol A. Food Addit Contam
14
, 737–740.
20. Brede C, P Fjeldal I, Skjevrak I & Herikstad H (2003)
Increased migration levels of bisphenol A from polycarbo-
nate bottles after dishwashing, boiling and brushing. Food
Addit Contam 20, 684–689.
21. Cairney PA & Alder EM (2001) A survey of information
given by health professional s, about bottle feeding, to first
time mothers in a Scottish population. Health Bull 59,
97–101.
22. Department of Health (2006) Birth to Five. London: COI.
23. Health Promotion Unit (2004) Food and nutrition guide-
lines for pre-school services. http://www.healthpromotion.
ie/publications/index.php?categoryID55 (accessed March
2007).
24. UNICEF (undated) Sterilising baby feeding equipment. http://
www.babyfriendly.org.uk/page.asp?page5115&category54
(accessed March 2007).
25. Ministry of Health (2000) Food and Nutrition Guidelines
for Healthy Infants and Toddlers (Aged 0–2 Years):
A Background Paper. http://www.moh.govt.nz/moh.nsf/
pagesmh/1274/$File/fnghit2.pdf (accessed October
2006).
26. Sosial-og helsedirectoratet (2005) Morsmelkerstatning. http://
www.shdir.no/ernaering/spedbarnser n_r ing/morsmelker
statning/ (accessed October 2006).
27. US Department of Agriculture (1994) USDA Infant Nutri-
tion and Feeding. A Reference Handbook for Nutrition and
1198 MJ Renfrew et al.
Health Counselors in the WIC and CSF Programs. Beltsville
MD: USDA Food and Nutrition Service.
28. Younger-Lewis C (editor) (2001) Canadian Medical
Association Complete Book of Mother and Baby Care.
Toronto: Dorling Kindersley Ltd.
29. Khan K, ter Reit G, Glanville J, Sowden AJ & Klieijnen J
(2001) Undertaking Systematic Reviews of Research on
Effectiveness, 2nd ed. Centre for Reviews and Dissemina-
tion Report no. 4. York: University of York.
30. Anderson J & Gatherer A (1970) Hygiene of infant-feeding
utensils. Practices and standards in the home. Br Med J 2, 20–23.
31. Clegg A, Duke A & Prosser-Snelling K (1977) Standards of
cleansing and sterilization of infant feeding utensils in the
home. J Int Med Res 5, 473–475.
32. Gatherer A (1978) A review of standards of infant hygiene
in the home. Nurs Times 74, 1684–1685.
33. Hargrove C, Temple A & Chinn P (1974) Formula
preparation and infant illness. A comparison between
‘clean preparation’ and terminal sterilization with neonates.
Clin Pediatr (Phila) 13, 1057–1059.
34. Hughes R, Sauvain K, Blanton L & DeLoache W (1987)
Outcome of teaching clean vs. terminal methods of formula
preparation. Pediatr Nurs 13, 275–276.
35. Jacob F (1985) Infant care; getting it right. Community
Outlook Dec, 20–21.
36. Vaughan V 3rd, Dienst R, Sheffield C & Roberts R (1962)
A study of techniques of preparation of formulas for infant
feeding. J Pediatr 61, 547–555.
37. Gatherer A & Wood N (1966) Home standards of
sterilization of infant feeding bottles and teats. Mon Bull
Minist Health Public Health Lab Serv 25, 126–130.
38. Wharton B & Berger H (1976) Problems of childhood.
Bottle-feeding. Br Med J 1, 1326–1331.
39. Lucas A & Cole T (1990) Breast milk and neonatal
necrotising enterocolitis. Lancet 336, 1519–1523.
40. Committee on Fetus and Newborn, American Academy of
Pediatrics (1961) Sterilization of milk mixtures for infants.
Pediatrics 28, 674–675.
41. Gerber M, Berlinger B & Karolus J (1983) Sterilization of
infant formula. Clin Pediatr (Phila) 22, 344–349.
Cleaning and sterilisation of infant feeding equipment 1199
