Semin Neonatol 2002; 7: 325–333
doi:10.1053/siny.2002.9125, available online at http://www.idealibrary.com on
Preventing nosocomial bloodstream infection in very
low birth weight infants
William H. Edwards
Department of Pediatrics,
Children’s Hospital at Dartmouth,
One Medical Center Drive, Lebanon,
New Hampshire 03756, USA
Nosocomial sepsis is a frequent complication of caring for very low birth weight
infants and incidence varies substantially among centres. Many cases are preventable.
An organized approach to understanding the epidemiology of nosocomial sepsis within
a unit, and implementing evidence-based practices can successfully reduce the incidence.
Diagnostic accuracy is important to limit excess empiric antibiotic therapy. Instituting a
hand hygiene program of education, monitoring, and consideration of waterless hand
disinfectants to avoid hand transmission of organisms is essential. An emphasis on
early achievement of enteral nutrition, preferably with human milk is important to
reduce unnecessary exposure to central catheters and parenteral nutrition. Use of
maximum sterile barrier precautions by personnel trained and skilled in central catheter
insertion, followed by meticulous care in preventing catheter hub contamination will
reduce the incidence of catheter related sepsis. Ultimately, the culture of the NICU
needs to shift from a focus on early detection of infection to one of prevention.
? 2002 Elsevier Science Ltd. All rights reserved.
Key words: infant, very low birth
weight; cross infection; sepsis; infant,
premature, diseases, infant, premature;
intensive care, neonatal; total quality
management; quality assurance,
Survival for very low birth weight (VLBW) infants
has steadily improved. From 1988 to 1996, the
National Institute of Child Health and Human
Development (NICHD) neonatal network reported
increased survival from 74% to 84% . Increased
morbidity has accompanied better survival rates,
however. Nosocomial infection is frequent, and is
not only a risk for mortality, but also adds to
length of hospital stay and costs of care [2–4].
Multiple studies have identified risk factors associ-
ated with nosocomial bloodstream infections.
Incidence of nosocomial sepsis is inversely related
to birth weight and gestational age, higher in
males, and directly correlates with severity of
illness scores, ventilator days, length of stay, corti-
costeroid use, use of central catheters, and
parenteral nutrition, especially intravenous lipids
Comparing incidence rates for nosocomial infec-
tions in neonatal intensive care units is made
more difficult by lack of standard definitions. The
Center for Disease Control’s National Nosocomial
Infections Surveillance (NNIS) system defines
laboratory-confirmed bloodstream infection as the
recovery of a recognized pathogen from one or
more blood cultures. Classification of common
skin contaminants (e.g. diptheroids, Bacillus sp.,
Propionibacterium sp., coagulase-negative staphylo-
cocci, or micrococci) recovered from culture as a
bloodstream infection requires symptoms (fever
>38?C, hypothermia <37?C,* apnea, or bradycar-
dia) and either two or more separate positive
blood cultures or at least one positive blood culture
from a patient with an intravascular line where
the physician institutes appropriate antimicrobial
therapy. Peripartum infections resulting from expo-
sure to maternal flora, such as those due to Group
B streptococci are considered to be nosocomial
*As defined in original publication.
Correspondence to: William H. Edwards, MD, Department of Pediatrics,
Children’s Hospital at Dartmouth, One Medical Center Drive, Lebanon,
New Hampshire 03756, USA. Tel.: 603 650-5828; Fax: 603 650-5458;
1084–2756/02/$-see front matter © 2002 Elsevier Science Ltd. All rights reserved.
infections, although separately designated as
‘maternally acquired’ [5,11]. The NICHD neonatal
research network defines sepsis as positive results
from one or more blood cultures, in the presence of
clinical signs or symptoms suggestive of infection,
and antibiotic treatment for 5 or more days. They
further designate sepsis as early onset (positive
culture obtained at <72 h age) or late onset .
The Vermont Oxford Neonatal Network database
definition for sepsis and/or meningitis is recovery
of a bacterial pathogen (from a pathogen list)
from a blood and/or cerebrospinal fluid culture.
Coagulase-negative staphylococcal sepsis requires
a positive blood culture or cerebrospinal fluid
culture and signs of generalized infection (such as
apnea, temperature instability, feeding intolerance,
worsening respiratory distress or hemodynamic
instability) and treatment for 5 or more days with
intravenous antibiotics. Early onset is defined as
positive cultures obtained on or before day of life
3 . Regardless of definition, there is great
variation among centres in the incidence of
bloodstream infections. The center-to-center varia-
bility in the incidence of late-onset sepsis in
VLBW infants at 12 NICHD centres ranged from
11.5% to 32.4% . The median incidence for late
bacterial sepsis for 352 participating centres in the
Vermont Oxford Network for the year 2000 was
21% (interquartile range 12% to 28%). The vari-
ation was even greater for babies with birth
weights of 501–750 grams (median 42%; inter-
quartile range 25–56%) . As with many out-
comes of VLBW infants, there is strong evidence
that variation among centres in the incidence of
nosocomial sepsis persists after adjusting for
known risk factors and severity of illness, sug-
gesting that some neonatal intensive care units
are more successful preventing nosocomial sepsis
than others . A systematic approach using
quality improvement techniques can successfully
[12,14,15]. This review will detail a variety of
strategies for preventing nosocomial sepsis in
VLBW infants along with the evidence support-
ing them. For this review, nosocomial sepsis
will exclude maternally acquired, or early-onset
sepsis. Other important nosocomial infection
problems such as meningitis, pneumonia, necrotiz-
ing enterocolitis, urinary tract infections and
addressed, although many of the principles dis-
cussed will apply to prevention of these problems
A schematic framework for understanding how
prevention strategies may work is shown in Figure
1. In some cases of nosocomial sepsis, a virulent
organism may gain direct access to the blood-
stream, such as by infusing a contaminated intra-
venous solution . However, in most cases
bacteraemia is preceded by a variable period of
colonization without signs of infection. Organisms
responsible for nosocomial sepsis are typically
those recovered from the skin and gastrointestinal
tract of non-infected infants, and the distribution
of organisms is remarkably consistent among
reports [3–6]. Gram-positive organisms account for
70–75% of cases, with over 50% caused by
coagulase-negative staphylococci. Other frequent
gram-positive organisms are coagulase-positive
staphylococci and enterococci. There is particular
concern that the incidence of enterococcal infec-
tions may be increasing, as well as resistance to
vancomycin [17–20]. Gram-negative organisms
most commonly causing nosocomial sepsis are
coliform organisms (Escherichia coli, Klebsiella sp. and
Enterobacter sp.) which colonize the gastrointestinal
W. H. Edwards
tract. Infections due to Pseudomonas sp. are rela-
tively less common, but may be difficult to control
due to persistence in reservoirs in the environment
[21,22]. Fungal organisms are commensal organ-
isms that colonize the skin and gastrointestinal
tract. Infections are related to disease severity and
presence of invasive therapies, and prolonged
exposure to antibiotics [23,24]. The incidence of
fungal sepsis in VLBW infants is 2–3% [3,13], but
has been reported to be as high as 20% for infants
weighing less than 1000 grams at birth . It is
difficult to dissociate the morbidity caused by
nosocomial infection from the underlying disease
severity associated with problems of extreme pre-
maturity. However, mortality rates associated with
gram-negative and fungal sepsis are much higher
than for gram-positive organisms. Case fatality
rate in the NICHD neonatal research network for
infections with all gram-positive organisms was
10.1%, compared to 39.6% for all gram-negatives
and 28.1% for fungi. Mortality associated with
Pseudomonas sepsis was 61.8% . A clear under-
standing of the profile of organisms causing noso-
comial sepsis in a neonatal intensive care nursery is
vital to prioritizing strategies for prevention.
Making an accurate diagnosis of sepsis in VLBW
infants is sometimes difficult. Some episodes of
sepsis have low quantitative counts of organisms in
the blood, although the actual incidence of low
colony count sepsis by organism type is not known
. The relationship between culture volume and
detection of sepsis in adults is clearly established
. Concern about excessive blood losses from
testing and technical difficulties obtaining sufficient
blood volume from arterial or venous punctures
often result in small volumes of blood for culture.
A reasonable recommendation is to obtain a mini-
mum of 1 ml of blood for culture . Positive
cultures obtained from indwelling arterial or
venous catheters may reflect colonization of the
catheter or the hub rather than bloodstream infec-
tion. To improve diagnostic accuracy, it is often
recommended to obtain two cultures—one from
the indwelling catheter and a second from direct
arterial or venous puncture. Concordance between
the two cultures, either positive or negative,
increases confidence in diagnosing or ruling out
sepsis. Discordant results from two cultures are
most often interpreted as the positive culture being
a contaminant . An alternative interpretation
could be that in low colony count sepsis the
detecting a true infection.
Limiting unnecessary empiric antibiotic therapy
should be a major motivation to accurately diag-
nose nosocomial sepsis. Frequent suspicion of sep-
sis along with the often subtle presenting clinical
signs result in liberal use of broad-spectrum anti-
biotics. Antibiotic use then increases the risk of
opportunistic infections in individual patients and
the risk of developing antibiotic resistance in
organisms over time [23,24,30,31]. Widespread
vancomycin use to empirically cover coagulase-
negative staphylococci may contribute to infec-
tions from more virulent gram-negative organisms
and the emergence of vancomycin resistant entero-
cocci [18–20,32]. Improvements in blood culture
systems with automated detection has reduced the
time to positive results . Empiric antibiotic
therapy pending culture results should be no more
than 48 h when such a system is in place.
One of the most important strategies for prevent-
ing nosocomial sepsis, at the same time most
difficult to study and quantify, is the role of unit
culture. During a quality improvement project to
reduce nosocomial sepsis rates in VLBW infants,
site visits were made to two NICUs with the
lowest incidence of nosocomial sepsis in the
Vermont Oxford Network . Two attributes of
these benchmark units were apparent: the staffs of
the units were both aware and proud of their low
nosocomial sepsis rate, and they believed noso-
comial infection was preventable. When an infant
developed an infection, it was considered a break-
down in care. Two conceptual models for a unit’s
beliefs about nosocomial sepsis are proposed in
Table 1, a model of entitlement, and one of
prevention. In the entitlement model, there is a
fatalistic belief that nosocomial sepsis is inevitable
due to factors beyond the control of the care
team. VLBW infants have inherent deficiencies in
immunity, and risks associated with necessary and
life-saving care, such as invasive lines, parenteral
nutrition and ventilator support are unavoidable.
The focus of the care team is on detecting early
signs of infection and starting treatment early. The
Preventing nosocomial bloodstream infection in very low birth weight infants
team may take pride in its ability to detect subtle
presenting clinical signs, and sepsis work-ups and
empiric courses of antibiotics pending culture
results are frequent. Ultimately, no connection is
made between a lapse in ideal care, for example
leading to colonization and infection of a central
line, and the event of sepsis. Lack of ownership is
further hampered by the interval between the care
failure and the downstream event, making it diffi-
cult to even suspect such a connection. In contrast
to the entitlement model, in the prevention model
the inherent risk due to impaired immunity is also
acknowledged, but the focus is on developing care
practices that minimize the risk. An episode of
nosocomial sepsis prompts a review to determine
whether any events or breaks in ideal care practices
might have contributed. Most important, belief
that nosocomial sepsis is preventable leads to
motivation to improve. Because nosocomial infec-
tions are multifactorial in origin (Fig. 1), it is
particularly important for all team members to
understand the unit’s goal of nosocomial infection
prevention, and their role in adhering to ideal care
practices and encouraging creativity in proposing
potential improvements in care.
Endemic strains of bacteria or fungi may persist in
NICUs over months to even decades, and are
transmitted to new patients by hand transmission
[34–36]. The importance of hand transmission
in nosocomial infections has been emphasized in
multiple recent reviews, and many guidelines exist
[37–40]. Evidence-based reviews of hand hygiene
have been recently published [41,42]. Health care
providers generally acknowledge the importance
of handwashing, but overestimate their own
compliance . If it is clearly important, but care
providers fail to do it, is there any hope that hand
hygiene can be improved? Several studies hold
promise. Better accessibility of sinks is associated
with improved handwashing, while installation of
automated sinks have little lasting benefit [44,45].
Waterless hand rub sanitizers have been shown to
be effective in decontaminating hands and are
generally well tolerated . Introducing these
products has led to improved hand decontamina-
tion in some, but not all trials [47–50]. Use of
waterless hand rubs may reduce time required for
traditional handwashing, making compliance more
likely and cost-effective . A variety of educa-
tional strategies have improved hand disinfection,
but without ongoing monitoring, the effect is
transient [45,52]. An innovative approach has been
taken of educating patients about the importance of
handwashing, and asking that they remind their
caregivers to wash. For adult medical and surgical
patients, compliance with handwashing was in-
creased by 34% and 50% in two studies [53,54]. All
patients asked nurses to wash hands, but only 35%
asked physicians . The possibility of including
parents as partners ensuring compliance with hand
decontamination guidelines in NICUs should be
Protecting the hands against irritation and break-
down is important. A small study showed that
nurses with hand irritation had greater numbers of
colonizing species . Recently, artificial nails
have been implicated as potential risk factors for
transmission of gram-negative bacteria, particularly
Pseudomonas aeruginosa, and should be prohibited
from use in the NICU [21,56,57].
Besides the widely held belief that better nu-
trition enhances immune function and helps resist
Table 1. Two conceptual models for nosocomial sepsis in VLBW infants
CausalityUnavoidable – inherent risk from
poor immune function and necessary
Chance or unavoidable
Preventable in most
Focus of care team
Why did it happen?
Motivation for improvement
Breakdown in ideal care
Challenge to continually
W. H. Edwards
infection, nutrition care practices may relate to
nosocomial infection risk in other ways. Noso-
comial sepsis incidence correlates with parenteral
nutrition, especially intravenous lipids, as well as
with the central lines often placed primarily for
giving parenteral nutrition [3,4,6–9]. Although it is
reasonable to assume that limiting exposure to
central lines and parenteral nutrition by earlier
establishment of enteral nutrition might reduce the
incidence of nosocomial sepsis, there are no specific
trials with that primary hypothesis. There is evi-
dence that early initiation of enteral feedings while
an umbilical artery catheter is in situ is safe, and
results in fewer days on parenteral nutrition . In
a systematic review of minimal enteral nutrition,
or ‘trophic’ feedings, defined as providing <=25
kcal/kg/day for >=5 days, infants in the study
groups reached full enteral feeds sooner, with fewer
days total days feedings were held. The incidence
of necrotizing enterocolitis was not significantly
different among groups. Nosocomial sepsis was not
a reported outcome .
The possibility that human milk feedings reduce
risk of nosocomial sepsis has been suggested in
several small studies and is reviewed more exten-
sively in this issue by Hanson [60–62]. Feeding
studies with pre-term mothers’ breast milk com-
pared to premature infant formula are difficult to
conduct due to the lack of random allocation, lack
of ability to blind care providers, and need to
supplement breast milk with formula when supply
is inadequate. el-Mohandes and co-workers consid-
ered infants to be human milk-fed if they received
only human milk as an enteral nutrient for >=1
week, or if human milk accounted for at least 40%
of their total enteral caloric intake. By that defini-
tion, 59 were fed human milk and 114 formula. The
odds ratio for sepsis in the human milk-fed group
compared with formula was 0.38 (95% confidence
interval 0.15–0.95, P=0.04). In the Schanler study,
the human milk group was defined as receiving an
average of >=50 ml/kg/day of human milk during
hospitalization. There were 62 infants in the human
milk group and 46 in the formula group. The
incidence of sepsis was not significantly different
(31% vs 48%, P=0.07), but the combined incidence
of sepsis and/or necrotizing enterocolitis was less in
the human milk group compared to the formula
group (31% vs 54%, P<0.01). Hylander and
co-workers compared VLBW infants receiving
any human milk (n=123) to those receiving only
formula (n=89). The human milk-fed infants
had a lower incidence of nosocomial sepsis or
meningitis than those fed formula (19.5% vs 32.6%,
Glutamine supplementation has been suggested
to potentially reduce nosocomial sepsis incidence in
VLBW infants. Neither a systematic review nor a
preliminary report of a large multicenter random-
ized clinical trial demonstrated any benefit of
glutamine supplementation in reducing nosocomial
A correlation between nosocomial sepsis in VLBW
infants and centrally placed lines has been reported
in numerous studies [5,7,9,65]. Indwelling catheters
are a particular factor for infection with coagulase-
negative staphylococci [36,66]. Since the introduc-
tion of percutaneously inserted central catheters,
the use of surgically placed catheters has decreased
in most NICUs. Practical issues such as relative
benefits of umbilical versus percutaneously placed
central lines, when to remove umbilical lines and
clear risk-benefit analysis of indications for placing
central lines need further study. Excellent reviews
of the evidence base for preventing central line
infections have been recently published [42,67].
Although these reviews analyse more adult studies,
many of the recommendations are relevant to
preventing catheter-related bloodstream infections
Strategies related to the insertion of central lines
include using maximal barrier precautions and
aseptic technique during placement and developing
a limited team of skilled persons to consistently
perform the procedure. In a randomized clinical trial
of 343 adult oncology patients, catheter-related
infections were 6.3 times more likely when only
sterile gloves and small drapes were used compared
with maximal sterile barrier precautions during
catheter insertion . Education of physicians-in-
training in catheter insertion also reduced primary
bloodstream and catheter-related infections by
28% (P<0.01) in adult ICUs . Using maximal
barrier precautions in a before and after analysis of
a quality improvement prevention program in
neonates was also associated with a decrease in
central venous catheter bacteraemias (RR 0.27, 95%
CI 0.15–0.51; P<0.001) .
The portals of entry for organisms causing
catheter-related sepsis are the skin insertion site
and the catheter hub. The importance of contami-
nation of the hub in the sequence of catheter
Preventing nosocomial bloodstream infection in very low birth weight infants
colonization leading to infection has been described
in several adult and neonatal studies [70–72].
Saltzman and co-workers studied 113 catheters in
88 neonates. In 10 of 28 episodes of catheter-
related sepsis, the organism isolated from blood
was isolated from a culture of the catheter hub
prior to the onset of clinical sepsis, and in 5
additional episodes, simultaneously. The common
sense preventive measure of limiting the number of
connecting hubs is suggested by a prospective
catheter-related sepsis were more common with
triple lumen than with single lumen catheters .
Carefully devised techniques of maintaining hub
sterility during line changes and active decontami-
nation of the catheter hub may be effective in
limiting hub colonization [15,74]. Saltzman and
co-workers developed an in vitro model to study
hub disinfection and emphasized the importance of
mechanical friction, which was 99% effective in hub
decontamination. The use of 70% alcohol com-
pleted the disinfection . There is promise that
with better understanding of the epidemiology
of catheter colonization, new designs may be
effective in reducing catheter-related sepsis [76,77].
Antibiotic impregnated catheters have strong evi-
dence for efficacy in preventing catheter-related
sepsis in adults, and research in developing similar
materials for use in neonates should be encouraged
Since by far the most common organism
associated with neonatal catheter-related sepsis is
coagulase-negative staphylococcus, a number of
clinical trials have been conducted using vancomy-
cin prophylaxis, and the practice has been the
subject of a critical review . Although effective
in reducing the incidence of coagulase-negative
staphylococcal sepsis, the lack of proven benefit
for other important clinical outcomes of mortality
and length of stay along with concern for the
development of vancomycin resistance in other
organisms are reasons for caution in recommending
widespread use of this strategy [20,78].
where episodes of
The skin of the premature infant is exceedingly
fragile, particularly in the first weeks after birth.
Multiple care practices are associated with skin
breakdown, including epidermal stripping from
tapes and adhesives, abrasions from rough ma-
terials and skin punctures for blood sampling.
Nopper and Lane in a study designed to measure
the effects of an emollient ointment on insensible
water loss, found an unexpected reduction in the
incidence of nosocomial sepsis in the group treated
with twice a day application of the ointment for
two weeks . A randomized clinical trial of
infants weighing 501–1000 grams at birth of simi-
lar design failed to confirm a benefit of reduced
nosocomial sepsis. The group receiving prophy-
lactic application of emollient ointment actually
had a higher incidence of nosocomial infection .
Additional reports have associated ointments with
infection. Campbell and co-workers in a case-
control study demonstrated a potential relationship
between the topical application of petrolatum oint-
ment and systemic candidiasis in infants weighing
<1500 grams . Ramsey and co-workers re-
ported contamination of Aquaphor?as a potential
source of nosocomial infection . Currently the
practice of prophylactic application of emollients
for an extended time to prevent infection cannot be
The incidence of nosocomial sepsis in VLBW
infants varies greatly among NICUs. Comparative
data from similar centres using the same definitions
are helpful in identifying problems and measur-
ing results of improvement initiatives. Quality
improvement methods are effective in reducing the
incidence of nosocomial sepsis [12,14,15]. Reasons
for developing nosocomial sepsis are multifactorial,
and may be unit specific. Specific targets for
improvement should be based on understanding
the epidemiology of nosocomial sepsis within the
unit’s own environment. Many practices with
strong evidence for efficacy are incompletely
implemented in most NICUs. Although practices of
using prophylactic antibiotics may be effective,
they should not be substitutes for a high quality
1 Lemons JA, Bauer CR, Oh W, et al. Very low birth weight
outcomes of the National Institute of Child health and
human development neonatal research network, January
1995 through December
Research Network. Pediatrics 2001; 107(1): E1.
W. H. Edwards
2 Gray JE, Richardson DK, McCormick MC, et al.
Coagulase-negative staphylococcal bacteremia among
very low birth weight infants: relation to admission illness
severity, resource use, and outcome. Pediatrics 1995;
3 Stoll BJ, Gordon T, Korones SB, et al. Late-onset
sepsis in very low birth weight neonates: a report
from the National Institute of Child Health and Human
Development Neonatal Research Network. The Journal of
Pediatrics 1996; 129(1): 63–71.
4 Makhoul IR, Sujov P, Smolkin T, et al. Epidemiological,
clinical, and microbiological characteristics of late-onset
sepsis among very low birth weight infants in Israel: a
national survey. Pediatrics 2002; 109(1): 34–39.
5 Gaynes RP, Edwards JR, Jarvis WR, et al. Nosocomial
infections among neonates in high-risk nurseries in
the United States. National Nosocomial Infections
Surveillance System. Pediatrics 1996; 98(3 Pt 1): 357–
6 Fanaroff AA, Korones SB, Wright LL, et al. Incidence,
presenting features, risk factors and significance of
late onset septicemia in very low birth weight infants.
The National Institute of Child Health and Human
Development Neonatal Research Network. The Pediatric
Infectious Disease Journal 1998; 17(7): 593–598.
7 Brodie SB, Sands KE, Gray JE, et al. Occurrence of
nosocomial bloodstream infections in six neonatal inten-
sive care units. The Pediatric Infectious Disease Journal 2000;
8 Freeman J, Platt R, Epstein MF, et al. Birth weight and
length of stay as determinants of nosocomial coagulase-
negative staphylococcal bacteremia in neonatal intensive
care unit populations: potential for confounding. American
Journal of Epidemiology 1990; 132(6): 1130–1140.
9 Avila-Figueroa C, Goldmann DA, Richardson DK, et al.
Intravenous lipid emulsions are the major determinant of
coagulase- negative staphylococcal bacteremia in very
low birth weight newborns. The Pediatric Infectious Disease
Journal 1998; 17(1): 10–17.
10 Stoll BJ, Temprosa M, Tyson JE, et al. Dexamethasone
therapy increases infection in very low birth weight
infants. Pediatrics 1999; 104(5): e63.
11 Gaynes RP, Horan TC. Surveillance of Nosocomial
Infections. In: Mayhall CG (ed.) Hospital Epidemiology and
Infections Control. Baltimore, MD: Williams & Wilkins,
12 Horbar JD, Rogowski J, Plsek PE, et al. Collaborative
quality improvement for neonatal intensive care. Pediatrics
2001; 107: 14–22.
13 Horbar JD (ed.). Vermont Oxford Network 2000 Database
Summary. Burlington, VT: Vermont Oxford Network,
14 Eggimann P, Harbarth S, Constantin MN, et al. Impact of
a prevention strategy targeted at vascular-access care on
incidence of infections acquired in intensive care. Lancet
2000; 355(9218): 1864–1868.
15 Maas A, Flament P, Pardou A, et al. Central venous
catheter-related bacteraemia in critically ill neonates: risk
factors and impact of a prevention programme. The
Journal of Hospital Infection 1998; 40(3): 211–224.
16 Clinical sepsis and death in a newborn nursery associated
with contaminated parenteral medications—Brazil, 1996.
MMWR. Morbidity and Mortality Weekly Report 1998;
17 Sohn AH, Garrett DO, Sinkowitz-Cochran RL, et al.
Prevalence of nosocomial infections in neonatal intensive
care unit patients: Results from the first national point-
prevalence survey. The Journal of Pediatrics 2001; 139(6):
18 McNeeley DF, Saint-Louis F, Noel GJ. Neonatal entero-
coccal bacteremia: an increasingly frequent event with
potentially untreatable pathogens. The Pediatric Infectious
Disease Journal 1996; 15(9): 800–805.
19 Lee HK, Lee WG, Cho SR. Clinical and molecular
vancomycin-resistant enterococci in a neonatal intensive
care unit. Acta Paediatrica 1999; 88(6): 651–654.
20 Recommendations for preventing the spread of vanco-
mycin resistance. Hospital Infection Control Practices
Advisory Committee (HICPAC). Infection Control and
Hospital Epidemiology 1995; 16(2): 105–113.
21 Foca M, Jakob K, Whittier S, et al. Endemic Pseudomonas
aeruginosa infection in a neonatal intensive care unit. The
New England Journal of Medicine 2000; 343(10): 695–700.
22 Grundmann H, Kropec A, Hartung D, et al. Pseudomonas
aeruginosa in a neonatal intensive care unit: reservoirs
and ecology of the nosocomial pathogen. The Journal of
Infectious Diseases 1993; 168(4): 943–947.
23 Makhoul IR, Kassis I, Smolkin T, et al. Review
of 49 neonates with acquired fungal sepsis: further
characterization. Pediatrics 2001; 107(1): 61–66.
24 Benjamin DK Jr, Ross K, McKinney RE Jr, et al. When
to suspect fungal infection in neonates: A clinical
comparison of Candida albicans and Candida parapsi-
losis fungemia with coagulase-negative staphylococcal
bacteremia. Pediatrics 2000; 106(4): 712–718.
25 Kaufman D, Boyle R, Hazen KC, et al. Fluconazole
prophylaxis against fungal colonization and infection in
preterm infants. The New England Journal of Medicine 2001;
26 Dietzman DE, Fischer GW, Schoenknecht FD. Neonatal
Escherichia coli septicemia--bacterial counts in blood. The
Journal of Pediatrics 1974; 85(1): 128–130.
27 Ilstrup DM, Washington II JA. The importance of volume
of blood cultured in the detection of bacteremia and
fungemia. Diagnostic Microbiology and Infectious Disease
1983; 1(2): 107–110.
28 Schelonka RL, Chai MK, Yoder BA, et al. Volume of blood
required to detect common neonatal pathogens. The
Journal of Pediatrics 1996; 129(2): 275–278.
29 Schifman RB, Strand CL, Meier FA, et al. Blood culture
contamination: a College of American Pathologists
Q-Probes study involving 640 institutions and 497134
specimens from adult patients. Archives of Pathology and
Laboratory Medicine 1998; 122(3): 216–221.
30 Van Houten MA, Uiterwaal CS, Heesen GJ, et al. Does
the empiric use of vancomycin in pediatrics increase the
risk for Gram-negative bacteremia? The Pediatric Infectious
Disease Journal 2001; 20(2): 171–177.
31 de Man P, Verhoeven BA, Verbrugh HA, et al. An
antibiotic policy to prevent emergence of resistant bacilli.
Lancet 2000; 355(9208): 973–978.
32 Karlowicz MG, Buescher ES, Surka AE. Fulminant late-
onset sepsis in a neonatal intensive care unit, 1988–1997,
Preventing nosocomial bloodstream infection in very low birth weight infants
and the impact of avoiding empiric vancomycin therapy.
Pediatrics 2000; 106(6): 1387–1390.
33 Garcia-Prats JA, Cooper TR, Schneider VF, et al. Rapid
detection of microorganisms in blood cultures of newborn
infants utilizing an automated blood culture system.
Pediatrics 2000; 105(3 Pt 1): 523–527.
34 Newport MT, John JF, Michel YM, et al. Endemic
Serratia marcescens infection in a neonatal intensive care
nursery associated with gastrointestinal colonization.
Pediatric Infectious Disease 1985; 4(2): 160–167.
35 Huebner J, Pier GB, Maslow JN, et al. Endemic nosocomial
transmission of Staphylococcus epidermidis bacteremia
isolates in a neonatal intensive care unit over 10 years.
The Journal of Infectious Diseases 1994; 169(3): 526–531.
36 Huebner J, Goldmann DA. Coagulase-negative staphylo-
cocci: role as pathogens. Annual Review of Medicine 1999;
37 Aiello AE, Larson EL. What is the evidence for a causal
link between hygiene and infections? The Lancet Infectious
Diseases 2002; 2(2): 103–110.
38 Garner JS, Favero MS. CDC guidelines for the prevention
and control of nosocomial infections. Guideline for
handwashing and hospital environmental control, 1985.
Supersedes guideline for hospital environmental control
published in 1981. American Journal of Infection Control
1986; 14(3): 110–129.
39 Larson EL. APIC guideline for handwashing and hand
antisepsis in health care settings. American Journal of
Infection Control 1995; 23(4): 251–269.
40 Ward V, Wilson J, Taylor L, et al. Preventing Hospital-
Acquired Infections: Clinical Guidelines. London: Public
Health Laboratory Service, 1997.
41 Pratt RJ, Pellowe C, Loveday HP, et al. The epic project:
developing national evidence-based guidelines for pre-
venting healthcare associated
Guidelines for preventing hospital-acquired infections.
Department of Health (England). The Journal of Hospital
Infection 2001; 47(Suppl): S3–82.
42 Shojania KG, Duncan BW, McDonald KM, et al (eds).
Making Health Care Safer: A Critical Analysis of Patient
Safety Practices. Evidence Report/Technology Assessment No.
43 (Prepared by the University of California at San Francisco–
Stanford Evidence-based Practice Center under Contract
Rockville, MD: Agency for Healthcare Research in
43 Harris AD, Samore MH, Nafziger R, et al. A survey on
handwashing practices and opinions of healthcare
workers. The Journal of Hospital Infection 2000; 45(4): 318–
44 Kaplan LM, McGuckin M. Increasing handwashing com-
pliance with more accessible sinks. Infection Control 1986;
45 Larson EL, Bryan JL, Adler LM, et al. A multifaceted
approach to changing handwashing behavior. American
Journal of Infection Control 1997; 25(1): 3–10.
46 Larson E, Silberger M, Jakob K, et al. Assessment of
alternative hand hygiene regimens to improve skin health
among neonatal intensive care unit nurses. Heart and Lung:
The Journal of Critical Care 2000; 29(2): 136–142.
47 Muto CA, Sistrom MG, Farr BM. Hand hygiene rates
unaffected by installation of dispensers of a rapidly acting
hand antiseptic. American Journal of Infection Control 2000;
48 BischoffWE, Reynolds
Handwashing compliance by health care workers: The
impact of introducing an accessible, alcohol-based hand
antiseptic. Archives of Internal Medicine 2000; 160(7):
49 Maury E, Alzieu M, Baudel JL, et al. Availability of an
alcohol solution can improve hand disinfection com-
pliance in an intensive care unit. American Journal of
Respiratory and Critical Care Medicine 2000; 162(1): 324–
50 Pittet D, Hugonnet S, Harbarth S, et al. Effectiveness of a
hospital-wide programme to improve compliance with
hand hygiene. Infection Control Programme. Lancet 2000;
51 Voss A, Widmer AF. No time for handwashing!?
Handwashing versus alcoholic rub: can we afford 100%
compliance? Infection Control and Hospital Epidemiology
1997; 18(3): 205–208.
52 Tibballs J. Teaching hospital medical staff to handwash.
The Medical Journal of Australia 1996; 164(7): 395–398.
53 McGuckin M, Waterman R, Porten L, et al. Patient
education model for increasing handwashing compliance.
American Journal of Infection Control 1999; 27(4): 309–
54 McGuckin M, Waterman R, Storr IJ, et al. Evaluation
of a patient-empowering hand hygiene programme in
the UK. The Journal of Hospital Infection 2001; 48(3):
55 Larson EL, Hughes CA, Pyrek JD, et al. Changes in
bacterial flora associated with skin damage on hands of
health care personnel. American Journal of Infection Control
1998; 26(5): 513–521.
56 Pottinger J, Burns S, Manske C. Bacterial carriage by
artificial versus natural nails. Abstracts-On-Line: The Society
for Healthcare Epidemiology of America 1989; 17(6): 340–
57 Moolenaar RL, Crutcher JM, San Joaquin VH, et al. A
prolonged outbreak of Pseudomonas aeruginosa in a
neonatal intensive care unit: did staff fingernails play a
role in disease transmission? Infection Control and Hospital
Epidemiology 2000; 21(2): 80–85.
58 Davey AM, Wagner CL, Cox C, et al. Feeding premature
infants while low umbilical artery catheters are in place: a
prospective, randomized trial. The Journal of Pediatrics
1994; 124(5 Pt 1): 795–799.
59 Tyson JE, Kennedy KA. Minimal enteral nutrition for
promoting feeding tolerance and preventing morbidity in
parenterally fed infants. Cochrane Database of Systematic
Reviews 2000(2); CD000504.
60 el-Mohandes AE, Picard MB, Simmens SJ, et al. Use of
human milk in the intensive care nursery decreases the
incidence of nosocomial sepsis. Journal of Perinatology
1997; 17(2): 130–134.
61 Schanler RJ, Shulman RJ, Lau C. Feeding strategies for
premature infants: beneficial outcomes of feeding fortified
human milk versus preterm formula. Pediatrics 1999;
103(6 Pt 1): 1150–1157.
62 Hylander MA, Strobino DM, Dhanireddy R. Human milk
feedings and infection among very low birth weight
infants. Pediatrics 1998; 102(3): E38.
W. H. Edwards
63 Poindexter BB, Ehrenkranz RA, Stoll BJ, et al. Parenteral
glutamine supplementation in ELBW infants: A multi-
center randomized clinical trial. Pediatric Research 2002;
51(4, Part 2): 317A.
64 Tubman TR, Thompson SW. Glutamine supplementation
for prevention of morbidity in preterm infants. Cochrane
Database of Systematic Reviews 2001(4); CD001457.
65 Benjamin DK Jr, Miller W, Garges H, et al. Bacteremia,
central catheters, and neonates: when to pull the line.
Pediatrics 2001; 107(6): 1272–1276.
66 Fleer A, Verhoef J, Pascual Hernandez A. Coagulase-
negative staphylococci as nosocomial pathogens in
neonates. The role of host defense, artificial devices, and
bacterial hydrophobicity. American Journal of Medicine
1986; 80(6B): 161–165.
67 Mermel LA. Prevention of intravascular catheter-related
infections. Annals of Internal Medicine 2000; 132(5): 391–
68 Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of
central venous catheter-related infections by using maxi-
mal sterile barrier precautions during insertion. Infection
Control and Hospital Epidemiology 1994; 15(4 Pt 1): 231–
69 Sherertz RJ, Ely EW, Westbrook DM, et al. Education of
physicians-in-training can decrease the risk for vascular
catheter infection. Annals of Internal Medicine 2000;
70 Linares J, Sitges-Serra A, Garau J, et al. Pathogenesis of
catheter sepsis: a prospective study with quantitative and
semiquantitative cultures of catheter hub and segments.
Journal of Clinical Microbiology 1985; 21(3): 357–360.
71 Salzman MB, Isenberg HD, Shapiro JF, et al. A prospective
study of the catheter hub as the portal of entry
for microorganisms causing catheter-related sepsis in
neonates. The Journal of Infectious Diseases 1993; 167(2):
72 Mueller-Premru M, Gubina M, Kaufmann ME, et al. Use
of semi-quantitative and quantitative culture methods
and typing for studying the epidemiology of central
parenteral nutrition. Journal of Medical Microbiology 1999;
73 McCarthy MC, Shives JK, Robison RJ, et al. Prospective
evaluation of single and triple lumen catheters in total
parenteral nutrition. JPEN. Journal of Parenteral and Enteral
Nutrition 1987; 11(3): 259–262.
74 Sitges-Serra A, Hernandez R, Maestro S, et al. Prevention
of catheter sepsis: the hub. Nutrition 1997; 13(4 Suppl):
75 Salzman MB, Isenberg HD, Rubin LG. Use of disinfect-
ants to reduce microbial contamination of hubs of vascu-
lar catheters. Journal of Medical Microbiology 1993; 31(3):
76 Segura M, Alvarez-Lerma F, Tellado JM, et al. A clinical
trial on the prevention of catheter-related sepsis using
a new hub model. Annals of Surgery 1996; 223(4): 363–
77 Mermel LA. New technologies to prevent intravascular
catheter-related bloodstream infections. Emerging Infectious
Diseases 2001; 7(2): 197–199.
78 Craft AP, Finer NN, Barrington KJ. Vancomycin for
[Systematic Review]. Cochrane Database of Systematic
Reviews 2002; 1.
79 Nopper AJ, Horii KA, Sookdeo-Drost S, et al. Topical
ointment therapy benefits premature infants. The Journal of
Pediatrics 1996; 128(5 Pt 1): 660–669.
80 Edwards WH, Conner JM, Soll RF, et al. The
Effect of Aquaphor? Original Emollient Ointment on
Nosocomial Sepsis Rates nd Skin Integrity in Infants of
Birth Weight 501 to 1000 Grams. Pediatric Research 2001;
49(4 part 2): 388A.
81 Campbell JR, Zaccaria E, Baker CJ. Systemic candidiasis in
extremely low birth weight infants receiving topical
petrolatum ointment for skin care: a case-control study.
Pediatrics 2000; 105(5): 1041–1045.
82 Ramsey K, Malone S, Fey P, et al. Aquaphor as a source of
colonization and subsequent bloodstream infections
among very low birthweight neonates. Abstracts-On-
Line: The Society for Healthcare Epidemiology of America
in preterm neonates
Preventing nosocomial bloodstream infection in very low birth weight infants