Original Article
Effect of Bathing on Skin Flora of Preterm Newborns
Maria Luzia Chollopetz da Cunha, RN, PhD
Renato S. Procianoy, MD, PhD
OBJECTIVE:
To assess the effect of bathing with water only or with mild pH neutral
soap and water on skin flora of preterm newborns.
METHOD:
Randomized blinded clinical trial that enrolled 73 preterm newborns
whose gestational age was between 28 and 35 weeks and birth weight
between 800 and 1800 g. At the age of 3 days the infants were randomly
assigned to a group that was given daily baths either with water only, or
with soap and water for 7 or more days. Antibiotic treatment delayed onset
of the trial in some preterm infants. On the final day of bathing axillary
bacterial cultures were obtained before and 30 minutes after bathing.
RESULTS:
At the time of axillary culturing, the mean age of the newborns was
19 days. Coagulase-negative Staphylococcus was the most prevalent
microorganism in both groups. No significant difference was found in the
count of microorganism colonies between both groups. The comparison
of the number of bacterial colonies between the time before and after
bath, carried out by repeated-measures ANOVA showed a significant
difference over time in the two groups, without a significant difference
between the two groups.
CONCLUSIONS:
Bathing preterm newborns with water only or with soap and water
produces similar effects on skin colonization of a preterm neonate. Both
are effective to decrease the number of colonies of Gram-positive and
Gram-negative bacteria.
Journal of Perinatology (2005) 25, 375–379. doi:10.1038/sj.jp.7211332
Published online 28 April 2005
Department of Pediatrics, Newborn Section, Universidade Federal do Rio Grande do Sul and
Hospital de Clı´nicas de Porto Alegre, Porto Alegre, RS, Brazil.
Financial assistance: This project was partially funded by the Hospital de Clı´nicas de Porto Alegre
Research Incentive Fund (FIPE-HCPA) Grant no 02318.
Address correspondence and reprint requests to Renato S. Procianoy, RN, PhD, Rua Tobias da
Silva 99 conj 302, Porto Alegre, RS 90570-020, Brazil.
Journal of Perinatology 2005; 25:375–379
r 2005 Nature Publishing Group All rights reserved. 0743-8346/05 $30
www.nature.com/jp
INTRODUCTION
Routine bathing may harm the neonatal skin because of the fragile
epidermis1 and allow increased colonization of environmentally
acquired bacteria. Most authors state that all soaps are irritating
and emphasize that their frequent use is harmful, because they
remove the lipid film from the skin surface.2–4 Physiologically, the
skin pH, which is neutral at birth, acidifies during the first week of
life, with a pH varying from 5.0 to 5.5.1,5–7 This ‘‘acid mantle’’
diminishes bacterial colonization and promotes the retention of
humidity at the skin barrier.8 Bathing may undo the ‘‘acid
mantle’’ elevating the skin pH.2,9
Mechanisms involved in skin colonization of preterm newborns
in a NICU have not been fully explained.10 The skin colonization
of a newborn in a NICU is consequence of multiple factors,
including routine baths. In a review of the literature it was found
that no study investigated the effect of bathing with water only or
bathing with mild pH neutral soap and water on the skin microbial
flora of preterm newborns.
Our hypothesis is that the type of bath may influence the
number of colonies and type of microorganisms present in the skin
flora. This study compares the type of microorganism and quantity
of colonies present on the skin of preterm newborns before and
after bathing.
PATIENTS AND METHODS
A randomized blind clinical trial was performed with preterm
newborns with gestational age between 28 and 35 weeks and birth
weight between 800 and 1800 g who were admitted to a single
neonatal intensive care unit between October 1, 2002 and
December 31, 2003. The evaluation of gestational age, performed
immediately after birth, was determined using the maternal dates,
and confirmed by clinical findings.11
Exclusion criteria were: preterm newborns with skin breakdown,
congenital malformations, children of HIV-positive mothers,
children of mothers with any infection of the STORCH group
during pregnancy, nonuse of the Kangaroo Mother Method
(skin-to-skin contact), mechanical ventilation, intravenous or
intra-arterial catheters and clinical suspicion or proven infection
at the time of taking the axillary cultures. The Kangaroo method
is routinely used in the NICU where the study was performed. Since
in this method there is skin-to-skin contact between newborn and
mother, then not using this method may lead to a different type of
skin colonization. Therefore, it was decided to exclude from the
study newborns in whom the Kangaroo method was not used. The
375
 da Cunha and Procianoy
use of invasive procedures such as mechanical ventilation and
intravenous or intra-arterial catheters, was an exclusion criteria
because of increased infection risk.
Classification of Newborns
All preterm neonates born during the data collection period, who
fulfilled the eligibility criteria, were randomly assigned to one of
the two groups. Group A consisted of preterm neonates who were
bathed daily with water only for 7 consecutive days, and Group B
consisted of preterm neonates who were given a daily bath with
water and liquid mild pH neutral soap for 7 consecutive days
(Table 1). By protocol the infants were bathed daily and once for 7
consecutive days without systemic antibiotic therapy then axillary
quantitative bacterial cultures were performed. Cultures were
collected between the 11th and 30th day of life. From the time of
assignment to the time of axillary culture, antibiotic-treated infants
received the assigned bath type.
The subjects were randomized in blocks of 10 preterm newborns
each, and each block contained five subjects from each group. The
purpose of this technique was to ensure that the number of
participants would be distributed equally among the study groups.
A 1-year period was estimated to complete the data collection of the
study sample. Taking into account that seasonal changes in
nosocomial colonization could occur during this period, it was
decided to perform randomization in blocks of 10. In this manner
temporal changes in colonization could be controlled. The
Computer Program for Epidemiologists (PEPI v.3.0)12 was used for
block randomization.
During the first 3 days of life, all preterm newborns were bathed
with liquid soap and water, according to the unit routine. These
baths were not taken into account in the study. Members of the
nursing team in the NICU administered the daily baths. Since the
control group was to be treated according to the unit routine, with
a daily bath, the experimental group (water only) also had a daily
bath.
Table 1 Composition of Liquid Mild pH Neutral Soap
Soap formula*
Texapon SBN (detergent)
Dehyton KB (cocamide)
Plantaren 2000 (detergent)
Glycerin (emolient)
Coperlan KDB (thickener)
Citric acid
Sodium chloride
Deionized water
pH ¼ 7
*Formula manipulated in the Industrial Pharmacy at HCPA.
376
Bathing and Skin Flora of Preterm Newborns
The technique of the sponge bath was similar for both groups
except the cleansing agent used. The same tap water was used in
both groups (pH ¼ 6.77). The sponge bath was performed using soft
cotton compresses, which were wetted only with water (group A) or
with liquid soap and water (group B). The bath began by washing
the newborn’s face with cotton compress soaked only in water and
afterwards dried, in both groups. Then the scalp was washed and
dried. Next the neck, abdomen, back and extremities were washed.
Finally, the genitals and the buttocks were washed. The newborn’s
whole body was dried, diapered, and returned to the incubator.
Procedures
After seven antibiotic-free bathing skin cultures were collected from
the right axilla. A single researcher, blind to the type of bath given
to the newborn, performed the collection procedure. The collection
method for a microbial skin culture was based on a technique
previously used by Franck et al.13 Two collections were performed,
in each preterm newborn, once immediately before a bath and
another 30 minutes after the bath. The cotton swab was dipped into
5 ml of tryptic soy broth culture media and then painted 10 times
on the right axilla in a same 2 cm2 area. The cotton swab was
painted on the axilla five times in the vertical direction and five
times in the horizontal one. The cotton swab was placed back into
the tryptic soy broth media and sent to the microbiology laboratory
immediately after each collection.
Blinded quantitative cultures of 0.05 ml of the tryptic soy broth
were performed using sheep blood agar culture plates. The number
of colony forming units (CFU) of each microorganism was
determined after 48 hours incubation. Identification of
microorganisms was based on the morphological analysis of the
colonies, on biochemical tests and on the use of an automated
mini APIs system (bioMérieux, Dulham, USA). The
microorganisms identified were coagulase-negative Staphylococcus,
Staphylococcus aureus, Gram-negative bacilli and Candida sp.
Sample Size
The reference used to calculate the sample size was the study by
Franck et al.,13 estimating that the difference in proportion
(microorganisms) between the groups was 34% (26% versus 60%).
A 5% (a ¼ 0.05) level of significance and a power of 80%
(b ¼ 0.2) were established.12 It was estimated that 66 subjects
would be needed, 33 preterm infants for each group.
Statistical Analysis
Since the data in CFU presented an asymmetrical distribution,
logarithmic transformation was used to normalize the data
distribution. The culture before bathing and the culture 30 minutes
after bathing of the groups were compared using the repeated-
measures ANOVA.
The categorical data involving the occurrence or not of the
different microorganisms studied were compared among the groups
Journal of Perinatology 2005; 25:375–379
Bathing and Skin Flora of Preterm Newborns da Cunha and Procianoy

using the w2 test and, when necessary, the exact Fisher’s test.
Demographic data were also compared between the groups using
the Student’ t and w2 tests. The level of statistical significance
accepted was a ¼ 0.05. The Statistical Package for Social Science
(SPSS version 10.0) was used to analyze the data.
Ethics
The Ethics Committee of the hospital where the study was
performed approved the research protocol, and all the preterm
neonates had the written consent of their parents to participate in
the project.
RESULTS
During the study period 394 newborns with gestational age <37
weeks were hospitalized in the NICU. Of these, only 106 fulfilled the
inclusion criteria (gestational age between 28 and 35 weeks and
birth weight between 800 and 1800 g). However, 10 were excluded
because they had not participated in the Kangaroo Mother Method
due to the severity and instability of their clinical picture, five had
malformations, four had a congenital infection and one died. The
others were included in the study.
In all, 86 preterm newborns were included in the study, 43 in
each group. In group A there were 10 exclusions, nine due to the
use of antibiotics such that they did not remain untreated for 7
days before 30 days of age, and one who died. In group B three
neonates were excluded, two because they had been treated for a
long time with antibiotics, and one who died. Thus, 73 premature
newborns distributed in groups A (n ¼ 33) and B (n ¼ 40)
participated in the analysis.
There was no difference in the two groups as to birth weight,
gestational age and age (days of life) in collecting skin cultures.
Skin cultures of newborns in both groups were obtained at an
average of 19 days of age. The number of patients not treated with
antibiotics and those who were treated with antibiotics before the
seven days ‘‘off’’ was similar in both groups (Table 2).
Organisms present before the bath showed a high prevalence
of coagulase-negative Staphylococcus in both groups and
overall no difference in incidence of any bacterial species
(Table 3).
As to the difference in the number of CFU between the culture
before bath and the culture 30 minutes after bath, the repeated-
measures ANOVA showed a significant difference in the CFU counts
of Gram-positive ( p<0.001) and Gram-negative ( p ¼ 0.032)
bacteria, indicating that the skin colonization decreased between
the culture before bath and the culture after bath. This change was
similar for both groups (Figures 1 and 2). No significant diference
in the CFU counts of Gram-positive ( p ¼ 0.13) and Gram-negative
( p ¼ 0.80) bacteria occurred between the water bathing and the
soap and water bathing (Figures 1 and 2).
The patients were analyzed after stratification according
to the use of antibiotics. This analysis evaluated the preterm
infants who were and were not treated with antibiotics comparing
the group that was bathed only with water with those bathed with
soap and water. No statistically significant difference was found
among the groups. Thus, use of antibiotics did not affect the
comparisons of water bathing when compared with soap and water
bathing.
Of the 73 preterm newborns, three had a positive blood culture
between the time of randomization and the time of axillary
culturing. Two Group A infants had positive blood cultures: one
coagulase-negative Staphylococcus and the other was
Staphylococcus aureus. The skin cultures of the former patient
was congruent with the blood culture (same strain), but the latter
patient was not congruent. In group B, the same strain of
coagulase-negative Staphylococcus was identified in one blood
culture and was also identified in the skin culture of the same
preterm newborn.
Table 3 Comparison of Axillary Microorganisms Present Before Bath
Microorganism Type of bath p
With water With water and
(n ¼ 33) soap (n ¼ 40)

Table 2 Characteristics of Patient Groups Coagulase-negative 29 (87.9%) 36 (90%) 0.77

Staphylococcus
Characteristics Type of bath p S. aureus 6 (18.2%) 12 (30.0%) 0.24
Klebsiela pneumoniae 2 (6.1%) 1 (2.5%) 0.59
Group A Group B with
Enterobacter sp. 1 (3.0%) 2 (5.0%) 0.99
with water water and soap
Eschericia coli 1 (3.0%) 0 (0.0%) 0.45
(n ¼ 33) (n ¼ 40)
Klebsiela oxytoca 1 (3.0%) 0 (0.0%) 0.45
Gestational age (weeks) 31±2 32±2 0.22 Stenotrophomonas maltophilia 1 (3.0%) 0 (0.0%) 0.45
Birth weight (g) 1356±270 1384±221 0.63 Acinetobacter sp. 1 (3.0%) 0 (0.0%) 0.45
Age at collection (days of life) 19±6 19±5 0.91 Serratia sp. 1 (3.0%) 0 (0.0%) 0.45
Use of antibiotics 16 (48.5%) 17 (42.5%) 0.78 Candida sp. 1 (3.0%) 0 (0.0%) 0.45

Data expressed as mean±SD or frequency (percentage). Data expressed in # patients and percentage.

Journal of Perinatology 2005; 25:375–379 377

 da Cunha and Procianoy
Figure 1. Difference in the number of colony forming units of Gram-
positive bacteria between the culture before bath and the culture 30
minutes after bath by repeated-measures ANOVA in log. p(time) <0.001,
p(interaction) ¼ 0.13.
Figure 2. Difference in the number of colony forming units of Gram-
negative bacteria between the culture before bath and the culture 30
minutes after bath by repeated-measures ANOVA in log. p(time) ¼ 0.032,
p(interaction) ¼ 0.80.
DISCUSSION
As in our study, others have demonstrated predominance of
coagulase-negative Staphylococcus on skin of preterm newborns
admitted.14–16 Franck et al.13 found the presence of this
378
Bathing and Skin Flora of Preterm Newborns
microorganisms in 93% of the cases when they investigated the
frequency of bathing and colonization of the skin of preterm
newborns. Savey et al.14 demonstrated positive axillary cultures for
S. aureus from 9.6% of preterm newborns in their study.14
However, the study performed by Cowan and Frost17 comparing the
use of soap versus detergent on skin flora of term newborns found
a prevalence of S. aureus between 18.35 and 20.3%. These results
are similar to ours.
Cowan and Frost17 demonstrated a prevalence of 1.3% of Gram-
negatives in skin flora of term infants, but Franck et al.13
demonstrated that 26% of preterm infants had Gram-negative
organisms in skin cultures. Our results are similar to those of
Franck et al. The difference between term and preterm infant
colonization may be due to the short period of time term infants
spend in the hospital hence preventing nosicomial infection or to
specific conditions of their skin such as being drier.18
Although no difference in the effect of water or soap and water
was noted on types of organisms or CFU, we did show a significant
decrease in CFU following bathing in both groups. However, no
significant effect difference was found between the groups. We are
unaware of a similar study in preterm infants in the literature. A
single study by Medves and O’Brien compared skin cultures of two
groups of term newborns whose admission bath was either water
only or soap and water. They showed that both conditions had a
minimal effect on the bacterial colonization of skin at birth.19
Gfatter et al.2 showed that bathing with soap triggers an
elevation in skin pH which interferes in physiological protection
(acid mantle) provoking a change in the composition of the
cutaneous bacterial flora and in the activity of enzymes on the
epidermis. For this reason, the use of soap once or twice a week is
recommended, considering that daily bath routine is responsible
for drying the skin, predisposing to breakdown.1 Lack of skin
integrity favors invasion of bacteria and fungi, and development of
systemic infection.20
In the present study, skin culture was not a good marker of
sepsis, since a small number of cases with positive blood culture
was found, that is, of the 73 preterm newborns, only three
developed sepsis since the time of randomization. Sample size was
not figured out for this purpose, although, according to Evans
et al.,15 the surface cultures of the skin are of limited value in
predicting the etiology of neonatal sepsis.
A limitation of our study is the lack of blinding the nurses who
were administering the baths. However, we believe that this did not
affect the results, because both the researcher who performed the
axillary cultures and the microbiologist who performed the
bacteriological studies were blinded to the type of agent used in
bathing.
In conclusion, our data demonstrate that bathing premature
infants results in reduced skin colonization by bacteria from before
to after the bath. Bathing with water or soap and water does not
affect this result. No difference in types of bacteria or colony counts
Journal of Perinatology 2005; 25:375–379
Bathing and Skin Flora of Preterm Newborns
were observed in the prebath skin culture after 1 week of baths
whether the infants were bathed with water or soap and water.
Acknowledgements
The authors thank Prof. Dr. Mario Wagner for his assistance with the statistical
analysis, and Dr. Suzana Barcellos for her assistance at Microbiology Laboratory.
References
1. Lund C, Kuller J, Lane A, Lott JW, Raines DA. Neonatal skin care: the
scientific basis for practice. J Obstet Gynecol Neonatal Nurs 1999;28:
241–54.
2. Gfatter R, Hackl P, Braun F. Effects of soap and detergents on skin surface
pH, stratum corneum hydration and fat content in infants. Dermatology
1997;195:258–62.
3. Baranda L, Gonzalo-Amaro R, Torres-Alvares B, Alvares C, Ramı́rez V.
Correlation between pH and irritant effect of cleansers marked for dry skin.
Int J Dermatol 2002;41:494–9.
4. Lodén M, Buraczewska I, Edlund F. The irritation potential and reservoir
effect of mild soaps. Contact Dermatitis 2003;49:91–6.
5. Fox C, Nelson D, Warechsm J. The timing of skin acidification in very low
birth weight infants. J Perinatol 1998;18:272–5.
6. Yosipovitch G, Maayan-Metzger A, Merlob P, Sirota L. Skin barrier properties
in different body areas in neonates. Pediatrics 2000;106:105–8.
7. Hoeger PH, Enzmann CC. Skin physiology of the neonate and young infant:
a prospective study of functional skin parameters during early infancy.
Pediatr Dermatol 2002;19:256–62.
8. Nix DH. Factors to consider when selecting skin cleansing products.
J Wound Ostomy Continence Nurs 2000;27:260–8.
Journal of Perinatology 2005; 25:375–379
da Cunha and Procianoy
9. Lund C. Prevention and management of infant skin breakdown. Nurs Clin
North Am 1999;34:907–20.
10. Saiman L, Ludington E, Dawson JD, Patterson JE, Rangel-Frausto S, Wiblin
T. Risk factors for Candida species colonization of neonatal intensive care
unit patients. Pediatr Infect Dis J 2001;20:1119–24.
11. Ballard JL, Wedig K, Wang L, Ellers-Walsman L, Lipp R. New Ballard
Score, expanded to include extremely premature infants. J Pediatr 1991;119:
417–23.
12. Abramson JH, Gahlinger P. Computer Programs for Epidemiologists: PEPI v.
3.0. Salt Lake City, UT: Sagebrush Press; 2000.
13. Franck LS, Quinn D, Zahr L. Effect of less frequent bathing of preterm
infants on skin flora and pathogen colonization. J Obstet Gynecol Neonatal
Nurs 2000;29:584–9.
14. Savey A, Fleurette J, Salle BL. An analysis of microbial flora of premature
neonates. J Hosp Infect 1992;21:275–89.
15. Evans ME, Schaffner W, Federspiel CF, Cotton RB, McKee T, Stratton CW.
Sensitivity, specificity, and predictive value of body surface cultures in a
neonatal intensive care unit. JAMA 1988;259:248–52.
16. D’Angio CT, McGowan KL, Baumgart S, Geme JS, Harris MC. Surface
colonization with coagulase-negative staphylococci in premature neonates.
J Pediatr 1989;114:1029–34.
17. Cowan ME, Frost MR. A comparison between a detergent baby bath
additive and baby soap on the skin flora of neonate. J Hosp Infect 1986;
7:91–5.
18. Chiller K, Selkin BA, Murakawa GJ. Skin microflora and bacterial infections
of the skin. J Invest Dermatol Symp Proc 2001;6:170–4.
19. Medves JM, O’Brien B. Does bathing newborns remove potentially harmful
pathogens from the skin? Birth 2001;28:161–5.
20. Rowen JL, Atkins JT, Levy ML, Baer SC, Baker CJ. Invasive fungal dermatitis
in the r1000-gram neonate. Pediatrics 1995;95:682–7.
379