countries reported the absence of both a nationwide and a
Almost all respondents (96%) from the 25 countries with-
out nationwide guidelines to prevent neonatal GBS infection
would give antibiotics to women with signs of infection during
labor. In case of preterm birth (?37 weeks), ruptured mem-
branes for ?18 or 24 h, respectively, 53, 54 and 65% of the
respondents would give antibiotics during labor. When GBS-
positive mothers were adequately treated during labor, ad-
ministration of antibiotics to infants without signs of infection
would be decided for by 48% of the respondents if preterm and
19% if term. When mothers were not or inadequately treated
during labor, 61% of the infants born before 37 weeks gesta-
tion and 35% of term infants would be treated with antibiotics
Discussion. The aim of the study was to establish the
incidence of GBS colonization in pregnant women and the
incidence of neonatal GBS infection and to determine policies
for prevention of neonatal GBS infection in Europe. The
response rate of 15% was low, but 29 of the 33 European
countries involved were represented by at least 1 respondent.
Therefore we believe that we have a reasonable impression of
The data collected show that in most European countries
the incidence of GBS colonization among pregnant women
varies between 10 and 20% and the incidence of neonatal GBS
infection ranges from 0.5 to 2 per 1000 live births. The
differences found may be partly explained by the use of
different data collection sources (regional, national), differ-
ences in definition of GBS sepsis used (probable and/or proven
cases) and differences in the culture methods used.
Nationwide guidelines for prevention have been issued in
only a few European countries thus far. About one-half of the
respondents reported the existence of local hospital-based
protocols in their institution, but the prevention practices
vary among doctors and institutions. The policy of giving
antibiotics to pregnant women with signs of infection during
delivery is widely accepted. However, only about one-half of
the respondents would give antibiotics in case of expected
preterm delivery and after prolonged rupture of the mem-
When pregnant women are identified as being colonized
with GBS, most respondents would not give antibiotics to
prevent early onset GBS infection to the infant in case of term
delivery and absence of neonatal signs of infection, whether or
not intrapartum antibiotic prophylaxis was given. However,
about one-half of the respondents would do so after preterm
delivery, especially if no or inadequate intrapartum prophy-
laxis was given.
Our study shows that only four European countries have a
nationwide guideline for prevention of early onset GBS infec-
tion. Because of the lack of epidemiologic data and uniform
early onset GBS infection prevention methods, a surveillance
study in European countries would be needed for determina-
tion of the most appropriate prevention policy.
Acknowledgment. This study was financed by Zorg
Onderzoek Nederland Project 2200004.
Monique A. J. M. Trijbels-Smeulders, M.D.
Louis A. A. Kolle ´e, M.D.
Albert H. Adriaanse, M.D.
Jan L. L. Kimpen, M.D.
Leo J. Gerards, M.D.
Department of Pediatrics
University Children’s Hospital
University Medical Center
Nijmegen (MAJMTS, LAAK)
Department of Obstetrics and Gynecology
Departments of Pediatrics (JLLK) and
Wilhelmina Children’s Hospital
University Medical Center
Accepted for publication Nov. 12, 2003.
Key words: Neonatal infection, group B Streptococcus, preven-
Address for reprints: M. Trijbels-Smeulders, M.D., Depart-
ment of Pediatrics, UMC St. Radboud, P.O. Box 9101, 6500 HB
Nijmegen, the Netherlands. Fax 31 243619123; E-mail
1. CDC. Prevention of perinatal group B streptococcal disease: a
public health perspective. MMWR 1996;45:1–24.
2. Schrag SJ, Zywicki S, Farley MM, et al. Group B streptococ-
cal disease in the era of intrapartum antibiotic prophylaxis.
N Engl J Med 2000;342:15–20.
3. CDC. Prevention of perinatal group B streptococcal disease:
revised guidelines from CDC. MMWR 2002;51:1–24.
4. Schrag SJ, Zell ER, Lynfield R, et al. A population-based
comparison of strategies to prevent early-onset group B
streptococcal disease in neonates. N Engl J Med 2002;347:
5. Poulain P, Betremieux P, Donnio PY, Proudhon JF, Karege
G, Giraud JR. Selective intrapartum anti-bioprophylaxy of
group B streptococci infection of neonates: a prospective
study in 2454 subsequent deliveries. Eur J Obstet Gynecol
Reprod Biol 1997;72:137–40.
6. Hoogkamp-Korstanje JAA, GerardsLJ, Cats BP. Maternal
carriage and neonatal acquisition of group B streptococci.
J Infect Dis 1982;145:800–3.
7. Trijbels-Smeulders M, Gerards LJ, Pasker-de Jong PCM, et
al. Epidemiology of neonatal group B streptococcal disease in
the Netherlands 1997–1998. Paediatr Perinat Epidem 2002;
8. Hastings MJ, Easmon CS, Neill J, Bloxham B, Rivers RP.
Group B streptococcal colonisation and the outcome of preg-
nancy. J Infect 1986;12:23–9.
9. Moses LM, Heath PT, Wilkinson AR, Jeffery HE, Isaacs D.
Early-onset group B streptococcal neonatal infection in Ox-
ford 1985–96. Arch Dis Child Fetal Neonatal Ed 1998;79:
10. Feikin DR, Thorsen P, Zywicki S, Arpi M, Westergaard JG,
Schuchat A. Association between colonization with group B
streptococci during pregnancy and preterm delivery among
Danish women. Am J Obstet Gynecol 2001;184:427–33.
11. Aavitsland P, Hoiby EA, Lystad A. Systemic group B strep-
tococcal disease in neonates and young infants in Norway
1985–94. Acta Paediatr 1996;85:104–5.
12. Juncosa T, Bosch J, Dopico E, et al. Neonatal infection by
Streptococcus agalactiae: multicenter study in the area of
Barcelona. Enferm Infecc Microbiol Clin 1998;16:312–15.
13. Trijbels-Smeulders MAJM, Adriaanse AH, Gerards LJ, Kim-
pen JLL. Prevention strategy for neonatal early-onset group-
B-streptococcal (GBS) disease in the Netherlands. Rev Med
AFFORDABLE DIAGNOSIS OF HUMAN
IMMUNODEFICIENCY VIRUS INFECTION IN INFANTS
BY P24 ANTIGEN DETECTION
The expense of PCR testing limits diagnosis of
HIV infection in infancy in low resource settings.
Vol. 23, No. 2, February 2004173
THE PEDIATRIC INFECTIOUS DISEASE JOURNAL
The ultrasensitive p24 antigen assay has been pro-
posed as an accurate substitute; however, its ability
to detect different HIV viral subtypes remains to be
determined. A sensitivity of 98.1% and specificity of
98.7% was obtained testing 203 samples from 24
HIV-infected and 66 uninfected infants born to HIV
subtype C-infected women.
Antenatal clinic prevalence rates of HIV infection in South
Africa were 25.9% in 2002, which means that more than
one-fourth of children currently born in South Africa are
vertically exposed to HIV.1In general guidelines for Preven-
tion of Mother to Child Transmission (PMTCT) programs in
low resource settings require all exposed children to be
followed to 12 months of age or older before their HIV
infection status can be determined by an HIV antibody test.
The persistence of maternal anti-HIV antibodies and the cost
and complexity of two HIV PCR tests precludes earlier
testing.2In South Africa ?280 000 HIV-exposed children per
annum, of whom as many as 90% may be HIV-uninfected,
must be followed for 12 months or more. A universal frustra-
tion of PMTCT programs in low resource settings is the lack
of capacity to achieve this follow-up, thereby missing the
opportunity to determine the HIV infection status of exposed
children.3, 4This prevents access to available medical care
known to improve the quality and quantity of life of HIV-
infected children and their families (e.g. prophylactic medi-
cation, treatment of intercurrent infections, nutritional and
emotional support). Assessment of the p24 antigen (Ag) assay
for diagnosis of HIV infection in infants is listed among the
urgent research priorities to improve pediatric HIV/AIDS
care in developing countries.5, 6
The value of the p24 Ag assay in this context has been
limited by its poor sensitivity.7, 8Recently a technically im-
proved version of the assay has been as sensitive and specific
as HIV PCR tests that detect viral nucleic acids. This ultra-
sensitive p24 Ag assay has been proposed for use in diagnosis
of pediatric HIV infection and in viral load monitoring of
patients receiving antiretroviral therapy.5, 8–14A proviso for
widespread introduction of the p24 Ag assay is the require-
ment for further validation of the assay in non-B viral
subtypes.5, 9, 13Viral subtype C accounts for 99% of HIV
infections in South Africa.15The performance of the p24 Ag
assay for early infant diagnosis in HIV subtype C is described.
Materials and methods. Infants born to HIV-infected
women attending Coronation Women and Children’s Hospital
in Johannesburg, South Africa were enrolled between Janu-
ary and October 2002 in a study to assess an affordable infant
diagnostic protocol in a low resource setting. The ethics
committee of the University of the Witwatersrand approved
the study. Mothers and infants each received a single dose of
nevirapine for PMTCT as standard practice at the hospital.16
Fewer than 3% of mothers reported breast-feeding their
infants; this was consistent with an HIV transmission rate of
?9% at 6 weeks and 3 months of age.16No infant received
antiretroviral therapy after the first 72 h of age.
The p24 Ag ultrasensitive enzyme-linked immunosorbent
assay (HIV-1 p24 Ag Ultra kit; PerkinElmer Life Sciences,
Turku, Finland) was performed on 203 plasma samples
belonging to 24 HIV-infected and 66 HIV-uninfected infants
(Table 1). The plasma samples were collected prospectively in
EDTA, separated and stored at –70°C within 8 h of collection.
HIV DNA PCR tests were analyzed with the Roche Amplicor
HIV-1 DNA test Version 1.5 (Roche Diagnostic Systems, Inc.,
Branchburg, NJ) at the time that the blood was sampled. The
final HIV infection status was established by 2 concordant
HIV DNA PCR tests done at 6 weeks and 3 months of age and
seroreversion at 12 months of age. When seroreversion had
not occurred by 12 months of age, an additional negative PCR
result was obtained on stored samples at 4 or 7 months of
age.2Laboratory staff performing the p24 Ag assay were
blinded to the PCR results and the HIV infection status of the
child. Viral load test results were not available for assessment
of the quantitative accuracy of the p24 Ag assay. HIV sub-
typing was performed on 16 (67%) of the 24 HIV-infected
children and 40 (44%) of the 90 mothers of these infants.
Results. The HIV DNA PCR test, regardless of the age at
which it was performed, concurred with the final HIV infec-
tion status of the child in all 90 patients. The HIV DNA PCR
results and final HIV infection status can therefore be used
interchangeably as the standard against which the p24 Ag
assay was assessed. Diagnostic sensitivity of the ultrasensi-
tive p24 Ag assay was based on 52 plasma samples from 24
HIV-infected infants at various ages (Table 1). One false
negative result was obtained in a sample taken at 6 weeks of
age, giving an overall sensitivity of 98.1%. Diagnostic speci-
ficity of the p24 Ag assay was evaluated on 151 samples from
66 HIV-uninfected infants at varying ages. Two false positive
TABLE 1. Results achieved with the p24 Ag assay in comparison with the known positive and negative HIV DNA PCR
results at different ages
The number of tests performed in each age group equals the number of patients tested except as noted. The majority (82%) of
assays were performed at or before 3 months of age.
No. of Positive
p24 Ag Tests/No.
of Positive PCR
No. of Negative
p24 Ag Tests/No.
of Negative PCR
6 wk (5.9 wk)*22/23†95.7 62/6210085
3 mo (3 mo) 20/20100 62/62 100100100
4 mo (4.1 mo) 2/2100 3/3 100 100100
7 mo (7.4 mo)7/710022/24§ 93.6 92.3 100
* Numbers in parentheses, median.
† 21 patients.
‡ Numbers in parentheses, percent.
§ 20 patients.
PPV, positive predictive value; NPV, negative predictive value.
174Vol. 23, No. 2, February 2004
THE PEDIATRIC INFECTIOUS DISEASE JOURNAL
results were obtained in samples taken at 7 months of age,
giving an overall specificity of 98.7%. The positive and nega-
tive predictive values were 96.2 and 99.3%, respectively.
Reanalysis of the 3 discordant samples and second aliquots of
thawed plasma for each of the 3 samples, analyzed in a
blinded manner, yielded correct results. In these instances
the problem was that the test reading differed minimally
from the cutoff reading, signifying a marginal negative or
positive p24 Ag test result. When HIV genotyping was per-
formed, all HIV-infected infants and all mothers of infants
tested were infected with subtype C virus.
Discussion. Local experience with previous versions of the
p24 Ag assay showing a poor sensitivity of 48% despite an
excellent specificity of 100% (data not shown) has been as
disappointing as reported elsewhere.7, 9, 14HIV-1 p24 Ag
detection in plasma is problematic because the p24 antigen
forms a complex with p24-specific antibodies resulting in
underdetection or false negatives.8, 9Boiling plasma before
antigen testing destroys the antigen-binding capacity of all
antibodies releasing p24 Ag from the immune complex. The
p24 Ag ultrasensitive assay assessed here uses heat dena-
turation and boosts p24 Ag detection by a signal amplifi-
cation step that further increases the sensitivity of the
assay. The improvement in technology has resulted in an
assay that rivals the performance of qualitative and quan-
titative HIV PCR tests.5, 8–14The technology, in compari-
son with PCR is less costly and less complex, involving less
expensive equipment already available in most laborato-
ries. Technically the p24 Ag assay is simpler to perform and
requires minimal training, making diagnosis and monitor-
ing of HIV infection more affordable and accessible in low
The reported sensitivity and specificity of the ultrasensi-
tive p24 Ag assay in the context of pediatric HIV diagnosis
ranges from 97 to 100% and from 99 to 100%, respective-
ly.5, 10, 11, 14Viral subtypes A, B, D and E have prevailed in
these studies.5, 10, 11, 14The sensitivity and specificity of 98.1
and 98.7%, respectively achieved in this study reaffirms how
well the ultrasensitive p24 Ag assay performs in comparison
with the current standard PCR test for infant diagnosis of
HIV. The data show that the ultrasensitive p24 Ag assay
detects viral subtype C as well as subtype B, the subtype that
has been most extensively evaluated.10The timing of HIV
testing in infancy to detect all cases of in utero and intrapar-
tum HIV infection has been subject to debate.2, 17Initial
studies concentrated on establishing the accuracy of the p24
assay in determining HIV infection status in HIV-exposed
infants of varying ages and did not concentrate on younger
infants.10, 11The efficacy of the ultrasensitive p24 Ag assay in
early infant diagnosis requires testing at defined ages in the
first 3 to 4 months of life with a view to establishing
guidelines for the ages at which infants should be tested
(Table 1).14In this cohort the majority of infants were tested
at 6 weeks or 3 months of age (Table 1), at which time the
specificity of the ultrasensitive p24 Ag assay was 100% with
a single false negative result at 6 weeks of age.
The technical improvements that have enabled the ultra-
sensitive p24 Ag assay to perform as well as PCR for infant
diagnosis can make early infant diagnosis of HIV a reality
and transform health care for HIV exposed children of sub-
Saharan Africa and other low resource settings worldwide.
Acknowledgments. We are grateful to all staff and pa-
tients who participated in the study. This research was
partially funded by the Bristol Myers-Squibb Secure the
Future Initiative RES105-01 and the National Health Labo-
Gayle G. Sherman, M.Med. (Haem)
Gwynneth Stevens, Ph.D.
Wendy S. Stevens, M.Med. (Haem)
Department of Molecular Medicine and
National Health Laboratory Service
University of the Witwatersrand
Johannesburg, South Africa
Accepted for publication Oct. 19, 2003.
Key words: Infant, human immunodeficiency virus, diagnosis,
p24 antigen, human immunodeficiency virus subtype C, develop-
Reprints not available.
1. Dorrington RE, Bradshaw D, Budlender D. HIV/AIDS profile
in the provinces of South Africa: indicators for 2002. Cape
Town: Centre for Actuarial Research, Medical Research
Council and the Actuarial Society of South Africa, 2002.
2. Centers for Disease Control. Revised guidelines for prophy-
laxis against Pneumocystis carinii pneumonia for children
infected with or perinatally exposed to human immunodefi-
ciency virus. MMWR 1995;44(RR-4):1–11.
3. African Network for the Care of Children Affected by HIV/
AIDS (ANECCA). Early diagnosis and care of HIV infected
children. Workshop report of 2003, Kampala, Uganda.
4. McCoy D, Besser M, Visser R, et al. [Updated March 7, 2002]
Interim findings on the national PMTCT Pilot Sites Lessons
and Recommendations. Available at: http://www.hst.org.za/
pubs/pmtct/pmtctinterim.htm. Accessed 27 May 2003.
5. Mascolini M. [Update April 2002] HIV monitoring, diagnostic
technology transfer to resource poor settings. Available at:
Accessed May 27, 2003.
6. Lepage P, Spira R, Kalibala S, et al. Care of human immu-
nodeficiency virus-infected children in developing countries:
International Working Group on Mother-to-Child Transmis-
sion of HIV. Pediatr Infect Dis J 1998;17:581–6.
7. Guay LA, Hom DL, Kabengera SR, et al. HIV-1 ICD p24
antigen detection in Ugandan infants: use in early diagnosis
of infection and as a marker of disease progression. J Med
8. Schu ¨pbach J, Bo ¨ni J, Tomasik Z, et al. Sensitive detection
and early prognostic significance of p24 antigen in heat-
denatured plasma of human immunodeficiency virus type
1-infected infants. J Infect Dis 1994;170:318–24.
9. Schu ¨pbach J, Tomasik Z, Nadal D, et al. Use of HIV-1 p24 as
a sensitive, precise and inexpensive marker for infection,
disease progression and treatment failure. Int J Antimicrob
10. Nadal D, Bo ¨ni J, Kind C, et al. Prospective evaluation of
amplification-boosted ELISA for heat-denatured p24 antigen
for diagnosis and monitoring of pediatric human immunode-
ficiency virus type 1 infection. J Infect Dis 1999;180:1089–95.
11. Lyamuya E, Bredberg-Råde ´n U, Massawe A, et al. Perfor-
mance of a Modified HIV-1 p24 antigen assay for early
diagnosis of HIV-1 infection in infants and prediction of
mother-to-infant transmission of HIV-1 in Dar es Salaam,
Tanzania. J Acquir Immune Defic Synd Hum Retrovir 1996;
12. Stephenson J. Cheaper HIV Drugs for poor nations bring a
new challenge: monitoring treatment. JAMA 2002;288:
13. Schu ¨pbach J. Measurement of HIV-1 p24 antigen by signal-
amplification-boosted ELISA of heat-denatured plasma is a
simple and inexpensive alternative to tests for viral RNA.
AIDS Rev 2002;4:83–92.
14. Sutthent R, Gaudart N, Chokpaibulkit K, et al. p24 antigen
detection assay modified with a booster step for diagnosis and
monitoring of human immunodeficiency virus type 1 infec-
tion. J Clin Microbiol 2003;41:1016–22.
15. van Harmelen JH, van der Ryst E, Loubser AS, et al. A
Vol. 23, No. 2, February 2004 175
THE PEDIATRIC INFECTIOUS DISEASE JOURNAL
predominantly HIV type 1 subtype C-restricted epidemic in Download full-text
South African urban populations. AIDS Res Hum Retrovi-
16. Guay LA, Musoke P, Fleming T, et al. Intrapartum and
neonatal single-dose nevirapine compared with zidovudine
for prevention of mother-to-child transmission of HIV-1 in
Kampala, Uganda: HIVNET 012 randomised trial. Lancet
17. Benjamin DK, Miller WC, Fiscus SA, et al. Rational testing of
the HIV-exposed infant. Pediatrics 2001;108:e3.
PSEUDOMONAS AERUGINOSA SEPTIC SHOCK
SECONDARY TO “GRIPE WATER” INGESTION
We report the case of a 9-month-old girl who
presented in septic shock after ingestion of a con-
taminated herbal supplement commonly used to
treat colic. Herbal supplements are widely used by
well-meaning parents for many common conditions.
Pediatricians should be aware that the variable
manufacturing and packaging conditions of herbal
supplements can lead to contamination with infec-
Herbal medicines are popular in many countries world-
wide, including the United States, for treatment of infant
colic and gassiness. Commercial preparations containing seed
oils from dill, fennel, anise, mint, ginger and other herbs are
widely believed to relieve abdominal distress in babies. Al-
though generally safe some herbal supplements can lead to
significant and potentially lethal complications for children.
We report a case of an infant who developed septic shock,
severe absolute leukopenia, neutropenia and subcutaneous
nodules caused by Pseudomonas aeruginosa after ingestion of
contaminated “Gripe Water,” an off-brand Ayurvedic medi-
cine brought from India.
Case report. A 9-month-old previously healthy infant
presented to our emergency department with fever and fuss-
iness for 2 days. The day before admission she began to vomit
repeatedly. On the day of admission, she became progres-
sively lethargic and was taken to the emergency department.
At presentation the infant was febrile (39.9°C), with a pulse of
181 beats/min, respiratory rate of 45 breaths/min, blood
pressure 94/50 mm Hg and oxygen saturation at 96% in room
air. The infant’s skin was mottled. She received immediate
fluid resuscitation and laboratory assessment for infection of
urine, blood and cerebrospinal fluid, after which, ceftriaxone
and vancomycin were administered intravenously. The initial
total leukocyte cell count was 1.6 ? 103/?l (1.6 ? 109/l) with
an absolute neutrophil count of zero. The hemoglobin was 8.2
g/dl (82 g/l). Analysis of electrolytes and pH showed hypona-
tremia (125 mmol/l), hypokalemia (2.2 mmol/l) and a non-
anion gap metabolic acidosis. The patient was admitted to the
pediatric intensive care unit.
Past history. The infant was a previously healthy full term
infant, born in the United States and cared for at home by her
mother. She received all immunizations according to sched-
ule. Growth and development were normal. The family his-
tory was unremarkable. There were no siblings, no one else
was ill at home and there were no pets.
Hospital course. On arrival to the pediatric intensive care
unit, the patient remained lethargic with abnormal vital
signs. Twelve hours into her hospital course she developed
several 1-cm erythematous subcutaneous nodules on her
extremities. Six hours later the blood culture contained
Gram-negative rods identified as P. aeruginosa. Anti-
Pseudomonas therapy (ceftazidime and tobramycin) was ini-
tiated. The subcutaneous nodules were in retrospect consis-
tent with those seen during Pseudomonas sepsis.
The parents revealed that the child had been given, since
early infancy, a daily preventative dose of an herbal solution
called chvarka, which had been purchased in India. Chvarka
is an Ayurvedic medicine version of gripe water, used by
parents around the world to treat or prevent infant colic.
Three days before onset of the infant’s illness, the parents
opened a new bottle of chvarka which had recently been
purchased in India. A sample of the chvarka was cultured,
and numerous bacterial colonies on the agar were identified
as P. aeruginosa the following day.
The infant’s clinical condition improved slowly. She was
discharged from the hospital in good condition after 3 weeks
of anti-Pseudomonas therapy.
Discussion. Gripe water is a traditional herbal remedy
said to effectively treat or prevent “flatulence, teething pain
and tummy upsets in babies.” Despite many enthusiastic
anecdotes no formal evaluation of gripe water has ever been
undertaken. First formulated in England in 1851, it became
commonplace for English nannies to use the product liberally.
The original Woodward’s Gripe Water contained 3.6% alcohol,
dill oil, sodium bicarbonate, sugar and water.1Over time
many companies have modified the ingredients in gripe
water. Typical ingredients in commercial formulations in-
clude some combination of “herbal oils” (e.g. dill, cardamom,
fennel and/or anise), sugar, bicarbonate and sometimes alco-
hol.2, 3Recipes for making gripe water at home are also
readily available over the Internet.4In 1993, the Food and
Drug Administration (FDA) ordered automatic detention of
all shipments of gripe water because it fit the definition of a
new drug without having FDA approval.5However, compa-
nies now market and sell gripe water in the US as a dietary
supplement, instead of a medicine, bypassing FDA regula-
tion. A typical single dose is 2.5 to 5 ml, depending on the age
of the baby. It can be repeated several times daily. The
product, from the United Kingdom, India and other countries,
can be purchased in many ethnic grocery stores. Domestic
gripe water can also be purchased in specialty stores and over
We were unable to find any previous reports of sepsis
caused by bacterial contamination of gripe water or docu-
mented cases of adverse affects from gripe water. To test
locally purchased products, we purchased gripe water made
by two Indian companies and one Egyptian company. Cul-
tures from the contents of the three bottles were sterile.
Because of the lack of regulation of dietary supplements
from the FDA, there can be an increased risk of contamina-
tion of improperly manufactured herbal liquid preparations.
Given the increased frequency of use of herbal supplements in
children, clinicians should be aware of the potential dangers
of ingesting products with loosely regulated manufacturing
David Sas, D.O., M.P.H.
Maria A. Enrione, M.D.
Richard H. Schwartz, M.D.
Departments of Pediatrics and Pediatric
Inova Fairfax Hospital for Children
Falls Church, VA
Accepted for publication Nov. 5, 2003.
Key words: Sepsis, Pseudomonas aeruginosa, gripe water,
176Vol. 23, No. 2, February 2004
THE PEDIATRIC INFECTIOUS DISEASE JOURNAL