PIDSP Journal 2011 Vol 12 No.

2
Copyright ® 2011 75

ETIOLOGY OF NEONATAL SEPSIS IN FIVE URBAN HOSPITALS IN THE

PHILIPPINES

AUTHORS: Cecilia C. Maramba-Lazarte, MD1, Mary Ann C. Bunyi, MD2, Elizabeth E. Gallardo, MD3,
Jonathan G. Lim, MD4, Joanne J. Lobo, MD5, Cecille Y. Aguilar, MD1
University of the Philippine- Philippine General Hospital1, Philippine Children's Medical Center 2,,St. Louis University- Hospital
of the Sacred Heart3, Chong Hua Hospital4, Davao Doctor's Hospital5
CORRESPONDENCE:
Cecilia C. Maramba-Lazarte
Email: [email protected]
KEYWORDS
neonatal sepsis, neonatal infection

ABSTRACT
The most frequent cause of mortality in neonates is infections which include sepsis, meningitis, pneumonia,
diarrhea and tetanus. Approximately 1.6 million deaths due to neonatal infections occur annually in
developing countries. The causes of sepsis vary from one geographic area to another. Thus, it is important
to create a database for the Philippines regarding the etiologies of neonatal sepsis.
Objectives: The study was undertaken to determine the most common bacterial pathogens of neonatal
sepsis in the Philippines and the antibiotic sensitivity patterns of these pathogens.
Methods: A multicenter surveillance and chart review was conducted in five hospital sites for a period of
six months- from July to December 2006. All neonates less than or equal to 28 days who had positive
cultures from a sterile site and or had fulfilled the SIRS criteria as defined by the International Pediatric
Sepsis Consensus Conference in the presence of suspected or proven infection were included in the study.
Results: A total of 289 neonates fulfilled the criteria of sepsis. More than 61% of the patients had early
onset neonatal sepsis. The most common site of infection was the lungs, (57%), followed by sepsis without
an obvious site of infection (35%).
In 50% of patients, an organism was identified; gram-negative bacteria were the dominant bacteria found
(Pseudomonas, Burkholderia, Klebsiella) which comprised 94% of the bacteria isolated. Resistance patterns
varied among the different urban study centers. The Pseudomonas and Burkholderia isolates from the
Philippine General Hospital (PGH) were sensitive to ceftazidime. All Klebsiella isolates from St. Louis
University (SLU) were resistant to third generation cephalosporins. In contrast, Klebsiella isolates from
PGH and Davao were sensitive to third generation cephalosporins but resistant to piperacillin-tazobactam.
The overall mortality rate was 11%. Ampicillin and an aminoglycoside were the most common regimens
used but were only successful in less than 50% of cases.
Conclusion: Gram-negative organisms comprised the majority of the neonatal infections, with
Pseudomonas and Burkholderia being the most prevalent. Resistance patterns differed among the different
centers. Continuing surveillance of infections is still needed in order choose the most appropriate empiric
therapy for each center.

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 76

INTRODUCTION The International Pediatric Sepsis

The most frequent cause of mortality in Consensus Conference8 revised the definitions
neonates is infections which include sepsis, of severe sepsis and septic shock for the
meningitis, pneumonia, diarrhea and tetanus. 1 pediatric population with the objective of
Approximately 1.6 million deaths due to facilitating the performance of successful
neonatal infections occur annually in clinical studies in children with sepsis. These
developing countries. Neonates are a new definitions shall be utilized in this study in
vulnerable population due to their immature order to conform to international standards
immune system. Other factors, such as being and to differentiate patients suspected of
premature, being low birth weight, presence of sepsis from those who are truly symptomatic.
respiratory problems, maternal infection, and Only symptomatic cases shall be included in
delivery room manipulations further put this study.
neonates at risk for developing sepsis. The OBJECTIVES
incidence of neonatal sepsis ranges from 7.1 to This study was undertaken to determine the
38 cases/1000 live births in Asia in contrast to most common bacterial pathogens of neonatal
1.5-3.5 cases/1000 live births in the United sepsis in the Philippines, the site of infection,
States1. The causes of sepsis vary from one based on clinical presentation and laboratory
geographic area to another. Thus, it is findings and the antibiotic sensitivity patterns
important to create a data base for the of these pathogens.
Philippines regarding the etiologies of neonatal MATERIALS AND METHODS
sepsis. Since antibiotic resistance is a national, A multicenter surveillance and chart review
as well as, a global problem, it is also necessary were conducted in five hospital sites for a
to have information on the susceptibility period of six months- from July to December
patterns of these bacteria. This has implications 2006. The study was conducted at the
with regard the choice of empiric therapy in Philippine General Hospital (PGH) in Manila,
suspected or proven cases thus giving optimal Philippine Children’s Medical Center (PCMC) in
medical treatment from the start of therapy. Quezon City, St. Louis University-Hospital of the
The three most common bacterial Sacred Heart (SLU) in Baguio, Davao Doctors’
pathogens of the neonatal period listed in Hospital (DDH) in Davao City and Chong Hua
foreign literature are of Group B beta- Hospital (CHH) in Cebu City. A mix of private
hemolytic streptococci, Escherichia coli, and and public hospitals was chosen to obtain data
Listeria monocytogenes2 but to date, no cases from patients across all socioeconomic levels.
were published for group B streptococci in the Permission from the administration and
Philippines. This supports data that group B research committees from each hospital was
streptococci are reported more frequently as a obtained prior to the study.
cause of neonatal sepsis in developed countries All neonates less than or equal to 28 days
than in developing areas.2 Previous Philippine who had positive cultures from a sterile site
studies on neonatal sepsis showed and/ or had fulfilled the SIRS criteria as defined
predominant pathogens were that of gram- by the International Pediatric Sepsis Consensus
negative bacilli with Enterobacter spp. and Conference8 in the presence of or as a result of
Klebsiella pneumoniae being the most common suspected or proven infection were diagnosed
organisms isolated.3,4,5,6,7 Pseudomonas spp., to have neonatal sepsis and were included in
Acinetobacter spp. and Candida spp. were also the study.
seen but with lesser incidences.

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 77

Definitions Blood cultures were performed routinely on

Systemic inflammatory response syndrome most neonates suspected of sepsis. Cultures
(SIRS) from other sites (endotracheal tube,
The presence of at least two of the following cerebrospinal fluid, urine) were performed
four criteria, one of which must be abnormal depending on the presenting signs and
temperature or leukocyte count: symptoms of the patients and were not routine
a) Core temperature of >38.5°C or <36°C. for all patients. Blood was cultured using brain
b) Tachycardia, defined as a mean heart rate >2 heart infusion broth according to standard
SD above normal for age in the absence of methods. The isolates were identified by
external stimulus, chronic drugs, or painful standard biochemical tests. Antibiotic
stimuli; or otherwise unexplained persistent resistance of the isolates was performed using
elevation over a 0.5 to 4 hr time period OR for the disk diffusion technique (Kirby-Bauer
children <1 yr old bradycardia, defined as a method) and susceptibility of the isolates was
mean heart rate <10th percentile for age in the interpreted according to the National
absence of external vagal stimulus, β-blocker Committee for Clinical Laboratory Standards
drugs, or congenital heart disease; or otherwise (NCCLS).
unexplained persistent depression over a 0.5-hr The antibiotics used for each patient were
time period. included as one of the items in the surveillance
c) Mean respiratory rate >2 SD above normal forms. First line antibiotics are defined as being
for age or mechanical ventilation for an acute the initial antibiotics given for at least 48 hours,
process not related to underlying while the Final antibiotics are defined as being
neuromuscular disease or the receipt of general the last antibiotics shifted to if the first-line
anesthesia. antibiotics were changed. An antibiotic course
d) Leukocyte count elevated or depressed for was deemed successful if the patient was not
age (not secondary to chemotherapy-induced shifted to another antibiotic and the patient
leukopenia) or >10% immature neutrophils. survived.
Infection RESULTS
A suspected or proven (by positive culture, Demographic Data
tissue stain, or polymerase chain reaction test) From July to December 2006, a total of 289
infection caused by any pathogen OR a clinical neonates admitted at the different study
syndrome associated with a high probability of centers fulfilled the criteria of sepsis. Fifty-five
infection. Evidence of infection includes per cent of the patients were recorded at PGH,
positive findings on clinical exam, imaging, or while the least number of patients came from
laboratory tests (e.g., white blood cells in a CHH in Cebu. More than half of the total
normally sterile body fluid, perforated viscus, patients were male (59%) and this male
chest radiograph consistent with pneumonia, predominance was consistently seen among all
petechial or purpuric rash, or purpura study centers. Approximately 90% were born
fulminans) in a medical facility, 75% of which in a hospital,
A standardized surveillance form was used the rest in a lying in-clinic.
and the same was accomplished by all A great majority of the neonates included in
investigators as each patient with a diagnosis of the study were of low birth weight (CHH-78%,
neonatal sepsis was identified. Data including SLU-69%, PGH-69%, DDH-55%, PCMC-64%). A
place of birth, age of gestation, site of infection, large number of the neonates with symptoms
bacteria isolated and specimen from which it of sepsis were premature (PCMC-51%, Davao-
was isolated, susceptibility pattern, antibiotic 65%, Cebu-56%). For PGH, only 29% of the
used, and outcome of the patient were also neonates were of premature while in SLU it was
included. 48%.

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 78

Table 1. Demographic data of neonatal sepsis patients.

Study Center No. of Patients (%) Sex Place of Birth
Male (%) Female (%) Home (%) Lying in (%) Hospital (%)
SLH 21 (7.2) 11 (52) 10 (48) 0 0 21(100)
CHH 18 (6.2) 14 (78) 4 (22) 1(6) 2(11) 15(83)
DDH 23 (7.9) 15 (65) 8(35) 1(5) 0 22(95)
PCMC 67 (23.1) 38(57) 29(43)
PGH 160 (55.3) 93(58) 67(42) 23(14) 9(6) 128(80)
Total 289 (100) 171(59) 118 (41)

Table 2. Onset of infection neonatal sepsis in five urban hospitals in the Philippines.
Born in hospital Born in Born at home, Born at Total early Total late
Within 72 hrs hospital, Within 72 hrs home onset onset
(%) After 72 hrs (%) (%) > 72 hrs (%) <72 hrs (%) >72 hrs (%)
SLU 21 (100) 0 0 0 21 (100) 0 (0)
CHH 2 (11) 13 (72) 1 (6) 2 (11) 3 (17) 15 (83)
DDH 8 (35) 14 (61) 0 (0) 1 (4) 8 (35) 15 (65)
PCMC 0 1(2) 51(76) 15(22) 51 (76) 16 (24)
PGH 82 (51) 53 (33) 12 (7.5) 13 (8) 94 (59) 66 (41)
Total 113 (39) 81 (28) 63 (22) 31 (11) 177 (61) 112 (39)

Table 3. Site of infection of neonatal sepsis in five urban hospitals in the Philippines.
Study Clinical sepsis Pneumonia Meningitis NEC Cellulitis Urinary Infective endo-
center without focus (%) tract carditis
SLU 4 16 0 0 0 0 0
CHH 11 5 0 0 1 1 0
DDH 12 6 1 1 0 3 1
PCMC 28 39 0 0 0 0 0
PGH 45 98 3 13 1 0 0
Total 100(35%) 164(57%) 4(1%) 14(5%) 2(<1%) 4(1%) 1(<1%)

Table 4. Number of isolates cultures from the different study centers.

Study center No. of patients No Organism isolated (+) Organism isolated
tested
SLH 13 7 (54%) 6 (46%)
CH 18 16 (89%) 2 (11%)
DDH 23 15 (65%) 8 (35%)
PCMC 64 63 (98 %) 1 (2%)
PGH 160 38 (24%) 122 (76%)
Total 278 139 (50%) 139 (50%)

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 79

Table 5. Organisms isolated in five urban hospitals in the Philippines.

Organism isolated Baguio CHH DDH PCMC PGH Number of % of total
isolates
Acinetobacter spp. 7 7 5%
Aeromonas spp. 1 1 0.7%
Alkaligenes spp. 5 5 3.6%
Burkholderia spp. 30 30 21.6%
Enterobacter spp. 1 3 2 6 4.3%
E. coli 1 1 1 3 2.1%
Haffnia alvei 1 1 0.7%
Klebsiella spp. 3 1 11 15 10.8%
Pseudomonas spp.
1 2 55 60 43.2%
(non-aeruginosa)
P. aeruginosa 2 2 1.4%
Salmonella spp. 1 1 0.7%
MRSE 1 5 6 4.3%
Enterococcus spp. 2 2 1.4%
Candida species 1 1 2 1.4%
Total 6 2 8 1 122 139 100%

Table 6. Organisms isolated and sensitivity pattern from pathogens found in St. Louis University
Hospital in patients diagnosed with neonatal sepsis.
Organism Isolated All isolates Sensitive All isolates Resistant Variable No. of
Isolates/No.
tested
Klebsiella oxytoca Cefepime Amikacin, Ampicillin, Ciprofloxacin (50% 3/3
Ceftriaxone, Piperacillin- sens)
Tazobactam, Meropenem,
Imipenem, Cotrimoxazole
Pseudomonas Amikacin, Ciprofloxacin, Cefuroxime, Ceftriaxone, 1/1
maltophilia Piperacillin-Tazobactam Imipenem
Enterobacter Imipenem Netilmicin, Amikacin, 1/1
cloacae Ceftriaxone. Ciprofloxacin
Aeromonas Netilmicin, Ceftriaxone, Ampicillin, Cefuroxime 1/1
hydrophilia Piperacillin-Tazobactam,
Imipenem

Table 7. Organisms isolated and sensitivity pattern from pathogens found in Chong Hua Hospital in
patients diagnosed with neonatal sepsis.
Organism Sensitive Resistant No. of Isolates/No.
Isolated tested
E. coli Gentamicin, Netilmicin, Amikacin, Cotrimoxazole 1/1
Cefuroxime, Ceftazidime,
Cefepime, Ciprofloxacin,
Piperacillin-Tazobactam,
Imipenem,
Staphylococcus Vancomycin, Clindamycin Cefazolin, Cefuroxime, 1/1
epidermidis Ciprofloxacin, Oxacillin,
Piperacillin-Tazobactam,
Erythromycin

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 80

Table 8. Organisms isolated and sensitivity pattern from pathogens found in Davao Doctors Hospital.
Organism Isolated Sensitive Resistant Variable No. of Isolates/No.
tested
Enterobacter cloacae Ciprofloxacin, Meropenem, Ampicillin, Gentamicin, Cefepime (66% sens), 3/3
Imipenem Amikacin, Cefuroxime, Chloramphenicol (66%
Ceftazidime sens), Cotrimoxazole
(66% sens)
Pseudomonas spp. Ceftazidime, Cefepime, Gentamicin, Amikacin, Chloramphenicol (50% 2/2
Ciprofloxacin, Piperacillin- Ampicillin, Cefuroxime, sens)
Tazobactam, Meropenem,
Imipenem, Cotrimoxazole,
Klebsiella ozanae Ceftazidime, Ciprofloxacin, Gentamicin, Amikacin, 1/1
Chloramphenicol, Ampicillin, Cefuroxime,
Meropenem, Imipenem Cefepime, Piperacillin,
Tazobactam, Cotrimoxazole
Escherichia coli Gentamicin, Amikacin, None 1/1
Ampicillin, Cefuroxime,
Ceftazidime, Cefepime,
Ciprofloxacin, Piperacillin,
Tazobactam, Imipenem,
Meropenem

Table 9. Organisms isolated and sensitivity pattern from pathogens found in Philippine Children’s
Medical Center, Quezon City.
Organism Sensitive Resistant No. of
Isolated Isolates/ No.
tested
Salmonella Gentamicin, Amikacin, Cefuroxime, Ceftriaxone, Cotrimoxazole 1/1
spp. Cefepime, Chloramphenicol, Ciprofloxacin,
Piperacillin-Tazobactam, Imipenem

life) with only 39% developed sepsis after dominant bacteria found (Pseudomonas spp,
72hrs. Early onset sepsis was more prevalent in Burkholderia spp., Klebsiella spp.), which
SLU, PCMC and PGH, while for DDH and CHH, comprised 94% of the bacteria isolated. Two
late onset infections predominated. isolates of Methicillin-resistant Staphylococcus
The most common site of infection was the aureus and two isolates of Enterococcus were
lungs, presenting as pneumonia 57%, the only gram-positive bacteria isolated. These
followed by sepsis without an obvious site of were from neonates admitted at PGH and CHH
infection (35%). Other sites were much less only.
frequent which included urinary tract infection, The study centers also had diverse gram
necrotizing enterocolitis, meningitis, cellulitis negative organisms which predominated in
and infective endocarditis. their area. Pseudomonas putida was the most
common isolate in PGH. The sole isolate of
Pathogens Isolated PCMC grew Salmonella spp.
In 50% of patients, an organism was
identified, but the isolation rates differed Antibiotic Resistance
among the centers. The PGH had the highest Resistance patterns varied among the
yield (76%), while PCMC had the lowest with different urban study centers. The
only one culture positive patient among 67 Pseudomonas and Burkholderia isolates from
patients. Gram negative bacteria were the PGH were fully sensitive to Ceftazidime. All

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 81

Klebsiella isolates from SLU were resistant to transferred from other parts of the country and
third generation cephalosporins. In contrast, most deliveries at this hospital were high risk
Klebsiella isolates from PGH and Davao were pregnancies. Consistent with previous studies,
sensitive to third generation cephalosporins but there were more male patients who were
resistant to piperacillin-tazobactam. Tables 7 to Across all centers, Gram-negative bacteria were
11 show the details of antibiotic resistance for the dominant pathogens, although the species
each study site. were varied. Pseudomonas spp. was the most
Antibiotic Usage common bacteria isolated followed by
Antibiotic usage differed slightly from in Burkholderia cepacia. This is consistent with
each institution. In Baguio, Cebu, Davao and reports from other countries in Asia and the
PCMC, first line drugs are a combination of Middle East. A study from Kashir, Iran showed
Ampicillin (Ampicillin-sulbactam occasionally 72.1% of the bacteria isolated from infants with
for Davao) and an aminoglycoside (amikacin or neonatal sepsis were Gram negative with
gentamicin). Unfortunately, in approximately Pseudomonas and Klebsiella as the most
half of these patients, their physicians shifted common organisms.10 In a more recent study
these antibiotics either due to perceived poor from Shiraz, Iran, Early onset sepsis (EOS) was
responses of the patients or receipt of the caused by E. coli, followed by Klebsiella which
culture and sensitivity results changing to a was different from Late onset sepsis (LOS)
drug which is sensitive. which was due to Coagulase negative
Outcome Staphylococcus followed by Enterococcus11.
The overall mortality rate was 14% with the In the Database of the Aga Khan University
highest mortality at PGH (19%) and the lowest Hospital of Pakistan, gram negative organisms
at PCMC (7.5%). The mortality rates of neonatal accounted for 60% (173 of 292) of all their
sepsis for the other centers were as follows: isolates in their newborn unit; with Klebsiella as
DDH-17%, CHH-17% and SLU-9.5%. the most common one found. 12 In the NICU of
DISCUSSION MKCG Hospital in Orissa, India 88.4 % (38 of 48)
The large number of patients collected in a were gram-negative bacilli with Klebsiella and
six-month period reflects the significance of E. coli being the most common ones.13 In
this disease. The number of patients per center another study from India, gram negative
reflects the general economic status of the organisms also predominated (58.5%).14 In a
citizens with large volumes in the public more recent from Pudcherry, India, the most
hospitals (PGH), and much fewer admissions in common etiology of EOS was Klebsiella
the private hospitals (St. Louis Hospital, and pneumoniae.15
Chong Hua Hospital). The PGH may have an
unusually large number because it is a national
referral center wherein patients may have been
diagnosed with sepsis. Early onset sepsis was
also higher in number than late onset sepsis
despite narrowing the definition of the latter to
the first three days of life.
Prior to the antibiotic era, the mortality from
septicemia was more than 90%. In the present
times, mortality remains high, between 20%-
40%. The mortality rate in this study was 14%
which may be due to these hospital having
adequate supportive facilities since they are all
tertiary hospitals.

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 82

Table 10. Top five Organisms isolated and sensitivity pattern from pathogens found in Philippine
General Hospital.
Organism Sensitive Resistant Variable No. of
Isolated Isolates/ No.
tested
Pseudomonas Ceftazidime Gentamicin, Amikacin, Cefuroxime (43% S), Cefotaxime (50% S), 55/1-25
putida Ampicillin, Cefepime (85% S), Ciprofloxacin (35% S), depending on
Piperacliin-Tazobactam (82% S), Meropenem antibiotic
(62% S), Imipenem (23% S),Cotrimoxazole
(88% S)
Burkholderia Cefotaxime, Cefepime, Amikacin, Imipenem Ampicillin (14% S), Ceftazidime (90% S) 16/3-11
mallei Piperacillin-Tazobactam, Ceftriaxone (75% S), Ciprofloxacin (38% S), depending on
Cotrimoxazole Meropenem (29% S) antibiotic
Burkholderia Cefuroxime, Ceftriaxone, Gentamicin, Amikacin, Meropenem (75% S), Imipenem (33% S) 8/1-5
cepacia Ceftazidime, Piperacillin- Ciprofloxacin depending on
Tazobactam, antibiotic
Cotrimoxazole
Acinetobacter Gentamicin, Amikacin, Ampicillin (20% S), Cefuroxime 50% S), 7/1-5
baumanii Ciprofloxacin, Ceftazidime (25% S), Ceftriaxone (50% S), depending on
Imipenem, Cefepime (25% S), Piperacillin-Tazobactam antibiotic
Cotrimoxazole (40% S), Meropenem (50% S)
Burkholderia Ceftazidime, Piperacillin- Gentamicin, Amikacin, Cefepime (66% S) 6/1-5
pseudomallei Tazobactam, Meropenem Netilmicin, depending on
Ceftriaxone, antibiotic

Table 13. First line drugs used and success rates in patients with neonatal sepsis according to hospital.
Drug SLU CHH DDH PCMC PGH All hospitals
No. Succ Succ No. Succ No. Succ No. Succ No. Succ
(%) (%) (%) (%) (%) (%)
Ampicillin 4 1 (25) 15 6 (40) 7 2 (29) 62 33(53) 31 15 (48) 119 57 (48)
Ampicillin- 0 0 9 3(33) 0 0 9 3(33)
Sulbactam
Gentamicin 19 6 (32) 0 5 2 (40) 62 3(53) 0 86 41 (48)
Amikacin 0 15 7 (47) 15 6 (40) 1 0 158 28(26) 189 82(43)
Cefotaxime 1 0 2 1 (50) 0 0 2 2(100) 0 5 5(100)
Piper-Tazo 0 0 2 1(50) 1 0 106 84(76) 106 29(27)
Meropenem 1 0 (0) 0 0 0 3 3(100) 1 0 5 3 (60)
Ceftazidime 0 1 1(100) 5 1 (20) 0 6 6(100) 11 5(45)
Cefepime 1 1 (100) 0 2 0 0 3 1 (33)
Oxacillin 0 1 0 (0) 0 0 4 3(75)
Metronidazole 0 1 1 (100) 1 1 (100) 0 14 11(79) 13 4(31)
Penicillin 13 4 (31) 0 0 0 1 1(100) 15 5 (36)
dominant organism.16 In the Neonatal Network
Traditionally, Group B streptococci (GBS), E. of Burgundy France, from preterm infants (<35
coli and Listeria monocytogenes are considered weeks) with early onset sepsis, gram negative
to be most common in the Western countries, bacilli was the most frequent organism isolated.
but this phenomenon too is changing in certain But for term and near-term infants, GBS
groups of patients. In the latest data from the followed by E. coli was most frequent.17
National Institute of Child Health (NICH) and Gram-positive organisms found in other
human Development Neonatal Research studies include GBS, S. epidermidis, S. aureus,
Network VLBW registry, gram-negative Enterococcus and S. pneumoniae. In our study,
organisms already predominate for early onset only S. epidermidis and Enterococcus were
sepsis (53%) with E. coli being the most found.

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 83

There is also a changing pattern of evaluation undergo reevaluation periodically to

antimicrobial resistance in many parts of the give the most effective treatment depending on
world. In India, gram negative pathogens the changes which occur.
isolated from neonates have high rates of Dr. Darmstadt has recommended the
resistance to almost all antibiotics (50% for following first line therapy in facility settings:
amikacin, 75% for gentamicin, >80% resistance for early and late sepsis: ampicillin and
to third generation cephalosporins, about 40% gentamicin; for early-onset meningitis:
resistance to piperacillin-tazobactam and about ampicillin plus gentamicin; and for late onset
20% resistance to imipenem).18 In Pakistan, a meningitis: ampicillin, gentamicin (or amikacin),
similar trend is being experienced with Gram- and/or cefotaxime.22 In Iran, some experts
negative bacilli having a resistance of >50% for advocate using a third generation
third generation cephalosporins, >50% cephalosporin for treating E. coli and Klebsiella
resistance to Ampicillin and >80% resistance to infections, and Vancomycin for Staphylococcal
gentamicin19. In a recent study from Jordan, infections.11
gram negative bacteria isolated from septic Ampicillin and an aminoglycoside are still
neonates were highly resistant to Ampicillin being used as first line drugs in most hospitals
(82.8-100%), Gentamicin (41.4-79.5%), in the Philippines. These antibiotics seem to be
Ceftazidime (46-50%) and Piperacillin- less useful in the hospitals studied based on
Tazobactam (21.4-54.3%) 20. Only Imipenem resistance rates and antibiotic outcomes seen.
and Ciprofloxacin showed good sensitivity. In Although the mortality was not high, the
the United States, resistance to ampicillin is the hospital stays of these patients were presumed
main concern (57%) for the gram negative to be longer because of the initial non-response
bacteria that they have isolated21. In Iran, to the regimen. This combination may still be
Klebsiella isolates from neonates showed 100% given to patients with risk factors of sepsis
resistance to ampicillin, 31% resistance to (PROM, maternal fever, etc) but with no signs
ceftriaxone, 46% resistance to Amikacin and and symptoms and are still awaiting results of
27% to gentamicin.10 Our study shows varying blood cultures or other laboratory indicators
resistance to the different groups of antibiotics for sepsis. For symptomatic patients (those
among the different hospitals. who strictly fulfill the criteria of neonatal sepsis
The Infectious Diseases and Tropical as defined in this study) the use of a third
Medicine Section of the Department of generation cephalosporin plus amikacin may be
Pediatrics of the PGH has an ongoing more cost-effective, although clinical trials may
surveillance since the mid 1990’s of nosocomial be needed to confirm this. Once culture and
infections of which many of the neonates born sensitivity results are available, antibiotics
at their institution are also included. There has should be adjusted according to this new
been an increasing trend of multi-drug resistant information, while taking into consideration the
gram negative bacilli in the neonatal intensive clinical response of the patient.
care unit. This is the reason that for certain The reason why there is a reluctance to
subsets of patients (low birth weight neonates promote third generation cephalosporins as
born at PGH), piperacillin-tazobactam plus an first line drugs is that although they are poor
aminoglycoside had been the first line regimen. inducers of beta-lactamase expression, they are
But due to this practice and other factors, such sensitive to these enzymes and are significantly
as improved infection control, the pattern associated with repressor mutations. The
seems to be changing for the better because importance of limiting its use only to
the gram negative bacteria now are more symptomatic patients (fulfilling the criteria of
susceptible to third generation cephalosporins. sepsis as defined in this study) should be
These regimens is has to is undergoing re- emphasized and duration of therapy must be

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 84

well-defined. Potentially septic patients who third generation cephalosporin (ceftazidime, or

are asymptomatic may be started with cefotaxime) plus an aminoglycoside should be
Ampicillin and an Aminoglycoside. In treatment started. Reevaluation should again be done in
protocols such as in France, antibiotic therapy 48-to-72 hours. Adequate supportive measures
is discontinued as soon as clinical infection is aside from antibiotics as well as good infection
resolved and blood CRP concentration is <10 control should not be forgotten. More research
mg/L. Their treatment duration is usually seven is also needed to determine the shortest
days for septicemia, 14 days for GBS meningitis effective duration of treatment.
and 21 days for gram negative bacillary
meningitis. Acknowledgements:
This research was supported in part, by an
CONCLUSIONS unrestricted study grant from Pan Pharma
Neonatal sepsis is still an important cause of Philippines. We would also like to thank all the
morbidity and mortality in our country. The pediatric consultants and residents from the
most common site of infections is the lungs, Saint Louis University Hospital, Chong Hua
presenting as pneumonia, followed by clinical Hospital, Davao Doctors Hospital, Philippine
sepsis without a focus. A changing pattern of Children’s Medical Hospital and Philippine
bacteria isolated as well as their resistance has General Hospital for their generous
been observed. Gram negative bacteria were cooperation.
the dominant bacteria found (Pseudomonas
spp, Burkholderia spp., Klebsiella spp.) which REFERENCES
comprised 94% of the bacteria isolated. Two 1. Vergano S, Sharland M, Kazembe P, Mwansambo C,
Heath P. Neonatal Sepsis: An International Perspective.
isolates of Methicillin resistant Staphylococcus Arch. Dis. Child Fetal Neonatal Ed. 2005; 90; 220-224.
aureus and two isolates of Enterococcus were
the only gram positive bacteria isolated. 2. Feigin RD, Cherry JD, Demmler G, Kaplan S. Textbook
th
Increasing resistance to commonly used of Pediatric Infectious Diseases 5 Edition 2004; 929-930.
antibiotics is seen. Ampicillin and an
3. Jesswani RS. Early Onset Sepsis in SGA Term Neonates
aminoglycoside were the most common in Davao Medical Center. PIDSP Journal 2001:5:26-31.
regimens used but were only successful in less
than 50% of cases. A third generation 4. Garcia D, Soriano. Etiology of Neonatal Sepsis at
cephalosporin plus an aminoglycoside might be Cardinal Santos Medical Center NICU, 1993-
a more cost-effective combination but its use 1999.unpublished
should be limited only to symptomatic patients 5. Asuncion. Etiologic Agents of 129 Cases of Septicemia
who fulfill the sepsis criteria. at the Makati Medical Center NICU, 1995, unpublished

RECOMMENDATIONS 6. Bravo L. Etiologic Agents of Late-Onset Neonatal

Surveillance should be continued because Sepsis at the Philippine Children’s Medical Center, 1995,
changes in pathogens as well as susceptibility unpublished
patterns are expected. For asymptomatic 7. Salaveria-Imperial, Neonatal Sepsis at the Philippine
neonates but with risk factors for sepsis General Hospital. PIDSP Journal 2002; 6: 5-9.
(prolonged rupture of membranes, maternal
fever, chorioamnionitis, etc.) ampicillin plus an 8. Goldstein B, Giroir B, Randolph A. International
aminoglycoside may be started and re-evaluate Pediatric Sepsis Consensus Conference: Definitions for
Sepsis and Organ Dysfunction in Pediatrics. Pediatric
its need in 48-to-72 hours. If the patient is Critical Care Med. 2005; 6: 1, 2-8.
asymptomatic and cultures are negative,
antibiotics should be discontinued. If a neonate 9. Ann L Anderson-Berry, MD Neonatal Sepsis at
is symptomatic (fulfills the sepsis criteria), a http://emedicine.medscape.com/article/978352-
overview

Downloaded from www.pidsphil.org

 PIDSP Journal 2011 Vol 12 No.2
Copyright ® 2011 85

explore simplified antimicrobial regimens for the

10. Movahedian AH, Moniri R and Mosayebi Z. Bacterial treatment of neonatal sepsis. Meeting Report, 2002.
Culture of Neonatal Sepsis. Iranian J Publ Health 2006;
35(4): 84-89.

11. Shahian M, Pishva N, Kalani M. Bacterial Etiology and

Antibiotic Sensitivity patterns of Early-Late Onset
Neonatal Sepsis among Newborns in Shiraz, Iran 2004-
2007. Iran J Med Sci 2010; 35 (4): 293-298.

12. Bhutta ZA, Yusf K, Neonatal Sepsis in Karachi: Factors

Determining Outcome and Mortality. J of Trop Ped 1997;
43: 65-70

13. Mahapatra A, Ghosh SK, Mishra S, Pattnaik D,

Pattnaik K. Mohanty SK. Enterobacter cloacae: A
predominant pathogen in neonatal septicemia. Indian J
Med Microbiol 2002; 20:110-112.

14. Agnihotri N, Kaistha N, Gupta V. Antimicrobial

Susceptibility of Isolates from Neonatal Septicemia. Jpn. J
Infect Dis 2004; 57: 273-275.

15. Ananthakrishnan S, Gunanseharan D. Etiology and

Risk Factors for Early Onset Neonatal Sepsis. Indian
Journal of Medical Microbiology 2009; 27: 279.

16. NICH Stoll BJ, Hansen NI, et al. Very low birth weight
preterms with early onset neonatal sepsis: the
predominance of gram negative infections continues in
the NICHD Neonatal Research Network, 2002-2003.
Pediatr Infect Dis J 2005; 24(7): 635-9

17. Labenne M, Michaut F, Gouyon b, et al. a Population-

based observational study of restrictive guidelines for
antibiotic therapy in early-onset neonatal infections.
Pediatr Infect Dis J 2007; 26(7): 593-599.

18. Isaacs D. Neonatal Sepsis: The antibiotic crisis. Indian

Pediatrics 2005; 42:9-13

19. S Rahman, A Hameed, M T Roghani, Z Ullah.

Multidrug resistant neonatal sepsis in Peshawar,
Pakistan. Arch Dis Child Fetal Neonatal Ed 2002; 87:F52–
F54

20. Khassawneh M, Khader Y, Abuqtaish N. Clinical

features of neonatal sepsis by resistant gram negative
bacteria, 2009; 51: 332-336.

21. Baltimore RS, Huie SM, Meek JI, Schuchat A, O Brien

K. Early-Onset Neonatal Sepsis in the Era of Group B
Streptococcal Prevention. Pediatrics 2001; 108; 1094-
1098

22. Department of Child and Adolescent Health and

Development World Health Organization. Meeting to

Downloaded from www.pidsphil.org