Infection Control Practices Reduce Nosocomial Infections and Mortality in Preterm Infants in Bangladesh

Abstract

The skin is a potential source for invasive infections in neonates from developing countries such as Bangladesh, where the level of environmental contamination is exceedingly high. A randomized controlled trial was conducted from 1998 to 2003 in the Special Care Nursery of a tertiary hospital in Bangladesh to test the effectiveness of topical emollient therapy in enhancing the skin barrier of preterm neonates less than 33 weeks of gestational age. In the initial months of the study, the infection and mortality rates were noted to be unacceptably high. Therefore, an infection control program was introduced early in the trial to reduce the rate of nosocomial infections.
After a comprehensive review of neonatal care practices and equipment to identify sources of nosocomial infections, a simple but comprehensive infection control program was introduced that emphasized education of staff and caregivers about measures to decrease risk of contamination, particularly hand-washing, proper disposal of infectious waste, and strict asepsis during procedures, as well as prudent use of antibiotics.
Infection control efforts resulted in declines in episodes of suspected sepsis (47%), cases of culture-proven (61%) sepsis, patients with a clinical diagnosis of sepsis (79%), and deaths with clinical (82%) or culture-proven sepsis (50%).
The infection control program was shown to be a simple, low-cost, low-technology intervention to reduce substantially the incidence of septicemia and mortality in the nursery.

Full text

Original Article
Infection Control Practices Reduce Nosocomial Infections and
Mortality in Preterm Infants in Bangladesh
Gary L. Darmstadt, MD, MS
A.S.M. Nawshad Uddin Ahmed, MBBS, FRCP
Samir K. Saha, PhD
MAK Azad Chowdhury, MBBS, FRCP
Muhammad Asif Alam, MBBS, MS
Mahamuda Khatun, MBBS
Robert E. Black, MD, MPH
Mathuram Santosham, MD, MPH
OBJECTIVE:
The skin is a potential source for invasive infections in neonates from
developing countries such as Bangladesh, where the level of
environmental contamination is exceedingly high. A randomized
controlled trial was conducted from 1998 to 2003 in the Special Care
Nursery of a tertiary hospital in Bangladesh to test the effectiveness of
topical emollient therapy in enhancing the skin barrier of preterm
neonates less than 33 weeks of gestational age. In the initial months of
the study, the infection and mortality rates were noted to be unacceptably
high. Therefore, an infection control program was introduced early in the
trial to reduce the rate of nosocomial infections.
STUDY DESIGN:
After a comprehensive review of neonatal care practices and equipment to
identify sources of nosocomial infections, a simple but comprehensive
infection control program was introduced that emphasized education of
staff and caregivers about measures to decrease risk of contamination,
particularly hand-washing, proper disposal of infectious waste, and strict
asepsis during procedures, as well as prudent use of antibiotics.
RESULTS:
Infection control efforts resulted in declines in episodes of suspected sepsis
(47%), cases of culture-proven (61%) sepsis, patients with a clinical
diagnosis of sepsis (79%), and deaths with clinical (82%) or culture-
proven sepsis (50%).
CONCLUSION:
The infection control program was shown to be a simple, low-cost, low-
technology intervention to reduce substantially the incidence of
septicemia and mortality in the nursery.
Journal of Perinatology (2005) 25, 331–335. doi:10.1038/sj.jp.7211283
Published online 17 February 2005
INTRODUCTION
Nosocomial infections are a principal cause of morbidity and
mortality among hospitalized neonates in developing countries.
1–3
The incidence of sepsis in premature infants in developing
countries is particularly high, estimated at 30 to 60%,
4,5
and
mortality is 40 to 70%.
6–11
Limited information is available,
however, from many developing countries on the impact of
strategies to prevent and control these infections.
Model infection control programs to detect and prevent
nosocomial infections and minimize the emergence and spread of
antibiotic-resistant strains within hospital nurseries have been
established in developed countries, and some countries in Latin
America have adapted these programs based on local needs and
resources.
12,13
Many aspects of these programs, however, are not
readily applicable to health-care settings with more limited
resources, such as many health facilities in south Asia and sub-
Saharan Africa, where the majority of global neonatal deaths
occur. Infection control programs should incorporate two broad
strategies fundamental to the control of nosocomial infections:
reducing bacterial transmission within the nursery through
infection control interventions, and reducing selective pressure for
the emergence of antibiotic resistance through promotion of
rational use of antibiotics and control of their misuse. Although
resource limitations will restrict the scope of infection control
programs in many settings, the ability to use antibiotics rationally
and to implement simple infection control procedures, such as
hand-washing, are not beyond the reach of health-care facilities
with even very limited resources.
In the course of implementing an intervention trial at Dhaka
Shishu Hospital, Bangladesh, to assess the impact of topical
Address correspondence and reprint requests to Gary L. Darmstadt, MD, Department of
International Health E8153, Bloomberg School of Public Health, Johns Hopkins University, 615
North Wolfe Street, Baltimore, MD 21205, USA.
Presented in part at 10th Asian Conference on Diarrhoeal Diseases and Nutrition, Dhaka,
Bangladesh, 2004; and Third National Annual Conference and Scientific Session, Bangladesh
Neonatal Forum, Dhaka, Bangladesh, 2002.
Department of International Health (G.L.D., R.E.B., M.S.), Bloomberg School of Public Health,
Johns Hopkins University, Baltimore, MD, USA; Saving Newborn Lives Initiative (G.L.D.), Save the
Children USA, Washington, DC, USA; Department of Neonatology (A.S.M.N.U.A.; M.A.K.A.C.),
Institute of Child Health, Dhaka Shishu Hospital, Dhaka, Bangladesh; Department of Microbiology
(M.A.A., M.K., S.K.S.), Institute of Child Health, Dhaka Shishu Hospital, Dhaka, Bangladesh
Journal of Perinatology 2005; 25:331–335
r
2005 Nature Publishing Group All rights reserved. 0743-8346/05 $30
www.nature.com/jp 331

emollient therapy to enhance skin barrier function on the
incidence of infections in preterm infants, the results of which will
be reported elsewhere,
14
we found alarmingly high rates of
nosocomial infections and mortality. This led to the development
and evaluation of the impact of a low-cost infection control
program, including antibiotic use guidelines and emphasis on
hygiene, on the incidence of nosocomial infections and neonatal
mortality rates among preterm infants in the Special Care Nursery.
METHODS
Study Setting and Patient Population
A randomized, controlled intervention trial was initiated in
December 1998 at Dhaka Shishu Hospital, Bangladesh, aimed to
reduce infections in preterm infants less than 33 weeks gestational
age through topical therapy with skin-barrier-enhancing
emollients.
15–17
Dhaka Shishu Hospital is the largest tertiary-level
pediatric hospital in Bangladesh, with 345 beds, including 16 cribs
and eight isolettes in the Special Care Nursery. Patients come to
Shishu Hospital from all over Bangladesh, although the principal
catchment area includes a population of approximately 16 million
from Dhaka and adjoining districts.
Gestational age of neonates admitted to the Hospital was
determined by maternal dates and Ballard and Dubowitz
criteria;
18,19
the average of the three values was used. Infants were
enrolled in the study if their average gestational age was less than
33 weeks, their chronological age was less than 72 hours, and
parental informed consent was obtained. Exclusion criteria
included: (1) admission for a major surgical procedure attended by
a high rate of infectious complications; (2) clinically evident skin
infection (confirmed by surface culture); (3) generalized skin
disease likely to produce a defect in epidermal barrier function; (4)
a structural defect of the skin involving greater than 5% body
surface area (e.g., congenital blistering disorder); (5) a major
congenital anomaly likely to predispose to infection; or (6) known
immunodeficiency.
During the initial phase of the trial, nosocomial infection and
mortality rates were found to be unacceptably high at 35 and 61%,
respectively, and the study Data Safety and Monitoring Board
encouraged the investigators to assess and address the situation.
Previously, epidemic outbreaks of nosocomial infections had
occurred in the Special Care Nursery due to Flavobacterium
meningosepticum, Salmonella typhi Group B, and Klebsiella
pneumoniae.
Infection Control Program
Given the high rate of nosocomial infections and mortality among
preterm neonates in the Special Care Nursery, an infection control
program was introduced in January 1999. In the first phase, a
review of neonatal care practices and equipment in the Nursery was
undertaken to identify potential sources of nosocomial infections,
including the staff, particularly their hands, and equipment used
in patient care, including incubators, nasal prongs, and suction
devices. Antibiotic use and intravenous catheter care were also
reviewed. Based on the findings of the review, educational sessions
were held with the nurses in the Nursery, who, in turn, held daily
sessions with patient family members (generally the mother, or, if
the mother was unavailable, another female relative) who resided
in the unit and assisted in caring for their own hospitalized
newborn. Emphasis was placed initially on personal hygiene of the
nurses and caregivers, particularly hand-washing, nail care, and
bathing. Emphasis was also placed on minimum handling of very
preterm infants, newborn bathing techniques to minimize risk for
hypothermia, and promotion of breastfeeding. Nurses were also
taught clean umbilical cord care and regular shifting of infant
posture and mild physiotherapy. Ward physicians were given
refresher training on early recognition and culturing of infants
with suspected sepsis, and appropriate antibiotic therapy.
In 2000, the Hospital formed an Infection Control Committee
comprising the Hospital Director, Epidemiologist, Head of
Microbiology, Nursing Supervisor, and a Staff Nurse. Further
measures the Committee took to prevent nosocomial infections
included:
1. Hand-washing: A freestanding wash basin containing Savlon
antiseptic was set at an accessible place near the sink in the
ward to further encourage hand-washing by doctors and nurses.
A poster was placed at the basin as a reminder of the ideal way
to wash hands.
2. Waste disposal: Colored buckets were placed in the ward; blue
buckets for normal waste and red for potentially infectious or
dangerous waste such as syringes, needles, broken tubes, etc.
3. Introduction of disposable needles: Private-paying patients
(20% of admissions) were asked to provide disposable syringes
and needles; for non-paying patients, disposable needles were
donated by the hospital, used and discarded, although in some
cases, due to lack of resources, syringes were used more than
once (after cleaning and boiling).
4. Visitation: Restrictions were placed on the number of visitors in
the wards, and visitors were required to wash their hands before
going to a patient’s bedside.
5. Cohorting: Surgical postoperative patients with wound infec-
tions were isolated from other patients.
6. Care of intravenous and urinary catheters: The staff was
trained in standard guidelines for use of these devices and an
infection control nurse was appointed to monitor their use.
7. Training and supervision: Weekly classes on infection control
procedures on the wards were conducted by the Nursing
Supervisor with nurses and caregivers. The head of the Infection
Control Committee also conducted classes on a regular basis
with the nurses, with special reference to specimen collection,
aseptic measures, and basics of microbiology related to
Darmstadt et al. Infection Control Reduces Neonatal Mortality
332 Journal of Perinatology 2005; 25:331–335

nosocomial infections. Practices were reinforced through daily
supervision by the Nursing Supervisor.
In a third stage of the infection control program begun in 2001, a
new Nursing Supervisor with interest and training in infection
control began to assemble groups of three to four mothers and other
caregivers on the ward to reinforce health and infection control
messages. A log book of these sessions was maintained, including
documentation of the issues discussed. Emphasis was further placed
on increasing awareness of the importance of infection control and
personal hygiene. Nurses inspected the mother’s hygiene, including
the fingernails, each day. Caregivers were given a clean gown each
day to wear at the bedside when handling the baby, and were given a
box in which to keep dirty items.
Hand-washing was highlighted as the most important measure
to prevent nosocomial infections. Caregivers and staff were
instructed to wash hands at the sink with soap and water, followed
by hexisol (2.5% v/v chlorhexidine gluconate solution in 70% w/w
isopropyl alcohol) before handling the babies. A towel at the sink
for use by caregivers was changed each day; prior to the
intervention, this was not regularly monitored. Study nurses and
physicians began to use disposable tissues, rather than the towel, to
dry their hands.
Additional actions taken as part of the third stage of the infection
control program included: (1) a separate suction device was
designated for each baby and placed at the bedside; (2) incubators
were cleaned with soap and water and Savlon, and dried in-between
patients and at least once per week if the baby was in the incubator
for a prolonged period; (3) emphasis was placed on keeping the
floors clean; (4) new bed mattresses were purchased without crevices
in which ‘‘dirt’’ could accumulate, and were cleaned with Savlon
every day; (5) nurse workload was restructured so that one nurse
would feed all the babies and a different nurse would change and
clean the beds, rather than sharing duties; (6) further instruction
and supervision was provided to ensure that invasive procedures
(e.g., venipuncture, catheter placement, lumbar puncture) were
performed using strict asepsis, including use of iodine and alcohol
(not just alcohol as practiced previously); and (7) babies with
clinically suspected or bacteriological-proven septicemia were
cohorted, if possible. Finally, judicious antibiotic use was reviewed,
including initial empiric use of ampicillin and gentamicin as first-
line treatment for suspected sepsis, prompt discontinuation of
antibiotics if cultures were negative and the baby was well clinically,
and appropriate and prompt adjustment of the antibiotic regimen
according to the sensitivity pattern of the isolate.
Outcomes
Nosocomial infection, or confirmed hospital-acquired sepsis, the
primary outcome measure, was defined as a noncontaminated
(e.g., Diphtheroids and Micrococcus spp. were excluded) positive
blood or cerebrospinal fluid (CSF) culture taken after 3 or more
days of hospitalization, where the culture did not grow the same
organism as the most recent previous culture in the same infant
(i.e., cultures that were positive on the basis of persistent infection
were excluded). If an infant had both a positive CSF and a positive
blood culture, or two positive blood cultures on the same day, only
one nosocomial infection on that day was counted.
Suspected sepsis was defined as any episode of illness for which
blood cultures were drawn. A patient was identified as having
clinical sepsis if, on the basis of clinical judgment, the attending
physician assigned sepsis as a final diagnosis at the end of
hospitalization.
Data Analysis
All analyses were conducted using STATA 7.0 statistical software
(Stata Corporation, College Station, TX, USA).
RESULTS
Nosocomial Infections
Epidemics. In 1998, before our study was initiated, nosocomial
outbreaks of group B Salmonella accounted for 58 documented
Figure 1. Number of episodes of suspected sepsis and confirmed sepsis
among preterm neonates hospitalized at Dhaka Shishu Hospital,
Bangladesh, during institution of infection control measures during
different time periods: (A) Review for potential sources of infections;
educational sessions with families and staff, emphasizing hygiene,
essential newborn care practices, and antibiotic use; (B) Infection
Control Committee formed; consolidation of infection control measures,
training and supervision; (C) Infection Control Nursing Supervisor
hired; further emphasis on education regarding hygiene and antibiotic
use.
Infection Control Reduces Neonatal Mortality Darmstadt et al.
Journal of Perinatology 2005; 25:331–335 333

cases of nosocomial infection in the neonatal ward, whereas 12
such cases occurred in 1999, and there were no outbreaks of
Salmonella infection in 2000. Similarly, outbreaks with K.
pneumoniae accounted for 62 infections in 1999, 14 in 2000, and
none in 2001.
Sporadic infections. Significant declines from 1999 to 2001
were found in cases of suspected sepsis [i.e., total number of
cultures obtained (Figure 1) and number of cultures obtained per
patient (data not shown) for suspected sepsis] and culture-proven
sepsis [i.e., total number of positive cultures (Figure 1) and
proportion of cultures obtained per patient that were positive (data
not shown)]. Number of patients diagnosed with clinical sepsis
(Figure 2) was also reduced. Mortality declined over the study
period, with reduced numbers of patients who died with clinical
sepsis and confirmed sepsis (Figure 3).
DISCUSSION
This study has important implications for infection control and
antibiotic use policies in neonatal care units in developing
countries. Despite the proven benefit of infection control
practices in developed country nurseries, there is little
documentation of successful infection control efforts in the
developing world, where rates of nosocomial infections remain
unacceptably high and contribute to high mortality rates due to
sepsis in health facilities. The high mortality and infection rates
and alarmingly high rates of antibiotic resistance that we
encountered among the nosocomial isolates in the Special Care
Nursery at our study site
2
led us to institute low-technology,
low-cost infection control procedures, with encouraging results
that may be widely applicable to other developing country
neonatal care settings.
We demonstrated that a relatively simple infection control
program, aimed at fundamental sources of nosocomial infection
identified through a systematic review of the care practices and
equipment in the nursery, can have a marked impact on
nosocomial infection and mortality rates in hospitalized preterm
infants. Some of the key elements of the program appeared to be
designation of an infection control nurse on the ward and
formation of a hospital Infection Control Committee to oversee
training activities and stimulate ownership of the problem on the
part of the staff; systematic review and planning to redress principal
potential sources of environmental contamination in the nursery;
regular training and supervision of nurses and caregivers on
routine infection control practices, particularly hand-washing; and
focused attention on judicious use of antibiotics. Little to no cost
was required to implement the program except for materials such
as cleansers and time for the various educational activities with
staff and caregivers. The will and efforts of key individuals,
particularly the Nursing Supervisor, to improve infection control
practices was pivotal in achieving success. Other nurseries
throughout the developing world could readily implement such a
program.
Acknowledgements
This study was supported by the Thrasher Research Fund, the Johns Hopkins
Family Health and Child Survival Cooperative Agreement with the United Sates
Agency for International Development, the Society for Pediatric Dermatology and
Save the Children USA through a grant from the Bill and Melinda Gates
Foundation. We thank the members of the Data Safety and Monitoring Board
(DSMB) for the parent intervention trial for their encouragement to pursue
infection control measures in the nursery. DSMB members included Kim
Mulholland, University of Melbourne, Barbara Stoll, Emory University, and
Figure 2. Number of preterm infants with clinical sepsis during
hospitalization at Dhaka Shishu Hospital, Bangladesh.
Figure 3. Number of patients who died with clinical sepsis and with
confirmed sepsis.
Darmstadt et al. Infection Control Reduces Neonatal Mortality
334 Journal of Perinatology 2005; 25:331–335

William Blackwelder (retired), National Institutes of Health (USA). We thank
Rachel Haws for editorial assistance in preparation of the manuscript, and
Maksuda Islam for microbiological technical assistance.
References
1. Ahmed A, Chowdhury M, Hoque M, Darmstadt G. Clinical and
bacteriological profile of neonatal septicaemia in a teriary level paediatric
hospital in Bangladesh. Indian Pediatr 2002;39:1034–9.
2. Ahmed A, Chowdhury M, Saha S, Darmstadt GL. Skin colonization of
newborns with multiply resistant bacteria: implications in outbreak of sepsis
in the nursery. Proceedings of the Third National Annual Conference on
Scientific Session Bangladesh Neonatal Forum 2002.
3. Ahmed A, Saha S, Chowdhury M, Law P, Ahmed S, Darmstadt G. Topical
skin barrier therapy and infection control measures improve survival of
hospitalized very low birth weight neonates. 10th Asian Conference on
Diarrhoeal Disease Nutrition 2003;89(Abstract 134).
4. Sharma A, Kutty K, Sabharwal U, Rathee S, Mohan H. Evaluation of
sepsis screen for diagnosis of neonatal septicemia. Indian J Pediatr
1993;60:559–63.
5. Puri J, Revathi G, Faridi M, Talwar V, Kumar A, Parkash B. Role of body
surface cultures in prediction of sepsis in a neonatal intensive care unit.
Ann Trop Pediatr 1995;15:307–11.
6. Bhakoo O. Neonatal bacterial infections at Chandigarh F a decade of
experience. Indian J Pediatr 1980;47:419–24.
7. Bhutta Z, Naqvi S, Muzaffar T, Farooqui B. Neonatal sepsis in Pakistan.
Acta Paediatr Scand 1991;80:596–601.
8. Stoll B. The global impact of neonatal infection. Clin Perinatol 1997;
24:1–21.
9. Das B, Mishra R, Mishra O, Bhargava V, Prakash A. Comparative outcome
of low birth weight babies. Indian Pediatr 1993;30:15–21.
10. Singh M, Paul VK, Deorari A, Ray D, Murali M, Sundaram K. et al.
Strategies which reduce sepsis related neonatal mortality. Indian J Pediatr
1988;55:955–60.
11. Bartlett A, Paz de Bocaletti M, Bocaletti M. Neonatal and early postneonatal
morbidity and mortality in a rural Guatemalan community: the importance
of infectious diseases and their management. Pediatr Infect Dis J
1991;10:752–7.
12. Ponce-de-Leon S. The needs of developing countries and the resources
required. J Hosp Inf 1991;18(Suppl A):376–81.
13. Saez-Llorens XCd, Wong M, Castano E, De Suman O, De Moros D, De
Atencio L. Impact of an antibiotic restriction policy on hospital expenditures
and bacterial susceptibilities: a lesson from a pediatric institution in a
developing country. Pediatr Infect Dis J 2000;19:200–6.
14. Darmstadt G, Saha S, Ahmed A, Chowdhury M, Law P, Ahmed S. et al.
Topical therapy with skin-barrier enhancing emollients prevents nosocomial
infections in preterm infants in Bangladesh: a randomized, controlled
clinical trial. Lancet 2005; in press.
15. Darmstadt GL, Mao-Qiang M, Chi E, Saha SK, Ziboh VA, Black RE, et al.
Impact of topical oils on the skin barrier: possible implications for neonatal
health in developing countries. Acta Paediatr 2002;91:1–9.
16. Darmstadt GL, Badrawi N, Law PA, Ahmed S, Bashir M, Iskander I. et al.
Topically applied sunflower seed oil prevents invasive bacterial infections in
preterm infants in Egypt: a randomized, controlled clinical trial. Pediatr
Infect Dis J 2004;23:719–25.
17. Darmstadt GL, Saha S, Ahmed A, Khatun M, Chowdhury M. The skin as a
potential portal of entry for invasive infections in neonates. Perinatol
2003;5:205–12.
18. Ballard J, Khoury J, Wedig K, Wang L, Ellers-Walsman B, Lipp R. New
Ballard Score, expanded to include extremely premature infants. J Pediatr
1991;119:417–23.
19. Dubowitz L, Dubowitz V, Goldberg C. Clinical assessment of gestational age
in the newborn infant. J Pediatr 1970;77:1–10.
Infection Control Reduces Neonatal Mortality Darmstadt et al.
Journal of Perinatology 2005; 25:331–335 335