International Journal of Drug Development and Research

Keywords

Neonatal sepsis, antibiotics, gram positive cocci, Neonatal Intensive Care Unit (NICU EOS, LOS) ,

Introduction

The human body has both local and systemic responses to microbes that cross epithelial barriers and attack underlying tissues. The basic signs of response to microbial invasion are fever, leukocytosis or leukopenia, tachypnea, and tachycardia. These are referred to as systemic inflammatory response syndrome (SIRS). In cases where SIRS is suspected or confirmed, the patient is said to have sepsis, and it's referred to as septicemia when the presence of microbes or their toxins is confirmed in blood cultures.(1) When these infections affect infants less than four weeks of age they are referred to as neonatal septicemia. Many factors render neonates more susceptible to infections.

Neutrophils or polymorphonuclear (PMN) cells are essential for efficient killing of bacteria. These cells are poor in chemotaxis and killing capability in neonates.(2) The diversity and the antimicrobial sensitivity of microbes causing sepsis may be different for each Neonatal Intensive Care Unit (NICU), and it may vary from time to time within the same unit.(3)

Types of Neonatal Sepsis

There are two types of neonatal sepsis, early-onset sepsis (EOS) and late-onset sepsis (LOS). They are categorized according to time of onset of the episode of bacteremia or disease (table 1). In EOS, the infection is transmitted from mother to neonate during the prenatal, intrapartum or postpartum periods.(4) Low birth weight and prematurity are risk factors for developing EOS, (5)and intrapartum fever, prolonged rupture of membranes for more than 18 hours, and vaginal delivery are factors which increase risk of maternal transmission of the infection.(6)

Causative Organisms of Neonatal Sepsis

Isolation of organisms from blood cultures is the gold standard for diagnosis of septicemia,(7) although not all cultures taken from septic neonates displayed a positive results.(8) In a study incidence rate of positive culture was only 1.9%, while 50% were clinically diagnosed and received antibiotics.(9)

The range of organisms that cause neonatal sepsis varies over time, from area to area, and from one hospital to another in the same city or district.(10) Gram negative organisms were the most common cause of neonatal sepsis in Europe and America in the 1960s.(11) Group B streptococcus (GBS) has became prevalent during the 1970s and coagulase negative Staphylococcus (CoNS) during the late 1980s and 1990s. Before 1996 GBS was the leading cause of EOS with incidence rate of 2 to 3 cases per 1,000 live births.(12) In 1999, due to introduction of the intrapartum antibiotic prophylaxis guidelines by the Centers for Disease Control and Prevention,(13) the rates of GBS disease have dramatically decreased by 70% to reach 0.5 cases per 1,000 live birth,(14) and further decreased to 0.32 cases per 1,000 live births in 2006.(15) The incidence rate of GBS in EOS did not change significantly in 2000s..(16) Another study in a developing country has shown gram negative organisms to be the major cause of neonatal sepsis. This might be due in part to over-the-counter sale of antibiotics.(17) A study from Nigeria showed almost equal contribution of gram negative and gram positive as causative agents for neonatal sepsis. The investigators found that 50.5% of isolates were gram negative bacilli (GNB) and 49.5% isolates were gram positive cocci (GPC).(18) While in a study in Busan, South Korea GPC was found to be the major causative agent of neonatal sepsis. The investigators found that over 60% of the isolates were GPC and around 32% were GNB.(19) In the Middle East, a study from Iraq had shown different patterns of isolates with GNB contributing 58% of isolates in a neonatal intensive care in Al-Anbar governorate, and 42% were GPC.(20) Investigators from the Indiansubcontinent viewed retrospectively patterns of bacterial isolates in neonatal intensive care unit in one hospital in Chandigarh, India over a period of five years.(21) The patterns of the isolates were almost similar to that of the Iraqi study with 58.5% of the identified isolates GNB and 41.5% GPC. This came in agreement with the recognized internationally move of pathogens causing neonatal sepsis, towards the dominance of GNB.(22) The dominance of GNB may be due in part to increase use of intrapartum antibiotics.(23) In Kuwait different patterns of distribution of GPC and GNB as neonatal sepsis causative agents were found. One study revealed more than 74% of isolates were GPC while GNB made up less than 10% of the isolates.(24) In Saudi Arabia the patterns of isolates in one neonatal intensive care unit in Riyadh was found to be similar to that in Kuwait, with 69% of isolates GPC while 25% were GNB.(25)

In the study from Saudi Arabia only 3% of GPC were GBS. However a study from Dubai, United Arab Emirates found GBS as the most common causative agent compromising 23% of the isolates.(26) The pattern of isolates of GBS were widely different between Saudi Arabia and United Arab Emirates reflecting differences in the causative agents of neonatal sepsis, and different from the pattern of isolates of GBS in United States.(27) In contrast, a study from Peshawar, Pakistan, investigators found no GBS among the isolates, while GNB were 70.4% and GPC were 29.5.(28) In a hospital in Jerusalem, Israel investigators found that 6.6% of the isolates were GBS, 61% were GPC and 33.1% were GN.(29) In Surrey, United Kingdom the rate of GBS were found to be 9.3% in addition to 15.8% GNB and 80.5% GPC.(30) Based on the above it's clear that patterns of bacterial isolates could differ from one place to another (table 1) and over time. Therefore studying the isolates from our own hospital would provide the most reliable information to manage the disease locally.

Methodology

Study Type: Retrospective, Descriptive

Study Site: Neonatal Intensive Care Unit (NICU) of Hamad Medical Corporation in Qatar, Doha

Study Period: Data was observed of relevant patients between August 2006 to June 2008.

Study population was selected using purposive sampling (non-probability sampling) which involves purposive or intentional selection of members of interest from a population to constitute a sample representing the population and exclude those who don’t suit purpose of study. The population of study was made up of all neonates who were admitted to the neonatal intensive care unit during the study period and who had culture-proven sepsis. Blood samples from the neonatal intensive care unit were collected by competent nurses according to a standard protocol and analyzed in Hamad Medical Corporation medical laboratory according to Hamad Medical Corporation policies and procedures, which are adopted from the Clinical and Laboratory Standards Institute (CLSI) recommendations.

Data Collection:

The data for the study was collected from the unit sepsis workup log-book, patients' medical records, patient's microbiological data release in electronic Medical Records (eMR) and microbiology department electronic records. The patient-sepsis characteristics were collected including patient identification number (HMC or HC number), sex, gestational age, birth weight, type of sepsis (earlyonset or late-onset) and presence of lines (peripheral or central). The patients' microbiological data including date of sepsis workup, microorganisms isolated and susceptibility patterns of isolated microorganisms for antibiotics commonly use in neonates were also collected. The data was collected from the above-mentioned sources using a well prepared data collection form.

Data Analysis:

Data was analyzed using Statistical Package for the Social Sciences (SPSS 16.0) package. Descriptive statistics have been calculated for all the collected variables.

Aim:

To investigate bacterial pathogens causing neonatal sepsis in the neonatal intensive care unit of Hamad Medical Corporation, Doha, Qatar.

Objectives:

1. Identify patients admitted to the neonatal intensive care unit between August 2006 and June 2008 who have culture-proven sepsis.

2. Collect patients' microbiological data; date of sepsis workup, microorganisms isolated and susceptibility of isolated microorganisms for antibiotics commonly use in neonates.

Results

During study period (August 2006 – June 2008), there were 27,601 live births at the Women’s Hospital, Hamad Medical Corporation, of which 3,497 (12.7%) were admitted to NICU (Figure 1). A total of 2,851 blood sample for blood culture were obtained, out of which 302 (10.6%) were positive. These positive blood cultures sampled obtained from 176 neonates, giving an overall incidence rate for sepsis of 6.4 per 1,000 live births, and 50 per 1,000 admissions (5%) to the NICU. The neonatal mortality rate during the study period was 5.3 per 1,000 live births (148/27,601) equivalent to 42 per 1,000 admissions. Thirty of the 148 deaths were in neonates who developed sepsis, resulting in case fatality rate of 17% (30/176).

Demographics of Study Population

The demographic data of neonates with positive blood culture are given in Table 2. Over sixty percent (60.8%) of the study population are male with a male-to-female ratio of 1.5:1. Preterm neonates made up more that 80% of the population.

Infection due to gram positive organisms represented 66% of the study population while 18% and 16% have developed fungal and gram negative sepsis respectively. Thirty neonates (17%) developed their first episode of sepsis at or before 3 days of life (early-onset sepsis) while the remaining 146 (83%) developed sepsis after three days of life (late-onset sepsis). Out of the study population, sixty four (36%) are extremely low birth weight (<1,000g).

Mortality

Thirty septic neonates died during hospitalization (Table 2). Neonates who developed fungal sepsis were nearly 10 times more likely to die during hospitalization than those who developed gram positive or gram negative sepsis (Relative risk (RR), 9.7; p-value < 0.05) (Table 9). Septic extremely low birth weigh (<1,000g) neonates were found 3 times more likely to die during hospitalization compared to larger neonates (RR, 2.75; p-value < 0.05). The relatively fewer death in neonates infected with gram positive sepsis resulted in a significantly lower relative risk of dying during hospitalization when compared with other organisms. However this mathematical fallacy must not be interpreted as a protective effect (RR, 0.27; p-value < 0.05). There was no significant increased risk of dying during hospitalization in septic males compared with female (p-value >0.05). On the other hand septic neonates who had central venous lines (CVL) were almost 5 times more likely to die during hospitalization compared to those with no CVL however this was not statistically significant (p-value > 0.05). Septic term neonates were more likely to survive to discharge when compared with septic preterm neonates but this also was not statistically significant (p-value > 0.05).

Organisms Isolated During Study Period

Multiple organisms were isolated in 8% (26) of culture specimens (Table 4). Approximately 10% of positive cultures were collected at or before 3 days of life while nearly 90% were collected after 3 days of life.

Table 5 shows the patterns of the different organisms isolated during the study period. Of 302 positive cultures, there were 320 organisms isolated. Twentyfour (8%) isolates were labeled by the laboratory as contaminant. Of 320 isolates 66% (211/320) were gram positive cocci, 18% (57/320) were fungi (candida), and approximately 16% (52/320) were gram negative bacilli. Only 10% (32/320) of the isolates were identified during early-onset sepsis, of which 25% (8/32) were gram negative bacilli, and 75% (24/32) were gram positive cocci. No fungi were isolated during early-onset episodes of sepsis. Onthe- other-hand, isolates identified during late-onset sepsis, include 65% (187/288) gram positive cocci, 20% (57/288) fungi, and 15% (44/288) gram negative bacilli. In late-onset sepsis coagulasenegative Staphylococcui (CoNS) was the leading organism representing 52% (151/288) of the isolates, followed by Candida albicans 10% (29/288), Candida parapsilosis 6.3% (18/288), and Staphylococcus aureus 4.9% (14/288). Eleven percent (17/151) of CoNS noted in late-onset sepsis is probable contaminant. The most commonly isolated CoNS was Staphylococcus Epidermidis 47% (71/151), followed by Staphylococcus haemolyticus 22.5% (34/151) and Staphylococcus capitis 12.5% (19/151).

In early-onset sepsis Group B streptococcus was the most commonly isolated organism representing 31% (10/32) of isolates, followed by Haemophilus influenzae which made up 12.5% (4/32). The incidence of Group B streptococcus in our hospital was found to be 0.36 per 1,000 live births. Table 6 displays the frequency of occurrence of episodes of positive culture per neonate in the study population. Two thirds (109/176) of the study population had 1 episode of positive blood culture, 22% (22/176) had 2 episodes, and 16% (28/176) had 3 episodes or more.

Discussion

The frequency of GPC and GNB isolated during our study are generally comparable to those reported by other studies. (2)(31)(16) The relatively low incidence of positive cultures at or before 3 days of life (10%) could be attributed to use of intrapartum chemoprophylaxis .(32),(33)

Eighteen percent of isolates in our study were fungi, mainly Candida spp. Our figures are high compared to other reports (21)(28). This could be due, in part to the high proportion of premature babies in our study population (82%).(34) In our study Candida albicans, Candida parapsilosis and Candida glabrata were the most commonly isolated fungi, comparable to reports from the national epidemiology of mycoses survey (NEMIS), United States.(35)

The incidence rates of EOS and LOS in our NICU are 1 and 5 cases per 1,000 live births. Our LOS incidence rates are similar to those quoted by the United States and Australia while the incidence rates of EOS are less.(14) Saudi Arabia reported high incidence rates of EOS (7.4 cases per 1,000 live births) compared with our figures.(36) Unlike reports from Pakistan, India, and GPC are the major causes of neonatal sepsis in our NICU. Similar etiology was observed in Kuwait (37) and Saudi Arabia. The most common organisms causing EOS and LOS in our NICU are GBS (31.3%) and CoNS (52%) respectively. These findings are comparable with results from United States (4), United Kingdom and Israel . While in the United States Escherichia coli is the second most common agent of EOS (23%) and made up 9% of LOS isolates (4), it's found in 3% of isolates in both EOS and LOS in our NICU. Neonatal group B streptococcal disease is very rare in Asia. The incidence rate in our NICU is 0.36 cases per 1,000 live births. It's comparable with incidence rates in the United States(19), Saudi Arabia, and Israel and less than rates quoted by United Kingdom, and United Arab Emirates. Our low GBS incidence is reflecting our advance antenatal care including intrapartum chemoprophylaxis according to CDC guidelines.(38) No fungi were cultured in our NICU during earlyonset episodes of sepsis, comparable to NICUs in United Kingdom, and United States. We revealed relatively high incidence of fungi isolates during lateonset episodes of sepsis (18%) compared with United Kingdom and United States. These episodes resulted in 26 (15%) cases of neonatal fungemia. This could be due in part to the fact that more than 80% of our study populations are preterm neonates, and more than 35% are ELBW (< 1,000g). Other risk factors for fungemia such as, use of third-generation cephalosporins or carbapenem, receiving two or more antibiotics, mechanical ventilation, parenteral nutrition for longer than 5 days, lipid emulsion longer than 7 days, and the use of H2-receptor antagonists are not uncommon in our NICU (39). Out of 148 neonatal deaths during the study period (5.3 cases per 1,000 live births), 30 developed sepsis, resulting in case-fatality rate of 17% (30/176). Casefatality rates reported in the literature varies between 2% and 50% (4). In this study we have correlated the case-fatality to factors playing a significant role in the occurrence of neonatal sepsis. Similar to results from 2 multi-center United States studies our data revealed that fungal sepsis was associated with increased mortality (p-value <0.05), however grampositive sepsis was not associated with increased mortality (p-value <0.05).(40) In contrast to results from the above-mentioned studies we found no risk of increased mortality with gram-negative sepsis. This could be attributed to the large sample size of these studies (1696 and 1299 neonates) compared with our study (176 neonates). One study found that sepsis-associated mortality was more likely to occur among male, black, or preterm.(40) We found that septic male neonates were not more likely to die during hospitalization; however this was not statistically significant. Septic ELBW (<1,000g) neonates were more likely to die during hospitalization (p-value <0.05). Both, gestational age and birth weight are well known determinants of neonatal mortality (41). In our study we revealed that septic term neonates were less likely to die during hospitalization (non-significant), however we retrieved no study for comparison. Moreover, we found that septic neonates who have central venous line were more likely to die during hospitalization (non-significant). Central venous lines or catheters are well known risk factor for neonatal sepsis, be it fungal or bacterial sepsis (42).

Conclusion

We concluded that, the patterns of isolates in our NICU are comparable to other Gulf Cooperation Council countries (GCC). These patterns require continuing surveillances for confirmation and monitoring of probable changes over time. Group B streptococcal infection incidence rates are low and similar to rates from United States. Relatively high incidence rates of fungal sepsis and high risk of fungal sepsis-associated mortality in our neonates necessitate an immediate implementation of strategies to reduce fungal sepsis rates. Incidence rates of sepsis and case-fatality rates in our NICU are comparable to the rates reported in Gulf Cooperation Council countries and within the international rates. This study provides a generalized idea of bacterial isolates in neonates, more studies of same kind would be useful to have a clear idea of pattern of bacterial infection in NICU.

Conflict of Interest

NIL

Source of Support:

NONE

Tables at a glance

Table 1 Table 2 Table 3 Table 4 Table 5 Table 6

References

1. Agnihotri, N., Kaistha, N. & Gupta, V. (2004) Antimicrobial susceptibility of isolates from neonatal septicemia. Jpn J Infect Dis, 57, 273-5.
2. Airede, A. I. (1992) Neonatal septicaemia in an African city of high altitude. J Trop Pediatr, 38, 189- 91.
3. Al-Eidan, F. A., Mcelnay, J. C., Scott, M. G., Kearney, M. P., troughton, K. E. & Jenkins, J. (1999) Sequential antimicrobial therapy: treatment of severe lower respiratory tract infections in children. J AntimicrobChemother, 44, 709-15.
4. AL-Zwaini, E. J. (2002) Neonatal septicaemia in the neonatal care unit, Al-Anbar governorate, Iraq. East Mediterr Health J, 8, 509-14.
5. Barlow, G. &Nathwani, D. (2005) Is antibiotic resistance a problem? A practical guide for hospital clinicians.Postgrad Med J, 81, 680-92.
6. Bizzarro, M. J. & Gallagher, P. G. (2007) Antibioticresistant organisms in the neonatal intensive care unit. SeminPerinatol, 31, 26-32.
7. Bizzarro, M. J., Raskind, C., Baltimore, R. S. & GALLAGHER, P. G. (2005) Seventy-five years of neonatal sepsis at Yale: 1928-2003. Pediatrics, 116, 595-602.
8. Bourchier, D. & Weston, P. J. (1997) Randomised trial of dopamine compared with hydrocortisone for the treatment of hypotensive very low birthweight infants. Arch Dis Child Fetal Neonatal Ed, 76, F174- 8.
9. CDC (1995) Recommendations for preventing the spread of vancomycin resistance. Recommendations of the Hospital Infection Control Practices Advisory Committee (HICPAC). MMWR Recomm Rep, 44, 1-13.
10. CDC (2002) Prevention of Perinatal Group B Streptococcal Disease: Revised Guidelines from CDC. MMWR MorbWkly Rep, 51(RR11):1-22.
11. Escobar, G. J., LI, D. K., Armstrong, M. A., Gardner, M. N., FOLCK, B. F., Verdi, J. E., Xiong, B. & Bergen, R. (2000) Neonatal sepsis workups in infants >/=2000 grams at birth: A populationbased study. Pediatrics, 106, 256-63.
12. Fauci, A., Braunwald, E., Kasper, D., Hauser, S., Longo, D., Jameson, J. &Loscalzo, J. (2008) Harrison's; Principles of Internal Medicine. 17th ed , London, McGraw-Hill Medical.
13. Garcia-Bustos, J. & Tomasz, A. (1990) A biological price of antibiotic resistance: major changes in the peptidoglycan structure of penicillin-resistant pneumococci. ProcNatlAcadSci U S A, 87, 5415-9.
14. Gomaa. H., U. E., &Rajaramb. U. (2001) Neonatal Septicemia in Al-Jahra Hospital, Kuwait: Etiologic Agents and Antibiotic Sensitivity Patterns. Med Principles and Pract, 10, 145-150.
15. Gould, I. & Van, J. (2004) Antibiotic Policies; Theory and Practice., NY, Kluwer Academic/Plenum Publishers.
16. Haque, K. N. (2006) Immuno-modulation in neonatal sepsis: intravenous immunoglobulin therapy in the prevention and treatment of neonatal sepsis: Is the answer, ‘Yes’,‘No’, or ‘Don’t know’? Haematologica Reports, 2(10):38-41.
17. Jain, A., Roy, I., Gupta, M. K., Kumar, M. &Agarwal, S. K. (2003) Prevalence of extendedspectrum beta-lactamase-producing Gram-negative bacteria in septicaemic neonates in a tertiary care hospital. J Med Microbiol, 52, 421-5.
18. Haque, K. N., Chagia, A. H. &Shaheed, M. M. (1990) Half a decade of neonatal sepsis, Riyadh, Saudi Arabia. J Trop Pediatr, 36, 20-3.
19. Harvey, R., Champe, P., Finkel, R., Cubeddu, L. & Clarke, M. (2008) Lippincott's Illustrated Reviews: Pharmacology, 4th Edition London, Lippincott's Williams & Wilkins.
20. Hawkey, P. M. (1998) The origins and molecular basis of antibiotic resistance. Bmj, 317, 657-60.
21. Hitomi, S., Kubota, M., Mori, N., Baba, S., Yano, H., Okuzumi, K. & Kimura, S. (2000) Control of a methicillin-resistant Staphylococcus aureus outbreak in a neonatal intensive care unit by unselective use of nasal mupirocin ointment. J Hosp Infect, 46, 123-9.
22. IREGBU, K. C., ELEGBA, O. Y. & BABANIYI, I. B. (2006) Bacteriological profile of neonatal Ful l Length Re s ea r chPape r Cov e r ed in Inde x Cope rnicuswith IC Value 4.68 fo r 2010 Afif Ahmed et al: Incidence of Bacterial Isolates from Blood Culture in the Neonatal Intensive Care Unit of Tertiary Care Hospital Int. J. Drug Dev. & Res., October-December 2012, 4 (4): 359-367 Covered in Scopus &Embase, Elsevier 366
23. septicaemia in a tertiary hospital in Nigeria. Afr Health Sci, 6, 151-4.
24. Jenson, H. B. & Pollock, B. H. (1998) The role of intravenous immunoglobulin for the prevention and treatment of neonatal sepsis. SeminPerinatol, 22, 50-63.
25. Jeong, I. S., jeong, J. S. & Choi, E. O. (2006) Nosocomial infection in a newborn intensive care unit (NICU), South Korea. BMC Infect Dis, 6, 103.
26. Kierans, W. J., Verhulst, L. A., Mohamed, J. & Foster, L. T. (2007) Neonatal mortality risk related to birth weight and gestational age in British Columbia. J ObstetGynaecol Can, 29, 568-74.
27. Knox, K., Lawson, W., Dean, B. & Holmes, A. (2003) Multidisciplinary antimicrobial management and the role of the infectious diseases pharmacist--a UK perspective. J Hosp Infect, 53, 85-90.
28. Kollef, M. H. & Fraser, V. J. (2001) Antibiotic resistance in the intensive care unit. Ann Intern Med, 134, 298-314.
29. Lesar, T. S. (2002) Prescribing errors involving medication dosage forms. J Gen Intern Med, 17, 579-87.
30. Malik, R. K., Montecalvo, M. A., Reale, M. R., LI, K., Maw, M., Munoz, J. L., Gedris, C., Van Horn, K., Carnevale, K. A., Levi, M. H. &Dweck, H. S. (1999) Epidemiology and control of vancomycin-resistant enterococci in a regional neonatal intensive care unit. Pediatr Infect Dis J, 18, 352-6.
31. Mcdonald, L. C., Walker, M., Carson, L., Arduino, M., Aguero, S. M., Gomez, P., Mcneil, P. & Jarvis, W. R. (1998) Outbreak of Acinetobacter spp. bloodstream infections in a nursery associated with contaminated aerosols and air conditioners. Pediatr Infect Dis J, 17, 716-22.
32. Parker, K., Buxton, L., Blumenthal, D. & BRUNTON, L. (2007) The Goodman and Gilman Manual of Pharmacology and Therapeutics, NY, McGraw-Hill Professional.
33. Paterson, D. L. &Bonomo, R. A. (2005) Extendedspectrum beta-lactamases: a clinical update. ClinMicrobiol Rev, 18, 657-86.
34. Peric, M., Bozdogan, B., Jacobs, M. R. &Appelbaum, P. C. (2003) Effects of an efflux mechanism and ribosomal mutations on macrolide susceptibility of Haemophilusinfluenzae clinical isolates. Antimicrob Agents Chemother, 47, 1017- 22.
35. Puopolo, K. M. & Madoff, L. C. (2007) Type IV neonatal early-onset group B streptococcal disease in a United States hospital. J ClinMicrobiol, 45, 1360-2.
36. Rahman, S., Hameed, A., Roghani, M. T. &Ullah, Z. (2002) Multidrug resistant neonatal sepsis in Peshawar, Pakistan. Arch Dis Child Fetal Neonatal Ed, 87, F52-4.
37. Rajbhandary, S. S., Marks, S. M. & Bock, N. N. (2004) Costs of patients hospitalized for multidrugresistant tuberculosis. Int J Tuberc Lung Dis, 8, 1012-6.
38. Saiman, L., Ludington, E., Dawson, J. D., Patterson, J. E., Rangel-Frausto, S., Wiblin, R. T., Blumberg, H. M., Pfaller, M., Rinaldi, M., Edwards, J. E., Wenzel, R. P. & Jarvis, W. (2001) Risk factors for Candida species colonization of neonatal intensive care unit patients. Pediatr Infect Dis J, 20, 1119-24.
39. Tom-Revzon., C. (2004) Strategic use of antibiotics in the neonatal intensive care unit. J Perinat Neonatal Nurs, 18, 241-58.
40. Tuppurainen, N. & Hallman, M. (1989) Prevention of neonatal group B streptococcal disease: intrapartum detection and chemoprophylaxis of heavily colonized parturients. ObstetGynecol, 73, 583-7.
41. UNICEF (2009) The State of the World’s Children 2009: Maternal and Newborn Health., [Online] Available at: https://www.unicef.org/sowc09/docs/SOWC09- FullReport-EN.pdf [Accessed 20 March 2009].
42. Vergnano, S., Sharland, M., Kazembe, P., Mwansambo, C. & Heath, P. T. (2005) Neonatal sepsis: an international perspective. Arch Dis Child Fetal Neonatal Ed, 90, F220-4.