SciELO - Scientific Electronic Library Online

vol.55 número3Frecuencia del hábito de fumar, su abandono, y las características socio-demográficas de los médicos de una facultad de medicinaEpidemiología de los estafilococos coagulasa negativos aislados de especímenes clínicos de sangre en el hospital universitario de West Indies índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados



  • No hay articulos citadosCitado por SciELO

Links relacionados

  • No hay articulos similaresSimilares en SciELO


West Indian Medical Journal

versión impresa ISSN 0043-3144

West Indian med. j. v.55 n.3 Mona jun. 2006 



Neonatal sepsis in very low birthweight infants at the university hospital of the West Indies


Sepsis neonatal en recién nacidos de peso rxtremadamente bajo al nacer en el hospital universitario de West Indies



H Trotman; Y Bell

Department of Obstetrics, Gynaecology and Child Health, The University of the West Indies, Kingston 7, Jamaica, West Indies





A retrospective review was conducted on the charts of all very low birthweight (VLBW) infants with culture proven sepsis admitted to the neonatal unit of the University Hospital of the West Indies (UHWI) during the period January 1, 1995 to December 31, 2000. During the study period, 22 VLBW infants were admitted to the neonatal unit with culture proven sepsis, 16 (73%) survived and 6 (27%) died. As birthweight and gestational age increased, outcome improved. There was no difference in survival based on age at presentation. Neonates with early onset disease had a significantly longer mean duration of rupture of membranes than those with late onset disease (p = 0.009) and babies with late onset disease had a significantly lower mean Hb level than those with early onset disease (p = 0.000). Predominant isolates were Klebsiella sp (10, 37%), Streptococcus Group D (4, 15%), Escherichia coli (3, 11%) and Group B Streptococcus (3, 11%). Klebsiella sp accounted for 8/13 (62%) of late onset infections. Complications included anaemia, thrombocytopenia, bleeding and multi-organ failure. Strategies aimed at prevention, such as limiting the excessive use of broad-spectrum empiric antibiotics and the periodic review and continuous reinforcement of infection control policies will help decrease the mortality and morbidity associated with nosocomial infection in the VLBW infant.


Se llevó a cabo un estudio retrospectivo de las estadísticas de todos los infantes de peso extrema-damente bajo al nacer (PEBN) con sepsis probada con cultivo, ingresados en la Unidad Neonatal del Hospital Universitario de West Indies en el período comprendido de enero 1 de 1995, a diciembre 31 de 2000. Durante el periodo en estudio, 22 infantes de PEBN fueron ingresados a la unidad neonatal con sepsis probada por cultivo. De estos, 16 (73%) sobrevivieron y 6 (27%) murieron. En la medida en que aumentaron el peso al nacer y la edad gestacional, mejoraron también los resultados. No hubo diferencias en cuanto a supervivencia sobre la base de la edad al momento de la presentación. Los infantes con infección neonatal temprana tuvieron una duración media significativamente mayor de la ruptura de membranas en relación con aquellos que presentaron infección neonatal tardía (p = 0.009), en tanto que los bebés con infección neonatal tardía presentaron un nivel de Hb significativamente más bajo en comparación con los que tuvieron infección temprana (p = 0.000). Los aislados predominantes fueron Klebsiella sp 10 (37%), Streptococcus del grupo D 4 (15%), Escherichia coli 3 (11%) y Strep-tococcus del grupo B 3 (11%). Klebsiella sp fue la causa de 8/13 (62%) de las infecciones neonatales tardías. Las complicaciones incluyeron anemia, trombocitopenia, sangramiento y fallo multiorgánico. Estrategias que apunten a la prevención – tal como limitar el uso excesivo de los antibióticos empíricos de amplio espectro, así como la revisión periódica y el refuerzo continuo de las políticas de control de infección – ayudarán a disminuir la mortalidad y la morbilidad asociadas con la infección de los recién nacidos de PEBN.




Sepsis is the most common cause of neonatal mortality in developing countries accounting for 30–50% of total neonatal deaths each year (1, 2). Neonates are deficient in humoral and cellular immunity; they produce immunoglobulins at a lower rate than adults (3). Transplacental maternal antibodies mediate humoral immunity primarily, hence very low birthweight (VLBW) premature infants are less likely to receive as many immunoglobulins as term infants. T–cell function is also less efficient in neonates (4). Complement function and phagocytic function inclusive of phagocytosis, phagocyte migration and toxin production are also deficient (5, 6).

The incidence of sepsis and its complications are therefore greater in VLBW infants and extremely premature babies. Although most premature babies survive today as a result of the availability of neonatal intensive care, they are frequently subjected to invasive procedures that increase the probability of infection.

Onset of infection within the first six days of life is thought to be primarily due to vertical transmission from mother-to-infant, while onset of infection at seven days of life or greater is more likely to be acquired through horizontal transmission. By virtue of the length of time VLBW infants may spend in the hospital setting, they are at prolonged risk for acquiring infection, particularly nosocomial infections. The identification of strategies to reduce infection in these infants will result in decreased mortality and mor-bidity.

The objective of this study was to describe the clinical presentation, causative organisms and outcome of bacterial sepsis in VLBW infants. The authors hypothesize that by virtue of the length of time VLBW infants spend on the neo-natal unit the organisms causing infection will be predominantly nosocomial in origin.



The University Hospital of the West Indies (UHWI), a university affiliated institution, is located in Kingston, the capital of Jamaica. The neonatal unit at the UHWI is a thirty-bed unit; neonates admitted to the unit are mainly inborn.

This study was a retrospective analysis of culture proven cases of sepsis in VLBW infants aged 0–30 days, admitted to the neonatal unit of the UHWI during the six-year period January 1, 1995 to December 31, 2000. Infants were identified from the neonatal unit admission logbook by the discharge diagnosis recorded, this diagnosis was then cross-checked with microbiology laboratory reports. The two in-vestigators reviewed the charts of all VLBW infants with culture proven sepsis and data on demographics, clinical presentation, causative organisms, sensitivity patterns and outcome were extracted using a data extraction sheet. During the study period, 395 VLBW infants were admitted to the neonatal unit.

All VLBW infants admitted to the unit with a presumptive diagnosis of sepsis had a full sepsis screen, which included blood culture, cerebrospinal fluid (CSF) culture, urine culture, a complete blood count and a chest radiograph (CXR). Blood for culture was taken from a peripheral vein after cleansing of the skin with an iodine solution. Samples of blood, CSF and urine were plated and analyzed according to standard bacteriological procedures. A full description of the methods of blood culture has been described from the Microbiology Department, UHWI (7). Infants were empirically commenced on a combination of a penicillin and an aminoglycoside, usually crystalline penicillin or amoxicillin and gentamicin. If the infant had previously been on these antibiotics within 72 hours of the sepsis screen, augmentin and amikacin were used.


Neonatal sepsis was defined as the presence of positive cultures, whether in the blood, CSF, or urine, associated with systemic clinical signs of infection such as fever, temperature instability, irritability, poor feeding and respiratory distress. Very low birthweight infants were those whose birthweight was less than 1500g. Early onset infection (EOD) were in-fections occurring less than 7 days of life while late onset infections (LOD) – occurred at 7–30 days of life. Death as a sequel of sepsis was defined as death occurring within 10 days of a positive culture. Complications of sepsis were complications occurring within seven days of a positive culture.

The UHWI/UWI Faculty of Medical Sciences Ethics Committee granted approval for this study to be conducted.


Descriptive analyses were performed. Continuous variables were expressed as means and compared by independent student t-test. Categorical variables were analyzed using the chi-square test with statistical significance taken at 5%.



During the study period 22 VLBW infants were admitted to the neonatal unit with culture proven sepsis, 20 (91%) were inborn. Sixteen (73%) infants survived and six (27%) died. There were 10 (45%) males of whom two (20%) died and 12 (55%) females of whom four (33%) died. There was no difference in outcome by gender. Table 1 shows the characteristics of neonates with culture positive sepsis by age of presentation.



A total of eight (38 %) neonates presented with early onset infection. Of these, seven (88%) presented within the first 24 hours of life, six (75%) babies with early onset infection survived. Thirteen (62%) neonates presented with late onset infection and of these nine (69%) survived. There was no difference in survival based on the age at presentation. The neonates with EOD had a significantly longer mean duration of rupture of membranes than those with LOD (p = 0.009). Babies with LOD had a significantly lower mean Hb level than those with EOD (p = 0.000).

As birthweight increased, outcome improved. Fifteen (68%) neonates with positive cultures had birthweights be-tween 1000 – 1499g, 13 (87%) survived; seven (32%) had birthweights less than 1000g, three (43%) survived (p = 0.05). There were 11 (50%) babies whose gestational ages were between 31–34 weeks, seven (32%) between 27–30 weeks and four (18%) less than 27 weeks gestation. Survival improved with increasing gestational age, all four (100%) babies less than 27 weeks gestation died while one (14%) baby 27–30 weeks gestation died and one (9%) baby 31–34 weeks gestation died (p = 0.001).

The most common presenting clinical features were respiratory distress – 7 (32%), poor feeding – 5 (23%), vomiting – 3 (14%), abdominal distension – 3 (14%), lethargy – 2 (9%) and irritability – 2 (9%).

A total of 27 isolates were obtained, 19 (70%) were from blood cultures and eight (30%) from urine cultures. Table 2 shows the organisms isolated during the study pe-riod. Predominant isolates were Klebsiella sp, Streptococcus Group D, Escherichia coli and Group B Streptococcus. Of the organisms isolated 18 (67%) were gram-negative organisms and 9(33%) were gram-positive organisms. Of the or-ganisms isolated from blood, 12 (63%) were gram-negative organisms and 7 (37%) were gram-positive organisms. While from the urine, 6 (75%) isolates were gram-negative organisms and two (25%) were gram-positive organisms. Kleb-siella sp accounted for 8/13 (62%) of the late onset infections.



Analysis of sensitivity patterns of the micro-organisms revealed that all GBS isolates were sensitive to penicillin. All Escherichia coli and Pseudomonas isolates were sensitive to gentamicin. However only six (60%) of the isolates of Kleb-siella sp were sensitive to gentamicin. Two (20%) of the Klebsiella sp isolates were resistant to the aminoglycosides, ceftazidime and ceftriaxone. These two isolates were only sensitive to meropenem.

Case fatality rate for VLBW infants with bacterial infections at UHWI was 27%. Case fatality rates were higher for infants weighing less than 1000g (57%) than for infants weighing greater than 1000g (13%). Gram-negative organisms accounted for four (66%) of the deaths. Table 3 shows the organisms causing infection in the non-survivors. Of the non-survivors, one infant developed disseminated intravascular coagulation and died before medical management could be instituted, two died from multisytstem organ failure and three died from respiratory failure.



Seven (32%) babies were anaemic and thrombocyto-penic, three (43%) of these developed bleeding. Three (14%) babies developed acute seizures as a sequel of sepsis, one of these babies had bleeding and went on to develop acute hy-drocephalus. Two (9%) VLBW infants with positive cultures developed multi-organ failure. Intensive care unit admission was required for four (18%) of the infants.



The mortality rate from sepsis for VLBW infants in this study is similar to that reported in the United States of America (USA) (8). However, in the USA, a greater percentage of babies less than 1000g survive and the weight distribution within the VLBW infants is different from that of this study where the weight distribution is skewed towards the larger babies who have better outcome. A similar cohort in the USA would have a lower mortality rate. The mortality rate, however, is lower than the rates reported in other developing countries where rates of 40–45% have been quoted (9, 10).

The finding of increased survival with increasing birthweight and gestational age is similar to previous studies (8, 9, 11–13). The predominance of LOD in these VLBW infants is not unexpected as by virtue of their immaturity they are more likely to have an extended hospital stay placing them at prolonged risk for infection. Koutouby et al and Moreno et al have also documented this finding (10, 14). The association of prolonged rupture of membranes with EOD has been shown in other studies (10, 15). The longer the duration of rupture of membranes, the longer is the period of time the unborn infant is exposed to ascending organisms from the mother’s genital tract predisposing to early onset of infection.

The lower haemoglobin levels in infants with LOD may be as a result of their more prolonged exposure to re-peated phlebotomy for blood tests in the neonatal unit and anaemia of prematurity. However, half of the infants were both anaemic as well as thrombocytopenic and some of them experienced bleeding. These are all markers of sepsis, parti-cularly thrombocytopenia, which has been shown to occur in gram-negative sepsis (16), which the majority of these in-fants had.

The common presenting symptoms are similar to those reported by Fanaroff et al (8). These illustrate the subtle and non-specific signs of sepsis in VLBW infants. Thus, the phy-sician must have a high index of suspicion for making the diagnosis of sepsis based on clinical signs in these infants.

Several studies from developed countries report coagu-lase negative staphylococcus as the predominant organism causing LOD in VLBW infants (17–19). In this study, this pattern was not seen. This may be as a result of the limited mechanical ventilation of infants during the study period and the inability to offer total parenteral nutrition in this setting. As such, there was not a high prevalence of indwelling central catheters, a major risk factor for coagulase negative staphylococcal infection. The finding of Klebsiella sp as the major pathogen in LOD is similar to studies from other developing countries (10, 20).

The resistance of Klebsiella sp to gentamicin and the emergence of multiple antibiotic resistant organisms have important implications in this setting, as Klebsiella sp was the most common organism causing nosocomial infection. This means that for infants with LOD, physicians must have a low threshold to switch from empiric antibiotic regimens to more potent antibiotics in the face of sub-optimal clinical improvement. Some parameters that may be used to guide the physician would be if there is persistence of the initial clinical signs of sepsis or the development of new signs after 48 – 72 hours of empiric antibiotics. Also, if there is wor-sening of laboratory markers of sepsis (anaemia, thrombocytopenia, leucopaenia, deranged PT/PTT) and if there is minimal improvement or deterioration in the clinical status of the infant.

This study has demonstrated an increased incidence of late onset infection in VLBW infants and this represents no-socomial infection. Strategies aimed at prevention of nosocomial infection such as limiting the excessive use of broad-spectrum empiric antibiotics, the development of a hand hy-giene programme and periodic review and continuous reinforcement of infection control policies must be instituted.

Development of a hand hygiene programme would in-clude provision of a scrub area at the entrance of the unit. All persons who have patient contact would be required on entering the unit to remove any jewellery eg watch, rings and bracelets and wash hands, fingernails and forearms with an antibacterial detergent for at least two minutes. Between handling of patients and after contact with contaminated surfaces, the hands should be washed for at least 15 seconds with soap and water. Every cubicle would be equipped with hands-free, hand-washing sinks so that every bed position is within six metres of a sink. A foot pedal would be used to operate soap dispensers and centre pull towel dispensers would be provided. Pictorial hand washing instructions would be provided above all sinks.

Surveillance for nosocomial infection would be carried out through review of patient charts for clinical evidence of infection and review of microbiologic culture reports. This is routinely done for the monthly perinatal mortality/morbidity conference. This meeting would be an ideal forum for the review and reinforcement of infection control policies. Fo-cussing on these strategies will help decrease the mortality and morbidity associated with hospital acquired infection in the VLBW infant.



1. Bang AT, Bang RA, Baitule SB, Reddy HM, Deshmukh MD. Effect of home based neonatal care and management of sepsis on neonatal mortality: field trial in rural India. Lancet 1999; 354: 1955–61.

2. Stoll BJ. The global impact of neonatal infection. Clin Perinatol 1997; 24: 1–21.

3. Wilson CB. Immnunologic basis for increased susceptibility of the neo-nate to infection. J Paediatr 1986; 108: 1–12.

4. Schelonka RL, Infante AJ. Neonatal immunology. Semin Perinatol 1998; 22: 2–14.

5. Berger M. Complement deficiency and neutrophil dysfunction as risk factors for bacterial infections in newborns and the role of granulocyte transfusion in therapy. Rev Infect Dis 1990; 12 suppl 4: S1401–9.

6. Shigeoka AO, Santos JL, Hill HR: Functional analysis of neutrophil granulocytes from healthy, infected, and stressed neonates. J Pediatr 1979; 95: 454–60.

7. MacFarlane DE, Narla VR. Bacteraemia at the University Hospital of the West Indies – a report of 222 cases. J Infect 1985; 10: 126–30.

8. Fanaroff AA, Korones SB, Wright LL, Verter J, Poland RL, Bauer CR et al. Incidence, presenting features, risk factors and significance of late onset septicemia in very low birthweight infants. The National Institute of Child Health and Human Development Neonatal Research Network. Pediatr Infect Dis J 1998; 17: 593–8.

9. Bhutta ZA, Yusuf K. Neonatal sepsis in Karachi: factors determining outcome and mortality. J Trop Pediatr 1997; 43: 65–70.

10. Koutouby A, Habibullah J. Neonatal sepsis in Dubai, United Arab Emirates J Trop Pediatr 1995; 41: 177–80.

11. Gladstone IM, Ehrenkranz RA, Edberg SC, Baltimore RS. A ten-year review of neonatal sepsis and comparison with the previous fifty – year experience. Pediatr Infect Dis J 1990; 9: 819–25.

12. Grauel EL, Halle E, Bollmann R, Buchholz P, Buttenberg S. Neonatal septicaemia incidence, etiology, and outcome: a 6-year analysis. Acta Paediatr Scand Suppl 1989; 360: 113 –9.

13. Tessin I, Trollfors B, Thiringer K. Incidence and aetiology of neonatal septicaemia and meningitis in western Sweden 1975–1986. Acta Pae-diatr Scand 1990; 79: 1023–30.

14. Moreno MT, Vargas S, Poveda R, Saez – Llorens X. Neonatal sepsis and meningitis in a developing Latin American country. Pediatr Infect Dis J 1994; 13: 516–20.

15. Benitz WE, Gould JB, Druzin ML. Risk factors for early-onset group B streptococcal sepsis: estimation of odds ratios by critical literature review. Pediatrics 1999; 103: 1275 e77.

16. Guida JD, Kunig AM, Leef KH, McKenzie SE, Paul DA. Platelet count and sepsis in very low birthweight neonates: is there an organism-specific response? Pediatrics 2003; 111: 1411–5.

17. Munson DP, Thompson TR, Johnson DE, Rhame FS, Vandrunen N, Ferrieri P. Coagulase-negative staphylococcal septicemia: experience in a newborn intensive care unit. J Pediatr 1982; 101: 602–5.

18. Baumgart S, Hall SE, Campos JM, Polin RA. Sepsis with coagulase- negative staphylococci in critically ill newborns. Am J Dis Child 1983; 137: 461–3.

19. Stoll BJ, Hansen N. Infections in VLBW infants: studies from the NICHD Neonatal Research Network. Semin Perinatol 2003; 27: 293–301.

20. Pawa AK, Ramji S, Prakash K, Thirupuram S. Neonatal nosocomial infection: profile and risk factors. Indian Pediatr 1997; 34: 297–302.



Dr H Trotman
Department of Obstetrics, Gynaecology and Child Health, Section of Child Health, The University of the West Indies
Kingston 7, Jamaica, West Indies
Fax: (876) 970-0329