SciELO - Scientific Electronic Library Online

vol.89 número5Experiencias, creencias y actitudes sobre donación de leche humana en mujeres de la provincia de AraucoCorrelación en la determinación de la edad ósea radiológica mediante el método de Greulich y Pyle versus la evaluación automatizada utilizando el software BoneXpert índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google


Revista chilena de pediatría

versión impresa ISSN 0370-4106

Rev. chil. pediatr. vol.89 no.5 Santiago oct. 2018 


Prolonged initial empirical antibiotic treatment and the risk of morbidity and mortality in very low birthweight infants

Diana Torres

Tomás Muñoz

Aldo Bancalari

Camilo Manríquez

1 Neonatal Fellowship Program, Department of Pediatrics, Faculty of Medicine, University of Concepción, Chile.

2 Department of Pediatrics, Faculty of Medicine, University of Concepción, Chile.

3 Neonatology Service, Guillermo Grant Benavente Hospital, Concepción, Chile.

4 Statistical Engineer of the Faculty of Medicine, University of Concepción, Chile.



The objective of this study is to evaluate the association between the duration of ini tial empirical antibiotic treatment and the subsequent development of late-onset sepsis, necrotizing enterocolitis (NEC) and death in very low birth weight (VLBW) infants.

Patients and Methods:

Quantitative, cross-sectional, analytical study of VLBW infants admitted to the neonatal ICU were included over a period of five years. Initial empirical antibiotic therapy was that which started im mediately after birth, without knowing the results of blood cultures. It was considered prolonged antibiotic therapy when the treatment duration was > 5 days. Perinatal variables, as well as the inci dence of late-onset sepsis, confirmed NEC and mortality were analyzed.


266 VLBW infants were studied, with an average gestational age and birth weight of 28.8 ± 2.5 weeks and 1.127 ± 264 g respectively. 213 infants received initial empiric antibiotic therapy (80.0%), which was prolonged in 67.6% of cases. All infants received two different antibiotics. 136 episodes of late-onset sepsis were described. The most common pathogens were coagulase-negative Staphylococcus and Staphylococcus aureus. Among the newborns with prolonged antibiotic therapy, there were 20 cases of confirmed NEC and 15 of the studied infants died (10.4%). When comparing the use of antibiotic therapy during > 5 days versus treatment less than 5 days duration, a statistically significant association was observed between prolonged antibiotic therapy and late-onset sepsis (p = 0.03) and confirmed NEC (p = 0.03), but not of mortality (p = 0.12).


The use of empirical antibiotic therapy for five days or more was associated with an increased risk of late-onset sepsis and NEC, but not of mortality in VLBW infants.

Keywords: Prolonged initial empirical antibiotic; late-onset sepsis; necrotizing enterocolitis; mortality


Antibiotics are among the most commonly used drugs in preterm newborns (NBs) hospitalized in neonatal intensive care units1,2,3. Exposure to antibiotics in very low birth weight (VLBW) neonates can lead to several alterations such as reduction of microbiota biodiversity, delay of normal colonization of the gastrointestinal tract, and/or proliferation of pathogenic organisms that may be resistant to antibiotics3-6.

The increase in potentially pathogenic organisms, the decrease in the normal intestinal flora, along with the deterioration in the intestinal epithelial barrier produces an abnormal colonization of the gastroin testinal tract; increasing the risk of invasion of the in testinal wall with translocation, and the production of inflammatory cytokines, all of which can trigger no socomial sepsis, necrotizing enterocolitis (NEC), and secondarily death3-8.

In recent years, some reports have described in VLBW neonates the association between prolonged initial empirical antibiotic treatment and subsequent adverse events9-13. These studies have shown that prolonged initial empirical antibiotic treatment >5 days in VLBW neonates without positive cultures increases the risk of developing late sepsis, NEC, and death in this group of patients9-13.

The hypothesis of this study was that prolonged initial empirical antibiotic therapy in VLBW neonates would be associated with an increase in the incidence of nosocomial sepsis, NEC, and/or death.

Since there is no national information on the effect of the prolonged empirical antibiotic use on VLBW neonates, the objective was to determine the associa tion between the duration of prolonged initial empirical antibiotic treatment and the subsequent deve lopment of late sepsis, NEC, and/or death in VLBW neonates whose postnatal cultures were negative.

Patients and Method


Quantitative, analytical cross-sectional study. All VLBW neonates admitted to the Neonatal Intensive Care Unit of the Neonatology Service of the Guiller mo Gran Benavente Hospital, Concepcion (HGGB), between January 1, 2009 and December 31, 2013 (five years) were included. A retrospective analysis of the clinical records of VLBW neonates with gestational age less than 32 weeks and/or birth weight less than 1,500 g, who survived more than seven days without deve loping early sepsis and who received initial empirical antibiotic treatment was performed. Neonates with major congenital anomalies, chromosomal abnorma lities, and also those transferred from another hospital and with early proven sepsis (positive blood cultures and/or CSF) were excluded.


Initial empirical antibiotic treatment was conside red to be that which was initiated since birth, without presenting clinical sepsis, and with negative blood and CSF cultures at 72 hours of life. Prolonged initial empirical antibiotic therapy was established when its du ration was five days or more. Clinical sepsis was consi dered when cultures were negative, but the NB showed three or more of the following signs: increased or de creased axillary temperature (< 36.5°C or > 37.5°C), lethargy, apnea, bradycardia, respiratory distress, ab dominal distension, blood in stool; increased oxygen requirements, prolonged capillary filling time, hypotension; associated with alterations in laboratory tests such as: hyperglycemia, thrombocytopenia <150.000/ mm3, leukopenia <5000/mm3 or leukocytosis >25.000/ mm3, CSF: >10 mg/dl14-15.

Early confirmed sepsis was considered in cases of clinical sepsis with positive blood cultures and/or CSF cultures in the first 72 hours of life, and late sepsis tho se with clinical sepsis plus positive blood cultures and/ or CSF cultures after 72 hours of life.

Only patients with confirmed NEC (stage >II ac cording to Bell classification16 were considered for the study. Neonatal mortality was defined as that which occurred during the first 28 postnatal days.

Statistical analysis

A univariate analysis was performed in which quan titative variables are presented as mean and standard deviations, while qualitative variables are presented in absolute and relative percentage frequency. In or der to determine the risk between prolonged duration (>5 days) and late sepsis, confirmed NEC, and neona tal death, we analyzed through logistic regression. A p<0.05 was considered significant using the statistical software SPSS version 19.0.

This study was approved by the Scientific Ethics Committee of the Concepción Health Service.


266 VLBW neonates were analyzed, with a gestatio nal age and mean birth weight ±SD of 28.8 ± 2.5 weeks and 1127 ± 264 g respectively. 131 of them were male (57.7%). Out of the total cases, 80.1% (213/266) recei ved empirical antibiotic therapy and 19.9% (53/266) did not receive initial antibiotics.

Out of the 213 VLBW neonates that received initial antibiotic therapy, it was prolonged in 67.6% of cases (144/213). (Table 1) shows the demographic characte ristics and perinatal history among neonates who received prolonged and not prolonged initial empirical antibiotic therapy. No differences in demographic or perinatal characteristics were observed between the two groups.

Table 1 Demographic and perinatal characteristics in 213 VLBW infants with prolonged and non-prolonged initial empirical antibiotic treatment. 

In neonates who received initial empirical antibio tics, ampicillin and amikacin were the most commonly used regimen (97%). Out of the 213 NBs with prolon ged initial empirical antibiotic therapy, 122 episodes of late sepsis were investigated, of which 80 (65.6%) were bacteriologically proven. 96 pathogenic agents were identified, the most frequent were coagulase-negative Staphylococci, Staphylococcus aureus, Enterobacter cloacae, Acinetobacter baumannii, and Pseudomonas aeruginosa (Table 2).

Table 2 Pathogens isolated in confirmed late onset sepsis. 

There were 20 confirmed NEC cases; of them, half required surgical intervention. Out of the 213 VLBW neonates with initial empirical antibiotic therapy, 15 (10.4%) died. There was a statistically significant association between the use of prolonged initial empirical antibiotic therapy and late sepsis (p = 0.03), with an OR of 1.88 ; and with the development of NEC (p = 0.03), with an OR of 9.71; but no evidence of increased mor tality (p = 0.12) (Table 3).

Table 3 Logistic regression of associated morbidities, according to prolonged treatment of empirical antibiotics. 


In the Neonatology Service of the HGGB, in the period analyzed approximately 80% of the VLBW neonates received initial empirical antibiotic thera py for an undetermined period of time. Antibiotics are among the most commonly prescribed medica tions in neonatal intensive care units1-2. According to Clark1 the antibiotics most frequently used in new borns as initial empirical therapy are ampicillin and gentamicin; similar to what was observed in the stu died population, with the exception of amikacin to replace gentamicin. The reason why in our center we use amikacin instead of gentamicin is the fact that the most common bacterial isolates are non-fermenting gram-negative bacilli and Pseudomonas aeruginosa, microorganisms that are more sensitive to amikacin than to gentamicin.

In this study, as in other publications1,3, it is confir med that a high percentage of preterm NBs who do not have confirmed early sepsis are unnecessarily exposed to prolonged antibiotic use.

The initial and prolonged empirical use of anti biotics in the neonatal population has previously bee

described as a possible risk factor in the development of late sepsis, NEC, and death9-13. Empirical administration of antibiotics may induce abnormal coloniza tion of the gastrointestinal tract of the neonate by in hibiting or eradicating protective bacteria and favoring the proliferation of potentially pathogenic micro-organisms4-8. Different studies have determined that the overgrowth of pathogenic species is higher after three days of exposure to antimicrobials9,11.

Cotten et al9 reported that exposure to prolonged initial empirical antibiotic therapy longer than or equal to five days with negative blood cultures was associated with an increased likelihood of the combined outcome of NEC and/or death (OR 1.50 [95% CI: 1.22 - 1.83]); only NEC (OR:1.34 [95% CI: 1.04 - 1.73]), and death (OR:1.86 [95% CI:1.45 - 2.39]) in extremely low birth weight neonates (<1000 g). On the other hand, Kuppala et al11 showed that the duration of initial antibiotic therapy longer than or equal to five days in preterm NBs less than 32 weeks of gestation, with negative blood cultures in the first week of life, was subsequently asso ciated with an increase in late sepsis, NEC and death11. Another case-control study10 evaluated if exposure to antibiotic therapy immediately after birth was an inde pendent risk factor for NEC and reported that the lon ger duration of antibiotic therapy was associated with a higher risk of bacteriologically proven late sepsis, and NEC10, consistent with the results of this study.

Ting et al17 also analyzed the rate of antibiotic use (antibiotic days/days of hospitalization) and reported a 10% increase in risk of adverse outcomes. They con sidered as adverse outcomes stage 3 or higher retino pathy of prematurity, mortality, and the compound outcome (mortality and/or significant morbidities, such as periventricular leukomalacia, bronchopul monary dysplasia, or retinopathy). Cantey et al18 also evaluated the impact of antibiotic exposure on VLBW neonates, finding that each additional day of antibio tics, after reaching clinical improvement, increased the risk of developing bronchopulmonary dysplasia, which was also more severe18. Similar studies in Egypt12, Australia13, and Mexico19 have also reported results consis tent with the above publications, confirming the risk of prolonged exposure to antibiotics in VLBW neonates. The results of these publications9-13,17-19 are similar to the findings of this study, in which it was observed that prolonged exposure (>5 days) to empirical antibiotics after birth increased the risk of late sepsis and NEC, OR=1,88, and OR=9,71 respectively. It should be no ted that in all these reports9-13,17-19, the studied popu lation has similar characteristics (very low or extreme low weight NBs). On the other hand, there is a high rate of positive blood cultures (65.6%) among the ca ses of late clinical sepsis in this study, where the most frequently isolated microorganism is the coagulase- negative Staphylococcus, which is consistent with what has been reported in other studies9,11,13.

Despite the solid evidence reported in the above studies9-13,17-19, the excessive and inadequate use of empirical antibiotics is a reality in many health centers. A clinical analysis with a cohort of 742 extremely low birth weight neonates with negative blood cultures re ported that 60% of the NBs received empirical anti biotics for more than three days20, concluding that the duration of initial empirical antibiotic therapy was an institutional decision and not based on obvious clini cal signs and symptoms of infection20.

In addition to the consequences that the prolonged exposure to antibiotics can have for NBs, this clinical practice implies an important cost for hospitals, both in terms of supplies and length of stay1,3,21-22. Based on international information and that obtained in the cu rrent study, it would not be reasonable or ethical to conduct a randomized controlled prospective study, but it would be possible to conduct a prospective, but not randomized, case-control study to determine if the use of empirical antibiotics prolonged for five days or more is associated with different morbidities and mortality. Therefore, the proper and justified use of antibiotics, especially in VLBW neonates, is a clinical as well as an institutional necessity.

In recent years there has been greater awareness re garding the responsible use of antibiotics, especially in VLBW neonates23. As a consequence, different strate gies have been developed in order to reduce the indication and duration of these treatments. One of them is the re-evaluation and adaptation of clinical guidelines for diagnosis and treatment of sepsis24 or the develo pment of predictive mathematical models22 that have succeeded in reducing the use and abuse of antibiotics.

Recently, strategies have also been implemented in Chile in order to adapt the use of antimicrobials in Neonatology units25, which has resulted in a reduction both in cases of late sepsis and in the associated costs25.

In conclusion, the use of initial empirical antibiotic therapy > 5 days was associated with an increased risk of late sepsis and NEC, no association was found with mortality. It is necessary to constantly evaluate and monitor the prudent use of initial empirical antibio tics in VLBW neonates and to timely discontinue the therapy if blood cultures are negative and the clinical condition of the neonate allows it.

Ethical Responsibilities

Human Beings and animals protection: Disclosure the authors state that the procedures were followed according to the Declaration of Helsinki and the World Medical Association regarding human experimenta tion developed for the medical community.

Data confidentiality: The authors state that they have followed the protocols of their Center and Local regulations on the publication of patient data.

Rights to privacy and informed consent: The authors have obtained the informed consent of the patients and/or subjects referred to in the article. This do cument is in the possession of the correspondence author.

Financial Disclosure: Authors state that no economic support has been asso ciated with the present study.

Conflicts of Interest: Authors declare no conflict of interest regarding the present study.


1. Hauge C, Stålsby Lundborg C, Mandaliya J, Marrone G, Sharma M. Up to 89% of neonates received antibiotics in cross-sectional Indian study including those with no infections and unclear diagnoses. Acta Paediatr. 2017;106:1674-83. [ Links ]

2. Clark RH, Bloom BT, Spitzer AR, Gerstmann DR. Reported medication use in the neonatal intensive care unit: data from a large national data set. Pediatrics. 2006;117:1979-87. [ Links ]

3. Cantey JB, Sánchez PJ. Prolonged antibiotic therapy for “culture-negative” sepsis in preterm infants: it’s time to stop!. J Pediatr. 2011;159:707-8. [ Links ]

4. Goldmann DA, Leclair J, Macone A. Bacterial colonization of neonates admitted to an intensive care environment. J Pediatr. 1978; 93:288-93. [ Links ]

5. Claude EC, Walker WA. Hypothesis: inappropriate colonization of the premature intestine can cause neonatal necrotizing enterocolitis. FASEB J. 2001; 15:1398-403. [ Links ]

6. Thompson PJ, Greenough A, Hird MF, Philpott-Howard J, Gamsu HR. Nosocomial bacterial infections in very low birth weight infants. Eur J Pediatr. 1992; 151:451-4. [ Links ]

7. Lin PW, Stoll BJ. Necrotising enterocolitis. Lancet. 2006; 368: 1271-83. [ Links ]

8. Gibbs K, Lin J, Holzman IR. Necrotising enterocolitis: the state of the science. Indian J Pediatr. 2007; 74:67-72. [ Links ]

9. Cotten CM, Taylor S, Stoll B, et al. Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics. 2009;123:58-66. [ Links ]

10. Alexander VN, Northrup V, Bizzarro MJ. Antibiotic exposure in the newborn intensive care unit and the risk of necrotizing enterocolitis. J Pediatr. 2011;159:392-7. [ Links ]

11. Kuppala V, Morrow A, Schibler K. Prolonged initial empirical antibiotic treatment is associated with adverse outcomes in premature infants. J Pediatr 2011;159: 720-5. [ Links ]

12. Abdel Ghany EA, Aliaa A. Empirical antibiotic treatment and the risk of necrotizing enterocolitis and death in very low birth weight neonates. Ann Saudi Med 2012;32:521-6. [ Links ]

13. Shah P, Nathan E, Doherty D, Patole S. Prolonged exposure to antibiotics and its associations in extremely preterm neonates-the Western Australian experience. J Matern Fetal Neonatal Med. 2013;26:1710-4. [ Links ]

14. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control. 1988;16:128-40. [ Links ]

15. Ottolini MC, Lundgren K, Mirkinson LJ, Cason S, Ottolini MG. Utility of complete blood count and blood culture screening to diagnose neonatal sepsis in the asymptomatic at risk newborn. Pediatr Infect Dis J. 2003;22:430-4. [ Links ]

16. Lee JS, Polin RA. Treatment and prevention of necrotizing enterocolitis. Semin Neonatol. 2003;8:449-59. [ Links ]

17. Ting JY, Synnes A, Roberts A, et al. Canadian Neonatal Network Investigators. Association Between Antibiotic Use and Neonatal Mortality and Morbidities in Very Low-Birth- Weight Infants Without Culture-Proven Sepsis or Necrotizing Enterocolitis. JAMA Pediatr. 2016;170:1181-7. [ Links ]

18. Cantey JB, Huffman LW, Subramanian A, et al. Antibiotic Exposure and Risk for Death or Bronchopulmonary Dysplasia in Very Low Birth Weight Infants. J Pediatr. 2017;181:289-93. [ Links ]

19. Briones-Lara E, Treviño-Báez J, Caballero-Trejo A, Iruegas-Maeda A, Palacios-Saucedo G, Ramírez-Rosalino M. Exposición prolongada a antibióticos y riesgo de sepsis tardía (ST) en neonatos de 1,000 a < 1,500 g: estudio de cohorte. Gac Med Mex. 2015;151:306-12. [ Links ]

20. Cordero L, Ayers LW. Duration of empiric antibiotics for suspected early-onset sepsis in extremely low birth weight infants. Infect Control Hosp Epidemiol. 2003; 24: 662-6. [ Links ]

21. Różańska A, Wójkowska-Mach J, Adamski P, et al. Antibiotic consumption in laboratory confirmed vs. non-confirmed bloodstream infections among very low birth weight neonates in Poland. Ann Clin Microbiol Antimicrob. 2017; 31:16:20. [ Links ]

22. Warren S, García M, Hankins C. Impact of neonatal early-onset sepsis calculator on antibiotic use within two tertiary healthcare centers. J Perinatol. 2017;37:394-397. [ Links ]

23. Ramasethu J, Kawakita T. Antibiotic stewardship in perinatal and neonatal care. Semin Fetal Neonatal Med. 2017;22:278-83. [ Links ]

24. Cotten CM. Antibiotic stewardship: reassessment of guidelines for management of neonatal sepsis. Clin Perinatol. 2015;42:195-206. [ Links ]

25. Urzúa S, Ferrés M, García P, Sánchez A, Luco M. Estrategias para reducir infecciones, uso de antimicrobianos y sus efectos en una unidad de neonatología. Revista Chilena de infectología, 2017;34: 99-107. [ Links ]

Received: August 07, 2017; Accepted: May 21, 2018

Correspondence: Aldo Bancalari Molina. E-mail:

Creative Commons License Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons