Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation

INTRODUCTION — Neonatal sepsis remains a major cause of neonatal mortality and morbidity in preterm and very low birth weight infants. Clinical care providers should have a high index of suspicion to diagnose infections in preterm and very low birth weight infants. Delays in initiating appropriate antibiotic therapy are associated with increased risk of morbidity and mortality.

This topic will review the treatment and prevention of bacterial sepsis in preterm infants. The clinical features and diagnosis of bacterial sepsis in preterm infants is discussed separately. (See "Clinical features and diagnosis of bacterial sepsis in preterm infants <34 weeks gestation".)

Sepsis in term and late preterm neonates is discussed separately. (See "Clinical features, evaluation, and diagnosis of sepsis in term and late preterm neonates" and "Management and outcome of sepsis in term and late preterm neonates".)

TERMINOLOGY — The following terms will be used throughout this topic:

●Preterm infants are those born at less than 34 weeks gestation.

●Late preterm infants (also called near-term infants) are those born between 34 and 36 completed weeks of gestation. Sepsis in late preterm infants is discussed in a separate topic review. (See "Management and outcome of sepsis in term and late preterm neonates".)

●Very low birth weight infants – Infants with birth weights <1500 g.

●Sepsis – Defined as isolation of a pathogenic bacterium from a blood culture.

•Early-onset sepsis (EOS) is defined as sepsis that occurs in the first 72 hours of life

•Late-onset sepsis (LOS) is defined as sepsis that occurs after the first 72 hours of life

●Health care-associated infections are defined as infections (eg, sepsis) acquired in the hospital while receiving treatment for other conditions [1].

TREATMENT — The management of sepsis in preterm infants consists of supportive care and antibiotic treatment, which includes initial empiric and organism-specific therapy. Although a variety of adjunctive interventions have been studied, none have been shown to improve the outcome of infants with sepsis.

The management of neonatal sepsis in preterm infants is reviewed here. The approach discussed below is generally consistent with guidelines published by the American Academy of Pediatrics (AAP) [2]. (See 'Society guideline links' below.)

Supportive care — Supportive care is initially focused on ensuring adequate systemic oxygenation and peripheral perfusion. In particular, aggressive resuscitative intervention is required in patients with fulminant sepsis, defined as severe sepsis or septic shock that is likely to result in death within 48 hours. (See "Clinical features and diagnosis of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Severe sepsis and septic shock' and "Neonatal shock: Etiology, clinical manifestations, and evaluation".)

General supportive care includes:

●Respiratory support with supplemental oxygen and/or noninvasive support (eg, continuous positive airway pressure [CPAP]), as needed. Severely affected patients may require invasive mechanical ventilation may be required. (See "Respiratory support, oxygen delivery, and oxygen monitoring in the newborn" and "Approach to mechanical ventilation in very preterm neonates".)

●Maintenance of a thermoneutral environment. (See "Overview of short-term complications in preterm infants", section on 'Prevention of hypothermia'.)

●Hemodynamic support with intravenous (IV) fluids and, if needed, vasoactive medications (eg, dopamine). (See "Fluid and electrolyte therapy in newborns" and "Neonatal shock: Management".)

Source control — When possible, measures should be undertaken to eradicate a focus of infection and/or eliminate ongoing microbial contamination because localized foci of infection (ie, abscess) may not respond to antibiotics alone. In particular, early removal of catheters that may be the foci of bacterial infection should be removed as early as possible. Delayed removal (>48 hours after diagnosis of sepsis) or failure to remove catheters are associated with an increased risk of complications (eg, end-organ damage and thrombocytopenia) and persistent bacteremia [3-5].

Antibiotic therapy — Antibiotic therapy includes initial empiric and organism-specific therapy. However, because of increasing antibiotic resistance, the choice of antibiotic agents should be carefully selected to reduce the likelihood of antibiotic-resistant bacteria within each neonatal intensive care unit.

Antibiotic resistance — There has been an increasing emergence of infections in preterm infants caused by antibiotic-resistant, gram-negative bacteria [6-8]. As an example, in studies from the National Institute of Child Health and Human Development (NICHD) Neonatal Research Network, approximately 80 percent of early-onset Escherichia coli infections were resistant to ampicillin [6,7,9-11]. Risk factors associated with acquisition of antibiotic-resistant bacteria include very low birth weight (birth weight <1500 g) and the exposure to third-generation cephalosporins.

Two mechanisms resulting in increased antibiotic resistance are production of the following [12-15] (see "Extended-spectrum beta-lactamases"):

●Chromosomally encoded or plasmid-derived AmpC beta-lactamases.

●Plasmid-mediated extended-spectrum beta-lactamases (ESBLs) [12-15]. Organisms that produce ESBLs, primarily E. coli and Klebsiella species, are resistant to penicillins, cephalosporins, and monobactams and can be resistant to aminoglycosides.

ESBL and AmpC beta-lactamase-producing organisms can be effectively treated using fourth-generation cephalosporins (eg, cefepime) [16,17] and carbapenems (eg, meropenem) [18,19]. Carbapenems (meropenem, imipenem) are a unique class of beta-lactam agents that are stable against most plasmid and chromosomal-mediated beta-lactamases and are effective against more than 90 percent of ESBL-producing Enterobacteriaceae. Meropenem is the preferred carbapenem in newborn infants as the safety profiles of other carbapenems have not been established in neonates [20]. Although gentamicin resistance is not common, the aminoglycosides amikacin or netilmicin, which are resistant to the aminoglycoside-modifying enzymes of the bacteria, can be used in patients who are infected with a gentamicin-resistant pathogen [21].

Empiric antibiotic therapy — Empiric antibiotic therapy should be initiated for infants with suspected sepsis once the evaluation has been completed because of the risk of death and severe morbidity (see 'Outcomes' below). The initial choice of parenteral antimicrobials for suspected sepsis in the preterm neonate is based on the likely pathogens, the susceptibility patterns of organisms in a particular nursery, and the presence of an apparent source of infection (eg, skin, joint, central line, or bone involvement).

Early-onset sepsis — The combination of ampicillin and gentamicin is effective in treating most of the common pathogens that cause early-onset sepsis (EOS) in preterm infants, such as group B streptococcus and E. coli [2].The choice of empiric antibiotics for EOS is generally the same as in term infants, which is discussed in greater detail separately (see "Management and outcome of sepsis in term and late preterm neonates", section on 'Early-onset sepsis'). However, dosing of these agents in preterm neonates differs from the dosing in term neonates, as summarized below. (See 'Antibiotic dosing' below.)

Because of the emergence of cephalosporin-resistant organisms, especially Enterobacter, Klebsiella, and Serratia species, the routine use of a third-generation cephalosporin to treat neonatal sepsis is generally discouraged, except for patients with suspected gram-negative bacterial meningitis because of its excellent penetration in the cerebrospinal fluid. In addition, severely ill preterm infants born under conditions that pose high risk for gram-negative EOS (eg, prolonged rupture of membranes with obstetric concern for intra-amniotic infection) may benefit from broader-spectrum empiric antibiotic coverage until culture results are known. (See "Bacterial meningitis in the neonate: Treatment and outcome", section on 'Empiric therapy'.)

Late-onset sepsis — In preterm infants with late-onset sepsis (LOS), the choice of empiric antibiotic therapy should be based on the likely organism and its pattern of antibiotic susceptibility for a particular nursery and the clinical setting.

The predominant causative organism for LOS in preterm infants is coagulase-negative staphylococci (CoNS); Staphylococcus aureus and gram-negative bacteria are also common pathogens (table 1). Thus, for most neonates with suspected LOS, we suggest a combination of vancomycin and gentamicin as initial empiric antibiotic therapy while awaiting isolation of the causative organism from culture and its antibiotic susceptibility. However, local antibiotic susceptibility patterns should be considered.

Alternative empiric regimens are used in select cases, based on additional clinical factors as follows:

●If meningitis is suspected, we add an expanded-spectrum cephalosporin (eg, cefotaxime [where available], ceftazidime, or cefepime). IV ceftriaxone should not be used in neonates, especially preterm neonates, due to the risk of adverse events (ie, increased risk of bilirubin neurotoxicity in neonates with hyperbilirubinemia and risk of adverse cardiopulmonary events when given concomitantly with IV calcium, including parenteral nutrition) [22]. (See "Bacterial meningitis in the neonate: Treatment and outcome", section on 'Empiric therapy'.)

●In neonates with a focal site infection, such as soft tissue, skin, joint, or bone involvement, nafcillin or oxacillin should be added to the empiric regimen to provide adequate coverage for S. aureus. (See "Staphylococcus aureus in children: Overview of treatment of invasive infections", section on 'Empiric antimicrobial therapy'.)

●If gram-negative infection is suspected or the course is fulminant, we add ceftazidime for coverage of a potential Pseudomonas infection.

In a retrospective study of 3339 neonates with S. aureus infection in 348 neonatal intensive care units, inadequate empiric antibiotic therapy (defined as not including ≥1 antibiotic with antistaphylococcal activity on day 1 of therapy) was associated with increased 30-day mortality (odds ratio 2.03, 95% CI 1.08-3.82) among infants with infection due to methicillin-resistant S. aureus (MRSA) but not among those with methicillin-sensitive S. aureus (MSSA) infection [23]. A similar study found that among neonates with LOS due to CoNS, vancomycin started on day one of therapy did not decrease 30-day mortality compared with delayed vancomycin therapy started after blood culture results, although it decreased the median duration of bacteremia by one day [24].

Antibiotic dosing — Our suggested dosing for IV ampicillin, vancomycin, and gentamicin is dependent on weight, GA, and postnatal age [20,25,26]. However, other centers may use different dosing regimens.

●Ampicillin – Ampicillin dosing for neonates ≤34 weeks GA is based on postnatal age [25,26]:

•Postnatal age ≤7 days – 50 mg/kg per dose IV every 12 hours

•Postnatal age >7 days – 75 mg/kg per dose IV every 12 hours

●Vancomycin – At the author's institution, initial vancomycin dosing in neonates ≤34 weeks GA is based on weight and postnatal age [25]:

•Weight <1200 g and postnatal age <28 days: 15 mg/kg/dose IV every 18 to 24 hours

•Weight 1200 to 2000 g and postnatal age <7 days: 15 mg/kg/dose IV every 12 to 18 hours

•Weight 1200 to 2000 g and postnatal age ≥7 days: 15 mg/kg/dose IV every 8 to 12 hours

•Weight >2000 g and postnatal age <7 days: 15 mg/kg/dose IV every 8 to 12 hours

•Weight >2000 g and postnatal age >7 days: 15 mg/kg/dose IV every 6 to 8 hours

For neonates receiving ongoing vancomycin therapy, trough levels should be monitored and dosing adjusted as needed to maintain the desired therapeutic target (typically 8 to 20 mcg/mL) [25].

Alternative dosing recommendations based on gestational age and serum creatinine are also available (refer to pediatric drug information for vancomycin) [20].

●Gentamicin – IV gentamicin dosing for preterm neonates is based on both GA and postnatal age [20,25].

•For neonates <30 weeks GA:

-Postnatal age ≤14 days – 5 mg/kg per dose IV every 48 hours

-Postnatal age >14 days – 5 mg/kg per dose IV every 36 hours

•For neonates 30 to 34 weeks GA:

-Postnatal age ≤14 days old – 5 mg/kg per dose IV every 36 hours

-Postnatal age >14 days old – 5 mg/kg per dose IV every 24 hours

Organism-specific therapy — Empiric antimicrobial therapy is altered based upon the isolated pathogen and its antimicrobial susceptibility pattern [15,27]. In general, we use the following organism-specific parenteral antibiotic therapy:

●CoNS – Vancomycin.

●S. aureus – Directed therapy is based on the sensitivity of the isolate to specific antibiotics. For infections caused by MSSA, therapy can be completed with nafcillin or oxacillin alone. MRSA infections are usually treated with vancomycin. (See "Staphylococcus aureus in children: Overview of treatment of invasive infections".)

●E. coli – Directed therapy is based on the sensitivity of the isolate to specific antibiotics. For infants with ampicillin-sensitive isolates, parenteral ampicillin can be used alone. In those with ampicillin-resistant infection, either an appropriate aminoglycoside (gentamicin) or an extended-spectrum cephalosporin (eg, cefotaxime, if available, or cefepime) can be used.

●Other gram negatives (eg, Klebsiella, Serratia, Enterobacter, Citrobacter) – Directed therapy is based on the sensitivity of the isolate to specific antibiotics. If the organism is susceptible, an extended-spectrum cephalosporin (eg, cefepime) can be used. However, these species are often ESBL-producing organisms. The drug of choice for ESBL-producing organisms is meropenem.

●Pseudomonas – Combination therapy of gentamicin, and ceftazidime or piperacillin/tazobactam.

Extremely low birth weight (ELBW; <1000 g) neonates who receive a full course of antibiotic therapy for culture-proven sepsis are at high risk for developing invasive Candida infections. Prophylactic antifungal therapy is suggested for these neonates, as discussed separately. (See "Candida infections in neonates: Treatment and prevention", section on 'Targeted prophylaxis'.)

Duration and response to therapy — Duration of antibiotic therapy is usually 10 to 14 days for uncomplicated bloodstream infections [28]. A longer duration of therapy is necessary in patients with meningitis (ie, 14 days for uncomplicated gram-positive meningitis and a minimum of three weeks for gram-negative meningitis), as discussed separately. (See "Bacterial meningitis in the neonate: Treatment and outcome", section on 'Duration'.)

For infants with negative cultures, the decision to continue or stop antibiotic therapy is individualized based on the clinical status of the neonate and the judgment of the attending neonatologist. In general, antibiotic therapy should be discontinued if cultures are sterile at 36 to 48 hours and there are no ongoing clinical concerns for sepsis. (See "Clinical features and diagnosis of bacterial sepsis in preterm infants <34 weeks gestation".)

As discussed below, antibiotic stewardship is an important measure to reduce the prevalence of antibiotic-resistant bacteria and fungal infections. A key part of such practice is to ensure timely discontinuation of antibiotics once bacterial infection has been deemed to be unlikely. Antibiotic stewardship protocols that include early discontinuation of antibiotics if cultures are negative at 36 hours (instead of 48 hours) have decreased the use of antibiotics without apparent adverse effects [29,30]. (See 'Infection control measures' below.)

Therapies with uncertain benefit — Several adjunctive immunomodulatory therapies have been evaluated for the treatment of neonatal sepsis, including intravenous immune globulin (IVIG), granulocyte transfusions, stimulating factors (ie, granulocyte colony-stimulating factor [G-CSF] and granulocyte-macrophage colony-stimulating factor [GM-CSF]), and pentoxifylline. None have been shown to conclusively improve outcomes of neonatal sepsis and we suggest not using them.

●IVIG – Based on the available evidence, we suggest not using IVIG for treatment of neonatal sepsis. It has been proposed that IVIG may be of benefit in preterm infants <32 weeks gestation with serious bacterial infection since most of the fetal transfer of maternal immunoglobulin occurs after 32 weeks gestation. However, several clinical trials have failed to demonstrate a clinical benefit of IVIG administration in neonates with suspected or confirmed sepsis [31,32].

●Granulocyte transfusions – There are limited data on granulocyte transfusion in neonatal sepsis. In a meta-analysis of four small trials that were conducted in the 1980s involving 79 infants with sepsis and neutropenia, treatment with granulocyte transfusion did not appear to have meaningful impact on morbidity or mortality [33]. The main adverse effects were pulmonary complications, which occurred in 4 percent of treated infants.

●G-CSF and GM-CSF – A meta-analysis of seven small trials (257 infants in total) that were conducted in the 1990s and early 2000s did not detect a significant reduction in mortality in neonates with sepsis who were treated with G-CSF or GM-CSF [34]. In a subgroup analysis of 97 neutropenic patients, mortality appeared to be lower in neonates who received G-CSF or GM-CSF (8 versus 26 percent; relative risk [RR] 0.34, 95% CI 0.12-0.92) [34]. However, the small number of events (17 deaths in total) precludes drawing any firm conclusion. In a follow-up report of the infants enrolled in one of these trials, neurodevelopmental outcomes, general health, and educational outcomes at age five years were similar in patients who received GM-CSF compared with placebo [35].

●Pentoxifylline – Pentoxifylline, a xanthine derivative, inhibits the release of tumor necrosis factor (TNF)-alpha, which is generally associated wⁱth systemic gram-negative infection. Limited data suggest that the addition of pentoxifylline to antibiotic therapy reduces mortality in neonates with sepsis. In a meta-analysis of six trials (416 neonates), pentoxifylline therapy was associated with a decrease in all-cause mortality during hospital stay (10 versus 17 percent; relative risk [RR] 0.57, 95% CI 0.35-0.93) [36]. However, the trials in the meta-analysis were small and most of them had important methodologic limitations (four of the six trials were judged to be at high risk of bias by the meta-analysis authors). Thus, the certainty of these findings is low. Larger well conducted trials are needed to confirm these findings before pentoxifylline can be recommended routinely in the treatment of neonatal sepsis.

PREVENTION — Prevention of neonatal sepsis due to health care-associated infections focuses primarily on infection control measures.

Infection control measures — Strategies to reduce nosocomial infections in the neonatal intensive care unit (NICU) include [37]:

●Hand hygiene – Hand hygiene remains one of the most effective methods for reducing healthcare-associated infections [38,39]. Detailed discussion of appropriate hand hygiene is provided in a separate topic review. (See "Infection prevention: Precautions for preventing transmission of infection", section on 'Hand hygiene'.)

●Catheter care – In neonates with central lines, strategies to prevent catheter-related blood stream infections (CRBSIs) include:

•Setting and adhering to institutional guidelines for the insertion and care of indwelling lines [40,41]. Guidelines should include the use of sterile technique and antiseptic agents at the site during line placement, daily monitoring of catheter sites, and redressing and cleaning the site on a weekly basis. Tubing used to infuse dextrose and amino acids should be replaced every four to seven days. Catheters should be removed promptly when they are no longer essential because the risk of infection generally increases with time; however, studies investigating the effect of dwell time on risk of infection with peripherally inserted central catheters have been contradictory [42,43]. We do not recommend routine use of antibiotic lock therapy for prevention of CRBSI in neonates. In a meta-analysis of three studies (271 infants), antibiotic lock therapy appeared to be effective in preventing CRBSI in neonates [44]. However, studies have not comprehensively assessed the effect of this intervention on development of resistant organisms. (See "Routine care and maintenance of intravenous devices".)

•Use of closed system of drug delivery [45].

•Promotion of early enteral feeding with breast milk, thereby reducing the need for or length of use of central venous lines and total parenteral nutrition.

●Avoidance of overcrowding, contact precautions, and cohorting – Other infection control measures include avoidance of overcrowding and use of contact precautions (ie, gown and gloves) when appropriate. In addition, when outbreaks occur within a NICU, cohorting patients and assigning dedicated nursing staff to such patients may reduce the spread. (See "Nosocomial infections in the intensive care unit: Epidemiology and prevention", section on 'Contact precautions, cohorting, and dedicated staff'.)

●Antibiotic stewardship – Antibiotic stewardship refers to judicious use of antibiotic therapy and is aimed at reducing the risk of antibiotic resistance and fungal infection [8,46]. This includes limiting antibiotic therapy to clinical situations in which bacterial infection is likely, discontinuing empirical therapy when a bacterial infection is not identified, and changing therapy to the narrowest spectrum based on susceptibility testing. In particular, restricting the use of third-generation cephalosporins may decrease the induction of extended-spectrum beta-lactamases (ESBLs). (See 'Antibiotic resistance' above.)

●Surveillance – Surveillance for infections with multidrug-resistant bacteria within the institution as a whole and within specific NICUs is critical for the early identification and control of epidemic outbreaks and endemic increases of resistant bacteria. The prevalence of isolation of multidrug-resistant bacteria (eg, MRSA, VRE, and ESBL-producing organisms) should be monitored, and these data should be disseminated to the clinical staff in the NICU as these data may impact the choice of empiric antibiotic therapy. (See "Nosocomial infections in the intensive care unit: Epidemiology and prevention", section on 'Surveillance'.)

●Implementing protocols and monitoring compliance – Continued quality improvement focused on increasing health care staff awareness and education, establishing common improvement goals, training, environmental care, and setting guidelines for patient care [47].

Potential prophylactic therapy — Prophylactic therapies that have been studied for prevention of sepsis in preterm neonates include lactoferrin and probiotics. Based on the available evidence, we suggest not using these therapies.

●Lactoferrin – Lactoferrin supplementation has been proposed as an intervention to prevent sepsis in preterm neonates. Lactoferrin is an iron-binding glycoprotein and a component of the mammalian innate response to infection. It is the major whey protein in colostrum, breast milk, tears, and saliva.

Based on the available evidence, it remains uncertain if lactoferrin supplementation reduces the risk of LOS in preterm infants. If there is an effect, it appears to be small. Thus, we suggest not routinely using lactoferrin in this setting.

In 2020 meta-analysis of eight trials (3575 infants), enteral lactoferrin supplementation decreased the rate of culture-proven bacterial LOS compared with placebo (13.9 versus 16.1 percent; relative risk [RR] 0.86, 95% CI 0.74-1.0) [48]. However, the finding was of borderline statistical significance and the absolute effect size was small (absolute risk difference 2.2 percent, 98% CI 0-4.2). In the largest trial, which involved >2200 neonates <32 weeks gestation randomized to enteral bovine lactoferrin or placebo (sucrose), the rate of LOS in the lactoferrin group was similar to that in the placebo group (29 versus 31 percent, respectively; relative risk [RR] 0.95, 95% CI 0.86-1.04); mortality rates were also similar (7 versus 6 percent; RR 1.05, 95% CI 0.66-1.68) [49].

●Probiotics – We suggest not routinely using probiotics (defined as live nonpathogenic microbial preparations that colonize the intestine) for the prevention of LOS in preterm infants. The efficacy of probiotics in preventing LOS and reducing mortality is unproven, and there are important concerns and uncertainties regarding appropriate dosing, strain selection, safety, and regulation of these products.

The two largest trials investigating the efficacy and safety of probiotics in preterm infants are the PiPS trial and the ProPrems trial [50,51]. PiPS was a multicenter trial involving >1300 neonates between 23 and 30 weeks gestation randomized to receive probiotic (Bifidobacterium breve BBG-001) or placebo (dilute formula) [50]. Rates of LOS were similar in both groups (11 versus 12 percent; RR 0.97, 95% CI 0.73-1.29). Rates of necrotizing enterocolitis (NEC) and mortality were also similar (NEC 9 versus 10 percent; RR 0.93, 95% CI 0.68-1.27; mortality 8 versus 9 percent; RR 0.93, 95% CI 0·67-1.30). ProPrems was an earlier multicenter trial involving 1099 preterm infants (gestational age <32 weeks and birth weight <1500 g) randomized to probiotic (consisting of a combination of Bifidobacterium infantis, Streptococcus thermophilus, and Bifidobacterium lactis) or placebo (maltodextrin) [51]. Rates of LOS were similar in both groups (13 and 16 percent, respectively); however, rates of NEC were slightly lower in the probiotic group compared with placebo (2 versus 4 percent, respectively).

In a 2016 meta-analysis of 37 trials (including both PiPS and ProPrems), probiotics were associated with a small but statistically significant reduction in the risk of LOS compared with placebo or no treatment (14 versus 16 percent; RR 0.86, 95% CI 0.78-0.94) [52]. The report did not include pooled estimates for other outcomes such as mortality and NEC. A separate meta-analysis found that probiotics reduced mortality (5 versus 7 percent; RR 0.72, 95% CI 0.57-0.92), with the effect most pronounced in studies with high proportions of exclusively breastfed neonates [53]. However, the investigators detected significant publication bias in favor of probiotics, which is an important limitation of these data.

Although the pooled data suggest probiotics may have a beneficial effect in reducing rates of LOS, the absolute effect size is small and it remains uncertain whether they reduce mortality. Other important uncertainties remain, including optimal probiotic strains, doses, and duration of therapy. Appropriate regulatory control of these products is another unresolved issue. In addition, rare but serious cases of probiotic-associated sepsis have been reported [54-57]. Given these uncertainties and concerns, we suggest not routinely using probiotics for the purpose of preventing LOS in preterm infants.

The use of probiotics for prevention of NEC is discussed in detail separately. (See "Neonatal necrotizing enterocolitis: Prevention", section on 'Probiotics'.)

OUTCOMES

Mortality — Neonatal sepsis remains a major contributor to mortality among preterm very low birth weight infants with reported mortality rates of approximately 20 to 30 percent for early-onset sepsis (EOS) and 15 to 25 percent for late-onset sepsis (LOS) [9,58-60]. Among extremely preterm neonates (variably defined in these studies as <26 to <28 weeks gestation), mortality rates as high as 40 percent for EOS and 30 percent for LOS have been reported [61,62].

The risk of mortality is greater in infants with gram-negative compared with gram-positive infections [7,63-68].

In a single-center study of 424 very low birth weight infants with LOS, factors that were independently associated with increased risk of mortality included [67]:

●Gram-negative or fungal pathogen

●Need for intubation

●Need for vasopressor therapy

●Hypoglycemia

●Thrombocytopenia

●Necrotizing enterocolitis (NEC)

Morbidity — Preterm infants with sepsis are at risk for both short- and long-term complications.

●Short-term complications – Sepsis increases the risk of patent ductus arteriosus, prolonged mechanical ventilation, bronchopulmonary dysplasia, NEC, and duration of hospital stay [58,60]. Gram-negative infections compared with gram-positive infections are associated with a greater morbidity and longer length of hospital stay [69]. (See "Overview of short-term complications in preterm infants".)

●Long-term complications – Sepsis is a risk factor for long-term neurodevelopmental impairment (NDI), including cognitive or motor delays, cerebral palsy, and/or sensory impairment (ie, vision or hearing loss) [61,70-74]. Sepsis affects neurodevelopment either by direct infection of the central nervous system or indirectly due to the impact of other associated comorbidities. Survivors of neonatal sepsis are also more likely to require home oxygen, tracheostomy, and/or gastrostomy compared with infants born at similar GA who did not have sepsis [59]. Long-term complications of preterm birth and risk factors for NDI are discussed in greater detail separately. (See "Overview of the long-term complications of preterm birth" and "Long-term neurodevelopmental impairment in infants born preterm: Epidemiology and risk factors".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Sepsis in newborn babies (The Basics)")

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sepsis in neonates" and "Society guideline links: Group B streptococcal infection in pregnant women and neonates".)

SUMMARY AND RECOMMENDATIONS

●Supportive care – General supportive care for preterm infants with bacterial sepsis includes respiratory and hemodynamic support and maintaining a thermoneutral environment. Catheters that may be foci of bacterial infection should be removed promptly once the diagnosis of sepsis is made (ideally within 48 hours). (See 'Supportive care' above.)

●Empiric antibiotic therapy – Prompt administration of empiric antibiotic therapy is critical for neonates with clinically suspected sepsis. The empiric regimen should provide broad coverage for the most likely pathogens (ie, for early-onset sepsis [EOS], group B streptococcus and Escherichia coli; for late-onset sepsis [LOS], coagulase-negative staphylococci [CoNs], Staphylococcus aureus, and gram-negative bacteria) (table 1). Local antibiotic susceptibility patterns should also be considered. Our general approach is as follows (see 'Empiric antibiotic therapy' above):

•For most preterm neonates with suspected EOS, we suggest ampicillin plus gentamicin as the empiric regimen rather than other regimens (Grade 2C). (See 'Early-onset sepsis' above and "Management and outcome of sepsis in term and late preterm neonates", section on 'Early-onset sepsis'.)

•For most preterm neonates with suspected LOS, we suggest vancomycin plus gentamicin (Grade 2C). Alternate regimens used in select clinical circumstances are discussed above. (See 'Late-onset sepsis' above.)

●Specific therapy – Antibiotic therapy is altered based upon isolation of the causative agent and its antimicrobial susceptibility pattern. (See 'Organism-specific therapy' above.)

●Duration of therapy – Duration of therapy depends on the results of blood culture and clinical course. For uncomplicated bloodstream infections, antibiotic therapy is continued for 10 to 14 days. For infants with negative cultures, the decision to continue or stop antibiotic therapy is individualized based on the clinical status of the neonate and the judgment of the attending neonatologist. In general, antibiotic therapy should be discontinued if cultures are sterile at 36 to 48 hours and there are no ongoing clinical concerns for sepsis. (See 'Duration and response to therapy' above.)

●No role for adjunctive immunomodulatory therapy – We suggest not using adjunctive immunomodulatory therapies (eg, intravenous immune globulin [IVIG], granulocyte transfusion, granulocyte and granulocyte-macrophage colony-stimulating factor, or pentoxifylline) (Grade 2C). (See 'Therapies with uncertain benefit' above.)

●Prevention – Prevention of neonatal sepsis due to health care-associated infections focuses primarily on infection control measures including hand hygiene, adherence to guidelines for the insertion and maintenance of indwelling lines, and antibiotic stewardship. We suggest not routinely using prophylactic therapies such as lactoferrin or probiotics (Grade 2C). (See 'Prevention' above.)

●Outcome – Neonatal sepsis remains a major cause for neonatal mortality and morbidity in preterm and very low birth weight infants. Reported mortality rates are approximately 25 percent for EOS and 15 to 20 percent for LOS. Gram-negative infections are associated with higher mortality. (See 'Outcomes' above.)