Predictors of Mortality and Severe Illness from Escherichia coli Sepsis in Neonates

doi:10.21203/rs.3.rs-3973889/v1

Background: Neonatal Escherichia coli (E. coli) sepsis is increasing. There is limited data on the factors which contribute to increased mortality and severity of illness in neonatal E. colisepsis.

Methods: Retrospective review of neonates (<30 days) admitted to a Level IV NICU in the United States from 2008 to 2022 with a diagnosis of E. coli bloodstream or cerebrospinal fluid infection was conducted. Primary outcome was defined as mortality from or severe illness during E. coliinfection (defined as need for inotropic support or metabolic acidosis).

Results: E. coli neonatal sepsis rate increased from 2008-2022 (average of 1.12 per 1,000 live births). The primary outcome, which occurred in 50% of cases, was independently associated with prematurity, neutropenia, and thrombocytopenia. Ampicillin resistance was not associated with the primary outcome.

Conclusion: GA, neutropenia, and thrombocytopenia but not ampicillin resistance, are associated with mortality or severe illness from E. colisepsis.

Health sciences/Diseases/Infectious diseases/Bacterial infection

Health sciences/Risk factors

antibiotic resistance

ampicillin

neonatal intensive care unit

Neonatal bacterial infections have significant morbidity and mortality with an estimated annual 6.3 million cases and 230,000 deaths globally.¹ In the United States, the incidence of neonatal bacterial sepsis varies between one and four per 1,000 live births (LBs).² Neonatal sepsis is categorized into early-onset (EOS, infants ≤ 72 hours old) and late-onset sepsis (LOS, infants ≥ 72 hours). EOS classically represents maternal-to-fetal transmission, while most LOS are nosocomial or community-acquired.³ As antenatal Group B Streptococcus (GBS) screening and intrapartum antibiotic prophylaxis (IAP) have led to significantly decreased GBS EOS, Escherichia coli (E. coli) is now one of the predominant organisms responsible for EOS.^{4, 5, 6, 7} Within the National Institute of Child Health and Development (NICHD) Neonatal Research Network (NRN) neonatal intensive care units (NICUs), 36.6% of EOS cases are due to E. coli (0.4 per 1 000 live births), compared to 30.2% from GBS.⁴ The Centers for Disease Control and Prevention (CDC) Active Bacterial Core surveillance data from 2005–2014 reported that 24.8% of EOS (0.2 per 1,000 live births) in the US were E. coli infections.⁸ The E. coli fraction of EOS cases increases to 58% (NRN) and 44% (CDC) for very low birth weight (VLBW, < 1500g BW) infants. E. coli is also the leading Gram-negative organism in LOS (5–12%) in preterm infants in the US.^{6, 9, 10} E. coli sepsis in neonates is associated with substantial mortality which is influenced by the gestational age (GA), birth weight (BW), and immune status of the neonate.

Although evaluation of suspected EOS has changed over time, the first-line drugs still remain ampicillin and gentamicin.¹¹ There has been growing concern regarding the emergence of multidrug-resistant E. coli strains and reports of increasing resistance to ampicillin and gentamicin.^{12, 13, 14} The rates of ampicillin and gentamicin resistance in E. coli isolates from US NICUs between 66–80% and 8–17% respectively.^{4, 15} The use of ineffective empiric antibiotics due to resistance results in inadequate infection control and disease progression to severe illness or death

There are inconsistent reports of association between risk of death among infants with E. coli EOS and ampicillin resistance.¹⁶ It is critical to identify the risk factors associated with mortality and severe illness from E. coli sepsis so that antibiotic stewardship programs and aggressive antibiotic regimens for neonates at risk can coexist in the NICU. We determined specific risk factors associated with mortality or severe illness in E. coli neonatal sepsis.

We conducted a retrospective review of all neonates ≤ 30 days old with E. coli positive blood or cerebrospinal fluid (CSF) culture admitted to Rainbow Babies and Children’s Hospital, a Level IV NICU in Cleveland, Ohio, from 01/01/2008 through 12/31/2022 with Institutional Review Board. Information was collected on infant and maternal clinical and demographic variables. Infants with congenital anomalies and conditions that would predispose to infection, such as immunodeficiency syndromes, were excluded from statistical analysis. Microbiological information included hours of life at culture collection, source (blood vs CSF), time to positivity, and antibiotic resistance profile. Infant clinical characteristics included outcome (discharged home vs death due to E. coli infection), complete blood counts (CBC) during sepsis, duration of antibiotic therapy, presence of metabolic acidosis (pH < 7.20 or base deficit > 10 mEq/L), and need for inotropic medication during the E. coli sepsis episode. Maternal characteristics included intra-amniotic infection (IAI), E. coli infection in the mother at any time during pregnancy, and maternal intrapartum antibiotic therapy.

We defined our primary outcome as either death or severe illness from E. coli infection. Death was attributed to E. coli infection if it happened at any point from collection of the blood or CSF culture till the end of antibiotic treatment for that particular episode of infection. Severe illness was defined as the need for inotropic medication or metabolic acidosis, as described above, during the episode of E. coli infection. The time of detection of sepsis, which was used to differentiate between EOS and LOS, was based on the time of collection of blood or CSF culture relative to time of birth.

Statistical analysis:

Charts which met inclusion and exclusion criteria were checked for completeness and consistency, and coded into RStudio. Descriptive data were obtained and summary statistics were created. These summary tables included the associated p-value stratified for the variable of interest. The primary outcome variable was death or severe illness defined as above. For all analyses, this composite outcome was used for descriptive statistics and logistic regression modeling. Bivariate analysis was performed to identify co-linearity between variables of interest. Those variables which were strongly associated (p < 0.05) with the outcome variable and relevant to our study were used. Multivariable logistic regression was performed using a backward stepwise method to identify key variables to improve the model's overall fit to the dataset. The model fitness was assessed using the goodness of fit test to estimate if the model fits the data appropriately. Odds ratios were identified and represented in an odds ratio plot with a 95% confidence interval. This odds ratio plot was used to identify and assess the direction of association between the predictor variables and the outcome variable. P-values were reported for each of the given variables in the multivariate logistic regression model. The generalized variance inflation factor (GVIF) was used to identify co-linearity between predictor variables numerically. Those variables with GVIF greater than five were excluded from the model despite these parameters adding significance to the model.

Demographics and incidence of E. coli sepsis:

In the 59 984 live-born infants from 2008 to 2022, there were 68 unique cases of monomicrobial E. coli bloodstream or CSF infection in neonates under 30 days old (1.12 cases per 1,000 LBs). E. coli sepsis incidence per 1,000 LBs has steadily increased at our center from 0.35 (7/19 716) in 2008 – 2012 to 1.03 (21/20 331) in 2013 – 2017 and to 2.01 (40/19 932) in the last five years (2018 – 2022) respectively. The incidence of E. coli sepsis in VLBW infants was 12.4 per 1,000 LBs. There were 38 cases of EOS and 30 cases of LOS. Median GA and BW of infants with E. coli sepsis were 31 2/7 wks (27 4/7 – 38 0/7 wks) and 1620 g (970 – 2835 g) respectively. There were three cases of E. coli meningitis; two grew E. coli from their blood cultures simultaneously. The median age of diagnosis was 31.5 h (1.9 – 173.8 h).

Maternal IAI:

Twenty-one infants (30.8%) were born to mothers with a documented IAI (maternal peripartum fever ≥ 38⁰C). EOS was significantly higher in these infants vs. LOS (p<0.001). Only two E. coli LOS cases occurred in infants born to mothers with IAI. There were no differences between rates of mortality, severity of illness, or the primary outcome between infants who were born to mothers with IAI vs those who were not.

Early and late-onset E. coli sepsis:

There were 38 EOS and 30 LOS cases during the 15-year study period (Table 3). Males were more likely to have LOS than females, who were more likely to have EOS. The median GA or BW was not different between EOS and LOS infants. Although more VLBW infants were in the EOS group than the LOS group (55% vs 33%, p=0.071), this did not reach statistical significance. There was no difference between ampicillin or gentamicin resistance between EOS and LOS E. coli isolates. Mortality was not different between EOS and LOS cases, and neither was a need for inotropic medication or severe metabolic acidosis.

Primary outcome of mortality or severe illness from E. coli sepsis:

Thirty-four (50%) infants out of the 68 diagnosed with E. coli sepsis met the primary outcome (Table 2). 20/34 (58.8%) of these infants had a BW < 1500 g (VLBW). Out of all VLBW infants diagnosed with E. coli sepsis, 64.5% experienced death or severe illness in comparison to 37.8% in infants with BW ≥ 1500 g who were diagnosed with E. coli sepsis (OR 3.299, 95% CI 1.447 – 7.775, p=0.028). The median age (IQR) at which E. coli infection was diagnosed for infants with the primary outcome was 31.5 h (2.3 – 162), with 20/34 (58.5%) of these infants having EOS. Twenty-five of the 34 infants (73.5%) had ampicillin-resistant E. coli, which was not statistically different from the degree of ampicillin-resistance in the entire cohort (69.1%), or in the infants who survived to discharge without severe E. coli sepsis (64.7%). Apart from GA (and its surrogate marker, BW), the lowest platelet count and the lowest ANC during the episode of sepsis were the only other variables significantly different between those infants who died or had severe illness from E. coli sepsis vs those who did not in our univariate analysis (Table 3). Infants who died or had a severe illness were also more likely to be thrombocytopenic or neutropenic during the episode of sepsis. There was no association between the primary outcome and sex, race, ampicillin or gentamicin resistance, mode of delivery, or maternal IAI. None of the three cases of E. coli meningitis died or had severe illness. The median antibiotic use duration for E. coli sepsis in those infants who survived (with or without severe illness) was 21 days.

In our multivariate logistic regression model, the combination of GA < 37 weeks, neutropenia, and thrombocytopenia had four times greater odds of an infant experiencing mortality or severe illness related to E. coli sepsis. All three of these variables were individually associated with an increased risk of mortality or severe illness (Figure 1). GA and BW were co-linear variables, so BW was not included in the multivariate model. Note that sex was used as a variable to improve the overall degrees of freedom, which increased the r-squared statistic.

Platelet counts and absolute neutrophil counts (ANCs):

Platelet counts and ANCs were available on 64/68 infants with E. coli sepsis, out of which 26 infants were thrombocytopenic (<150,000/µL). Those who died or had a severe illness were significantly more likely to be thrombocytopenic as compared to those who did not (56% vs. 21%, p=0.003), with a lower median platelet count during the episode of sepsis (95 (68 – 164) vs. 248 (190 – 291), p<0.001, values in 1,000/µL).

Those who died or had a severe illness were also significantly more likely to be neutropenic (<1,500/µL) as compared to those who did not (39% vs. 12%, p=0.022), with a lower median (IQR) ANC during the episode of sepsis (1.9 (0.8 – 5.4) vs. 4.1 (2.8 – 7), p=0.022, values in 1,000/µL).

Empiric antibiotic resistance:

Antibiotic susceptibility data were available for 67 of the 68 E. coli isolates. Ampicillin resistance was 70% overall and similar when including only VLBW infants. Twenty-eight out of 38 (74%) E. coli EOS isolates and 19/29 (66%) LOS isolates were ampicillin-resistant. Three out of 38 (8%) E. coli EOS cases and 2/29 (7%) LOS cases were gentamicin resistant. Ampicillin and gentamicin resistance did not vary over time. There was no significant difference in our primary outcome between ampicillin-resistant and susceptible cases (25/47, 53.2% vs 9/20, 45.0%, p=0.43). Ampicillin resistance did not differ between VLBW infants and those with BW ≥ 1500 g (22/30, 73% vs. 25/37, 68%, p=0.91), EOS and LOS cases (28/38, 74% vs. 19/29, 66%, p=0.36), infants who met primary outcome vs. those who did not (25/34, 74% vs. 22/33, 67%, p=0.5), and those who died vs. those who survived (12/17, 71% vs 35/50, 70%, p=0.96).

E. coli neonatal sepsis accounts for a more significant proportion of cases since the widespread adoption of GBS IAP prophylaxis and subsequent reduction in GBS EOS cases.^{4, 17} This is reflected in our study, where the incidence rate of E. coli sepsis has increased almost three-fold from 2008–2012 to 2018–2022, despite a similar number of births throughout the entire period. We hypothesize this is predominantly attributable to a global increase in E. coli infection in the community, particularly of resistant strains.^{18, 19} This could lead to increased maternal colonization, leading to more EOS cases, and increased nosocomial transmission, leading to more LOS cases. Although the mortality from E. coli disease did not increase in our cohort during this period, the increase in the number of cases is still of concern as the median duration of antibiotic use in infants who survived after E. coli sepsis was 21 days. Recent epidemiologic studies have shown that every day of antibiotic exposure in neonates increases mortality and risk of bronchopulmonary dysplasia.^{20, 21} In addition, neonatal EOS and LOS may lead to long-term neurodevelopmental impairments in survivors, both in humans and in animal models of disease.^{22, 23, 24} This is one of the few studies to report the factors associated with the composite outcome of mortality or severe illness from E. coli sepsis.

In our univariate analysis, the outcomes of mortality, severe illness, or a combination of either were not associated with sex, maternal race, maternal IAI, maternal intrapartum antibiotic use, delivery type, timing of disease onset, and ampicillin resistance. GA, or BW as a surrogate of GA, thrombocytopenia, and neutropenia, were the only variables significantly different in infants who died or had severe illness, both in the univariate analysis and the multivariate logistic regression model. GA and BW are two variables that have been consistently associated with mortality in neonatal sepsis.^{25, 26} Thrombocytopenia is more common in Gram-negative and fungal neonatal sepsis than Gram-positive sepsis. It is associated with increased odds of mortality, especially in Gram-negative sepsis in VLBW neonates.^{27, 28} Thrombocytopenia in neonates has been associated with other morbidities, such as IVH and NEC.^{27, 29} Thrombocytopenia and neutropenia were both significantly associated with mortality or severe illness in our cohort, both in the univariate analysis and the multivariate logistic regression. Intubation and the need for mechanical ventilation have also often been associated with increased mortality in neonatal sepsis. However, we did not include this variable in our analysis as a majority of these were preterm neonates who were on mechanical ventilation due to their lung disease. We did not have information on the episode-related need for intubation or escalation of respiratory support.

Antibiotic resistance to E. coli strains has been increasing. Recent cohort studies from the NICHD NRN and the CDC point to a more than 80% resistance to ampicillin within E. coli strains isolated from neonates.^{4, 8} We found a similar rate of ampicillin resistance of 70% in our cohort, and this did not change over the 15 years we studied. This is in stark contrast to adult studies, where E. coli is the most commonly isolated bacterial pathogen in culture-positive community-onset sepsis, but the rates of ampicillin resistance are lower and in the 40–50% range.^{30, 31}

Antimicrobial resistance was also similar in VLBW and non-VLBW infants, and we did not find any relationship between ampicillin resistance and death, severe illness, or the composite outcome of either in our study. These data are reassuring, especially in the era of multi-drug resistant E. coli, drug shortages, and concern regarding the continued efficacy of ampicillin and gentamicin as first-line empiric antibiotic therapy for EOS in neonates.³² Neonates with antibiotic-resistant E. coli sepsis may also require prolonged hospitalization, which increases the risk of exposure to healthcare-associated infections, including other antibiotic-resistant pathogens. Antimicrobial resistance also leads to longer antibiotic duration and use of multiple antibiotics, which are independently associated with mortality and morbidity in neonates. Our data point toward the hypothesis that antimicrobial resistance and host-pathogen interaction are possibly exclusive of each other. Mortality and severity of illness are a function of bacterial pathogenicity and host susceptibility. Preterm neonates are the susceptible host in this context due to their immature developing immune system, which has a dysregulated immune response. Bacterial pathogenicity, which is one of the key drivers of the exaggerated immune response in preterm neonates, leads to endotoxin-mediated tissue damage and hypoxia, metabolic acidosis, cardiorespiratory compromise, organ dysfunction, and ultimately death. Our data lends credibility to the recent AAP consensus statement on the continued use of ampicillin and gentamicin for EOS and warrants caution when attributing causes of sepsis-related death to resistant bacteria.^{33, 34} There is a trend towards increased ampicillin resistance in the neonates who met the primary outcome; we cannot comment whether this would have reached statistical significance with a larger sample size.

A major strength of this study is the extended study period. The incidence of E. coli neonatal sepsis is relatively low at any major US NICU. Most studies include NRN, CDC, or VON datasets, which do not account for geographic variations in E. coli pathogenicity patterns. We also use a composite outcome of death or severe illness to identify infant and maternal risk factors. As neonatal care becomes better at reducing the outcome of death, there is an increased burden of morbidities in the survivors and it is critical to include this in studies on sepsis.^{35, 36} It is reassuring that ampicillin or gentamicin resistance was not associated with the primary outcome in our study. The rates of ampicillin and gentamicin resistance seen in our population are similar to the rates reported by the recent NRN study, and this rate was similar across EOS and LOS cases as well in VLBW infants who were at the highest risk of death or severe illness.⁴ Although this is from a single center, these data suggest that ampicillin and gentamicin remain effective as empiric antibiotics for neonatal sepsis.

The main drawback of this study is that this is a single-center experience; thus, the number of E. coli cases is insufficient to draw conclusions that can be applied to other centers. The laboratory system at our institution instituted new microbiological techniques with enhanced antimicrobial inactivation to improve the isolation of microorganisms in 2016. We acknowledge that this upgrade might partially account for the increased E. coli incidence in the last five years. We do not have follow-up data on the surviving infants, and hence we cannot speculate on their long-term neurodevelopmental outcomes. However, studies show that hypotension and the need for inotropes are associated with neurodevelopmental impairment (NDI) in preterm infants.^{20, 37}

This study adds to the current wealth of knowledge of E. coli sepsis in the neonatal population and the relationship between various host and pathogen factors contributing to poor outcomes.

Conflicts of interests:

The authors declare no conflicts of interest. This study was approved by the Institutional Review Board of University Hospitals, Cleveland, Ohio, USA (IRB #20221171) and a waiver of consent was granted. The study was performed in accordance with the Declaration of Helsinki.

Li J, Xiang L, Chen X, Li S, Sun Q, Cheng X, et al. Global, regional, and national burden of neonatal sepsis and other neonatal infections, 1990-2019: findings from the Global Burden of Disease Study 2019. Eur J Pediatr 2023, 182(5): 2335-2343.
Shane AL, Sanchez PJ, Stoll BJ. Neonatal sepsis. Lancet 2017, 390(10104): 1770-1780.
Cortese F, Scicchitano P, Gesualdo M, Filaninno A, De Giorgi E, Schettini F, et al. Early and Late Infections in Newborns: Where Do We Stand? A Review. Pediatr Neonatol 2016, 57(4): 265-273.
Stoll BJ, Puopolo KM, Hansen NI, Sanchez PJ, Bell EF, Carlo WA, et al. Early-Onset Neonatal Sepsis 2015 to 2017, the Rise of Escherichia coli, and the Need for Novel Prevention Strategies. JAMA Pediatr 2020, 174(7): e200593.
Baltimore RS, Huie SM, Meek JI, Schuchat A, O'Brien KL. Early-onset neonatal sepsis in the era of group B streptococcal prevention. Pediatrics 2001, 108(5): 1094-1098.
Hornik CP, Fort P, Clark RH, Watt K, Benjamin DK, Jr., Smith PB, et al. Early and late onset sepsis in very-low-birth-weight infants from a large group of neonatal intensive care units. Early Hum Dev 2012, 88 Suppl 2(Suppl 2): S69-74.
Lukacs SL, Schrag SJ. Clinical sepsis in neonates and young infants, United States, 1988-2006. J Pediatr 2012, 160(6): 960-965 e961.
Schrag SJ, Farley MM, Petit S, Reingold A, Weston EJ, Pondo T, et al. Epidemiology of Invasive Early-Onset Neonatal Sepsis, 2005 to 2014. Pediatrics 2016, 138(6).
Flannery DD, Edwards EM, Coggins SA, Horbar JD, Puopolo KM. Late-Onset Sepsis Among Very Preterm Infants. Pediatrics 2022, 150(6).
Greenberg RG, Kandefer S, Do BT, Smith PB, Stoll BJ, Bell EF, et al. Late-onset Sepsis in Extremely Premature Infants: 2000-2011. Pediatr Infect Dis J 2017, 36(8): 774-779.
Gray JW, Ubhi H, Milner P. Antimicrobial treatment of serious gram-negative infections in newborns. Curr Infect Dis Rep 2014, 16(2): 400.
Bizzarro MJ, Dembry LM, Baltimore RS, Gallagher PG. Changing patterns in neonatal Escherichia coli sepsis and ampicillin resistance in the era of intrapartum antibiotic prophylaxis. Pediatrics 2008, 121(4): 689-696.
Didier C, Streicher MP, Chognot D, Campagni R, Schnebelen A, Messer J, et al. Late-onset neonatal infections: incidences and pathogens in the era of antenatal antibiotics. Eur J Pediatr 2012, 171(4): 681-687.
Alarcon A, Pena P, Salas S, Sancha M, Omenaca F. Neonatal early onset Escherichia coli sepsis: trends in incidence and antimicrobial resistance in the era of intrapartum antimicrobial prophylaxis. Pediatr Infect Dis J 2004, 23(4): 295-299.
Flannery DD, Akinboyo IC, Mukhopadhyay S, Tribble AC, Song L, Chen F, et al. Antibiotic Susceptibility of Escherichia coli Among Infants Admitted to Neonatal Intensive Care Units Across the US From 2009 to 2017. JAMA Pediatr 2021, 175(2): 168-175.
Bergin SP, Thaden JT, Ericson JE, Cross H, Messina J, Clark RH, et al. Neonatal Escherichia coli Bloodstream Infections: Clinical Outcomes and Impact of Initial Antibiotic Therapy. Pediatr Infect Dis J 2015, 34(9): 933-936.
Miselli F, Cuoghi Costantini R, Creti R, Sforza F, Fanaro S, Ciccia M, et al. Escherichia coli Is Overtaking Group B Streptococcus in Early-Onset Neonatal Sepsis. Microorganisms 2022, 10(10).
Schlackow I, Stoesser N, Walker AS, Crook DW, Peto TE, Wyllie DH, et al. Increasing incidence of Escherichia coli bacteraemia is driven by an increase in antibiotic-resistant isolates: electronic database study in Oxfordshire 1999-2011. J Antimicrob Chemother 2012, 67(6): 1514-1524.
Vihta KD, Stoesser N, Llewelyn MJ, Quan TP, Davies T, Fawcett NJ, et al. Trends over time in Escherichia coli bloodstream infections, urinary tract infections, and antibiotic susceptibilities in Oxfordshire, UK, 1998-2016: a study of electronic health records. Lancet Infect Dis 2018, 18(10): 1138-1149.
Doucette SM, Kelly EN, Church PT, Lee S, Shah V, Canadian Neonatal Network I, et al. Association of inotrope use with neurodevelopmental outcomes in infants <29 weeks gestation: a retrospective cohort study. J Matern Fetal Neonatal Med 2022, 35(25): 6044-6052.
Flannery DD, Dysart K, Cook A, Greenspan J, Aghai ZH, Jensen EA. Association between early antibiotic exposure and bronchopulmonary dysplasia or death. J Perinatol 2018, 38(9): 1227-1234.
Lieblein-Boff JC, McKim DB, Shea DT, Wei P, Deng Z, Sawicki C, et al. Neonatal E. coli infection causes neuro-behavioral deficits associated with hypomyelination and neuronal sequestration of iron. J Neurosci 2013, 33(41): 16334-16345.
Mukhopadhyay S, Puopolo KM, Hansen NI, Lorch SA, DeMauro SB, Greenberg RG, et al. Impact of Early-Onset Sepsis and Antibiotic Use on Death or Survival with Neurodevelopmental Impairment at 2 Years of Age among Extremely Preterm Infants. J Pediatr 2020, 221: 39-46 e35.
Stoll BJ, Hansen NI, Adams-Chapman I, Fanaroff AA, Hintz SR, Vohr B, et al. Neurodevelopmental and growth impairment among extremely low-birth-weight infants with neonatal infection. JAMA 2004, 292(19): 2357-2365.
Lukacs SL, Schoendorf KC, Schuchat A. Trends in sepsis-related neonatal mortality in the United States, 1985-1998. Pediatr Infect Dis J 2004, 23(7): 599-603.
Thatrimontrichai A, Premprat N, Janjindamai W, Dissaneevate S, Maneenil G. Risk Factors for 30-Day Mortality in Neonatal Gram-Negative Bacilli Sepsis. Am J Perinatol 2020, 37(7): 689-694.
Ree IMC, Fustolo-Gunnink SF, Bekker V, Fijnvandraat KJ, Steggerda SJ, Lopriore E. Thrombocytopenia in neonatal sepsis: Incidence, severity and risk factors. PLoS One 2017, 12(10): e0185581.
Levit O, Bhandari V, Li FY, Shabanova V, Gallagher PG, Bizzarro MJ. Clinical and laboratory factors that predict death in very low birth weight infants presenting with late-onset sepsis. Pediatr Infect Dis J 2014, 33(2): 143-146.
Guida JD, Kunig AM, Leef KH, McKenzie SE, Paul DA. Platelet count and sepsis in very low birth weight neonates: is there an organism-specific response? Pediatrics 2003, 111(6 Pt 1): 1411-1415.
Rhee C, Kadri SS, Dekker JP, Danner RL, Chen HC, Fram D, et al. Prevalence of Antibiotic-Resistant Pathogens in Culture-Proven Sepsis and Outcomes Associated With Inadequate and Broad-Spectrum Empiric Antibiotic Use. JAMA Netw Open 2020, 3(4): e202899.
Daneman N, Fridman D, Johnstone J, Langford BJ, Lee SM, MacFadden DM, et al. Antimicrobial resistance and mortality following E. coli bacteremia. EClinicalMedicine 2023, 56: 101781.
Weissman SJ, Stoll B. Ampicillin and Gentamicin in Infants With Suspected Sepsis: Long Live Amp and Gent-But for How Long? JAMA Pediatr 2021, 175(2): 131-132.
Puopolo KM, Benitz WE, Zaoutis TE, Committee On F, Newborn, Committee On Infectious D. Management of Neonates Born at </=34 6/7 Weeks' Gestation With Suspected or Proven Early-Onset Bacterial Sepsis. Pediatrics 2018, 142(6): e20182896.
Puopolo KM, Benitz WE, Zaoutis TE, Committee On F, Newborn, Committee On Infectious D. Management of Neonates Born at >/=35 0/7 Weeks' Gestation With Suspected or Proven Early-Onset Bacterial Sepsis. Pediatrics 2018, 142(6): e20182894.
Stoll BJ, Hansen NI, Bell EF, Walsh MC, Carlo WA, Shankaran S, et al. Trends in Care Practices, Morbidity, and Mortality of Extremely Preterm Neonates, 1993-2012. JAMA 2015, 314(10): 1039-1051.
Horbar JD, Carpenter JH, Badger GJ, Kenny MJ, Soll RF, Morrow KA, et al. Mortality and neonatal morbidity among infants 501 to 1500 grams from 2000 to 2009. Pediatrics 2012, 129(6): 1019-1026.
Osborn DA, Evans N, Kluckow M, Bowen JR, Rieger I. Low superior vena cava flow and effect of inotropes on neurodevelopment to 3 years in preterm infants. Pediatrics 2007, 120(2): 372-380.

Table 1: Descriptive information on infants diagnosed with E. coli bloodstream or cerebrospinal fluid infection from 2008 – 2022. GA, gestational age; BW, birth weight; IQR, interquartile range; IAI, intra-amniotic infection.

Demographic/Clinical Characteristic
Median GA in weeks (IQR)	31 2/7 (27 4/7 – 38 0/7)
Median BW in grams (IQR)	1622 (970 – 2835)
Sex, N (%) Female Male	25 (36.8%) 43 (63.2%)
Race, N (%) Black White Other	37 (54.4%) 28 (41.2%) 3 (4.4%)
Delivery type, N (%) Vaginal Cesarean section	47 (69.1%) 21 (30.9%)
Maternal IAI	20/68 (29.4%)
Median hours of life at which blood culture was obtained (IQR)	31.5 (1.9 – 173.8)
Ampicillin resistance, N (%)	48/68 (70.6%)
Gentamicin resistance, N (%)	6/68 (8.8%)
Mortality	17/68 (25%)
Need for inotropic medication or severe metabolic acidosis	31/68 (45.6%)
Median days of antibiotics in those who survived (IQR)	21 (14 – 21)

Table 2: Characteristics of infants who died or had severe illness from episode of E. coli sepsis vs. those who survived till discharge without severe illness from E. coli sepsis. GA, gestational age; BW, birth weight; VLBW, very low birth weight; IQR, interquartile range; IAI, intra-amniotic infection; ANC, absolute neutrophil count; EOS, early onset sepsis. *1 infant had missing data for antibiotic susceptibility.

Characteristic	Died or had severe illness from E. coli sepsis (N=34)	Survived to discharge without severe illness from E. coli sepsis (N=34)	P value
Median GA (IQR)	28.5 (25.6, 32.5)	33.8 (30.0, 38.5)	0.023
Median BW (IQR)	1105 (775, 2165)	1975 (1350, 2889)	0.012
Number of VLBW neonates (%)	20 (59%)	11 (32%)	0.028
Sex, N (%) Male Female	20 (59%) 14 (41%)	22 (65%) 12 (35%)	0.52
Race, N (%) White Black Other	15 (44%) 18 (53%) 1 (2.9%)	13 (38%) 19 (56%) 2 (5.9%)	0.99
Delivery type, N (%) C-section Vaginal Unknown	11 (32%) 23 (68%)	9 (27%) 23 (70%) 1 (3%)
Presence of maternal IAI, N (%)	8 (25%)	11 (35%)	0.42
Ampicillin resistance, N (%)	25/34 (74%)	22/33* (67%)	0.51
Gentamicin resistance, N(%)	3/34 (8.8%)	2/33* (6.1%)	>0.9
Lowest ANC, median (IQR)	1.9 (0.8, 5.4)	4.1 (2.8, 7.0)	0.022
Lowest platelet count, median (IQR)	95 (68, 164)	248 (190, 291)	< 0.001
EOS (%)	59%	53%	0.62
Time of collection of blood culture, median hours (IQR)	32 (2,162)	34 (2,234)	0.91
Time to positivity of blood culture, median hours (IQR)	9.67 (8.24, 11.41)	9.59 (8.66, 10.90)	0.74
Thrombocytopenic, N (%)	19 (56%)	7 (21%)	0.003
Neutropenic, N (%)	12 (35%)	4 (12%)	0.023

Table 3: Characteristics of infants with E. coli early-onset sepsis (EOS) and late-onset sepsis (LOS). EOS was defined as positive blood or CSF culture within 72 hours of birth, and any positive cultures which were obtained after the 72-hour period were classified as LOS. GA, gestational age; BW, birth weight; VLBW, very low birth weight; IAI, intra-amniotic infection.

	EOS (N=38)	LOS (N=30)	P value
GA	29.6 (27.4,34.0)	32.0 (25.4, 38.3)	0.51
BW	1380 (1021, 2682)	1775 (791, 2889)	0.51
GA < 37 weeks	29/38 (76%)	19/30 (63.3%)	0.24
VLBW (%)	55%	33%	0.071
Sex Female	19/38 (50%)	24/30 (80%)	0.011
Race Black White Other	22 16 0	12 15 3	0.22
Delivery type C-Section	16/38 (42%)	4/30 (13%)	0.012
Maternal IAI Yes	18/38 (47%)	2/30 (6.7%)	< 0.001
Ampicillin resistance R	28/38 (74%)	19/30 (66%	0.53
Gentamicin resistance R	3/38 (7.9%)	2/30 (6.9%)	> 0.9
Mortality	10 (26%)	7 (23%)	0.84
Need for inotropic medication or severe metabolic acidosis	20 (53%)	11 (37%)	0.23

There is NO conflict of interest to disclose.

Predictors of Mortality and Severe Illness from Escherichia coli Sepsis in Neonates

Status:

Journal Publication

Version 1

Abstract

Figures

INTRODUCTION

METHODS

Statistical analysis:

RESULTS

DISCUSSION

Declarations

Conflicts of interests:

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1