medicina

Article
Predictors of Mortality in Early Neonatal Sepsis:
A Single-Center Experience
Marija Jovičić 1 , Miloš N. Milosavljević 2, * , Marko Folić 3,4 , Radiša Pavlović 3 and Slobodan M. Janković 2,4

1 Institute of Neonatology, 11000 Belgrade, Serbia; mexicomara@gmail.com

2 Department of Pharmacology and Toxicology, Faculty of Medical Sciences, University of Kragujevac,
34000 Kragujevac, Serbia; slobnera@gmail.com
3 Department of Pharmacy, Faculty of Medical Sciences, University of Kragujevac, 34000 Kragujevac, Serbia;
markof@medf.kg.ac.rs (M.F.); rpavlovic@medf.kg.ac.rs (R.P.)
4 Clinical Pharmacology Department, University Clinical Centre Kragujevac, 34000 Kragujevac, Serbia
* Correspondence: milosavljevicmilos91@gmail.com

Abstract: Background and Objectives: Early neonatal sepsis is associated with a significant mortality
rate despite modern treatment strategies. Our aim was to identify risk factors contributing to the
occurrence of death in newborns with early neonatal sepsis. Materials and Methods: We conducted a
retrospective cross-sectional study that included newborns with early sepsis who received care in the
intensive and semi-intensive care units at the Institute of Neonatology, Belgrade, Serbia. Newborns
with early neonatal sepsis who died comprised the case group, whereas those who survived made up
the control group. The diagnostic and therapeutic approach to the septic condition was carried out
independently of this study, according to valid hospital protocols and current good practice guidelines.
The influence of a large number of variables on the examined dichotomous outcome, as well as the
mutual interaction of potential predictor variables, was examined by binary logistic regression.
Results: The study included 133 pregnant women and 136 newborns with early neonatal sepsis,
of which 51 (37.5%) died, while the remaining 85 newborns (62.5%) survived. Newborns who died
had a statistically significantly lower birth weight compared to those who survived (882.8 ± 372.2 g
vs. 1660.9 ± 721.1 g, p = 0.000). Additionally, compared to newborns who survived, among the
deceased neonates there was a significantly higher proportion of extremely preterm newborns (74.5%
vs. 22.4%, p = 0.000). The following risk factors for the occurrence of death in early neonatal sepsis
Citation: Jovičić, M.; Milosavljević,
M.N.; Folić, M.; Pavlović, R.;
were identified: low birth weight, sepsis caused by gram-negative bacteria, and the use of double-
Janković, S.M. Predictors of Mortality inotropic therapy and erythrocyte transfusion during the first week. Conclusions: Pediatricians should
in Early Neonatal Sepsis: A pay special attention to infants with early neonatal sepsis in whom any of the identified risk factors
Single-Center Experience. Medicina are present in order to prevent a fatal outcome.
2023, 59, 604. https://doi.org/
10.3390/medicina59030604 Keywords: neonatal sepsis; newborn; neonatology; pediatricians; case-control studies; logistic models
Academic Editor: Johannes Mayr

Received: 24 February 2023

Revised: 10 March 2023 1. Introduction
Accepted: 15 March 2023
Early neonatal sepsis, which is commonly described as sepsis that occurs during the
Published: 18 March 2023
first 72 h after delivery, is still a significant factor in neonatal mortality [1]. The causative
agents of the infection are microorganisms that colonize the maternal genitourinary tract.
Group B Streptococcus (GBS) remains the most common cause of early neonatal sepsis
Copyright: © 2023 by the authors. apparent in term infants, while Escherichia coli is the most common cause of early sepsis in
Licensee MDPI, Basel, Switzerland. premature infants [1,2]. Some of the risk factors for early neonatal sepsis identified so far are:
This article is an open access article prematurity, low birth weight, congenital anomalies, low Apgar score, invasive procedures
distributed under the terms and during pregnancy, virulence of microorganisms, premature rupture of fetal membranes,
conditions of the Creative Commons a sibling with history of an invasive GBS infection, chorioamnionitis, elevated maternal
Attribution (CC BY) license (https:// body temperature, poor prenatal care, low socioeconomic status, abuse of psychoactive
creativecommons.org/licenses/by/ substances, and black race [2–7].
4.0/).

Medicina 2023, 59, 604. https://doi.org/10.3390/medicina59030604 https://www.mdpi.com/journal/medicina

Medicina 2023, 59, 604 2 of 11

The gold standard for the diagnosis of sepsis is a positive blood culture finding, where
care must be taken with the volume of blood drawn, as an insufficient amount of blood
drawn for blood culture cannot detect low levels of bacteremia [8]. The amount of blood
taken for a blood culture is particularly important for diagnosing early neonatal sepsis.
Thus, it is considered that the amount of blood in 1 milliliter provides a sensitivity of about
90% in diagnosing early neonatal sepsis [9]. The timely, prompt, and adequate administra-
tion of antibiotics in neonatal sepsis is essential. However, the prolonged use of antibiotics
in infants who are not suffering from sepsis delays the identification of other conditions that
are similar to sepsis, and is associated with the emergence of bacterial resistance, fungal
infections, late-onset sepsis, and necrotizing enterocolitis [10–13]. Hence, clinicians are con-
tinually searching for new markers that might provide high selectivity and specificity for
the diagnosis of early neonatal sepsis, such as inflammatory markers, cytokines, different
hematological parameters, and indicators for determining nutritional status [14–21].
Early neonatal sepsis is associated with a significant mortality rate of up to 54% [1],
despite modern treatment strategies. Therefore, our aim was to identify risk factors con-
tributing to the occurrence of death in newborns with early neonatal sepsis.

2. Materials and Methods

2.1. Study Design and Study Population
This study was a retrospective cross-sectional study. The study included newborns
with early sepsis who received care in the intensive and semi-intensive care units at the
Institute of Neonatology, Belgrade, Serbia, between 2013 and 2015. The study included two
groups of participants: a case group composed of newborns with early sepsis who died,
and a control group made up of newborns with sepsis, and other characteristics similar
those of the cases, who survived hospital care and were released as cured. The diagnostic
and therapeutic approach to the septic condition was carried out independently of this
study, according to valid hospital protocols and current good practice guidelines related to
the central topic of this research.
Newborns who met the following inclusion criteria participated in the study: (1) sepsis
identified within the first 72 h of life and (2) positive blood culture and/or cerebrospinal
fluid (CSF) results. The criteria for excluding the patient from the clinical trial were:
(1) the presence of major congenital anomalies; (2) incomplete medical documentation;
and (3) violation of the study protocol. The study used a so-called “convenient” sample,
and in order to reduce the influence of the researcher’s bias, the sample was consecutive,
i.e., it integrated all newborns who met the inclusion criteria and were treated in the period
from 1 January 2013 to 1 January 2015.

2.2. Variables Measured in the Study

The death outcome in newborns with early neonatal sepsis was determined by review-
ing their medical records. In the domain of maternal factors with a potential significant
influence on the outcome of interest, the following variables were analyzed: the method
of conception (natural or in vitro fertilization), method of delivery (vaginal or caesarean
section), singleton/multiple pregnancy, and the mother’s history of previous births, history
of previous miscarriages, prenatal conditions/diseases (chronic diseases, i.e., diseases and
conditions related to pregnancy), age, and available microbiological analyses (cervical
and vaginal smear findings, urine culture findings). The effect of the following newborn-
related variables was examined on the dependent variable: gender, body weight at birth,
gestational age at birth, Apgar score in the 1st minute, body temperature on admission
to Institute of Neonatology, verified microbiological cause of early neonatal sepsis with
antibiogram, biochemical and clinical parameters (average value of blood pH on the 1st
day, average value of base excess on the 1st day, average value of the fraction of inspired
oxygen during the 1st day, average value of systolic blood pressure on the 1st day, average
value of diastolic blood pressure on the 1st day, average value of mean blood pressure on
the 1st day, leukocytes on the 1st day, thrombocytes on the 1st day, hemoglobin on the
Medicina 2023, 59, 604 3 of 11

1st day, albumins on the 1st day, urea on the 3rd day, creatinin on the 3rd day, bilirubin
on the 2nd day, the presence of convulsions during the 1st week, the presence of pneu-
mothorax/pneumomediastinum/pulmonary interstitial emphysema during the 1st week),
conducted therapeutic measures (resuscitation measures, the use of ibuprofen for the treat-
ment of ductus arteriousus, the use of surfactant), the use of double-inotropic therapy,
the use of phototherapy, initial antibiotic therapy, protein intake during the 1st day, glucose
intake during the 1st day, fluid intake during the 1st day, erythrocyte transfusion during
the 1st week, thrombocyte transfusion during the 1st week, and plasma transfusion during
the 1st week.

2.3. Statistical Data Analysis

Statistical data analysis was performed using the SPSS software package, version 18
(SPSS Inc., Chicago, IL, USA). The methods of descriptive statistics were used to show
the basic characteristics of the newborns. Mean values ± standard deviation with range
(minimum–maximum) were used for continuous variables, and relative frequency was
used for categorical variables. The distribution of numerical data was tested for normality
using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Student’s t-test for independent
samples determined the existence of a statistically significant difference between the com-
pared groups in the values of continuous variables, provided that the distribution is normal;
otherwise, the nonparametric alternative Mann–Whitney U test was used. Comparisons
between categorical variables were made using the χ2 test (or Fisher’s exact probability
test for the low frequency of certain categories). If the probability of the null hypothesis
was lower than 5% (p = 0.05), the difference was considered statistically significant. The in-
fluence of a large number of variables on the examined dichotomous outcome, as well as
the mutual interaction of potential predictor variables, was examined by binary logistic
regression, and the results are presented in the form of a raw and adjusted odds ratio (OR)
with the associated 95% confidence interval.

3. Results
3.1. Characteristics of Pregnant Women
The study included 133 pregnant women with a mean (SD) age of 30.5 (5.7) years.
The basic clinical characteristics of the pregnant women are shown in Table 1.

Table 1. The basic clinical characteristics of the pregnant women.

Variable Mean ± Standard Deviation (Range) or Number (%)

Conception
Natural 114 (85.7%)
Assisted reproduction 19 (14.3%)
Number of fetuses
Singleton pregnancy 96 (72.2%)
Multiple pregnancy 37 (27.8%)
Number of previous births: 0/1/2/3/4/5 59 (42.7%)/51 (38.3%)/13 (9.8%)/5 (3.8%)/4 (3.0%)/1 (0.7%)
History of miscarriages 27 (20.3%)
Threatened premature labor 36 (27.1%)
Premature rupture of membranes 33 (24.8%)
Amniotic fluid pathology 26 (19.5%)
Acute and chronic diseases
Urogenital infection/chorioamnionitis 39 (28.3%)
Hypertension 24 (18.0%)
Diabetes mellitus 5 (3.8%)
Any smoking during pregnancy 12 (9.0%)

3.2. Characteristics of Newborns

In total, the study included 136 newborns with early neonatal sepsis (both twins in
three mothers developed early neonatal sepsis), of which 51 (37.5%) died (cases), while
Medicina 2023, 59, 604 4 of 11

the remaining 85 newborns (62.5%) survived (controls). The basic characteristics of the
newborns at birth are shown in Table 2. Compared to the control group, the newborns
who died had statistically significantly lower values of birth weight and Apgar score in the
first minute. Additionally, among the cases, there was a significantly higher proportion of
extremely and very preterm newborns compared to the control group.

Table 2. The basic characteristics of the newborns at birth.

Cases (n = 51) Controls (n = 85)

Variable Mean ± Standard Deviation Mean ± Standard Deviation Test Value and p Value
(Range) or Number (%) (Range) or Number (%)
Gender
Female 23 (45.1%) 34 (40.0%)
Male 28 (54.9%) 51 (60.0%) χ2 = 0.163, p = 0.686

Birth weight (g) 882.8 ± 372.2 (400–2850) 1660.9 ± 721.1 (650–3600) U = 507.500, p = 0.000 *
Time of delivery 1

Extremely preterm 38 (74.5%) 19 (22.4%)

Very preterm 12 (23.5%) 26 (30.6%)
Moderate preterm 0 (0%) 15 (17.6%)
χ2 = 43.368, p = 0.000 *
Late preterm 0 (0%) 18 (21.2%)
Early term 1 (2%) 5 (5.9%)
Full term 0 (0%) 2 (2.4%)
Delivery method
Vaginal 27 (53%) 38 (44.7%)
Caesarean section 24 (47.1%) 47 (55.3%) χ2 = 1.993, p = 0.369

Developmental level 2
SGA 7 (13.7%) 12 (14.1%)
AGA 43 (84.3%) 70 (82.4%) χ2 = 0.867, p = 0.285
LGA 1 (2.0%) 3 (3.5%)
Ponderal index 2.2 ± 0.4 (1.2–3.3) 2.2 ± 0.3 (1.3–3.0) U = 2109.500, p = 0.794
Apgar score at first minute
8–10 1 (2.0%) 21 (24.7%)
4–7 16 (31.4%) 41 (48.2%) χ2 = 24.288, p = 0.000 *
0–3 34 (66.7%) 23 (27.1%)
ABO blood group type
A 23 (46.0%) 34 (40.0%)
B 8 (16.0%) 11 (12.9%)
χ2 = 1.383, p = 0.710
AB 4 (8.0%) 6 (7.1%)
O 15 (30.0%) 34 (40.0%)
Rh blood group type
Rh− 8 (15.7%) 11 (12.9%)
Rh+ 43 (84.3%) 74 (87.1%) χ2 = 0.042, p = 0.622

* Statistically significant. 1 Extremely preterm—<28 weeks; very preterm—28 0/7 to 31 6/7 weeks; moderate
preterm—32 0/7 to 33 6/7 weeks; late preterm—34 0/7 to 36 6/7 weeks; early term—37 0/7 to 38 6/7 weeks;
full term—39 0/7 to 40 6/7 weeks. 2 SGA—small for gestational age; AGA—appropriate for gestational age;
LGA—large for gestational age.

All newborns included in this study were diagnosed with early neonatal sepsis and all
had a positive blood culture. The most important biochemical and clinical characteristics
of the newborns by group are shown in Table 3, while the presentation of the applied
therapeutic measures is given in Table 4.
Medicina 2023, 59, 604 5 of 11

Table 3. The biochemical and clinical characteristics of the newborns.

Cases (n = 51) Controls (n = 85)

Variable Mean ± Standard Deviation Mean ± Standard Deviation Test Value and p Value
(Range) or Number (%) (Range) or Number (%)
Cause of sepsis
Gram-negative bacteria 1 47 (92.5%) 45 (52.9%)
2
χ2 = 20.642, p = 0.000 *
Gram-positive bacteria 4 (7.8%) 40 (47.1%)
Average value of blood pH on the
7.26 ± 0.11 (6.86–7.48) 7.32 ± 0.08 (7.07–7.46) U = 1407.500, p = 0.001 *
1st day
Average value of base excess on the
−6.35 ± 3.78 (−21.2–2.67) −4.64 ± 2.73 (−12.0–1.57) U = 1494.500, p = 0.002 *
1st day (mmol/L)
Body temperature on admission
35.52 ± 0.80 (32.90–36.90) 36.15 ± 0.54 (34.70–37.30) U = 1035.500, p = 0.000 *
(◦ C)
Average value of FiO2 3 during the
49.7 ± 15.6 (27.0–100.0) 44.6 ± 13.8 (21.0–85.0) U = 1745.500, p = 0.058
1st day
Average value of systolic blood
51.84 ± 10.87 (26.00–83.00) 59.06 ± 11.71 (42.00–92.00) U = 1335.000, p = 0.000 *
pressure on the 1st day (mmHg)
Average value of diastolic blood
25.61 ± 9.46 (8.00–59.00) 31.24 ± 8.68 (16.00–60.00) U = 1278.500, p = 0.000 *
pressure on the 1st day (mmHg)
Average value of mean blood
36.87 ± 9.65 (15.00–65.00) 43.80 ± 9.91 (27.00–81.00) U = 1222.500, p = 0.000 *
pressure on the 1st day (mmHg)
Leukocytes on the 1st day (×109 /L) 19.36 ± 10.96 (4.30–68.90) 20.42 ± 13.56 (3.60–94.60) U = 2107.500, p = 0.787
Thrombocytes on the 1st day
181.78 ± 59.74 (49.00–310.00) 209.05 ± 61.45 (88.00–348.00) U = 1694.500, p = 0.021 *
(×109 /L)
Hemoglobin on the 1st day (g/L) 169.86 ± 29.80 (88.00–221.00) 182.33 ± 28.81 (86.00–250.00) U = 1684.000, p = 0.030 *
Albumin on the 1st day (g/L) 25.82 ± 5.03 (12.00–42.00) 30.51 ± 5.08 (19.00–42.00) U = 1026.500, p = 0.000 *
Urea on the 3rd day (mmol/L) 9.73 ± 2.49 (5.20–16.80) 6.43 ± 3.61 (1.20–18.9) U = 668.500, p = 0.000 *
Creatinin on the 3rd day (mmol/L) 95.41 ± 25.91 (49.00–166.00) 74.89 ± 21.63 (23.00–161.00) U = 842.000, p = 0.000 *
Bilirubin on the 2nd day (µmol/L) 76.91 ± 28.29 (17.00–139.00) 88.97 ± 34.51 (27.00–229.00) U = 1764.000, p = 0.100
Convulsions during 1st week
Yes 33 (64.7%) 18 (21.2%)
χ2 = 23.946, p = 0.000 *
No 18 (35.35) 67 (78.8%)
Pneumothorax/Pneumomediastinum/
Pulmonary interstitial emphysema
Yes 10 (19.6%) 7 (8.2%)
χ2 = 3.132, p = 0.077
No 41 (80.4%) 78 (91.8%)
* Statistically significant. 1 Klebsiella, Acinetobacter, Pseudomonas, Escherichia coli, Serratia, Citrobacter. 2 Staphylococcus,
Enterococcus. 3 The fraction of inspired oxygen.

Table 4. Conducted therapeutic measures in newborns.

Cases (n = 51) Controls (n = 85)

Variable Mean ± Standard Deviation Mean ± Standard Deviation Test Value and p Value
(Range) or Number (%) (Range) or Number (%)
Resuscitation measures
Yes 40 (78.4%) 47 (55.3%)
χ2 = 17.047, p = 0.001 *
No 11 (21.6%) 38 (44.7%)
Medicina 2023, 59, 604 6 of 11

Table 4. Cont.

Cases (n = 51) Controls (n = 85)

Variable Mean ± Standard Deviation Mean ± Standard Deviation Test Value and p Value
(Range) or Number (%) (Range) or Number (%)
Ibuprofen for the treatment of
ductus arteriousus
Yes 16 (31.4%) 19 (22.4%)
χ2 = 0.926, p = 0.336
No 35 (68.6%) 66 (77.6%)
Surfactant
Yes 22 (43.1%) 20 (23.5%)
χ2 = 4.859, p = 0.027 *
No 29 (56.9%) 65 (76.5%)
Double-inotropic therapy
Yes 41 (80.4%) 17 (20.0%)
χ2 = 45.093, p = 0.000 *
No 10 (19.6%) 68 (80.0%)
Phototherapy
Yes 41 (80.4%) 76 (89.4%)
χ2 = 1.405, p = 0.236
No 10 (19.6%) 9 (10.6%)
Initial antibiotic therapy
Ampicillin + Gentamycin 11 (21.6%) 27 (31.8%)
Ampicillin + Amikacin 39 (76.5%) 55 (64.7%) χ2 = 2.091, p = 0.352

Ampicillin + Meropenem 1 (2.0%) 3 (3.5%)

Protein intake during the 1st
0.5 ± 0.6 (0–3.3) 0.4 ± 0.5 (0–2.0) U = 1871.000, p = 0.160
day (g/kg)
Glucose intake during the 1st
5.1 ± 1.3 (1.0–8.7) 5.0 ± 1.2 (0.8–8.4) U = 2148, p = 0.930
day (mg/kg/min)
Fluid intake during the 1st
99.0 ± 17.4 (58.0–156.0) 85.7 ± 17.1 (13.0–139.0) U = 1142.500, p = 0.000 *
day (mL/kg)
Erythrocyte transfusion
during the 1st week
Yes 47 (92.2%) 22 (25.9%)
χ2 = 53.394, p = 0.000 *
No 4 (7.8%) 63 (74.1%)
Thrombocyte transfusion
during the 1st week
Yes 29 (56.9%) 10 (11.8%)
χ2 = 29.530, p = 0.000 *
No 22 (43.1%) 75 (88.2%)
Plasma transfusion during the
1st week
Yes 40 (78.4%) 33 (38.8%)
χ2 = 18.549, p = 0.000 *
No 11 (21.6%) 52 (61.2%)
* Statistically significant.

3.3. Risk Factors for Fatal Outcome in Early Neonatal Sepsis

The results of both univariate and multivariate binary logistic regression from the last
step with satisfactory goodness of fit (Cox and Snell R square 0.546, Nagelkerke R2 0.746,
Hosmer–Lemeshow Chi-square 9.268, df = 8, p = 0.320, overall model accuracy of 89.6%)
with adjustment for potential confounders are shown in Table 5. The variables entered in
the multivariate analysis were: birth weight, average value of blood pH on the first day,
Medicina 2023, 59, 604 7 of 11

body temperature on admission, double-inotropic therapy, convulsions during the first

week, erythrocyte transfusion during the first week, plasma transfusion during the first
week, and blood culture. After adjustment for potential confounders and other independent
variables, the following risk factors for the occurrence of death in early neonatal sepsis
were identified: low birth weight, sepsis caused by gram-negative bacteria (GNB), and the
use of double-inotropic therapy and erythrocyte transfusion during the first week.

Table 5. Crude and adjusted odds ratios (OR) of the risk factors for fatal outcome in early neona-
tal sepsis.

Univariate Model Multivariate Model

Risk Factors Crude OR with 95% CI Adjusted # OR with 95% CI
p p
0.996 (0.995–0.998) 0.998 (0.996–1.000)
Birth weight
p = 0.000 * p = 0.046 *
0.001 (0.000–0.067) 0.010 (0.000–3.334)
Average value of blood pH on the 1st day
p = 0.001 * p = 0.121
0.212 (0.107–0.421) 0.641 (0.265–1.552)
Body temperature on admission
p = 0.000 * p = 0.324
16.400 (6.857–39.222) 9.186 (2.451–34.432)
Double-inotropic therapy
p = 0.000 * p = 0.001 *
6.824 (3.144–14.812) 2.126 (0.621–7.283)
Convulsions during 1st week
p = 0.000 * p = 0.230
33.648 (10.866–104.198) 5.279 (1.147–24.290)
Erythrocyte transfusion during the 1st week
p = 0.000 * p = 0.033 *
5.730 (2.582–12.717) 1.290 (0.320–5.198)
Plasma transfusion during the 1st week
p = 0.000 * p = 0.721
10.444 (3.455–31.569) 7.071 (1.147–43.584)
Blood culture
p = 0.000 * p = 0.035 *
p: statistical significance; CI: confidence interval; *: statistically significant; #: adjusted for average value of blood
pH on the first day, body temperature on admission, convulsions during first week, and plasma transfusion
during the first week.

4. Discussion
Sepsis in the pediatric population has been extensively studied in recent years. With more
than 1.2 million cases per year [22] and 25% of fatal outcomes worldwide [23], it remains
among the leading causes of childhood mortality. It is important to acknowledge that
84% of infant deaths due to sepsis are preventable [24], and the understanding of its risk
profile may contribute to the development of risk-prediction models aimed at improving
the overall survival of this group of patients.
The main objective of the present study was to identify potential predictors for the fatal
outcome of sepsis in newborns. Our results showed birth weight to be a protective factor.
The incidence of a fatal outcome increases as birth weight decreases. In a systematic review
with a meta-analysis of 240 studies, the highest case fatality rate of 24% was observed in
the group of patients with very low birth weight (<1500 g), while the incidence rate was the
lowest, at 15%, in infants with a birth weight ≥2500 g [24]. The average birth weight in our
case group was 882.8 ± 372.2, indicating the possibility of a high mortality rate. In addition
to the incidence rate, one of the studies analyzed in the systematic review noted a more than
sixfold-increased risk of a fatal outcome in newborns with a low birth weight [25]. Another
major risk factor in the same study was the time of delivery, showing a strong association
between prematurity and severe sepsis onset. Newborns who are prematurely born have a
relatively immature immune system, are more susceptible to infections, and are at higher
risk for mortality [26]. Given that 93 percent of all of our patients were preterm and that we
had a higher number of extremely preterm and very preterm newborns compared to the
Medicina 2023, 59, 604 8 of 11

control group, the time of delivery could also be associated with a fatal outcome alongside
birth weight, irrespective of the lack of significance in our results. A similar explanation
was proposed in a prospective cohort study of 172 newborns with sepsis. The newborns
with extremely low birth weight, defined as a weight of 1000 g, had the highest mortality
rate (65.7%) compared to other birth-weight categories of patients [27]. This population
born before the 29th week of gestation also had a higher risk of the development of other
conditions already proven to be significantly associated with a higher mortality rate [28],
which could contribute to the increased frequency of fatal outcomes in our case group of
patients. According to our results, a 1 g increase in birth weight would decrease sepsis-
related death risk by 0.2%.
Recently, the requirement for inotropic support was assessed to predict the mortality
rate in newborns with sepsis. In a study population similar to ours, weighing under 1500 g
and born before 32 weeks of gestation, the mortality risk was shown to be 22 times higher
in a group of newborns requiring inotropic support [29]. The pooled OR for the need for
vasoactive agents obtained from 20 studies analyzed in a recently published meta-analysis
was 6.5 [23]. It has been reported that newborns with sepsis who are at high risk of death
develop multiple cardiovascular, respiratory, immune, and renal failures [30], and that the
need for inotropic drugs to support the cardiovascular system is significantly increased
in this group of patients [31]. The development of such multiorgan failure can be utilized
to predict newborn death rates. The Neonatal Sequential Organ Failure Assessment Score
(nSOFA) is a recently introduced tool for the evaluation of the presence and progression
of specific signs of multiorgan failure. One of its determinants for the evaluation of the
cardiovascular system is, precisely, the utilization of two or more inotropes. The mortality
rate was significantly higher in the group of newborns with an nSOFA score >4 [31].
Our results also support the association between double-inotropic treatment and the risk
of a fatal outcome, suggesting the importance of the timely recognition of multiorgan
dysfunction to reduce sepsis mortality. The nSOFA scoring system or some other tool for
organ failure assessment is needed in routine practice.
The progression to dysfunction of different organs due to sepsis, lower oxygen delivery,
and a rapid decline in circulating red blood cells (RBC) and hemoglobin are some of the
indications for transfusions in preterm newborns to prevent shock and a fatal outcome [32].
More than 90% of preterm infants are likely to develop anemia of prematurity [33], which
is significantly associated with a fatal outcome if sepsis is present [27]. Furthermore, very
low-birth-weight preterm infants with sepsis were shown to be 3.22 times more likely to
receive RBC transfusions [34]. Nearly 90% of newborns who weigh less than 1000 g may
require an average of five RBC transfusions while they are in the hospital [35]. In addition to
sepsis, our patients in the case group had considerably lower hemoglobin levels than those
in the control group, which may have been the main factor in the need for RBC transfusions.
However, although with the aim to prevent the poor outcomes of sepsis, erythrocyte
transfusion is also associated with the development of life-threatening conditions including
necrotizing enterocolitis, bronchopulmonary dysplasia, and intraventricular hemorrhage
that could diminish the benefits of its utilization [33]. The same authors found that the
mortality rate among preterm newborns receiving erythrocytes was 50% higher during
the first 28 days of life. The results of another study showed an association between a
fatal outcome after 28 days of life and the number of transfusions. When comparing
newborns who received three or more erythrocyte transfusions to those who received
fewer than three, the relative risk of death was 89% higher [34]. It is possible that our
patients developed some of the aforementioned conditions, which increased the rate of
fatal outcomes compared to the control group. The other reasons for adverse conditions
after erythrocyte transfusion may be related to the donor, the time of the erythrocyte
bioactive substance accumulation, or artificial additives [33]. However, there are studies
that reported no difference in the incidence of mortality or multiple-organ dysfunction
in newborns who received fresh blood cells compared to those who received the oldest
blood units available [32]. Given the inconsistency, future studies with sufficient power
Medicina 2023, 59, 604 9 of 11

are needed to confirm the association between erythrocyte transfusion and poor outcomes
from sepsis in newborns.
In our study, the most common cause of sepsis was GNB, with Klebsiella, Acinetobacter,
Pseudomonas, Escherichia coli, Serratia, and Citrobacter being most frequently identified.
This conclusion is in line with the research-based data. Among hospitalized neonates who
are more likely to experience septic shock and die, the rate of GNB sepsis rises. In this
case-control study analyzing patients’ data over 27 years, the frequency of GNB sepsis
was 66%, and a fatal outcome was observed in 33% of newborns compared to 20% in
newborns with non-GNB sepsis. The main reasons for increased death in the case group
were a higher risk of septic shock and a lack of adequate medication [36]. Similarly, GNB
sepsis ended fatally in 67.7% (Acinetobacter), 58.2% (Klebsiella), and 58.3% (Pseudomonas)
of patients, with multiple antibiotic resistance as the only confirmed predictor of fatal
outcome [37]. The main contributors to a fatal outcome in newborns with Acinetobacter
sepsis were septic shock, with odds of 41.38, and inadequate antibiotic treatment, showing
a 10-times-higher risk [38]. According to a recent systematic review, antibiotic resistance
caused by insufficient empiric care is a serious concern in the rising prevalence of GNB
sepsis in infants [39]. It is possible that empiric antibiotic treatment, multidrug resistance,
and the development of septic shock in our population are the reasons for the sevenfold
higher likelihood of a fatal outcome in newborns with GNB sepsis.
Our analysis was limited by the data being collected from a single institution, and the
ratio between the groups. Nevertheless, we managed to include a total of 136 newborns
over a period of 3 years, a ratio between groups of almost 1:2, which we consider sufficient
data for analysis and valid conclusions. The absence of data on antibiotic resistance, which
could cast additional light on the connection between GNB sepsis and mortality rates,
is another shortcoming.

5. Conclusions
The fatal outcome of sepsis in newborns was associated with birth weight, double-
inotropic treatment, gram-negative bacteria, and erythrocyte transfusion. Our results also
revealed the importance of additional tools for timely diagnosis, organ failure assessment,
and the appropriate guidance for treatment options that may decrease multidrug resistance
and the mortality rate of sepsis in newborns.

Author Contributions: M.J., M.F. and S.M.J. designed and planned the study. M.J. was responsible
for data acquisition. M.N.M. and S.M.J. performed statistical data analysis. M.J., M.N.M. and R.P.
wrote the manuscript. M.F. edited the manuscript. All authors have read and agreed to the published
version of the manuscript.
Funding: This study was partially financed by grant No 175007 from the Serbian Ministry of Educa-
tion, Science and Technological Development.
Institutional Review Board Statement: The ethical approval was received from the Ethical Commit-
tee of the Institute of Neonatology, Belgrade, Serbia (No. 986/3, data of approval 7 May 2019).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the
study. Written informed consent was obtained from the patient(s) to publish this paper.
Data Availability Statement: Data are available upon request from corresponding authors.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Shane, A.L.; Sánchez, P.J.; Stoll, B.J. Neonatal sepsis. Lancet 2017, 390, 1770–1780. [CrossRef] [PubMed]
2. Simonsen, K.A.; Anderson-Berry, A.L.; Delair, S.F.; Davies, H.D. Early-onset neonatal sepsis. Clin. Microbiol. Rev. 2014, 27, 21–47.
[CrossRef] [PubMed]
3. Adair, C.E.; Kowalsky, L.; Quon, H.; Ma, D.; Stoffman, J.; McGeer, A.; Robertson, S.; Mucenski, M.; Davies, H.D. Risk factors
for early-onset group B streptococcal disease in neonates: A population-based case-control study. CMAJ 2003, 169, 198–203.
[PubMed]
Medicina 2023, 59, 604 10 of 11

4. Møller, M.; Thomsen, A.C.; Borch, K.; Dinesen, K.; Zdravkovic, M. Rupture of fetal membranes and premature delivery associated
with group B streptococci in urine of pregnant women. Lancet 1984, 2, 69–70. [CrossRef]
5. Faxelius, G.; Bremme, K.; Kvist-Christensen, K.; Christensen, P.; Ringertz, S. Neonatal septicemia due to group B streptococci–
perinatal risk factors and outcome of subsequent pregnancies. J. Perinat. Med. 1988, 16, 423–430. [CrossRef] [PubMed]
6. Herbst, A.; Källén, K. Time between membrane rupture and delivery and septicemia in term neonates. Obstet. Gynecol. 2007, 110,
612–618. [CrossRef]
7. Stoll, B.J.; Hansen, N.I.; Sánchez, P.J.; Faix, R.G.; Poindexter, B.B.; Van Meurs, K.P.; Bizzarro, M.J.; Goldberg, R.N.; Frantz, I.D., III;
Hale, E.C.; et al. Early onset neonatal sepsis: The burden of group B Streptococcal and E. coli disease continues. Pediatrics 2011,
127, 817–826. [CrossRef]
8. Schelonka, R.L.; Chai, M.K.; Yoder, B.A.; Hensley, D.; Brockett, R.M.; Ascher, D.P. Volume of blood required to detect common
neonatal pathogens. J. Pediatr. 1996, 129, 275–278. [CrossRef]
9. Nizet, V.; Klein, J.O. Bacterial sepsis and meningitis. In Infectious Diseases of the Fetus and Newborn Infant, 7th ed.; Remington, J.S.,
Klein, J.O., Wilson, C.B., Nizet, V., Maldonado, Y.A., Eds.; Elsevier: Philadelphia, PA, USA, 2011; pp. 222–275.
10. Kuppala, V.S.; Meinzen-Derr, J.; Morrow, A.L.; Schibler, K.R. Prolonged initial empirical antibiotic treatment is associated with
adverse outcomes in premature infants. J. Pediatr. 2011, 159, 720–725. [CrossRef]
11. Weese-Mayer, D.E.; Fondriest, D.W.; Brouillette, R.T.; Shulman, S.T. Risk factors associated with candidemia in the neonatal
intensive care unit: A case-control study. Pediatr. Infect. Dis. J. 1987, 6, 190–196. [CrossRef]
12. Saiman, L.; Ludington, E.; Pfaller, M.; Rangel-Frausto, S.; Wiblin, R.; Dawson, J.; Blumberg, H.M.; Patterson, J.E.; Rinaldi, M.;
Edwards, J.E.; et al. Risk factors for candidemia in Neonatal Intensive Care Unit patients. The National Epidemiology of Mycosis
Survey study group. Pediatr. Infect. Dis. J. 2000, 19, 319–324. [CrossRef] [PubMed]
13. Piantino, J.H.; Schreiber, M.D.; Alexander, K.; Hageman, J. Culture Negative Sepsis and Systemic Inflammatory Response
Syndrome in Neonates. Neuroreviews 2013, 14, e294–e305. [CrossRef]
14. Weirich, E.; Rabin, R.L.; Maldonado, Y.; Benitz, W.; Modler, S.; Herzenberg, L.A. Neutrophil CD11b expression as a diagnostic
marker for early-onset neonatal infection. J. Pediatr. 1998, 132, 445–451. [CrossRef] [PubMed]
15. Du, J.; Li, L.; Dou, Y.; Li, P.; Chen, R.; Liu, H. Diagnostic utility of neutrophil CD64 as a marker for early-onset sepsis in preterm
neonates. PLoS ONE 2014, 9, e102647. [CrossRef] [PubMed]
16. Genel, F.; Atlihan, F.; Gulez, N.; Kazanci, E.; Vergin, C.; Terek, D.T.; Yurdun, O.C. Evaluation of adhesion molecules CD64, CD11b
and CD62L in neutrophils and monocytes of peripheral blood for early diagnosis of neonatal infection. World J. Pediatr. 2012,
8, 72–75. [CrossRef]
17. Küster, H.; Weiss, M.; Willeitner, A.E.; Detlefsen, S.; Jeremias, I.; Zbojan, J.; Geiger, R.; Lipowsky, G.; Simbruner, G. Interleukin-
1 receptor antagonist and interleukin-6 for early diagnosis of neonatal sepsis 2 days before clinical manifestation. Lancet 1998, 352,
1271–1277. [CrossRef]
18. Hornik, C.P.; Benjamin, D.K.; Becker, K.C.; Li, J.; Clark, R.; Cohen-Wolkowiez, M.; Smith, P.B. Use of the complete blood cell count
in early-onset neonatal sepsis. Pediatr. Infect. Dis. J. 2012, 31, 799–802. [CrossRef]
19. Chitra, W.; Lubis, B.M.; Siregar, O.R.; Hasibuan, B.S.; Lubis, I.N.D.; Nafianti, S. Role of Modified Hematologic Scoring System and
Platelet Indices in Diagnosing Neonatal Sepsis. Perinatology 2022, 23, 22–28.
20. Li, T.; Li, X.; Wei, Y.; Dong, G.; Yang, J.; Yang, J.; Fang, P.; Qi, M. Predictive Value of C-Reactive Protein-to-Albumin Ratio for
Neonatal Sepsis. J. Inflamm. Res. 2021, 14, 3207–3215. [CrossRef]
21. Li, T.; Qi, M.; Dong, G.; Li, X.; Xu, Z.; Wei, Y.; Feng, Y.; Ren, C.; Wang, Y.; Yang, J. Clinical Value of Prognostic Nutritional Index in
Prediction of the Presence and Severity of Neonatal Sepsis. J. Inflamm. Res. 2021, 14, 7181–7190. [CrossRef]
22. Weiss, S.L.; Peters, M.J.; Alhazzani, W.; Agus, M.S.D.; Flori, H.R.; Inwald, D.P.; Nadel, S.; Schlapbach, L.J.; Tasker, R.C.; Argent,
A.C.; et al. Surviving Sepsis Campaign International Guidelines for the Management of Septic Shock and Sepsis-Associated
Organ Dysfunction in Children. Pediatr. Crit. Care Med. 2020, 21, e52–e106. [CrossRef] [PubMed]
23. Menon, K.; Schlapbach, L.J.; Akech, S.; Argent, A.; Biban, P.; Carrol, E.D.; Chiotos, K.; Chisti, M.J.; Evans, I.V.R.; Inwald, D.P.; et al.
Pediatric Sepsis Definition Taskforce of the Society of Critical Care Medicine. Criteria for Pediatric Sepsis-A Systematic Review
and Meta-Analysis by the Pediatric Sepsis Definition Taskforce. Crit. Care Med. 2022, 50, 21–36. [CrossRef] [PubMed]
24. Gan, M.Y.; Lee, W.L.; Yap, B.J.; Seethor, S.T.; Greenberg, R.G.; Pek, J.H.; Tan, B.; Hornik, C.P.; Lee, J.H.; Chong, S.L. Contemporary
Trends in Global Mortality of Sepsis Among Young Infants Less Than 90 Days: A Systematic Review and Meta-Analysis. Front.
Pediatr. 2022, 10, 890767. [CrossRef] [PubMed]
25. Liang, L.; Kotadia, N.; English, L.; Kissoon, N.; Ansermino, J.M.; Kabakyenga, J.; Lavoie, P.M.; Wiens, M.O. Predictors of Mortality
in Neonates and Infants Hospitalized With Sepsis or Serious Infections in Developing Countries: A Systematic Review. Front.
Pediatr. 2018, 6, 277. [CrossRef]
26. Kumar, S.K.; Bhat, B.V. Distinct mechanisms of the newborn innate immunity. Immunol. Lett. 2016, 173, 42–54. [CrossRef]
27. Kolesnichenko, S.I.; Kadyrova, I.A.; Lavrinenko, A.V.; Zhumadilova, Z.A.; Avdienko, O.V.; Vinogradskaya, Y.V.; Fominykh, Y.A.;
Panibratec, L.G.; Akhmaltdinova, L.L. Mortality Risk Factors of Early Neonatal Sepsis During COVID-19 Pandemic. Infect. Drug
Resist. 2022, 15, 6307–6316. [CrossRef]
28. Stoll, B.J.; Hansen, N.; Fanaroff, A.A.; Wright, L.L.; Carlo, W.A.; Ehrenkranz, R.A.; Lemons, J.A.; Donovan, E.F.; Stark, A.R.; Tyson,
J.E.; et al. Changes in pathogens causing early-onset sepsis in very-low-birth-weight infants. N. Engl. J. Med. 2002, 347, 240–247.
[CrossRef]
Medicina 2023, 59, 604 11 of 11

29. Goh, G.L.; Lim, C.S.E.; Sultana, R.; De La Puerta, R.; Rajadurai, V.S.; Yeo, K.T. Risk Factors for Mortality From Late-Onset Sepsis
Among Preterm Very-Low-Birthweight Infants: A Single-Center Cohort Study From Singapore. Front. Pediatr. 2022, 9, 801955.
[CrossRef]
30. Wynn, J.L.; Kelly, M.S.; Benjamin, D.K.; Clark, R.H.; Greenberg, R.; Benjamin, D.K., Jr.; Smith, P.B. Timing of Multiorgan
Dysfunction among Hospitalized Infants with Fatal Fulminant Sepsis. Am. J. Perinatol. 2017, 34, 633–639. [CrossRef]
31. Wynn, J.L.; Polin, R.A. A neonatal sequential organ failure assessment score predicts mortality to late-onset sepsis in preterm
very low birth weight infants. Pediatr. Res. 2020, 88, 85–90. [CrossRef]
32. Spinella, P.C.; Tucci, M.; Fergusson, D.A.; Lacroix, J.; Hébert, P.C.; Leteurtre, S.; Schechtman, K.B.; Doctor, A.; Berg, R.A.;
Bockelmann, T.; et al. Effect of Fresh vs Standard-issue Red Blood Cell Transfusions on Multiple Organ Dysfunction Syndrome in
Critically Ill Pediatric Patients: A Randomized Clinical Trial. JAMA 2019, 322, 2179–2190. [CrossRef]
33. Crawford, T.M.; Andersen, C.C.; Hodyl, N.A.; Robertson, S.A.; Stark, M.J. The contribution of red blood cell transfusion to
neonatal morbidity and mortality. J. Paediatr. Child Health 2019, 55, 387–392. [CrossRef]
34. dos Santos, A.M.; Guinsburg, R.; de Almeida, M.F.; Procianoy, R.S.; Marba, S.T.; Ferri, W.A.; Rugolo, L.M.; Lopes, J.M.; Moreira,
M.E.; Luz, J.H.; et al. Factors associated with red blood cell transfusions in very-low-birth-weight preterm infants in Brazilian
neonatal units. BMC Pediatr. 2015, 15, 113. [CrossRef] [PubMed]
35. Ali, N. Red blood cell transfusion in infants and children—Current perspectives. Pediatr. Neonatol. 2018, 59, 227–230. [CrossRef]
[PubMed]
36. 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, 689–694. [CrossRef] [PubMed]
37. Peters, L.; Olson, L.; Khu, D.T.K.; Linnros, S.; Le, N.K.; Hanberger, H.; Hoang, N.T.B.; Tran, D.M.; Larsson, M. Multiple antibiotic
resistance as a risk factor for mortality and prolonged hospital stay: A cohort study among neonatal intensive care patients with
hospital-acquired infections caused by gram-negative bacteria in Vietnam. PLoS ONE 2019, 14, e0215666. [CrossRef] [PubMed]
38. Thatrimontrichai, A.; Tonjit, P.; Janjindamai, W.; Dissaneevate, S.; Maneenil, G.; Phatigomet, M. Risk Factors Associated With
30-Day Mortality Among Neonates With A. baumannii Sepsis. Pediatr. Infect. Dis. J. 2021, 40, 1111–1114. [CrossRef]
39. Wen, S.C.H.; Ezure, Y.; Rolley, L.; Spurling, G.; Lau, C.L.; Riaz, S.; Paterson, D.L.; Irwin, A.D. Gram-negative neonatal sepsis in low-
and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis.
PLoS Med. 2021, 18, e1003787. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.