Abstract
Background
Timely diagnosis and effective management are critical for improving outcomes in neonatal bloodstream infections (BSIs), particularly in resource-limited settings. This study aimed to determine clinical signs and laboratory markers associated with BSIs among neonates at Tertiary hospital in Mwanza, Tanzania.
Methods
A hospital-based cross-sectional study was conducted, among neonates treated for suspected BSI from January 2023 to June 2024. Clinical signs of sepsis were systematically documented. Blood samples were collected for culture, serial C-reactive protein (CRP), and complete blood count at admission, 24 h, and 72 h. Data were analyzed using R software version 4.4.
Results
A total of 327 neonates with suspected BSI were enrolled, with a median age of 1 day [interquartile range (IQR): 1–5 days]. The case-fatality rate was 9.2% (30/327). BSIs were confirmed in 53/327 (16.2%) neonates. Among confirmed cases, 39/53 (73.5%) were caused by Gram-negative bacteria, predominantly Klebsiella pneumoniae and Acinetobacter spp. Neonates with confirmed BSI had a significantly longer mean hospital stay than those with culture-negative sepsis (10.1 ± 8.0 days vs. 6.6 ± 6.6 days; p < 0.001). Elevated CRP was an independent predictor of BSI (p < 0.001).
Conclusion
BSIs remain a significant cause of neonatal morbidity and mortality in this setting. Gram-negative organisms predominated and were associated with worse outcomes. A combination of clinical features, early warning signs, and CRP may aid in the early identification of neonates at high risk of BSIs and support timely initiation of appropriate management in resource limited settings.
Keywords: C-reactive protein, gram negative bloodstream infection, low-resource settings, neonates, sepsis
Introduction
Sepsis remains a major global health threat, with neonates bearing the highest burden of disease and mortality (1). The prevalence and mortality of neonatal bloodstream infections (BSI) are approximately twofold higher in low- and middle-income countries (LMICs) than in high-income countries (HICs) (2, 3). The World Health Organization (WHO), through Sustainable Development Goal (SDG) 3.2, has emphasized the need to improve neonatal survival, targeting a reduction in neonatal mortality to at least 12 per 1,000 live births by 2030 (4). Achieving this goal in LMICs would require substantial improvements in the timely recognition and appropriate management of BSI (5).
In Tanzania, the prevalence of laboratory-confirmed BSI among neonates varies widely, ranging from 11% at Bugando Medical Centre (BMC), Mwanza, to as high as 70% at Mwananyamala Hospital in Dar es Salaam (6, 7). At BMC specifically, the prevalence has increased from 11% in 2018 to 59% in 2022 among neonates admitted to the neonatal intensive care unit and nursery unit. Despite routine use of blood culture in these studies, a significant proportion of clinically suspected cases yield no identifiable pathogen (6, 8).
Furthermore, there is currently no standardized local guidelines to guide decisions on antibiotic initiation. A recent study by Khan N (9) reported that strict adherence to national and international guidelines may lead to a doubling of antibiotic prescriptions, highlighting the urgent need for context-specific, practical diagnostic tools (9). The etiological spectrum of neonatal BSI varies globally, with multidrug-resistant Gram-negative bacteria being predominant in Tanzania and other LMIC neonatal units. In particular, Klebsiella pneumoniae is frequently reported as the leading pathogen, posing significant challenges to effective treatment. Although blood culture remains the gold standard for BSI diagnosis, it is often unavailable in many LMIC settings and requires 48–96 h to yield results, potentially delaying appropriate management (10–12). In HICs, predictive models incorporating perinatal risk factors, vital signs, and laboratory biomarkers have been developed to facilitate early initiation of appropriate empirical therapy (13, 14). Biomarkers such as elevated C-reactive protein (CRP) and abnormalities in complete blood count (CBC) have been shown to predict both early and late onset neonatal sepsis in Tanzania and other settings (15, 16). However, readily measurable clinical parameters including blood glucose, temperature, oxygen saturation, respiratory rate, and heart rate remain underutilized in neonatal care in Tanzania. Importantly, abnormalities in these vital signs have been shown to identify neonates at high risk of clinical deterioration and may support timely and targeted management (6, 15, 17–21).
C-reactive protein (CRP) and complete blood count (CBC) are technically simple, cost-effective, and have a short turnaround time, and they do not require advanced laboratory infrastructure or highly specialized personnel. In this study, we conducted a comprehensive analysis of bloodstream infections (BSI) at the Department of Pediatrics and Child Health, Bugando Medical Centre (BMC), a major healthcare provider in the Lake Victoria region of Tanzania.
Methodology
Study design, area and duration
This was a hospital-based cross-sectional study conducted at Bugando Medical Centre (BMC), Tanzania, within the Department of Pediatrics and Child Health, specifically in the neonatal units. BMC is a zonal referral and teaching hospital affiliated with the Catholic University of Health and Allied Sciences, with a bed capacity of approximately 950. It serves the Lake Zone, with an estimated population of 16 million people from the regions of Kagera, Geita, Mwanza, Shinyanga, Simiyu, and Mara. The neonatal unit is divided into the neonatal intensive care unit (NICU) and the special care nursery. The NICU admits approximately five neonates per week, while the special care nursery receives an average of 17 admissions per week.
Study population
The study included neonates admitted from 15th January 2023 to 30th June 2024 with clinical signs of sepsis from birth to 5 days of life. The clinical signs and symptoms included any of the following: temperature instability (hypothermia and fever), irritability, lethargy, respiratory symptoms (e.g., tachypnea, grunting, hypoxia), poor feeding, tachycardia, convulsion, jaundice, poor perfusion, and hypotension (22).
Definitions
Bloodstream infection (BSI) was defined as the presence of clinical signs of sepsis accompanied by isolation of a causative bacterial pathogen in blood culture. Body temperature was measured in degrees Celsius using a digital thermometer (Tempo Max, Spengler, France). Hypothermia was defined as an axillary temperature <36.5°C on admission, while fever was defined as a temperature >37.5°C. Oxygen saturation (SpO₂) was measured using a pulse oximeter (UT 100, Germany), and hypoxia was defined as an oxygen saturation of <90% in both term and preterm neonates. Random blood glucose was measured using a haemoglucometer (Accu-Chek® Instant Blood Glucose Meter, Roche® GmbH, Germany). Hypoglycemia was defined as a blood glucose level <2.5 mmol/L, while hyperglycemia was defined as >8.3 mmol/L. Heart rate was assessed using a pulse oximeter (UT 100, Germany), with a normal range defined as 100–160 beats per minute. Respiratory rate was determined by counting breaths per minute, with a normal range of 40–60 breaths per minute. Preterm infants were defined as neonates born before 37 completed weeks of gestation (21).
Coagulase-negative staphylococcal (CoNS) infection was defined as the presence of a single positive blood culture for CoNS in conjunction with evidence of systemic inflammation, indicated by an elevated C-reactive protein (CRP ≥10 mg/L) and/or an abnormal white blood cell (WBC) as previously described (23, 24).
Sample size and sampling procedure
A minimum sample size of 295 neonates with suspected BSIs was calculated using the Kish–Leslie formula (25). An estimated prevalence of 25.9% for confirmed BSI, derived from the average of previous studies conducted at Bugando Medical Centre, Tanzania, was used in the calculation (11, 20).
Data collection procedure
Patient recruitment and specimen collection
Patient were screened by pediatricians (principal investigator and research assistants) using the guidelines published by the WHO Young Infants Study group (22) that require assessment of clinical signs and symptoms relevant to BSIs such as fever, convulsions, difficulty in breathing, jaundice etc.
Patients were screened and enrolled within five days of life (at admission, at birth or when developing signs and symptoms of BSI). Clinical information such as history of premature delivery, place of delivery, mode of delivery, birth weight, birth weight, vitals etc. were collected during enrollment. Additionally, physical examinations were done to all included neonates at the time of enrollment. In addition, blood samples were collected for CRP, CBC and culture.
Data was collected through interviews with caregivers and information on the child progress during hospitalization was obtained from medical charts. Following blood sample collection, appropriate treatments following the Tanzania neonatal guideline were initiated using ampicillin and gentamicin as first line, cefotaxime as second line and meropenem as third line (21) and the neonates were followed up until discharge or death.
Clinical samples and laboratory procedures
Approximately 3 mL of venous blood was collected from all neonates with clinical sepsis at enrollment using plain and Ethylenediaminetetraacetic acid (EDTA) BD Vacutainer tubes (BD Vacutainer, Nairobi, Kenya) for the determination of C-reactive protein (CRP) and complete blood count (CBC), respectively, as previously described (16, 26). The turnaround time for these tests was approximately two hours. Laboratory parameters were interpreted according to established criteria (27). Serial measurements of CRP and CBC were performed at admission, and at 24 and 72 h post-admission.
For blood culture, approximately 1 mL of venous blood was inoculated into 10 mL of in-house prepared tryptone soy broth (TSB) and incubated aerobically at 37°C for 18–24 h upon receipt in the laboratory. Blind subcultures were performed onto in-house prepared 5% sheep blood agar (SBA), MacConkey agar (MCA), and MacConkey agar supplemented with cefotaxime (30 µg/mL) (Oxoid, UK), followed by aerobic incubation at 37°C for 18–24 h. If no growth was detected, broth cultures were re-incubated and sub-cultured at 48 h, 96 h, and finally on day 7.
Bacterial identification and antimicrobial susceptibility testing were performed as previously described (28, 29). Classification of isolates as true pathogens or contaminants was conducted using standardized criteria described by Murni et al. (24). Quality control procedures were strictly followed to ensure the sterility and reliability of culture media and laboratory processes.
Data management and analysis
Data analysis was performed using R software (version 4.4). Descriptive statistics, including frequencies and proportions, were used to summarize demographic, maternal, and clinical characteristics of the study participants. To compare mean values of vital signs and hematological indices between neonates with and without sepsis, data normality was assessed using the Shapiro–Wilk test. As the normality assumption was not met, the Wilcoxon rank-sum test was applied for comparisons between groups. Univariate analysis was conducted using the Chi-square test or Fisher's exact test, as appropriate, to identify factors associated with bloodstream infection (BSI). Variables with potential associations were further included in a multivariate analysis to assess independent predictors while accounting for confounding factors. Statistical significance was set at p < 0.05.
Ethical consideration
The study got ethical clearance from the Joint Bugando Medical Centre/Catholic University of Health and Allied Sciences ethics and review committee with certificate no CREC/586/2022 and from the National Health Research Ethics Review Committee (NatHREC) with certificate number protocol NIMR/HQ/R.8a/Vol 1X/4318. Permission to conduct the study was granted by the BMC hospital authority. Written informed consent was sought from all parents/guardians of the neonates after full disclosure of the protocol. All neonates who had sepsis during the study were managed according to the BMC BSI protocol.
Results
Social demographic and clinical characteristics of the enrolled participants
A total of 327 (31.0%) neonates with clinical signs of bloodstream infection (BSI) were admitted between January 2023 and June 2024 out of 1,056 total neonatal admissions. Of these, 155 (47.4%) were male. The median age at admission was 1 day [interquartile range (IQR): 1–5 days]. Among the neonates, 138 (42.2%) were preterm. Clinical characteristics at admission included inability to breastfeed in 28 (8.6%) neonates, convulsions in 20 (6.1%), fever (temperature >37.5°C) in 30 (9.1%), and oxygen saturation <90% in 50 (15.2%). Regarding maternal factors, the majority of mothers (222/327, 67.9%) reported antibiotic use during the last month of pregnancy. Prolonged premature rupture of membranes (PROM ≥18 h) occurred in 30 (9.2%) cases, and 60 (18.3%) mothers had meconium-stained amniotic fluid during labor (Table 1).
Table 1.
Social demographic and clinical characteristics of the 327 neonates with clinical sepsis admitted at bugando medical center, mwanza, Tanzania from January 2023 to June 2024a.
| Variable | All N (%) | Pat. with BSI N (%) | Pat. without BSI N (%) | P |
|---|---|---|---|---|
| Sex | ||||
| Female | 155 (47.4) | 23 (14.8) | 132 (85.2) | 0.63 |
| Male | 172 (52.6) | 30 (17.4) | 142 (82.6) | |
| Age (days) | 1.54 (0.98) | |||
| Gestational Age | ||||
| ≤36 6/7 weeks | 138 (42.2) | 23 (16.7) | 115 (83.3) | 0.97 |
| ≥37 weeks | 189 (57.8) | 30 (15.9) | 159 (84.1) | |
| Birth weight | ||||
| ≥ 2,500gm | 203 (61.9) | 30 (14.8) | 94 (46.3) | |
| 1,500 to 2,500gm | 107 (32.6) | 19 (17.8) | 88 (82.2) | 0.47 |
| 1,000 to 1,499gm | 17 (5.5) | 4 (23.5) | 13 (76.5) | |
| Clinical symptoms | ||||
| Delayed crying | 54 (16.5) | 7 (13.0) | 47 (87.0) | 0.55 |
| Difficulty in breathing | 233 (71.3) | 43 (18.5) | 190 (82.6) | 0.46 |
| Irritability | 12 (3.7) | 3 (25) | 9 (75) | 0.42 |
| Jaundice | 49 (14.9) | 6 (12.2) | 43 (87.8) | 0.53 |
| Cyanosis | 48 (14.7) | 9 (18.8) | 39 (81.2) | 0.67 |
| Inability to breastfeed | 28 (8.6) | 4 (14.3) | 24 (85.7) | 0.98 |
| Convulsions | 20 (6.1) | 3 (15) | 17 (85) | 1.0 |
| Axilla temperature | ||||
| ≤36.5°C | 104 (31.8) | 20 (19.2) | 84 (80.8) | |
| ≥37.5°C | 30 (9.1) | 2 (6.7) | 28 (93.3) | 0.4202 |
| 36.5°C to 37.5°C | 193 (59.1) | 31 (16.1) | 162 (83.9) | |
| Oxygen Saturation | ||||
| ≥90% | 277 (84.8) | 43 | 234 | 0.5605 |
| <90% | 50 (15.2) | 10 | 40 | |
| History of fever during pregnancy | ||||
| Yes | 55 (16.8) | 10 (18.2) | 45 (81.2%) | 0.16 |
| Maternal intrapartum fever | ||||
| Yes | 9 (2.8) | 2 (22.2) | 7 (77.8) | 0.8143 |
| Antibiotic usage during pregnancy | ||||
| Yes | 105 (32.1) | 85 (81.0) | 20 (19.0) | 0.2820 |
| Intrapartum antibiotics usage | ||||
| Yes | 131 (40.1) | 24 (18.3) | 107 (81.7) | 0.4251 |
| Duration of rupture of the membrane | ||||
| Less than 18 h | 297 (90.8) | 50 | 247 | 0.4410 |
| More than 18 h | 30 (9.2) | 3 | 27 | |
| Meconium-stained liquor | ||||
| Yes | 60 (18.3) | 13 (21.7) | 47 (78.3) | 0.2820 |
| Mode of delivery | ||||
| SVDb | 141 (56.9) | 117 (83) | 24 (17) | 0.8446 |
| C/Sc | 186(43.1) | 157(84.4) | 29(15.6) | |
Table 1 describe the baseline characteristics of neonates with and without bloodstream infection (BSI). Values are presented as number and percentage, N (%) for categorical variables and mean (± standard deviation) for continuous variables. Comparisons between neonates with and without bloodstream infection were performed using the Chi-square test for categorical variables with expected cell counts ≥5 and the Fisher's exact test for variables with small expected counts (< 5), including maternal intrapartum fever and duration of rupture of membranes >18 h. A p < 0.05 was considered statistically significant.
SVD, Spontaneous Vertex delivery;.
C/S, Caesarian Section.
Among the 274 neonates with culture-negative sepsis, the majority (268, 97.8%) received antibiotic treatment. Overall, antibiotics were initiated at the time of enrollment in 321 (98.2%) of all enrolled neonates. The most commonly prescribed regimen was a combination of ampicillin and gentamicin (269, 83.8%), followed by cefotaxime (52, 16.2%) (Figure 1).
Figure 1.
A flow chart shows 327 neonates with clinical sepsis and 53(16.2%) had positive culture results. Antibiotic treatment was initiated in 268(97.8%) neonates with negative culture results.
Factors associated with bloodstream infections
Table 2 summarizes the admission vital signs (temperature, heart rate, respiratory rate, and random blood glucose) stratified by bloodstream infection (BSI). Overall, there were no significant differences in vital signs between neonates with culture-confirmed BSI and those with negative blood culture results, except for oxygen saturation. Neonates with BSI had significantly lower oxygen saturation levels at admission compared to those without BSI (p = 0.021). In addition, there were significant differences in admission leukocyte count (p = 0.031), platelet counts (p = 0.004) and C-reactive protein (p < 0.001) levels between neonates with culture-confirmed bloodstream infection (BSI) and those without confirmed sepsis. Neonates with culture-positive BSI had higher CRP levels, lower leukocyte counts, and lower platelets than those without confirmed infection (Table 2). On multivariate analysis, only CRP remained independently associated with BSI (OR = 1.05, 95% CI: 1.03–1.08, p < 0.001).
Table 2.
Comparing vital signs and hematological parameters of the neonates with and without BSI admitted at the neonatal unit in bugando medical center in mwanza, Tanzania.
| Variable | BSI Mean (±SD) | No BSI Mean (±SD) | P-value |
|---|---|---|---|
| Temperature ( °C, mean/SD) | 36.5 (0.73) | 36.7 (0.74) | 0.38 |
| Heart Rate (bpm, mean/SD) | 140.7 (13.39) | 143.9 (15.21) | 0.4066 |
| Respiratory Rate (breath, mean/SD) | 59.7 (13.32) | 59.8 (13.85) | 0.9999 |
| Random Blood Glucose (mmol/L, mean/SD) | 5.0 (1.78) | 5.0 (2.40) | 0.4888 |
| SPO2(%, mean/SD) | 92.6849 (5.28) | 94.4340 (5.85) | 0.0209 |
| Hemoglobin level (g/dL) | 16.5 | 16.7 | 0.444 |
| Monocyte (x109/L | 1.2 | 1.5 | 0.144 |
| Lymphocyte (x109/L) | 3.7368 | 3.5922 | 0.678 |
| Leucocyte (x109/L) | 12.5664 | 14.4127 | 0.031 |
| Neutrophils (x109/L) | 7.6242 | 8.9912 | 0.068 |
| Platelets (x109/L) | 191.4200 | 230.2422 | 0.004 |
| C Reactive Protein (mg/L) | 45.5924 | 9.3605 | <0.001 |
| Neutrophil to Lymphocyte ratio (NLR) | 3.2 | 3.3 | 0.058 |
Gram-negative pathogens dominate the pathogenic spectrum of BSI
Of the 327 neonates with clinical signs of sepsis, positive blood culture was found in 53 (16.1%, 95% CI: 12.0–20.0). Most of the isolated bacteria were Gram-negative (39/53, 73.6%). Klebsiella pneumoniae, Coagulase-negative Staphylococcus (CoNS), and Acinetobacter spp. were the most frequently isolated bacteria (Table 3).
Table 3.
Distribution of bacterial organisms isolated from neonates with BSI.
| Organism | Frequency | Percent |
|---|---|---|
| Gram-positive Cocci | ||
| CoNS | 11 | 20.8% |
| Staphylococcus aureus | 2 | 3.8% |
| Streptococcus spp. | 1 | 1.9% |
| Total | 14 | 26.4% |
| Gram-negative Bacilli—Enterobacteriaceae | ||
| Escherichia coli | 2 | 3.8% |
| Klebsiella pneumoniae | 25 | 47.2% |
| Enterobacter aerogenes | 2 | 3.8% |
| Total | 29 | 54.7% |
| Gram-negative Bacilli non-fermenters | ||
| Acinetobacter spp. | 8 | 15.1% |
| Pseudomonas aeruginosa | 1 | 1.9% |
| Unidentified Gram-negative bacteria | 1 | 1.9% |
| Total | 10 | 18.9% |
| Total | 53 | 100.0% |
Outcome among neonates admitted at bugando medical centre
Regarding treatment outcomes, the overall mortality rate was 9.2% (30/327). The mean duration of hospital stay was significantly longer among neonates with culture-positive BSI than those with culture-negative results (10.1 ± 8.0 days vs. 6.6 ± 6.6 days, p = 0.0009). The majority of neonates (62.7%) were hospitalized for 1–6 days, while 5.3% (16/327) stayed for more than 21 days. Mortality was higher among neonates with culture-positive BSI (15.1%, 8/53) than those with culture-negative results (8.0%, 22/274), although this difference was not statistically significant (p = 0.06). Furthermore, no deaths were observed among neonates with Gram-positive BSI, whereas 20.5% (8/39) of those with Gram-negative BSI died.
Discussion
Neonatal morbidity and mortality remain major public health challenges, contributing substantially to under-five mortality, particularly in low- and middle-income countries (LMICs). This study assessed the prevalence, as well as clinical and hematological predictors, of bloodstream infections (BSI) among neonates. The overall mortality was 9.2%, with culture-confirmed BSI identified in 16.1% of neonates. The majority of isolates (73.6%) were Gram-negative bacteria, and elevated C-reactive protein (CRP) was independently associated with BSI.
The proportion of BSI (16.1%) observed in this study was lower than that reported in previous studies conducted in the same setting (11, 15) and other parts of East Africa (30, 31) but higher than reports from high-income countries (32, 33). The relatively lower prevalence in our study may reflect improvements in neonatal care and the implementation of national neonatal care guidelines in Tanzania (34) which have strengthened case management at lower-level health facilities. However, the widespread use of pre-referral antibiotics, as recommended by WHO guidelines, and the relatively small blood volume (1 mL) collected for culture may have reduced the diagnostic yield, thereby underestimating the true burden of BSI (35).
In this study, Klebsiella pneumoniae, coagulase-negative staphylococci (CoNS), and Acinetobacter spp. were the most frequently isolated pathogens which is consistent with previous findings from similar settings (36). Gram-negative bacteria formed the majority of the isolates in our study as previously reported in the same setting (11, 15). Klebsiella pneumoniae was the leading pathogen, accounting for approximately half of culture-confirmed BSI cases, in line with reports from this setting and other LMIC settings (6, 37–39).
The persistence of Klebsiella pneumoniae in neonatal units may be attributed to its ability to form biofilms, survive in hospital environments, and acquire antimicrobial resistance (40). In many LMIC settings, recurrent clonal outbreaks have been linked to overcrowding, limited infection prevention and control (IPC) capacity, and inadequate microbiological and genomic surveillance (41, 42). To better understand transmission dynamics and circulating strains, whole-genome sequencing (WGS) of isolates from clinical and environmental sources is recommended to inform targeted IPC strategies.
Clinically, neonates with BSI did not differ significantly in most vital signs compared to those without BSI, except for lower oxygen saturation at admission. Hypoxemia may represent an early warning sign of clinical deterioration, as previously reported [13,29]. Subtle abnormalities in physiological parameters such as oxygen saturation, temperature, and blood glucose may precede overt clinical signs of sepsis (19). When combined with laboratory markers such as CRP and platelet count, these indicators can support early identification and prompt management of neonatal sepsis, particularly in resource-limited settings.
Consistent with prior studies (15, 43) neonates with BSI had a significantly longer hospital stay compared to those without BSI. Mortality was higher among neonates with culture-positive infections and was observed exclusively among those with Gram-negative BSI, highlighting the clinical severity associated with these pathogens. This study has several limitations. First, the small blood volume (1 mL) collected for culture may have reduced pathogen detection rates. Second, prior antibiotic exposure among referred neonates may have further lowered culture positivity. Third, routine screening for Group B Streptococcus (GBS) and assessment of additional inflammatory markers such as procalcitonin and interleukin-6 were not performed, which may have limited comprehensive evaluation of neonatal sepsis.
Conclusion
Bloodstream infections remain a significant cause of neonatal morbidity and mortality in this setting. Elevated CRP was an independent predictor of BSI, while hypoxemia and thrombocytopenia were important clinical and laboratory findings associated with BSI. Gram-negative organisms, particularly Klebsiella pneumoniae, predominated and were associated with worse outcomes. In resource-limited settings, a combination of clinical features, early warning signs, and accessible laboratory markers such as CRP may aid in the early identification of neonates at high risk of BSI and support timely initiation of appropriate management.
Acknowledgments
We would like to express our sincere gratitude to all study participants and their caregivers for their invaluable contribution to this research. We are especially grateful to the research assistants for their dedication and commitment throughout the data collection process. We also extend our appreciation to the staff of the Departments of Pediatrics and Microbiology for their cooperation and support. Finally, we thank the administration of Bugando Medical Centre for granting permission to conduct this study.
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. The work was funded by the Else Kröner-Fresenius Foundation within the framework of the Else Kröner Center Würzburg-Mwanza.
Footnotes
Edited by: Thomas S Murray, Yale University, United States
Reviewed by: Dr Sayan Kumar Das, Tripura Medical College & Dr. B.R. Ambedkar Memorial Teaching Hospital, India
Ganga Moorthy, Duke University Hospital, United States
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.
Ethics statement
The studies involving humans were approved by Joint Bugando Medical Centre/Catholic University of Health and Allied Sciences ethics and review committee. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants' legal guardians/next of kin.
Author contributions
DM: Validation, Resources, Project administration, Data curation, Methodology, Conceptualization, Visualization, Writing – review & editing, Funding acquisition, Investigation, Supervision, Writing – original draft. FM: Formal analysis, Software, Writing – review & editing. OK: Conceptualization, Supervision, Writing – review & editing. CH: Supervision, Conceptualization, Writing – review & editing. SM: Writing – review & editing, Conceptualization, Supervision.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author SM declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
- 1.Sturrock S, Sadoo S, Nanyunja C, Le Doare K. Improving the treatment of neonatal sepsis in resource-limited settings: gaps and recommendations. Res Rep Trop Med. (2023) 14:121–34. 10.2147/RRTM.S410785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE, et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet. (2015) 385(9966):430–40. 10.1016/S0140-6736(14)61698-6 [DOI] [PubMed] [Google Scholar]
- 3.Li J, Shen L, Qian K. Global, regional, and national incidence and mortality of neonatal sepsis and other neonatal infections, 1990–2019. Front Public Health. (2023) 11:1139832. 10.3389/fpubh.2023.1139832 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.World Health Organization. Targets of Sustainable Development Goal 3 [press release] (2001). Available online at: https://www.who.int/data/gho/data/themes/topics/sdg-target-3_2-newborn-and-child-mortality (Accessed April 9, 2026).
- 5.Celik IH, Hanna M, Canpolat FE, Mohan P. Diagnosis of neonatal sepsis: the past, present and future. Pediatr Res. (2022) 91(2):337–50. 10.1038/s41390-021-01696-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Marando R, Seni J, Mirambo MM, Falgenhauer L, Moremi N, Mushi MF, et al. Predictors of the extended-spectrum-beta lactamases producing Enterobacteriaceae neonatal sepsis at a tertiary hospital, Tanzania. Int J Med Microbiol. (2018) 308(7):803–11. 10.1016/j.ijmm.2018.06.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Majigo M, Makupa J, Mwazyunga Z, Luoga A, Kisinga J, Mwamkoa B, et al. Bacterial aetiology of neonatal sepsis and antimicrobial resistance pattern at the regional referral Hospital, Dar es Salam, Tanzania; a call to strengthening antibiotic stewardship program. Antibiotics. (2023) 12(4):767. 10.3390/antibiotics12040767 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kovacs D, Silago V, Msanga DR, Mshana SE, Seni J, Oravcova K, et al. The hospital environment versus carriage: transmission pathways for third-generation cephalosporin-resistant bacteria in blood in neonates in a low-resource country healthcare setting. Sci Rep. (2022) 12(1):8347. 10.1038/s41598-022-11626-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khan N, Chimhini G, Shrestha SK, Cortina-Borja M, Chimhuya S, Zailani G, et al. Assessing the use of neonatal sepsis guidelines and antibiotic prescription with large-scale prospective data from Zimbabwe and Malawi. J Pediatric Infect Dis Soc. (2025) 14(4):piaf017. 10.1093/jpids/piaf017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Moorthy GS, Rubach MP, Sechu A, Mbwasi R, Peter N, Kalu IC, et al. Clinical characteristics, antimicrobial resistance, and mortality of neonatal bloodstream infections in northern Tanzania, 2022–2023. PLoS One. (2025) 20(3):e0319816. 10.1371/journal.pone.0319816 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Silago V, Kovacs D, Msanga DR, Seni J, Matthews L, Oravcová K, et al. Bacteremia in critical care units at bugando medical centre, mwanza, Tanzania: the role of colonization and contaminated cots and mothers’ hands in cross-transmission of multidrug resistant gram-negative bacteria. Antimicrob Resist Infect Control. (2020) 9(1):1–14. 10.1186/s13756-020-00721-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Thomson KM, Dyer C, Liu F, Sands K, Portal E, Carvalho MJ, et al. Effects of antibiotic resistance, drug target attainment, bacterial pathogenicity and virulence, and antibiotic access and affordability on outcomes in neonatal sepsis: an international microbiology and drug evaluation prospective substudy (BARNARDS). Lancet Infect Dis. (2021) 21(12):1677–88. 10.1016/S1473-3099(21)00050-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bomrah S, Uddin M, Upadhyay U, Komorowski M, Priya J, Dhar E, et al. A scoping review of machine learning for sepsis prediction- feature engineering strategies and model performance: a step towards explainability. Critical Care. (2024) 28(1):180. 10.1186/s13054-024-04948-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sokou R, Ioakeimidis G, Piovani D, Parastatidou S, Konstantinidi A, Tsantes AG, et al. Development and validation of a sepsis diagnostic scoring model for neonates with suspected sepsis. Front Pediatr. (2022) 10:1004727. 10.3389/fped.2022.1004727 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kayange N, Kamugisha E, Mwizamholya DL, Jeremiah S, Mshana SE. Predictors of positive blood culture and deaths among neonates with suspected neonatal sepsis in a tertiary hospital, mwanza-Tanzania. BMC Pediatr. (2010) 10(1):39. 10.1186/1471-2431-10-39 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chacha F, Mirambo MM, Mushi MF, Kayange N, Zuechner A, Kidenya BR, et al. Utility of qualitative C- reactive protein assay and white blood cells counts in the diagnosis of neonatal septicaemia at bugando medical centre, Tanzania. BMC Pediatr. (2014) 14(1):248. 10.1186/1471-2431-14-248 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bangi V, Syamala Devi S. Neonatal sepsis: a risk approach. J Dr NTR Univ Health Sci. (2014) 3(4):254–8. 10.4103/2277-8632.146632 [DOI] [Google Scholar]
- 18.Wang B, Wang Q-M, Li D-X. An analysis of predictive factors for severe neonatal infection and the construction of a prediction model. Infect Drug Resist. (2023) 19:3561–74. 10.2147/IDR.S408126 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Duncan KD, McMullan C, Mills BM. Early warning systems: the next level of rapid response. Nursing (Brux). (2012) 42(2):38–44; quiz 5. 10.1097/01.NURSE.0000410304.26165.33 [DOI] [PubMed] [Google Scholar]
- 20.Msanga DR, Parpia F, Konje ET, Hokororo A, Mshana SE. High mortality among premature neonates with positive blood culture neonatal sepsis in a tertiary hospital, Tanzania: a call for action. Children. (2021) 8(11):1037. 10.3390/children8111037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tanzania Ministry of Health CD, Gender, Elderly and Children. National Guideline For Neonatal Care and Establishment Of Neonatal Care Unit;1 (2019).
- 22.Group WYIS. Bacterial etiology of serious infections in young infants in developing countries: results of a multicenter study. Pediatr Infect Dis J. (1999) 18(10):S17–22. 10.1097/00006454-199910001-00004 [DOI] [PubMed] [Google Scholar]
- 23.Henry E, Christensen RD. Reference intervals in neonatal hematology. Clin Perinatol. (2015) 42(3):483–97. 10.1016/j.clp.2015.04.005 [DOI] [PubMed] [Google Scholar]
- 24.Murni IK, Duke T, Daley AJ, Kinney S, Soenarto Y. True pathogen or contamination: validation of blood cultures for the diagnosis of nosocomial infections in a developing country. J Trop Pediatr. (2018) 64(5):389–94. 10.1093/tropej/fmx081 [DOI] [PubMed] [Google Scholar]
- 25.Kish L. Survey Sampling. New York: John Wiley & Sons; (1965). [Google Scholar]
- 26.López Sastre JB, Pérez Solís D, Roqués Serradilla V, Fernández Colomer B, Coto Cotallo GD. Evaluation of procalcitonin for diagnosis of neonatal sepsis of vertical transmission. BMC Pediatr. (2007) 7(1):9. 10.1186/1471-2431-7-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Newman TB, Puopolo KM, Wi S, Draper D, Escobar GJ. Interpreting complete blood counts soon after birth in newborns at risk for sepsis. Pediatrics. (2010) 126(5):903–9. 10.1542/peds.2010-0935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Murray Baron PT. Manual of Clinical Microbiology. Washington DC: American Society of Microbiology Press; (1999). [Google Scholar]
- 29.Matthew AF, William A, Michael N. Clinical and Laboratory Standards Institute (CLSI). Performance Standards for Antimicrobial Susceptibility Testing. 17th Informational Supplement. Wayne (PA): CLSI (2007). p. 44–57. CLSI document M100-S17. [Google Scholar]
- 30.Okube OT, Komen M. Prevalence and predictors of neonatal sepsis among neonates admitted at the newborn unit of kenyatta national hospital, Nairobi, Kenya. Open J Obstet Gynecol. (2020) 10(9):1216–32. 10.4236/ojog.2020.1090114 [DOI] [Google Scholar]
- 31.Katugume B, Muzungu J, Okello N, Kigongo E, Namutebi DA. Prevalence of neonatal sepsis and associated factors among neonates admitted in the neonatal intensive care unit at lira regional referral hospital, northern Uganda. PLoS One. (2025) 20(1):e0315794. 10.1371/journal.pone.0315794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kuld R, Krauth A, Kühr J, Krämer J, Dittrich R, Häberle L, et al. Possible rates of detection of neonatal sepsis pathogens in the context of microbiological diagnostics in mothers–real world data. Geburtshilfe Frauenheilkd. (2023) 83(11):1382–90. 10.1055/a-2091-0856 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Fleischmann C, Reichert F, Cassini A, Horner R, Harder T, Markwart R, et al. Global incidence and mortality of neonatal sepsis: a systematic review and meta-analysis. Arch Dis Child. (2021) 106(8):745–52. 10.1136/archdischild-2020-320217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ministry of Health CD, Gender, Elderly and Children. National Guideline for Neonatal Care and Establishment of Neonatal Care Unit. 1st ed. Dodoma: The United Republic Of Tanzania; (2019). p. 286. [Google Scholar]
- 35.Connell TG, Rele M, Cowley D, Buttery JP, Curtis N. How reliable is a negative blood culture result? Volume of blood submitted for culture in routine practice in a children's hospital. Pediatrics. (2007) 119(5):891–6. 10.1093/ofid/ofz179 [DOI] [PubMed] [Google Scholar]
- 36.Golińska E, Kozień Ł, Tomusiak-Plebanek A, Kędzierska J, Dorycka M, Lauterbach R, et al. Epidemiology of neonatal sepsis in two neonatal intensive care units in krakow, Poland in 2016–2017 years. BMC Infect Dis. (2023) 23(1):827. 10.1186/s12879-023-08836-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mshana SE, Hain T, Domann E, Lyamuya EF, Chakraborty T, Imirzalioglu C. Predominance of Klebsiella pneumoniaeST14 carrying CTX-M-15 causing neonatal sepsis in Tanzania. BMC Infect Dis. (2013) 13(1):466. 10.1186/1471-2334-13-466 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Chandrakoshi G, Kumar S. The role of hematological parameters in diagnosis of clinically suspected cases of neonatal septicemia. Int J Biol Med Res. (2018) 9(3):6389–92. [Google Scholar]
- 39.Das SK, Prajapati N. A retrospective study on the outcome of sepsis among neonates admitted to the neonatal intensive care unit at a tertiary care hospital in western India. J Med Soc. (2024) 38(1):63–8. 10.4103/jms.jms_79_23 [DOI] [Google Scholar]
- 40.Michaelis C, Grohmann E. Horizontal gene transfer of antibiotic resistance genes in biofilms. Antibiotics. (2023) 12(2):328. 10.3390/antibiotics12020328 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Banerjee T, Bhattacharjee A, Upadhyay S, Mishra S, Tiwari K, Anupurba S, et al. Long-term outbreak of Klebsiella pneumoniae & third generation cephalosporin use in a neonatal intensive care unit in north India. Indian J Med Res. (2016) 144(4):622–9. 10.4103/0971-5916.200900 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Phillips LT, Bates M, Coffin SE, Foster-Nyarko E, Kapasa M, Machona S, et al. Transmission dynamics of Klebsiella pneumoniae in a neonatal intensive care unit in Zambia before and after an infection control bundle. PLOS glob Pub Health. (2026) 6(2):e0005965. 10.1371/journal.pgph.0005965 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sisay EA, Mengistu BL, Taye WA, Fentie AM, Yabeyu AB. Length of hospital stay and its predictors among neonatal sepsis patients: a retrospective follow-up study. Int J Gen Med. (2022) 15:8133. 10.2147/IJGM.S385829 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.