Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis

Stefani Miranda; Aminuddin Harahap; Dominicus Husada; Muhammad Reza

doi:10.12688/f1000research.158689.2

Home Browse Risk factors for mortality due to neonatal sepsis: a systematic review...

ALL Metrics

Views

Downloads

Get PDF

Get XML

Export

▬

✚

Systematic Review

Revised

Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis

[version 2; peer review: 1 approved with reservations]

Stefani Miranda ¹, Aminuddin Harahap², Dominicus Husada^3,4, Muhammad Reza⁵

PUBLISHED 23 Jul 2025

Author details Author details

¹ Drs.Titus Uly Kupang Police Hospital, Jalan Nangka 84, Kupang, East Nusa Tenggara, 85112, Indonesia
² Department of Child Health, Dr. Ramelan Navy Central Hospital, Jalan Gadung No.1, Surabaya, East Java, 60244, Indonesia
³ Department of Child Health, Faculty of Medicine, Universitas Airlangga, Surabaya, Indonesia
⁴ Dr. Soetomo Academic General Hospital, Jalan Prof. Dr. Moestopo 6-8, Surabaya, East Java, 60286, Indonesia
⁵ Department of Child Health, Waluyo Jati General Hospital, Jalan Dr. Soetomo 1, Probolinggo, East Java, 67282, Indonesia

Stefani Miranda
Roles: Conceptualization, Data Curation, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Resources, Software, Validation, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Aminuddin Harahap
Roles: Resources, Supervision, Writing – Review & Editing

Dominicus Husada
Roles: Conceptualization, Formal Analysis, Methodology, Resources, Supervision, Validation, Writing – Review & Editing

Muhammad Reza
Roles: Formal Analysis, Methodology, Resources, Supervision, Validation, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

Abstract

Background

Certain risk factors have been shown to increase the mortality of patients with neonatal sepsis. This study aimed to determine the risk factors for neonatal sepsis-related mortality.

Methods

Google Scholar, MEDLINE, ProQuest, ScienceDirect, and Scopus databases were searched to identify relevant literature from 2014 to 2023. Observational analytical studies in English that reported the risk factors for neonatal sepsis mortality were chosen. We assessed the risk of bias by using the checklists of the Joanna Briggs Institute. Fixed-effect models were used when the number of included studies was <5; otherwise, random-effects models were employed. Heterogeneity was evaluated using the I² statistic. Publication bias was assessed using a funnel plot, and a sensitivity analysis was performed. Statistical significance was set at P <0.05. Analyses were conducted using the RevMan 5.4.1.

Results

Twelve out of 40,587 articles included a total of 2232 patients. The majority of patients were male (50.4%–70%). The I² statistics showed no heterogeneity across studies for sex, gestational age, birth weight, or requirement for inotropic support. History of invasive ventilation (OR = 35.06 [16.84–72.99]), requirement for inotropic support (OR = 18.04 [8.38–38.81]), low 1^st minute Apgar score (OR = 4.93 [2.1–11.58]), convulsive (OR = 4.69 [2.03–10.82]), poor feeding (OR = 3.95 [2.12–7.33]) episodes, preterm birth (OR = 3.63 [2.78–4.74]), low birth weight (OR = 3.02 [1.58–5.75]), early onset sepsis (OR = 2.52 [1.74–3.64]), and lethargy (OR = 2.14 [1.5–3.04]) were associated with neonatal sepsis mortality.

Conclusions

A history of invasive ventilation use, requirement for inotropic support, low 1^st minute Apgar score, convulsions, poor feeding episodes, preterm birth, low birth weight, early onset sepsis, and lethargy were identified as significant risk factors for neonatal sepsis mortality. Clinicians must be vigilant to improve outcomes and prevent death.

Keywords

meta-analysis, mortality, neonatal sepsis, risk factors, systematic review.

Corresponding author: Stefani Miranda

Competing interests: No competing interests were disclosed.

Grant information: The author(s) declared that no grants were involved in supporting this work.

Copyright: © 2025 Miranda S et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Miranda S, Harahap A, Husada D and Reza M. Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis [version 2; peer review: 1 approved with reservations]. F1000Research 2025, 13:1532 (https://doi.org/10.12688/f1000research.158689.2) First published: 19 Dec 2024, 13:1532 (https://doi.org/10.12688/f1000research.158689.1) Latest published: 23 Jul 2025, 13:1532 (https://doi.org/10.12688/f1000research.158689.2)

Revised Amendments from Version 1

We've made some minor changes to the paper, particularly in the "Statistical analysis" section, due to the typographical error.

See the authors' detailed response to the review by Samuel R Neal

Background

Neonates with infections face significant challenges, as the outcomes for this population can be quite poor.¹ Neonatal infections, particularly systemic infections known as neonatal sepsis, remain a prevalent concern.² Neonatal sepsis is characterized by a systemic inflammatory response syndrome (SIRS) triggered by infection within the first four weeks of life.^3,4 Due to the non-specific nature of initial clinical signs—such as tachycardia, fever, and drowsiness—neonatal sepsis can often be misdiagnosed as other neonatal conditions, including metabolic diseases, respiratory distress syndrome, and intracranial hemorrhage.³ Furthermore, clinical manifestations such as respiratory distress, hypothermia, cyanosis, apnea, convulsions, and prolonged capillary refill time are closely associated with mortality related to neonatal sepsis, possibly linked to metabolic derangements and cardiovascular collapse.^5,6

Neonatal sepsis is a major health concern in neonatal health care units worldwide.⁴ Neonatal infections account for more than one-third (36%) of all neonatal mortalities worldwide.^7,8 The mortality rate of neonatal sepsis ranges from 9% to 65%.⁹ Sepsis is the main cause of neonatal mortality, accounting for more than one million deaths each year worldwide.^7,8 According to the World Health Organization (WHO), 2.4 million neonates die from neonatal sepsis,¹⁰ with developing countries accounting for 40% of these deaths.¹¹ In developed countries, mortality rates range from 5% to 20%, resulting in significant disability.^12–15

All organ systems undergo discernible physiological transitions and learn to respond to various external stimuli during the neonatal period, which means that this period is a period of high exposure, as they perform tasks necessary for survival outside the womb.¹⁶ The immaturity of the immune system at this time, especially in premature neonates, lends different clinical, physical, and outcome characteristics to infections compared to other age groups. Furthermore, inherent factors such as poorly established and immature skin barriers, mucosal defense mechanisms, and blood-brain barriers contribute to the increased vulnerability of neonates to infection. Therefore, neonates are more susceptible to various pathogens.¹⁷

Preterm birth, low birth weight, and prolonged rupture of membranes are known risk factors that increase the likelihood of neonatal sepsis-related mortality.^3,18 In addition to these risk factors, studies have shown that low socioeconomic conditions can further increase the likelihood of neonatal sepsis mortality. This could be due to poor access to healthcare facilities or a lack of awareness of preventive measures for neonatal sepsis.¹⁹ Clinicians should recognize these risk factors and take measures to prevent, diagnose, and treat neonatal sepsis to reduce morbidity and mortality. Herein, we performed a systematic review and meta-analysis to assess the risk factors of mortality in patients with neonatal sepsis.

Methods

This systematic review and meta-analysis were conducted following the reporting guidelines provided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) in 2020.²⁰ Our research protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO), with registration number CRD42023456164.

Data sources and search strategy

Five electronic databases, namely Google Scholar, MEDLINE, ProQuest, ScienceDirect, and Scopus, were independently searched by three authors (SM, AH, and MR) to explore relevant literature published in English between 2014 and 2023. The databases were searched using the following keywords and their variations: “risk factors,” “neonates,” “sepsis,” and “mortality” (Additional File 1). In addition, back referencing was conducted by reviewing the bibliographies of the included studies to identify potentially relevant literature.

Eligibility criteria

Articles were included if they met the following criteria: 1) human studies, 2) original articles, 3) neonate studies (0–28 days), 4) observational analytic studies (cohort studies, case-control studies, and cross-sectional studies), 5) were published in English, and 6) reported on one or more risk factors for neonatal sepsis mortality. Anonymous reports, case reports, citations without abstracts and/or full texts, commentaries, duplicate publications, editorials, letters to the editor, reviews, and qualitative studies were excluded. In cases where duplicate reports or studies were shared, the analysis included the first published article with the largest patient sample.

The selection of studies was carried out independently by three authors (SM, AH, and MR) following a three-step process based on predetermined inclusion and exclusion criteria. The first step involved screening the article titles. Titles approved by either author were then used in the second step. In the second step, the abstracts of the articles selected from the first step were screened. The abstracts approved by the two authors were then moved on to the third step. The final step involved screening the full text of the studies. The full texts of the studies selected in the second step were reviewed by all three authors. Studies approved by all three authors were included in this meta-analysis. Any disagreements were resolved by consensus discussion with the fourth author (DH).

Study quality assessment

Three authors (SM, AH, and MR) independently assessed the risk of bias in the included studies using the Joanna Briggs Institute (JBI) Checklist. Any disagreements regarding the risk of bias in particular studies were resolved through consensus-based discussions with the fourth author (DH). The quality of the studies was determined based on a previous meta-analysis, in which studies with a JBI score of over 70% were considered to be of high quality, those with a score between 50% and 70% were classified as medium quality, and those with a score below 50% were considered to be of low quality.²¹

Data extraction

All selected articles underwent a double data entry process using Microsoft Excel spreadsheets. Two authors (SM and AH) independently used a predesigned data extraction form to extract the data. Any discrepancies between the authors were resolved through consensus-based discussions with the fourth author (DH). The data collection encompassed the authors’ names, country, year of publication, study design, study period, sampling method, sample size, number of deceased patients, and number and percentage of analyzed risk factors.

Statistical analysis

We used the odds ratio to determine the correlation between neonatal sepsis mortality and its risk factors. Random-effects and fixed-effects models were used in this meta-analysis. Fixed-effect models were used when the number of included studies was less than five²²; otherwise, random-effects models were implemented. The degree of heterogeneity in effect sizes was evaluated using I² statistics.^23,24 I² values ranged from 0 to 100%, with a value of 0% indicating no heterogeneity. The I² value of 0%–25% indicates no heterogeneity, 25%–50% indicates moderate heterogeneity, 50%–75% indicates large heterogeneity, and 75%–100% indicates extreme heterogeneity.^23,25,26 This was visually represented in a forest plot. The overall effects were calculated using the combined Z-value. A p-value of significance was set at P < 0.05.²⁵ Sensitivity analysis was conducted by removing each study and recalculating the p-value for the remaining studies.²⁷ Funnel plots were used to assess the presence of publication bias among the risk factor groups,^28–30 with a minimum of six studies.³¹ All analyses were performed using RevMan 5.4.1 (Cochrane Collaboration, Oxford, UK).³²

Results

Search results

A total of 40,587 articles were identified on the basis of the search criteria from the selected databases. After removing the duplicates, 32,566 articles remained. Furthermore, 29,162 articles were excluded after reviewing their titles and abstracts. Of the remaining 3,404 articles, 3,394 were excluded because they either had citations without full text and/or abstracts, lacked necessary statistical data, did not provide patients’ demographic characteristics, or did not include mortality data. Two articles were obtained through back-referencing after excluding 16 from a total of 18 articles. Ultimately, 12 articles were included in the meta-analysis.^{15,17,18,33–41} The process of selecting relevant literature is shown in Figure 1.

Figure 1. Flowchart of the literature selection process.

Study characteristics

The risk of bias was assessed using the JBI checklists. All the articles included in the study scored above 70% on the JBI checklist, indicating that they were of high quality ( Figure 2).^{15,17,18,33–41} The articles were published between 2017 and 2023. Most of the studies were case-control studies conducted in different regions, ranging from Indonesia to Ethiopia. In total, there were 2232 patients involved in the studies, with varying sample sizes of neonates ranging from 50 to 455. The proportion of males in the study samples ranged from 50.4% to 70%. The number of deceased neonates varied from 21 to 174 (Table extended data).^{15,17,18,33–41}

Figure 2. Risk of bias of included studies.

(a1, a2) Case-control studies. (b1, b2) Cohort studies. (c1, c2) Cross-sectional study.

Mortality-related risk factors

In this study, we collected 14 effect sizes from the 12 studies.^{15,17,18,33–41} Our analysis revealed an increased risk of mortality in neonates with sepsis who had a history of using invasive ventilation (OR = 35.06 [16.84–72.99]), required inotropic support (OR = 18.04 [8.38–38.81]), had a low 1^st minute Apgar score (OR = 4.93 [2.1–11.58]); convulsive (OR = 4.69 [2.03–10.82]), poor feeding (OR = 3.95 [2.12–7.33]) episodes; preterm birth (OR = 3.63 [2.78–4.74]), low-birth-weight (OR = 3.02 [1.58–5.75]), early onset sepsis (OR = 2.52 [1.74–3.64]), and lethargy (OR = 2.14 [1.5–3.04]) ( Figure 3).

Figure 3. Forest plots show the association between risk factors and neonatal sepsis mortality.

(a) Gender. (b) Onset of sepsis. (c) Birth weight. (d) Gestational age. (e) Mode of delivery. (f ) Admission type. (g) Low 1^st minute Apgar score. (h) Hypothermia episodes. (i) Convulsive episodes. (j) Jaundice episodes. (k) Lethargy episodes. (l) Poor feeding episodes. (m) The use of invasive ventilation. (n) Requirement for inotropic support.

Heterogeneity, sensitivity analysis, and publication bias

The I² statistics for sex, birth weight, gestational age, and requirement for inotropic support did not show heterogeneity among the studies ( Figure 3). Our sensitivity analysis revealed that the overall estimates for the onset of sepsis, birth weight, gestational age, a low 1^st minute Apgar score, having convulsive, poor feeding episodes, the use of invasive ventilation, and the requirement for inotropic support were not influenced by any single study. The funnel plot for admission and convulsive episodes displayed a symmetrical distribution, suggesting no significant publication bias in the included studies ( Figure 4).

Figure 4. Funnel plots of meta-analysis including.

(a) Gender. (b) Onset of sepsis. (c) Gestational age. (d) Mode of delivery. (e) Admission type. (f ) Convulsive episodes.

Discussion

Our analysis showed that history of invasive ventilation, requirement for inotropic support, low 1^st minute Apgar score, convulsive, poor feeding episodes, preterm birth, low birth weight, early onset sepsis, and lethargy was identified as significant risk factors for neonatal sepsis mortality ( Figure 3).

As proven in this study, neonates who experienced early onset sepsis (EOS) were approximately two times more at risk of mortality. This finding could be attributed to the rapid and severe progression of the condition leading to septic shock and eventual death, commonly seen in cases of EOS.⁴² Nevertheless, a few studies have documented a higher mortality rate due to late-onset sepsis (LOS).^43,44 The increased mortality rates observed in the LOS group were attributed to the performance of invasive procedures, extended hospital stay, and prolonged administration of antibiotics.^40,45,46 Neonates with sepsis and a history of invasive ventilation were 35 times more at risk for death, as observed in this meta-analysis.

Neonates with sepsis and a history of poor feeding were at risk of mortality four times. Breast milk, as a source of vitamin A and antibodies, helps fight infections in neonates. Sepsis in poorly fed neonates has been shown to engender hypoglycemia through the inhibition of gluconeogenesis, lactic acidosis, and increased glucose requirements. Thus, the risk of death was high in this group of neonates.¹⁷ Neonates with sepsis who developed multiple organ failure were at a high risk of death. These neonates require inotropic drugs to support their cardiovascular system.¹⁸ In our study, neonates who required inotropic support were 18 times at risk of mortality. The Apgar score is clinically used to evaluate hemodynamic impairment. It was performed in the 1^st minute as a screening tool to determine the need for early intervention. Low Apgar scores were correlated with increased mortality in neonates without congenital anomalies.⁴⁷ Neonates with sepsis a low 1^st minute Apgar score were five times more at risk for mortality in this study.

As observed in this study, preterm birth and low birth weight (LBW) septic neonates increased the risk of neonatal sepsis mortality by three times each. The prevalence of sepsis was higher in preterm and LBW neonates, with a mortality rate of 17.6%. Preterm neonates with sepsis face higher complication risks due to their immature immune systems, leading to impaired humoral immunity and vulnerability in their early days.⁴⁸ Mortality in this group of neonates is due to insufficient immunoglobulin synthesis, complement systems (C3 and C5), phagocytosis, and opsonization.⁴⁹ The risk of mortality in preterm neonates with sepsis increases as birth weight decreases.⁵⁰ LBW neonates are more vulnerable to sepsis due to the insufficiency of the immune system and the number of invasive and therapeutic procedures performed, leading to nosocomial infections.⁵¹

Neonates with sepsis who presented with lethargy were twice at risk for mortality, while those who had convulsive episodes were four times at risk of death. Our conclusions are in accordance with those of previous studies.^5,46 Similar to the observations of other authors, we found that sepsis neonates with lethargic and convulsive episodes had significantly higher mortality rates, which may be attributed to cardiovascular collapse and metabolic derangements.^41,52–55 Structural brain lesions, such as brain infarctions and hemorrhage, are aggravated by convulsive episodes, which govern the overall physiological and hemodynamic stability. Other conditions such as acute neonatal encephalopathy can also increase mortality in patients with sepsis¹⁷

Our study has some limitations. Some of the limitations are: 1) The inclusion criteria of the studies, which were published exclusively in English, may limit their representativeness; and 2) Data regarding the risk of various adverse outcomes were derived from a limited number of studies. There is a need for large prospective studies with appropriate controls to further explore these issues.

Conclusions

In summary, we found through a meta-analysis that neonates with sepsis who had a history of using invasive ventilation, required inotropic support, had a low 1st minute Apgar score, had convulsive and poor feeding episodes, preterm birth, low birth weight, early onset sepsis, and lethargy were more at risk for death. Clinicians must be alert about this risk factor and employ early and personalized treatment strategies to increase effectiveness and minimize mortality.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Data availability statement

Underlying data

No data are associated with this article.

Extended data

Figshare: Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis. The project contains the following extended data:

• Additional File 1. DOI: https://doi.org/10.6084/m9.figshare.27690711.⁵⁶
• Figure 1 Flowchart of the literature selection process. DOI: https://doi.org/10.6084/m9.figshare.27685539.⁵⁷
• Table extended data: Characteristics of the studies. DOI: https://doi.org/10.6084/m9.figshare.27686955.⁵⁸

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Reporting guidelines

Figshare: PRISMA checklist for ‘Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis’. The project contains the following reporting guidelines:

• PRISMA_2020_checklist_Miranda. DOI: https://doi.org/10.6084/m9.figshare.27706389.⁵⁹
• Figure 1 Flowchart of the literature selection process. DOI: https://doi.org/10.6084/m9.figshare.27685539.⁵⁷

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Acknowledgements

Not applicable.

References

1. Odabasi IO, Bulbul A: Neonatal sepsis. Sisli Etfal Hastan. Tıp Bul. 2020; 54(2): 142–158. PubMed Abstract | Publisher Full Text
2. Pace E, Yanowitz T: Infections in the NICU: neonatal sepsis. Semin. Pediatr. Surg. 2022 Aug 1; 31(4): 151200. Publisher Full Text
3. Salim RF, Sobeih AA, Abd El Kareem HM: Evaluation of the clinical value of circulating miR-101, miR-187 and miR-21 in neonatal sepsis diagnosis and prognosis. Egypt. J. Med. Hum. Genet. 2020 Dec 9; 21(1): 12. Publisher Full Text
4. Chaudhry S, Haroon F, Irfan Waheed KA, et al.: Blood lactate levels and lactate clearance as predictors of mortality in neonatal sepsis. J. Ayub Med. Coll. Abbottabad. 2022; 34(3): 438–441. Publisher Full Text Reference Source
5. Jajoo M, Kapoor K, Garg L, et al.: To study the incidence and risk factors of early onset neonatal sepsis in an out born neonatal intensive care unit of India. J. Clin. Ne onatol. 2015; 4(2): 91. Publisher Full Text
6. Bharad RV, Singh CS, Singh LR: Risk factors and immediate outcome of early onset neonatal sepsis. J. Med. Sci. Clin. Res. 2017 Apr 30; 05(04): 21050–21056. Publisher Full Text Reference Source
7. Nordberg V, Iversen A, Tidell A, et al.: A decade of neonatal sepsis caused by gram-negative bacilli-a retrospective matched cohort study. Eur. J. Clin. Microbiol. Infect. Dis. 2021 Sep 24; 40(9): 1803–1813. PubMed Abstract | Publisher Full Text | Free Full Text
8. Shane AL, Sánchez PJ, Stoll BJ: Neonatal sepsis. Lancet. 2017 Oct 14; 390(10104): 1770–1780. Publisher Full Text
9. López UOJ, Buriticá HHM: Lethality by neonatal sepsis, risk factors and microbiological characteristics. Andes Pediatr. 2021 Oct 2; 92(5): 690–698. PubMed Abstract | Publisher Full Text Reference Source
10. World Health Organization: Newborn mortality.2022 [cited 2023 Sep 20]. Reference Source
11. Zelellw DA, Dessie G, Worku Mengesha E, et al.: A systemic review and meta-analysis of the leading pathogens causing neonatal sepsis in developing countries. Ekwunife OI, editor. Biomed. Res. Int. 2021 Jun 5; 2021: 1–20. Publisher Full Text Reference Source
12. Bakhuizen SE, de Haan TR , Teune MJ, et al.: Meta-analysis shows that infants who have suffered neonatal sepsis face an increased risk of mortality and severe complications. Acta Paediatr. 2014 Dec; 103(12): 1211–1218. PubMed Abstract | Publisher Full Text
13. Kabwe M, Tembo J, Chilukutu L, et al.: Etiology, antibiotic resistance and risk factors for neonatal sepsis in a large referral center in Zambia. Pediatr. Infect. Dis. J. 2016 Jul; 35(7): e191–e198. PubMed Abstract | Publisher Full Text
14. Wynn JL, Wong HR, Shanley TP, et al.: Time for a neonatal-specific consensus definition for sepsis. Pediatr. Crit. Care Med. 2014 Jul; 15(6): 523–528. PubMed Abstract | Publisher Full Text | Free Full Text
15. Vizcarra-Jiménez D, Copaja-Corzo C, Hueda-Zavaleta M, et al.: Predictors of Death in Patients with Neonatal Sepsis in a Peruvian Hospital. Trop. Med. Infect. Dis. 2022 Oct 31; 7(11): 342. Publisher Full Text Reference Source
16. Kliegman RM, St. Geme JW III: Nelson textbook of pediatrics. 21st ed.Behrman RE, editor. Philadelphia: Elsevier Inc.; 2020; p. 53.
17. Bekele T, Merga H, Tesfaye T, et al.: Predictors of mortality among neonates hospitalized with neonatal sepsis: a case control study from southern Ethiopia. BMC Pediatr. 2022 Jan 3; 22(1): 1. PubMed Abstract | Publisher Full Text | Free Full Text
18. Jovičić M, Milosavljević MN, Folić M, et al.: Predictors of mortality in early neonatal sepsis: a single-center experience. Medicina (B Aires). 2023 Mar 18; 59(3): 604. PubMed Abstract | Publisher Full Text | Free Full Text Reference Source
19. Bohanon FJ, Lopez ON, Adhikari D, et al.: Race, income and insurance status affect neonatal sepsis mortality and healthcare resource utilization. Pediatr. Infect. Dis. J. 2018 Jul; 37(7): e178–e184. PubMed Abstract | Publisher Full Text | Free Full Text
20. Page MJ, McKenzie JE, Bossuyt PM, et al.: The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar 29; 372: n71. Publisher Full Text
21. Pimsen A, Kao CY, Hsu ST, et al.: The effect of advance care planning intervention on hospitalization among nursing home residents: a systematic review and meta-analysis. J. Am. Med. Dir. Assoc. 2022 Sep; 23(9): 1448–1460.e1. PubMed Abstract | Publisher Full Text
22. Tufanaru C, Munn Z, Stephenson M, et al.: Fixed or random effects meta-analysis? Common methodological issues in systematic reviews of effectiveness. Int. J. Evid. Based Healthc. 2015 Sep; 13(3): 196–207. Publisher Full Text Reference Source
23. Higgins JP, Thompson SG, Deeks JJ, et al.: Measuring inconsistency in meta-analyses. BMJ (Online). 2003 Sep 7; 327(2): 557–560. Publisher Full Text
24. Huedo-Medina TB, Sánchez-Meca J, Marín-Martínez F, et al.: Assessing heterogeneity in meta-analysis: Q statistic or I 2 Index? Psychol. Methods. 2006 Jun; 11(2): 193–206. Publisher Full Text
25. Clephas PRD, Heesen M: Interpretation of meta-analyses. Interventional. Pain Med. 2022; 1(June): 100120. Publisher Full Text
26. Pei Y, Tian Q, Zhang L, et al.: Exploring the major sources and extent of heterogeneity in a genome-wide association meta-analysis. Ann. Hum. Genet. 2016 Mar 20; 80(2): 113–122. PubMed Abstract | Publisher Full Text | Free Full Text
27. Tawfik GM, Dila KAS, Mohamed MYF, et al.: A step by step guide for conducting a systematic review and meta-analysis with simulation data. Trop. Med. Health. 2019 Dec 1; 47(1): 46. PubMed Abstract | Publisher Full Text | Free Full Text
28. Fernández-Castilla B, Declercq L, Jamshidi L, et al.: Visual representations of meta-analyses of multiple outcomes: extensions to forest plots, funnel plots, and caterpillar plots. Methodology. 2020 Dec 22; 16(4): 299–315. Publisher Full Text
29. Egger M, Davey Smith G, Schneider M, et al.: Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997 Sep 13; 315(7109): 629–634. PubMed Abstract | Publisher Full Text | Free Full Text
30. Peters JL, Sutton AJ, Jones DR, et al.: Comparison of two methods to detect publication bias in meta-analysis. JAMA. 2006 Oct 22; 295(6): 676–680. Publisher Full Text
31. Tang JL, Liu JL: Misleading funnel plot for detection of bias in meta-analysis. J. Clin. Epidemiol. 2000 May; 53(5): 477–484. Reference Source
32. The Cochrane Collaboration: Review Manager (RevMan).2022.
33. Bandyopadhyay T, Kumar A, Saili A, et al.: Distribution, antimicrobial resistance and predictors of mortality in neonatal sepsis. J. Neonatal-Perinatal Med. 2018 Jul 5; 11(2): 145–153. PubMed Abstract | Publisher Full Text
34. Nyenga AM, Mukuku O, Mutombo AK, et al.: Predictors of mortality in neonatal sepsis in a resource-limited setting. J. Adv. Pediatr. Child Health. 2021 Jun 16; 4(1): 057–061. Publisher Full Text
35. Goh GL, Lim CSE, Sultana R, et al.: Risk factors for mortality from late-onset sepsis among preterm very-low-birthweight infants: a single-center cohort study from Singapore. Front. Pediatr. 2022 Jan 31; 9(January): 1–9. Publisher Full Text
36. Rasyida IA, Pratama DC, Chamida FM: Prognostic factors of neonatal sepsis mortality in developing country. Pediatr. Infect. Vaccine. 2023; 30(1): 12. Publisher Full Text
37. ElHaie OMA, El Sabbagh GG, Elsayed A s: Predictors of mortality in outborns with neonatal sepsis: a prospective observational study. Benha Med. J. 2023; 20(20): 216–222.
38. Harum NA, Utomo M, Aditiawarman, et al.: The correlation between Apgar score and gestational age with neonatal sepsis and associated mortality. Jurnal Widya Medika. 2021; 7(2): 141–156. Reference Source
39. Meshram R, Gajimwar V, Bhongade S: Predictors of mortality in outborns with neonatal sepsis: a prospective observational study. Niger. Postgrad. Med. J. 2019; 26(4): 216–222. PubMed Abstract | Publisher Full Text
40. Gosai D, Shah BH, Jyothi S: Predictors of mortality in neonatal septicemia in a tertiary care centre. Int. J. Contemp. Pediatrics. 2020 Sep 21; 7(10): 2037. Publisher Full Text Reference Source
41. Tareen Z, Jirapradittha J, Sirivichayakul C, et al.: Factors associated with mortality outcomes in neonatal septicemia in Srinagarind hospital, Thailand. J. Med. Sci. Clin. Res. 2017 Aug 19; 05(08): 26672–26678. Reference Source
42. Gomella TL, Eyal FG, Bany-Mohammed F: Gomella’s neonatology: management, procedures, on-call problems, diseases, and drugs. 8th ed.New York: McGraw Hill; 2020; pp. 1175–1189.
43. Stoll BJ, Hansen N, Fanaroff AA, et al.: Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics. 2002 Aug 1; 110(2 Pt 1): 285–291. Publisher Full Text Reference Source
44. Dawodu A, al Umran K , Twum-Danso K: A case control study of neonatal sepsis: experience from Saudi Arabia. J. Trop. Pediatr. 1997 Apr 1; 43(2): 84–88. PubMed Abstract | Publisher Full Text
45. Agarwal R, Deorari A, Vinod P, et al.: AIIMS protocols in neonatology. 2nd ed.Delhi: Noble Vision; 2019.
46. Jumah DS, Hassan MK: Predictors of mortality outcome in neonatal sepsis. Med. J. Basrah Univ. 2007 Jun 28; 25(1): 11–18. Publisher Full Text Reference Source
47. Gudayu TW: Proportion and factors associated with low fifth minute Apgar score among singleton newborn babies in Gondar University referral hospital; North West Ethiopia. Afr. Health Sci. 2017 Mar; 17(1): 1–6. PubMed Abstract | Publisher Full Text | Free Full Text
48. Greco E, Lupia E, Bosco O, et al.: Platelets and multi-organ failure in sepsis. Int. J. Mol. Sci. 2017 Oct 20; 18(10): 2200. PubMed Abstract | Publisher Full Text | Free Full Text Reference Source
49. Iliodromiti S, Mackay DF, Smith GCS, et al.: Apgar score and the risk of cause-specific infant mortality: a population-based cohort study. Lancet. 2014 Nov 15; 384(9956): 1749–1755. PubMed Abstract | Publisher Full Text
50. Hornik CP, Fort P, Clark RH, 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 May; 88(Suppl 2): S69–S74. PubMed Abstract | Publisher Full Text | Free Full Text
51. Utomo MT: Neonatal sepsis in low birth weight infants in Dr. Soetomo general hospital. Indonesian J. Trop. Infect. Disease. 2010 May 3; 1(2): 86. Publisher Full Text Reference Source
52. Meshram RM, Kadu KS: Risk factors of mortality in septicemic extramural neonates: an experience from resource limited setting of central India. Paediatr. Indones. 2024 Apr 4; 64(3): 209–217. Publisher Full Text Reference Source
53. Zakariya BP, Bhat BV, Harish BN, et al.: Risk factors and predictors of mortality in culture proven neonatal sepsis. Indian J. Pediatr. 2012 Mar 14; 79(3): 358–361. PubMed Abstract | Publisher Full Text
54. Shaw CK, Shaw P, Malla T, et al.: The clinical spectrum and outcome of neonatal sepsis in a neonatal intensive care unit at a tertiary care hospital in Western Nepal: january 2000 to december 2005 - A retrospective study. ‎Eastern J. Med. 2012; 17(3): 119–125.
55. Ogunlesi TA, Ogunfowora OB: Predictors of mortality in neonatal septicemia in an underresourced setting. J. Natl. Med. Assoc. 2010 Oct; 102(10): 915–922. PubMed Abstract | Publisher Full Text
56. Miranda S, Harahap A, Husada D, et al.: Additional File 1. figshare. 2024 [cited 2024 Nov 13]. Reference Source
57. Miranda S, Harahap A, Husada D, et al.: Figure 1 Flowchart of the literature selection process. Figshare. 2024 [cited 2024 Nov 14]. Reference Source
58. Miranda S, Harahap A, Husada D, et al.: Characteristics of the studies. figshare. 2024 [cited 2024 Nov 14]. Reference Source
59. Miranda S, Harahap A, Husada D, et al.: PRISMA_2020_checklist_Miranda. figshare. 2024 [cited 2024 Nov 14]. Reference Source

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 19 Dec 2024

Author details Author details

Aminuddin Harahap
Roles: Resources, Supervision, Writing – Review & Editing

Dominicus Husada
Roles: Conceptualization, Formal Analysis, Methodology, Resources, Supervision, Validation, Writing – Review & Editing

Muhammad Reza
Roles: Formal Analysis, Methodology, Resources, Supervision, Validation, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

The author(s) declared that no grants were involved in supporting this work.

Article Versions (2)

version 2

Revised

Published: 23 Jul 2025, 13:1532

https://doi.org/10.12688/f1000research.158689.2

version 1

Published: 19 Dec 2024, 13:1532

https://doi.org/10.12688/f1000research.158689.1

© 2025 Miranda S et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

	Views	Downloads
F1000Research	-	-
PubMed Central Data from PMC are received and updated monthly.	-	-

Citations

SEE MORE DETAILS

CITE

how to cite this article

Miranda S, Harahap A, Husada D and Reza M. Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis [version 2; peer review: 1 approved with reservations]. F1000Research 2025, 13:1532 (https://doi.org/10.12688/f1000research.158689.2)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Version 1

VERSION 1

PUBLISHED 19 Dec 2024

Views

Reviewer Report 28 May 2025

Samuel R Neal, The University of Edinburgh College of Medicine and Veterinary Medicine,, Edinburgh, UK

Approved with Reservations

https://doi.org/10.5256/f1000research.174322.r383246

Summary

In this systematic review and meta-analysis, the authors aimed to determine risk factors associated with mortality from neonatal sepsis and estimate their effect sizes. They included 12 studies in their final analysis and identified nine risk factors associated with increased risk of mortality from neonatal sepsis. This is a relevant and important research question, and literature syntheses are increasingly needed given the volume of neonatal sepsis research.

I caveat this review with the fact that I am not an expert in the statistical methods of meta-analysis, and therefore I have not reviewed the individual statistical models. However, I have several major concerns from other aspects of this review.

Major issues

The search methodology appears robust as described, and agrees with the published protocol on PROSPERO. However, I am sceptical about some of the authors findings.

From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?
There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.
How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Other issues

It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?
How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.
Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?
Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?
Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?
The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?
I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.
On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?
On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Partly
Are sufficient details of the methods and analysis provided to allow replication by others?

Partly
Is the statistical analysis and its interpretation appropriate?

Partly
Are the conclusions drawn adequately supported by the results presented in the review?

Partly
If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

References

1. Borenstein M, Higgins JP, Hedges LV, Rothstein HR: Basics of meta-analysis: I2 is not an absolute measure of heterogeneity.Res Synth Methods. 2017; 8 (1): 5-18 PubMed Abstract | Publisher Full Text
2. Vizcarra-Jiménez D, Copaja-Corzo C, Hueda-Zavaleta M, Parihuana-Travezaño EG, et al.: Predictors of Death in Patients with Neonatal Sepsis in a Peruvian Hospital.Trop Med Infect Dis. 2022; 7 (11). PubMed Abstract | Publisher Full Text

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Neonatal sepsis, global health, diagnostic tests, clinical prediction modelling, systematic reviews

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Author Response 07 Aug 2025

Stefani Miranda, Drs.Titus Uly Kupang Police Hospital, Jalan Nangka 84, Kupang, East Nusa Tenggara, 85112, Indonesia

07 Aug 2025

Author Response

Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved ... Continue reading Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved our manuscript. We have incorporated the suggested changes and provided detailed responses to each comment as follows:

Major issues

Reviewer response: From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?

Author response: We had already contacted the study authors and waited to receive their response for a period of time, but we were still unable to obtain the full versions of the articles.

Reviewer response: There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.

Author response: Our search did not focus on low/middle-income (LMIC) or resource-constrained countries. It was simply a coincidence that the studies analyzed were predominantly from LMICs. Other studies were excluded from the meta-analysis because they did not match the study's eligibility criteria.

Reviewer response: How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Author response: We solely included clinical risk factors to reduce the study's heterogeneity.

Other issues

Reviewer response: It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?

Author response: This is the first meta-analysis to highlight risk factors for neonatal sepsis-related mortality. This meta-analysis provides information on clinical factors that lead to mortality in neonates with sepsis. Therefore, whenever we encounter such a situation, we can make decisions to prevent deterioration and reduce the mortality. The prior studies on neonatal sepsis and/or mortality were original research that is analyzed in this meta-analysis.

Reviewer response: How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.

Author response: The definition of neonatal sepsis in this meta-analysis is in accordance with each individual study's definition.

Reviewer response: Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?

Author response: Individual risk factor definitions were also in accordance with the definitions used in each individual study. In meta-analysis, we do not modify or directly interfere with the definitions used in individual research. However, it interprets, standardizes, or harmonizes them in order to make studies comparable, which can have an impact on how the results are sorted and understood. The degrees of heterogeneity in this study were assessed using I² statistics.

Reviewer response: Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?

Author response: We chose the 2014-time limit for our searches because we only wanted to include the previous ten years of original research in our analysis to keep their novelty.

Reviewer response: Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?

Author response: The meta-analysis was completed and published in 2024. We did not include more literature because this is not a living systematic review and meta-analysis.

Reviewer response: The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?

Author response: For other databases that did not use MeSH terms, we used keywords and their variations, as described in the 'Data sources and search strategy' section of this study.

Reviewer response: I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.

Author response: Thank you for your constructive feedback.

Reviewer response: On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?

Author response: We apologize for the mistake. We have already revised it to the correct one. We imply "a p-value of significance was set at P < 0.05."

Reviewer response: On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Author response: Thank you for the advice. We have already revised it accordingly.

Please inform us if further revisions are necessary. Thank you.
Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved our manuscript. We have incorporated the suggested changes and provided detailed responses to each comment as follows:

Major issues

Reviewer response: From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?

Author response: We had already contacted the study authors and waited to receive their response for a period of time, but we were still unable to obtain the full versions of the articles.

Reviewer response: There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.

Author response: Our search did not focus on low/middle-income (LMIC) or resource-constrained countries. It was simply a coincidence that the studies analyzed were predominantly from LMICs. Other studies were excluded from the meta-analysis because they did not match the study's eligibility criteria.

Reviewer response: How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Author response: We solely included clinical risk factors to reduce the study's heterogeneity.

Other issues

Reviewer response: It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?

Author response: This is the first meta-analysis to highlight risk factors for neonatal sepsis-related mortality. This meta-analysis provides information on clinical factors that lead to mortality in neonates with sepsis. Therefore, whenever we encounter such a situation, we can make decisions to prevent deterioration and reduce the mortality. The prior studies on neonatal sepsis and/or mortality were original research that is analyzed in this meta-analysis.

Reviewer response: How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.

Author response: The definition of neonatal sepsis in this meta-analysis is in accordance with each individual study's definition.

Reviewer response: Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?

Author response: Individual risk factor definitions were also in accordance with the definitions used in each individual study. In meta-analysis, we do not modify or directly interfere with the definitions used in individual research. However, it interprets, standardizes, or harmonizes them in order to make studies comparable, which can have an impact on how the results are sorted and understood. The degrees of heterogeneity in this study were assessed using I² statistics.

Reviewer response: Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?

Author response: We chose the 2014-time limit for our searches because we only wanted to include the previous ten years of original research in our analysis to keep their novelty.

Reviewer response: Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?

Author response: The meta-analysis was completed and published in 2024. We did not include more literature because this is not a living systematic review and meta-analysis.

Reviewer response: The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?

Author response: For other databases that did not use MeSH terms, we used keywords and their variations, as described in the 'Data sources and search strategy' section of this study.

Reviewer response: I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.

Author response: Thank you for your constructive feedback.

Reviewer response: On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?

Author response: We apologize for the mistake. We have already revised it to the correct one. We imply "a p-value of significance was set at P < 0.05."

Reviewer response: On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Author response: Thank you for the advice. We have already revised it accordingly.

Please inform us if further revisions are necessary. Thank you.
Competing Interests: We do not have any competing interests. Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 07 Aug 2025

Stefani Miranda, Drs.Titus Uly Kupang Police Hospital, Jalan Nangka 84, Kupang, East Nusa Tenggara, 85112, Indonesia

07 Aug 2025

Author Response

Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved ... Continue reading Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved our manuscript. We have incorporated the suggested changes and provided detailed responses to each comment as follows:

Major issues

Reviewer response: From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?

Author response: We had already contacted the study authors and waited to receive their response for a period of time, but we were still unable to obtain the full versions of the articles.

Reviewer response: There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.

Author response: Our search did not focus on low/middle-income (LMIC) or resource-constrained countries. It was simply a coincidence that the studies analyzed were predominantly from LMICs. Other studies were excluded from the meta-analysis because they did not match the study's eligibility criteria.

Reviewer response: How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Author response: We solely included clinical risk factors to reduce the study's heterogeneity.

Other issues

Reviewer response: It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?

Author response: This is the first meta-analysis to highlight risk factors for neonatal sepsis-related mortality. This meta-analysis provides information on clinical factors that lead to mortality in neonates with sepsis. Therefore, whenever we encounter such a situation, we can make decisions to prevent deterioration and reduce the mortality. The prior studies on neonatal sepsis and/or mortality were original research that is analyzed in this meta-analysis.

Reviewer response: How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.

Author response: The definition of neonatal sepsis in this meta-analysis is in accordance with each individual study's definition.

Reviewer response: Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?

Author response: Individual risk factor definitions were also in accordance with the definitions used in each individual study. In meta-analysis, we do not modify or directly interfere with the definitions used in individual research. However, it interprets, standardizes, or harmonizes them in order to make studies comparable, which can have an impact on how the results are sorted and understood. The degrees of heterogeneity in this study were assessed using I² statistics.

Reviewer response: Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?

Author response: We chose the 2014-time limit for our searches because we only wanted to include the previous ten years of original research in our analysis to keep their novelty.

Reviewer response: Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?

Author response: The meta-analysis was completed and published in 2024. We did not include more literature because this is not a living systematic review and meta-analysis.

Reviewer response: The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?

Author response: For other databases that did not use MeSH terms, we used keywords and their variations, as described in the 'Data sources and search strategy' section of this study.

Reviewer response: I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.

Author response: Thank you for your constructive feedback.

Reviewer response: On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?

Author response: We apologize for the mistake. We have already revised it to the correct one. We imply "a p-value of significance was set at P < 0.05."

Reviewer response: On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Author response: Thank you for the advice. We have already revised it accordingly.

Please inform us if further revisions are necessary. Thank you.
Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved our manuscript. We have incorporated the suggested changes and provided detailed responses to each comment as follows:

Major issues

Reviewer response: From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?

Author response: We had already contacted the study authors and waited to receive their response for a period of time, but we were still unable to obtain the full versions of the articles.

Reviewer response: There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.

Author response: Our search did not focus on low/middle-income (LMIC) or resource-constrained countries. It was simply a coincidence that the studies analyzed were predominantly from LMICs. Other studies were excluded from the meta-analysis because they did not match the study's eligibility criteria.

Reviewer response: How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Author response: We solely included clinical risk factors to reduce the study's heterogeneity.

Other issues

Reviewer response: It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?

Author response: This is the first meta-analysis to highlight risk factors for neonatal sepsis-related mortality. This meta-analysis provides information on clinical factors that lead to mortality in neonates with sepsis. Therefore, whenever we encounter such a situation, we can make decisions to prevent deterioration and reduce the mortality. The prior studies on neonatal sepsis and/or mortality were original research that is analyzed in this meta-analysis.

Reviewer response: How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.

Author response: The definition of neonatal sepsis in this meta-analysis is in accordance with each individual study's definition.

Reviewer response: Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?

Author response: Individual risk factor definitions were also in accordance with the definitions used in each individual study. In meta-analysis, we do not modify or directly interfere with the definitions used in individual research. However, it interprets, standardizes, or harmonizes them in order to make studies comparable, which can have an impact on how the results are sorted and understood. The degrees of heterogeneity in this study were assessed using I² statistics.

Reviewer response: Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?

Author response: We chose the 2014-time limit for our searches because we only wanted to include the previous ten years of original research in our analysis to keep their novelty.

Reviewer response: Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?

Author response: The meta-analysis was completed and published in 2024. We did not include more literature because this is not a living systematic review and meta-analysis.

Reviewer response: The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?

Author response: For other databases that did not use MeSH terms, we used keywords and their variations, as described in the 'Data sources and search strategy' section of this study.

Reviewer response: I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.

Author response: Thank you for your constructive feedback.

Reviewer response: On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?

Author response: We apologize for the mistake. We have already revised it to the correct one. We imply "a p-value of significance was set at P < 0.05."

Reviewer response: On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Author response: Thank you for the advice. We have already revised it accordingly.

Please inform us if further revisions are necessary. Thank you.
Competing Interests: We do not have any competing interests. Close
Report a concern

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 19 Dec 2024

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1
Version 2 (revision) 23 Jul 25
Version 1 19 Dec 24	read

Samuel R Neal, The University of Edinburgh College of Medicine and Veterinary Medicine,, Edinburgh, UK

Comments on this article

All Comments(0)

Add a comment

Browse by related subjects

Back to all reports

Reviewer Report

14 Views

28 May 2025 | for Version 1

Samuel R Neal, The University of Edinburgh College of Medicine and Veterinary Medicine,, Edinburgh, UK

14 Views Cite this report Responses(1)

Approved With Reservations

From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?
There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.
How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Other issues

It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?
How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.
Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?
Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?
Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?
The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?
I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.
On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?
On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Are the rationale for, and objectives of, the Systematic Review clearly stated?

Partly
Are sufficient details of the methods and analysis provided to allow replication by others?

Partly
Is the statistical analysis and its interpretation appropriate?

Partly
Are the conclusions drawn adequately supported by the results presented in the review?

Partly
If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)

Not applicable

References

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Neonatal sepsis, global health, diagnostic tests, clinical prediction modelling, systematic reviews

Respond to this report

Responses (1)

Author Response

07 Aug 2025

Stefani Miranda, Drs.Titus Uly Kupang Police Hospital, Jalan Nangka 84, Kupang, East Nusa Tenggara, 85112, Indonesia

Point-by-point response to Reviewer’s comments

We would like to express our gratitude to the Reviewer for the constructive and insightful comments. We believe that these comments have significantly improved our manuscript. We have incorporated the suggested changes and provided detailed responses to each comment as follows:

Major issues

Reviewer response: From Figure 1, 2259 records were excluded for having no abstracts and/or full texts available. This is two-thirds of all records that were sought for retrieval, which seems exceptionally high. Are you able to provide more details as to why this is the case and the challenges you faced? Were study authors contacted to obtain missing information as stated in your protocol?

Author response: We had already contacted the study authors and waited to receive their response for a period of time, but we were still unable to obtain the full versions of the articles.

Reviewer response: There is no mention in your review that your aims were limited to low/middle-income (LMIC) or resource-constrained countries, or that you limited your searches to only include studies published in this context. However, except Singapore, all 12 included studies were conducted in LMICs. I find this surprising, particularly given the breath of literature on many neonatal sepsis-related topics in high-income countries and the frequent lack of robust evidence in neonatal sepsis from LMICs. Can you confirm that your searches were not limited to LMICs and that you did not identify any studies on risk factors for mortality in neonatal sepsis from high-income countries except Singapore? Regardless, this is worth inclusion in your discussion.

Author response: Our search did not focus on low/middle-income (LMIC) or resource-constrained countries. It was simply a coincidence that the studies analyzed were predominantly from LMICs. Other studies were excluded from the meta-analysis because they did not match the study's eligibility criteria.

Reviewer response: How was a ‘risk factor’ defined and were all risk factors mentioned in each study included or did you only include selected risk factors (and if so, how were these selected)? I haven’t reviewed all included studies individually but, for example, the study by Vizcarra-Jiménez et al. (2022) also identified thrombocytopenia, elevated creatinine levels, and lack of exclusive breastfeeding as risk factors for mortality in sepsis. Why were these risk factors not included in your review?

Author response: We solely included clinical risk factors to reduce the study's heterogeneity.

Other issues

Reviewer response: It would be helpful to contextualise this review in terms of the current literature. For example, what work (if any) has already been done by other authors to synthesise risk factors for mortality from neonatal sepsis? Have previous narrative and/or systematic reviews already been published? Why is your current systematic review necessary and what does it add to the current literature?

Author response: This is the first meta-analysis to highlight risk factors for neonatal sepsis-related mortality. This meta-analysis provides information on clinical factors that lead to mortality in neonates with sepsis. Therefore, whenever we encounter such a situation, we can make decisions to prevent deterioration and reduce the mortality. The prior studies on neonatal sepsis and/or mortality were original research that is analyzed in this meta-analysis.

Reviewer response: How was neonatal sepsis defined for the purpose of your review? Did you include studies that used any definition of sepsis (as defined by the individual study authors), or did you only include studies with culture-positive sepsis (or any other specific criteria). This is not mentioned in the manuscript or additional files but is critically important to the interpretation of your findings, especially given the lack of a consensus definition for neonatal sepsis. It also feeds directly into your risk of bias assessment (where the items include “exposure measured in a standard, valid and reliable way” and “same criteria used for identification of cases and controls”.) It would be helpful to include these definitions in the ‘Characteristics of the studies’ table and to discuss them in the main body.

Author response: The definition of neonatal sepsis in this meta-analysis is in accordance with each individual study's definition.

Reviewer response: Similarly, you do not describe how the individual risk factors were defined and measured in each study. For example, lethargy and poor feeding are highly subjective and neonatal seizures can be difficult to distinguish from other movements. Were the definitions of individual risk factors the same or different between studies and how might this influence your findings?

Author response: Individual risk factor definitions were also in accordance with the definitions used in each individual study. In meta-analysis, we do not modify or directly interfere with the definitions used in individual research. However, it interprets, standardizes, or harmonizes them in order to make studies comparable, which can have an impact on how the results are sorted and understood. The degrees of heterogeneity in this study were assessed using I² statistics.

Reviewer response: Why did you choose the lower time limit of 2014 for your searches? Did something change in 2014 that could be expected to influence the results, or was this an arbitrary, pragmatic decision?

Author response: We chose the 2014-time limit for our searches because we only wanted to include the previous ten years of original research in our analysis to keep their novelty.

Reviewer response: Similarly, the searches in this review are now two years old. Was any consideration given to updating the searches before publication? Are you aware of any additional studies that have since been published?

Author response: The meta-analysis was completed and published in 2024. We did not include more literature because this is not a living systematic review and meta-analysis.

Reviewer response: The search strategy you provide in Additional File 1 only includes MeSH terms. How was this strategy adapted to other databases, for example Google Scholar, which does not use MeSH terms?

Author response: For other databases that did not use MeSH terms, we used keywords and their variations, as described in the 'Data sources and search strategy' section of this study.

Reviewer response: I’m not sure you have correctly interpreted the I-squared statistic. I-squared measures the proportion of total variability that is due to between-study heterogeneity, rather than the magnitude of any between-study heterogeneity itself. See, for example: https://doi.org/10.1002/jrsm.1230. As mentioned earlier, I am not an expert in the statistical methods of meta-analysis and am happy to be corrected by someone more expert.

Author response: Thank you for your constructive feedback.

Reviewer response: On page 4 (‘Statistical analysis’) you write: “A p-value of less significance was set at P < 0.05". What do you mean by a ‘p-value of less significance’ or is this an error in the direction of the inequality symbol (i.e. p < 0.05 is significant)?

Author response: We apologize for the mistake. We have already revised it to the correct one. We imply "a p-value of significance was set at P < 0.05."

Reviewer response: On page 4 (‘Statistical analysis’) you write: “We used the odds ratio to determine the correlation…”. I think this would be better worded as “association”, as odds ratios measure the strength of association between an exposure and outcome, rather than correlation, which has a specific statistical definition.

Author response: Thank you for the advice. We have already revised it accordingly.

Please inform us if further revisions are necessary. Thank you.

View more View less

Competing Interests

We do not have any competing interests.

Alongside their report, reviewers assign a status to the article:

Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions

[1] 1. Odabasi IO, Bulbul A: Neonatal sepsis. Sisli Etfal Hastan. Tıp Bul. 2020; 54(2): 142–158. PubMed Abstract | Publisher Full Text

[2] 2. Pace E, Yanowitz T: Infections in the NICU: neonatal sepsis. Semin. Pediatr. Surg. 2022 Aug 1; 31(4): 151200. Publisher Full Text

[3] 3. Salim RF, Sobeih AA, Abd El Kareem HM: Evaluation of the clinical value of circulating miR-101, miR-187 and miR-21 in neonatal sepsis diagnosis and prognosis. Egypt. J. Med. Hum. Genet. 2020 Dec 9; 21(1): 12. Publisher Full Text

[4] 4. Chaudhry S, Haroon F, Irfan Waheed KA, et al.: Blood lactate levels and lactate clearance as predictors of mortality in neonatal sepsis. J. Ayub Med. Coll. Abbottabad. 2022; 34(3): 438–441. Publisher Full Text Reference Source

[5] 5. Jajoo M, Kapoor K, Garg L, et al.: To study the incidence and risk factors of early onset neonatal sepsis in an out born neonatal intensive care unit of India. J. Clin. Ne onatol. 2015; 4(2): 91. Publisher Full Text

[6] 6. Bharad RV, Singh CS, Singh LR: Risk factors and immediate outcome of early onset neonatal sepsis. J. Med. Sci. Clin. Res. 2017 Apr 30; 05(04): 21050–21056. Publisher Full Text Reference Source

[7] 7. Nordberg V, Iversen A, Tidell A, et al.: A decade of neonatal sepsis caused by gram-negative bacilli-a retrospective matched cohort study. Eur. J. Clin. Microbiol. Infect. Dis. 2021 Sep 24; 40(9): 1803–1813. PubMed Abstract | Publisher Full Text | Free Full Text

[8] 8. Shane AL, Sánchez PJ, Stoll BJ: Neonatal sepsis. Lancet. 2017 Oct 14; 390(10104): 1770–1780. Publisher Full Text

[9] 9. López UOJ, Buriticá HHM: Lethality by neonatal sepsis, risk factors and microbiological characteristics. Andes Pediatr. 2021 Oct 2; 92(5): 690–698. PubMed Abstract | Publisher Full Text Reference Source

[10] 10. World Health Organization: Newborn mortality.2022 [cited 2023 Sep 20]. Reference Source

[11] 11. Zelellw DA, Dessie G, Worku Mengesha E, et al.: A systemic review and meta-analysis of the leading pathogens causing neonatal sepsis in developing countries. Ekwunife OI, editor. Biomed. Res. Int. 2021 Jun 5; 2021: 1–20. Publisher Full Text Reference Source

[12] 12. Bakhuizen SE, de Haan TR , Teune MJ, et al.: Meta-analysis shows that infants who have suffered neonatal sepsis face an increased risk of mortality and severe complications. Acta Paediatr. 2014 Dec; 103(12): 1211–1218. PubMed Abstract | Publisher Full Text

[13] 13. Kabwe M, Tembo J, Chilukutu L, et al.: Etiology, antibiotic resistance and risk factors for neonatal sepsis in a large referral center in Zambia. Pediatr. Infect. Dis. J. 2016 Jul; 35(7): e191–e198. PubMed Abstract | Publisher Full Text

[14] 14. Wynn JL, Wong HR, Shanley TP, et al.: Time for a neonatal-specific consensus definition for sepsis. Pediatr. Crit. Care Med. 2014 Jul; 15(6): 523–528. PubMed Abstract | Publisher Full Text | Free Full Text

[15] 15. Vizcarra-Jiménez D, Copaja-Corzo C, Hueda-Zavaleta M, et al.: Predictors of Death in Patients with Neonatal Sepsis in a Peruvian Hospital. Trop. Med. Infect. Dis. 2022 Oct 31; 7(11): 342. Publisher Full Text Reference Source

[16] 16. Kliegman RM, St. Geme JW III: Nelson textbook of pediatrics. 21st ed.Behrman RE, editor. Philadelphia: Elsevier Inc.; 2020; p. 53.

[17] 17. Bekele T, Merga H, Tesfaye T, et al.: Predictors of mortality among neonates hospitalized with neonatal sepsis: a case control study from southern Ethiopia. BMC Pediatr. 2022 Jan 3; 22(1): 1. PubMed Abstract | Publisher Full Text | Free Full Text

[18] 18. Jovičić M, Milosavljević MN, Folić M, et al.: Predictors of mortality in early neonatal sepsis: a single-center experience. Medicina (B Aires). 2023 Mar 18; 59(3): 604. PubMed Abstract | Publisher Full Text | Free Full Text Reference Source

[19] 19. Bohanon FJ, Lopez ON, Adhikari D, et al.: Race, income and insurance status affect neonatal sepsis mortality and healthcare resource utilization. Pediatr. Infect. Dis. J. 2018 Jul; 37(7): e178–e184. PubMed Abstract | Publisher Full Text | Free Full Text

[20] 20. Page MJ, McKenzie JE, Bossuyt PM, et al.: The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar 29; 372: n71. Publisher Full Text

[21] 21. Pimsen A, Kao CY, Hsu ST, et al.: The effect of advance care planning intervention on hospitalization among nursing home residents: a systematic review and meta-analysis. J. Am. Med. Dir. Assoc. 2022 Sep; 23(9): 1448–1460.e1. PubMed Abstract | Publisher Full Text

[22] 22. Tufanaru C, Munn Z, Stephenson M, et al.: Fixed or random effects meta-analysis? Common methodological issues in systematic reviews of effectiveness. Int. J. Evid. Based Healthc. 2015 Sep; 13(3): 196–207. Publisher Full Text Reference Source

[23] 23. Higgins JP, Thompson SG, Deeks JJ, et al.: Measuring inconsistency in meta-analyses. BMJ (Online). 2003 Sep 7; 327(2): 557–560. Publisher Full Text

[24] 24. Huedo-Medina TB, Sánchez-Meca J, Marín-Martínez F, et al.: Assessing heterogeneity in meta-analysis: Q statistic or I 2 Index? Psychol. Methods. 2006 Jun; 11(2): 193–206. Publisher Full Text

[25] 25. Clephas PRD, Heesen M: Interpretation of meta-analyses. Interventional. Pain Med. 2022; 1(June): 100120. Publisher Full Text

[26] 26. Pei Y, Tian Q, Zhang L, et al.: Exploring the major sources and extent of heterogeneity in a genome-wide association meta-analysis. Ann. Hum. Genet. 2016 Mar 20; 80(2): 113–122. PubMed Abstract | Publisher Full Text | Free Full Text

[27] 27. Tawfik GM, Dila KAS, Mohamed MYF, et al.: A step by step guide for conducting a systematic review and meta-analysis with simulation data. Trop. Med. Health. 2019 Dec 1; 47(1): 46. PubMed Abstract | Publisher Full Text | Free Full Text

[28] 28. Fernández-Castilla B, Declercq L, Jamshidi L, et al.: Visual representations of meta-analyses of multiple outcomes: extensions to forest plots, funnel plots, and caterpillar plots. Methodology. 2020 Dec 22; 16(4): 299–315. Publisher Full Text

[29] 29. Egger M, Davey Smith G, Schneider M, et al.: Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997 Sep 13; 315(7109): 629–634. PubMed Abstract | Publisher Full Text | Free Full Text

[30] 30. Peters JL, Sutton AJ, Jones DR, et al.: Comparison of two methods to detect publication bias in meta-analysis. JAMA. 2006 Oct 22; 295(6): 676–680. Publisher Full Text

[31] 31. Tang JL, Liu JL: Misleading funnel plot for detection of bias in meta-analysis. J. Clin. Epidemiol. 2000 May; 53(5): 477–484. Reference Source

[32] 32. The Cochrane Collaboration: Review Manager (RevMan).2022.

[33] 33. Bandyopadhyay T, Kumar A, Saili A, et al.: Distribution, antimicrobial resistance and predictors of mortality in neonatal sepsis. J. Neonatal-Perinatal Med. 2018 Jul 5; 11(2): 145–153. PubMed Abstract | Publisher Full Text

[34] 34. Nyenga AM, Mukuku O, Mutombo AK, et al.: Predictors of mortality in neonatal sepsis in a resource-limited setting. J. Adv. Pediatr. Child Health. 2021 Jun 16; 4(1): 057–061. Publisher Full Text

[35] 35. Goh GL, Lim CSE, Sultana R, et al.: Risk factors for mortality from late-onset sepsis among preterm very-low-birthweight infants: a single-center cohort study from Singapore. Front. Pediatr. 2022 Jan 31; 9(January): 1–9. Publisher Full Text

[36] 36. Rasyida IA, Pratama DC, Chamida FM: Prognostic factors of neonatal sepsis mortality in developing country. Pediatr. Infect. Vaccine. 2023; 30(1): 12. Publisher Full Text

[37] 37. ElHaie OMA, El Sabbagh GG, Elsayed A s: Predictors of mortality in outborns with neonatal sepsis: a prospective observational study. Benha Med. J. 2023; 20(20): 216–222.

[38] 38. Harum NA, Utomo M, Aditiawarman, et al.: The correlation between Apgar score and gestational age with neonatal sepsis and associated mortality. Jurnal Widya Medika. 2021; 7(2): 141–156. Reference Source

[39] 39. Meshram R, Gajimwar V, Bhongade S: Predictors of mortality in outborns with neonatal sepsis: a prospective observational study. Niger. Postgrad. Med. J. 2019; 26(4): 216–222. PubMed Abstract | Publisher Full Text

[40] 40. Gosai D, Shah BH, Jyothi S: Predictors of mortality in neonatal septicemia in a tertiary care centre. Int. J. Contemp. Pediatrics. 2020 Sep 21; 7(10): 2037. Publisher Full Text Reference Source

[41] 41. Tareen Z, Jirapradittha J, Sirivichayakul C, et al.: Factors associated with mortality outcomes in neonatal septicemia in Srinagarind hospital, Thailand. J. Med. Sci. Clin. Res. 2017 Aug 19; 05(08): 26672–26678. Reference Source

[42] 42. Gomella TL, Eyal FG, Bany-Mohammed F: Gomella’s neonatology: management, procedures, on-call problems, diseases, and drugs. 8th ed.New York: McGraw Hill; 2020; pp. 1175–1189.

[43] 43. Stoll BJ, Hansen N, Fanaroff AA, et al.: Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics. 2002 Aug 1; 110(2 Pt 1): 285–291. Publisher Full Text Reference Source

[44] 44. Dawodu A, al Umran K , Twum-Danso K: A case control study of neonatal sepsis: experience from Saudi Arabia. J. Trop. Pediatr. 1997 Apr 1; 43(2): 84–88. PubMed Abstract | Publisher Full Text

[45] 45. Agarwal R, Deorari A, Vinod P, et al.: AIIMS protocols in neonatology. 2nd ed.Delhi: Noble Vision; 2019.

[46] 46. Jumah DS, Hassan MK: Predictors of mortality outcome in neonatal sepsis. Med. J. Basrah Univ. 2007 Jun 28; 25(1): 11–18. Publisher Full Text Reference Source

[47] 47. Gudayu TW: Proportion and factors associated with low fifth minute Apgar score among singleton newborn babies in Gondar University referral hospital; North West Ethiopia. Afr. Health Sci. 2017 Mar; 17(1): 1–6. PubMed Abstract | Publisher Full Text | Free Full Text

[48] 48. Greco E, Lupia E, Bosco O, et al.: Platelets and multi-organ failure in sepsis. Int. J. Mol. Sci. 2017 Oct 20; 18(10): 2200. PubMed Abstract | Publisher Full Text | Free Full Text Reference Source

[49] 49. Iliodromiti S, Mackay DF, Smith GCS, et al.: Apgar score and the risk of cause-specific infant mortality: a population-based cohort study. Lancet. 2014 Nov 15; 384(9956): 1749–1755. PubMed Abstract | Publisher Full Text

[50] 50. Hornik CP, Fort P, Clark RH, 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 May; 88(Suppl 2): S69–S74. PubMed Abstract | Publisher Full Text | Free Full Text

[51] 51. Utomo MT: Neonatal sepsis in low birth weight infants in Dr. Soetomo general hospital. Indonesian J. Trop. Infect. Disease. 2010 May 3; 1(2): 86. Publisher Full Text Reference Source

[52] 52. Meshram RM, Kadu KS: Risk factors of mortality in septicemic extramural neonates: an experience from resource limited setting of central India. Paediatr. Indones. 2024 Apr 4; 64(3): 209–217. Publisher Full Text Reference Source

[53] 53. Zakariya BP, Bhat BV, Harish BN, et al.: Risk factors and predictors of mortality in culture proven neonatal sepsis. Indian J. Pediatr. 2012 Mar 14; 79(3): 358–361. PubMed Abstract | Publisher Full Text

[54] 54. Shaw CK, Shaw P, Malla T, et al.: The clinical spectrum and outcome of neonatal sepsis in a neonatal intensive care unit at a tertiary care hospital in Western Nepal: january 2000 to december 2005 - A retrospective study. ‎Eastern J. Med. 2012; 17(3): 119–125.

[55] 55. Ogunlesi TA, Ogunfowora OB: Predictors of mortality in neonatal septicemia in an underresourced setting. J. Natl. Med. Assoc. 2010 Oct; 102(10): 915–922. PubMed Abstract | Publisher Full Text

[56] 56. Miranda S, Harahap A, Husada D, et al.: Additional File 1. figshare. 2024 [cited 2024 Nov 13]. Reference Source

[57] 57. Miranda S, Harahap A, Husada D, et al.: Figure 1 Flowchart of the literature selection process. Figshare. 2024 [cited 2024 Nov 14]. Reference Source

[58] 58. Miranda S, Harahap A, Husada D, et al.: Characteristics of the studies. figshare. 2024 [cited 2024 Nov 14]. Reference Source

[59] 59. Miranda S, Harahap A, Husada D, et al.: PRISMA_2020_checklist_Miranda. figshare. 2024 [cited 2024 Nov 14]. Reference Source

Risk factors for mortality due to neonatal sepsis: a systematic review and meta-analysis

Abstract

Background

Methods

Results

Conclusions

Keywords

Revised Amendments from Version 1

Background

Methods

Data sources and search strategy

Eligibility criteria

Study quality assessment

Data extraction

Statistical analysis

Results

Search results

Figure 1. Flowchart of the literature selection process.

Study characteristics

Figure 2. Risk of bias of included studies.

Mortality-related risk factors

Figure 3. Forest plots show the association between risk factors and neonatal sepsis mortality.

Heterogeneity, sensitivity analysis, and publication bias

Figure 4. Funnel plots of meta-analysis including.

Discussion

Conclusions

Declarations

Ethics approval and consent to participate

Consent for publication

Data availability statement

Underlying data

Extended data

Reporting guidelines

Acknowledgements

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Browse by related subjects

Competing Interests Policy

Stay Updated