Composition of early life leukocyte populations in preterm infants with and without late-onset sepsis.

BACKGROUND: Composition of leukocyte populations in the first month of life remains incompletely characterised, particularly in preterm infants who go on to develop late-onset sepsis (LOS). AIM: To characterise and compare leukocyte populations in preterm infants with and without LOS during the first month of life. STUDY
DESIGN: Single-centre prospective observational cohort study. PARTICIPANTS: Infants born <30 weeks gestational age (GA). OUTCOME MEASURES: Peripheral blood samples were collected at 1, 7, 14, 21 and 28 days of life. Leukocyte populations were characterised using 5-fluorophore-6-marker flow cytometry. Absolute leukocyte counts and frequency of total CD45+ leukocytes of each population were adjusted for GA, birth weight z-scores, sex and total leukocyte count.
RESULTS: Of 119 preterm infants enrolled, 43 (36%) had confirmed or clinical LOS, with a median onset at 13 days (range 6-26). Compared to infants without LOS, the adjusted counts and frequency of neutrophils, basophils and non-cytotoxic T lymphocytes were generally lower and immature granulocytes were higher over the first month of life in infants who developed LOS. Specific time point comparisons identified lower adjusted neutrophil counts on the first day of life in those infants who developed LOS more than a week later, compared to those without LOS, albeit levels were within the normal age-adjusted range. Non-cytotoxic T lymphocyte counts and/or frequencies were lower in infants following LOS on days 21 and 28 when compared to those who did not develop LOS.
CONCLUSION: Changes in non-cytotoxic T lymphocytes occurred following LOS suggesting sepsis-induced immune suppression.

Introduction

Late-onset sepsis (LOS; onset >72 hours of age) remains a frequent cause of morbidity and mortality in preterm infants (born <37 weeks gestation age (GA)) [1]. Risk is inversely related to GA and birth weight and LOS occurs in over 20% infants born ≤32 weeks GA [2]. Late-onset sepsis and the associated inflammation contribute to increased mortality and adverse outcomes, particularly impaired neurodevelopment [3].

Leukocytes play important roles in initiating inflammatory responses, stimulating complement pathways and activating adaptive immune responses. However, composition of early life leukocyte populations remain incompletely characterised, particularly in preterm infants who go on to develop LOS [4, 5]. Most data on neonatal leukocyte numbers and/or frequencies are derived from cord blood studies, from which immune parameters may be poorly reflective of postnatal immunity [6]. Moreover, leukocyte data are routinely obtained from manual blood films and automated blood count analysers and lack information on relative leukocyte frequencies [7, 8]. Consequently, there are few reports on leukocyte numbers and proportions preceding and following neonatal sepsis [4]. In this prospective observational study, we characterised the principal leukocyte populations during the first month of life in very preterm infants (born <30 weeks GA) with and without LOS. We hypothesised that infants who subsequently developed LOS would have differential leukocyte counts and proportions compared to uninfected infants, both preceding and following sepsis.

Methods

Study population

This prospective, longitudinal observational study of innate immune system development during the first month of life in very preterm infants born <30 weeks GA was approved by the Women and Newborn Health Services Human Research Ethics Committee at King Edward Memorial Hospital, Perth, Australia (1627/EW) [9-11]. Written, informed consent was obtained from parents prior to their child’s participation. Infants born at <30 weeks GA without major congenital anomalies were eligible for recruitment to this study between July 2009 and October 2011.

In accordance with local clinical guidelines, blood culture was performed if infants had clinical signs suggestive of LOS, including increased ventilation or oxygen requirements, lethargy, decreased perfusion, temperature instability, vomiting, feed intolerance, increased apnoea and/or bradycardia. LOS was defined as follows: ‘Confirmed LOS’—positive blood culture and a C-reactive protein (CRP) of >15 mg/L within 72 h of blood culture. ‘Clinical LOS’—negative blood culture and a CRP of >15 mg/L within 72 h of blood culture [9, 12]. ‘Any LOS’—infants with confirmed and clinical LOS. Infants with a Gram-positive blood culture (coagulase-negative staphylococci n = 2 and Enterococcus faecalis n = 1), 2–4 serial CRPs of <15 mg/L within 72 h of blood culture sampling and the absence of sustained clinical features of LOS were considered contaminants and analysed with infants that did not develop confirmed or clinical LOS (‘no LOS’).

Relevant demographic and clinical data until discharge from the neonatal intensive care unit were extracted from electronic patient databases. Placental histology was examined as part of routine care for infants born <30 weeks GA, using an adaption of a widely accepted semi-quantitative scoring system, as previously described [13].

Whole blood sampling

Peripheral venous blood samples (0.8 mL) were collected into Lithium-heparin tubes (BD Biosciences, North Ryde, NSW, Australia) on days 1 (n = 119), 7 (n = 111), 14 (n = 109), 21 (n = 111) and 28 (n = 111) of life and processed within 4 h of collection to characterise immune development using a number of assays [9-11].

Leukocyte immunostaining for flow cytometry

For this study, 50 μL whole blood samples were stained with leukocyte marker antibodies using an established 5-fluorophore-6-marker surface staining method, as previously described [14]. In brief, whole blood was incubated with a cocktail of monoclonal mouse anti-human antibodies: CD2-APC, CD36-PE, CD16-APC-H7, CD45-AmCyan, CD19-V450 and rat anti-human chemoattractant receptor-homologous molecule expressed on T helper type 2 (Th2) cells (CRTh2)-AF647 (all BD Biosciences). Red blood cells were lysed with BD FACS Lysing Solution before fixation in Stabilizing Fixative and transfer to TruCOUNT Tubes (all BD Biosciences).

Flow cytometry analysis

Flow cytometry analysis was performed as previously described [9]. Briefly, data was acquired in real-time on a triple-laser eight-colour FACSCanto II flow cytometer with FACSDiva software (BD Biosciences). Flow cytometer fluorescence intensities were monitored weekly with SPHERO Rainbow Calibration Beads (SheroTech Inc, Lake Forest, IL, USA). An established gating strategy, validated in neonatal cord blood and critically ill newborns, was used to identify principal leukocyte populations using FlowJo version 8 software (Tree Star Inc., Ashland, OR, USA) [Cytometry A. 2007 ">14-16]. Leukocytes were separated from debris using positivity for the common leukocyte antigen CD45. Through successive gating, the following populations were identified: neutrophils (SSChigh/CD16+), immature granulocytes (SSChigh/CD16-/CD2-/CRTh2-), eosinophils (SSChigh/CD16-/CRTh2+), basophils (SSCint/CD16-/CRTh2+), CD16- classical monocytes (SSCint/CD16-/CD2-/CRTH2-/CD19-/CD36+), CD16+ non-classical monocytes (SSCint/CD16+/CD2-/CRTH2-/CD19-/CD36+), B lymphocytes (SSClow/CD16-/CD2-/CRTH2-/CD36-/CD19+), non-cytotoxic (CD16-) T lymphocytes (SSClow/CD16-/CD2+/CRTH2+) and cytotoxic (CD16+) T lymphocytes and natural killer (NK) cells (SSClow/CD16+/CD2+/CRTH2+), as shown in S1 Fig in S1 File. Based on the surface markers used, we speculate that the neutrophil population consists mainly of mature neutrophils and less-mature banded cells (CD16dim), the immature granulocytes includes promyelocytes, myelocytes and metamyelocytes, the CD16+ cytotoxic T/NK lymphocytes contains mostly effector CD8+ T cells, and the CD16- non-cytotoxic T lymphocytes are mainly CD4+ T cells [17-23]. Other cells in the T cell compartment, such as γδ T and NKT cells are likely to be included in the non-cytotoxic and cytotoxic T/NK lymphocyte populations [24].

Statistical analysis

Continuous data were summarised as median and interquartile ranges (IQR, 25th-75th) and categorical data with frequency distributions. Group (any LOS or no LOS) allocation was made based on the whole study period (i.e. infants who developed LOS during the study period were assigned to the LOS group from Day 1). Univariate comparisons of demographic and clinical data between the any LOS and no LOS groups, and confirmed LOS and clinical LOS groups, were performed using the Mann-Whitney test for continuous outcomes and Fisher’s exact test for categorical outcomes. Absolute leukocyte counts were calculated according to BD TruCOUNT instructions by dividing the number of events in the population of interest by the total bead count and multiplying the number of beads per test, divided by the sample volume with dilutions accounted for [9]. Descriptive summaries, prior to analysis, of absolute leukocyte populations counts and frequency of total CD45+ leukocytes data are presented in S1 Table in S1 File. Leukocyte count and frequency data were transformed to the natural logarithm for analysis to fulfil the assumption of normality of the residuals. Mixed linear regression analysis using a random effects model and a group by time interaction was performed to assess the effect of LOS on leukocyte population levels, while accounting for the repeated measures at time points. All models were adjusted for known confounders, GA, birth weight z-score, sex and total leukocyte count [2]. For weekly analysis, post-hoc pairwise differences in adjusted counts between each time point were calculated within LOS and no LOS groups. Analogous analyses assessed temporal changes in leukocyte counts and frequencies, and the ratio of neutrophil counts to lymphocyte counts between groups. All post-hoc pairwise comparisons were Bonferroni-corrected to maintain an overall alpha error rate of 0.05. Model parameters were back transformed and presented graphically as geometric means and their 95% confidence intervals (CI). Stata (version 16, StataCorp, College Station, TX, USA) statistical software was used for data analysis. Prism 8 for macOS (GraphPad, La Jolla, CA, USA) was used to generate figures. P values <0.05 were considered statistically significant.

Results

Clinical characteristics of study cohort

Of the 129 very preterm infants recruited to this study, infants who suffered from early-onset sepsis (n = 4) or necrotising enterocolitis (n = 5) during the study period or had insufficient samples (n = 1) were excluded from analysis. Clinical characteristics of the analysed cohort (n = 119) are shown in Table 1. Consistent with known risk factors for LOS, preterm infants with any LOS (confirmed (n = 28) and clinical (n = 15)) were younger and smaller at birth and required more respiratory support compared to infants without LOS (n = 76; Table 1) [2]. The median age at the onset of LOS was postnatal day 13 (IQR 25th-75th 10–18 days; range 6–26 days). Three infants, all without LOS, died on days 2 and 3 due to pulmonary haemorrhage (n = 2) and extreme prematurity (n = 1). Data collected prior to death were included in the analysis and leukocyte populations were within the range of uninfected infants. Most of the confirmed LOS episodes (n = 23, 82.1%) were caused by Gram-positive bacteria, predominantly Coagulase-negative staphylococci (n = 19, 82.6%), followed by Enterococcus faecalis (n = 2), Staphylococcus aureus (n = 1) and Bacillus sphaericus (n = 1). Gram-negative infections (n = 5, 17.9%) were caused by Escherichia coli (n = 3), Enterobacter cloacae (n = 1) and Klebsiella pneumoniae (n = 1). The clinical characteristics of infants with confirmed and clinical LOS were similar except for a higher maximum CRP in the confirmed LOS group (S2 Table in S1 File).

Table 1

Basic clinical characteristics of the study cohort.

	Any LOS	No LOS	P value
	n = 43	n = 76
Gestational age (weeks)	26.0 (25.3–28.3)	27.7 (26.6–29.1)	0.0005
Birth weight (grams)	790 (660–955)	1010 (810–1240)	<0.0001
Birth weight z-score	-0.05 (-0.75–0.32)	0.21 (-0.54–0.64)	0.105
Males	22 (51.2)	43 (56.6)	0.702
Small for gestational age	4 (9.3)	4 (5.2)	0.455
Multiple birth	9 (20.9)	15 (19.7)	>0.999
Caesarean section	25 (58.1)	42 (55.3)	0.848
ROM	17 (39.5)	24 (31.6)	0.425
Antenatal steroids	40 (93.0)	74 (97.4)	0.350
Apgar <7 at 5 min	12 (27.9)	11 (14.5)	0.093
Chorioamnionitis^	18/40 (45.0)	31/62 (50.0)	0.687
Mechanical ventilation	41 (95.3)	69 (90.8)	0.485
Duration (hours)	306 (41–889)	21 (11–101)	<0.0001
CPAP	43 (100)	73 (96.1)	0.552
Duration (hours)	879 (643–1237)	761 (174–1130)	0.025
IVH (grade III/IV)	4 (9.3)	4 (5.3)	0.458
Chronic lung disease	15 (34.9)	12 (15.8)	0.023
Postnatal dexamethasone	8 (18.6)	3 (3.9)	0.005
ROP (stage III/IV)^	2/40 (5.0)	0/72 (0)	0.126
Length of NICU stay (days)	101 (69–136)	63 (49–84)	<0.0001
Mortality to 28 days	0 (0)	3 (4.0)	0.552
Age at LOS onset (days)	13 (10–18)	-	-
Highest CRP (mg/L)*	43 (25–73)	-	-

Data are expressed as median (IQR) or n (%), as appropriate.

ROM, rupture of membranes >24 hours prior to delivery; CPAP, continuous positive airway pressure; IVH, intraventricular haemorrhage; ROP, retinopathy of prematurity; CRP, C-reactive protein.

^from available reports.

* Within 72 hour of blood culture.

–data not available or applicable.

Leukocyte populations during the first month of life

The adjusted counts and frequencies of neutrophil, basophil, immature granulocytes and non-cytotoxic T cells were generally different between infants with and without LOS during the first month of life (Figs 1 and 2; S3 and S4 Tables in S1 File). Neutrophil counts and frequencies were unchanged between days 1–14 and then decreased thereafter in the no LOS group, but remained lower and unchanged over time in the any LOS group. Basophil counts and frequencies peaked by day 7 and decreased thereafter in the no LOS group, yet levels remained the same after increasing from days 1 to 7 in the any LOS group. Non-cytotoxic T lymphocyte counts and frequencies increased with age during the first month of life, but were lower in the any LOS group compared to the no LOS group. Immature granulocyte counts and frequencies were unchanged on days 1–7 in both the any and no LOS groups, but decreased thereafter in the no LOS group and remained higher in any LOS until day 28. The temporal trends in counts and frequencies of classical monocytes remained unchanged during the first month of life for both the any and no LOS groups (Figs 1 and 2; S3 and S4 Tables in S1 File). Non-classical monocytes, eosinophils and cytotoxic T/NK and B lymphocytes increased with age and were generally similar between the any and no LOS groups (Figs 1 and 2; S3 and S4 Tables in S1 File).

Fig 1

Absolute leukocyte counts.

The absolute counts of total leukocyte, granulocyte, monocyte and lymphocyte populations from peripheral blood of preterm infants with any LOS (confirmed and clinical; closed circle) and no LOS (open circle). Analysis was performed by mixed linear regression analysis using a random effects model that was adjusted for known confounders, GA, birth weight z-score, sex and total leukocyte count. Data shown as geometric mean with 95% confidence intervals. Arrows indicate the median age of LOS onset of 13 days in infants in the any LOS group. *Comparison between preterm infants with any LOS and no LOS at the specified time point; level of significance indicated by symbols, e.g. *P < 0.05.

Fig 2

Leukocyte frequency.

The percentage of total CD45+ granulocyte, monocyte and lymphocyte populations from preterm infants with any LOS (confirmed and clinical; closed circle) and no LOS (open circle). Analysis was performed by mixed linear regression analysis using a random effects model that was adjusted for known confounders, GA, birth weight z-score, sex and total leukocyte count. Data shown as geometric mean with 95% confidence intervals. Arrows indicated the median age of LOS onset of 13 days in infants in the any LOS group. *Comparison between preterm infants with any LOS and no LOS at the specified time point; level of significance indicated by symbols, e.g. *P < 0.05.

Leukocyte populations at each time point

Neutrophil and non-cytotoxic T lymphocyte counts and/or frequencies were associated with LOS. Adjusted neutrophil counts were lower in infants with any LOS than those with no LOS on day 1 (mean 1.57x109/L vs 2.26x109/L, p = 0.011), but were similar at all other time points (Figs 1 and 2). Non-cytotoxic T lymphocyte counts and/or frequencies were lower in the any LOS group compared to the no LOS group on days 21 (mean 1.17x109/L vs 1.47x109/L, p = 0.048; 22.3% vs 27.7%, p = 0.021) and 28 (mean 23.9% vs 30.3%, p = 0.010; Figs 1 and 2). In contrast, counts and frequencies of immature granulocytes, basophils, eosinophils, classical and non-classical monocytes and cytotoxic T/NK and B lymphocytes were similar between the no LOS and any LOS groups at all time points (Figs 1 and 2).

Neutrophil-to-lymphocyte count ratio

There were no differences in the ratios of neutrophils-to-all-lymphocytes and to lymphocyte sub-populations between the any LOS and no LOS groups (Fig 3).

Fig 3

Neutrophil-to-lymphocyte ratios.

The ratio of neutrophils to a) all lymphocytes, b) non-cytotoxic T lymphocytes, c) B lymphocytes, and d) cytotoxic T/NK lymphocytes, from preterm infants with any LOS (confirmed and clinical; closed circle) and no LOS (open circle). Analysis was performed by mixed linear regression analysis using a random effects model that was adjusted for known confounders, GA, birth weight z-score, sex and total leukocyte count. Data shown as geometric mean with 95% confidence intervals. Arrows indicate the median age of LOS onset of 13 days for infants in the any LOS group.

Confirmed and clinical LOS

Leukocyte populations and neutrophil-to-lymphocyte ratios were similar between infants with confirmed and clinical LOS except for an elevated frequency in non-cytotoxic T lymphocytes on day 7 in the confirmed LOS group (mean 24.6% vs 17.8%, p = 0.031).

Discussion

To our knowledge, this is the first study to characterise preterm infant peripheral blood principal leukocyte populations from a single sample taken weekly during the first month of life. We show that neutrophils, basophils and non-cytotoxic T lymphocytes were generally lower and immature granulocytes were higher over the first month of life in infants who developed LOS, compared to infants without LOS. Specific time point comparisons revealed that infants with LOS had significantly lower neutrophils on the first day of life preceding LOS and non-cytotoxic T lymphocytes were decreased following LOS.

We extend the findings of previous studies that examined the development of monocyte and basophil populations to one week of age by characterising these populations through the first month of life, a time when the risk of sepsis is the highest [25-28]. The counts/frequencies of the granulocyte and monocyte populations observed in our study are consistent with available previous data, obtained mainly from manual blood films and automated analyses, and some flow cytometry-based methods [8, 25, 28–32]. We found counts/frequencies of non-cytotoxic T lymphocytes remained relatively stable during the first month of life, whereas cytotoxic T/NK cells and B lymphocytes increased with age, similar findings to previous longitudinal studies that have characterised the T cell compartment in preterm infants [28, 32–34].

Leukocyte populations in infants with and without LOS

We report that infants who later developed LOS had lower neutrophil counts on day 1 of age, albeit within the normal age-adjusted range [29]. Neutropenia, compounded by functional impairments, has been linked to sub-optimal microbial colonisation and LOS susceptibility in murine models of sepsis [35-38]. However, in human infant studies a link between early-onset neutropenia and LOS has not been confirmed [39-41]. Our observation of lower neutrophil counts in infants who subsequently develop LOS is transient and potentially affected by maternal factors (e.g. pre-eclampsia, placental insufficiency) that are associated with early-onset neutropenia in newborns [41].

Basophil counts and frequencies were elevated until day 14 and declined thereafter in infants without LOS, whereas they remained consistently lower in those who developed LOS. The role of basophils in human LOS remains unclear, with most data collected from animal models. Basophil-deficient mice exhibit reduced bacterial clearance and increased mortality in sepsis models, indicating an important role for basophils in initiating an immune response to bacterial infection [42]. Our data of lower basophil counts in infants who developed LOS support a contribution of these cells to LOS susceptibility.

The elevated ratio of non-cytotoxic T to cytotoxic T/NK lymphocytes observed in our study is in keeping with the Th2-skewed response reported in early infancy [32, 33]. This skewing increases tolerance to pathogens, which is important for early life commensal colonisation, but may contribute to susceptibility to infection in infants [43]. Interestingly, we found that following LOS on days 21 and 28, preterm infants had reduced counts and frequencies of non-cytotoxic T lymphocytes compared to infants who did not develop LOS. Depleted circulating lymphocyte populations is a characteristic of sepsis-induced immunosuppression in adults, and is reported at post-mortem in preterm and term infants and children following fatal sepsis [44-46]. Taken together, these data suggest a degree of immune suppression following LOS in our cohort of preterm infants.

Compared to infants without LOS, immature granulocytes remained elevated near the time of LOS, a finding consistent with previous neonatal sepsis studies [23].

Neutrophil-to-lymphocyte ratio

An elevated ratio of neutrophils to lymphocytes is a proposed biomarker for systemic inflammation in neonatal sepsis with an area under the curve of 0.78, sensitivity of 0.73 and specificity of 0.79 from receiver operating characteristic analysis [47]. We found that the ratio of neutrophils to lymphocytes was similar between preterm infants with and without LOS, indicating this ratio is a poor predictor of LOS in our cohort. The discrepancy in the ratio of neutrophils to lymphocytes in septic and non-septic infants between our study and those published, suggests that elevated ratios may not necessarily be present in the days or weeks leading up to and after LOS, likely due to the rapid inflammatory changes in response to infection [9]. Additionally, our leukocyte counts were adjusted known risk factors for LOS, which may refine the difference in neutrophil-to-lymphocyte ratio between infants with and without LOS.

Study strengths and limitations

The strengths of this study include the longitudinal prospective sample collection through the first month of life and detailed analysis of leukocyte populations from peripheral blood in real time, both of which are unusual in preterm infant studies. Our analyses were adjusted for known risk factors for LOS [2]. We acknowledge that there are some limitations of the study, including the generalised classification of leukocytes and lack of specific populations, particularly in the T cell compartment (e.g. CD4, CD8, regulatory T cells and other innate-like T cells), that are known to play specific roles in the immune response to bacterial infection [31, 33, 34, 48, 49]. The sample size is modest, reflecting the difficulties of repeated standardised sampling in this population, which limits relating our laboratory findings to clinical outcomes and the influence of maternal factors and antibiotic therapy that may influence leukocyte populations. Moreover, the investigation into potential differences between Gram-positive and Gram-negative infections was limited by the modest number of infants with LOS particularly due to organisms other than Coagulase-negative staphylococci. Sample collection at the time of sepsis evaluation was not permitted due to ethical reasons, limiting analyses to weekly time points. The lack of functional data precludes a mechanistic interpretation of the relationship of leukocyte counts and frequencies to LOS. The correlation between the 5-fluorophore-6-marker flow cytometry method and manual/automated counts for some cell populations (e.g. basophils, monocytes) in blood samples from newborn infants is low and interpreting the counts in this study as a reference range should be done with caution [15, 16].

Conclusions

This study adds to the limited knowledge of the composition of leukocyte populations in preterm infant, particularly for those who develop LOS. Decreased non-cytotoxic T lymphocytes following LOS may reflect immunosuppression. Future studies further characterising leukocyte populations, particularly immunosuppressive populations (e.g. regulatory T lymphocytes and myeloid-derived suppressor cells), using additional markers and advanced techniques, such as high dimensionality analysis, that allow for in-depth detection beyond conventional gating, will provide complementary information on early life immune development in infants.

(PDF)

Click here for additional data file.

3 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Markus Sperandio

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating in your Funding Statement:

(This study was funded by grants from the National Health and Medical Research Council of Australia (#572548, www.nhmrc.gov.au), the Western Australia Telethon Channel 7 Trust and the Western Australia Department of Health.

DB is supported by a National Health and Medical Research Council of Australia Investigator Grant (#1175744, www.nhmrc.gov.au) and their research at the Murdoch Children's Research Institute is supported by the Victorian Government's Operational Infrastructure Support Program.

TS is supported by a Raine Foundation and Western Australia Department of Health Clinician Research Fellowship (rainefoundation.org.au).

JH is supported by a University Postgraduate Award and a Wesfarmers Centre of Vaccines and Infectious Diseases Scholarship (www.infectiousdiseases.telethonkids.org.au)

The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

3. Thank you for stating the following in the Funding Section of your manuscript:

(This study was supported by the National Health and Medical Research Council of Australia project grant (#572548), the Western Australia Department of Health, and the Western Australia Telethon Channel 7 Trust. TS is supported by a Raine Foundation and Western Australia Department of Health Clinician Research Fellowship. DB is supported by a National Health and Medical Council of Australia Investigator Grant (GTN1175744). Research at the Murdoch Children’s Research Institute is supported by the Victorian Government’s Operational Infrastructure Support Program. JH is supported by a University Postgraduate Award and postgraduate scholarship from the Wesfarmers Centre of Vaccines and Infectious Diseases. The funding sponsors had no role in the study design, collection, interpretation or analysis of data, writing the report or the decision to submit the article for publication.)

We note that you have provided funding information that is currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

(This study was funded by grants from the National Health and Medical Research Council of Australia (#572548, www.nhmrc.gov.au), the Western Australia Telethon Channel 7 Trust and the Western Australia Department of Health.

DB is supported by a National Health and Medical Research Council of Australia Investigator Grant (#1175744, www.nhmrc.gov.au) and their research at the Murdoch Children's Research Institute is supported by the Victorian Government's Operational Infrastructure Support Program.

TS is supported by a Raine Foundation and Western Australia Department of Health Clinician Research Fellowship (rainefoundation.org.au).

JH is supported by a University Postgraduate Award and a Wesfarmers Centre of Vaccines and Infectious Diseases Scholarship (www.infectiousdiseases.telethonkids.org.au).

The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.)

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

Please carefully address all concerns raised by the reviewers.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a clearly designed study to characterize the circulating leukocyte subsets in preterm neonates. The authors performed weekly FACS analyses during the first month of life and found differences between those infants that developed late onset sepsis compared to those who did not develop late onset sepsis.

Major criticism

• The staining panel is relatively simplified. Why did the authors not look at B1 B cells or other subsets that are highly interesting in the newborn? The atypical selection of leukocyte populations should be explained.

• Why was LOS defined retrospectively? It would have been much better to define this a priori.

• Definition of clinical LOS is missing: Which clinical signs were documented?

Minor comments

• The authors mention that they collected 0,8 ml blood samples but they used only 50µl for staining. How does this come?

• How many infants would have been eligible and how many infants were recruited. Were ther potential sources of a biased recruitment?

• The authors should discuss whether the administration of antibiotics may have an influence on leukocyte subpopulations.

• I do not understand the sense of the arrows in Figure 1.

Reviewer #2: The manuscript "Composition of early life leukocyte populations in preterm infants with and without late-onset sepsis" by Hibbert et al. present a descriptive single centre cohort study of 119 preterm infants below 30 weeks of gestation. The study analysed leukocyte composition of infants during the first month of life. Although data exist on this topic, the manuscript impresses by the rather large number of infants included and the systematic longitudinal follow up.

Thus, the analysis of this cohort is of interest, and with the thoughts below, the authors should be encouraged to answer some more of possibly relevant questions:

1) All Figures: For better understanding: please explain in more detail the group allocation at the time points given. Is “No LOS” at a certain time point all subjects having had no LOS until this time point or during the whole study period or only until this time point? And includes “LOS” at a time points all subjects having had LOS before or up to this time point or during the whole study period? If group allocation has been made on basis of the whole study period, the authors should analyse the data allocating the subjects to “No Los” and “LOS” before and after sepsis.

2) For all Figures: please explain “adjustment” in more detail.

3) Page 9, line 164: what about other confounders: steroids, mode of delivery, antibiotics?

4) Table 1 and page 10, line 185: age of LOS is given as 13 (10-18) in the table, and 13 (6-26) in the text, please correct if necessary.

5) What about the influence of pathogens? The authors should be encouraged to analyse the difference of Gram-positive and Gram-negative sepsis and especially of CONS-sepsis.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Christian Gille, MD

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

1 Feb 2022

Reviewer 1

This is a clearly designed study to characterize the circulating leukocyte subsets in preterm neonates. The authors performed weekly FACS analyses during the first month of life and found differences between those infants that developed late onset sepsis compared to those who did not develop late onset sepsis.

R1. Major criticism
The staining panel is relatively simplified. Why did the authors not look at B1 B cells or other subsets that are highly interesting in the newborn? The atypical selection of leukocyte populations should be explained.

Response: We acknowledge that the antibody panel used in this study is dated however this study was conducted between 2009 and 2011, a time when this panel was relatively novel. We agree that it would be ideal to have used a more sophisticated panel that included additional leukocyte sub-populations. However, a limited number of fixable fluorochromes and available channels (6-channels) on our flow cytometer were available at the time of the study. This 6-marker/5-colour surface marker panel for an extended white cell differential was therefore the best option at the time, especially in acutely ill individuals. Beckman Coulter since adapted this panel as their CytoDiff reagent, which has been validated in various health conditions (PMID: 21875393; 21779183), including infant cord blood and peripheral blood from critically ill newborns (PMID: 22862853; 27834108). The limitations of this panel have been raised in the discussion on page 28. We agree that a range of more recently identified innate lymphocytes and adaptive regulatory cells would be of great interest for future studies in the field, potentially with higher-throughput sequencing-based technologies. However, we still believe that our study adds important knowledge of postnatal extended leukocyte subset development in very preterm infants.

R1.1. Why was LOS defined retrospectively? It would have been much better to define this a priori.

Response: For this observational study, we followed our local NICU guidelines (blood culture positivity and elevated, serially measured C-reactive protein) for identifying late-onset sepsis. Therefore, late-onset sepsis was defined at the time of sepsis based on pre-defined criteria and not retrospectively.

We have removed ‘retrospectively’ (line 93 of the original submission) from the methods section to avoid confusion.

R1.2. Definition of clinical LOS is missing: Which clinical signs were documented?

Response: Clinical LOS was defined as a negative blood culture and a CRP of > 15mg/L within 72 hours of blood culture (line 96 of the original submission). Blood culture was performed if infants had clinical signs suggestive of LOS, including increased ventilation or oxygen requirements, lethargy, decreased perfusion, temperature instability, vomiting, feed intolerance, increased apnoea and/or bradycardia.

The Methods section has been updated to include this information.

Minor comments
R1.3. The authors mention that they collected 0,8 ml blood samples but they used only 50µl for staining. How does this come?

R1.3 Response: Thank you for raising this point. The samples in the manuscript were part of a prospective, longitudinal observational cohort study of innate immune system development in very preterm infants. The 0.8 mL of blood collected in this study was used in several assays for characterising immune development, including the assays reported in PMID: 31960030, 31578034, and 29746597. In addition to these published assays, several assays have not been published yet, including additional FACS panels for monocytes/DCs and T lymphocytes, as well as several whole blood pathogen phagocytosis and killing assays. Collectively, these analyses use a total of 0.8 mL of blood. We are currently analysing these data for manuscript preparation in 2022.

R1.4. How many infants would have been eligible and how many infants were recruited. Were there potential sources of a biased recruitment?

R1.4 Response: All infants born < 30 weeks gestational age who were admitted to the NICU were eligible for recruitment unless major congenital anomalies were known. Of the 398 eligible infants during the study period, 129 were recruited. The basic demographic and clinical data were similar between the eligible infants who were and were not recruited to the study (GA 26.9 vs 26.83 weeks, p=0.76; birth weight 942 vs 962 g, p=0.54; males 54 vs 52%, p=0.62; caesarean section 57% vs 65%, p=0.914). Therefore, there is no indication of biased recruitment.

Please see table in PDF appended to the end of the PDF.

R1.5. The authors should discuss whether the administration of antibiotics may have an influence on leukocyte subpopulations.

R1.5 Response: We agree that this is a topic of interest since some evidence from neonatal animal models of infection shows exposure to antibiotics, particularly exposure in utero, can alter the function, phenotype, and frequency of leukocytes (PMID 27036912, 22705104; 33178650). However, our study was not designed or powered to assess the potential influence of antibiotic use on leukocyte counts and therefore is outside the scope of this manuscript.

The ‘strengths and limitations’ section has been updated.

R1.6. I do not understand the sense of the arrows in Figure 1.

R1.6 Response: Thank you for raising that the illustration is unclear. The arrows in Figure 1 represent the median age (i.e., 13 days) that infants in the ‘any LOS’ group developed any LOS.

The figure legend has been updated to make this clearer.

Reviewer 2

The manuscript "Composition of early life leukocyte populations in preterm infants with and without late-onset sepsis" by Hibbert et al. present a descriptive single centre cohort study of 119 preterm infants below 30 weeks of gestation. The study analysed leukocyte composition of infants during the first month of life. Although data exist on this topic, the manuscript impresses by the rather large number of infants included and the systematic longitudinal follow up.
Thus, the analysis of this cohort is of interest, and with the thoughts below, the authors should be encouraged to answer some more of possibly relevant questions:

R2.1. All Figures: For better understanding: please explain in more detail the group allocation at the time points given. Is “No LOS” at a certain time point all subjects having had no LOS until this time point or during the whole study period or only until this time point? And includes “LOS” at a time points all subjects having had LOS before or up to this time point or during the whole study period? If group allocation has been made on basis of the whole study period, the authors should analyse the data allocating the subjects to “No Los” and “LOS” before and after sepsis.

R2.1 Response: Thank you for your input and consideration. Group allocations were made based on the whole study period, that is, infants that developed LOS during the study period were assigned to the LOS group from Day 1 so that the trajectory of their immune development before and after the LOS episode could be tracked. Peripheral blood was collected for this study was on days 1, 7, 14, 21 and 28 of life. Due to ethical and clinical care reasons, we could not collect an additional sample at the time of sepsis onset and therefore were unable to perform leukocyte counts at the time of sepsis. The primary aims of this study were to i) characterise the innate immune responses in a cohort of very preterm infants during the period of greatest risk of LOS, and ii) evaluate if any immunological features preceding an episode of LOS may predict the risk of LOS occurring. Therefore, we elected to allocate infants to their respective groups for the whole study period. As expected from our own and published data, 61% of LOS cases occurred between days 7 and 14, a further 31% between days 15 and 21, and only two infants each between days 1-7 and 22-28, respectively. Therefore, the differences observed between infants with and without LOS after day 7 are valid for 92% of infants. However, we agree that data leading up to, during, and after sepsis are important, and we feel future studies should consider this approach which will further advance our knowledge in this area.

The Statistical Analysis section on page 9 has been updated with this information.

R2.2. For all Figures: please explain “adjustment” in more detail.

R2.2 Response: Our statistical models (mixed linear regression analysis using random effects) were adjusted for known confounders, GA, birth weight z-score, sex and total leukocyte count (page 9 under the Statistical Analysis section).

The figure legends on pages 13 and 14 have been updated to include this information.

R2.3. Page 9, line 164: what about other confounders: steroids, mode of delivery, antibiotics?

R2.3 Response: We thank the reviewers for this suggestion however we do not think it is necessary to include these confounders in our model for the following reasons:

i) Over 90% of mothers of infants in the any LOS and no LOS groups received antenatal steroids (Table 1; 93% and 97%, p=0.35), therefore we would not expect this to influence the results. Moreover, antenatal steroids are not associated with an increased risk of LOS when gestational age is corrected for and with the proven benefit of antenatal steroids, it would be unethical to withhold treatment from women at risk of preterm birth (PMID: 33368142, 26142898, 1171763730646235). Lastly, it would not be feasible to recruit a sufficiently large number of very preterm infants not exposed to antenatal steroids.

ii) The percent of caesarean sections was similar between the any LOS and no LOS groups (caesarean section 58% and 55%, p=0.85), as shown in Table 1; therefore, we would not expect the mode of delivery to affect our results.

iii) Postnatal steroid use was higher in our cohort of infants with any LOS compared to the no LOS infants (18% and 4%, p=0.005). However, postnatal steroids were given after the study period and therefore would not affect our results. Consistent with other published neonatal studies a higher percentage of infants in the any LOS who received postnatal steroids is expected because infants who develop LOS are typically younger, smaller and more unwell than uninfected infants (PMID: 12165580). Moreover, LOS itself drives inflammation and increases the risk of chronic lung disease, which is treated using postnatal steroids (PMID: 31664055, 10585987). The ‘strengths and limitations’ section has been updated.

R2.4. Table 1 and page 10, line 185: age of LOS is given as 13 (10-18) in the table, and 13 (6-26) in the text, please correct if necessary.

R2.4 Response: Thank you for your careful review. At first glance, these numbers look contradictory; however, the text reports the range and the table reports the interquartile range (25th – 75th).

The text under the Results section on page 10 has been updated to include the interquartile range to avoid confusion.

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.

17 Feb 2022

Composition of early life leukocyte populations in preterm infants with and without late-onset sepsis

PONE-D-21-37106R1

Dear Dr. Currie,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Markus Sperandio

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors have sufficiently addressed the comments raised by both reviewers.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed all reviewer's comments. Although the study has some weaknesses, it still adds new insights that are worth to be reported.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

21 Feb 2022

PONE-D-21-37106R1

Composition of early life leukocyte populations in preterm infants with and without late-onset sepsis

Dear Dr. Currie:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Dr. Markus Sperandio

Academic Editor

PLOS ONE