If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The neutrophil-to-lymphocyte ratio (NLR) is easily calculated from the full blood count, which is routinely performed.
NLR shows promise as a low-cost marker of placental inflammation who deliver preterm without clinical signs of infection.
NLR is equally feasible in a global health context where the burden of preterm birth is highest.
Chorioamnionitis is present in up to 70% of spontaneous preterm births and is associated with poor maternal, fetal and neonatal outcomes.
To explore the relationship between the neutrophil-to-lymphocyte ratio and histological chorioamnionitis in women who delivered preterm with no clinical signs or symptoms of infection.
This was a retrospective analysis of a cohort of women who delivered spontaneously between 16 and 36+6 weeks at a tertiary UK hospital. Only women with placental histology and no signs of clinical infection were included. The neutrophil-to-lymphocyte ratio was calculated from a full blood count sample taken routinely within 24 h of delivery. The neutrophil-to-lymphocyte ratio was also calculated from first trimester booking bloods (<13 + 6 weeks) in a subgroup. Placental histopathology was categorised as either inflammatory (i.e. histologic chorioamnionitis, with or without evidence of fetal inflammatory response) or non-inflammatory (vascular pathology or a normal placenta).
169 women had available placental pathology and were included in the analysis. 70 % (118/169) had confirmed placental inflammation. The mean neutrophil-to-lymphocyte ratio was significantly raised in this group compared to those with normal (n = 24) or vascular (n = 27) pathology (inflammatory neutrophil-to-lymphocyte ratio 9.81 vs non-inflammatory neutrophil-to-lymphocyte ratio 6.53, p = 0.002.
The delivery neutrophil-to-lymphocyte ratio had an area under the receiver operating characteristic curve of 0.69 (0.60 to 0.78) for predicting placental inflammation. A raised neutrophil-to-lymphocyte ratio (>6) was associated with an odds ratio of 5.2 (95 % CI 2.55 to 10.56) for histological chorioamnionitis, with a sensitivity of 80 % and negative predictive value of 86 %. A higher cut-off of 9 had a negative predictive value of 79 % for fetal inflammatory response.
A raised neutrophil-to-lymphocyte ratio is associated with a 5-fold increased risk of histological chorioamnionitis in women who delivered early without signs or symptoms of infection. It was also raised at the time of preterm labour compared to the first trimester. A full blood count is an almost universal investigation in women admitted in preterm labour, often repeated, making this inexpensive and non-invasive ratio a useful additional antenatal biomarker in women admitted in spontaneous preterm labour at risk of subclinical chorioamnionitis and its associated poor outcomes.
Preterm birth is the leading cause of neonatal and infant mortality, as well as a major contributor to lifetime disability. Intrauterine inflammation and infection is frequently implicated in the onset of premature parturition, particularly in babies born before 30 weeks [
] including preterm birth, neonatal intraventricular haemorrhage, cerebral palsy, bronchopulmonary dysplasia, early-onset neonatal sepsis, necrotising enterocolitis and death. While acute HCA is evidence of a maternal host response, coexistent chorionic vasculitis, umbilical phlebitis or funisitis represents the fetal inflammatory response (FIR), and even after adjustment for gestational age is associated with a higher rate of neonatal morbidity and multiorgan fetal involvement [
], the hallmark of intrauterine inflammation and infection. The only antenatal diagnostic test is amniocentesis which carries risks of miscarriage or preterm birth and is not commonly used within clinical practice. In suspected chorioamnionitis iatrogenic preterm delivery is often undertaken to minimize maternal and fetal morbidity and therefore antenatal diagnosis would improve risk assessment and avoid unnecessary intervention, or conversely potentially harmful attempts to prolong the pregnancy. Neonatal morbidity and mortality can occur with HCA in the absence of overt clinical signs of maternal infection (C-reactive protein, white cell count or raised fetal heart rate) [
], posing a significant clinical conundrum. Like other obstetric conditions such as pre-eclampsia, it is unlikely that a single predictive marker will be diagnostic therefore non-invasive biomarkers are highly sought after.
The systemic host response to inflammation and infection causes neutrophilia and lymphocytopenia, resulting in a raised neutrophil-to-lymphocyte ratio (NLR). This easily measured marker has been used as a parameter of stress in critically ill patients [
] and when monitoring treatment response. It is a reproducible, non-invasive and inexpensive marker of sub-clinical inflammation and infection. The aim of this study was to investigate the association between the NLR calculated from a venous full blood count within 24 h of delivery and infective placental histology in women who delivered preterm spontaneously with no clinical signs or symptoms of infection.
Material and methods
This was a retrospective analysis of 169 asymptomatic women who delivered spontaneously between 16+0- and 36 + 6-weeks’ gestation at a London tertiary hospital over a thirteen-year period. Women were included in this study if they had placental histology and a full blood count sample taken within 24 h of birth. Patients with maternal pyrexia (>38 °C), WCC > 15, CRP > 5 or a clinical diagnosis of chorioamnionitis were excluded from the analysis.
Using Redline’s criteria, placental pathology was characterised according to histological lesions signifying either inflammatory, vascular or no placental pathology [
]. The criteria for each category are summarised in Supplemental Table 1. The histological results were then sub-categorised as either inflammatory or non-inflammatory (vascular or normal pathology). Inflammatory histology was categorised into chorioamnionitis with or without evidence of fetal inflammatory response.
Absolute neutrophil and lymphocyte counts were collected via the electronic patient records from each woman’s booking visit (<13+6 weeks, where available), and within 24 h of preterm birth.
Statistical analysis was performed using Stata 14.0. Descriptive statistics were used to describe the study population. The NLR was calculated and compared between women with and without placental inflammation using Pearson’s independent chi square coefficient. Receiver operating characteristic (ROC) curves were generated and compared. This data was used to determine optimal cut-offs, balancing test sensitivity and specificity. Predictive statistics were carried out to determine if NLR accurately predicted placental inflammation using the determined thresholds.
This study was exempt from requiring ethical approval under the UK Health and Social Care Act 2012, which states that research involving anonymised routinely collected clinical data is excluded from research ethics committee review. As per hospital policy, placental analysis was completed for all women for the following indications: perinatal loss due to miscarriage or stillbirth, unexpected admission to Neonatal Intensive Care Unit, requested by Fetal Medicine Unit following antenatal scan, fetal growth restriction, placental abruption or consultant request (ie. preterm birth). Placental histology reports were routinely available as part of normal clinical care. Pathologists were blinded to clinical presentations, except gestation at delivery, and were all fully competent in both gross and microscopic placental evaluation. This part of the study formed an institutional audit into the need for placental pathology.
From a cohort of 216 women who delivered spontaneously preterm and had placental histology, 169 did not have clinical signs or symptoms of infection and had had a full blood count sample available within 24 h of delivery. 87 % (147/169) also had first trimester bloods available for analysis. Demographic, obstetric and background characteristics for study participants are described in Table 1. The average gestation of delivery was 25+6 weeks (SD 5+6 weeks); 38 % (65/169) of pregnancies ended in late miscarriage (16 to 23+6 weeks), 3 % in stillbirth and 11 % resulted in neonatal death.
Table 1Maternal demographics and characteristics.
Whole cohort (n = 169)
Inflammatory placenta (n = 118)
Non-inflammatory placenta (n = 51)
Gestational age at delivery (weeks days)
Age (mean, SD)
Ethnicity (N, %) Black White Asian Other
88 (56 %)42 (27 %)20 (13 %)6 (4 %)
67 (60 %)27 (24 %)13 (12 %)4 (4 %)
21 (47 %)15 (33 %)7 (16 %)2 (4 %)
Obstetric Risk Factors
Previous Preterm Birth
122 (78 %)
84 (76 %)
38 (84 %)
Previous Late Miscarriage
71 (46 %)
54 (49 %)
17 (38 %)
Previous Cervical Surgery
10 (6 %)
8 (7 %)
2 (4 %)
6 (4 %)
4 (4 %)
2 (4 %)
Other Risk Factors
Recurrent urinary tract infections
4 (2 %)
2 (2 %)
2 (4 %)
3 (2 %)
2 (2 %)
1 (2 %)
Data are mean (standard deviation) or n (%) unless otherwise specified. Late miscarriage is defined at fetal demise between 14 and 24 weeks gestation. N/A = lack of data or inapplicable to that cohort. *UTI: urinary tract infections.
A total of 70 % (118/169) of pregnancies had confirmed placental inflammation, of whom 42 % (50/118) had FIRS diagnosed histopathologically (Fig. 1). There were no cases of funisitis without HCA.
Placental inflammation was seen more frequently in late miscarriage and the extremes of preterm birth (Fig. 2), while normal or vascular placental histology were more frequently observed in the late preterm birth period (inflammatory histology and gestational age linear regression R2 = -0.869, non-inflammatory histology and gestational age R2 = 0.756).
There was no significant difference in the neutrophil count (6.31 versus 6.32, 95 % CI −1.08 to 1.10, p = 0.99) lymphocyte count (1.83 versus 1.96, 95 % CI −0.08 to 0.34, p = 0.21) or NLR (3.77 versus 3.51, 95 % CI −1.04 to 0.52, p = 0.51) from first trimester bloods from women with and without placental inflammation at time of delivery (Table 2). NLR did not predict inflammatory placental histology at delivery (AUC ROC 0.41, 95 % CI 0.31 to 0.51).
Table 2Leukocyte differential count and NLR early in pregnancy and at the time of preterm birth.
Serum inflammatory markers
n = 169 (mean, SD)
Inflammatory placenta n = 118 (mean, SD)
Non-inflammatory placenta n = 51 (mean, SD)
95 % confidence interval (p value)
1st TRIMESTER1 BLOODS
−1.08 to 1.10 (p = 0.99)
−0.08 to 0.34 (p = 0.21)
−1.04 to 0.52 (p = 0.51)
−4.23 to -0.51 (p = 0.01)
0.02 to 0.52 (p = 0.03)
−5.35 to -1.21 (p = 0.002)
Differences between neutrophils, lymphocytes and NLR are presented in Table III.
Mean NLR was significantly raised at delivery compared to first trimester bloods (NLR 8.84 (SD 6.13) versus 3.69 (2.27), 95 % CI = 4.1 to 6.2, p < 0.0001).
At delivery the neutrophil count (12.32 versus 9.95, 95 % CI −4.32 to −0.51, p = 0.01), lymphocyte count (1.54 versus 1.81, 95 % CI 0.02 to 0.52, p = 0.03) and NLR (9.81 versus 6.51, 95 % CI −5.35 to −1.21, p = 0.002) were all significantly higher in women who delivered with placental inflammation. Delivery NLR showed strongest clinical utility in predicting placental inflammation in women without signs of infection [AUC ROC 0.69 (95 % CI 0.60 to 0.78)] (Table 3 and Fig. 3). Absolute values of NLR were higher in women with advanced placental inflammation (confirmed FIRS), although not statistically significant (9.81 versus 10.55, p = 0.56).
Table 3Leukocyte differential count and neutrophil-to-lymphocyte ratio for predicting placental inflammation in women delivering preterm without signs or symptoms of infection.
ROC curve analysis was used to determine an optimal cut off for predicting placental inflammation with and without FIRS, balancing test sensitivity and specificity. An optimal cut off of 6 had a sensitivity of 80.0 %, with a negative predictive value (NPV) of 85.7 %. For FIRS, a higher cut off of 9 had a similar NPV of 79.2 %, confirming utility to aid ruling out FIRS (Table 4).
Table 4Accuracy of the neutrophil-to-lymphocyte ratio to predict placental inflammation with or without fetal inflammatory response in women delivering preterm without signs or symptoms of infection.
To ensure differences in gestational age at delivery did not influence our findings we investigated the prediction of placental inflammation at clinically important time points. NLR retained clinical utility within gestational age subgroups (AUC of 0.69 (95 % CI 0.52 to 0.86) in late miscarriage 16 to 23 + 6, AUC of 0.70 (95 % CI 0.56 to 0.85) between 22 and 28 weeks).
NLR, a commonly used marker of subclinical inflammation outside of the pregnant population, was significantly raised in women with HCA at the time of preterm birth in the absence of clinical signs of infective morbidity (maternal pyrexia, raised WCC or CRP) compared to women with normal or vascular placental pathology.
A full blood count is routinely, often repeatedly, measured in pregnancy, even in a global-health context, thus NLR has potential to provide additional information universally, without additional tests being carried out. It is reproducible and inexpensive and shows potential as an additional marker in the challenging context of subclinical inflammation and infection in women presenting in preterm labour (HCA when NLR > 6 or FIRS when NLR > 9).
NLR was significantly raised in all women delivering preterm compared to first trimester counts, regardless of aetiology. Other investigators have shown that NLR is not raised during term labour [
], therefore our findings suggest NLR may be related to preterm labour, even in the absence of an infective cause. These findings need to be confirmed in future studies. Nonetheless, the ratio was significantly higher in preterm women with placental inflammation, confirming NLR as a potential tool to discriminate aetiology and aid in management of preterm labour and delivery, particularly useful when there is an absence of obvious clinical signs or symptoms of infection, or other morbidities.
A large proportion (70 %) of women who delivered early had confirmed inflammation on placental histology; incidence was inversely related to gestational age (R2 = −0.87). This suggests that infection is a more common aetiological factor at earlier gestations, as previously described [
The diagnosis of subclinical chorioamnionitis poses a significant clinical conundrum in obstetrics. Currently, obstetricians must balance the risk and benefits of prolonging gestation and avoiding unnecessary complications of prematurity against the risks of maternal and/or overwhelming fetal or early neonatal sepsis based on poor predictive markers. Positive amniotic fluid cultures and inflammatory biomarkers (e.g. IL-6) [
] which would not be easily translatable in low-middle income settings, is not acceptable to women with risks of its own. Placental histology, still considered the gold standard in diagnosing chorioamnionitis, has a role in guiding postnatal care of the newborn but is only available retrospectively after delivery. Maternal serological markers may serve as an inexpensive and reproducible antenatal surrogate [
], however are often only reflective of late stage infection and in nearly-one third of cases chorioamnionitis diagnosed postnatally is asymptomatic in the mother making a simple non-invasive test to guide management desirable.
Like other obstetric conditions, such as pre-eclampsia, it is unlikely that a single marker will be sufficient to determine diagnosis, appropriate timing of delivery and predict likely outcomes. However NLR may prove a feasible clinical tool together with other maternal blood inflammatory markers and could be used to direct the need for more precise diagnostic inflammatory tests or fetal and placental imaging [
Previous studies have correlated a raised NLR with adverse pregnancy outcomes. In a small study of 37 patients, Jung et al (2015) showed that NLR (AUC 0.73) and IL-8 concentrations in amniotic fluid (AUC 0.72) were associated with earlier gestation at delivery (sPTB < 32 weeks’), and Lee et al (2017) showed an association with poor neonatal outcomes including necrotizing enterocolitis (NEC), even after adjustment for birth weight and gestational age (26 neonates diagnosed with NEC) [
]. Song et al (2018) observed worse pregnancy outcomes (early gestational age at delivery and poorer neonatal survival rates) in women with raised NLR at time of repeat cerclage with prolapsed membranes (n = 26) [
]. While both examine important clinical outcomes and confirm NLRs promise as a reliable predictor of pregnancy outcome, they are limited by their small sample size and need to be replicated in larger prospective studies.
Kim et al suggested NLR may be a predictive marker for HCA, with greater diagnostic usefulness than CRP in women who delivered preterm [
] however to the best of our knowledge, this is the first study which shows that NLR is equally valid in well patients (maternal temperature < 38 °C, WCC < 15 and CRP < 5) where the diagnostic dilemmas are more challenging.
Additionally while previous work has demonstrated that the NLR is significantly elevated in preterm patients compared to term delivery [
], we demonstrated that NLR is increased at preterm delivery compared to the first trimester.
While we did not have data to exclude blood samples potentially affected by antenatal corticosteroids, we were able to show that clinical utility was retained (AUC 0.69) in women with late miscarriage (16 to 23+6 weeks’), when steroids are not clinically indicated. This cohort allowed us to draw some limited conclusions about NLR at different gestational ages, however future trials will be needed to evaluate NLRs ability to prospectively predict outcome across all gestational ages in pregnancy, as well as validate cut offs.
NLR is a simple ratio calculated from a full blood count, a routine blood test universally performed, often repeatedly, during pregnancy. A raised NLR indicates a 5-fold increased risk of placental inflammation possibly proving itself to be a simple diagnostic alert to possible infective morbidity in women without signs or symptoms of infection. In a global health context where most preterm births happen but tests are scarce, NLR may prove a feasible clinical tool. NLR could also be used to direct the need for more precise diagnostic inflammatory tests as they are developed. Future trials will be needed to evaluate its ability to predict outcome prospectively in pregnancy.
No specific funding was obtained for this research. AER is partly funded by Wellbeing of Women (Registered Charity no: 239281) and by the CLAHRC South London (NIHR). This research was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy's and St Thomas' NHS Foundation Trust and King's College London and/or the NIHR Clinical Research Facility. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A. Supplementary data
The following are the Supplementary data to this article: