Presepsin levels in neonatal cord blood are not influenced by maternal SARS-CoV-2 infection

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of the global ongoing coronavirus disease 2019 (COVID-19) pandemic. COVID-19 may affect different organs and systems, but it mostly affects the respiratory system, where it can induce symptoms ranging from normal cold to severe respiratory distress. The condition is more severe and lethal in older adults and especially in those who have pre-existing comorbidities [1].

Pregnancy is a particular condition, in which women undergo physiological and immunological changes in order for their immune systems to be able to tolerate pregnancy [2]. Furthermore, viral infections, such as SARS-CoV-2, could influence innate immune responses during pregnancy [3]. Given the higher incidence of obstetrical complications in pregnant women infected with SARS-CoV-2 [4], neonates born to infected mothers with a more severe clinical course may have a worse outcome, mainly due to neonatal morbidity and mortality associated with prematurity [5]. Moreover, neonates infected with SARS-CoV-2 can range from mild to severe manifestations, up to multisystem inflammatory syndrome (MIS-C) [6].

Several biomarkers have been studied over the course of this pandemic in relation to COVID-19, ranging from C-reactive protein to troponin [7, 8].

Presepsin (P-SEP) is a small soluble regulatory peptide generated from a soluble cluster of differentiation 14 (CD14), that modulates immune responses through the interaction with T and B cells [9]. The principal ligand of CD14 is bacterial lipopolysaccharide (LPS), and this explains why P-SEP seems to be accurate in discriminating bacterial sepsis, whereas it does not show the same accuracy in the diagnosis of viral forms from current literature. The use of P-SEP as a novel biomarker for early diagnosis, risk stratification, and prognosis prediction appeared promising not only in septic adult patients [10], but also in neonates [11], [12], [13].

The same role has been studied in adult patients with SARS-CoV-2, with higher values of P-SEP found in patients with a more severe clinical course, as well a longer hospitalization [14], [15], [16].

The aim of this work was to assess if P-SEP values in cord blood of neonates born to positive COVID-19 mothers were influenced by maternal viral infection.

Materials and methods

Results

During the study period, P-SEP values were assessed in 119 neonates born to mothers with a documented SARS-CoV-2 infection (Table 1). Of the mothers, 43 (36.1%) had a symptomatic disease, according to the WHO classification [19]. Seventy-eight neonates (65.6%) were vaginally delivered and forty-one (3.4%) by caesarean section.

Three neonates tested positive (2.5%) at the first swab performed at less than 24 h since birth, but all were negative at 48 h of age (indeterminate infection). The other two cases (1.7%) had a confirmed positive nasopharyngeal test (possible infection, since it was from a non-sterile sample.

We found a median P-SEP value of 455 pg/mL (IQR 292–762) in umbilical cord blood samples collected after the delivery. We found no significant differences in P-SEP values between term neonates (median: 461 pg/mL; IQR 292–771) and preterm neonates (median: 404 pg/mL; IQR 307–613). We observed that P-SEP values were not influenced by weight for gestational age (AGA vs SGA/LGA), or delivery mode (vaginal delivery vs. vacuum-assisted delivery/elective caesarean section/emergency caesarean section). In particular, we found no significant differences between neonates born to symptomatic COVID-19 mothers (median: 445 pg/mL; IQR 287–910) and neonates born to asymptomatic COVID-19 mothers (median: 458 pg/mL; IQR 302–716) (Figure 1).

Figure 1:

Cord P-SEP values in neonates born to symptomatic COVID-19 mothers and in neonates born to asymptomatic COVID-19 mothers.

Twenty-eight neonates (23.5%) had adverse neonatal outcomes during hospital stay (Table 2); a significantly higher incidence was observed among preterm neonates rather than in term neonates (91.7 vs 15.9% respectively, p < 0.001). P-SEP values did not significantly differ between neonates without adverse neonatal outcomes (median: 485 pg/mL; IQR 291–793) and those who have suffered them (median: 379 pg/mL; IQR 302–500) (p = 0.307).

We also found no significant differences in adverse neonatal outcomes and length of stay among those who had a P-SEP value higher than 95° centile or higher than 75° centile and those without. However, neonates with a length of stay greater than 5 days had a significantly higher P-SEP value at birth (median: 610 pg/mL; IQR 319–1,086) rather than neonates who have been discharged up to 4 days of life (median: 339 pg/mL; IQR 245–515) (p < 0.001).

We had no cases of confirmed SARS-CoV2- infection in this cohort. The three neonates with a positive nasopharyngeal swab at birth and indeterminate SARS-CoV-2 infection had a higher P-SEP value (median: 1,644 pg/mL; IQR 1045–2,599) than the two neonates with possible SARS-CoV-2 infection who later became positive at 48 h (median: 998 pg/mL; IQR 904–1,093) (Figure 2). Other 114 uninfected neonates had a median value of 439 pg/mL (IQR 291–708).

Figure 2:

Cord P-SEP values in the three neonates with positive nasopharyngeal swab at birth and indeterminate SARS-CoV-2 infection and in the two neonates with possible SARS-CoV-2 infection who later became positive at 48 h.

Among the neonates of this cohort, we had no cases of culture-proven bacterial early-onset sepsis. Only a full-term newborn with a high value of P-SEP (2,458 pg/mL) had a clinically suspected early-onset sepsis and received empiric antibiotic treatment, which was stopped when blood cultures resulted to be negative. He was regularly discharged at 7 days of life.

Discussion

In this study, we prospectively assessed presepsin values in umbilical cord blood in a cohort of neonates exposed to SARS-CoV-2 at the time of delivery. We found that P-SEP values in neonatal cord blood are not influenced by maternal SARS-CoV-2 infection, even in those neonates born to mothers with COVID-19 symptoms.

In recent years, P-SEP has gained increasing attention for its diagnostic accuracy in detecting neonatal sepsis [20, 21]. Interestingly, P-SEP seems to be not influenced by maternal and neonatal variables [11], which can bias the interpretation of the biomarker in symptomatic neonates or in neonates at risk for infection. It is well known that an elevation in values of C-reactive protein (CRP, the most widely used marker of sepsis in neonatology) is not necessarily diagnostic for sepsis but can physiologically occur in case of different conditions such as maternal intrapartum fever, premature rupture of membranes, vaginal delivery, fetal distress or perinatal asphyxia, respiratory distress syndrome [22, 23]. Similarly, procalcitonin (PCT) values show a physiologic increase 24–48 h after birth and decreases to normal values after 3 days [22]. We confirm that in our cohort P-SEP values were not affected by weight for gestational age and delivery mode.

Considering that immune system dysregulation and cytokines release are pivotal pathogenetic mechanisms of COVID-19 [24, 25], in severe cases driving to a frank production of proinflammatory cytokines and markers [26], it was important to exclude the influence of the maternal infectious state in modifying the P-SEP values of the neonates, as we have made.

Conversely, Zhu et al. described a mild increase in neonatal values of CRP and PCT in some neonates born to mothers with COVID-19, without evidence of sepsis [27].

Similarly to Zaninotto’s findings in adult patients with COVID-19 [14], we found that neonates with a longer length of stay had significantly higher P-SEP values at birth, although we did not use a specific cut-off as in adults. Indeed, available neonatal reference ranges for P-SEP are related to healthy term neonates and preterm neonates without sepsis [18], and the definitive threshold in neonates is yet to be determined when there is something to possibly affect values.

The main limitation of our study is the lack of neonates born to mothers with SARS-CoV-2 infection who also had an overlapped early-onset bacterial sepsis, to confirm the diagnostic accuracy of P-SEP in case of bacterial infection when an infant is born in the context of a maternal immune system dysregulation. A second limitation is the lack of evaluation of neonatal P-SEP values according to maternal symptoms severity.

Conclusions

In this study, for the first time, we reported cord P-SEP values in an adequate sample of term and preterm neonates born to mothers with COVID-19, without any culture-proven bacterial infection that could have increased them. Cord P-SEP values appeared to be similar in neonates born to symptomatic COVID-19 mothers and in neonates born to asymptomatic ones. We found no differences between neonates with adverse neonatal outcomes and those without; the only neonate with suspected early-onset sepsis had a high value of cord P-SEP. Therefore, considering the need for a feasible biomarker of EOS not influenced by maternal characteristics and hours of life (such as CRP and PCT), we postulate that P-SEP can be used to rule out bacterial sepsis even in neonates born to mothers with COVID-19 at the time of delivery.

However, further studies are still needed to explain the mechanisms of P-SEP increase in neonates exposed to perinatal maternal SARS-CoV-2 infection or with an indeterminate/possible SARS-CoV-2 infection in the same neonates.