Home Print this page Email this page Small font sizeDefault font sizeIncrease font size
Users Online: 1457
About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Advertise Login 

 Table of Contents  
Year : 2018  |  Volume : 7  |  Issue : 3  |  Page : 146-150

Use of a single C-reactive protein level in decision-making during neonatal sepsis evaluation

1 Department of Pediatrics, Division of Neonatology, Albany Medical Center, Albany, NY, USA
2 Department of Medical Education, Albany Medical College, Albany, NY, USA
3 Department of Pediatrics, Division of Infectious Diseases, Albany Medical Center, Albany, NY, USA

Date of Web Publication2-Aug-2018

Correspondence Address:
Dr. Rubia Khalak
Department of Pediatrics, Division of Neonatology, Albany Medical Center, MC.101, 43 New Scotland Avenue, Albany, NY 12208
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jcn.JCN_32_18

Rights and Permissions

Background: Use of a single C-reactive protein (CRP) level has been studied in the pediatric population, but limited information is available for the neonatal, particularly Neonatal Intensive Care Unit (NICU) infants. Objective: The objective of this study is to determine if a single CRP level in the context of other laboratory and clinical parameters, can assist in decision-making for antibiotic management. Materials and Methods: We reviewed the medical records of infants admitted to a large regional perinatal center NICU over a 2-year period. Infants in whom a CRP level had been measured were divided into sepsis-treated group if antibiotic therapy was instituted for ≥7 days or the no sepsis group if antibiotics were discontinued after 48 h. Characteristics of delivery, general characteristics of the infant and data at the time of sepsis evaluation were collected. This was a powered study to detect at least 10% difference in the proportion of those with CRP <8 mg/L versus elevated CRP levels. Results: There were 87 infants with CRP levels in the sepsis-treated group and 106 infants in the no sepsis group. Infants in the sepsis group had a significantly lower gestational age and birth weight but a significantly higher median CRP compared to infants in the no sepsis group (37.5 mg/L vs. 18.1 mg/L, P = 0.0016). When infants were evaluated based on CRP level, we found that 67% of the infants with CRP <8 mg/L had their antibiotics discontinued compared to 43% of the infants with CRP ≥8 mg/L (P = 0.002). Conclusion: A single CRP level done at the time of a neonatal sepsis evaluation can assist in decision-making regarding the continuance of antibiotic therapy. The use of CRP among infants undergoing sepsis evaluation in the NICU significantly affected the decision of medical providers to discontinue antimicrobial agents. This is consistent with the American Academy of Pediatrics Committee on Fetus and Newborn recommendation that bacterial sepsis is unlikely with normal CRPs and antibacterial agents may be discontinued.

Keywords: C-reactive protein, neonates, sepsis evaluation

How to cite this article:
Khalak R, Malhotra A, Santos RP. Use of a single C-reactive protein level in decision-making during neonatal sepsis evaluation. J Clin Neonatol 2018;7:146-50

How to cite this URL:
Khalak R, Malhotra A, Santos RP. Use of a single C-reactive protein level in decision-making during neonatal sepsis evaluation. J Clin Neonatol [serial online] 2018 [cited 2020 Feb 24];7:146-50. Available from: http://www.jcnonweb.com/text.asp?2018/7/3/146/238404


Neonatal sepsis evaluation, a common occurrence in the Neonatal Intensive Care Unit (NICU), is usually comprised a complete blood count (CBC), blood culture and possibly cerebrospinal fluid (CSF), and urine cultures. When an infant appears ill the clinical findings such as temperature instability, respiratory distress, apnea, bradycardia, and hypotension are nonspecific.[1],[2] Sometimes, especially if noted in isolation, these findings are transient, and no infection is confirmed. If an organism is isolated from a sterile site such as blood or CSF, the treatment course is straightforward. For the majority of remaining infants, the nonspecific symptoms resolve at some point after sepsis evaluation, but antibiotics have already been started. Often an organism is not isolated from the various specimens collected, bringing uncertainty of when to discontinue antibiotic treatment. The use of biomarkers, such as C-reactive protein (CRP), can assist in sepsis management decision-making.[3],[4],[5],[6] The use of rapid blood testing for CRP has become common in pediatrics to help evaluate infection in children and increasingly more common during the evaluation of sepsis in neonates.[7] Benitz et al. concluded that infection was very unlikely if CRP levels were <10 mg/L.[8]

Chiesa et al.[9] established normative values along with the effect of gestational age (GA) on CRP. The researchers found that the regardless of sampling time or GA, CRP increases after birth. As with some other more recent studies,[10],[11] each infant was assessed only once rather than obtaining serial samples. Other biomarkers such as procalcitonin (PCT) have also been studied but not found to be as correlative as CRP.[12],[13],[14] Although studied in children, the use of a single CRP value for evaluative purposes has not been well studied in the neonatal population and even less so in premature infants. Studies in the pediatric population have shown that single CRP determination in association with clinical signs of sepsis including apnea, tachycardia or bradycardia, tachypnea, irritability, or hypotension[2],[15] can be used in antibiotic decision management of treatment course.

In this study, we set out to determine if a single CRP value could be used for the discontinuation of antibiotic therapy. We hypothesize that a single CRP when obtained during a neonatal sepsis evaluation may influence decision-making regarding the discontinuation of antibiotics.

[TAG:2]Materials and Methods[/TAG:2]

This retrospective cohort study was approved by the Institutional Review Board. Maternal and neonatal data were collected from the Electronic Medical Record of all infants admitted to NICU from January 1, 2013, to December 31, 2014. Two infant groups were established: proven sepsis/presumed sepsis group (sepsis treated) in which infants were treated with at least a 7-day antibiotic course and a ruled-out sepsis infant group (no sepsis) that had antibiotics discontinued after 48 h. Sepsis evaluations could occur at any time during NICU hospitalization, thereby capturing early and late-onset sepsis. Data collection included general characteristics of delivery including maternal GBS status, presence of prolonged rupture of membranes (PROM) and mode of delivery and general characteristics of the infant including GA, birth weight (BW), gender, and 5 min Apgar. The variables collected for evaluation of neonatal sepsis included age in days at the beginning of neonatal sepsis evaluation, recent surgery within the month before evaluation, presence of apnea or bradycardia within 24 h of evaluation, presence and type of respiratory support at the time of sepsis evaluation, CBC within 24 h of evaluation, blood, urine, and CSF culture findings within 48 h of evaluation. A power analysis was performed to detect at least 10% difference in the proportion (two-sided) of those with normal versus elevated CRP level with an α = 0.05 and β = 0.10. This resulted in a sample size of at least 85 infants for the no sepsis group with antibiotics discontinued and another 85 infants in the sepsis-treated group for whom antibiotics were given for at least a 7-day course.

Statistical analysis of nonuniformly distributed data of BW and GA was done using the Mann—Whitney test. Fisher's exact test was used to compare PROM, gender, 5 min Apgar, resuscitation in delivery room, presence and type of respiratory support, presence of apnea or bradycardia, CBC, blood, urine, and CSF culture results between the two infant groups of sepsis treated and no sepsis. Sensitivity and specificity calculations were done using three CRP cutoffs, <8, <25, and <50 mg/L, the common thresholds reported in the literature. The level of significance was established at the P < 0.05 level. Statistical analyses were done using SPSS software (Version 23, IBM Corporation, Armonk, NY, USA).


Data were collected for 193 infants; 87 infants were in the sepsis-treated group and 106 infants in the no sepsis group. Of the 1574 infants admitted to the NICU during the study period, 1167 infants (74%) had a sepsis evaluation done at some point during their NICU hospitalization. Of these infants, 17% had a CRP done as part of their sepsis evaluation. More than half of the infants in this CRP were determined to have no sepsis and had their antibiotics discontinued [Figure 1]. Of the 1167 infants who underwent a sepsis evaluation, 1043 infants were in the no sepsis group having antibiotics discontinued, 89.4%. Of the sepsis treated infants, 87% had presumed sepsis and of these the majority had early-onset sepsis, and 43 infants compared to 33 with late-onset sepsis. In contrast, infants with confirmed sepsis were twice as likely to have late-onset sepsis as early-onset sepsis. Infants in the sepsis-treated group had a significantly lower GA (31.7 weeks, standard deviation [SD] 6.1 vs. 34.1 weeks, SD 5.6; P = 0.004).
Figure 1: Flow diagram of the Neonatal Intensive Care Unit infants with sepsis evaluation

Click here to view

Our data also showed differences in BW and 5 min Apgar when sepsis treated and no sepsis groups were compared [Table 1]. Measures of mean WBC count (17.6 SD 14.4 vs. 13.6 SD 6.6; P = 0.02) were higher in the sepsis-treated group of infants. Comparison of the age in days at the time of sepsis evaluation showed infants in the no sepsis group were evaluated at a mean of 14 days of age, SD 21 days and the sepsis-treated group infants were evaluated at a mean of 17 days, SD 26 days; P = 0.46. Other variables assessed such as gender, PROM, and mode of delivery were not different between the two groups of infants. Clinical findings such as bradycardia, apnea, presence or absence of respiratory support were also compared between the sepsis and no sepsis group of infants [Figure 2]. Most notably, there were more infants on ventilatory respiratory support in the sepsis-treated group (61 infants vs. 46 infants, P = 0.003). Infants on respiratory support of high-flow nasal cannula (HFNC) or nasal continuous positive airway pressure (NCPAP) were less likely to be treated for sepsis (7 infants vs. 28 infants, P = 0.0012).
Table 1: Comparison of infants treated for sepsis and infants without sepsis

Click here to view
Figure 2: Clinical signs noted with sepsis evaluation

Click here to view

Infants in the sepsis-treated group had a higher median CRP compared to infants in the no sepsis group (37.5 mg/L vs. 18.1 mg/L, P = 0.0016). When infants were evaluated based on CRP level, we found that 67% of the infants with CRP <8 mg/L had their antibiotics discontinued compared to 43% of the infants with CRP ≥8 mg/L (P = 0.002). Increasing the CRP threshold to ≥50 mg/L increased the number of infants who were continued on an antibiotic treatment course (P = 0.001). There was no difference in the incidence of apnea and bradycardia events when comparisons were made between infants treated with full course of antibiotics versus those who had antibiotics discontinued. We determined sensitivity and specificity for CRP values using the cutoffs of 8, 25, and 50 mg/L. Increasing the cutoff values for CRP increased the sensitivity and the positive predictive value. With a CRP value of <8 mg/L, the sensitivity was 0.57 and specificity was 0.67 [Table 2].
Table 2: Comparison of C-reactive protein cutoff values

Click here to view


In this study, we found that the almost 90% of NICU infants had their antibiotics discontinued with their sepsis evaluation. Seventeen percent of the infants with a sepsis evaluation had a CRP level assessed to assist in decision-making of antibiotic management, typically to reassure medical team regarding the discontinuation of antibiotics with a negative blood culture. The no sepsis group of infants who had a CRP done with their sepsis evaluation had a significantly lower median CRP compared to the sepsis treated infants, consistent with the premise that the low CRP level was one piece of data used to discontinue antibiotics. We also noted that two-thirds of the infants with CRP <8 mg/L had their antibiotics discontinued, this again shows a good use of CRP laboratory data for clinical decision-making.

Protein sepsis markers, such as CRP, should be used in concert with clinical signs and findings to make the diagnosis of neonatal sepsis and formulate a plan for management.[14],[15],[16],[17] CRP has been used to monitor response to infection and to assist in ruling out infection.[18] A study using a single CRP level to guide in which children antibiotics could safely be discontinued showed that a CRP <10 mg/dL was acceptable.[10] Most of these studies have been done in children, term, or near-term neonates using the lower CRP level cutoffs of <8 mg/dL or <10 mg/dL. We have shown that a single CRP level with other findings such as WBC level and respiratory support level can be used for decision-making of antibiotics during a neonatal sepsis evaluation even in preterm infants.

The Committee of Fetus and Newborn of the American Academy of Pediatrics (AAP) in their statement on infants with suspected or proven sepsis have expressed agreement in the utility of inflammatory biomarkers.[5] Although other biomarkers such as PCT may also be used, CRP has been described as a later but more specific marker of infection.[16],[17] Our study found that sepsis-treated infants not only had a lower GA and BW but also a higher mean WBC count. This finding is consistent with a higher risk of infection in low BW infants. However, variables such as PROM and mode of delivery did not differ between the sepsis treated and no sepsis groups of infants. This may be a result of maternal antibiotic treatment even without confirmed chorioamnionitis. As noted only 17% of the infants had a CRP done as part of their sepsis evaluation. We hypothesize that this number reflects those infants in whom there was a more serious concern of sepsis. With the possibility of negative cultures yet strong clinical concerns at the time of sepsis evaluation, a CRP level was obtained. Previously, without CRP assessment, many of these infants would have received a presumed sepsis antibiotic treatment course. Echoing the AAP guidelines, a recent study found that in infants <28 days of age with suspected sepsis and certain clinical signs such as hypotension and bradycardia along with CRP were more likely to have confirmed sepsis.[15]

As continues to be shown, clinical correlation with laboratory markers is imperative.[12],[18] If concentrations of CRP remain low, infection is unlikely and discontinuing antibiotic therapy is reasonable.[16],[18] Our data showed that there was a progressive improvement in sensitivity of a single CRP value as the cutoff was increased. We conclude that a single CRP level obtained during neonatal sepsis evaluation of even preterm infants can be helpful in assisting with the decision to discontinue antibiotics in the presence of a low CRP value. This is consistent with what has been published in the literature using two successive CRP values.

The use of a single CRP value has been shown in the pediatric and term neonatal populations but not well documented in the premature infant. Various clinical signs have been noted as having an association with sepsis, including apnea, tachycardia or bradycardia, tachypnea, irritability, or hypotension.[2],[19],[20] When some of these signs, particularly, heart rate or respiratory changes are considered in concert with CRP levels; strong association with sepsis has been reported.[15] This same research group also found bradycardia to be a more predictive sign of sepsis in preterm infants. In our study, we did not see a difference in apneic or bradycardic episodes for infants treated for sepsis versus the no sepsis group. However, we found that infants on respiratory support of HFNC or NCPAP were more likely to have their antibiotics discontinued with a sepsis evaluation whereas infants on ventilator support were more likely to complete a full course of antibiotics with their sepsis evaluation. Infants on ventilator support are presumably more ill at baseline and more susceptible to infection. Of note, 43% of the infants in the no sepsis group were on a ventilator, yet did have their antibiotics discontinued. The evaluation of cord blood for acute phase reactant biomarkers from preterm infants with sepsis has shown elevated levels of CRP.[20] The researchers found that the measurements of PCT were not different in the infants with sepsis compared to those without sepsis. The authors concluded that CRP could be used to predict early onset neonatal sepsis. It has been recently proposed that using biomarkers in conjunction, particularly PCT and CRP (early and late-phase biomarkers) could assist in decision-making of a sepsis diagnosis.[21],[22] Use of these markers to discontinue antibiotic therapy would aid in promoting antibiotic stewardship in both preterm and term neonates.

Our study was limited by its retrospective nature. A prospective study could have been designed to capture specific reasons why medical providers chose to continue antibiotics particularly in the setting of culture-negative sepsis and whether specific clinical or laboratory data or a combination of findings influenced the decision-making. A strength of this study is the large number of sepsis evaluations captured with available CRP values at a single academic center. This allowed us to conduct a powered study to detect a ≥10% difference when comparing CRP levels. The study results confirm our hypothesis that a single CRP level in conjunction with clinical findings can allow for determination and assistance with management and treatment of neonatal sepsis.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Bonadio WA, Hennes H, Smith D, Ruffing R, Melzer-Lange M, Lye P, et al. Reliability of observation variables in distinguishing infectious outcome of febrile young infants. Pediatr Infect Dis J 1993;12:111-4.  Back to cited text no. 1
Bekhof J, Reitsma JB, Kok JH, Van Straaten IH. Clinical signs to identify late-onset sepsis in preterm infants. Eur J Pediatr 2013;172:501-8.  Back to cited text no. 2
Weitkamp JH, Aschner JL. Diagnostic use of C-reactive protein (CRP) in assessment of neonatal sepsis. Neoreviews 2005;6:e508-15.  Back to cited text no. 3
Hedegaard SS, Wisborg K, Hvas AM. Diagnostic utility of biomarkers for neonatal sepsis — A systematic review. Infect Dis (Lond) 2015;47:117-24.  Back to cited text no. 4
Polin RA, Committee on Fetus and Newborn. Management of neonates with suspected or proven early-onset bacterial sepsis. Pediatrics 2012;129:1006-15.  Back to cited text no. 5
Ng PC. Diagnostic markers of infection in neonates. Arch Dis Child Fetal Neonatal Ed 2004;89:F229-35.  Back to cited text no. 6
Shah BA, Padbury JF. Neonatal sepsis: An old problem with new insights. Virulence 2014;5:170-8.  Back to cited text no. 7
Benitz WE, Han MY, Madan A, Ramachandra P. Serial serum C-reactive protein levels in the diagnosis of neonatal infection. Pediatrics 1998;102:E41.  Back to cited text no. 8
Chiesa C, Natale F, Pascone R, Osborn JF, Pacifico L, Bonci E, et al. C reactive protein and procalcitonin: Reference intervals for preterm and term newborns during the early neonatal period. Clin Chim Acta 2011;412:1053-9.  Back to cited text no. 9
Lacaze-Masmonteil T, Rosychuk RJ, Robinson JL. Value of a single C-reactive protein measurement at 18 h of age. Arch Dis Child Fetal Neonatal Ed 2014;99:F76-9.  Back to cited text no. 10
Beltempo M, Viel-Thériault I, Thibeault R, Julien AS, Piedboeuf B. C-reactive protein for late-onset sepsis diagnosis in very low birth weight infants. BMC Pediatr 2018;18:16.  Back to cited text no. 11
Stein M, Schachter-Davidov A, Babai I, Tasher D, Somekh E. The accuracy of C-reactive protein, procalcitonin, and s-TREM-1 in the prediction of serious bacterial infection in neonates. Clin Pediatr (Phila) 2015;54:439-44.  Back to cited text no. 12
Lapillonne A, Basson E, Monneret G, Bienvenu J, Salle BL. Lack of specificity of procalcitonin for sepsis diagnosis in premature infants. Lancet 1998;351:1211-2.  Back to cited text no. 13
Blommendahl J, Janas M, Laine S, Miettinen A, Ashorn P. Comparison of procalcitonin with CRP and differential white blood cell count for diagnosis of culture-proven neonatal sepsis. Scand J Infect Dis 2002;34:620-2.  Back to cited text no. 14
Ohlin A, Björkqvist M, Montgomery SM, Schollin J. Clinical signs and CRP values associated with blood culture results in neonates evaluated for suspected sepsis. Acta Paediatr 2010;99:1635-40.  Back to cited text no. 15
Philip AG, Mills PC. Use of C-reactive protein in minimizing antibiotic exposure: Experience with infants initially admitted to a well-baby nursery. Pediatrics 2000;106:E4.  Back to cited text no. 16
Hofer N, Zacharias E, Müller W, Resch B. An update on the use of C-reactive protein in early-onset neonatal sepsis: Current insights and new tasks. Neonatology 2012;102:25-36.  Back to cited text no. 17
Hisamuddin E, Hisam A, Wahid S, Raza G. Validity of C-reactive protein (CRP) for diagnosis of neonatal sepsis. Pak J Med Sci 2015;31:527-31.  Back to cited text no. 18
Griffin MP, Lake DE, O'Shea TM, Moorman JR. Heart rate characteristics and clinical signs in neonatal sepsis. Pediatr Res 2007;61:222-7.  Back to cited text no. 19
Wynn JL. Defining neonatal sepsis. Curr Opin Pediatr 2016;28:135-40.  Back to cited text no. 20
Mithal LB, Palac HL, Yogev R, Ernst LM, Mestan KK. Cord blood acute phase reactants predict early onset neonatal sepsis in preterm infants. PLoS One 2017;12:e0168677.  Back to cited text no. 21
Gilfillan M, Bhandari V. Biomarkers for the diagnosis of neonatal sepsis and necrotizing enterocolitis: Clinical practice guidelines. Early Hum Dev 2017;105:25-33.  Back to cited text no. 22


  [Figure 1], [Figure 2]

  [Table 1], [Table 2]

This article has been cited by
1 Potential biomarkers for diagnosing neonatal sepsis
Ankita Sharma,Anup Thakur,Chitra Bhardwaj,Neelam Kler,Pankaj Garg,Manvender Singh,Sangeeta Choudhury
Current Medicine Research and Practice. 2020;
[Pubmed] | [DOI]


Similar in PUBMED
  Search Pubmed for
  Search in Google Scholar for
Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded485    
    Comments [Add]    
    Cited by others 1    

Recommend this journal