Clinical Chemistry
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 49: 60-68, 2003; 10.1373/49.1.60
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (37)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chiesa, C.
Right arrow Articles by Pacifico, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chiesa, C.
Right arrow Articles by Pacifico, L.
Related Collections
Right arrow Evidence Based Laboratory Medicine and Test Utilization
Right arrow Proteomics and Protein Markers
Right arrow Endocrinology and Metabolism
(Clinical Chemistry. 2003;49:60-68.)
© 2003 American Association for Clinical Chemistry, Inc.

C-Reactive Protein, Interleukin-6, and Procalcitonin in the Immediate Postnatal Period: Influence of Illness Severity, Risk Status, Antenatal and Perinatal Complications, and Infection

Claudio Chiesa1,2a, Gabriella Pellegrini3, Alessandra Panero2, John F. Osborn4, Fabrizio Signore5, Marcello Assumma3 and Lucia Pacifico1,2

1 National Research Council, 00161 Rome, Italy.
Institutes of
2 Pediatrics and
4 Hygiene, "La Sapienza" University of Rome, 00161 Rome, Italy.
Divisions of
3 Neonatology and
5 Obstetrics, S. Camillo Hospital, 00152 Rome, Italy.

aAddress correspondence to this author at: Institute of Pediatrics, "La Sapienza" University of Rome, Viale R. Elena, 324 00161 Rome, Italy. Fax 39-06-4997-9216; e-mail Claudio.Chiesa{at}Uniroma1.it.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Studies of the diagnostic accuracy of most laboratory tests for early-onset neonatal sepsis have yielded variable results. We investigated whether some of this variation might be attributable to differences in population baseline severity and risk status as well as to specific ante- and perinatal variables, independent of the presence of neonatal infection.

Methods: The Score for Neonatal Acute Physiology (SNAP) was used to define illness severity, with SNAP Perinatal Extension (SNAP-PE) used to define the combined physiologic and perinatal mortality risk. A total of 134 ill newborns (19 with early-onset infection and 115 with no infection) were available for simultaneous analysis of the association of SNAP, SNAP-PE, and maternal and perinatal variables with C-reactive protein (CRP), interleukin-6 (IL-6), and procalcitonin (PCT) concentrations at birth and at 24 and 48 h of life.

Results: Early-onset neonatal infection was associated with significant increases in CRP, IL-6, and PCT concentrations at all three time points, independent of illness severity. However, among babies without infection, higher SNAP and SNAP-PE scores were associated with higher IL-6 concentrations at birth. Certain maternal or perinatal variables altered IL-6 and PCT values in the infected as well as in the uninfected neonates. However, if different cutoff points were used at any of the three neonatal ages, PCT sensitivity and specificity were greater than those of CRP or IL-6.

Conclusions: Illness severity and risk status are unlikely to interfere with the use of CRP and PCT for detection of early-onset neonatal sepsis. In contrast, the diagnostic value of IL-6 at birth may be altered by physiologic severity and risk indexes. The reliability of CRP, IL-6, and PCT for the diagnosis of early-onset neonatal infection requires specific cutoff values for each evaluation time point over the first 48 h of life.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Diagnosis of systemic infection is a routine challenge in neonatal intensive care settings, where the distinction between infectious and noninfectious causes of inflammation is often difficult. Faced with the uncertainty of currently available laboratory aids for the diagnosis of neonatal sepsis (1), we asked whether interpretation of the many studies in that field might have been hampered by the severity of the underlying illnesses that modulate the host response. We therefore investigated whether baseline illness severity and risk status might affect circulating concentrations of new and traditional inflammatory mediators, such as interleukin-6 (IL-6),1 procalcitonin (PCT), and C-reactive protein (CRP), independent of infection in the period 0–48 h of age. This is the critical decision time for most neonatal sepsis "work-ups", especially in the era of early newborn discharge (2). The study also provided a unique opportunity to analyze in the same patient population which maternal and perinatal factors might confound the interpretation of all three laboratory aids, regardless of diagnosis of early-onset (<=48 h of life) infection.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Our study population consisted of 219 consecutive critically ill inborns who were enrolled over a 6-month period (July to December 2001) in the delivery room of two major maternity hospitals (Saint Camillo and Policlinico Umberto I°, Rome, Italy). This subgroup was selected because all such infants are routinely admitted to the level III neonatal intensive care units (NICUs) of the same maternity hospitals. The attending neonatologists from the two participating NICUs performed all care from the delivery room or admission until discharge. We excluded from enrollment: (a) 15 preadmission deaths, including 9 delivery room deaths and 6 moribund infants for whom few tests and vital signs were obtained at the time of NICU admission; (b) 3 babies with inevitably lethal abnormalities; and (c) 16 babies whose cord blood was not sampled for all three study markers. The final population enrolled was 185 neonates, none of whom died during the 48-h study period.

Illness severity was measured using the Score for Neonatal Acute Physiology (SNAP) (3)(4), which uses the worst recorded values of >24 routinely measured physiologic variables (mean blood pressure, heart rate, respiratory rate and temperature, blood gas data, blood chemistries and blood counts, urine output, presence of seizures, apnea, stool guaiac, base excess, PO2/fraction of inspired oxygen ratio, and oxygenation index) during the first 24 h of stay in the NICU. The degree of derangement from physiologic normal is given a weighted score of 0 points for no derangement, 1 point for a clearly abnormal result that merits careful monitoring, 3 points for a severe derangement that requires immediate intervention, and 5 points for an acute life-threatening value. SNAP is the weighted sum of derangements across all organ systems; its contribution to overall severity, therefore, is proportionate (5). SNAP Perinatal Extension (SNAP-PE) supplements the SNAP with additional scoring for three potent perinatal mortality risks [i.e., birth weight, low Apgar score (<7 at 5 min), and small for gestational age (<5th percentile) status], all of which are independent of physiologic derangements (6). Thus, SNAP-PE score represents the combined physiologic and perinatal mortality risk. Data for SNAP and SNAP-PE scores were collected prospectively by the attending neonatologists.

In addition to SNAP and SNAP-PE, ante- and intrapartum data were recorded for every newborn and included maternal age, race, parity, prenatal care, multiple pregnancy, preexistent or pregnancy-related diseases (hypertension, diabetes, preeclampsia), medications during pregnancy, drug abuse, prenatal steroid exposure, maternal group B streptococcus (GBS) colonization, duration and characteristics of rupture of amniotic membranes, clinical evidence of chorioamnionitis and antibiotic treatment, placental abnormalities, fetal presentation, abnormalities in intrapartum fetal heart monitoring, mode of delivery, duration of active labor, use of anesthesia, and birth asphyxia. Infants were considered asphyxiated at birth if immediately after birth they needed endotracheal positive-pressure ventilation and/or cardiopulmonary resuscitation.

Institutional Review Board approval in both hospitals is not required for studies involving blood sampling from symptomatic individuals, but informed consent to perform the investigation was obtained from the infants’ parents or guardians. Over the study period, investigators implemented in both NICUs routine blood sampling for CRP as well as for IL-6 and PCT at 24 and 48 h of life. However, insufficient blood samples were obtained from 31 infants, leaving 154 NICU newborns potentially available for analysis.

At the time of initial evaluation, chest radiographs and blood samples from peripheral veins were obtained for every NICU patient for routine culture of aerobic and anaerobic bacteria, ureaplasmas, and Mycoplasma hominis. Cultures of cerebrospinal fluid and urine were performed when appropriate. In addition, tracheal samples from intubated patients were cultured for mycoplasmas and, when obtained within 8 h after birth, for aerobic bacteria (7). Screening for IgM antibodies to Toxoplasma, rubella virus, cytomegalovirus, herpes simplex virus, and Treponema pallidum was done in the first few postnatal days if congenital infection seemed likely.

Designation of infection status had to be made retrospectively because all neonates were considered at risk for, or demonstrated clinical evidence of, infection. Those of us (G.P., A.P., and L.P., all nationally certified in Pediatrics and Neonatology) responsible for classifying the infants as septic vs nonseptic were blind to the CRP, IL-6, and PCT values. The risk factors for infection included maternal fever, prolonged rupture of membranes (PROM) for >=18 h; maternal GBS colonization; uterine tenderness; foul smelling, cloudy amniotic fluid; preterm labor; and intrapartum antimicrobial administration. The 154 NICU neonates were classified as belonging to one of the following three groups: group 1, early-onset infection; group 2, no infection; and group 3, uncertain. Group 1 was composed of 19 (9 term and 10 preterm) infants, 11 of whom had a positive culture of one or more blood specimens drawn within the first 48 h of life in association with clinical signs of infection. Organisms isolated from blood cultures included GBS (n = 7) or Escherichia coli (n = 4). The remaining eight patients with negative body fluid cultures had definite, persistent clinical signs of sepsis prompting >=5 days of antibiotic treatment (8) (with the exception of one infant who died on the third day of life) and had certain historic and clinical factors associated with increased risk for infection. Definite clinical signs of infection were defined as the presence of three or more of the following categories of clinical signs derived from a validated sepsis score: (a) temperature instability (hypothermia, hyperthermia); (b) respiratory (grunting, intercostal retractions, apnea, tachypnea, cyanosis); (c) cardiovascular (bradycardia, tachycardia, poor perfusion, hypotension); (d) neurologic (hypotonia, lethargy, seizures); and (e) gastrointestinal (feeding intolerance, abdominal distension) (9). All eight infants with strong evidence of infection (or "clinical septicemia") were delivered of mothers who had received antibiotics before delivery because of clinical evidence of chorioamnionitis (n = 4), vaginal cultures persistently positive for GBS (n = 2), or preterm PROM (n = 2). Five of the eight infants also had radiographic findings consistent with sepsis (including pulmonary infiltrates or pleural effusions).

Group 2 was composed of 115 symptomatic babies who had negative body fluid cultures, were apparently well within 24–48 h, and had a benign clinical course until discharge. Infants belonging to this category received antibiotic treatment for 3 days or less.

Group 3 included 20 infants in whom systemic infection could be neither confirmed nor excluded. Fifteen had negative body fluid cultures and fewer than three categories of clinical signs, and 5 babies each had one blood culture yielding only typical skin or upper respiratory flora and either had no clinical signs of infection or recovered without specific therapy. Group 3 patients who were excluded from the study analysis had geometric mean CRP concentrations of 4.1 [95% confidence interval (CI), 3.0–5.5] mg/L, 5.6 (3.9–8.3) mg/L, and 5.2 (3.5–7.8) mg/L at birth, 24, and 48 h of age, respectively. The corresponding PCT values were 0.27 (0.12–0.58) µg/L, 15.4 (6.3–37.6) µg/L, and 5.6 (2.6–12.0) µg/L, and the IL-6 values were 40.3 (14.5–111.6) ng/L, 13.4 (5.2–32.7) ng/L, and 7.3 (4.4–13.2) ng/L.

A total of 134 NICU newborns (of whom 19 had early-onset infection and 115 no infection) were available for investigating the associations of SNAP and SNAP-PE scores as well as maternal and perinatal factors with each study marker in the infected and uninfected neonate at birth and 24 and 48 h of life.

il-6, pct, and crp measurements
Umbilical cord and postnatal serum samples for duplicate IL-6 (100 µL) and PCT (40 µL) determinations were stored in small aliquots at -70 °C until analysis. IL-6 concentrations were measured by an enzyme-linked immunoassay with a detection limit <1 ng/L (Endogen) (10). The within-run imprecision (CV; n = 12) was 5.4% at 10.24 ng/L and 4.6% at 400 ng/L, whereas the day-to-day imprecision (15 determinations) was 3.2% at 10.24 ng/L and 2.7% at 400 ng/L. PCT was determined by the LUMItest PCT method (BRAHMS Diagnostica) as described in detail elsewhere (11)(12). The lower limit of detection was 0.1 µg/L and the within-run imprecision (CV; n = 20) was 6.3% at 0.4 µg/L and 2.7% at 43.2 µg/L, whereas the day-to-day imprecision (40 determinations) was 13.4% at 0.5 µg/L and 7.1% at 34.2 µg/L. CRP was measured by rate nephelometry using a Beckman Array System protein analyzer (C-reactive protein reagent set 449760; Beckman Instruments) (10). The lower limit of detection was 4 mg/L, and the within-run imprecision (CV; n = 20) was 3.7% at 17.8 mg/L and 1.1% at 79.4 mg/L, whereas the day-to-day imprecision (20 determinations) was 3.9% at 7.4 mg/L and 3.0% at 90.8 mg/L.

statistical analysis
The measured CRP and IL-6 values were distributed with a long tail to the right (positive skew), but their logarithms were approximately normally distributed. Thus, all comparisons of CRP, IL-6, and PCT values and all regression analyses were done after logarithmic transformation. Consequently, all quoted mean values are geometric means with 95% CIs, and regression coefficients have been exponentiated to obtain ratios of geometric means. The means of infected and uninfected neonates were compared by the t-test.

We used log-linear regression to investigate the association between disease severity and concentrations of the markers CRP, IL-6, and PCT. We first regressed the logarithm of the concentration on the SNAP score. We then repeated the regression with a dummy variable to indicate the presence or absence of infection and repeated it again to take into account the possibility that the relationship between the log of the concentration and the SNAP score was not linear. The SNAP score was categorized in three intervals (0–9, 10–19, and >=20), and the regression was repeated using two dummy variables to represent SNAP category (SNAP score 0–9 being the reference category) and an additional dummy variable to indicate the presence or absence of infection.

Each of the three markers was measured three times, at birth and at 24 and 48 h of life; thus, a total of nine associations were investigated. For every association, the regression analyses produced six regression coefficients (one for the first regression, two for the second, and three for the third regression); thus, this part of the analysis yielded 54 significance tests of regression coefficients. This analysis was repeated with the SNAP-PE score. Thus, 108 statistical significance tests were performed. If a significance (P value) of 0.05 were considered to indicate evidence of an association, 5% of the results would be expected to be statistically significant even when the null hypothesis was true. This problem could be partially resolved, for example, by applying the Bonferroni correction to the observed levels of significance, but in the present analysis, there were further complications because the tests were clearly not independent. For this reason, the correction was not applied, but this implies that a plausible explanation of an otherwise unexpected statistically significant result at the P = 0.05 level could be just chance (13). For this reason, only the results of tests significant at P <=0.01 are highlighted in Tables 2–6 , although significance at P <=0.05 can be deduced from the CIs.


View this table:
[in this window]
[in a new window]
  		Table 2. Independent multiplicative effects of infection and either SNAP or SNAP-PE score on CRP concentration at 0, 24, and 48 h of life.1


View this table:
[in this window]
[in a new window]
  		Table 3. Independent multiplicative effects of infection and either SNAP or SNAP-PE score on PCT concentration at 0, 24, and 48 h of life.1


View this table:
[in this window]
[in a new window]
  		Table 4. Independent multiplicative effects of infection and either SNAP or SNAP-PE score on IL-6 concentration at 0, 24, and 48 h of life.1


View this table:
[in this window]
[in a new window]
  		Table 5. Independent multiplicative effects of infection and antepartum/intrapartum variables on CRP, IL-6, and PCT at 0, 24, and 48 h of life.1


View this table:
[in this window]
[in a new window]
  		Table 6. Performance of CRP, IL-6, and PCT measurements at 0, 24, and 48 h of life in diagnosis of early-onset neonatal sepsis.

The role of ante- and intrapartum variables [preexistent or pregnancy-related diseases (hypertension, diabetes, preeclampsia), medications during pregnancy, drug abuse, prenatal steroid exposure, maternal GBS colonization, duration and characteristics of rupture of amniotic membranes, clinical evidence of chorioamnionitis and antibiotic treatment, placental abnormalities, fetal presentation, abnormalities in intrapartum fetal heart monitoring, mode of delivery, duration of active labor, use of anesthesia, and the need for delivery room intubation and/or cardiopulmonary resuscitation] was investigated by calculating the multiple linear regressions of the logarithms of the concentrations of the markers CRP, IL-6, and PCT on the variables and the presence of infection. The exponentials of the regression coefficients are estimates of the independent multiplicative effects of the variables.

At each of the three neonatal ages, the sensitivity and specificity of CRP, IL-6, and PCT for the diagnosis of early-onset neonatal sepsis were calculated. The ROC curve was used to determine the optimal cutoff points that maximized the sum of the sensitivity and specificity of each test. This is equivalent to maximizing the Youden index (14). The curves were obtained by plotting sensitivity on the y axis against the false-positive rate (1 - specificity) on the x axis for all possible cutoff values of the tests.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The maternal and neonatal characteristics of the study population are presented in Table 1 . There were no significant differences between infected and uninfected infants with respect to gestational age [mean (SD), 33.6 (6.1) vs 33.8 (3.8) weeks], birth weight [2220 (1160) vs 2141 (859) g], male gender (13 of 19 vs 70 of 115), and small for gestational age status (1 of 19 vs 8 of 115). However, infected infants had a lower 5-min Apgar score [6.8 (1.8) vs 7.8 (1.4); P <0.01] and a higher admission score of illness severity [32.3 (15.6) vs 6.2 (12.0); P <0.0001] than did uninfected infants. The distributions of crude CRP, IL-6, and PCT values among the infected (group 1) and uninfected (group 2) patients at the three neonatal ages are shown in panels A, B, and C of Fig. 1 , respectively. Among infected babies, CRP, IL-6, and PCT values were not significantly different between those with confirmed and those with clinical sepsis at any of the three time points (data not shown).


View this table:
[in this window]
[in a new window]
  		Table 1. Maternal and neonatal characteristics.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Distribution of crude CRP (A), IL-6 (B), and PCT (C) among 19 infected (Group 1) and 115 uninfected (Group 2) babies at the three neonatal ages 0, 24, and 48 h.

{blacktriangleup} indicates 10 infants; • indicates 1 infant. Dashed lines indicate detection limits of assays; solid lines indicate medians (where not shown, the medians are lower than the detection limits).

When the data were analyzed after a logarithmic transformation, geometric mean CRP values were significantly higher in the infected than in the uninfected neonates at birth [12.0 (95% CI, 7.1–20.3) vs 3.3 (3.1–3.5) mg/L; P <0.0001], at 24 h [37.0 (23.6–58.0) vs 4.7 (4.1–5.4) mg/L; P <0.0001], and at 48 h of age [33.5 (18.1–62.2) vs 4.5 (4.0–5.1) mg/L; P <0.0001]. Likewise, geometric mean PCT values were significantly higher in the infected than in the uninfected babies at birth [3.79 (1.70–8.43) vs 0.22 (0.18–0.27) µg/L; P <0.0001], at 24 h [255.2 (164.3–396.4) vs 5.60 (4.28–7.30) µg/L; P <0.0001], and at 48 h of age [100.9 (65.3–156.0) vs 2.15 (1.69–2.71) µg/L; P <0.0001]. At birth, geometric mean IL-6 values were 622.7 (203.4–621.9) ng/L in the sepsis group compared with 19.9 (16.1–26.6) ng/L in the nonsepsis group (P <0.0001). At 24 h, geometric mean IL-6 values still were higher in septic than in nonseptic infants [45.0 (12.7–158.8) vs 11.6 (8.8–15.0) ng/L; P <0.01]. However, at 48 h of age, there were no significant differences in geometric mean IL-6 values between infected and uninfected babies [14.9 (5.6–39.5) vs 8.9 (7.3–11.7) ng/L].

Given that the SNAP-PE score is calculated by adding perinatal factors to those that determine the SNAP score, the two scores were highly correlated; the Pearson coefficient of linear correlation was 0.945, which implied that the associations between the markers CRP, IL-6, and PCT and the SNAP-PE score were likely to be similar to those found for the SNAP score. The log CRP value was positively associated with both SNAP and SNAP-PE at each of the three time points. However, this association with SNAP and SNAP-PE disappeared when the presence of infection was included in the regression model. The presence of infection increased the log CRP values independently of the SNAP and SNAP-PE scores. When the analysis was confined to the babies without infection, we found no association between SNAP or SNAP-PE and the log CRP value. Similarly, there was no association between SNAP or SNAP-PE and the log CRP values for the babies with infection. The data in Table 2Up imply that only the presence of infection affects the log CRP value. The relationship between log PCT and SNAP score was generally similar to that found for CRP (Table 3Up ). At all three times, the log PCT was significantly increased for babies with infection independent of the severity scores. However, PCT increased 12.5- to 36.7-fold during the course of infection (Table 3Up ), whereas CRP increased 3.5- to 8.0-fold (Table 2Up ). The log IL-6 value was associated with the presence of infection at birth and at 24 h, independent of the SNAP and SNAP-PE scores (Table 4Up ). Among the babies without infection, there was an association between log IL-6 and SNAP and SNAP-PE scores, particularly at birth, but this was not evident among infected babies.

Shown in Table 5Up are the factors by which the concentrations of the markers were multiplied in the presence of ante- and intrapartum complications. No variable was found to be independently associated with the log CRP value. PCT response was independently associated at birth with clinical amnionitis, at 24 h with delivery room intubation, and at both 24 and 48 h of age with preeclampsia. The only variable independently associated with IL-6 at time 0 was delivery room intubation. The increases associated with these variables were relatively small compared with those observed for the presence of infection, particularly in the case of PCT.

The sensitivities and specificities of CRP, IL-6, and PCT in early-onset neonatal sepsis are shown in Table 6Up . The cutoff value that maximized the sum of the sensitivity and specificity for CRP was >=4 mg/L at birth, whereas at both 24 and 48 h of life, it was >=10 mg/L. The cutoff values for IL-6 were >=200, >=30, and >=20 ng/L at birth, at 24 h, and at 48 h of age, respectively. At birth, at 24 h, and at 48 h, the PCT cutoff values were >=1, >=100, and >=50 µg/L, respectively.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The systemic inflammatory response (SIR) that follows tissue injury or infection orchestrates a wide variety of systemic and metabolic responses that, in most instances, serve to restore homeostasis. In critically ill newborns, the diagnostic repertoire for identifying infectious SIR is poor. The newborn infant responds to a wide variety of noxious stimuli (infectious, metabolic, respiratory, traumatic) with a limited repertoire of stereotyped reactions (15). Many of the manifestations of sepsis have their counterparts in noninfectious neonatal disorders. In view of the relatively poor predictive value of clinical variables and SIR for microbial infection and mortality of such patients, we evaluated the value of inflammatory markers whose concentrations may correspond to infection status.

The reliability of most laboratory markers, including CRP, IL-6, and PCT, for the differential diagnosis of infectious vs noninfectious SIR has been assessed in highly diverse groups of ill neonates with a mixture of diagnoses and conditions and has yielded variable results (1). It may be that some of this variation among published reports might reflect differences in baseline severity and risk status as well as specific factors acting during gestation and delivery independently of the presence of infection. There is wide variation in clinical severity among NICUs. Admissions may range from critically ill infants with multiple organ-system failure to mildly ill term infants with transient problems related to the birth process or to healthy premature infants who require technologic support until mature. Moreover, comparability of risk scores among institutions may be diminished by variation in disease severity for a given diagnosis.

In the context of newborn intensive care, birth weight and gestational age have generally been used as simple proxies for more elaborate measures of illness severity. Although birth weight and gestational age may be adequate for some purposes, they do not completely account for variations in the severity of illness. The SNAP score, designed to permit objective quantification of clinical severity in NICUs, is a validated, objective physiology-based score that captures the fundamental construct of newborn admission illness severity (3). It is clinically intuitive: the higher the SNAP, the sicker the infant (2). It has been prospectively validated in multiple NICUs covering very different population groups and found be highly correlated with other indicators of illness severity (3)(16)(17)(18). The advantage of the SNAP score is that it is based on the physiologic status of the baby within a given time frame (usually the first 24 h of admission) so that variations in diagnosis (or therapy) have little impact on the score itself. From this, SNAP-PE captures SNAP physiology scores, combining them with other risk factors to provide an overall risk of mortality (5).

The findings reported here indicate that baseline illness severity and risk status are a prominent component of both CRP and PCT responses at birth and at 24 and 48 h of life only in the presence of infection (Tables 2Up and 3Up ). Although it remains to be verified that PCT is synthesized in hepatocytes, our results may complement those of Nijsten et al. (19), who found that PCT and acute-phase proteins such as CRP are induced by similar pathways.

It is known that IL-6 is an important mediator of host response to stress and infection (20). Our results show that among babies without infection, the higher the SNAP and SNAP-PE scores, the greater the IL-6 response at birth. Therefore, the diagnostic value of IL-6 at birth may be altered by the physiologic severity and risk indexes. Our observations are similar to those made by deWerra et al. (21) in adult patients with septic or cardiogenic shock; the authors found that IL-6 concentrations were increased in both patient populations. In rodents, catecholamines and epinephrine in particular have been shown to increase IL-6 production (22)(23).

Previous studies have sought to identify which pre- and perinatal complications would mimic or mask alterations in the CRP response caused by infection but have yielded conflicting results (1)(24). Kushner et al.(25) found that maternal fever during labor or delivery, PROM, perinatal asphyxia, and other problems not resulting from infection are associated with increased CRP in umbilical cord blood. In a NICU population with wide-ranging differences in age, CRP was increased in those with a history of PROM, fetal distress, or chorioamnionitis (26). Ainbender et al. (27) identified fetal distress, low 1-min Apgar score, gestational diabetes, maternal drug addiction, maternal fever, and PROM as confounding factors in the diagnosis of early-onset neonatal infection. Schouten-Van Meeteren et al. (28), however, did not observe differences in CRP concentrations between perinatally asphyxiated infants and a control group at 12–24 h after birth. It is evident from our data that clinical amnionitis and the need for delivery room intubation may significantly increase CRP concentrations during the immediate postnatal period. However, neither variable appeared to be an independent factor that could confound interpretation of CRP measurement in the diagnosis of early-onset infection.

Although previous studies found that inflammatory cytokines, including IL-6, are released in uninfected infants with perinatal complications, the results obtained to date remain discordant (20)(29)(30). Our results are consistent with those of Jokic et al. (30), who showed that length of labor is associated with increased cord blood concentrations of cytokines, including IL-6, in uninfected full-term neonates. Our results are also consistent with those of Singh et al. (31), who showed that IL-6 concentrations in cord blood increased with clinical chorioamnionitis despite a lack of evidence of infection in the neonates. However, in our study, there was no association between these variables and IL-6 response even after adjustment for the presence of infection. In contrast, birth asphyxia was the only variable that independently increased cord IL-6 concentrations. The data available on the role of IL-6 in the injury response to hypoxia ischemia or asphyxia are currently limited (32).

Few studies have assessed whether maternal and perinatal complications may affect PCT values in the early postnatal period. In a previous study we demonstrated that maternal GBS colonization and rupture of membranes >=18 h may significantly affect PCT concentrations at birth and at 24 and 48 h of life in a group of healthy neonates (12). Thus it is not surprising that in the presence of clinical amnionitis, PCT concentrations at birth were significantly increased independent of the presence of infection. Our results are therefore partially consistent with those of Janota et al. (33), who reported a significant increase in serum PCT concentrations within 72 h of age in preterm infected and uninfected newborns born to mothers with chorioamnionitis. As suggested by Gendrel and Bohuon (34), hypoxemia may be responsible for increased PCT values in neonates. Similarly, we have found that birth asphyxia may affect PCT concentrations. Finally, a novel finding of our study is that neonates born to mothers with preeclampsia had higher PCT concentrations at both 24 and 48 h of life (Table 5Up ). Although certain variables altered PCT values in the infected as well as in the uninfected neonates, when different cutoff points were used at the three neonatal ages, PCT specificity for infection was greater than that for CRP or IL-6. Compared with the increases in PCT caused by ante- and perinatal events, the magnitude of PCT response to infection was much greater at each of the three neonatal ages.

The sensitivity of CRP in initial determinations for the detection of early-onset neonatal infection has been reported to vary from 35% to 65%, increasing to 74–97% by the time of the third determination. CRP specificity varied in initial determination from 92% to 96% and decreased to 76–86% for the third measurement (35)(36)(37)(38). IL-6 in umbilical cord blood has been reported to be consistently increased in newborn infants who develop sepsis within 48 h of birth, with specificities ranging from 88% to 93% (39). The sensitivity of PCT in initial determinations for the diagnosis of early-onset neonatal sepsis has been reported to vary from 61% to 85%, increasing to 72–100% within the subsequent 24 h. PCT specificity varied in initial determinations from 50% to 97% and ranged between 63% and 97% within the next 24 h (11)(40)(41). Possible factors accounting for such wide ranges include different CRP (1.5–20 mg/L), IL-6 (25–150 ng/L), and PCT (0.5–5 µg/L) cutoff points; different measurement methods; variations in study design or in definition of infection; and wide-ranging differences in postnatal age (hours to weeks) or inaccuracies in reporting it (1)(24)(39)(42). In 1980, Philip and Hewitt (43) suggested that, for simplicity, the same laboratory cutoffs for markers used in the diagnosis of neonatal sepsis should be applied during the first postnatal week. Our present findings suggest that failure to recognize the appropriate cutoff concentration for each time point of evaluation over the first 48 h of life may confound interpretation of what constitutes a "true negative" and a "true positive" value in the diagnosis of early-onset neonatal infection. In our study, when we used different cutoff values at each of the three ages, the sensitivities and specificities of PCT were greater than those obtained for CRP or IL-6.


   Footnotes
 
1 Nonstandard abbreviations: IL-6, interleukin-6; PCT, procalcitonin; CRP, C-reactive protein; NICU, neonatal intensive care unit; SNAP, Score for Neonatal Acute Physiology; SNAP-PE, SNAP-Perinatal Extension; GBS, group B streptococcus; PROM, prolonged rupture of membranes; CI, confidence interval; and SIR, systemic inflammatory response.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Weinberg GA, Powel KR. Laboratory aids for diagnosis of neonatal sepsis. Remington JS Klein JO eds. Infectious diseases of the fetus and newborn infant 2001:1327-1344 WB Saunders Philadelphia. .
  2. Escobar GJ. The neonatal "sepsis work-up": personal reflections on the development of an evidence-based approach toward newborn infections in a managed care organization. Pediatrics 1999;103:360-373.
  3. Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA. Score for neonatal acute physiology: a physiologic severity index for neonatal intensive care. Pediatrics 1993;91:617-623.[Abstract/Free Full Text]
  4. Richardson DK, Tarnow-Mordi WO. Measuring illness severity in newborn intensive care. J Intensive Care Med 1994;9:20-33.
  5. Richardson D, Tarnow-Mordi WO, Lee SK. Risk adjustment for quality improvement. Pediatrics 1999;103:255-265.
  6. Richardson DK, Phibbs CS, Gray JE, McCormick MC, Workman-Daniels K, Goldmann DA. Birth weight and illness severity: independent predictors of neonatal mortality. Pediatrics 1993;91:969-975.[Abstract/Free Full Text]
  7. Sherman MP, Goetzman BW, Ahlfors CE, Wennberg RP. Tracheal aspiration and its clinical correlates in the diagnosis of congenital pneumonia. Pediatrics 1980;65:258-263.[Abstract/Free Full Text]
  8. Stoll BJ, Gordon T, Korones SB, Shankaran S, Tyson JE, Bauer CR, et al. Early-onset sepsis in very low birth weight neonates: a report from the National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1996;129:72-80.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  9. Tollner U. Early diagnosis of septicaemia in the newborn: clinical studies and sepsis score. Eur J Pediatr 1982;138:331-337.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  10. Chiesa C, Signore F, Assumma M, Buffone E, Tramontozzi P, Osborn JF, et al. Serial measurements of C-reactive protein and interleukin-6 in the immediate postnatal period: reference intervals and analysis of maternal and perinatal confounders. Clin Chem 2001;47:1016-1022.[Abstract/Free Full Text]
  11. Chiesa C, Panero A, Rossi N, Stegagno M, De Giusti M, Osborn JF, et al. Reliability of procalcitonin concentrations for the diagnosis of sepsis in critically ill neonates. Clin Infect Dis 1998;26:664-672.[Web of Science][Medline] [Order article via Infotrieve]
  12. Assumma M, Signore F, Pacifico L, Rossi N, Osborn JF, Chiesa C. Serum procalcitonin concentrations in term delivering mothers and their healthy offspring: a longitudinal study. Clin Chem 2000;46:1583-1587.[Abstract/Free Full Text]
  13. Armitage P, Berry G. Comparison of several groups. Statistical methods in medical research 1987:186-213 Blackwell Scientific Publications Oxford. .
  14. Armitage P, Berry G. Statistical methods in epidemiology. Statistical methods in medical research 1987:455-483 Blackwell Scientific Publications Oxford. .
  15. Sáez-Llorens X, McCracken GH, Jr. Perinatal bacterial diseases. Feigin RD Cherry JD eds. Textbook of pediatric infectious diseases 1998:892-926 WB Saunders Philadelphia. .
  16. Escobar GJ, Fischer A, Li DK, Kremers R, Armstrong MA. Score for neonatal acute physiology: validation in three Kaiser Permanente neonatal intensive care units. Pediatrics 1995;96:918-922.[Abstract/Free Full Text]
  17. Petridou E, Richardson DK, Dessypris N, Malamitsi-Puchner A, Mantagos S, Nicolopoulos D, et al. Outcome prediction in Greek neonatal intensive care units using a score for neonatal acute physiology (SNAP). Pediatrics 1998;101:1037-1044.[Abstract/Free Full Text]
  18. Pollack MM, Koch MA, Bartel DA, Rapoport I, Dhanireddy R, El-Mohandes AA, et al. A comparison of neonatal mortality risk prediction models in very low birth weight infants. Pediatrics 2000;105:1051-1057.[Abstract/Free Full Text]
  19. Nijsten MW, Olinga P, The TH, de Vries EG, Koops HS, Groothuis GM, et al. Procalcitonin behaves as a fast responding acute phase protein in vivo and in vitro. Crit Care Med 2000;28:458-461.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  20. Miller LC, Isa S, LoPreste G, Schaller JG, Dinarello CA. Neonatal interleukin-1ß, interleukin-6, and tumor necrosis factor: cord blood levels and cellular production. J Pediatr 1990;117:961-965.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  21. de Werra I, Jaccard C, Corradin SB, Chioléro R, Yersin B, Gallati H, et al. Cytokines, nitrite/nitrate, soluble tumor necrosis factor receptors, and procalcitonin concentrations: comparisons in patients with septic shock, cardiogenic shock, and bacterial pneumonia. Crit Care Med 1997;25:607-613.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  22. Takaki A, Huang QH, Somogyvari-Vigh A, Arimura A. Immobilization stress may increase plasma interleukin-6 via central and peripheral catecholamines. Neuroimmunomodulation 1994;1:335-342.[Medline] [Order article via Infotrieve]
  23. Liao J, Keiser JA, Scales WE, Kunkel SL, Kluger MJ. Role of epinephrine in TNF and IL-6 production from isolated perfused rat liver. Am J Physiol 1995;268:R896-R901.
  24. Jaye DL, Waites KB. Clinical applications of C-reactive protein in pediatrics. Pediatr Infect Dis J 1997;16:735-747.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  25. Kushner I, Sweet AY, Yen-Watson B, Ribich WN, Merk J. Significance of C-reactive protein (CRP) in cord blood [Abstract]. Pediatr Res 1973;3:403.
  26. Forest JC, Lariviere F, Dolce P, Masson M, Nadeau L. C-Reactive protein as biochemical indicator of bacterial infection in neonates. Clin Biochem 1986;19:192-194.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  27. Ainbender E, Cabatu EE, Guzman DM, Sweet AY. Serum C-reactive protein and problems of newborn infants. J Pediatr 1982;101:438-440.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  28. Schouten-Van Meeteren NYN, Rietveld A, Moolenaar AJ, Van Bel F. Influence of perinatal conditions on C-reactive protein production. J Pediatr 1992;120:621-624.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  29. Hata T, Kawamura T, Inada K, Fujiwaki R, Ariyuki Y, Hata K, et al. Interleukin-6, interleukin-8 and granulocyte elastase in newborns with fetal distress. Gynecol Obstet Invest 1996;42:174-177.[Web of Science][Medline] [Order article via Infotrieve]
  30. Jokic M, Guillois B, Cauquelin B, Giroux JD, Bessis JL, Morello R, et al. Fetal distress increases interleukin-6 and interleukin-8 and decreases tumour necrosis factor-{alpha} cord blood levels in noninfected full-term neonates. BJOG 2000;107:420-425.[Medline] [Order article via Infotrieve]
  31. Singh B, Merchant P, Walker CR, Kryworunchko M, Diaz-Mitoma F. Interleukin-6 expression in cord blood of patients with clinical chorioamnionitis. Pediatr Res 1996;39:976-979.[Web of Science][Medline] [Order article via Infotrieve]
  32. Sävman K, Blennow M, Gustafson K, Tarkowski E, Hagberg H. Cytokine response in cerebrospinal fluid after birth asphyxia. Pediatr Res 1998;43:746-751.[Web of Science][Medline] [Order article via Infotrieve]
  33. Janota J, Stranák Z, Belohlávková S, Mudra K, Simák J. Postnatal increase of procalcitonin in premature newborns is enhanced by chorioamnionitis and neonatal sepsis. Eur J Clin Invest 2001;31:978-983.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  34. Gendrel D, Bohuon C. Procalcitonin, a marker of bacterial infection. Infection 1997;25:133-134.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  35. Benitz WE, Han MY, Madan A, Ramachandra P. Serial serum C-reactive protein levels in the diagnosis of neonatal infection. Pediatrics 1998;102:e41.[Abstract/Free Full Text]
  36. Messer J, Eyer D, Donato L, Gallati H, Matis J, Simeoni U. Evaluation of interleukin-6 and soluble receptors of tumor necrosis factor for early diagnosis of neonatal infection. J Pediatr 1996;129:574-580.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  37. Pourcyrous M, Bada HS, Korones SB, Baselski V, Wong SP. Significance of serial C-reactive protein responses in neonatal infection and other disorders. Pediatrics 1993;92:431-435.[Abstract/Free Full Text]
  38. Berger C, Uehlinger J, Ghelfi D, Blau N, Fanconi S. Comparison of C-reactive protein and white blood cell count with differential in neonates at risk for septicaemia. Eur J Pediatr 1995;154:138-144.[Web of Science][Medline] [Order article via Infotrieve]
  39. Mehr S, Doyle LW. Cytokines as markers of bacterial sepsis in newborn infants: a review. Pediatr Infect Dis J 2000;19:879-887.[Web of Science][Medline] [Order article via Infotrieve]
  40. Franz AR, Kron M, Pohlandt F, Steinbach G. Comparison of procalcitonin with interleukin 8, C-reactive protein and differential white blood cell count for the early diagnosis of bacterial infections in newborn infants. Pediatr Infect Dis J 1999;18:666-671.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  41. Lapillonne A, Basson E, Monneret G, Bienvenu J, Salle BS. Lack of specificity of procalcitonin for sepsis diagnosis in premature infants. Lancet 1998;351:1211-1212.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  42. Chiesa C, Pacifico L, Mancuso G, Panero A. Procalcitonin in pediatrics: overview and challenge. Infection 1998;26:236-241.[Web of Science][Medline] [Order article via Infotrieve]
  43. Philip AGS, Hewitt JR. Early diagnosis of neonatal sepsis. Pediatrics 1980;65:1036-1041.[Abstract/Free Full Text]



The following articles in journals at HighWire Press have cited this article:


Home page
 Arch. Dis. Child. Fetal Neonatal Ed.Home page
U K Mishra, S E Jacobs, L W Doyle, and S M Garland
Newer approaches to the diagnosis of early onset neonatal sepsis.
Arch. Dis. Child. Fetal Neonatal Ed., May 1, 2006; 91(3): F208 - F212.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
S. D. Carrigan, G. Scott, and M. Tabrizian
Toward Resolving the Challenges of Sepsis Diagnosis
Clin. Chem., August 1, 2004; 50(8): 1301 - 1314.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
C. Chiesa, A. Panero, J. F. Osborn, A. F. Simonetti, and L. Pacifico
Diagnosis of Neonatal Sepsis: A Clinical and Laboratory Challenge
Clin. Chem., February 1, 2004; 50(2): 279 - 287.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (37)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chiesa, C.
Right arrow Articles by Pacifico, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chiesa, C.
Right arrow Articles by Pacifico, L.
Related Collections
Right arrow Evidence Based Laboratory Medicine and Test Utilization
Right arrow Proteomics and Protein Markers
Right arrow Endocrinology and Metabolism

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS