| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
National Research Council, 00161 Rome, Italy.
2
Institute of Pediatrics, La Sapienza University of Rome, 00161 Rome, Italy.
3
Institute of Hygiene, La Sapienza University of Rome, 00161 Rome, Italy.
4
Division of Obstetrics, S. Camillo Hospital, 00152 Rome, Italy. \ %
5
Division of Neonatology, S. Camillo Hospital, 00152 Rome, Italy.
aAddress correspondence to this author at: Institute of Pediatrics, La Sapienza University of Rome, Viale Regina Elena, 324 00161 Rome, Italy. Fax 39-06-49-218-480; e-mail Claudio.Chiesa{at}Uniroma1.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Methods: CRP and IL-6 values were prospectively obtained for 148 healthy babies (113 term, 35 near-term) at birth and at 24 and 48 h of life, and from their mothers at delivery.
Results: Upper reference limits for CRP at each neonatal age were
established. At birth, CRP was significantly lower than at 24 and
48 h of life. Rupture of membranes 
18 h, perinatal distress, and
gestational hypertension significantly affected the neonatal CRP
dynamics, but at specific ages. There was no correlation between CRP
concentrations in mothers and their offspring at birth. The IL-6 values
observed in the delivering mothers and in their babies at all three
neonatal ages were negatively associated with gestational age. In the
immediate postnatal period, IL-6 dynamics for term babies were
significantly different from those for near-term babies. Maternal IL-6
concentrations correlated with babies’ IL-6 concentrations only for
term deliveries. Apgar score had a significant effect on babies’ IL-6
values at birth.
Conclusions: The patterns of CRP and IL-6 responses in the healthy neonate should be taken into account to optimize their use in the diagnosis of early-onset neonatal sepsis.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
48 h
of life) neonatal infection have prompted the development of several
screening tests, including C-reactive protein
(CRP),1
a very commonly used marker. A delay of at least several hours
is intrinsic to the cascade of events leading to increased serum CRP;
therefore, the predictive value of CRP improves with time and is best
between 24 and 48 h after infection is suspected
(1)(2). Serial measurements are therefore
recommended (3)(4)(5)(6)(7). However, although serial CRP
measurements have an excellent negative predictive value for the
presence of infection, the reverse does not hold true, especially for
culture-confirmed early-onset infection (1). On the other
hand, the use of CRP in the first few days of life is complicated by a
nonspecific, 2- to 3-day increase that may greatly reduce the positive
predictive value of CRP determinations (3)(8). In recent years, the search for diagnostic tests for early-onset sepsis in newborns has turned to cytokines, alone or in combination with CRP, based on the premise that their increases in response to infection may precede that of CRP (9). As a cytokine, interleukin-6 (IL-6) is the major inducer of hepatic protein synthesis including CRP. Increased IL-6 is a potentially useful diagnostic marker of early-onset septicemia, but reports in the literature are conflicting, quoting sensitivities of 69–100% and specificities of 36–93% (9). Such wide ranges are the result of using different cutoff points for abnormal values (25–150 ng/L), when and how many samples are collected, how patients and controls are selected, and how reference values are defined and identified.
As a prerequisite for analyzing the CRP and IL-6 responses associated with infectious and noninfectious conditions during the immediate postnatal period, there is a need to establish the "normal" dynamics of both variables in the healthy neonate, counterbalanced by a greater awareness of the maternal and perinatal factors that may affect them. To this end, we report prospective, simultaneous measurements of both CRP and IL-6 in healthy babies at birth and at 24 and 48 h of life, and in their mothers at delivery. We also sought to identify maternal and perinatal factors that could confound the interpretation of these measurements.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
35 weeks of
gestation singleton newborns with birth weights appropriate for
gestational age and with normal results on physical examination at
birth that implied no need of empiric management. The mother-infant
pair was included in the study if all of the following criteria were
met: (a) a blood sample was collected at the time of
delivery from the mother; (b) a blood sample was obtained at
birth and at 24 and at 48 h after birth from each infant for
measurement of CRP and IL-6; (c) the neonate had a
continuous uncomplicated hospital stay and was discharged as healthy
from the well-baby nursery on day 3 (vaginal births) or 4 (cesarean
births); and (d) the neonate had normal assessments at the
1-week follow-up visit (for newborns who were born at 35–36 weeks
gestational age only) and the 2- and 4-week follow-up visits. We
excluded parturients with multiple preexistent or pregnancy-related
noninfectious complications, or with clinically evident intraamniotic
infection (10).
All antepartum and intrapartum data were collected prospectively and
included maternal age, preexistent or pregnancy-related diseases, mode
of delivery, use of anesthesia, duration of active labor
(11), interval between rupture of membranes and delivery,
maternal group B streptococcus (GBS) colonization, intrapartum
antimicrobial administration, and abnormalities in intrapartum fetal
heart monitoring. The institutional policy was to give intrapartum
penicillin or broad-spectrum antibiotics to all women identified as GBS
carriers at 35–37 weeks of gestation. If the results of GBS cultures
were not known at the onset of labor or rupture of membranes,
intrapartum antimicrobial prophylaxis was administered if either or
both of the following risk factors were present: preterm labor at <37
weeks of gestation, and rupture of membranes 18 h before delivery
(12). Neonatal data included gestational age, birth weight,
gender, and Apgar scores at 1 and 5 min. Gestational age was
established on the basis of best obstetric estimate, including last
menstrual period and first or second trimester ultrasonography.
blood collection and CRP and IL-6 measurements
Maternal serum was sampled at the time of delivery, and fetal
serum was obtained from the umbilical vein at birth. Maternal,
umbilical cord, and postnatal blood (100 µL, to allow a double
determination) samples for IL-6 measurements were stored in small
aliquots at -70 °C until analysis. IL-6 concentrations were
measured in duplicate by an enzyme-linked immunoassay (Endogen)
sensitive to <1 ng/L. For the IL-6 assay, both the inter- and
intraassay CVs are <10% according to the manufacturer. Measurements
of serum CRP concentrations were available in all study subjects within
a few hours after blood collection. CRP was measured by rate
nephelometry using a Beckman Array System protein analyzer (C-reactive
protein reagent set 449760; Beckman Instruments). According to the
manufacturer, this CRP assay has a detection limit of 4 mg/L and has
inter- and intraassay CVs <4% at both low and high concentrations.
statistical analysis
The observed 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 and IL-6 values and
all regression analyses were done after the observed values were
log-transformed. Consequently, all quoted mean values are geometric
means with 95% confidence intervals (CIs), and regression coefficients
are exponentiated to obtain ratios of geometric means.
There are two common ways of estimating percentiles from observed frequency distributions. The first is to arrange the data in ascending order and to identify the value of the variable that cuts off the desired percentage. The second method requires that the distribution has some known or supposed mathematical form, e.g., that the logarithms are distributed normally. In this case, the mean and SD of the distribution can be estimated from the observed data, and the percentiles can be obtained using tables of the standard normal distribution. Both methods have advantages and disadvantages. The first method has the advantages that no assumption is necessary about the distributional form of the data and that is it is nonparametric, but it has a disadvantage in that estimates of the 95th and 97.5th percentiles may have large sampling errors if the total sample size is not large. The second method has the advantage of being more precise but only if the distributional form is correctly known. In the present study, an additional difficulty in applying the second method was that many of the observed values of CRP and IL-6 were less than the minimum concentrations detectable by the instruments used to analyze the blood samples; thus, with these values missing, it was impossible to see the shape of the whole distribution. If it can be assumed that the distributions are log-normal, the observed values can be used to reconstruct the whole distribution using normal scores (not z-scores). Clearly, the method will fail if the true distribution is not the one that is assumed. Both methods were applied to the observed data, and fortunately the results were very similar. Only the nonparametric direct estimates are reported.
Multiple linear regression analyses were performed to explore the association between CRP and IL-6 responses in the mother-infant pairs and the following variables: gestational age, pregnancy-induced hypertension, gestational diabetes, intrapartum fetal distress, maternal GBS colonization, type of delivery (spontaneous vaginal, elective cesarean section, emergency cesarean section, and induced delivery), duration of active labor, interval (hours) between rupture of the membranes and delivery, intrapartum antimicrobial administration, and use of epidural or general anesthesia. The relationships between the CRP and IL-6 concentrations in the mother and the concentrations in the baby at birth and at 24 and 48 h of life were investigated by Spearman rank correlations. All statistical tests were considered significant if P was <0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
37 weeks of
gestation) and 35 near-term (35–36 weeks of gestation) were available
for simultaneous analysis of CRP and IL-6.
The characteristics of mothers and their healthy infants are summarized
in Table 1
. The reference intervals that were established for both CRP and
IL-6 at the three fixed neonatal ages included 148 neonates who were
not necessarily free of history of maternal and intrapartum
complications but whose postnatal clinical course from birth to the
4-week follow-up visit was unremarkable, implying therefore, no need of
management (including antimicrobial treatment) throughout this study
period. The distribution of crude CRP values among the delivering
mothers and their healthy offspring at the three neonatal ages is shown
in Fig. 1A
. Nineteen (12.8%), 54 (36.4%), and 55 (37.1%) of the
healthy newborns had detectable concentrations of CRP at birth and 24
and 48 h of life, respectively, which implies that the median
values are undetectable. Table 2
lists the 90th, 95th, and 97.5th percentiles of CRP values for
each neonatal age.
|
|
|
When the data were analyzed after a logarithmic transformation, the
geometric mean CRP values were significantly lower in cord blood than
in blood samples taken at 24 h (P <0.0001) and at
48 h (P <0.0001), with no significant change from 24
to 48 h (P = 0.6; Table 2
). Fifty-one (34.4%) of
the parturient mothers had detectable CRP. There was no correlation
between CRP concentrations in parturient mothers and those found in
their offspring at birth (r = 0.058; P
= 0.48).
By regression analysis, CRP response at birth was negatively associated
with the Apgar score obtained 5 min (P = 0.01) after
birth and positively associated with rupture of membranes for 18 h
or longer (P = 0.001). The babies’ mean CRP
concentration at birth was increased by a factor of 1.50 (95% CI,
1.32–2.03) if the 5-min Apgar score was 8 and by a factor of 1.32
(95% CI, 1.07–1.61) if the time from rupture of membranes was 18 h.
In contrast, the mean CRP response at 24 h was positively
associated with pregnancy-induced hypertension (P =
0.0038). The babies’ mean CRP concentration at 24 h was increased
by a factor of 1.43 (95% CI, 1.01–2.03) if the mother had gestational
hypertension. At 48 h of life, the babies’ mean CRP response was
not significantly affected by any of the variables identified from
maternal and perinatal history. The maternal mean CRP concentration was
increased by a factor of 1.06 (95% CI, 1.01–1.12; P =
0.006) per hour of active labor, by a factor of 1.62 (95% CI,
1.07–2.49, P = 0.022) if the mother had GBS
colonization, and by a factor of 1.54 (95% CI, 1.06–2.24;
P = 0.021) if the time from rupture of membranes was
18 h.
Fig. 1, B and C
, shows the distribution of crude IL-6 values in the
term and near-term delivering mothers and in their healthy offspring at
the three neonatal ages. Table 3
lists the 90th and 95th percentiles of IL-6 values in the term
and near-term infants for each neonatal age. The reason for dividing
the infants into subgroups in Fig. 1, B and C
, and Table 3
was the
finding by regression analysis that gestational age was negatively
associated with IL-6 values obtained from the mothers as well as the
babies at all of the three neonatal ages. Multiple regression analysis
of the independent effects of gestational age and birth weight on IL-6
values revealed that only gestational age had a significant effect. The
regression coefficients (SE) for gestational age were -0.192 (0.078;
P = 0.014) for delivering mothers, -0.281 (0.084;
P = 0.001) for babies at birth, -0.135 (0.057;
P = 0.020) for babies 24 h of age, and -0.147
(0.068; P = 0.032) for babies 48 h of age. As
expected, the geometric mean concentration of IL-6 in the near-term
mothers was 2.57 (95% CI, 1.49–4.46; P = 0.001) times
higher than that of the term mothers. The geometric mean IL-6
concentration in the near-term babies at birth and at 24 and 48 h
of life was 6.40 (95% CI, 3.61–11.50; P <0.0001), 2.38
(95% CI, 1.41–3.99; P = 0.001), and 2.40 (95% CI,
1.49–3.88; P <0.001) times higher, respectively, than the
concentration observed in the term babies.
|
We therefore examined, within each of the two gestational age groups,
whether neonatal IL-6 concentrations changed between the three times
and whether IL-6 concentration at birth correlated with maternal IL-6
at delivery. Among the term babies, the geometric mean IL-6 values were
significantly lower at birth than at 24 h of life (P
<0.0001), with no significant change from 24 to 48 h of life
(P = 0.12; Table 3
). In contrast, within the subgroup
of near-term babies, the geometric mean IL-6 concentrations at birth
were not significantly different from those found at 24
(P = 0.52) and 48 (P = 0.30) h of life
(Table 3
). IL-6 concentrations obtained from term babies at birth
correlated with maternal IL-6 concentrations (rank correlation
coefficient, 0.52; P <0.0001), whereas those obtained from
near-term babies at birth did not correlate with those found in their
mothers at delivery (rank correlation coefficient, -0.004;
P = 0.98).
Using multiple regression analysis, we investigated whether the
remaining variables had a significant effect on IL-6 response in
delivering mothers as well as in their healthy offspring at any of the
three set times. After adjusting for the effect of gestational age, we
found that the mothers’ IL-6 concentrations were increased by a factor
of 1.17 (95% CI, 1.03–1.34; P = 0.022) per hour of
active labor, whereas the babies’ IL-6 concentrations were increased
at birth by a factor of 5.04 (95% CI, 2.44–21.4; P =
0.027) if the 5-min Apgar score was 8. In contrast, none of the
remaining potential confounding factors had a statistically significant
effect on the babies’ IL-6 concentrations at 24 and 48 h of life.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To our knowledge, this is the first longitudinal study of CRP dynamics in a large sample of healthy newborns during the immediate postnatal period. In the majority of published reports (2)(3)(4)(6)(7)(13)(14)(15)(16)(17), upper limits for CRP during the neonatal period have been established in symptomatic uninfected patients. In that vein, the literature has conflicting reports as to the cutoff points (1.5–20 mg/L) for the upper limits of CRP. Possible sources of heterogeneity were wide-ranging differences in postnatal age or inaccuracies in reporting it, single vs serial determinations, different sample sizes, and different measurement methods (18). However, there are fewer cross-sectional studies of upper limits for CRP in healthy newborns (5)(19)(20). Again, among these studies, there is some degree of heterogeneity, based on different measurement methods and different sampling times. Thus, the fact that there are no established CRP reference intervals in the neonatal period can explain the wide range of reported CRP sensitivities (47–100%) and specificities (6–97%) for detection of neonatal sepsis (8). The present study demonstrates the physiologic CRP changes that occur over the first 48 h of life. In view of this dynamic behavior, previously reported CRP cutoff points may not be appropriate at certain neonatal ages.
The novel finding observed is that in our cohort of healthy neonates,
rupture of membranes 18 h, perinatal distress, or gestational
hypertension significantly affected the CRP concentrations during the
immediate postnatal period, but at specific ages. The present findings
become important with the very latest clinical applications of CRP.
Philip and Mills (21) established a CRP value 10 mg/L in
the presence of one (or more) clinical sign(s) or one (or more) risk
factor(s) for infection as the clinical pathway for transferring a
neonate from the well-baby nursery to the NICU and starting
antimicrobial therapy. Franz et al. (22)(23), on
the other hand, considered a CRP value >10 mg/L in the presence of one
(or more) clinical sign(s) as a criterion to make a diagnosis of early-
as well as of late-onset clinical septicemia in NICU babies. However,
it is clear that the patterns of CRP response in the immediate
postnatal period should be taken into account to optimize the use of
CRP in both the clinical setting of NICU and the well-baby nursery.
It is known that CRP does not cross the placenta (8). Accordingly, the present study confirms that serum concentrations of CRP and its production in the mother and fetus/newborn are independent of one another; however, the same stimulus may be operating concurrently in each. The major obstetric condition in which determination of maternal serum CRP concentrations might be clinically useful is chorioamnionitis (24). Considerable variation in the sensitivity and specificity of this testing is reported by different authors (25). This may be a result of the use of different cutoff points for abnormal values (12–40 mg/L), different measurement methods, or different gold standards. It is evident from our data that some confounding factors per se should also be taken into account to optimize the use of maternal serum CRP.
To our knowledge, this is the first longitudinal study of IL-6 dynamics in a large sample of healthy newborns during the 48-h period after birth. Recently, De Jongh et al. (26) found that the physiologic concentration of IL-6 in the cord blood may be affected by duration of gestation. In the present study, at any of the three fixed ages, near-term healthy infants had significantly higher IL-6 concentrations than term healthy infants, demonstrating that a gestational age-dependent effect on the normal IL-6 values might be seen over the initial 48 h of life. These higher IL-6 concentrations in near-term healthy neonates may be the result of subclinical perinatal infections, which are frequent with preterm labor, although none of these babies became symptomatic or received antibiotic treatment. There is no reason to believe that near-term infants would have been more prone to mistaken assignment of healthy status than term infants born in the same clinical conditions. On the other hand, preterm delivery can be more stressful for the unprepared fetus, which can induce an increase in serum IL-6, probably via the stimulation of IL-6 production by adrenal gland cells (27). This may also explain the negative correlation of cord blood IL-6 values with the Apgar score.
We have also shown that the kinetics of IL-6 during the first 48 h of life in healthy infants are different in the near-term infant compared with kinetics in the term neonate, suggesting different physiologic processes. In the term neonate, the surge of IL-6 at 24 h of age probably reflects a physiological stress reaction induced at birth. Similar data regarding IL-6 dynamics in healthy term neonates have also been reported recently (28). In contrast, in the near-term neonate, umbilical cord IL-6 was already increased at the time of birth, suggesting that a physiological stress reaction had begun before birth.
Although placental cells produce IL-6 and other proinflammatory cytokines, the evidence that these cytokines cross the placenta remains unclear (29). In the present study, the positive association between IL-6 values from term mothers and their healthy offspring at birth suggests that IL-6 may cross the placental barrier. However, the finding of higher IL-6 concentrations in cord sera of near-term babies compared with the corresponding maternal samples argues against the dependency of neonatal serum IL-6 concentrations on maternal IL-6 concentrations in the setting of preterm delivery. Taken together, the present data demonstrate the effects of development on serum IL-6 reference intervals and dynamics, as well as on the relationship between maternal and cord blood concentrations during the 48-h period after birth.
Another question addressed in this study was whether the IL-6 concentration in maternal serum was affected by pregnancy-related disorders, mode of delivery, use and type of anesthesia, duration of active labor, and stressful labor. The results confirm that a longer labor is correlated with increased serum IL-6 (and CRP) production in near-term as well as in term delivering mothers (30)(31), suggesting that IL-6 and the physical activity of labor are interrelated. Finally, at delivery, near-term mothers had IL-6 values that differed significantly from those for term mothers, demonstrating again the potential confounding effect of gestational age. However, the design of the present study does not allow any conclusion regarding whether IL-6 has an etiologic role as precipitator of human preterm delivery.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1-antitrypsin and haptoglobin in neonatal septicemia. Acta Paediatr Scand 1983;72:679-683.[Web of Science][Medline]
[Order article via Infotrieve]
, and -6 and prostaglandins in vaginal/cervical fluids of pregnant women before and during labor. J Clin Endocrinol Metab 1993;77:805-815.[Abstract]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  E Rodriguez, J Guevara, A Paez, E Zapata, M. Collados, T. Fortoul, R Lopez-Marure, F Masso, and L. Montano The altered expression of inflammation-related molecules and secretion of IL-6 and IL-8 by HUVEC from newborns with maternal inactive systemic lupus erythematosus is modified by estrogens Lupus, December 1, 2008; 17(12): 1086 - 1095. [Abstract] [PDF]
|
|
|
|
 |
|
 E K Chung, R L Miller, M T Wilson, S J McGeady, and J F Culhane Antenatal risk factors, cytokines and the development of atopic disease in early childhood Arch. Dis. Child. Fetal Neonatal Ed., January 1, 2007; 92(1): F68 - F73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kaufman and K. D. Fairchild Clinical Microbiology of Bacterial and Fungal Sepsis in Very-Low-Birth-Weight Infants Clin. Microbiol. Rev., July 1, 2004; 17(3): 638 - 680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 A. Madan, M. M. Adams, and A. G. S. Philip Frequency and Timing of Symptoms in Infants Screened for Sepsis: Effectiveness of a Sepsis-Screening Pathway Clinical Pediatrics, January 1, 2003; 42(1): 11 - 18. [Abstract] [PDF]
|
|
|
|
|
|
C. Chiesa, G. Pellegrini, A. Panero, J. F. Osborn, F. Signore, M. Assumma, and L. Pacifico C-Reactive Protein, Interleukin-6, and Procalcitonin in the Immediate Postnatal Period: Influence of Illness Severity, Risk Status, Antenatal and Perinatal Complications, and Infection Clin. Chem., January 1, 2003; 49(1): 60 - 68. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ishibashi, Y. Takemura, H. Ishida, K. Watanabe, and T. Kawai C-Reactive Protein Kinetics in Newborns: Application of a High-Sensitivity Analytic Method in Its Determination Clin. Chem., July 1, 2002; 48(7): 1103 - 1106. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
indicates 10
subjects; • indicates 1 subject. Dashed lines,
detection limits of assays; solid lines, medians.