Main

The effect of labor on maternal and neonatal immunity has been studied since 1969 (1). Prior reports have shown decreased percentages of T lymphocytes expressing CD2, CD3, and CD4 and increased numbers of natural killer cells bearing the markers CD16 and CD56 in vaginally delivered infants when compared with infants born by cesarean section without labor (2–5).

Labor results in a neutrophilia, commonly observed in the early neonatal period (6,7). The effects of labor on neonatal neutrophil function have been debated. Results of various studies have shown either no effect or increased neutrophil function after labor (7–11). One recent study demonstrated that clinical signs of fetal distress, such as fetal acidemia and low Apgar scores, are associated with decreased neonatal neutrophil function (8). This study specifically examined the duration of labor on neutrophil function, and a negative correlation was found (8).

The present study was designed to test the hypothesis that both gestational age and the duration of labor are independent determinants of neonatal neutrophil function in preterm infants. Neonatal neutrophil function was measured indirectly by quantifying the percentage of neutrophils expressing leukocyte cell surface antigens associated with neutrophil activation during the immune response. We hypothesized that longer lengths of labor and advancing gestational age both contribute to increased expression of antigens on the neutrophil cell surface. We measured the percentage of neutrophils that expressed CD11b and CD11c (subunits of adhesion molecules), CD15 (a selectin molecule), CD33 (present on precursor neutrophils), and CD66b (contributes to respiratory burst activity) in a group of preterm uninfected neonates studied on the first day of life. We then correlated the expression of these antigens with a number of demographic and clinical factors including gestational age, mode of delivery, and length of active labor.

METHODS

The study was approved by the New England Medical Center Human Investigation Review Committee. Written informed consent was not required for obtaining otherwise discarded peripheral blood. Our analysis was based on 400 µL of blood obtained from discarded blood remaining in the venipuncture tubing and syringe or from the hematology sample after completion of the CBC. Samples were obtained from venous or arterial puncture or from an indwelling catheter based on individual patient access at the time of the laboratory draw.

Study population. Inborn premature infants admitted to the NICU at New England Medical Center during a 16-mo period that began in May 1995 who had blood drawn at the time of admission were eligible for enrollment in the study. Prematurity was defined as a gestational age of less than 37 wk. The onset of labor was defined as follows: documented uterine contractions every 2 to 3 min with ruptured membranes; or, in the absence of ruptured membranes, documented uterine contractions every 2 to 3 min with cervical dilation greater than or equal to 2 cm or 80% effacement (12). Inclusion criteria were the following: low risk for sepsis on admission (membranes ruptured less than 18 hours, no maternal fever), normal results on physical examination, and normal CBC. Infants with a positive blood culture or longer than 3 d of antibiotic treatment were subsequently disqualified. Infants were enrolled within the first 12 h of life, at the time blood was obtained for admission laboratory studies, which included a CBC and culture.

Data recorded for the study population included gestational age, birth weight, gender, mode of delivery, duration of active labor, indication for operative delivery if applicable, length of ruptured membranes, evidence of maternal fever, maternal laboratory data including group B streptococcus culture and total white blood cell count, maternal antibiotics before delivery, antenatal steroid exposure, Apgar scores at 1 and 5 min, infant CBC result, blood culture results, length of antibiotic therapy, length of hospitalization, and outcome parameters such as bronchopulmonary dysplasia, retinopathy of prematurity, patent ductus arteriosus, necrotizing enterocolitis, and intraventricular hemorrhage.

Determination of antigen expression. We measured leukocyte cell surface antigens known as cluster of differentiation antigens (CD antigens) present on the neutrophil plasma membrane. The antigens that were quantified are present on granulocytes and include CD11b, CD11c, CD15, CD33, and CD66b. Table 1 describes these antigens and their functions.

Table 1 CD antigens studied as markers for neutrophil activation
Full size table

Blood specimens for neutrophil antigen determination were collected from discarded peripheral blood samples obtained from infants undergoing NICU admission blood work. Samples were placed in Microtainer tubes containing EDTA. They were kept at room temperature and stained within 12 h of sampling.

Fifty microliters of whole blood were incubated with 10 µL of antigen-specific fluorescently conjugated MAb in each tube as follows: tube 1, FITC-conjugated antimouse IgG1 (Becton-Dickinson, San Jose, CA; catalog No. 349041) and PE-conjugated antimouse IgG1 (Becton-Dickinson; catalog No. 349043) as controls; tube 2, LeucoGATE, FITC-conjugated anti-CD45 and PE-conjugated anti-CD14 (Becton-Dickinson; catalog No. 340040); tube 3, FITC-conjugated anti-CD15 (Becton-Dickinson; catalog No. 347423) and PE-conjugated anti-CD13 (Becton-Dickinson; catalog No. 347837); tube 4, FITC-conjugated anti-CD66b (Immunotech, Marseilles, France; catalog No. 0531) and PE-conjugated anti-CD11b (Becton-Dickinson; catalog No. 347557); tube 5 FITC-conjugated anti-CD33 (Immunotech; catalog No. 1135) and PE-conjugated anti-CD11c (Becton-Dickinson; catalog No. 347637).

Samples were incubated at room temperature in the dark for 15 min, diluted with a 1% BSA in PBS solution and centrifuged. The supernatant was aspirated and the cell pellet was resuspended in 2 mL of FACS Lysing Solution (Becton-Dickinson; catalog No. 349202), mixed in a vortex agitator, and allowed to incubate for 10 min in the dark. The sample was centrifuged and the supernatant decanted. The remaining cell pellet was resuspended, washed with 1% BSA/PBS, and again centrifuged and decanted. The remaining cell pellet was fixed in 500 µL of a solution of 2% paraformaldehyde in 1% BSA/PBS.

Flow cytometric analysis. Samples were analyzed on a Becton-Dickinson FACStar Plus flow cytometer equipped with an Ion Laser Technology 5500A air-cooled laser (ILT, Salt Lake City, UT). Fluorescence emission was detected with a 530/30 nm band pass filter for FITC and a 575/26 nm band pass filter for PE. Electronic compensation was used between fluorescence channels to remove spectral overlap. The instrument was aligned using DNA-Check alignment fluorospheres (Coulter, EPICS Division, Hialeah, FL; lot No. 1642). Optimization of fluorescence sensitivity and initial compensation were also performed using Cali BRITE beads (Becton-Dickinson; catalog No. 349502). Finally, compensation was verified using biologic controls stained with CD3 FITC (Becton-Dickinson; catalog No. 349201) and CD19 PE (Becton-Dickinson; catalog No. 92-009). Samples were gated on forward scatter versus side scatter to exclude cell debris and cell aggregation. FITC and PE emissions were collected with a log amplifier, and immunofluorescence data were displayed on a four-decade logarithmic scale. Cytometric data were analyzed using Lysis II acquisition and analysis software. Neutrophil gates were set based on staining with CD45/CD14 and then verified using characteristic appearance of forward scatter and side scatter. Quadrants were set to include 99% of the nonspecific background staining with mouse IgG1 in the negative gate. Results were expressed as a percentage of cells staining with the antibody.

Statistical analysis. Statistical analysis was performed using statistical software including SPSS (SPSS Inc., Chicago, IL). Univariate analysis of neutrophil antigen expression versus mode of delivery and gender was tested by means of the t test. A two (steroids) × two (antibiotics) 2-way ANOVA was used to study the effect, alone and in combination, of these drugs on the CD antigen values. A bivariate regression analysis weighted for the number of measurements at each gestational age was used to evaluate the independent effects of gestational age and labor length on CD antigen expression. Univariate regression analysis was used to evaluate the relationship between CD antigen expression, 5-minute Apgar score, gender, and birth weight. Power analysis with an α = 0.05 was performed. Power was expressed as 1 - β. Because of the number of cases studied, the interaction between gestational age and labor length was not evaluated.

RESULTS

Study population. Sixty-eight infants who met the inclusion criteria were enrolled in the study. The mean (SD) percentages of antigen expression for the overall series before analysis of dependent variables were the following: CD11b, 84.2% (16.2); CD11c, 75.0% (19.5); CD15, 82.6% (14.0); CD33, 16.6% (18.6); and CD66b, 42.8% (29.6). The clinical characteristics of the study population are listed in Table 2. The mean length of labor for the entire study population was 5.8 h (range 0-27 h). All neonatal blood cultures were negative. Antenatal exposure to steroids and antibiotics, infant gender, 5-minute Apgar score, and weight at delivery had no statistically significant effect on the expression of neutrophil antigens.

Table 2 Clinical characteristics of the study population (68 infants)
Full size table

Antigen expression versus mode of delivery. To assess the effect of mode of delivery on neutrophil antigen expression, a univariate analysis of each of the CD antigens to mode of delivery was performed. Table 3 reports the mean percentage of the CD antigen expression for each group (cesarean section and vaginal delivery). The univariate analysis showed no significant correlation between antigen expression and mode of delivery. The p values of the t test ranged between 0.191 and 0.638. Because of the small difference in the mean values of the CD antigens in the two groups, a low power (1 - β) was observed (from 0.11 to 0.36).

Table 3 Mean (SD) CD antigen expression as a function of mode of delivery
Full size table

Antigen expression versus labor length and gestational age. To assess the effect of labor length and gestational age on neutrophil antigen expression, a bivariate analysis of the expression of each of the CD antigens as a function of labor length and gestational age was performed.

Bivariate analysis of each CD antigen showed a positive correlation with labor length. Gestational age correlated only with CD15 antigen expression. Results are reported in Table 4. Power (1 - β) ranged from 0.61 to 0.95 for the labor length, and a value of 0.74 was reported for the correlation between gestational age and CD15 antigen.

Table 4 Bivariate regression analysis
Full size table

DISCUSSION

For almost 30 years it has been known that the process of birth affects both the neonatal and maternal white blood cell counts (1,13). The mechanism behind this leukocytosis is less clear. The benefit or harm of labor to the neonatal immune system has been debated. The results of the present study indicate that longer lengths of labor are associated with increased expression of neutrophil antigens whose up-regulation is associated with neutrophil activation during the immune response. This implies that, at least in preterm infants, longer lengths of labor are associated with increased neonatal neutrophil activation. The mode of delivery does not appear to be an important factor in neonatal neutrophil activation. Therefore, preterm infants born after a precipitous vaginal delivery do not have increased expression of neutrophil antigens because they have not had the physiologic changes associated with labor. Although length of labor is associated with neonatal neutrophil activation, it is probably not the direct cause of it. Humoral substances, such as prostaglandins and cytokines, which are released into mother's circulation during labor and then cross the placenta into the fetal circulation have a more likely role in directly activating the neonatal neutrophil. In previous studies, labor length has been shown to correlate with increased cytokine and prostaglandin production in the maternal circulation (14).

The results of this study also indicate that the expression of neutrophil antigens on the first day of life is in general, independent of gestational age in preterm infants at less than 37 wk of gestation. This was an unexpected finding because, in fetuses, a positive linear correlation has been observed between gestational age and neutrophil antigen expression in cordocentesis samples (15). The relationship between gestational age and neutrophil antigen expression is potentially masked at birth by a much larger effect of labor on antigen expression. CD15, a ligand for adhesion molecules, did show correlation with gestational age even when corrected for labor length, which supports the speculation that the effect of gestational age on all neutrophil antigen expression exists but is masked by the effects of labor. We suspect that after the first few days of life the effects of gestational age again become apparent in neonatal neutrophil antigen expression.

Premature infants can mount an immune response. The process of labor can activate neutrophils in infants as young as 24 wk of gestation. The known immaturity of the neonatal immune system cannot be explained by neutrophil cell surface antigen expression. We have shown that premature infants have the ability to express antigens associated with cell adhesion, migration, and respiratory burst activity in response to labor.

We speculate that labor plays a role in preparing the fetal immune system for the transition to the extrauterine environment. Premature infants who have been exposed to labor may be protected from infection in the neonatal period by a more activated immune system. However, premature infants with activated immune systems may also be at increased risk for complications of prematurity thought to be of inflammatory origin or mediated by oxygen free radicals such as bronchopulmonary dysplasia or retinopathy of prematurity. We are currently evaluating the outcome of this study population with respect to the complications of prematurity. In addition, now that we know the gestational age is not a major factor in the expression of neutrophil cell surface antigens, we are performing further studies evaluating the effect of labor length versus no labor on a group of low-risk full-term infants. Further knowledge acquired may be important in evaluating the potential complications of elective repeat cesarean section delivery without labor.