Inflammatory cytokines in intrauterine growth retardation
Abstract
Background. To evaluate maternal serum levels of two inflammatory cytokines in women with intrauterine growth retardation (IUGR), while studying separately women with or without placental insufficiency.
Methods. The study comprised 14 women with IUGR and Doppler-defined placental insufficiency, 14 women with IUGR without placental insufficiency, and 28 healthy pregnant women as a control group. Tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) were measured using a commercially available kit. The Kruskal–Wallis test and the corrected Mann–Whitney U-test were used.
Results. There was a statistically significant difference in TNF-α levels among the three studied groups (p = 0.03). Women with IUGR and placental insufficiency showed statistically significant higher serum levels of TNF-α[2.2 pg/mL (1.3–4.1)] and a higher rate of detectable TNF-α[85.7% (12/14)] than those in the control group [0 pg/mL (0–2.7) and 32.1% (9/28)] (p = 0.01 and p = 0.001, respectively). On the contrary, there was no difference in either the TNF-α level [1.4 pg/mL (0–4.9)] or the rate of detectable TNF-α[57.1% (8/14)] between women with IUGR without placental insufficiency and women in the control group. The levels of IL-6 were similar in the three studied groups.
Conclusion. TNF-α is increased in women with IUGR and placental insufficiency but normal in those with IUGR and normal placental perfusion. We suggest that elevations of TNF-α could be a specific phenomenon of certain subsets of IUGR, identifying cases with placental dysfunction.
Intrauterine growth retardation (IUGR) is one of the main causes of perinatal morbidity and mortality. However, the mechanisms mediating impaired fetal growth remain unclear. IUGR is a heterogeneous condition that includes abnormal situations, most of them related to placental insufficiency and physiological but misunderstood events leading to healthy small-for-gestational-age babies. Therefore, there is a need to distinguish between different subgroups of IUGR related to different risks of perinatal morbidity.
An increased decidual cellular immunity limiting trophoblastic invasion and leading to failure of placentation has been reported to be involved in the pathogenesis of several pathologic conditions such as spontaneous abortion, preeclampsia and IUGR (1,2). It may be hypothesized that placental insufficiency in cases of IUGR could be caused by an immunological phenomenon. There are many previous reports studying the role of immunological mechanisms in the genesis of spontaneous abortion and preeclampsia, but only a few about IUGR. In women with preeclampsia, plasma and amniotic fluid elevations of some inflammatory cytokines (3) and increased production of these cytokines by peripheral blood mononuclear cells (4) have been related to a role of abnormal immune activation in the pathophysiology of this disease. There are a few papers (5–8) that have studied the levels of inflammatory cytokines in women with IUGR and reached contradictory conclusions, but none of them has studied separately cases with or without placental insufficiency. This is the aim of the present study.
Materials and methods
A prospective study was carried out at the University Hospital of Puerto Real, Cadiz. Fifty-six women in three groups were studied: 1) 14 women diagnosed with IUGR and Doppler-defined placental insufficiency (pulsatility index in umbilical artery greater than 90th percentile for gestation); 2) 14 women with IUGR and normal umbilical blood flow measurements; and 3) 28 age-, parity- and gestational-age-matched healthy pregnant women as a control group. No cases of multiple pregnancies were included in the study. IUGR was defined, first, by a fetal abdominal circumference below the 10th percentile for gestation and, second, at delivery all infants had a weight below the 10th percentile for gestation. All infants were chromosomally and anatomically normal and none of them showed signs of fetal infection in the neonatal period. Women admitted to the hospital because of IUGR were eligible. They were consecutive women for each previously defined group of IUGR. Exclusion criteria were the diagnosis of fetal malformations and the presence of some comorbid disease in the mothers, including infectious diseases. Women in the control group were chosen from our antenatal clinic. All women were normotensive and all of them agreed to sign an informed consent to participate in the study.
Venous blood was collected by venepuncture into 10-mL silicone-coated Vacutainer blood-collecting tubes containing no additives between 0700 and 0900 h in fasting state. Blood was allowed to clot at room temperature and was then centrifuged for 20 min at 2000 g. Aliquots of the serum were then stored at − 80 °C until they were required for the assay.
Tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) were determined using a commercially available immunoassay (R&D Systems, Inc, Minneapolis, USA). Duplicates were assayed in every experiment. A monoclonal antibody specific for each cytokine was coated onto a microtiter plate provided in the kit. Standards with known concentrations of each cytokine and samples were pipetted into wells, and cytokines were bound by an immobilized antibody during a first incubation at room temperature. After washing away any unbound proteins, an enzyme-linked polyclonal antibody specific for each cytokine was added to the wells and incubated again at room temperature. After washing to remove unbound antibody reagent, a substrate solution was added to the wells and a color developed in proportion to the amount of cytokine bound in the initial step. The color development was stopped after 20 min and the intensity of the color measured by a microtiter plate reader. We initially performed an assay in which the minimum detectable dose was 0.7 pg/mL for IL-6 and 4.4 pg/mL for TNF-α. However, we found that in a substantial proportion of cases neither cytokine was detected. We therefore performed a second assay using high sensitivity kits in which the minimum detectable dose was 0.094 pg/mL for IL-6 and 0.18 pg/mL for TNF-α. Intra- and interassay coefficients of variability were 4.2% and 5.4% for TNF-α and 1.7% and 2% for IL-6, respectively. Immediate serum separation after obtaining the samples and only one freeze-thawing cycle ensured high stability of both cytokines (9).
Using previously reported data (5) with the rate of detectable TNF-α as the main outcome, we calculated that a sample of 52 women was needed to find a significant difference with an α level of 0.05 and a power (1 − β) of 80%. We enrolled 56 women to account for withdrawals. Statistical analysis was performed using the SPSS computer statistics program. Distributions were checked by histogram and the Kolmogorov–Smirnov test. When a variable was normally distributed, data were presented as mean and standard deviation and comparisons were made using the one-way analysis of variance (anova) test. In cases of nonparametric variables data were shown as median and interquartile range and comparisons were made using the Kruskal–Wallis test. As both of the studied cytokines were distributed in a nonparametric manner, after the Kruskal–Wallis test we used the corrected Mann–Whitney U-test to compare groups with each other. Quantitative variables were expressed as numbers and percentages. To compare proportions, the χ2-test and Fisher's exact test (when expected numbers were less than 5) were used. Statistical significance was set at the 95% level (p < 0.05).
Results
Demographics and general obstetric outcomes are shown in Table I. As expected by definition, birthweight was significantly lower in the IUGR groups than in controls.
| Control | IUGR with PI | IUGR without PI | ||
|---|---|---|---|---|
| (n = 28) | (n = 14) | (n = 14) | p | |
| Age (years) | 29.0 ± 4.7 | 29.5 ± 6.9 | 30.1 ± 5.1 | NS |
| Gestational age at study (weeks) | 37.3 ± 1.5 | 36.4 ± 2.7 | 37.8 ± 1.8 | NS |
| Nulliparous (n, %) | 12 (42.8) | 6 (42.8) | 6 (42.8) | NS |
| Gestational age at delivery (weeks) | 40.3 ± 1.8 | 36.8 ± 2.9 | 38.2 ± 1.9 | <0.001 |
| Newborn weight (g) | 3441.0 ± 480.3 | 2034.5 ± 559.3 | 2317.1 ± 333.1 | <0.001 |
| Cesarean section (n, %) | 10 (35.7) | 6 (42.8) | 3 (21.4) | NS |
- IUGR, intrauterine growth retardation; PI, placental insufficiency
In addition, gestational age at delivery was lower in cases with IUGR plus placental insufficiency than in women in the control group. As shown in Table II, there was a statistically significant difference in TNF-α levels among the three studied groups (p = 0.03). We found that women with IUGR and placental insufficiency had statistically significantly higher levels of TNF-α than those in the control group (p = 0.01). On the other hand, there was no significant difference in TNF-α level between women with IUGR without placental insufficiency and women in the control group. Similarly, there was a statistically significant difference in the rate of detectable TNF-α among groups (p = 0.004). This rate was significantly higher in women with IUGR and placental insufficiency than in controls 85.7% (12/14) vs. 32.1% (9/28) (p = 0.001). On the other hand, the difference between women without IUGR and placental insufficiency and controls [57.1% (8/14) vs. 32.1% (9/28)] did not reach statistical significance (Fig. 1).
| TNF-α (pg/mL) | IL-6 (pg/mL) | |||
|---|---|---|---|---|
| Median (interquartile range) | Number /percentage detected | Median (interquartile range) | Number /percentage detected | |
| Control group (n = 28) | 0 (0–2.7) | 9/32.1 | 3.1 (2.6–6.3) | 28/100 |
| IUGR with PI (n = 14) | 2.2 (1.3–4.1)* | 12/85.7† | 3.0 (1.0–6.2) | 14/100 |
| IUGR without PI (n = 14) | 1.4 (0–4.9) | 8/57.1 | 2.3 (1.5–3.1) | 14/100 |
- IUGR, intrauterine growth retardation; PI, placental insufficiency
- * p = 0.01;
- † p = 0.001
Variation in the rate of detectable TNF-α between women with IUGR with or without placental insufficiency (PI) and women in the control group.
Serum levels of IL-6 were similar in the three studied groups.
TNF-α was detected in all those cases with severe placental insufficiency showing either absent end-diastolic flow or reverse flow (six cases), and the levels of this cytokine were higher 3.6 pg/mL (2.2–31.2) than those in the control group (p = 0.03). No difference in the IL-6 levels [1.44 pg/mL (1–5.68)] was found.
Discussion
Previous studies evaluating TNF-α levels in cases of IUGR have reported contradictory results. Two studies (6,7) investigated amniotic fluid TNF-α levels at the second trimester of gestation before IUGR was clinically present. In one of them (6), elevated levels were associated with small-for-gestational-age at birth. In the other (7), the levels were similar between patients subsequently delivering small-for-gestational-age neonates and controls. Another study (8) reported increased amniotic fluid TNF-α levels when IUGR was present.
Two studies (5,10) evaluated maternal and fetal plasma TNF-α levels in pregnancies associated with small-for-gestational-age newborns. Both of them studied only newborns with “idiopathic” growth retardation and decreased (5) or normal (10) plasma TNF-α levels were found.
Finally, two studies found increased placental expression of TNF-α and other inflammatory cytokines in cases of IUGR (11,12).
Similar to the TNF-α studies, IL-6 levels have been found to be increased in women with IUGR (8), even though other studies have not confirmed these findings (10,11).
IUGR is a heterogeneous condition that includes a wide variety of situations ranging from physiological small-for-gestational-age babies to abnormal conditions, including cases of fetal malformations, infections or placental insufficiency. This could explain the differences in the results of previous studies.
We found increased maternal serum levels of TNF-α in cases of placental insufficiency. In women with preeclampsia, increased expression of proinflammatory cytokines was associated with placental hypoxia (14,15). Along the same line, Holcberg et al. (12) found that increased TNF-α secretion in placentas of intrauterine growth-restricted fetuses was related to enhanced vasoconstriction of the fetal placental vascular bed. In addition, when they perfused normal placentas with angiotensin II, a potent vasoconstrictor, TNF-α production was increased. This supports our finding of elevated TNF-α levels in cases of IUGR with increased umbilical artery resistance.
In conclusion, TNF-α is increased in women with IUGR and placental insufficiency but it is normal in those with IUGR and normal placental perfusion. Our study suggests that elevations of TNF-α and possibly other inflammatory cytokines could be a specific phenomenon of certain subsets of IUGR, identifying cases with placental dysfunction. However, further studies including a larger number of cases could help to definitively clarify the relationships between cytokines and placental insufficiency and to evaluate the predictive value of TNF-α or other inflammatory cytokines to differentiate between physiological and pathological cases of IUGR.
