Vaginal Lactoferrin Administration before Genetic Amniocentesis Decreases Amniotic Interleukin-6 Levels

Adverse outcomes of pregnancy, such as abortion and premature delivery, are a consequence of cytokine imbalance. Therefore, in obstetrics special interest arouse concerning interleukin-6 (IL-6), an activator of acute phase responses and a lymphocyte stimulator, that over the years has been assigned both pro- and anti-inflammatory characteristics [1, 2]. Indeed, an increased amniotic fluid (AF) IL-6 concentration was reported in patients with pregnancy loss following second trimester amniocentesis [3], and the same correlation was subsequently confirmed for preterm delivery and fetal growth restriction [4, 5], thus suggesting that reducing the cytokine level might improve the outcome of pregnancy by rebalancing the inflammatory response. In this regard, we have shown that ampicillin is able to decrease Il-6 release by WHISH cells, a human amnion-derived cell line, as well as the AF cytokine concentration in women undergoing second trimester amniocentesis [6]. In the present study we focused on lactoferrin (LF), an 80-kDa iron-binding glycoprotein of the transferrin family provided with anti-inflammatory and antimicrobial properties, which is able to decrease IL-6 in several experimental conditions [7]. LF is physiologically present in AF [8, 9], and it is frequently administered to pregnant women for treating iron deficiency anemia [10].

Considering the safety of the drug administration during pregnancy, its eventual efficacy to decrease AF IL-6 could represent a useful tool in preventing abortion secondary to amniocentesis.

We performed an open-label clinical study enrolling 60 Caucasian pregnant patients undergoing genetic amniocentesis at the 16th gestational week at the Obstetric Unit of Ferrara University from March 2011 to March 2012. The inclusion criteria were: singleton physiological pregnancy and maternal age as the only indication to fetal karyotyping. The exclusion criteria were: consumption of drugs interfering with the immune system, previous miscarriages, pregnancy at risk for maternal or fetal disease, lactose intolerance. The research was in accordance with the ethical principles of the Declaration of Helsinki. The Local Ethics Committee approved the study design. All pregnant women gave their written informed consent. A vaginal compound containing 300 mg of LF (Difesan, Progine Farmaceutici, Firenze, Italy) was randomly administered to patients 4 or 12 h prior to amniocentesis, in order to obtain 3 groups:

• group A: 20 untreated patients;

• group B: 20 patients treated 4 h before amniocentesis;

• group C: 20 patients treated 12 h before amniocentesis.

A total amount of 20 ml AF was withdrawn for karyotype analysis, microbiological culture, α-fetoprotein (18 ml) and IL-6 assay (2 ml).

Our protocol entails using 22-gauge needles provided with a 24-gauge tip, in order to reduce the diameter of the hole in the amniotic membrane and the consequent risk of AF leakage.

A questionnaire was administered to the patients, in order to check for any complication (vaginal bleeding, uterine contraction, rupture of the membranes) within 7 days following the procedure.

Fresh AF specimens (2 ml) were centrifuged at 3,000 g, 4°C for 10 min. The proinflammatory cytokine IL-6 was measured and quantified in the AF supernatants (100 µl), aliquoted and stored at -70°C until analysis, by an enzyme-linked immunoabsorbent assay using a Quantikine Elisa Human IL-6 Kit (R & R Systems, Minneapolis, Minn., USA), according to the manufacturer's instructions. For all assays, full or diluted (1:10 to 1:100) AF samples (tested in duplicates) were done to fall within the range of the standard curves as previously described [11]. The interassay and intra-assay coefficients of variation were <10% for all the analytes. The lower detection limit for human IL-6 was 3.12 pg/ml. A zero value was assigned to samples below these limits.

Data were analyzed using SPSS, version 20.0 (SPSS Inc., Chicago, Ill., USA). Nonparametric analysis by the Kruskal-Wallis test was considered appropriate for the sample size. The Behrens-Fisher test was used to verify the difference among mean IL-6 values in the 3 groups. Differences with p < 0.05 were considered statistically significant.

The mean maternal age in the whole study population was 37.5 ± 2.1 years (minimum 35, maximum 42). This parameter was not different among the 3 groups (p = 0.5) and distributed as follows: 37.8 (±2.4), 37.7 (±2.3) and 37.1 (±2.1) years, respectively, for groups A, B and C.

A normal karyotype was registered in all cases. The AF microbiological cultures were negative in all the samples. No complication (vaginal bleeding, uterine contractions and ruptured membranes) was registered within 7 days after amniocentesis. The course of pregnancy was normal in all patients, ending in spontaneous delivery at term.

The comparison of the distribution of the IL-6 values among the 3 groups showed a highly significant difference (p = 0.001). Significance was further confirmed by the Behrens-Fisher test. In particular, the IL-6 group B mean value was shown to be significantly lower compared with those of the other two groups, with a significance of p = 0.006 and p = 0.03 compared with groups C and A, respectively. In contrast, there was no significant difference between group A and group C IL-6 mean values (table 1).

These differences are graphically represented in figure 1, where value distribution shows a particularly narrow box-and-whiskers plot width in group B compared with the other two.

It is well known that some of the cytokines involved in the inflammatory response also regulate implantation and labor. Indeed, an inflammatory microenvironment appears to be required for adequate tissue remodeling during the early phase of physiological pregnancy [12]. In contrast, the second trimester, along with a large part of the third, is characterized by the prevalence of anti-inflammatory signals. Again the scale plate hangs down towards an inflammatory pattern near term, for inducing uterine contractions and cervical dilatation [13]. Therefore, it can be argued that cytokines and prostanoids are not inflammatory by themselves, as in normal conditions such compounds do regulate important physiological functions. It is rather their imbalance that triggers inflammation [14]. Ominous complications of the second and third trimesters are related to an increased level of cytokines able to induce prostanoid release, thus triggering uterine contraction and impairing uteroplacental blood flow. Either abortion or premature delivery has been related to increased inflammatory and decreased anti-inflammatory cytokines. Such changes are frequently ascribed to bacterial mediation [15, 16],but an increased IL-6 concentration in the fetal compartment (AF and fetal blood) plays itself a relevant role even in the absence of any evidence of infection [17]; accordingly, it has been shown to predict abortion secondary to amniocentesis [3]. The mechanism by which the cytokine is able to directly trigger uterine contractions appears to be the activation of the arachidonic acid cascade, with consequent production of prostaglandin E₂[18, 19, 20]. As for possible causes other than infection for an increased release of cytokines able to trigger gestational inflammation, it has been shown that genetic proinflammatory cytokine polymorphisms are associated with a higher risk of preterm birth [21, 22, 23, 24].Whatever the cause, IL-6 elevation cannot be considered to be a benign state [25], suggesting a search for suitable therapies aimed at reducing its concentration, in order to prevent dangerous prostaglandin release. Accordingly, anticytokine therapy has recently been suggested for preterm labor as well [26]. As mentioned above, ampicillin is able to directly inhibit amniotic IL-6 release both in vitro and in women undergoing genetic amniocentesis [6]. Furthermore, the drug significantly decreases basal as well as stimulated prostaglandin E₂ release from amnion tissue [27] while other classes of antibiotics not only lack this ability, but also impair that of ampicillin when administered together [28]. Based on the evidence above, and considering that the 1.4% abortion rate following amniocentesis [29] might at least in part derive from a preexistent inflammatory gestational condition (whether or not related to infection), we felt it was justified to administer ampicillin to all women undergoing invasive prenatal diagnostic procedures. We started this strategy many years ago, and our total abortion rate for genetic amniocentesis is 0.3%, i.e. below the natural incidence observed between 16 and 24 weeks. Other authors suggest a different approach for preventing abortion secondary to amniocentesis [30]. Indeed, considering that the intrauterine environment might not be sterile, they tested the capacity of a macrolide antibiotic in reducing the abortion rate following the procedure. Their study is quite remarkable due to the extremely high number of cases, although it raises concern as regards some of the data. For instance, in spite of quoting a 1.08 natural abortion rate [31] following amniocentesis, the authors reported only a 0.28% incidence in the control group, compared with 0.03% in treated patients. In other words, it appears that, in their hands, the invasive procedure in itself was provided with a protective efficacy against pregnancy loss, a quality hard to recognize even for highly skilled operators! Apart from such a questionable datum, further doubts arise from considering that, although macrolides provide protection against sensitive germs, they are not only unable to directly decrease amniotic prostaglandin release, but also their addition completely abolishes this capability of ampicillin [28].Therefore, given the dual action of ampicillin against IL-6 and prostaglandin E₂, along with the broad spectrum bactericide efficacy, one wonders why the large trial mentioned above tested a macrolide instead!

However, waiting for the question to be answered, it seems also appropriate to search for less harmful drugs compared to antibiotics, provided with the same protective efficacy, along with a decreased risk of side effects. This was the purpose of the present study, testing the efficacy of vaginal administration of LF in decreasing the amniotic concentration of IL-6 in women undergoing genetic amniocentesis.

LF is a major component of the mammalian innate immune system. Its protective effects range from direct antimicrobial activities against a large panel of microorganisms, including bacteria, viruses, fungi and parasites, to anti-inflammatory and anticancer action. These activities rely upon LF capable of binding iron as well as the protein interactions with molecular and cellular components of both host and pathogens [32, 33]. Interestingly, endogenous AF LF normally increases after 31 weeks of gestation. However, levels above 2.5 μg/ml before this gestational age are considered highly suggestive of intra-amniotic infections [34, 35], thus indicating that endogenous LF release represents a natural protective response of the organism.

Moreover, the exogenous bovine form, administered for treating iron deficiency anemia, has been reported to decrease IL-6 concentration in the maternal serum [7].

With the aim to counteract the inflammatory cascade triggered by the tissue lesion that an invasive diagnostic procedure produces even in expert hands, in the present study we tested the efficacy of exogenous LF, administered throughout the vaginal route to decrease the amniotic level of IL-6 in patients undergoing genetic amniocentesis. Our data show that 300 mg of the compound significantly reduce the cytokine concentration as early as 4 h after administration. However, no significant difference is found between mean IL-6 values after 12 h and those of the controls. Such a behavior is probably due to the short half-life of LF, and it represents a further expression of the effectiveness of the drug. Indeed, besides a short resistance to peptic degradation (about 17 min), oral LF shows a very short half-life (about 8 min) and a poor bioavailability [36]. It is noteworthy that along with the decrease in the values, their distribution becomes particularly narrow 4 h after LF administration, to widen and increase again with time, due to the short half-life of the drug. Therefore, the opportunity of several daily doses may be suggested in order to maintain a low level of the cytokine.

We consider such an action an interesting finding for a compound that is also a natural component of the human body, although we are aware that our study population was restricted to cases with a normal pregnancy. Whether or not the same result could be obtained in patients with inflammatory syndrome remains to be ascertained. On the other hand, we know that even the most experienced operator is unable to avoid the risk of abortion produced by tissue damage derived from invasive prenatal diagnosis. Evidence reported in the literature indicates that such a risk is likely of an inflammatory nature, and therefore it must be prevented by the administration of anti-inflammatory drugs. In this respect, besides ampicillin and cephalosporin that represent the best antibiotic choice, transvaginal LF could be a good alternative. In this regard, it will be necessary to ascertain whether, in addition to inhibiting amniotic IL-6 release, the compound also reduces prostaglandin E₂ synthesis, thereby enhancing its effectiveness against abortion. If so, it would seem more appropriate to entrust the task of a general anti-inflammatory action to the natural compound, reserving the antibiotic for cases of infection.

In conclusion, in patients undergoing genetic amniocentesis, the capacity of transvaginal LF to decrease amniotic IL-6 levels may represent a useful tool against the increased risk of abortion that the procedure entails.