Materials and Methods

This is a secondary data analysis of previously published data that aimed to compare the fetal plasma concentration of adenosine from normal singleton pregnancies (n = 27) and that from patients with preeclampsia (n = 39). Preeclampsia was defined in the presence of gestational hypertension (systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg on 2 determinations at least 6 hours apart) and proteinuria (≥300 mg in a 24-hour urine collection). Patients with multiple gestations, chronic hypertension, diabetes mellitus, and renal disease were excluded from the study. Patients with preeclampsia were subclassified according to the impedance to blood flow in both uterine arteries (as determined by Doppler ultrasound scan) into groups of patients with (n = 25) and without (n = 14) abnormal uterine artery Doppler velocimetry (UADV). UADV was measured before cord blood sampling, as previously described. Abnormal UADV was defined as a mean pulsatility index 2 SDs below the mean for gestational age. Cord blood samples were taken at the time of UADV, as previously described, and a stop solution was used to prevent adenosine degradation. Fetal plasma adenosine concentration was determined with the use of high-performance liquid chromatography. The detection limit was at least 10 μmol/L, and the intra- and interassay coefficients of variation were 4.8% and 6.7%, respectively. All patients signed consent forms that had been approved by the local Human Investigation Committee, as previously described.

Data are presented as mean ± SEM. Comparisons were performed with the unpaired t test. The statistical packages that were used were SPSS (version 12.0; SPSS Inc, Chicago, IL) and MedCalc (version 7.4.4.1; MedCalc Software, Mariakerke, Belgium). A probability value of < .05 was considered significant.

Results

There were no significant differences in the maternal age between patients with normal pregnancies and those with preeclampsia with and without abnormal UADV. Similarly, there were no significant differences in the gestational age at blood sampling between normal pregnancies (33 ± 1.3 weeks’ gestation) and patients with preeclampsia with (33 ± 1.1 weeks’ gestation) or without (34 ± 0.8 weeks’ gestation) abnormal UADV. In contrast, patients with preeclampsia with abnormal UADV delivered significantly earlier (35 ± 1 weeks’ gestation) than normal pregnant patients (37 ± 1 weeks’ gestation) and those with preeclampsia with normal UADV (38 ± 1.2 weeks’ gestation; P = .03).

The fetal plasma concentrations of adenosine did not change with gestational age. Fetal plasma concentrations of adenosine were significantly higher in patients with preeclampsia than those in patients with normal pregnancies (mean, 1.35 ± 0.09 μmol/L vs 0.52 ± 0.06 μmol/L; P < .0001). The fetal plasma concentrations of adenosine in patients with preeclampsia with abnormal UADV were significantly higher than those from normal pregnancies (mean, 1.78 ± 0.15 μmol/L vs 0.52 ± 0.06 μmol/L; P < .0001) ( Figure ). In contrast, the fetal plasma concentrations of adenosine from patients with preeclampsia with normal UADV were not significantly different from those of normal pregnancies (mean, 0.58 ± 0.14 μmol/L vs 0.52 ± 0.06 μmol/L; P = .6; Figure ).

Fetal plasma adenosine according to uterine artery Doppler velocimetry ( UADV )

Espinoza. High fetal plasma adenosine concentrations in preeclampsia. Am J Obstet Gynecol 2011 .

Comment

The results of this study indicate that patients with preeclampsia have higher fetal plasma concentrations of adenosine than uncomplicated pregnancies. Moreover, it is the subset of patients with preeclampsia with high impedance to blood flow in the uterine arteries that account for these differences. The observations that patients with preeclampsia with sonographic evidence of chronic uteroplacental ischemia have higher fetal plasma concentrations of adenosine than normal pregnancies are novel. Previous studies indicated that, among patients with preeclampsia, those with abnormal UADV had higher fetal plasma adenosine concentrations than those with normal impedance to blood flow in the uterine arteries, regardless of their acid-base status.

Adenosine is a ubiquitous nucleoside that is produced by dephosphorylation of adenosine triphosphate. Hypoxic tissues produce adenosine that has the tendency to restore the balance between oxygen supply and demand. This nucleoside is a potent vasodilator in the heart, brain, and skeletal muscle, whereas, to the contrary, it is a vasoconstrictor in the kidney and lungs, where it acts through the generation of a thromboxane/endoperoxide product. Thus, adenosine increases oxygen supply by increasing blood flow in some tissues. Adenosine also has suppressive actions in metabolic control, inhibiting energy consumption and conserving energy supply ; it has been termed a retaliatory metabolite . For instance, intravenous infusion of adenosine lowers oxygen consumption in anesthetized adults by 13% ± 4%. In human fetuses, adenosine counterbalances some of the effects of high levels of circulating catecholamines at birth and depresses whole-body oxygen use in fetal sheep. In simulated birth of fetal sheep, plasma adenosine concentration falls 50% after cord occlusion, and the change is accompanied by marked elevations in whole-body oxygen use, which is a result that is consistent with the concept of adenosine modulation of fetal metabolism.

Other means by which adenosine maintains tissue oxygenation include the stimulation of angiogenesis and the proliferation and migration of endothelial cells by mechanisms that are dependent and independent of VEGF. In the placenta, adenosine signaling appears to be central to the increased production of sFlt-1, which is an antiangiogenic factor that has been implicated in the pathophysiology of preeclampsia.

Clinical and experimental evidence suggest that uteroplacental ischemia may result in maternal hypertension and increased circulating concentrations of antiangiogenic factors (sVEGFR-1 and soluble endoglin among others). Evidence that supports a role for placental ischemia includes (1) reduced uterine perfusion in pregnant nonhuman primates and rats is associated with hypertension and increased placental expression of antiangiogenic factors; (2) cytotrophoblast that is cultured under hypoxic conditions up-regulates the messenger RNA expression and production of sVEGFR-1 in the supernatant; (3) increased expression of sFlt-1 of human placental is mediated by hypoxia inducible factor-1; (4) among patients with preeclampsia, the higher the impedance to blood flow in the uterine arteries (a surrogate marker of chronic uteroplacental ischemia) the higher the maternal plasma concentration of antiangiogenic factors; and (5) histologic lesions suggestive of chronic trophoblast ischemia have been associated with preeclampsia and include decidual arteriolopathy, central villi infarction, and hypermaturity of villi in “classic preeclampsia,” severe villous edema in “Mirror Syndrome” and “avascular” villi in pregnancies that are complicated with mole and partial mole. Collectively, this evidence indicates that chronic uteroplacental ischemia may be associated with angiogenic imbalances during pregnancy. However, it would seem doubtful that reproductive evolution would allow chronic trophoblast ischemia to lead into an antiangiogenic state to compromise the survival of both the mother and the fetus. This apparent paradox could be explained by the concept of the fetal-maternal conflict, wherein fetal growth and development may proceed at the expense of the maternal wellbeing. Using this framework, we speculated that chronic uteroplacental ischemia would limit the amount of substrates that are available for fetal growth that would then, in turn, signal the placental release of antiangiogenic factors to increase the maternal blood pressure and compensate for limited blood flow to the placental and fetal tissues. The magnitude of angiogenic imbalances, gene-environment interaction, and other factors would determine whether a patient with chronic trophoblast ischemia would experience preeclampsia, fetal growth restriction, both, or any of the other intermediate phenotypes. Chronic trophoblast ischemia appears to be less relevant in the pathophysiology of late-onset preeclampsia (>34 weeks’ gestation). Indeed, the latter frequently is associated with fetuses that are adequate or large for gestational age. It is possible that, in late-onset preeclampsia, an increased fetal demand for substrates that surpass the placental ability to sustain fetal growth may induce fetal signaling for placental over-production of antiangiogenic factors and subsequent “compensatory” maternal hypertension.

It is possible that, in the context of chronic uteroplacental ischemia, the fetus may use the adenosine system and/or other signaling mechanisms to increase the maternal blood pressure in an attempt to increase uteroplacental blood flow. An elegant in vitro study provided compelling evidence in support of this view. In that study, the authors determined the adenosine concentrations in fetal venous perfusates with the use of isolated dual-perfused human placental cotyledons where the fetal compartment and the intervillous space were perfused under controlled conditions. The authors reported that cessation of “maternal” perfusion was associated with a 2- to 6-fold increase in fetal venous perfusate concentrations of adenosine and a concomitant increase in fetoplacental perfusion pressure. Furthermore, perfusate pressure and the concentration of adenosine in the fetal compartment returned to baseline levels on reperfusion of the “maternal” circuit. A more recent study that used placental explants from rats indicates that exogenous adenosine administration significantly increases the concentration of sFlt-1 in the supernatants in normoxic conditions and that the addition of dipyridamole (an adenosine transporter antagonist that increases extracellular adenosine concentration) to cell cultures leads to a 1.6-fold increase in the concentrations of sFlt-1 in the supernatants. Moreover, although hypoxia was associated with a 2-fold increase in the concentrations of sFlt-1 in the supernatant, blockade of adenosine signaling (use of a nonspecific adenosine receptor antagonist) blunted the hypoxic effect on the concentrations of sFlt-1 and VEGF to a level similar to normoxic conditions. These results indicate that adenosine signaling is important for placental induction of sFlt-1 under both normoxic and hypoxic conditions.

It is appropriate to note that not all observations support a role for increased circulating fetal adenosine in the placental over-expression and release of antiangiogenic factors. Adenosine is thought to function largely as a paracrine mediator, because it is broken down quickly in the human circulation; its half-life is <10 seconds in adults ; thus, its actions are likely to be confined to locations where it is released, such as the placenta. Also, the hypertension that characterizes preeclampsia is unlikely to be brought about by adenosine, but rather solely by secondary mediators, because adenosine itself is predominantly a vasodilator in most tissues (the placenta and lungs are notable exceptions). It is also possible that changes in circulating fetal adenosine may be a consequence of pathologic processes in preeclampsia, rather than an initiator of them.

Previous reports indicate that the maternal plasma concentration of adenosine in patients with preeclampsia is also significantly higher than that of uncomplicated pregnancies. However, the circulating concentration of adenosine in preeclamptic patients is, on average, 2.3 times lower than their fetal counterparts. Studies that have used human placental perfusates indicate that placental uptake of adenosine is a Na-independent, carrier-mediated process. Moreover, studies that have used dual-perfused guinea pig placenta indicate that only 5-20% of adenosine metabolic products reach the fetal circulation and that maternal adenosine normally does not reach the fetal circulation. Thus, it is unlikely that circulating fetal plasma adenosine is of maternal origin.

Limitations of this study include a limited sample size; therefore, data collection was not powered for this secondary data analysis. Because sample analysis was done several years ago, more sensitive and contemporary techniques may determine whether adenosine or other nucleosides participate in fetal signaling in patients with preeclampsia without sonographic evidence of chronic uteroplacental ischemia.

In summary, the results of the present study demonstrate that patients with preeclampsia with sonographic evidence of chronic uteroplacental ischemia have high fetal plasma concentrations of adenosine. Although we recognize the limitations mentioned earlier, the balance of observations that have been reviewed herein suggest collectively, that fetal signaling may participate in the mechanism of disease in patients with preeclampsia with chronic uteroplacental ischemia.