The use of SC at the Onset of Preeclampsia: Clinical Studies
To begin with, a meta-analysis pooled the results of preclinical and clinical studies, pointing out some pitfalls in the translation of SC use from rodents to humans: preclinical studies used higher doses (from 4 to 100 mg/kg/day, so at least four times the dose used in clinical trials) and at an earlier stage of pregnancy, when placentation is still ongoing. It should be stressed that prolongation of pregnancy is a major aim since it decreases both morbidity at early and late terms of gestation[52,53] and cost. However, this is a debatable criterion for outcome since it is dependent on the decision of the investigators to move delivery forward. The ideal (but more complex) criterion would be the preeclampsia-free duration of pregnancy in the case of its introduction before the onset of clinical manifestations. Other PDE5 inhibitors could also be of interest. Vardenafil, for instance, showed an even better vasodilatory effect on umbilical arteries from preeclamptic pregnancies and an increase in the secretion of PlGF by endothelial cells.[44,55] Tadalafil might be advantageous, as it appears not to cross the placental barrier and was recently found to be safe in a small safety trial (although only eight women were included).
In 2003, years before any in vitro or in vivo evidence, Downing et al. hypothesized that PDE5 inhibitors would be beneficial in preeclampsia. This was based on the idea that accumulation of cGMP would increase vasodilation and NO production would prevent oxidative stress. The first double-blind, placebo-controlled clinical trial of the use of SC in preeclampsia was published 6 years later. Patients were included if they had early-onset preeclampsia (defined here as hypertension plus proteinuria >500 mg/day between 24 and 34 weeks of gestation). The primary outcome was the prolongation of pregnancy. Gestational age at randomization was 31 + 4 weeks in the SC group and 29 + 0 in the placebo group. The study found no significant difference between groups. Nonetheless, the size of groups (20 subjects each) was probably too small to draw a firm conclusion, with SC, whose half-life is <4 h, given only once a day at a very low dose compared with that used in preclinical studies (20 mg/day, with a progressive increase up to 80 mg if the drug was well tolerated). Importantly, in this study, where sildenafil use never exceeded 15 days and where mean gestational age at inclusion was ~30 weeks, there were no safety issues regarding the neonates. In particular, surfactant-deficient lung disease and mortality were comparable between groups. This trial revealed a good SC tolerance profile during preeclampsia and a significant antihypertensive effect: mean diastolic blood pressure decreased from 88 mmHg at randomization to 80 at delivery, while it rose from 90 to 96 in the placebo group.
The same group performed a parallel study in which they used pressure myography to analyze the relaxation of the arteries extracted from the umbilical cord and the omentum of the study subjects. They found no difference between groups, showing that the drug had no effect on ex vivo samples.
Trapani et al. recently published similar results in another placebo-controlled trial with a larger cohort of subjects. Inclusion criteria were similar (early-onset preeclampsia), but with a higher dose and a better distribution of SC: 50 mg every 8 h. While in the first study most women were nulliparous, the population of this trial was more heterogeneous, with half being multiparous. Mean gestational age was also ~30 weeks. A 4-day increase in the duration of pregnancy was observed in the treated group, as well as a significant decrease in mean arterial pressure (from 116.4 ± 5.1 to 100.3 ± 5.6 mmHg at 24 h); a reduced pulsatile index in the uterine and umbilical arteries was also noted (of a modest magnitude compared with the numbers typically observed in healthy pregnancies) without affecting the velocimetric profile of the foetal middle cerebral artery. Outcomes for the baby after delivery were similar to the placebo group, arguing in favour of a selective maternal effect.[61,62] Respiratory distress syndrome in particular was equivalent between groups, as was death.
Disturbing Results From the STRIDER Trial
The STRIDER trial (Sildenafil TheRapy in dismal prognosis early onset fetal growth restriction) was meant to present further data on the tolerance and relevance of the prolonged use of sildenafil in early pregnancy (between 22 and 30 weeks of gestation) complicated by IUGR. The UK arm of the trial, conducted in 19 foetal medical units, was published in February 2018. It failed to demonstrate a beneficial effect of low-dose SC (25 mg three times per day) to prolong pregnancy: time to delivery was 17 days in women exposed to SC versus 18 days in the placebo group. The safety profile of SC was good overall, and perinatal mortality and morbidity were statistically comparable (including the use of surfactant in neonates). However, on 24 July 2018, the lead investigator of the Dutch arm, Dr Wessel Ganzevoort, gave a public interview to de Volskrant, a Dutch daily morning newspaper, in which he reported that SC was associated with an increased risk of pulmonary hypertension in newborn babies (17 cases versus 3 in the placebo group) and increased mortality (19 versus 8, of which 11 versus 0 had pulmonary issues). This information led to the suspension of the Canadian arm of the trial and was widely reported in the international nonspecialist press.
Nephrol Dial Transplant. 2019;34(11):1819-1826. © 2019 Oxford University Press