Sildenafil for the Treatment of Preeclampsia, an Update

Should We Still Be Enthusiastic?

Noémie Simon-Tillaux; Edouard Lecarpentier; Vassilis Tsatsaris; Alexandre Hertig


Nephrol Dial Transplant. 2019;34(11):1819-1826. 

In This Article

Relevance of SC for Preeclampsia: Some Promising Preclinical Studies

At present, no specific drug targets sFlt-1, which is why a drug targeting the downstream effects in the preeclamptic maternal endothelium is promising. SC acts through a highly specific inhibition of PDE5, the enzyme that consumes cGMP by converting it into guanosine-5'-triphosphate. cGMP is the first of the second messengers in the NO pathway[23] (Figure 1). The drug was developed in the 1990s as a vasodilatory agent to improve heart failure, but in dose–response Phase I studies it was shown to potentialize the effects of glyceryl trinitrate and to induce penile erection. Eventually erectile dysfunction became the primary indication of SC, with pulmonary arterial hypertension also an indication, but to a much lesser extent (the alveolar endothelium is dependent on NO to become distended in response to inspiration).[23] Although PDE5 is expressed in almost all tissues, SC selectively acts on only some vascular beds, resulting in a modest reduction in blood pressure in humans, a desirable profile in pregnancy, as a sudden systemic blood pressure drop is deleterious for the foetus.[23] PDE5 inhibitors are thus an interesting alternative to previous strategies that aimed to manipulate the NO pathway in preeclamptic women. NO donors such as glyceryl trinitrate or isosorbide dinitrate or NO precursors such as L-arginine have not proven to be entirely satisfactory.[22] PDE inhibitors were tested ex vivo in myometrial biopsies and induced vasodilation in arteries dissected out from preeclamptic subjects, without any effect on placental or omental arteries.[25,42] In a rat model of preeclampsia induced by suramin (a potent antiangiogenic drug), SC rescued the inhibitory effect on the relaxation induced by nitro-L-arginine methyl ester (L-NAME; a NOS inhibitor) or 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (a guanylyl cyclase inhibitor, the enzyme that produces cGMP) on thoracic aorta explants.[43] This efficacy was confirmed ex vivo on umbilical arteries isolated from the placenta of preeclamptic pregnancies.[44] In several studies in pregnant rats exposed to L-NAME, the effect of SC on blood pressure control, foetal weight and survival, microalbuminuria (along with a lower quantity of urinary messenger RNA of podocin and nephrin, used here as markers of glomerular injury) has been demonstrated.[36–40] In this model, the pups had impaired learning, also restored by SC, suggesting a protective effect not only on the maternal endothelium but also in the offspring.[45] Furthermore, mice carrying a heterozygous deletion of the eNOS gene, affected by a mild hypertensive phenotype, displayed better arterial pressure control and foetal outcome under SC.[46] SC has also been reported to be effective in COMT-deficient mice (a model where preeclampsia is mild but not directly caused by inhibition of NO). It had no impact on blood pressure—and even increased albuminuria—but it improved foetal growth by 20%, most probably by reducing placental arterial resistance.[37]

The efficacy of SC might actually go beyond NO metabolism. When given in nonpregnant COMT-deficient mice, SC decreases the serum accumulation of fatty acids and its oxidation intermediates, restoring circulating levels of kynurenine, a precursor of nicotinamide adenine dinucleotide, a rate-limiting compound of oxidative phosphorylation.[47] These results are relevant because mitochondrial dysfunction (on oxidative phosphorylation) is evident in the placenta of preeclamptic women.[48] SC was also found to be of benefit in another model of preeclampsia induced by the clamping of uterine arteries, where the placenta is hypoxic and sFlt-1 is high; of note, the drug had little effect on blood pressure in control rats.[49] Gillis et al.,[50] however, described the real antihypertensive effect of SC in Dahl salt-sensitive rats, along with decreased proteinuria, decreased uterine artery resistance and a better outcome in pups (increased foetal survival, size and weight).

Timely mechanistic data were provided by the work of Burke et al.,[19] where SC attenuated sFlt-1-induced preeclampsia and restored a normally low sensitivity to AngII in pregnant mice. By blunting VEGF-driven phosphorylation (hence activity) of eNOS, sFlt-1 increases oxidative stress because NO would normally act as a scavenger of oxidants. In preeclamptic vessels, AngII sensitivity was increased by L-NAME and reversed by SC, confirming the pivotal role played here by NO. Together, these are encouraging data for measurement of the clinical efficacy of SC in humans.