Skip main navigation
×

Sildenafil During Pregnancy

A Preclinical Meta-Analysis on Fetal Growth and Maternal Blood Pressure
Originally publishedhttps://doi.org/10.1161/HYPERTENSIONAHA.117.09690Hypertension. 2017;70:998–1006

Abstract

Sildenafil is a new approach to treat fetal growth restriction (FGR) and preeclampsia. We performed a systematic meta-analysis to evaluate effects of sildenafil. Our search identified 22 animal studies (mouse, rat, rabbit, sheep, and guinea pigs) and 2 human randomized controlled trials. Data were pooled using ratio of means and mean differences with 95% confidence intervals for fetal growth and maternal blood pressure, respectively. Meta-regression analyses were performed for study-related factors that might affect efficacy of sildenafil, including the model used (healthy pregnancy versus FGR/preeclampsia) and route of administration. Dose–response curves with dose per metabolic weight (mg/kg0.75 per 24 hours) were fitted using splines. Our analyses show that sildenafil increases fetal growth during FGR/preeclampsia pregnancy compared with healthy pregnancy (1.10 [1.06–1.13] versus 1.03 [0.99–1.06]; P=0.006). There was no significant effect on fetal growth in the absence of FGR/preeclampsia. Effects were similar among different species and largest after oral and continuous administration. There was a positive relation between dose and fetal growth up to a human equivalent dose of ≈450 mg/d. A significant blood pressure–lowering effect of sildenafil is present during FGR/preeclampsia pregnancy only (−19 [−25 to −13] mm Hg; P<0.01), with the effect size being highly dependent on baseline blood pressure and without effect in the absence of hypertension. This meta-analysis supports that sildenafil improves fetal growth and maternal blood pressure regulation during FGR and preeclampsia pregnancy. The greatest beneficial effects on fetal growth are with dosages greater than those currently used in human studies.

Introduction

Fetal growth restriction (FGR) and preeclampsia complicate ≈5% of all pregnancies.1 Both disorders often originate from impaired development of uteroplacental vasculature during the first half of pregnancy, which can lead to placental insufficiency. FGR might develop secondary to suboptimal oxygen and nutrient transport to the fetus. In addition, the clinical manifestations of preeclampsia (maternal hypertension and proteinuria/other organ disturbances) are thought to develop in response to the release of (hypoxic) placental factors by the insufficient placenta. Because there is not yet an effective therapy for these disorders, they are managed by careful monitoring of fetal and maternal condition and timely delivery when they seem to be compromised. The direst consequences are in the early preterm period when prematurity and growth restriction implicate major risks of perinatal morbidity and mortality.2

A new approach to treat FGR and preeclampsia is administration of sildenafil. Sildenafil is a phosphodiesterase 5 (PDE5) inhibitor, which enhances NO-mediated effects by inhibiting cyclic guanosine monophosphate degradation. During pregnancy, NO activity increases, mediating essential vascular adaptation, such as reducing maternal peripheral vascular resistance3 and creating the low-resistance/high-caliber uteroplacental unit needed to provide adequate blood flow to the fetus.4 Increasing NO availability during pregnancy might overcome deficits in placental and systemic NO reported in FGR and preeclampsia and thereby improve placental function and maternal endothelial function.5 Moreover, sildenafil might counterbalance vascular dysfunction because of augmented vasoconstrictors and antiangiogenic factors.6,7 Because no pronounced maternal and fetal side effects were observed in pregnant women using sildenafil for pulmonary artery hypertension,8 transition of sildenafil from preclinical to clinical studies was established for the indications FGR and preeclampsia. The first prospective studies and small clinical trials show beneficial effects of maternal oral sildenafil treatment on fetal growth9 and maternal blood pressure (BP)10,11 at fairly low dosages. The international collaborative STRIDER (Sildenafil Therapy in Dismal Prognosis Early-Onset Fetal Growth Restriction) initiative consists of 3 large randomized controlled trials (NCT02277132, NCT02442492, and ACTRN1261200058483), and 2 more in preparation that will evaluate the effect of sildenafil on fetal growth in FGR pregnancies.12 Although clinical trials are already underway, no recent systematic review with effect-size estimation has been performed on all the preclinical and clinical studies in pregnancy. Such systematic reviews might help to improve the design of clinical studies, thereby resulting in better health care and more transparent translational medicine.13 Therefore, we performed a meta-analysis on all available studies in which effects of sildenafil on fetal growth and maternal BP are reported. Our primary aim was to identify modifiable factors contributing to effect size in FGR or preeclampsia pregnancies by performing subanalyses based on study designs (eg, dose and route of administration).

Methods

Literature Search

On November 24, PubMed, Embase, and the Cochrane library databases were searched separately for animal studies and human RCTs, published before November 1, 2016, using prenatal treatment with sildenafil and reporting fetal growth or maternal BP (search string Table S1 in the online-only Data Supplement).

Inclusion and Exclusion of Studies

Two investigators (N.D.P. and F.T.) independently screened the articles in title abstract and full text. Studies performed in healthy pregnant animals and animal models for preeclampsia or FGR were selected when the effect of sildenafil on fetal or birth weight or maternal BP was reported after prenatal administration of sildenafil in the presence of an untreated control group. Studies were excluded in case of (1) missing data, (2) combined interventions with other means in addition to sildenafil, or (3) outcome measurements after single dose or single-day (nonchronic) administration of sildenafil. The RCT by Trapani et al10 was excluded from the BP analyses because these measurements took place 24 hours after the first administration. In addition, at later time points, there is a large difference in the use of additional antihypertensive drugs between the groups, which hinders reliable comparison of mean differences (MDs) in BP.

Data Extraction

Data were extracted by the 2 independent investigators (N.D.P. and F.T.). A third investigator (H.G.) was consulted if there was no consensus on inclusion or data extraction. We extracted data on study characteristics, including species, animal model (FGR/preeclampsia), strain, and maternal weight. For administration strategy, the dose, route, duration, and scheme were extracted. Maternal weight was used to calculate the dose of sildenafil in mg·kg·d when dose was presented in mg/d in the original article. If maternal weight was not provided, we contacted the corresponding author. In the absence of a response, we estimated this based on species, strain, and age of the animal (specified as such in Table 1 and Tables S2 and S3). For maternal BP, we extracted information on measurement techniques, BP type, and gestational day of measurement. Graph digitizers were used if data were only graphically presented. If medians were reported, we estimated mean and SD with Hozo formulas.14 SEM and pooled SEM were converted to SDs by using reported group size. The corresponding author was contacted if data were incomplete or if there was no access to full text of the article.

Table 1. Overview of the Main Study Characteristics of Included Studies With Specification of Study Population, Sildenafil Administration, and Reported Outcomes

Author (y)Study PopulationSildenafil AdministrationOutcomes Reported
SpeciesModelPregnancy (Healthy, FGR, or PE)RouteDose, mg·kg·dDuration (Part/Full Gestation)Fetal GrowthMaternal BP
Abou-Tarboush (2011)28Mouse (SWR/J)WTHealthyOral6.5, 13, 19.5, 26, 32.5, and 403 d at 3 different intervalsx
Dilworth (2013)47Mouse (C57Bl/6J)WT, Igf2−/−Healthy, FGROral50*Partx
Lopez-Tello (2016)48Rabbit (NZ×CW)50% protein restrictionFGROral5Partx
Luong (2011)27Rat (SD)WTHealthySC100Partx
Motta (2015)49Mouse (BALBc)WT, L-NAMEHealthy, PE+FGROral10Partx
Nassar (2012)50Rat (SD)WT, L-NAMEHealthy, PE+FGROral15Partx
Oyston (2016)51Sheep (RC)embolization of uterine arteriesFGRSC1.9Partx
Ozdegirmenci (2011)46Rat (Wistar)WTHealthyOral15Partx
Refuerzo (2006)52Rat (Long-Evans)HypoxiaFGROral90Partx
Sanchez-Aparicio (2008)53Guinea pigPerinatal asphyxia by cord clampingHealthyOral0.05 and 0.5Partx
Satterfield (2010)54Sheep (Suffolk)WT, 50% diet restrictedHealthy, FGRSC0.94 and 1.88*Partx
Trapani (2016)10HumanNAPE (+60% FGR)Oral2*Partx
Burke (2016)6Mouse (CD1)sFlt-1 adenovirusPE+FGROral100Partxx
George (2013)55Rat (SD)WT, RUPPHealthy, PE+FGROral45Partxx
Gillis (2016)56Rat (SS-Dahl rat)Spontaneous PEPE+FGROral50Partxx
Herraiz (2012)57Rat (Wistar)WT, L-NAMEHealthy, PEOral4Fullxx
Pellicer (2011)58Rat (Wistar)WTHealthyOral4Fullxx
Ramesar (2010)59Rat (SD)L-NAMEPE+FGRSC10Partxx
Roberts (2016)60Mouse (C57BL/6J)WT, eNOS+/−Healthy, PEOral100*Fullxx
Sasser (2010)31Rat (SD)WTHealthyOral10, 50, and 90Partxx
Stanley (2012)61Mouse (C57BL/6J)WT, COMT−/−Healthy, PE+FGROral50*Partxx
Samangaya (2009)11HumanNAPE (+30% FGR)Oral0.78*Partxx
Turgut (2008)62Rat (Wistar)SuraminPEIP5Partxx
Cauli (2010)63Rat (Wistar)WT, L-NAMEHealthy, PEOral4Fullx

BP indicates blood pressure; FGR, fetal growth restriction; NA, not applicable; PE, preeclampsia; and WT, wild type.

*Fixed dose of sildenafil during pregnancy, not adjusted for gestational weight gain. mg·kg·d represents the dose at start of administration.

Dose in mg·kg·d was calculated using estimated mean maternal weight.

Statistical Analysis

Data were analyzed using R software (version 3.1.1), using the metafor package.15 For difference in fetal body weight, we used the ratio of means (ROM), calculated by meansildenafil /meancontrol as outcome measure.16 MD (meansildenafil−meancontrol) was used as effect-size measurement for BP. Random- and mixed-effects models were fitted using restricted maximum likelihood estimation. Because some studies included multiple cohorts with different dosages of sildenafil, multilevel models were used to account for within-study clustering of data points.17,18 Data are reported as estimated average effect (µ) with 95% confidence interval throughout. In primary analyses, unexplained interstudy heterogeneity (I2) is reported, and a value of >50% was considered significant.19 Meta-regression was used to examine the effects of moderator variables on heterogeneity in studies with FGR/preeclampsia pregnancies using the mouse, rat, and human studies. Factors related to study design (species; percentage of growth restriction), administration strategy (timing in pregnancy; route; scheme, dose), baseline BP, and BP type were investigated. For dose–response relationships, differences in dosing between species were corrected by applying a linear scaling exponent of 0.75,20 resulting in dose in mg per metabolic body weight per 24 hours (mg/kg0.75 per 24 hours). This approach was verified in a subset of studies where pharmacokinetic data for sildenafil in the investigated species and administration route was available. In this case, the area under the curve per 1 mg/kg sildenafil after a single dose was used as a normalization factor. Human equivalent dosing was attained by converting doses in mg·kg·d to mg/d using an estimated maternal weight of 75 kg. Nonlinear dose–response models were applied using restricted cubic splines. Two-sided P value <0.05 was considered significant.

Methods in the online-only Data Supplement: assessment of quality and publication bias.

Results

Study Selection and Characteristics

The literature search on the use of sildenafil in animal studies resulted in a total of 459 hits of which 22 met the inclusion criteria. In a separate search for human RCTs with chronic use of sildenafil, we identified 2 studies (flowchart, Figure S1). Table 1 summarizes the study characteristics of the 24 studies. Data on fetal growth were extracted from 23 studies, and data on maternal BP from 12 studies (Tables S2 and S3).

Effect of Sildenafil on Fetal Growth

The meta-analysis shows that sildenafil significantly improves fetal growth in the FGR/preeclampsia group but not in the healthy pregnant groups (1.10 [1.06–1.13] versus 1.03 [0.99–1.06]; P=0.006; Figure 1). Within the FGR/preeclampsia studies, the study by Gillis et al was identified as an outlier (Figure S2). When this study was excluded, the estimate changed only slightly (ROM 1.09 versus 1.10).

Figure 1.

Figure 1. Meta-analysis of sildenafil treatment on fetal growth. Forest plot showing the effect of sildenafil on fetal growth according to the model used (healthy pregnancy vs fetal growth restriction [FGR]/preeclampsia [PE]). Data are presented as ratio of means (ROM) and 95% confidence intervals (CIs). Same dose administered during gestational day: a7 to 9, b10 to 12, and c13 to 15.

Because our primary aim was to investigate effects of sildenafil on FGR, we excluded study cohorts with healthy pregnancies from further analysis. Meta-regression within FGR/preeclampsia studies shows that differences in species and administration strategies did not significantly influence efficacy of treatment (Figure S3). However, a trend toward a stronger effect was observed when the drug was administered for the full duration of pregnancy and when it was administered orally and continuously. No correlation was found between degree of growth restriction and the sildenafil effect on fetal growth (Figure S4).

Dose–response curves were created using the dose in mg per metabolic body weight per 24 hours (mg/kg0.75 per 24 hours) because the doses of sildenafil per kg body weight differed ≤2 orders of magnitude between rodents and large animals. Across species, the dose–response relationship shows a positive relationship between dose and fetal growth, reaching a plateau effect at an ROM of 1.10, at a human equivalent dose of 467 mg/d (Figure 2). Because most studies used a high dose with the intent of establishing a positive effect, doses were likely in the upper region of the dose–response range. Nevertheless, we observed smaller effects on offspring body weight at the lower end of this dose range. Because the curve fit in meta-analysis started at a ROM of 1.04 at a human equivalent dose of 60 mg/d, the effect of lower dosages was extrapolated to 0 mg/d, which per definition is at ROM=1. Only 8 studies assessed placental function in the presence of FGR/preeclampsia (Table S4). The variety of techniques and outcome variables did not allow a useful meta-analysis.

Figure 2.

Figure 2. Dose−response curve of sildenafil treatment on fetal growth in fetal growth restriction/preeclampsia models. Dose−response curve fitted using regional polynomial spline and showing the effect of sildenafil dose in mg per kg metabolic weight per day (mg/kg0.75 per 24 h) on fetal growth. Effect size of fetal growth is presented as ratio of means (ROM); 95% confidence intervals are represented by the grey area. Crossed dot, human study; diagonal striped area, dose range with positive relationship between dose and effect size.

Effect of Sildenafil on Maternal BP

The meta-analyses for maternal BP show that sildenafil significantly decreases BP during FGR/preeclampsia pregnancy (MD, −19 [−25 to −13] mm Hg; P<0.01), whereas no significant BP-lowering effects were observed during healthy pregnancies (MD, −4 [−12 to 2] mm Hg; Figure 3). By performing outlier analyses, we identified the study by Gillis et al as having a large effect size (Figure S5). Exclusion of this study did not affect the observed effect size. In the meta-regression analysis using the FGR/preeclampsia pregnancies, we did not observe significant differences in BP lowering between species, type of BP measured, or strategy of administration (Figure S6).

Figure 3.

Figure 3. Meta-analysis of prenatal sildenafil treatment on maternal blood pressure (BP). Forest plot showing the effect of sildenafil on maternal BP according to the model used (healthy pregnancy vs fetal growth restriction [FGR]/preeclampsia [PE]). Data are presented as ratio of means and 95% confidence intervals (CIs). DBP indicates diastolic BP; MAP, mean arterial pressure; RE, random effects; and SBP, systolic BP.

In further analysis, the BP-lowering effect was observed to be linearly associated with baseline BP with almost no BP-lowering effect in the absence of hypertension (Figure 4A). Baseline BP could explain nearly all heterogeneity between studies. The dose–response curve shows a steep decrease in BP at lower dosages up to ≈90 mg human equivalent daily dose (Figure 4B) and decreases only slightly further at higher dosages. We must note that a limited number of studies investigated intermediate-range dosages between 90 and 800 mg/d. We were, therefore, not able to pinpoint an inflection point, at which a plateau is reached.

Figure 4.

Figure 4. Factors influencing the effect of sildenafil treatment on maternal blood pressure (BP) growth in fetal growth restriction/preeclampsia models. The importance of baseline BP in determining BP-lowering effect for studies reporting on mean arterial pressure (MAP) and systolic BP (SBP; A). Dose–response curve shows the effect of sildenafil dose in mg per kg metabolic weight per day (mg/kg0.75 per 24 h) on maternal BP (B). Effect size on maternal BP is presented as mean difference in BP. 95% confidence intervals are represented by the grey area. Crossed dot, human study; open dot, outlier excluded based on >2 SD.

Quality Assessment of Studies and Publication Bias

Figure S7A and S7B shows quality assessment of the studies. We did not find publication bias for either fetal growth or maternal BP (Figure S8A and S8B).

Discussion

This meta-analysis confirms that administration of sildenafil increases fetal growth and decreases maternal BP during compromised pregnancies. Our findings proved to be consistent, despite considerable species differences between studies. Sildenafil did not significantly increase fetal growth in the absence of FGR and preeclampsia, and no BP-lowering effect could be observed in the absence of FGR, preeclampsia, and hypertension.

To our knowledge, this is the first meta-analysis providing a complete overview of all literature on effects of sildenafil on fetal growth and maternal BP. To be able to pool data from preclinical and clinical studies, we attempted to identify outcome measures that could be adapted across animal species to the human situation. These included the ROM for fetal growth to provide accessible effect size that is scalable across 3 to 4 orders of magnitude in body weight and metabolic weight for interspecies drug-dosage conversion. These adaptations were proven valid because we found low heterogeneity between studies and no significant differences in the outcome between species. For maternal BP, no adaptations were required because we found the MDs to be directly translatable across species and similar for systolic BP and mean arterial pressure. With our approach, we created a rich source of information, which could be used to study factors contributing to efficacy of sildenafil.

The results of our meta-analysis show that sildenafil has the potential to improve fetal growth during suboptimal intrauterine growth conditions with a maximum potential gain in fetal weight of ≈10%. Improvement of growth is thought to be because of an increase in uteroplacental blood flow by sildenafil leading to higher availability of oxygen and nutrients to the growing fetus10 and is attributed to inhibition of PDE5 in the uterine vasculature, including uterine arteries21 and smaller vessels in myometrium and placenta.22,23 We were not able to assess whether the effect of sildenafil on fetal growth is indeed accompanied by increased uteroplacental blood flow because only 8 studies assessed placental function, and method and timing of assessment was variable (Table S4). Improved fetal growth secondary to enhanced placental perfusion is supported by the finding that sildenafil does not protect against FGR in nonplacental species, such as the chicken.24

Our meta-analysis also shows that sildenafil reduces maternal BP during FGR/preeclampsia pregnancy but not during uncomplicated pregnancy. Not unexpectedly, we found that the BP-lowering effect is totally dependent on initial BP levels with almost no BP-lowering effect in the absence of hypertension. Absence of BP-lowering effects of sildenafil were reported in normotensive controls in various models of hypertension in male rodents.25,26 It is reassuring that this can be confirmed in normotensive pregnant animals because inducing hypotension may compromise uterine flow. BP-lowering actions might be mediated by systemic NO-mediated vasodilatation and by interference with circulating antiangiogenic factors that affect maternal endothelium.6,7 By these actions, sildenafil can protect the maternal vasculature and might prolong the time to pregnancy termination on maternal indication, as already shown in human RCTs.10,11 In the human RCT by Samangaya et al,11 lower BP at 2 weeks after administration was observed in the sildenafil group. In addition, the RCT by Trapani et al10 showed that the need for additional antihypertensive drug or an increase in the initial dosage of alpha methyldopa after randomization was higher in the placebo group (58%) compared with the sildenafil group (32%).

The dose–response curve shows that the maximum fetal growth effect is reached at doses of sildenafil corresponding to a human equivalent dose of ≈450 mg/d. This suggests that there is room for improvement in human studies by increasing the dose currently used, which is around 75 mg/d. However, this should be mentioned with caution because we were unable to correct for differences in pharmacokinetics of sildenafil between species. The allometric scaling method applied here only corrects for body weight and is unable to correct for species differences, such as differences in phase I and II drug metabolism during pregnancy, drug efflux transporter affinity, gestational weight gain, plasma volume expansion, blood flow distribution, and placental microvascular structure. These factors, many of which are affected by pregnancy disorders, such as preeclampsia, are likely to contribute to the pharmacokinetics of sildenafil. Using the limited pharmacokinetic data available, we created a dose–response curve with doses corrected for differences in bioavailability and half life between species, in which we do not observe large differences compared with the original dose–response curve (Figure S9). Still, for future studies, it is advisable to perform dose finding in compromised pregnancies in humans, to optimize fetal growth. The finding that the highest BP-lowering effect is reached at a lower dose of sildenafil compared with the dose with the maximum effect on fetal growth suggests that optimal treatment doses might well differ per indication.

With limited studies available on safety of sildenafil during pregnancy and because sildenafil passes the placenta,11,27 we recommend that this should go hand in hand with detailed study on pharmacokinetics and direct and long-term effects on offspring. Adverse effects of sildenafil that might occur at high doses include embryotoxicity28 and side effects, such as headaches that get more severe.29 Another adverse effect might be a reduction of the required plasma volume expansion during pregnancy by PDE5 inhibition in the kidney.30 Although reduced plasma volume expansion during sildenafil is suggested to occur in pregnant SD rats,31 it remains unclear whether this also occurs in human pregnancy. Because reduced plasma volume expansion might compromise uteroplacental blood flow, it is essential that future studies assess plasma volume expansion, by change in hematocrit, to assess the safety of PDE5, particularly when sildenafil is used in patients with compromised plasma volume expansion.

Our meta-regression shows that administration during full gestation has slightly more beneficial effects on fetal growth compared with administration during late gestation. This might be explained by the fact that NO is also important during early placentation, for endovascular invasion of the cytotrophoblast,32 immune modulation,33 and angiogenesis.34 Unfortunately, optimal timing of administration in human pregnancy is difficult to deduce from animal studies because of large differences in gestational duration and placental and fetal developmental stages between species.35 High-risk preventive treatment, as currently advised with low dose aspirin, could be a strategy when safety data for the first trimester become available and prediction of FGR/preeclampsia early in pregnancy gets more accurate.36 Another finding from the meta-regression analyses is that there are no significant differences in the effect of sildenafil between different routes of administration. Note that we were not able to adjust for differences in bioavailability of sildenafil via different administration routes. Previous studies comparing bioavailability after oral and systematic dosing have shown that sildenafil bioavailability is reduced by the hepatic first pass,37,38 suggesting a lower bioavailability after oral administration compared with intraperitoneal and subcutaneous administration. However, we found oral administration, used in most of the studies, achieves promising effects, which might be because the major metabolite of sildenafil, N-Desmethyl sildenafil, is an active compound with the same selectivity and comparable half life as sildenafil. In humans, the half life of sildenafil is 3 to 4 hours,39 which might lead to noncontinuous exposure during interval administration, compromising optimal growth, and indeed we observe a trend toward a favorable effect on growth after continuous administration. For this reason, it has been suggested to use tadalafil—a PDE5-inhibitor with faster onset and longer action.40,41 An alternative would be a slow-release oral drug.

The strength of a meta-analysis of all preclinical and clinical studies is that it supplies a rich source of useful information,13 particularly because of the variation in study design.42 Although analyzing animal models for such approaches has proven useful, it also has limitations. Available animal models for FGR/preeclampsia certainly do not perfectly match the human situation. Spontaneous placental insufficiency is rare in animals; thus models are used based on exogenous interventions (environmental triggers, surgery, or administration of drugs) and genetic modifications. All have their strengths (resemblances) and weaknesses (differences) with the human situation, which make direct translation of the findings difficult.43,44 Because we do not have the perfect animal model for FGR/preeclampsia, our premise is that pooled analysis on studies containing a large variety of species and models makes our conclusions on the effects of sildenafil more robust. For this very reason, some opinion articles suggest that potential therapeutic agents for treating FGR/preeclampsia should be tested in >1 animal model before translation to the clinic.45 Theoretically, variation in the model might result in different drug responses. We were unable to perform a detailed comparison in effect size between different models because of the large variety of models, with only few (between 1 and 4) studies reporting on effects in the same model. However, an effect of species did not explain variation in outcomes. Outlier analysis showed that the study by Gillis et al had an extremely high effect size for both fetal growth and maternal BP. This is an interesting finding because this study is the only study using a spontaneous model, that develops a combined phenotype of FGR and preeclampsia resembling the human situation, making this model a promising model for preeclampsia. Whether the findings of Gillis et al better reflect the effects that will be observed in humans remains an open question. Another limitation is that our dose–response curves required extrapolation because there are no studies using low doses of sildenafil. Finally, this meta-analysis only includes the 2 most reported outcome parameters, whereas other outcomes could be useful to guide drug choice. For example, potential beneficial effects of sildenafil on offspring brain and cardiovascular system have been reported.24,46

Perspectives

In summary (Table 2), our meta-analysis confirms that sildenafil is a promising treatment for FGR and preeclampsia because it improves fetal growth and reduces maternal BP. By including all preclinical and clinical studies, our analysis contained a large variation in study design. We show that in the presence of placental insufficiency (FGR/preeclampsia), sildenafil can increase fetal growth by ≈10%. Because we observe that optimal fetal growth might be reached at a daily dose about an order of magnitude higher than the dose currently used, we recommend human dose-finding studies for optimal fetal growth. This should go hand in hand with obtaining additional safety data in human pregnancy, especially on plasma volume expansion during pregnancy and on long-term fetal effects. Continuous exposure to PDE5 inhibition by administration of sildenafil or long-acting PDE5 inhibitors seems promising.

Table 2. Summary of Findings

Summary of Findings
Main conclusions
 We show that chronic administration of sildenafil improves fetal growth in normal and FGR/preeclampsia pregnancy, with a larger effect on fetal growth in the presence of FGR and preeclampsia.
 By combining clinical and preclinical studies, our analysis contained a large variation in study design that allows identification of factors affecting the efficacy of sildenafil, which is useful in a field where progress is slow because of safety and ethical concerns.
 Optimal fetal growth seems to be reached at a dose that is about an order of magnitude higher than the daily dose currently used in human studies.
 Sildenafil reduces maternal BP only in the presence of initially elevated BP.
Recommendations for future studies
 Dose-finding studies for optimal fetal growth with
  Mechanistic indicators, such as placental blood flow
 Additional safety data
  Pharmacokinetics and plasma volume expansion
  Long-term offspring effects
 Study efficacy of long-acting PDE5 inhibitors
BP indicates blood pressure; FGR, fetal growth restriction; and PDE5, phosphodiesterase 5.

Footnotes

*These authors contributed equally to this work.

The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.117.09690/-/DC1.

Correspondence to Nina D. Paauw, Division of Woman and Baby, Department of Obstetrics, University Medical Centre Utrecht, Postbus 85090, 3508 AB Utrecht, The Netherlands. E-mail

References

  • 1. Chaddha V, Viero S, Huppertz B, Kingdom J. Developmental biology of the placenta and the origins of placental insufficiency.Semin Fetal Neonatal Med. 2004; 9:357–369. doi: 10.1016/j.siny.2004.03.006.CrossrefMedlineGoogle Scholar
  • 2. Bernstein IM, Horbar JD, Badger GJ, Ohlsson A, Golan A. Morbidity and mortality among very-low-birth-weight neonates with intrauterine growth restriction. The Vermont Oxford Network.Am J Obstet Gynecol. 2000; 182(1pt 1):198–206.CrossrefMedlineGoogle Scholar
  • 3. Williams DJ, Vallance PJ, Neild GH, Spencer JA, Imms FJ. Nitric oxide-mediated vasodilation in human pregnancy.Am J Physiol. 1997; 272(2pt 2):H748–H752.MedlineGoogle Scholar
  • 4. Krause BJ, Hanson MA, Casanello P. Role of nitric oxide in placental vascular development and function.Placenta. 2011; 32:797–805. doi: 10.1016/j.placenta.2011.06.025.CrossrefMedlineGoogle Scholar
  • 5. Oyston CJ, Stanley JL, Baker PN. Potential targets for the treatment of preeclampsia.Expert Opin Ther Targets. 2015; 19:1517–1530. doi: 10.1517/14728222.2015.1088004.CrossrefMedlineGoogle Scholar
  • 6. Burke SD, Zsengellér ZK, Khankin EV, et al.. Soluble fms-like tyrosine kinase 1 promotes angiotensin II sensitivity in preeclampsia.J Clin Invest. 2016; 126:2561–2574. doi: 10.1172/JCI83918.CrossrefMedlineGoogle Scholar
  • 7. Ramesar SV, Mackraj I, Gathiram P, Moodley J. Sildenafil citrate decreases sFlt-1 and sEng in pregnant l-NAME treated Sprague-Dawley rats.Eur J Obstet Gynecol Reprod Biol. 2011; 157:136–140. doi: 10.1016/j.ejogrb.2011.03.005.CrossrefMedlineGoogle Scholar
  • 8. Cartago RS, Alan PA, Benedicto J. Pregnancy outcomes in patients with severe pulmonary hypertension and Eisenmenger syndrome treated with sildenafil monotherapy.Obstet Med. 2014; 7:40–42. doi: 10.1177/1753495X13514403.CrossrefMedlineGoogle Scholar
  • 9. von Dadelszen P, Dwinnell S, Magee LA, Carleton BC, Gruslin A, Lee B, Lim KI, Liston RM, Miller SP, Rurak D, Sherlock RL, Skoll MA, Wareing MM, Baker PN; Research into Advanced Fetal Diagnosis and Therapy (RAFT) Group. Sildenafil citrate therapy for severe early-onset intrauterine growth restriction.BJOG. 2011; 118:624–628. doi: 10.1111/j.1471-0528.2010.02879.x.CrossrefMedlineGoogle Scholar
  • 10. Trapani A, Gonçalves LF, Trapani TF, Vieira S, Pires M, Pires MM. Perinatal and hemodynamic evaluation of sildenafil citrate for preeclampsia treatment: a randomized controlled trial.Obstet Gynecol. 2016; 128:253–259. doi: 10.1097/AOG.0000000000001518.CrossrefMedlineGoogle Scholar
  • 11. Samangaya RA, Mires G, Shennan A, Skillern L, Howe D, McLeod A, Baker PN. A randomised, double-blinded, placebo-controlled study of the phosphodiesterase type 5 inhibitor sildenafil for the treatment of preeclampsia.Hypertens Pregnancy. 2009; 28:369–382. doi: 10.3109/10641950802601278.CrossrefMedlineGoogle Scholar
  • 12. Ganzevoort W, Alfirevic Z, von Dadelszen P, Kenny L, Papageorghiou A, van Wassenaer-Leemhuis A, Gluud C, Mol BW, Baker PN. STRIDER: sildenafil therapy in dismal prognosis early-onset intrauterine growth restriction–a protocol for a systematic review with individual participant data and aggregate data meta-analysis and trial sequential analysis.Syst Rev. 2014; 3:23. doi: 10.1186/2046-4053-3-23.CrossrefMedlineGoogle Scholar
  • 13. Ritskes-Hoitinga M, Leenaars M, Avey M, Rovers M, Scholten R. Systematic reviews of preclinical animal studies can make significant contributions to health care and more transparent translational medicine.Cochrane Database Syst Rev. 2014; 3:ED000078.Google Scholar
  • 14. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample.BMC Med Res Methodol. 2005; 5:13. doi: 10.1186/1471-2288-5-13.CrossrefMedlineGoogle Scholar
  • 15. Vliechtbauer W. Conducting meta-analyses in R with metafor package.J Stat Softw. 2010; 36:1–48.Google Scholar
  • 16. Friedrich JO, Adhikari NK, Beyene J. The ratio of means method as an alternative to mean differences for analyzing continuous outcome variables in meta-analysis: a simulation study.BMC Med Res Methodol. 2008; 8:32. doi: 10.1186/1471-2288-8-32.CrossrefMedlineGoogle Scholar
  • 17. Konstantopoulos S. Fixed effects and variance components estimation in three-level meta-analysis.Res Synth Methods. 2011; 2:61–76. doi: 10.1002/jrsm.35.CrossrefMedlineGoogle Scholar
  • 18. van Kesteren CF, Gremmels H, de Witte LD, Hol EM, Van Gool AR, Falkai PG, Kahn RS, Sommer IE. Immune involvement in the pathogenesis of schizophrenia: a meta-analysis on postmortem brain studies.Transl Psychiatry. 2017; 7:e1075. doi: 10.1038/tp.2017.4.CrossrefMedlineGoogle Scholar
  • 19. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses.BMJ. 2003; 327:557–560. doi: 10.1136/bmj.327.7414.557.CrossrefMedlineGoogle Scholar
  • 20. West GB, Brown JH, Enquist BJ. A general model for the origin of allometric scaling laws in biology.Science. 1997; 276:122–126.CrossrefMedlineGoogle Scholar
  • 21. Goulopoulou S, Hannan JL, Matsumoto T, Ergul A, Webb RC. Augmented dilation to nitric oxide in uterine arteries from rats with type 2 diabetes: implications for vascular adaptations to pregnancy.Am J Physiol Heart Circ Physiol. 2014; 306:H610–H618. doi: 10.1152/ajpheart.00588.2013.CrossrefMedlineGoogle Scholar
  • 22. Wareing M, Myers JE, O’Hara M, Baker PN. Sildenafil citrate (Viagra) enhances vasodilatation in fetal growth restriction.J Clin Endocrinol Metab. 2005; 90:2550–2555. doi: 10.1210/jc.2004-1831.CrossrefMedlineGoogle Scholar
  • 23. Wareing M, Myers JE, O’Hara M, Kenny LC, Taggart MJ, Skillern L, Machin I, Baker PN. Phosphodiesterase-5 inhibitors and omental and placental small artery function in normal pregnancy and pre-eclampsia.Eur J Obstet Gynecol Reprod Biol. 2006; 127:41–49. doi: 10.1016/j.ejogrb.2004.06.014.CrossrefMedlineGoogle Scholar
  • 24. Itani N, Skeffington KL, Beck C, Giussani DA. Sildenafil therapy for fetal cardiovascular dysfunction during hypoxic development: studies in the chick embryo.J Physiol. 2017; 595:1563–1573. doi: 10.1113/JP273393.CrossrefMedlineGoogle Scholar
  • 25. Yaguas K, Bautista R, Quiroz Y, Ferrebuz A, Pons H, Franco M, Vaziri ND, Rodriguez-Iturbe B. Chronic sildenafil treatment corrects endothelial dysfunction and improves hypertension.Am J Nephrol. 2010; 31:283–291. doi: 10.1159/000279307.CrossrefMedlineGoogle Scholar
  • 26. Cavalcanti CO, Alves RR, de Oliveira AL, Cruz JC, de França-Silva MS, Braga VA, Balarini CM. Inhibition of PDE5 restores depressed baroreflex sensitivity in renovascular hypertensive rats.Front Physiol. 2016; 7:15. doi: 10.3389/fphys.2016.00015.CrossrefMedlineGoogle Scholar
  • 27. Luong C, Rey-Perra J, Vadivel A, et al.. Antenatal sildenafil treatment attenuates pulmonary hypertension in experimental congenital diaphragmatic hernia.Circulation. 2011; 123:2120–2131. doi: 10.1161/CIRCULATIONAHA.108.845909.LinkGoogle Scholar
  • 28. Abou-Tarboush FM, Abdel-Samad MF, Al-Meteri MH. Developmental toxicity of orally administered sildenafil citrate (Viagra) in SWR/J mice.Saudi J Biol Sci. 2011; 18:135–139. doi: 10.1016/j.sjbs.2010.12.007.CrossrefMedlineGoogle Scholar
  • 29. Tsertsvadze A, Yazdi F, Fink HA, MacDonald R, Wilt TJ, Bella AJ, Ansari MT, Garritty C, Soares-Weiser K, Daniel R, Sampson M, Moher D. Oral sildenafil citrate (viagra) for erectile dysfunction: a systematic review and meta-analysis of harms.Urology. 2009; 74:831–836.e8. doi: 10.1016/j.urology.2009.04.026.CrossrefMedlineGoogle Scholar
  • 30. West CA, Sasser JM, Baylis C. The enigma of continual plasma volume expansion in pregnancy: critical role of the renin-angiotensin-aldosterone system.Am J Physiol Renal Physiol. 2016; 311:F1125–F1134. doi: 10.1152/ajprenal.00129.2016.CrossrefMedlineGoogle Scholar
  • 31. Sasser JM, Baylis C. Effects of sildenafil on maternal hemodynamics and fetal growth in normal rat pregnancy.Am J Physiol Regul Integr Comp Physiol. 2010; 298:R433–R438. doi: 10.1152/ajpregu.00198.2009.CrossrefMedlineGoogle Scholar
  • 32. Khalil A, Hardman L, O Brien P. The role of arginine, homoarginine and nitric oxide in pregnancy.Amino Acids. 2015; 47:1715–1727. doi: 10.1007/s00726-015-2014-1.CrossrefMedlineGoogle Scholar
  • 33. Kniotek M, Boguska A. Sildenafil can affect innate and adaptive immune system in both experimental animals and patients.J Immunol Res. 2017; 2017:4541958. doi: 10.1155/2017/4541958.CrossrefMedlineGoogle Scholar
  • 34. Pyriochou A, Zhou Z, Koika V, Petrou C, Cordopatis P, Sessa WC, Papapetropoulos A. The phosphodiesterase 5 inhibitor sildenafil stimulates angiogenesis through a protein kinase G/MAPK pathway.J Cell Physiol. 2007; 211:197–204. doi: 10.1002/jcp.20929.CrossrefMedlineGoogle Scholar
  • 35. Malassiné A, Frendo JL, Evain-Brion D. A comparison of placental development and endocrine functions between the human and mouse model.Hum Reprod Update. 2003; 9:531–539.CrossrefMedlineGoogle Scholar
  • 36. Montagnana M, Danese E, Lippi G, Fava C. Blood laboratory testing for early prediction of preeclampsia: chasing the finish line or at the starting blocks?Ann Med. 2017; 49:240–253. doi: 10.1080/07853890.2016.1255350.CrossrefMedlineGoogle Scholar
  • 37. Nichols DJ, Muirhead GJ, Harness JA. Pharmacokinetics of sildenafil after single oral doses in healthy male subjects: absolute bioavailability, food effects and dose proportionality.Br J Clin Pharmacol. 2002; 53(suppl 1):5S–12S.CrossrefMedlineGoogle Scholar
  • 38. Shin HS, Bae SK, Lee MG. Pharmacokinetics of sildenafil after intravenous and oral administration in rats: hepatic and intestinal first-pass effects.Int J Pharm. 2006; 320:64–70. doi: 10.1016/j.ijpharm.2006.04.005.CrossrefMedlineGoogle Scholar
  • 39. Walker DK, Ackland MJ, James GC, Muirhead GJ, Rance DJ, Wastall P, Wright PA. Pharmacokinetics and metabolism of sildenafil in mouse, rat, rabbit, dog and man.Xenobiotica. 1999; 29:297–310. doi: 10.1080/004982599238687.CrossrefMedlineGoogle Scholar
  • 40. Sakamoto M, Osato K, Kubo M, Nii M, Tanaka H, Murabayashi N, Umekawa T, Kamimoto Y, Ikeda T. Early-onset fetal growth restriction treated with the long-acting phosphodiesterase-5 inhibitor tadalafil: a case report.J Med Case Rep. 2016; 10:317. doi: 10.1186/s13256-016-1098-x.CrossrefMedlineGoogle Scholar
  • 41. Downing JW, Ramasubramanian R, Johnson RF, Minzter BH, Paschall RL, Sundell HW, Engelhardt B, Lewis R. Hypothesis: selective phosphodiesterase-5 inhibition improves outcome in preeclampsia.Med Hypotheses. 2004; 63:1057–1064. doi: 10.1016/j.mehy.2004.03.042.CrossrefMedlineGoogle Scholar
  • 42. Papazova DA, Oosterhuis NR, Gremmels H, van Koppen A, Joles JA, Verhaar MC. Cell-based therapies for experimental chronic kidney disease: a systematic review and meta-analysis.Dis Model Mech. 2015; 8:281–293. doi: 10.1242/dmm.017699.CrossrefMedlineGoogle Scholar
  • 43. Cushen SC, Goulopoulou S. New models of pregnancy-associated hypertension.Am J Hypertens. 2017. doi: 10.1093/ajh/hpx063CrossrefMedlineGoogle Scholar
  • 44. LaMarca B, Amaral LM, Harmon AC, Cornelius DC, Faulkner JL, Cunningham MW. Placental ischemia and resultant phenotype in animal models of preeclampsia.Curr Hypertens Rep. 2016; 18:38. doi: 10.1007/s11906-016-0633-x.CrossrefMedlineGoogle Scholar
  • 45. Sibley CP. Treating the dysfunctional placenta.J Endocrinol. 2017; 234:R81–R97. doi: 10.1530/JOE-17-0185.CrossrefMedlineGoogle Scholar
  • 46. Ozdegirmenci O, Kucukozkan T, Akdag E, Topal T, Haberal A, Kayir H, Oter S, Akyol M, Uzbay T. Effects of sildenafil and tadalafil on ischemia/reperfusion injury in fetal rat brain.J Matern Fetal Neonatal Med. 2011; 24:317–323. doi: 10.3109/14767058.2010.492061.CrossrefMedlineGoogle Scholar
  • 47. Dilworth MR, Andersson I, Renshall LJ, Cowley E, Baker P, Greenwood S, Sibley CP, Wareing M. Sildenafil citrate increases fetal weight in a mouse model of fetal growth restriction with a normal vascular phenotype.PLoS One. 2013; 8:e77748. doi: 10.1371/journal.pone.0077748.CrossrefMedlineGoogle Scholar
  • 48. Lopez-Tello J, Arias-Alvarez M, Jimenez-Martinez MA, Barbero-Fernandez A, Garcia-Garcia RM, Rodriguez M, Lorenzo PL, Torres-Rovira L, Astiz S, Gonzalez-Bulnes A, Rebollar PG. The effects of sildenafil citrate on feto-placental development and haemodynamics in a rabbit model of intrauterine growth restriction [published online ahead of print May 23, 2016].Reprod Fertil Dev. doi: 10.1071/RD15330. http://www.publish.csiro.au/rd/RD15330.Google Scholar
  • 49. Motta C, Grosso C, Zanuzzi C, Molinero D, Picco N, Bellingeri R, Alustiza F, Barbeito C, Vivas A, Romanini MC. Effect of sildenafil on pre-eclampsia-like mouse model induced by L-NAME.Reprod Domest Anim. 2015; 50:611–616. doi: 10.1111/rda.12536.CrossrefMedlineGoogle Scholar
  • 50. Nassar AH, Masrouha KZ, Itani H, Nader KA, Usta IM. Effects of sildenafil in Nω-nitro-L-arginine methyl ester-induced intrauterine growth restriction in a rat model.Am J Perinatol. 2012; 29:429–434. doi: 10.1055/s-0032-1304823.CrossrefMedlineGoogle Scholar
  • 51. Oyston C, Stanley JL, Oliver MH, Bloomfield FH, Baker PN. Maternal administration of sildenafil citrate alters fetal and placental growth and fetal- placental vascular resistance in the growth- restricted ovine fetus.Hypertension. 2016; 68:760–767. doi: 10.1161/HYPERTENSIONAHA.116.07662.LinkGoogle Scholar
  • 52. Refuerzo JS, Sokol RJ, Aranda JV, Hallak M, Hotra JW, Kruger M, Sorokin Y. Sildenafil citrate and fetal outcome in pregnant rats.Fetal Diagn Ther. 2006; 21:259–263. doi: 10.1159/000091352.CrossrefMedlineGoogle Scholar
  • 53. Sánchez-Aparicio P, Mota-Rojas D, Nava-Ocampo AA, Trujillo-Ortega ME, Alfaro-Rodríguez A, Arch E, Alonso-Spilsbury M. Effects of sildenafil on the fetal growth of guinea pigs and their ability to survive induced intrapartum asphyxia.Am J Obstet Gynecol. 2008; 198:127.e1–e127.e6. doi: 10.1016/j.ajog.2007.06.068.CrossrefGoogle Scholar
  • 54. Satterfield MC, Bazer FW, Spencer TE, Wu G. Sildenafil citrate treatment enhances amino acid availability in the conceptus and fetal growth in an ovine model of intrauterine growth restriction.J Nutr. 2010; 140:251–258. doi: 10.3945/jn.109.114678.CrossrefMedlineGoogle Scholar
  • 55. George EM, Palei AC, Dent EA, Granger JP. Sildenafil attenuates placental ischemia-induced hypertension.Am J Physiol Regul Integr Comp Physiol. 2013; 305:R397–R403. doi: 10.1152/ajpregu.00216.2013.CrossrefMedlineGoogle Scholar
  • 56. Gillis EE, Mooney JN, Garrett MR, Granger JP, Sasser JM. Sildenafil treatment ameliorates the maternal syndrome of preeclampsia and rescues fetal growth in the dahl salt-sensitive rat.Hypertension. 2016; 67:647–653. doi: 10.1161/HYPERTENSIONAHA.115.06071.LinkGoogle Scholar
  • 57. Herraiz S, Pellicer B, Serra V, Cauli O, Cortijo J, Felipo V, Pellicer A. Sildenafil citrate improves perinatal outcome in fetuses from pre-eclamptic rats.BJOG. 2012; 119:1394–1402. doi: 10.1111/j.1471-0528.2012.03430.x.CrossrefMedlineGoogle Scholar
  • 58. Pellicer B, Herraiz S, Cauli O, Rodrigo R, Asensi M, Cortijo J, Serra V, Morcillo E, Felipo V, Simón C, Pellicer A. Haemodynamic effects of long-term administration of sildenafil in normotensive pregnant and non-pregnant rats.BJOG. 2011; 118:615–623. doi: 10.1111/j.1471-0528.2010.02839.x.CrossrefMedlineGoogle Scholar
  • 59. Ramesar SV, Mackraj I, Gathiram P, Moodley J. Sildenafil citrate improves fetal outcomes in pregnant, L-NAME treated, Sprague-Dawley rats.Eur J Obstet Gynecol Reprod Biol. 2010; 149:22–26. doi: 10.1016/j.ejogrb.2009.11.005.CrossrefMedlineGoogle Scholar
  • 60. Roberts RP, Refuerzo JS, Ferrari F, Ontiveros AE, Tamayo EH, Sibai BM, Blackwell SC, Longo M. Sildenafil treatment in a nonsevere hypertensive murine model lowers blood pressure without reducing fetal growth.Am J Obstet Gynecol. 2016; 215:386.e1–386.e8. doi: 10.1016/j.ajog.2016.05.002.CrossrefGoogle Scholar
  • 61. Stanley JL, Andersson IJ, Poudel R, Rueda-Clausen CF, Sibley CP, Davidge ST, Baker PN. Sildenafil citrate rescues fetal growth in the catechol-O-methyl transferase knockout mouse model.Hypertension. 2012; 59:1021–1028. doi: 10.1161/HYPERTENSIONAHA.111.186270.LinkGoogle Scholar
  • 62. Turgut NH, Temiz TK, Bagcivan I, Turgut B, Gulturk S, Karadas B. The effect of sildenafil on the altered thoracic aorta smooth muscle responses in rat pre-eclampsia model.Eur J Pharmacol. 2008; 589:180–187. doi: 10.1016/j.ejphar.2008.04.034.CrossrefMedlineGoogle Scholar
  • 63. Cauli O, Herraiz S, Pellicer B, Pellicer A, Felipo V. Treatment with sildenafil prevents impairment of learning in rats born to pre-eclamptic mothers.Neuroscience. 2010; 171:506–512. doi: 10.1016/j.neuroscience.2010.08.065.CrossrefMedlineGoogle Scholar

Novelty and Significance

What Is New?

  • This is the first meta-analysis providing a complete overview of studies on the effect of sildenafil on fetal growth and maternal blood pressure.

  • By combining clinical and preclinical studies, our analysis contained a large variation in study design that allows identification of factors affecting the efficacy of sildenafil.

What Is Relevant?

  • We confirm that sildenafil improves fetal growth and reduces maternal blood pressure.

  • Optimal fetal growth is reached at a dose that is about an order of magnitude higher than the daily dose currently used in human studies.

  • Sildenafil does not lower blood pressure in the absence of hypertension.

Summary

This meta-analysis supports that sildenafil improves fetal growth and maternal blood pressure regulation during pregnancy and might help guide the design of (pre) clinical studies.