OBJECTIVE: To estimate the effects of pravastatin on the altered vascular function in a mouse model of preeclampsia induced by overexpression of soluble fms-like tyrosine kinase-1 (sFlt-1).
METHODS: Pregnant CD1 mice, at day 8 of gestation, were randomly allocated to injection using the tail vein of the adenovirus carrying sFlt-1 (109 plaque-forming units in 100 microliters; sFlt-1 group) or mFc (109 plaque-forming units) as virus control, and then to receive pravastatin (Pra; 5 mg/kg/d) dissolved in drinking water or control. The mice in four groups (sFlt-1, sFlt-1-pravastatin, mFc, and mFc-pravastatin; n=4–6 per group) were killed at day 18 of gestation and 2-mm segments of carotid artery were used for vascular reactivity studies. Serum sFlt-1 levels were also measured by enzyme-linked immunosorbent assay.
RESULTS: Mice in the sFlt-1 group had the highest responses to phenylephrine. Treatment with pravastatin decreased the contractile responses to phenylephrine (maximal effect [mean±standard error of the mean] 137.35± 27.70 compared with 42.24±8.76; P=.006) for sFlt-1 compared with sFlt-1-pravastatin, respectively. There were no differences in the contractile responses to thromboxane A2. The vasorelaxant responses to acetylcholine were significantly highest in the mFc-pravastatin group, with a maximal effect of 108.37±5.25 compared with 89.77±3.96 in the mFc group (P=.008), and those with sodium nitroprusside were not different across the four groups. Serum sFlt-1 levels were not different at baseline (day 8) but were significantly lower in sFlt-1-pravastatin compared with sFlt-1 at day 18 (59.42±5.31 compared with 102.59±15.15 ng/mL; P=.01).
CONCLUSION: Pravastatin improved the vascular reactivity in this murine model of preeclampsia by decreasing sFlt-1 levels. Statins should be evaluated for the prevention of the vascular abnormalities of preeclampsia.
Administration of pravastatin in a mouse model of preeclampsia ameliorates the vascular reactivity by decreasing the production of the antiangiogenic soluble fms-like tyrosine kinase-1.
From the Department of Obstetrics and Gynecology, University of Texas Medical Branch, Galveston, Texas.
Supported by the Department of Obstetrics & Gynecology, University of Texas Medical Branch, Galveston, Texas.
Presented in Plenary II Session at the 30th Annual Meeting of the Society of Maternal-Fetal-Medicine, February 1–6, 2010, Chicago, Illinois.
Corresponding author: Maged M. Costantine, MD, Department of Obstetrics and Gynecology, University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-0587; e-mail email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
Preeclampsia remains one of the major causes of maternal and neonatal morbidities. It complicates 5–8% of pregnancies, and the clinical manifestations that lead to its diagnosis appear only in the later stages of the disease.1 It is currently thought that preeclampsia is secondary to early pregnancy imbalances in proangiogenic and anti-angiogenic factors such as vascular endothelium growth factor (VEGF), placental growth factor, soluble fms-like tyrosine kinase (sFlt-1), and soluble endoglin.2,3 This is associated with defective trophoblast invasion and remodeling of the spiral arterioles4 leading to a hypoxic placenta, excessive release of vasoactive factors and cytokines, and endothelial dysfunction, which then triggers the clinical stage of the maternal syndrome characterized by hypertension, proteinuria, and other manifestations of end-organ damage.4
Although preeclampsia is unique to a pregnant state, it shares biologic and pathologic plausibility with adult cardiovascular diseases, as well as many risk factors.5 Endothelial dysfunction and inflammation are fundamental for the initiation and progression of both.4,6 Whereas attempts for primary and secondary prevention of preeclampsia have failed, except for a small benefit from low-dose aspirin,7 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors, or statins, have been shown to be essential and successful in primary and secondary prevention of cardiovascular mortality and other cardiovascular morbidities through pleiotropic and lipid-lowering actions.8,9
Thus, our objective in this study was to estimate the effects of a hydrophilic 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitor, pravastatin, on the altered vascular function in preeclampsia using a well-characterized animal model of preeclampsia induced by overexpression of sFlt-1.10,11
MATERIALS AND METHODS
Pregnant CD1 mice were obtained from Charles River (Wilmington, MA). The study protocol and all procedures were approved by the institutional Animal Care and Use Committee of the University of Texas Medical Branch, Galveston, Texas. The mice were housed separately in temperature- and humidity-controlled quarters with constant light/dark cycles of 12 hours/12 hours. They were provided with food and water ad libitum. Regular maintenance and care were provided by certified personnel and veterinary staff according to the guidelines of the Animal Care and Use Committee. Surgical procedures were performed according to the Animal Care and Use Committee guidelines under anesthesia with ketamine (Ketalar; Parke-Davis, Morris Plains, NJ) and xylazine (Gemini; Rugby, Rockville Center, NY). The animals were killed at day 18 of gestation by carbon dioxide (CO2) inhalation in accordance with the Animal Care and Use Committee and the American Veterinary Medical Association guidelines.
The preparation of adenovirus carrying sFlt-1 and mFc fragments, as well as the generation and validation of the preeclampsia model, has been described in detail elsewhere.10,11 In brief, we used a replication-deficient recombinant adenovirus vector that leads, after a single injection, to hepatocyte transduction, which allows sustained in vivo transgene expression by using the cellular machinery and secretion into the systemic circulation.12 Pregnant CD1 mice at day 8 of gestation were randomly allocated to injection via the tail vein with either adenovirus carrying sFlt-1 (109 plaque-forming units in 100 microliters; sFlt-1 group) or adenovirus carrying the murine immunoglobulin G2α Fc fragment (109 plaque-forming units in 100 microliters; mFc virus control group). We, as well as others, have previously shown that transfection of pregnant mice with adenovirus carrying sFlt-1 leads to an increase in circulating levels of sFlt-1, hypertension, and a preeclampsia-like picture.10,11
From the next day (day 9) and until the animals were killed, mice from both groups were randomly assigned to receive pravastatin (Sigma-Aldrich, St Louis, MO) in their drinking water (sFlt-1-pravastatin and mFc-pravastatin). The control groups (sFlt-1 and mFc) received water only. Pravastatin was dissolved in water at a concentration that gave a dose of 5 mg/kg/d based on the daily water consumption of pregnant mice, which was predetermined in a preliminary study and as previously described.13 There was no difference in the drinking behavior or amount of daily water intake across the four groups.
After animals were killed at day 18 of gestation, 2-mm segments of the right carotid arteries were dissected and mounted on a wire myograph (Model 610M; J.P. Trading I/S, Aarhus, Denmark) over 25-micron tungsten wires. The preparations were bathed in Krebs solution and maintained at 37°C, pH of approximately 7.4. Mixture of 95% oxygen and 5% carbon dioxide was bubbled continuously through the solution. The force was continuously recorded using an isometric force transducer and analyzed using PowerLab system and Chart 5 data acquisition and playback software (AD Instruments, Castle Hill, Australia). After stabilization of the tone, the vessels were contracted twice with 60 mmol/L potassium chloride (KCl) for 7 minutes to enhance reproducibility of responses. The second response to KCl was used as reference in the calculations of the responses to the contractile agents. To ensure endothelial function, the response to a single concentration of acetylcholine (ACh; 10−6 mmol/L) in vessels precontracted with phenylephrine (10−6 to 3×10−6 mmol/L) was determined. Only arteries demonstrating a substantial response to ACh (more than 70–80% of relaxation) and constriction to high KCl were used for experimental studies.
After 1-hour equilibration in Krebs solution, relaxant responses to cumulative concentrations of the endothelium-dependent vasorelaxant ACh (10−10 to 10−5 mmol/L) and the endothelium-independent vasorelaxant sodium-nitroprusside (10−10 to 10−5 mmol/L) were evaluated after precontraction of the vessels with phenylephrine (10−7 to 10−6 mmol/L) to produce matching contractions in the study groups. In addition, contractile responses to cumulative concentrations of the α1-adrenergic agonist phenylephrine (10−10 to 10−5 mmol/L), in the presence and absence of the nonselective nitric oxide synthase inhibitor Nω-nitro-L-arginine methyl ester (l-NAME, 10−4 mmol/L), and contractile responses to thromboxane A2 (TXA2; 10−10 to 10−5 mmol/L) were recorded.
Blood (100 microliters) was obtained from the CD1 mice at baseline (day 8), at midpregnancy (day 14), and at killing (day 18). It was centrifuged at 3,000 rpm for 10 minutes, and serum was collected and stored at −80°C until time of testing. The sFlt-1 sera levels were measured using mouse soluble VEGF R1 enzyme-linked immunosorbent assay per the manufacturer's guidelines (R&D Systems, Minneapolis, MN). All assays were run in duplicate.
Potassium chloride, acetylcholine hydrochloride, phenylephrine hydrochloride, sodium-nitroprusside, and l-NAME were obtained from Sigma-Aldrich (St. Louis, MO) and TXA2 from Cayman Chemical (Ann Arbor, MI). The composition of Krebs solution was NaCl 119 mmol/L, KCl 4.7 mmol/L, KH2PO4 1.2 mmol/L, NaHCO3 25 mmol/L, MgCl2 1.2 mmol/L, CaCl2 2.5 mmol/L, ethylenediaminetetraacetic acid 0.026 mmol/L, glucose 11 mmol/L. Drugs were diluted with the Krebs solution.
For the vascular reactivity studies, the response to the second KCl dose was used as a reference to calculate the percent contraction achieved by phenylephrine (plus or minus l-NAME) and TXA2. However, phenylephrine precontraction was used to measure percent relaxation induced by ACh and sodium-nitroprusside. The area under the concentration response curves (AUC), logarithm of the concentration producing 50% of the maximal effect, and the maximal effect were calculated. The sFlt-1 sera levels were calculated using a standard curve derived from known concentrations of the recombinant protein.
Analysis was performed using SigmaPlot (Systat 11.0, Chicago, IL) and Prism 4 (GraphPad Software Inc, LA Jolla, CA). Data are reported as mean plus or minus standard error of the mean or median with interquartile range when specified. The Shapiro-Wilk test was used to assess for normality; if passed and with equal variance, one-way analysis of variance with post hoc Holm-Sidak method was used; otherwise we used Kruskal-Wallis analysis with post hoc Dunn test. Holm-Sidak14 is a step-down test for pairwise comparisons that sets the critical level according to Sidak correction of Bonferroni inequality. A two-tailed P<.05 was considered statistically significant.
Maternal weights at baseline (day 8: 30.81±1.11 g, 29.77±1.12 g, 29.77±1.39 g, 29.33±0.72 g; P=.78), at killing (day 18: 54.18±1.75 g, 51.11±1.55 g, 49.33±3.37 g, 49.92±3.21; P=.39), average number of pups (median [interquartile range ]: 12 [12–14], 12 [11–13], 11 [10–12], 13 [10–14]; P=.66) and average total pup weight (18.45±1.60 g, 17.75±1.39 g, 15.53±1.38 g, 15.69±1.96 g; P = .49) were not significantly different across the four groups of mice (mFc, sFlt-1, mFc-pravastatin, and sFlt-1-pravastatin, respectively).
Data from vascular reactivity experiments for maximal effect and AUC are summarized in Table 1. Mice in the sFlt-1 group had the highest responses to phenylephrine (Fig. 1A). Treatment with pravastatin decreased the contractile responses to phenylephrine in sFlt-1-pravastatin compared with the sFlt-1 group (maximal effect P=.006, AUC P=.04, logarithm of the concentration producing 50% of the maximal effect P=.02). Although the differences in the response to phenylephrine became statistically nonsignificant with the addition of l-NAME (Fig. 1B), the sFlt-1-pravastatin group still had lower responses compared with sFlt-1. There were no differences in the contractile responses to TXA2, although treatment with pravastatin decreased the contractile responses in the sFlt-1 group, but the difference was not statistically significant (P=.06).
The vasorelaxant responses to ACh were significantly highest in the mFc-pravastatin group, with maximal effect of 108.37±5.25 compared with 89.77±3.96 in the mFc group; P=.008 (Fig. 1C). There were no differences in the vasorelaxant responses to sodium-nitroprusside across all four groups (Fig. 1D).
Serum sFlt-1 levels were not different at baseline (day 8). At midgestation, sFlt-1 levels were significantly higher in the sFlt-1 group compared with the mFc group (89.69±21.01 compared with 29.79±0.83 ng/mL, P=.009) and also higher, although not statistically significant, when compared with the sFlt-1-pravastatin group (70.99±8.78 ng/mL, P=.38). Treatment with pravastatin significantly lowered the sFlt-1 serum levels at day 18 in the sFlt-1-pravastatin group compared with sFlt-1 (59.42±5.31 compared with 102.59±15.15 ng/mL, P=.01), and in the mFc-pravastatin mice compared with mFc, but the latter difference did not reach statistical significance (50.30±5.31 compared with 72.29±8.18 ng/mL, P=.053) (Fig. 2).
We found that treatment of pregnant mice destined to develop preeclampsia with pravastatin improves their vascular function as assessed by in vitro carotid artery vascular reactivity experiments. The mechanism behind this improvement appears to be endothelium dependent and involves reduction in the serum sFlt-1 level.
Pravastatin is a competitive inhibitor of 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase, which is the enzyme catalyzing the early rate-limiting step in cholesterol biosynthesis, the conversion of 3-hydroxy-3-methyl-glutaryl-coenzyme A to mevalonate.15 It is a polar hydrophilic compound (the most hydrophilic among the current statins) and thus does not cross the placenta.15 Its efficacy in preventing cardiovascular morbidity and mortality was proven in multiple studies,16,17 with some showing benefit across a wide range of cholesterol levels.16,17 This suggests a role that bypasses low-density lipoprotein cholesterol lowering to other pleiotropic effects such as anti-inflammatory, up-regulation of endothelial nitric oxide synthase, inhibition of smooth muscle cells proliferation, decrease in oxygen free radical formation as well as immunomodulatory effects.18
Statins are FDA pregnancy category X. This indicates that “studies in animals or human beings have demonstrated fetal abnormalities or there is evidence of fetal risk based on human experience or both, and the risk of the use of the drug in pregnant women clearly outweighs any possible benefit.”19 This classification was based on the fact that there are no indications that warrant use of a statin in pregnancy (no benefit to outweigh any risk); atherosclerosis is a long-standing disease, and thus discontinuation during pregnancy may not influence the long-term course of the disease. It was also based on theoretical concern of potential teratogenicity secondary to inhibition of cholesterol synthesis in pregnancy, in addition to small case series showing some teratogenic effects. Animal and human data have provided mixed results;20,21 however, the more recent epidemiologic studies do not suggest that statins are a major teratogen.20 Moreover, data on teratogenicity may be specific to particular statins. As opposed to the more lipophilic compounds such as simvastatin or lovastatin, pravastatin is the most hydrophilic statin, is the least potent sterol synthesis inhibitor, and has not been associated or linked with teratogenic effects or other adverse pregnancy outcomes such as growth restriction or miscarriages.22,23
Our study has some limitations. Our results apply only to this animal model of preeclampsia, which is based on the angiogenic imbalances early in pregnancy. Other models have been suggested and include surgical, pharmacologic, or genetic manipulations, such as reduction of uteroplacental perfusion by uterine artery ligation, alteration of vascular function by impairment in the nitric oxide system by knocking out the endothelial nitric oxide synthase gene, chronic blockade of the endothelial-derived relaxing factor, or use of a nitric oxide synthase inhibitor such as l-NAME. Most animal models lack some features of preeclampsia in humans, thus making extrapolation or application of our results clinically even more challenging.
Our initial guess for potential mechanisms to explain our findings of improved vascular reactivity with the use of pravastatin was the ability of statins to up-regulate endothelial nitric oxide synthase.24 However, our results show that this may not be the sole mechanism. With the addition of l-NAME, a nonselective inhibitor of nitric oxide synthase, the responses to phenylephrine in the sFlt-1-pravastatin group were still much lower compared with those of the sFlt-1 animals not treated with pravastatin, although the difference did not reach statistical significance. Moreover, a nitric oxide pathway may not explain the observed reduction in the sFlt-1 serum levels observed in our study.
In vivo and in vitro studies show that statins induce, in a concentration- and time-dependent manner, the transcription and expression of heme oxygenase-1 in endothelial24 and vascular smooth muscle26 cells. Heme oxygenase-1 is an inducible antioxidant defense protein with cytoprotective and anti-inflammatory properties.26,27 It is essential for heme oxidation to biliverdin in a three-step process that liberates carbon monoxide (CO) and iron. Patients with preeclampsia have significantly decreased CO concentrations in their exhaled breath, signifying either decreased levels or decreased activity of heme oxygenase-1.28 They also have elevated serum levels of sFlt-1 and soluble endoglin.2,3 Recently, activation/up-regulation of the heme oxygenase-1/CO pathway with simvastatin was found to inhibit sFlt-1 and soluble endoglin release from endothelial cells and placental explants.29 However, statins have been shown to stimulate VEGF transcription and expression through up-regulation of hypoxia-inducible factor-1α in endothelial cells.30,31 Whether these described mechanisms apply to our study, and thus can be used to explain our results, is speculative at this point and requires more investigation and verification. Other potential mechanisms that were not evaluated in this study involve the pleiotropic effects of statin on free oxygen radical formation and cytokine production, as well as immunomodulatory and other anti-inflammatory effects.
In conclusion, we found that pravastatin improved the vascular reactivity in this murine model of preeclampsia by decreasing sFlt-1 levels. Statins should be evaluated for prevention of the vascular abnormalities of preeclampsia. We also believe that pregnancy should be considered the window to a woman's future health, and those women with constitutionally determined and unfavorable vascular and metabolic profiles need to be identified and treated in an effort to decrease their future risk of cardiovascular disease.
1. Diagnosis and management of preeclampsia and eclampsia. ACOG Practice Bulletin No. 33. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99:159–67
2. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672–83.
3. Levine RJ, Lam C, Qian C, Yu KF, Maynard SE, Sachs BP, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006;355:992–1005.
4. Redman CW, Sargent IL. Latest advances in understanding preeclampsia. Science 2005;308:1592–4.
5. Berends AL, de Groot CJ, Sijbrands EJ, Sie MP, Benneheij SH, Pal R, et al. Shared constitutional risks for maternal vascular-related pregnancy complications and future cardiovascular disease. Hypertension 2008;51:1–8.
6. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005;352:1685–95.
7. Barton JR, Sibai BM. Prediction and prevention of recurrent preeclampsia. Obstet Gynecol 2008;112(2 pt 1):359–72.
8. Brugts JJ, Yetgin T, Hoeks SE, Gotto AM, Shepherd J, Westendorp RG, et al. The benefits of statins in people without established cardiovascular disease but with cardiovascular risk factors: meta-analysis of randomised controlled trials. BMJ 2009;338:b2376.
9. Mills EJ, Rachlis B, Wu P, Devereaux PJ, Arora P, Perri D. Primary prevention of cardiovascular mortality and events with statin treatments: a network meta-analysis involving more than 65,000 patients. J Am Coll Cardiol 2008;52:1769–81.
10. Lu F, Longo M, Tamayo E, Maner W, Al-Hendy A, Anderson G, et al. The effect of over-expression of sFlt-1 on blood pressure and the occurrence of other manifestations of preeclampsia in unrestrained conscious pregnant mice. Am J Obstet Gynecol 2007;196:396.e1–7.
11. Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649–58.
12. Wei K, Kuhnert F, Kuo CJ. Recombinant adenovirus as a methodology for exploration of physiologic functions of growth factor pathway. J Mol Med 2008;86:161–9.
13. Elahi MM, Cagampang FR, Anthony FW, Curzen N, Ohri SK, Hanson MA. Statin treatment in hypercholesterolemic pregnant mice reduced cardiovascular risk factors in their offspring. Hypertension 2008;51:939–44.
14. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979;6:65–70.
15. Hatanaka T. Clinical pharmacokinetics of pravastatin: mechanisms of pharmacokinetic events. Clin Pharmacokinet 2000;39:397–412.
16. Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, et al. Prevention of coronary heart disease with Pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995;333:1301–7.
17. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. N Engl J Med 1998;339:1349–57.
18. Shaw SM, Fildes JE, Yonan N, Williams SG. Pleiotropic effects and cholesterol-lowering therapy. Cardiology 2009;112:4–12.
19. U.S. Food and Drug Administration. Drug bulletin. Fed Reg 1980;44:37434–67.
20. Kazmin A, Garcia-Bournissen F, Koren G. Risks of statin use during pregnancy: a systematic review. J Obstet Gynaecol Can 2007;29:906–8.
21. Taguchi N, Rubin ET, Hosokawa A, Choi J, Ying AY, Moretti ME, et al. Prenatal exposure to HMG-CoA reductase inhibitors: effects on fetal and neonatal outcomes. Reprod Toxicol 2008;26:175–7.
22. Díaz-Zagoya JC, Asenjo-Barrón JC, Cárdenas-Vázquez R, Martínez F, Juárez-Oropeza MA. Comparative toxicity of high doses of vastatins currently used by clinicians in CD-1 male mice fed with a hypercholesterolemic diet. Life Sci 1999;65:947–56.
23. Edison RJ, Muenke M. Mechanistic and epidemiologic considerations in the evaluation of adverse birth outcomes follow-ing gestational exposure to statins. Am J Med Genet A 2004;131:287–98.
24. Endres M, Laufs U, Huang Z, Nakamura T, Huang P, Moskowitz MA, et al. Stroke protection by 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors mediated by endothelial nitric oxide synthase. Proc Natl Acad Sci U S A 1998;95:8880–5.
25. Grosser N, Hemmerle A, Berndt G, Erdmann K, Hinkelmann U, Schürger S, et al. The antioxidant defense protein heme oxygenase 1 is a novel target for statins in endothelial cells. Free Radic Biol Med 2004;37:2064–71.
26. Lee TS, Chang CC, Zhu Y, Shyy JY. Simvastatin induces heme oxygenase-1: a novel mechanism of vessel protection. Circulation 2004;110:1296–302.
27. Duckers HJ, Boehm M, True AL, Yet SF, San H, Park JL, et al. Heme oxygenase-1 protects against vascular constriction and proliferation. Nat Med 2001;7:693–8.
28. Kreiser D, Baum M, Seidman DS, Fanaroff A, Shah D, Hendler I, et al. End tidal carbon monoxide levels are lower in women with gestational hypertension and pre-eclampsia. J Perinatol 2004;24:213–217.
29. Cudmore M, Ahmad S, Al-Ani B, Fujisawa T, Coxall H, Chudasama K, et al. Negative regulation of soluble Flt-1 and soluble endoglin release by heme oxygenase-1. Circulation 2007;115:1789–97.
30. Chen SD, Hu CJ, Yang DI, Nassief A, Chen H, Yin K, et al. Pravastatin attenuates ceramide-induced cytotoxicity in mouse cerebral endothelial cells with HIF-1 activation and VEGF upregulation. Ann N Y Acad Sci 2005;1042:357–64.
31. Nishimoto-Hazuku A, Hirase T, Ide N, Ikeda Y, Node K. Simvastatin stimulates vascular endothelial growth factor production by hypoxia-inducible factor-1alpha upregulation in endothelial cells. J Cardiovasc Pharmacol 2008;51:267–73.
© 2010 by The American College of Obstetricians and Gynecologists.
Figure. No caption available.