If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Anesthesia, Mount Sinai Hospital, University of Toronto, Toronto ONDepartment of Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto ON
Department of Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto ONThe Lunenfeld-Tannenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto ON
Department of Obstetrics and Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto ONThe Lunenfeld-Tannenbaum Research Institute, Mount Sinai Hospital, University of Toronto, Toronto ON
Early identification of women at risk of developing early-onset severe preeclampsia (sPE) is a key objective in obstetrics. An elevated ratio of serum soluble fms-like tyrosine kinase (sFlt-1) to placenta-like growth factor (PlGF) (sFlt-1/PlGF ratio) precedes overt hypertension. The longitudinal relationship between this biomarker, maternal hemodynamics, and maternal serum uric acid during the pre-clinical phase is unknown.
Study Design
We followed 20 normotensive women at high risk of developing sPE from 20 weeks until delivery or 34 weeks’ gestation. Non-invasive hemodynamic monitoring using bioreactance technology was performed at 20 to 22, 24 to 26, 28 to 30, and 32 to 34 weeks’ gestation. Serum uric acid, sFlt-1, and PlGF were measured simultaneously.
Results
Six of 20 women (30%) delivered before 33 weeks with sPE and had significantly higher mean total peripheral resistance (TPR), higher serum uric acid, and higher sFlt-1/PlGF ratios at 24 weeks’ gestation than unaffected individuals. The area under the curve, cut-off values, and sensitivity and specificity to predict sPE at 24 weeks were as follows: TPR 0.84, 1250 dyne.s.cm-5, 80%, 93%; sFlt-1/PlGF ratio 0.94, 55, 100%, 93%; and serum uric acid 0.99, 255 μmol/L, 100%, 93%. TPR and sFlt-1 were positively correlated in the sPE group before antihypertensive treatment (r=0.65, P=0.01). Serum uric acid correlated with both sFlt-1 (r=0.65, P=0.003) and sFlt-1/PlGF ratio (r=0.54, P=0.02).
Conclusion
A combination of non-invasive determination of TPR together with measurement of serum uric acid may identify a subset of clinically high-risk women with evolving sPE, independent of the determination of the sFlt-1/PlGF ratio. The predictive ability of this integrated approach needs to be assessed in a larger cohort of women to further confirm its utility.
Résumé
Objectifs
L’identification précoce des femmes exposées à des risques d’en venir à présenter une prééclampsie grave d’apparition précoce (PEg) constitue un objectif clé en obstétrique. La constatation d’un rapport sFlt-1 (tyrosine kinase 1 de type fms soluble) - PlGF (facteur de croissance placentaire) (rapport sFlt-1/ PlGF) élevé précède celle d’une hypertension patente. La relation longitudinale entre ce marqueur biologique, l’hémodynamique maternelle et le taux sérique maternel d’acide urique au cours de la phase préclinique est inconnue.
Devis de l’étude
Nous avons suivi, de 20 semaines jusqu’à l’accouchement ou 34 semaines de gestation, 20 femmes normotendues exposées à un risque élevé d’en venir à présenter une PEg. Un monitorage hémodynamique non effractif (faisant appel à la technologie de la bioréactance) à été mené à 20-22, à 24-26, à 28-30 et à 32-34 semaines de gestation. Les taux sériques d’acide urique, de sFlt-1 et de PlGF ont été mesurés simultanément.
Résultats
Six des 20 femmes (30 %) ont accouché avant 33 semaines et présentaient une PEg; elles présentaient une résistance vasculaire périphérique totale (RVPT) moyenne considérablement plus élevée, un taux sérique d’acide urique accru et des rapports sFlt-1/PlGF plus élevés à 24 semaines de gestation que les femmes non affectées. La surface sous la courbe, les valeurs seuils et la sensibilité / spécificité pour ce qui est de la prévision de la PEg à 24 semaines étaient les suivantes : RVPT=0,84, 1 250 dyne.s.cm-5, 80 %, 93 %; rapport sFlt-1/PlGF=0,94, 55, 100 %, 93 %; et taux sérique d’acide urique=0,99, 255 μmol/l, 100 %, 93 %. La RVPT et la sFlt-1 présentaient une corrélation positive au sein du groupe PEg avant la mise en œuvre d’un traitement antihypertenseur (r=0,65, P=0,01). Le taux sérique d’acide urique présentait une corrélation tant avec la sFlt-1 (r=0,65, P=0,003) qu’avec le rapport sFlt-1/PlGF (r=0,54, P=0,02).
Conclusion
La combinaison « détermination non effractive de la RVPT-mesure du taux sérique d’acide urique » pourrait permettre l’identification d’un sous-ensemble de femmes qui sont exposées à un risque élevé sur le plan clinique et qui présentent une PEg évolutive, sans devoir avoir recours à la détermination du rapport sFlt-1/PlGF. La valeur prédictive de cette approche intégrée doit être évaluée auprès d’une cohorte de femmes de plus grande envergure afin d’en confirmer davantage l’utilité.
Risk factors for preeclampsia in healthy nulliparous women: a prospective multicenter study. The National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units.
It is defined as new onset hypertension (≥140/90 mmHg) after 20 weeks’ gestation with or without proteinuria (≥0.3 g in 24 hours), and may be considered severe in the presence of one or more of the following: blood pressure≥160/110 mmHg on two occasions six hours apart, proteinuria>5 g in 24 hours or 3+, oliguria<500 mL in 24 hours, cerebral or visual disturbances, cyanosis or pulmonary edema, epigastric or right upper quadrant pain, impaired liver function, thrombocytopenia, or fetal growth restriction.
and is characterized by high systemic vascular resistance and low cardiac output. This hemodynamic profile is typical in women with overt sPE and may be observed in the latent phase of disease.
Maternal serum uric acid is elevated in women with sPE, although its predictive value has been questioned in systematic reviews of the available evidence.
More recently, interest has focused on the detection of an imbalance in the ratio of placental-derived pro- and anti-angiogenic growth factors, in particular the performance of the ratio of maternal serum soluble fms-like tyrosine kinase and placenta-like growth factor. A relationship between this ratio and impaired maternal arterial vasodilatation has also been suggested.
We hypothesized that serial hemodynamic assessment of normotensive women at high risk of developing sPE, accompanied by measurement of the sFlt-1/PlGF ratio and uric acid in maternal blood, would be capable of providing greater understanding of the mechanisms involved in the development of overt sPE.
METHODS
Women were recruited to the study at 19 weeks’ gestation from our Maternal-Fetal Medicine Division Placenta Clinic. This clinic provides antenatal care to pregnant women at risk of developing preeclampsia or intrauterine growth restriction. Informed written consent was sought before women were enrolled. For inclusion in the study women were required to have a singleton pregnancy with a blood pressure<140/90 mmHg and normal fetal anatomical ultrasound (including normal fetal biometry and amniotic fluid). In addition, women were required to have the following characteristics:
1.
one or more medical risk factors or obstetric risk factors for preeclampsia or IUGR,
2.
elevated mean uterine artery Doppler pulsatility index, and
3.
two abnormal maternal integrated prenatal screening test values (either biochemical test values or abnormal placental morphology as defined by shape or texture).
Medical risk factors were factor V Leiden mutation; prothrombin gene mutation; elevated IgG anti-cardiolipin antibodies (>15 GPL) or a presence of lupus anticoagulant; elevated BMI (>30 kg/m2); advanced maternal age (≥40 years); and autoimmune conditions (systemic lupus erythematosus, mixed connective tissue disorder). Obstetric risk factors were previous stillbirth at>20 weeks’ gestation, extreme preterm delivery at<32 weeks’ gestation, major placental abruption, previous preeclampsia, or previous infant with birth weight<10th centile for gestational age. Elevated mean uterine artery Doppler pulsatility index was defined as a value above the 95th centile at 19 to 22 weeks’ gestation.
Abnormal maternal integrated prenatal screening test values were as follows:
1.
Biochemical: pregnancy-associated placental protein-A<0.35 MoM at 11 to 13 weeks, alpha-fetoprotein>2.0 MoM, dimeric inhibin assay>3.0 MoM or total human chorionic gonadotropin>4.0 MoM at 16 to 19 weeks, as currently used in our clinic;
2.
Morphological: placental shape was considered abnormal if maximum thickness was>4.0 cm or>50% of maximum length, maximum placental length was<10.0 cm, or placental texture was abnormal (i.e., placenta contained≥1 echogenic cystic lesions or a “jelly-like” appearance due to lack of normal villous development).
Exclusion criteria were multifetal pregnancy, any major fetal abnormality, uterine contractions, ruptured membranes, rescue cerclage for a cervix<2.0 cm long, BP>140/90 mmHg, chronic renal disease including nephrotic syndrome, and use of anti-hypertensive drugs or heparin.
Research visits were conducted on the same day as ambulatory appointments at the Placenta Clinic. Maternal demographics (age, weight [pre-pregnancy and current], height, gravidity and parity, history of smoking, alcohol and drug use, and current medications) were recorded upon recruitment.
Hemodynamic assessment was performed with non-invasive cardiac output monitoring using bioreactance technology (NICOM, Cheetah Medical, Vancouver, WA). Trans-thoracic bioreactance is a technique of non-invasive continuous cardiac output monitoring. It is based on an analysis of relative phase shifts of an alternating electrical current that occur when this current traverses the thoracic cavity. These phase shifts are tightly correlated with stroke volume.
Heart rate is obtained from the electrocardiogram reading, and cardiac output is calculated. Blood pressure is taken using an integrated non-invasive cuff, and total peripheral vascular resistance is derived mathematically from the above data.
Hemodynamic monitoring was performed with the patient semi-recumbent and slightly tilted to the left to avert aorto-caval compression. Four double NICOM electrodes were attached, two below the clavicle in the mid-clavicular line and two at the costal margin in the mid-clavicular line. A non-invasive blood pressure cuff was also placed on the patient’s upper arm to measure brachial artery pressure at 90-second intervals. Hemodynamic monitoring was commenced at this point and continued for 15 minutes. Baseline hemodynamic readings were obtained at 20 to 22 weeks’ gestation. Follow-up hemodynamic assessments were performed in three windows, at 24 to 26 weeks, 28 to 30 weeks, and 32 to 34 weeks, in tandem with clinically indicated visits (normally every 2 weeks). These visits were continued until 34 weeks or delivery, whichever came first.
Maternal blood samples were obtained at each research visit after the hemodynamic monitoring was completed. Samples were immediately sent for complete blood count and measurement of uric acid, urea, creatinine, serum electrolytes, and liver function tests as part of the routine Placenta Clinic maternal assessment. A further serum sample was stored at −800 °C for the ELISA measurement of PlGF and sFlt-1 (R&D Systems Inc., Minneapolis, MN).
All women had standard high-risk antenatal care provided by two high-risk staff obstetricians, initially with visits every two weeks that included comprehensive ultrasound examinations (estimated fetal weight, biophysical profile, and umbilical artery Doppler velocimetry). The need for increased maternal surveillance (more frequent visits or hospital admission) and antihypertensive therapy was guided by clinical assessment, blood pressure recordings, and trends in standard laboratory tests (platelet count, aspartate transaminase level, serum uric acid, and urinary protein/creatinine ratio). Antihypertensive treatment, normally using oral labetalol, was commenced when blood pressure reached≥150/100 mmHg. An increase in ultrasound surveillance during antihypertensive therapy sought evidence of reduced fetal growth accompanied by low amniotic fluid volume or abnormal umbilical artery Doppler waveforms. The additional hemodynamic data obtained with the NICOM did not influence clinical management. The sFlt-1/PlGF ratios were calculated in a batched manner following completion of the study.
Maternal and fetal outcomes were determined in a standard manner from hospital electronic records. Blood pressure at delivery was classified using the American College of Obstetricians and Gynecologists criteria for the diagnosis of preeclampsia.
Small for gestational age was defined as a birth weight below the 10th percentile. Intrauterine growth restriction was defined as a birth weight below the 3rd percentile for gestational age.
Serum samples for PlGF and s-Flt-1 assays were obtained at each hemodynamic assessment and stored at −800 °C for analysis. Analysis was performed in one batch by ELISA. The assay operator was blind to the clinical outcomes of the participants. The participants and physicians were unaware of the PlGF and sFlt-1 assay results.
As this was a pilot study, we recruited a convenience sample of 21 patients. The cohort was divided into three groups (sPE, IUGR, and normal) according to the clinical diagnosis. Continuous data were presented as medians (interquartile ranges) unless otherwise stated. Categorical data were presented as absolute values and percentages. Three group comparisons were conducted using the Kruskal-Wallis one-way analysis of variance. Two group comparisons were conducted using the Mann-Whitney U test. Proportions were compared using the chi-squared test (or Fisher exact test where appropriate). Hemodynamic and biomarker trends over time were displayed using line charts. Statistical analysis was not carried out on the serial data because of the small sample size. A receiver operating characteristic analysis was performed to determine the ability of total peripheral resistance and biomarkers measured at 24 weeks’ gestation to predict the development of later severe preeclampsia. Only TPR data obtained before antihypertensive therapy was used in the receiver operating characteristic analysis. We used SPSS v.20 (IBM Corp., Armonk NY) to conduct the analysis, and we accepted a P value of<0.05 as statistically significant.
Institutional research ethics board approval was obtained (Mount Sinai Hospital Research Ethics Board, REB number 10–0198-A).
RESULTS
Twenty-one women consented to participate in the study. One was subsequently excluded as she delivered before her first follow-up appointment. Maternal age at delivery, pre-pregnancy weight, height, BMI, gravidity, and parity in each group are shown in Table 1. No significant differences were noted across the groups. The results of integrated prenatal screening tests and placental ultrasound (placental size and shape and uterine artery Doppler studies) are summarized in Table 2. No significant differences were noted, although the mean uterine artery pulsatility index in the sPE group approached significance (P=0.05); the alternative method of interpreting uterine artery Doppler velocimetry (the presence or absence of an early diastolic notch in the waveform) showed no significant difference in proportion. Two women had been smokers but stopped in early pregnancy. All women denied alcohol or drug consumption during pregnancy. Twelve women had a complex obstetrical history, and four of these women had a history of preeclampsia in previous pregnancies. Six women had medical risk factors for preeclampsia, and four of these women developed sPE in the index pregnancy.
Table 1Patient demographics.
Data presented as medians (interquartile ranges)
Normal pregnancy n=9
Pregnancy with IUGR n=5
Pregnancy with severe preeclampsia n=6
P
Age at delivery, years
37 (33 to 38)
33 (32 to 42)
33 (31 to 39)
0.65
Pre-pregnancy weight, kg
64 (60 to 72)
80 (59 to 101)
63 (56 to 72)
0.65
Height, cm
160 (155 to 174)
165 (161 to 169)
161 (152 to 164)
0.44
BMI, kg/m2
25.6 (22.0 to 30.0)
28.5 (21.8 to 37.5)
25.8 (20.7 to 35.5)
0.74
Gravidity, n
4 (2 to 5)
2 (1 to 4)
3 (2 to 3)
0.26
Parity, n
1 (1 to 2)
0 (0 to 1)
1 (0 to 2)
0.17
Kruskal-Wallis one-way ANOVA was used to compare the medians
Six of the 20 women completing the study (30%) developed early onset sPE, five women (25%) developed IUGR without hypertension, and nine women (45%) remained normotensive and delivered healthy infants at term. In the sPE group, the median gestation at the onset of hypertension was 28+3 weeks.
There were significant differences between the groups in terms of gestation and birth weight. The median gestational ages at delivery in the normal and IUGR groups were 37 (36 to 39) weeks and 37 (36 to 37) weeks, respectively, while the sPE group delivered at a median 29 (27 to 31) weeks (P=0.002). The median birth weight in the normal group was 2720 g (2110 to 3008); in the IUGR group it was 1920 g (1740 to 2040), and in the sPE group 970 g (555 to 1295) (P=0.001). Of the six women who developed sPE, one delivered an infant whose birth weight was below the third centile, and three delivered infants whose weight was between the third and tenth centiles. Serial hemodynamic data collected for each variable are shown in Figure 1. The percentage changes from baseline over time for the three groups are shown in Figure 2.
Figure 1Serial hemodynamic data for each variable recorded
The hemodynamic data and absolute values of the biomarkers studied at 24 weeks’ gestation are displayed as medians and interquartile ranges in Table 3. Levels of TPR, PlGF, sFlt-1, uric acid, and the sFlt-1/PlGF ratio were all significantly different in women who later developed sPE compared with women who remained normotensive (normal and IUGR groups combined). The ability of the tests at 24 weeks to identify those women who subsequently developed sPE is summarized in Table 4. Serum uric acid was the strongest predictor; a cut-off value for uric acid of 255 μmol/L achieved an area under the curve value of 0.99 (95% CI 0.94 to 1.0), with 100% sensitivity and a 7% false-positive rate (1-specificity).
Table 3Hemodynamic parameters and biomarkers in women with severe preeclampsia versus normal and IUGR groups combined at 24 weeks’ gestation. Results are displayed as medians (interquartile ranges)
In the six women who were normotensive at the time of sampling but who subsequently developed sPE, sFlt-1 levels positively correlated with TPR (r=0.65; P=0.01) (Figure 3). In contrast, serum uric acid, PlGF, and the sFlt-1/PlGF ratio did not correlate with TPR in these women. In the women who did not develop sPE, TPR correlated positively with PlGF (r=0.47; P=0.001), and negatively with the sFlt-1/PlGF ratio (r=−0.29; P=0.03). Serum uric acid correlated positively with sFlt-1 (r=0.65; P=0.003) and the sFlt-1/PlGF ratio (r=0.54; P=0.02) but not with PlGF.
Figure 3Correlation between sFlt-1 and TPR with 95% CI in the six women who developed severe preeclampsia during the phase prior to the onset of hypertension
Our study demonstrates that the serial integration of non-invasive maternal hemodynamics with maternal circulating levels of uric acid, sFlt-1, and PlGF, including the sFlt-1/PlGF ratio, measured at 24 weeks, can identify women with evolving subclinical severe preeclampsia. In addition, we demonstrated the serial changes of the hemodynamic profile in women with and without evolving severe preeclampsia. Women destined to have normal outcomes showed the expected fall in TPR to 30 weeks’ gestation, followed by a slow but steady physiologic increase in the third trimester.
The group destined to develop IUGR showed an increase in TPR from 20 weeks’ gestation with a fall in cardiac output due to falling stroke volume. This has been previously documented by Bamfo et al., who used echocardiography to show a falling maternal cardiac output due to decreased SV in normotensive women with IUGR compared to normotensive women with appropriate fetal growth.
However, we found that women destined to develop sPE exhibited a rapid and marked increase in TPR with a fall in cardiac output. Interestingly, the decline in cardiac output was due to a fall in heart rate while SV remained relatively stable. These observations have important implications for the treatment of early onset preeclampsia, since drugs with a predominant β-blocker action (such as labetalol) are commonly used by obstetricians and internists; their use could result in a further fall in maternal cardiac output.
Over 80% of women who develop preeclampsia do so near term, and they have no greater perinatal morbidity than normotensive women.
Women developing preeclampsia near term have normal to high cardiac output on hemodynamic assessment; as an illustration, a randomized control trial of atenolol to suppress elevated cardiac output from 24 weeks reduced the incidence of preeclampsia from 18% to 3.8%, but resulted in a mean reduction in birth weight of 440 g.
In contrast, the small proportion of women who are at risk of early-onset severe disease, with associated placental pathology and IUGR, have the opposite hemodynamic profile, with a low cardiac output and high vascular resistance. This difference has previously been documented
but not utilized as we have now done, by integration with maternal blood tests, allowing accurate early identification of the most vulnerable women when clinically well.
Our results are in keeping with those of recent echocardiographic assessment studies that have shown that differences in hemodynamic profiles between women who will or will not develop preeclampsia are present in the early stages of the second trimester.
Prospective observational study of serial cardiac output by transthoracic echocardiography in healthy pregnant women undergoing elective caesarean delivery.
Khaw et al. demonstrated higher TPR and lower cardiac output at 11 to 14 weeks’ gestation in women who subsequently developed fetal growth restriction (<10th centile) with or without preeclampsia.
Furthermore, women who remained normotensive with fetal growth restriction had a significantly lower SV than women with normal pregnancies. Studies have documented decreased maternal systolic function in women with both sPE and IUGR
In contrast to echocardiography, NICOM is operator independent, very simple to use, and is easily performed in the outpatient setting. Our study indicates that this is a novel investigational tool that may be very useful in revealing the evolving maternal vasculopathy that precedes overt hypertension and could in turn guide therapeutic intervention.
Our integrated approach is important practically, because it is applicable in an outpatient setting except for the potential need to introduce real-time reporting of the sFlt-1/PlGF ratio test. In a much larger study, Rana et al. showed that the (retrospective) use of the sFlt-1/PlGF ratio test gave optimal test characteristics to predict severe preeclampsia, although their real-time measurement of serum uric acid performed well.
We found that the best maternal blood predictor of severe preeclampsia was in fact measurement of serum uric acid, a simple automated biochemical test that is a fraction of the cost of the sFlt-1/PlGF ratio test, even on an automated analyzer. Our findings support recent conclusions that measurement of serum uric acid deserves further comparison with more expensive tests of placentally derived factors that directly regulate vascular function. The concept of using the ratio of the anti-angiogenic (sFlt-1) and pro-angiogenic (PlGF) blood markers as a predictor of severe preeclampsia is attractive. PlGF, a member of the VEGF family, is a pro-angiogenic peptide secreted by the placenta and circulates in high concentration during normal pregnancy. PlGF is also a potent vasodilator, especially of human uterine arteries, through effects on nitric oxide and prostaglandin I2.
Its target receptor is membrane bound fms-like tyrosine kinase. In its soluble form (sFlt-1), this protein binds circulating PlGF in the serum, thereby reducing the effective circulating concentration of active PlGF. During the second trimester in a normotensive pregnancy, the PlGF concentrations are high and the sFlt-1 concentrations low, creating a pro-angiogenic state.
This increase begins approximately five weeks before the onset of clinical disease. Parallel with the increase in the sFlt-1 level, there are decreases in free PlGF levels which may be attributed to binding by sFlt-1. Our serial data indicate that maternal circulating sFlt-1 levels correlate positively with maternal TPR in normotensive women who will subsequently develop sPE. Noori et al. performed serial assessments of maternal vascular function using brachial artery flow-mediated dilatation and measurements of PlGF, sFlt-1, and their ratio.
Given the inhibitory action of sFlt-1 on PlGF, our data support a mechanistic role for these factors in the evolution towards severe preeclampsia.
Nevertheless, our data suggest that the much simpler rapid assay of serum uric acid deserves more attention. Some studies have assessed uric acid as a predictor of all forms of preeclampsia, but with mixed results.
Recently, serum uric acid levels have been shown to identify women who may develop preeclampsia among women referred for assessment of gestational hypertension.
Uric acid levels are generally raised once the clinical disease is manifest, and the level of increase above normal correlates with the severity of maternal and fetal risk.
Our study is the first to detect a positive correlation between serum uric acid, sFlt-1, and elevated TPR, and supports the concept that increased uric acid concentrations are part of the pathogenesis of the vasculopathy rather than a mere marker of disease.
This study has some limitations. Although it was prospective, this was a small pilot study; therefore, the interpretation of our results is limited by sample size, even though 30% of the cohort developed severe preeclampsia. In addition, we were not blinded to the hemodynamic profile of the women during the study period.
CONCLUSION
We have shown that the combination of non-invasive determination of TPR and uric acid levels may identify a subset of clinically high-risk women with evolving sPE, independent of determining the sFlt-1/PlGF ratio. The predictive ability of this integrated approach needs to be assessed in a larger cohort of women to further confirm its utility.
ACKNOWLEDGEMENTS
Funding for this study was provided by the Rose Torno Chair, Mount Sinai Hospital (to John Kingdom), and the Canadian Institutes for Health Research.
REFERENCES
Sibai B.M.
Gordon T.
Thom E.
Caritis S.N.
Klebanoff M.
McNellis D.
et al.
Risk factors for preeclampsia in healthy nulliparous women: a prospective multicenter study. The National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units.
Prospective observational study of serial cardiac output by transthoracic echocardiography in healthy pregnant women undergoing elective caesarean delivery.
30511-9&id=gr1.jpg){kind=link}
30511-9&id=gr2.jpg){kind=link}
30511-9&id=gr3.jpg){kind=link}