JCDR - Register at Journal of Clinical and Diagnostic Research
Journal of Clinical and Diagnostic Research, ISSN - 0973 - 709X
Biochemistry Section DOI : 10.7860/JCDR/2014/7901.4533
Year : 2014 | Month : Jul | Volume : 8 | Issue : 7 Full Version Page : CC04 - CC07

Association of Lipid Profile and Uric Acid with Pre-eclampsia of Third Trimester in Nullipara Women

Vibhuti Agarwal1, Bharat Kumar Gupta2, Abhishek Vishnu3, Mamtatyagi4, Shiprasolanki5, Jas Kiran6

1 Post Graduate Student, Department of Biochemistry, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.
2 Professor, Department of Biochemistry, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.
3 Assistant Professor, Department of Anaesthesia, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.
4 Professor, Department of Obstetrics & Gynecology, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.
5 Post Graduate Student, Department of Biochemistry, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.
6 Assistant Professor, Department of Biochemistry, Subharti Medical College, Swami Vivakanand Subharti University, Meerut, India.


NAME, ADDRESS, E-MAIL ID OF THE CORRESPONDING AUTHOR: Dr. Vibhuti Agarwal, Post Graduate Student, Department of Biochemistry, Subharti Medical College, Swami Vivakanand Subharti University, Bye Pass Road, Meerut- 250005, India.
Phone: 919760036256, 919412701534,
E-mail: anchitbharat@hotmail.com;
Abstract

Background: Pre-eclampsia (PE) affects approximately 3% of all pregnancies worldwide, with onset of symptoms in the late divond or third trimester, commonly after 32nd week. It is common in nulliparous women. To avoid complications it is necessary to diagnose it in advance, but the available tools are unable to clinch the diagnosis of preeclampsia effectively in majority.

Aim: To find out an association of lipid profile and uric acid with PE in nullipara pregnant women in third trimester.

Materials and Methods: One hundred nulliparous pregnant women in their third trimester of 18-35 years were divided into; 50 pre-eclamptics of study group and 50 non pre-eclamptic in control group; further subdivided according to age, 18-26 and 27-35 yrs. Diagnosis was confirmed as per the standard criteria. Lipid profile and uric acid levels were estimated by Vitros 250 dry chemistry analyser. Data was analysed statistically by student t-test at p<0.01 level of significance.

Results: TC, LDL-c and VLDL-c levels in the study group as a whole and in the patients between 18-26 years were significant; HDL-c levels in the patients between 27-35 years were significant while TG and uric acid levels in all the three study groups were significant.

Conclusion: Total cholesterol, LDL-c, VLDL-c, triglycerides (TGs) and uric acid levels were raised in preeclampsia and statistically significant; while HDL-c levels were raised in these patients but statistically non-significant, it can be concluded that there exists an association in lipid profile and uric acid with PE therefore dyslipidemia and raised uric acid levels are the features of PE in nullipara pregnant women in their third trimester.

Keywords

Introduction

Pre-eclampsia PE affects approximately 3% of all pregnancies worldwide [1], with onset of symptoms in the late second or third trimester, most commonly after the 32nd week. PE is a multi-system disorder of pregnancy, which is characterised by new onset hypertension and proteinuria that develop after 20 weeks of gestation in previously normotensive women [2,3].

The disorder has a higher incidence among nulliparous women, in women who conceive with assisted reproduction techniques, and in women affected by autoimmune disorders, reflecting the probable influence of an “inexperienced” or dysregulated maternal immune system in its emergence [4,5]. On the other hand, women with pre-existing metabolic, vascular or renal disease are especially at increased risk for superimposed PE [6].

The mechanism of causation of preeclampsia is not well understood. The available tools for its diagnosis are effective only when the disease sets in, and in many cases at this stage; it becomes difficult to avoid complications. It is necessary to diagnose this condition in advance so that the future complications of mother as well as fetus may be prevented.

Early pregnancy dyslipidemia is associated with an increased risk of PE. In pregnancy, lipolysis of TG-rich lipoproteins is reduced because of decreased lipolytic activities of the mother. In PE, the vascularization of the fetoplacental unit may be impaired, resulting in yet-undefined compensatory mechanisms that may further increase synthesis of maternal Triglyceride (TG) levels. In addition, the decreased catabolism of TG-rich lipoproteins by reduced placental uptake and the putative concomitant decrease of lipoprotein lipolysis results in the accumulation of TG-rich remnant lipoproteins in the maternal circulation. Remnant lipoproteins may induce platelet activation and endothelial dysfunction, thus leading to the major clinical symptoms of PE [7].

There are several potential origins for uric acid in preeclampsia; abnormal renal function, increased tissue breakdown, acidosis and increased activity of the enzyme xanthine oxidase/dehydrogenase [8].

We have done an extensive review of the literature relating lipid profile and uric acid with PE. Several studies we found were conducted outside India, very few studies conducted on Indian population are available; so we decided, “To find out any possible correlation between lipid profile and uric acid level with PE in nullipara pregnant women in their third trimester”.

Objective

To find out an association of lipid profile and uric acid with PE in nullipara pregnant women in third trimester.

Materials and Methods

Nulliparous pregnant women in their third trimester (25-36 weeks of gestation), attending Obstetrics and Gynecology Department, Chhatrapati Shivaji Subharti Hospital were enrolled and divided into study group (Pre-eclamptic) and control group (Normotensive with no proteinuria). Each group had 50 subjects. Clearance was obtained from the (IEC) and informed consent was obtained from every subject.

Patients with Diabetes Mellitus, Gestational Diabetes, Chronic hypertension (hypertension arising before 20 weeks gestation), Coronary artery disease, Hepatitis B infections, Human immunodeficiency virus (HIV) infection, Chronic obstructive pulmonary disease (COPD), Complicated and/or multiple pregnancies, women consuming antihypertensive medication, aspirin and / or corticosteroids, Undiagnosed patients and patients positive for urinary protein with urinary tract infection were not included in the study.

General information(s), detailed history, complete general and systemic examination and routine investigations were done in all subjects.

Sample was collected from the antecubital vein, aseptically through venipuncture. Hemolysed samples were discarded. Either the blood sample was immediately processed or stored in a fridge at 2-8°C. In some events the serum was separated, labeled and stored at -70°C. These samples were thawed prior to analysis.

PE was diagnosed as a blood pressure level ≥ 140/90 mm of Hg; confirmed by at least two blood pressure measurements six hours apart and proteinuria ≥2+ by dipstick in “clean-catch-midstream” random sample of urine, collected at the time of enrollment.

Lipid profile including TC, TG, HDL-c, LDL-c, VLDL-c and Serum uric acid levels were investigated in all the subjects.

The serum was processed in Vitros-250 auto analyser using readymade dry chemistry kits procured from Ortho-Clinical Diagnostics, Johnson & Johnson, USA. Internal and external quality controls were followed as per NABL accreditation program. Levi-Jenning Plot was fed in Vitros 250 auto analyser and strictly followed. Samples giving readings above or below two SD were reprocessed / discarded.

All the data recorded was compiled and statistically analysed by using Mean, Standard deviation and Students t-test.

Results

The present study was conducted on 100 nulliparous pregnant women in their third trimester between the ages 18 to 35 years; 50 were patients of PE in study group and 50 normotensive non-pre-eclamptic women in control group. PE was diagnosed on the basis of history, clinical examination, blood pressure findings and presence of proteinuria. Subjects were distributed as; between 18-26 and 27-35 years.

In the study group 43 (86%) were in the age group 18-26 years and 7 (14%) were between 27-35 years of age while in the control group distribution of subject was 46 (92%) and 4 (8%) respectively.

Results and observations are given in [Table/Fig-1,2,3,4,5and6].

Total cholesterol level and its statistical significance in different groups

Distribution of CasesGroupsMean ± SD (mg/dl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)201.86±25.73.0000p<0.01 (significant)
Control (n=46)165.13±19.30
27-35 Yrs.Study (n=7)192.42±38.12.2390p>0.01 (N.S.)
Control (n=4)170±20.84
TOTALStudy (n=50)200.54±27.50.0000p<0.01 (significant)
Control (n=50)165.52±19.24

HDL cholesterol level and its statistical significance in different groups

Distribution of CasesGroupsMean ± SD (mg/dl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)37.18±10.95.2897p>0.01 (N.S.)
Control (n=46)35.13±6.49
27-35 Yrs.Study (n=7)42.85±12.41.0001p<0.01 (significant)
Control (n=4)32±7.11
TOTALStudy (n=50)37.98±11.21.0951p>0.01 (N.S.)
Control (n=50)34.88±6.527

LDL cholesterol level and its statistical significance in different groups

Distribution of CasesGroupsMean ± SD (mg/dl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)117±25.75.0046p<0.01 (significant)
Control (n=46)103.17±19.30
27-35 Yrs.Study (n=7)106.85±16.72.0985p>0.01 (N.S.)
Control (n=4)110.50±14.88
TOTALStudy (n=50)116±24.82.0074p<0.01 (significant)
Control (n=50)104±19

VLDL cholesterol level and its statistical significance in different groups

Distribution of CasesGroupsMean ± SD (mg/dl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)47.25±7.79.0000p<0.01 (significant)
Control (n=46)26.71±2.23
27-35 Yrs.Study (n=7)51.57±8.05.7201p>0.01 (N.S.)
Control (n=4)27.50±1.73
TOTALStudy (n=50)47.86±7.89.0000p<0.01 (significant)
Control (n=50)26.78±2.19

Triglyceride level and its statistical ignificance in different groups

Distribution of CasesGroupsMean ± SD (mg/gl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)236.48±39.06.0000p<0.01 (significant)
Control (n=46)133.60±11.72
27-35 Yrs.Study (n=7)257.85±40.29.0001p<0.01 (significant)
Control (n=4)138±9.89
TOTALStudy (n=50)239.50±39.54.0000p<0.01 (significant)
Control (n=50)134±11.56

Uric acid level and its statistical significance in different groups

Distribution of CasesGroupsMean ± SD (mg/dl)Probability of Unpaired t-Testp-value/ Significance
18-26 Yrs.Study (n=43)8.195±1.170.0000p<0.01 (significant)
Control (n=46)4.658±0.839
27-35 Yrs.Study (n=7)8.385±0.944.0011p<0.01 (significant)
Control (n=4)4.425±1.084
TOTALStudy (n=50)8.222±1.1346.0000p<0.01 (significant)
Control (n=50)4.64±0.8506

Discussion

Definition of PE has changed with time. The initial international classification and definition of the hypertensive disorders of pregnancy compiled by Davey et al., defined it to be having a diastolic blood pressure of 90 mm Hg on two occasions, or 110 mm Hg on a single occasion [9].

In PE the glomerular barrier is certainly altered and there is an increased excretion of protein including albumin. When total protein excreted (TPE) exceeds 1 g/24 hours or 1+ by dipstick, tubular protein reabsorption will be saturated and individual proteins excretion rates will be related to their molecular weights (Davison) [10].

We selected nullipara women for the present study as nulliparity is not only have high risk of PE but is the most common maternal risk factor which can easily be assessed just by the history. Duckitt et al., concluded in their study that nulliparity is not only the most common maternal risk factor but it has been shown to almost triple the risk of PE [11].

It was found that TC, LDL-c and VLDL-c levels in the study group as a whole and in the patients between 18-26 years were statistically significant at p<0.01 level of significance; while HDL-c levels were found raised in all the three subgroups in patients of PE compared to normotensive pregnant women and were found to be statistically significant only in patients between 27-35 years.

It was also observed that TG and uric acid levels were found raised in all the three subgroups in patients of PE as compared to normotensive pregnant women and that these levels in all the three study groups were statistically significant.

Wladimiroff et al., [12] and Cleusen et al., [13] obtained similar findings for lipid profile in pre-eclamtic and normotensive pregnancies.

Increased levels of TG with reduced HDL-C have been observed in our study. Enquobahrie et al., [14] and Gractacose et al., [15]; in their respective studies concluded that hypertriglyceridemia is probably a consequence of competition between chylomicrons and very low-density lipoprotein (VLDL) cholesterol for the lipoprotein lipase.

The principle modulator of this hypertriglyceridemia is oestrogen as pregnancy is associated with hyperoestrogenaemia. Oestrogen induces hepatic biosynthesis of endogenous TG which is carried by VLDL [16]. This process may be modulated by hyper insulinism found in pregnancy [17]. Serum TG concentration increases much more significantly in pre-eclamptic pregnancy. Our study also corroborated with the findings of Equobohrie et al., [14] and Cekmen et al., [18].

The above mentioned interactions along with increased endothelial TG accumulation may result in endothelial cell dysfunction [19]. Increased TG found in pregnancy induced hypertension including PE, is likely to be deposited in predisposed vessels, such as the uterine spiral arteries and contributes to the endothelial dysfunction both directly and indirectly through generation of small dense LDL [20].

In the present study, serum VLDL-c level was also significantly higher (p < 0.01) in the third trimester of pre-eclamptic pregnancy which is probably due to hypertriglyceridemia leading to enhanced entry of VLDL that carries endogenous TG into circulation. VLDL levels were also found increased in pre-eclamptic in the studies conducted by Kokio et al., and Teichmann et al., this is probably due to increased VLDL lipoproteins which accumulate over the maternal vascular endothelium, particularly those of uterine and renal vessels. The VLDL–c level as reported by some researchers might increases upto 2.5 folds at term over the pre-pregnancy level [21,22].

A significant higher level of LDL-c level, as was found in the present study, was also reported by Grratacos et al. and wakatsuki et al., in third trimester of pre-eclamptic pregnancy, may be attributed to hyperestrogenacmia [23].

Lipid metabolism is altered during pregnancy and is characterised by normal or even low cholesterol during early pregnancy and hypertriglyceridaemia in late pregnancy. The anabolic phase of early pregnancy produces metabolic changes that encourage lipogenesis and fat storage in preparation for the catabolic phase of late pregnancy in which there is rapid fetal growth. The insulin resistance of pregnancy increases lipolyis in adipose tissue, leading to an enhanced flux of fatty acids to the liver. This promotes the synthesis of VLDLs and as a result, increased TG concentrations. In addition, insulin resistance reduces the activity of lipoprotein lipase an insulin dependent enzyme that is responsible for VLDL clearance from plasma. Therefore, VLDL remains in the plasma for longer and ultimately leads to accumulation of low density lipoprotein (LDL) [24].

In the past, uric acid was only considered as being the cause of gout; however, it is now increasingly thought of as a factor that can alter endothelial cell function. Hence, interest has been renewed in uric acid as a possible initiator of the maternal response in PE, rather than it being just a ‘bystander’ marker of the disorder. Over the past 10–20 years, a number of biochemical markers have been proposed to predict which women will develop preeclampsia. These markers include human chorionic gonadotropin, α-fetoprotein, urinary calcium, antiphospholipid antibodies, urinary kallikrein and homocysteine. In the past 5–10 years, additional factors, such as soluble endoglin, P1GF and cellular fibronectin have also been proposed as biomarkers for PE [25].These markers were identified based on the pathophysiological abnormalities associated with PE, including placental dysfunction, endothelial and coagulation activation and systemic inflammation. However, data regarding the reliability of these markers as predictors of the development of PE have been inconsistent, as most markers lack the specificity and sensitivity needed for routine clinical use [26].

It was observed from the present study that uric acid levels were not only found raised in all the three subgroups in patients of PE as compared to normotensive pregnant women but they were highly significant also. Lilliana Susanana et al., have made similar observations of higher levels of uric acid in serum of patients of PE which were statistically significant in their study which compared these levels in essential hypertensives, normal and pre-eclamptic women during pregnancy in their third trimester [27].

Serum uric acid is one of the parameters used in early diagnosis of preeclampsia. An elevated level of uric acid reflects the degree of placental cell destruction as well as severity of preeclampsia [28].This may lead to decrease in the renal tubular excretion. Altered renal handling of urate clearance may be due to renal dysfunction and increased xanthine oxidase activity. Thus hyperuricemia in PE is primarily due to decreased renal clearance and increased tubular reabsorption of uric acid, because of the reduction in glomerular filtration rate [29,30].

AC Martin et al., concluded that increased uric acid concentrations may be part of the pathogenesis of the clinical syndrome rather than a marker of PE. Alternatively, production of uric acid might increase in pre-eclamptic women as part of an appropriate response to inflammation. The known role of uric acid as a scavenger of oxygen free radicals supports this theory [31].

Uric acid has important role in vascular damage and oxidative stress. Hyperuricemia may also reflect impaired endothelial integrity and contribute to the pathogenesis of PE. Hence, early estimation of serum uric acid might reduce systemic complications and maternal deaths due to PE [28].

Limitations

Single sample was tested for lipid profile and uric acid levels.

Results were not compared with non pregnant women.

Recommendations

1. Sample should be drawn serially at definite intervals in third trimester and tested for lipid profile and uric acid levels.

2. A group including normal healthy non-pregnant women in same age should also be included in the study and their levels of lipid profile and uric acid be compared with the existing two groups.

Conclusion

Total cholesterol, LDL-c, VLDL-c, TG and uric acid levels were raised in pre-eclamptic and statistically significant; while HDL-c levels were raised in these patients but statistically non-significant, it can be concluded that there exists an association in lipid profile and uric acid with pre-eclamptic therefore dyslipidemia and raised uric acid levels are the features of pre-eclamptic in nullipara pregnant women in their third trimester.

References

[1]World Health Organization (WHO) World health report: Make every mother and child count 2005 GenevaWHO:63  [Google Scholar]

[2]Redman CW, Sargent IL, Latest advances in understanding preeclampsia Science 2005 308:1592-94.  [Google Scholar]

[3]Sibai B, Dekker G, Kupferminc M, PE Lancet 2005 365:785-99.  [Google Scholar]

[4]Saito S, Shiozaki A, Nakashima A, Sakai M, Sasaki Y, The role of the immune system in preeclampsia Mol Aspects Med 2007 28:192-209.  [Google Scholar]

[5]Sargent IL, Borzychowski AM, Redman CW, Immunoregulation in normal Pregnancy and PE: an overview Reprod Biomed Online 2006 13:680-86.  [Google Scholar]

[6]Catov JM, Ness RB, Kip KE, Olsen J, Risk of early or severe preeclampsia related to pre-existing conditions Int J Epidemiol 2007 36:412-19.  [Google Scholar]

[7]Enquobahrie DA, Williams MA, Butler CL, Frderick IO, Miller RS, Luthy DA, Maternal plasma lipid concentration in early pregnancy and risk of preeclampsia Am J Hypertens 2004 17(7):574-81.  [Google Scholar]

[8]Johnson RJ, Kang DH, Feig D, Kivlighn S, Kanellis J, Watanabe S, Is there a pathogenetic role for uric acid in hypertension and cardiovascular and renal disease? Hypertension 2003 41:1183e90  [Google Scholar]

[9]Davey DA, Macgillivary I, The classification and definition of the hypertensive disorders of pregnancy Am J ObstetGynecol 1988 158:892-98.  [Google Scholar]

[10]Davison JM, Renal function during normal pregnancy and the effect of renal disease and PE The kidney in pregnancy 1986 Martinus Nijhoff, Boston:65-80.In: Andreucci VE (ed)  [Google Scholar]

[11]Duckitt K, Harrington D, Risk factors for PE at antenatal booking: systematic review of controlled studies BMJ 2005 330:565  [Google Scholar]

[12]Waldimiroff JW, VandenElzen HJ, Cohen–Overbeck TE, De Bruijin AJ, Grobbee DE, Serum Lipids in early pregnancy and risk of preeclampsia Br. J. obstet Gynecol 1996 103:107-22.  [Google Scholar]

[13]Cleusen T, Djurovic S, Henriksen T, Dyslipdemiaa early second trimester is mainly a feature of women with early onset preeclampsia Br. J. Obstet Grynecol 2001 108:1081-87.  [Google Scholar]

[14]Enquobahrie DA, Williams MA, Butter CL, Frederick IO, Miller RS, Luthy DA, Maternal Plasma Lipid concentrations in early pregnancy and risk of preeclampsia Am.J. hypertens 2004 17(7):574-81.  [Google Scholar]

[15]Grratacos E, Casals E, Gomez O, Llurba E, Merdader I, Cararach V, Increased susceptibility to low density lipoprotein oxidation in women with a history of preeclampsia British Journal of Obst & Gynae 2003 110(4):400-04.  [Google Scholar]

[16]Glueck CJ, Fallet RW, Scheel D, Effects of oesterogenic compounds on triglyceride kinetics Metabolism 1975 24:537-45.  [Google Scholar]

[17]Adegoke OA, Lyare EE, Gbenebitse SO, Fasting plasma glucose and cholesterol levels in pregnant Nigerian Women Niger Postgrad Med J 2003 10(1):32-6.  [Google Scholar]

[18]Cekmen MB, Erbagci AB, Balat A, Duman C, Maral H, Ergen K, Plasma lipid and lipoprotein concentrations in pregnancy induced hypertension Clin. Biochem 2003 36(7):575-81.  [Google Scholar]

[19]Mikhail MS, Basu J, Palan PR, Furgiusle J, Romney SL, Anyaegbunam A, Lipid profile in women with preeclampsia: relationship between plasma triglyceride levels and severity of preeclampsia J. Assoc. Acad. Minor Phys 1995 6(1):43-5.(1995)  [Google Scholar]

[20]Sattar N, Bendomir A, Berry C, Shepherd J, Greer IA, Packard CJ, Lipoprotein subfraction concentrations in preeclampsia: pathogenic parallels to atherosclerosis Obstet Gyneco 1997 89(3):403-08.  [Google Scholar]

[21]Kokia E, Barkai G, Reichman B, Segal P, Goldman B, Mashiach S, Maternal serum lipid profile in pregnancies complicated by hypertensive discorders J. Parinat. Med 1990 18(6):473-78.  [Google Scholar]

[22]Teichmann AT, Wieland H, Cremer P, Knolow G, Mehle U, Serum lipid and lipoprotein concentrations in pregnancy and at onset of labour in normal and complicated pregnancies caused by hypertensive gestosis and fetal growth retardation Geburtshilfe Frauenheilkd, Germany 1988 48(3):134-39.  [Google Scholar]

[23]Wakatsuki A, Ikenoue N, Okatani Y, Shinohara K, Fukaya T, Lipoprotein particles in preeclampsia: Susceptibility to oxidative motidification Obstet. Gynecol 2000 96(1):55-9.  [Google Scholar]

[24]Silliman K, Shore V, Forte TM, Hypertriglceridema. During late pregnancy in associated with the formation of small dense low – density lipoproteins and the presence of large buoyant high density lipoproteins Metabolism 1994 43:1035-41.  [Google Scholar]

[25]Carty DM, Delles C, Dominiczak AF, Novel biomarkers for predicting preeclampsia Trends Cardiovasc. Med 2008 18:186-94.  [Google Scholar]

[26]Cnossen J, S.Are tests for prediction PE good enough to make screening viable? A review of reviews and critical appraisal Acta Obstet. Gynecol. Scand 2009 88:758-65.  [Google Scholar]

[27]Liliana Susana Voto, Ricardo Elia, Darbon-Grosso Francisco Uranga Imaz, Miguel Margulies, Uric acid levels: a useful index of the severity of preeclampsia and perinatal prognosis J. Perinat. Med 1988 16:123-27.  [Google Scholar]

[28]Bargale A, Ganu J, Trivedi D, Nagane N, Mudaraddi R, Sagare A, Serum Hs-CRP and uric acid as indicator of severity in Preeclampsia IJPBS 2011 2(3):340-45.  [Google Scholar]

[29]Sahu S, Daniel M, Abraham R, Vedavalli R, Senthilvel V, Study of uric acid and nitric oxide concentrations in preeclampsia and normal pregnancy Int J Biol Med Res 2011 2(1):390-93.  [Google Scholar]

[30]Dane B, Kayaoglu Z, Dane C, Batmaz G, Kiray M, Doventas Y, The relationship between elevated maternal uric acid level and bilateral early diastolic notching at uterine arteries at second trimester and pregnancy complications Perinatal Journal 2011 19(2):64-70.  [Google Scholar]

[31]Annable C. Martin, Could uric acid have a pathogenic role in PE? Nature reviews nephrology 2011 6:744-48.  [Google Scholar]