vcam in preeclampsia.pdf

Ž .International Journal of Gynecology & Obstetrics 63 1998 115]121

Article

Ž .Vascular cell adhesion molecule-1 VCAM-1 andleukocyte activation in pre-eclampsia and eclampsia

E. Budak, R. Madazli, M.F. AksuU, A. Benian, A. Gezer, N. Palit, F. Yildizfer

Department of Obstetrics and Gynecology, Cerrahpasa Medical Faculty, Uni ersity of Istanbul, Istanbul, Turkey

Received 23 February 1998; received in revised form 1 July 1998; accepted 16 July 1998

Abstract

Objecti e: To evaluate the levels of VCAM-1 in pre-eclampsiareclampsia as a possible marker of leukocyteactivation and endothelial damage. Methods: We performed a case-control study on 25 healthy pregnant women and35 patients diagnosed as pre-eclampsia or eclampsia which were randomly selected. Peripheral venous blood samples

Ž .were obtained and serum levels of VCAM-1 were measured by enzyme-linked immunoassay ELISA . Results: Inpre-eclampsiareclampsia, VCAM-1 levels were higher than in normal pregnancy. Serum concentrations of VCAM-1were significantly higher in severe pre-eclampsia and eclampsia compared to mild pre-eclampsia or healthy controls.ROC analysis detected that VCAM-1 G450 ngrml had a sensitivity of 0.79 and a specificity of 0.90 in detectingsevere pre-eclampsia and eclampsia. Conclusion: This is the first study to correlate VCAM-1 levels with severity ofdisease in pre-eclampsia. Our findings indicate that increasing levels of soluble VCAM-1 are present in thecirculation of patients with severe pre-eclampsiareclampsia compared to mild pre-eclampsia or healthy pregnantwomen. Elevated VCAM-1 levels may represent a possible mechanism by which endothelial cells attract leukocytesand cause endothelial cell damage. Q 1998 International Federation of Gynecology and Obstetrics

Keywords: Vascular cell adhesion molecule-1; Endothelial damage; Pre-eclampsia

U Corresponding author. Tel.: q90 212 5861514; fax: q90 212 586 1514; e-mail: [email protected]

0020-7292r98r$ - see front matter Q 1998 International Federation of Gynecology and ObstetricsŽ .P I I S 0 0 2 0 - 7 2 9 2 9 8 0 0 1 3 8 - 6

( )E. Budak et al. r International Journal of Gynecology & Obstetrics 63 1998 115]121116

1. Introduction

Increasing evidence suggests that generalizedendothelial cell damage and dysfunction are ma-jor pathophysiologic features of pre-eclampsiaw x1]3 . Although, the mechanisms of endothelialactivation and dysfunction are yet unclear, thereis substantial evidence suggesting that leukocyte-

w xinstigated damage contribute to this process 4]6 .Recently, studies have attempted to identifymarkers of leukocyte activation and leukocyteinitiated endothelial damage in pre-eclampsiaw x7]9 .

VCAM-1 is a cell adhesion molecule and amember of the immunoglobulin supergene familyw x10 . VCAM-1 has a single chain glycoproteinstructure and functions as a transmembrane re-ceptor in vascular endothelial cell membranes.

Ž .Very late antigen-4 VLA-4 , a a b integrin, is4 1the counter ligand of VCAM-1 and is expressedby neutrophils, monocytes, lymphocytes, ba-

w xsophils, eosinophils and certain tumor cells 11,12 .Cytokines and other pro-inflammatory substancesstimulate the production of endothelial cell adhe-sion molecules and their transport to the cell

w xmembrane 13,14 . Concurrently, after inductionby chemostimulants or specific signals, leukocytesbecome activated and adhere to endothelial cells,

w xby binding to VCAM-1 15 . Leukocytes becomefirmly attached to vascular endothelium andrapidly migrate into tissues through gaps withinthe endothelial lining to take part in inflamma-tory processes. When the primary insult is exertedfor a prolonged duration, as in chronic inflamma-tion or autoimmune disorders, leukocyte activa-tion and recruitment to the involved area mayaccentuate the initial insult and ultimately may

w xbecome a primary cause of tissue injury 4 .It is believed that VCAM-1 can be utilized as a

marker to monitor endothelial and leukocyte acti-w xvation 9,16]18 . VCAM-1 levels have been de-

monstrated to correlate with disease state pro-w xgression 19]22 . The purpose of the present study

was to measure circulating VCAM-1 levels innormal pregnant women and pre-eclampsia, andto determine if disease severity in pre-eclampsiais correlated with serum VCAM-1 concentrations.

2. Materials and methods

2.1. Patient selection

The study population consisted of 25 womenwith normal pregnancy and 35 women with pre-eclampsiareclampsia who were diagnosed andtreated in our department. The mean age andgestational weeks of the groups studied were simi-lar. The exclusion criteria were the presence ofactive infectious disease or medical complicationsincluding autoimmune disorders, diabetes melli-tus and inflammatory conditions. The hyperten-sive patients were subclassified as mild pre-

Ž . Ž .eclampsia ns16 , severe pre-eclampsia ns16Ž . w xand eclampsia ns3 23 .

2.2. Blood samples and VCAM-1 measurements

Single peripheral venous blood samples wereobtained from all patients included in the study.Blood samples from pre-eclamptic and eclampticwomen were drawn during labor or within 2 weeksof delivery between 28 and 41 weeks as soon asthe diagnosis was established. Samples from thenormotensive group were obtained between 26and 40 weeks gestation. The samples were storedat y308C until the analyses were performed.Serum levels of human soluble VCAM-1Ž .h-sVCAM-1 were quantified by an ELISA tech-nique according to the manufacturer’s instruc-

Ž .tions Boehringer Mannheim, Germany .

2.3. Statistical analysis

Data analysis was performed using the SPSSŽstatistical software package SPSS Inc., Chicago,

.USA . The results were evaluated utilizing theMann]Whitney U-test, correlation and regressionanalysis. Test performance was evaluated by re-ceiver-operating characteristic analysis using theGraphROC program for Windows.

3. Results

There was no significant difference betweenthe distribution of patients in relation to gestatio-

( )E. Budak et al. r International Journal of Gynecology & Obstetrics 63 1998 115]121 117

nal age at which blood samples were obtained.VCAM-1 levels did not differ in relation to gesta-tional age in the normotensive groupŽ 2 .r s0.0004 . The levels of VCAM-1 were sig-nificantly higher in pre-eclampsia and eclampsia,

Ž .compared to normal pregnant women P-0.001 .When VCAM-1 levels of patients in pre-eclampsiareclampsia were further analyzed in

Žrelation to disease severity mild pre-eclampsia,.severe pre-eclampsia and eclampsia different re-

Ž .sults were obtained Fig. 1 . Interestingly, al-though VCAM-1 levels were elevated in mild

Ž .pre-eclampsia 366"78 compared to normalŽ .pregnant women 326"102 , the difference was

Ž .not statistically significant P)0.05 . However,when VCAM-1 levels of severe pre-eclampsiaŽ . Ž .513"125 and eclampsia patients 649"49were compared to control or mild pre-eclampsiapatients, a highly significant increase was foundŽ .P-0.001 . There was a significant direct linear

correlation with the increments of diastolic bloodpressure and VCAM-1 levels in the hypertensive

Ž 2 . Ž .group P-0.001, r s0.69 Fig. 2 , but not inthe normal pregnancy group. VCAM-1 levels wereincreased with diastolic blood pressure elevation,most marked above 110 mmHg, which also corre-sponds to severe pre-eclampsia and eclampsia.

The ROC curves for mild pre-eclampsia, severepre-eclampsia plus eclampsia and all pre-

Ž .eclamptic patients were constructed Fig. 3 toexpress the relationship between sensitivity andspecificity of VCAM-1 levels for different sub-groups of pre-eclampsia patients in relation to

Ždisease severity mild pre-eclampsia: area underthe curves0.61, S.E.s0.06; severe pre-eclampsiaplus eclampsia: area under the curve s0.87,S.E.s0.05; all pre-eclamptic patients: area under

.the curves0.76, S.E.s0.06 . The diagnostic valueof VCAM-1 was statistically significant in predict-

Žing severe pre-eclampsia and eclampsia Ps

Fig. 1. Scattergram display of distribution of VCAM-1 concentrations in relation to disease severity in pre-eclampsia. IndividualVCAM-1 levels are shown for 25 healthy pregnant women, 16 mild pre-eclampsia, 16 severe pre-eclampsia and three eclampsiapatients.


Fig. 2. VCAM-1 levels of patients in the hypertensive group in relation to diastolic blood pressure elevation.

. Ž .0.001 , but not mild pre-eclampsia P)0.05 .ROC curve analysis showed that the most optimalcut-off level for severe pre-eclampsia was 450

Ž .ngrml sensitivity s0.79, specificitys0.90 . Forpredicting severe disease, the test efficiency wascalculated as 86%, positive likelihood ratio as8.09 and negative likelihood ratio as 4.28.

4. Discussion

The mechanisms causing endothelial activationand dysfunction are poorly defined in pre-eclampsia. The extent to which leukocyte activa-tion and leukocyte-initiated damage are impor-tant in the pathophysiology of pre-eclampsia isnot well understood. Some studies have demon-strated that leukocyte activation is increased in

pre-eclampsia, and some markers used to monitorŽleukocyte activation e.g. defensin, lactoferrin, se-

. w xlectin were found to be elevated 5]8 .Soluble isoforms of vascular adhesion molecules

are found in the circulation. It has been proposedthat levels of circulating adhesion molecules maybe useful markers for disease activity and that

w xthey also may have physiological effects 16 . Ithas also been shown that adhesion molecules playkey roles in the evolution of some disorders char-acterized by vasculitis and chronic inflammation.Increased levels of VCAM-1 have been detectedin Wegener’s granulomatosis, systemic lupus ery-thematosus, rheumatoid arthritis, inflammatory

w xbowel disease and atherosclerosis 19]22 . Levelsof VCAM-1 were correlated to disease activity,being significantly higher during active comparedwith inactive disease and frequently normalizingwith clinical remission.


Fig. 3. Receiver-operating characteristic curves of sensitivity and specificity of severe pre-eclampsia plus eclampsiassevere diseaseŽ . Ž . Ž .curve 1 , mild and severe pre-eclampsia plus eclampsia curve 2 and mild pre-eclampsia curve 3 in relation to the reference curveŽ .curve 4 . The ROC curve for severe disease is clearly superior to other curves. This difference is particularly apparent in the areaof high specificity, where the curve is higher and more to the left than that of the other curves.

w xLyall et al. 9 were the first to show thatsoluble VCAM-1 was elevated in the serum ofpre-eclamptic patients. Later, they demonstratedthat increased cytokine levels were closely corre-

w xlated with elevated levels of VCAM-1 14 . Raynorw xet al. 24 studied VCAM-1 levels of 78 healthy

pregnant women between 5 and 39 weeks gesta-tion. They revealed that mean concentrations ofVCAM-1 were significantly higher before 20

Ž .weeks gestation 407"134 ngrml than at orŽ .after 20 weeks 346"111 ngrml . Recently,

w xAustgulen et al. 18 studied the levels of solubleŽadhesion molecules VCAM-1, ICAM-1 and E-

.Selectin in pre-eclamptic women, healthy preg-nant women and healthy non-pregnant women.They demonstrated that all three adhesionmolecules were significantly elevated in pre-eclamptic pregnancies, whereas serum levels innormal pregnancy and non-pregnant healthy indi-

viduals did not differ, and no differences wereobserved in relation to gestational age or labor.

w xLyall et al. 9 had also studied the same adhesionmolecules but they detected significant elevationsonly in VCAM-1 levels. Interestingly, when Haller

w xet al. 17 studied the effect of pre-eclampticpatients’ serum on the expression of adhesion

Ž .molecules ICAM-1, VCAM-1 on cultured en-dothelial cells, they detected a significant increasein ICAM-1 expression but not VCAM-1. Most

w xrecently, Kraus et al. 25 published their resultson ICAM-1, VCAM-1, E-selectin and PECAM-1Ž .platelet endothelial cell adhesion molecule-1levels in pre-eclampsia and healthy pregnantwomen. Consistent with the results of Austgulen

w xet al. 18 , they showed that concentrations of allfour soluble adhesion molecules were significantlyelevated in pre-eclampsia. Furthermore, Kraus et

w xal. 25 found significantly elevated levels of


VCAM-1 and ICAM-1 in the plasma of pregnantwomen who subsequently developed pre-eclampsia 3]15 weeks before the onset of clinicalsymptoms.

Our objective was to study VCAM-1 levels inpre-eclampsiareclampsia; and specifically to eval-uate VCAM-1 levels in relation to disease sever-ity. Consistent with previous reports, we observedsignificantly elevated levels of VCAM-1 in ourstudy population. However, when we subclassifiedthe hypertensive group into mild pre-eclampsia

Žand severe pre-eclampsia plus eclampsia severe.disease , we found that VCAM-1 levels increased

with disease severity. Although elevated levelswere detected in mild pre-eclampsia, the differ-ence was not significant, either a reflection of amore benign process or due to a small samplesize. However, serum concentrations of VCAM-1were significantly higher in severe pre-eclampsiaand eclampsia compared to mild pre-eclampsia orhealthy controls. VCAM-1 levels G450 ngrmlwere calculated to be the most optimal cut-offvalue in predicting severe disease in pre-eclampticpatients. These data suggest that VCAM-1 is morevaluable in predicting severe pre-eclampsia andeclampsia, but not mild pre-eclampsia, due tohigher values observed in these patients com-pared to mild pre-eclampsia patients. Previousstudies have documented high VCAM-1 levels inpre-eclampsia patients but have not specificallyanalyzed their results in relation to disease sever-ity. If we assume that the extent of endothelialactivation and damage is more pronounced insevere disease and the prognosis of mild pre-eclampsia is relatively benign compared to severepre-eclampsia and eclampsia, this is not a surpris-ing finding.

VCAM-1 levels reported in different studiesseem to differ in similar patient populations. Sincedifferent immuno-assays can yield different re-sults on the same samples due to the use ofdifferent antibodies and difference in the calibra-tion standards used, we do not give too muchweight to absolute values. Additionally, what ismeasured as freely available VCAM-1 in the cir-culation misses a significant portion of shedVCAM-1 that is unavailable due to adhesion to

w xreceptor-bearing cells 16 . Therefore it is most

appropriate to compare trends within studiesrather than absolute values among studies untilthere is assay standardization.

5. Conclusion

This study shows that circulating levels ofVCAM-1 are increased in pre-eclampsia. Serumconcentrations correlate with disease severity;VCAM-1 is significantly higher in severe pre-eclampsiareclampsia compared to mild pre-eclampsia or healthy pregnant women. The clini-cal value of measuring levels of soluble VCAM-1has not been established. However, recently per-formed studies and our observation suggest thatthis assay may be valuable in identifying patientsat risk for severe disease and also in the clinicalprediction of outcome. Additional prospectivestudies are warranted to capture the utility ofVCAM-1 assay.

Acknowledgements

This work was supported by the Research Fundof the University of Istanbul, Project NumberT-333r190397.

References

w x1 Redman CW. Current topic: pre-eclampsia and the pla-centa. Placenta 1997;12:301]308.

w x2 Friedman SA, Taylor RN, Roberts JM. Pathophysiologyof pre-eclampsia. Clin Perinatol 1991;18:661]682.

w x3 Roberts JM, Taylor RN, Musci TJ, Rodgers GM, HubelCA, McLaughlin MK. Pre-eclampsia: an endothelial celldisorder. Am J Obstet Gynecol 1989;161:1200]1204.

w x4 Harlan JM. Consequences of leukocyte-vessel wall inter-actions in inflammatory and immune reactions. SeminThromb Hemost 1987;13:434]444.

w x5 Greer IA, Dawes J, Johnston TA, Calder AA. Neu-trophil activation is confined to the maternal circulationin pregnancy-induced hypertension. Obstet Gynecol1991;78:28]32.

w x6 Greer IA, Haddad NG, Dawes J, Johnstone FD, CalderAA. Neutrophil activation in pregnancy-induced hyper-tension. Br J Obstet Gynaecol 1989;96:978]982.

w x7 Prieto JA, Panyutich AV, Heine RP. Neutrophil activa-tion in pre-eclampsia. Are defensins and lactoferrinelevated in pre-eclamptic patients? J Reprod Med1997;42:29]32.

w x8 Butterworth BH, Greer IA, Liston WA, Haddad NG,


Johnston TA. Immunocytochemical localization of neu-trophil elastase in term placenta decidua and my-ometrium in pregnancy-induced hypertension. Br JObstet Gynaecol 1991;98:929]933.

w x9 Lyall F, Greer IA, Boswell F, Macara LM, Walker JJ,Kingdom JC. The cell adhesion molecule, VCAM-1, isselectively elevated in serum in pre-eclampsia: does thisindicate the mechanism of leucocyte activation? Br JObstet Gynaecol 1994;101:485]487.

w x10 Osborn L, Hession C, Tizard R, Vassallo C, LuhowskyjS, Chi-Rosso G et al. Direct expression cloning of vascu-lar cell adhesion molecule 1, a cytokine-induced en-dothelial protein that binds to lymphocytes. Cell1989;59:1203]1211.

w x11 Frenette PS, Wagner DD. Adhesion molecules-Part 1. NEngl J Med 1996;334:1526]1529.

w x12 Reinhardt PH, Elliott JF, Kubes P. Neutrophils canadhere via alpha4beta1-integrin under flow conditions.Blood 1997;89:3837]3846.

w x13 Iademarco MF, Barks JL, Dean DC. Regulation ofvascular cell adhesion molecule-1 expression by IL-4and TNF-alpha in cultured endothelial cells. J ClinInvest 1995;95:264]271.

w x14 Greer IA, Lyall F, Perera T, Boswell F, Macara LM.Increased concentrations of cytokines interleukin-6 andinterleukin-1 receptor antagonist in plasma of womenwith pre-eclampsia: a mechanism for endothelial dys-function? Obstet Gynecol 1994;84:937]940.

w x15 Adams DH, Lloyd AR. Chemokines: leucocyte recruit-ment and activation cytokines. Lancet 1997;349:490]495.

w x16 Gearing AJH, Newman W. Circulating adhesionmolecules in disease. Immunol Today 1993;14:506]512.

w x17 Haller H, Ziegler EM, Homuth V et al. Endothelialadhesion molecules and leukocyte integrins in pre-eclamptic patients. Hypertension 1997;29:291]296.

w x18 Austgulen R, Lien E, Vince G, Redman CW. Increasedmaternal plasma levels of soluble adhesion moleculesŽ .ICAM-1, VCAM-1, E-selectin in pre-eclampsia. Eur JObstet Gynecol Reprod Biol 1997;71:53]58.

w x19 Wellicome SM, Kapahi P, Mason JC, Lebranchu Y,Yarwood H, Haskard DO. Detection of a circulatingform of vascular cell adhesion molecule-1: raised levelsin rheumatoid arthritis and systemic lupus erythemato-sus. Clin Exp Immunol 1993;92:412]418.

w x20 Janssen BA, Luqmani RA, Gordon C et al. Correlationof blood levels of soluble vascular cell adhesionmolecule-1 with disease activity in systemic lupus ery-thematosus and vasculitis. Br J Rheumatol 1994;33:1112]1116.

w x21 Jones SC, Banks RE, Haidar A et al. Adhesion moleculesin inflammatory bowel disease. Gut 1995;36:724]730.

w x22 Stegeman CA, Tervaert JW, Huitema MG, de Jong PE,Kallenberg CG. Serum levels of soluble adhesionmolecules intercellular adhesion molecule 1, vascularadhesion molecule 1, and E-selectin in patients withWegener’s granulomatosis. Relationship to disease activ-ity and relevance during follow-up. Arthritis Rheum1994;37:1228]1235.

w x23 ACOG. Hypertension in pregnancy. ACOG TechnicalBulletin, Number 219, January 1996.

w x24 Raynor BD, Parthasarathy S. Maternal serum vascularcell adhesion molecule concentration during pregnancy.J Society Gynecol Invest 1997;4:78]80.

w x25 Krauss T, Kuhn W, Lakoma C, Augustin HG. Circulat-ing endothelial cell adhesion molecules as diagnosticmarkers for the early identification of pregnant womenat risk for development of pre-eclampsia. Am J ObstetGynecol 1997;177:443]449.

Documents

vcam in preeclampsia.pdf