* Department of Pathology and Renal Division, Department of Medicine, and Renal, Vascular and Molecular Medicine Divisions, Departments of Medicine, Obstetrics & Gynecology, Beth Israel Deaconess Medical Center, and Harvard Medical School, Boston, Massachusetts
Correspondence: Dr. Isaac E. Stillman, Department of Pathology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215. Phone: 617-667-5959; Fax: 617-667-7120; E-mail: istillma{at}bidmc.harvard.edu
Preeclampsia is a pregnancy-specific disorder that complicatesapproximately 5% of all pregnancies, making it perhaps the mostcommon glomerular disease in the world. It is characterizedby new-onset hypertension and proteinuria, in association witha characteristic glomerular lesion, endotheliosis. "Glomerularendotheliosis" represents a specific variant of thrombotic microangiopathythat is characterized by glomerular endothelial swelling withloss of endothelial fenestrae and occlusion of the capillarylumens. Associated thrombosis is unusual. Recent evidence suggeststhat this unusual glomerular lesion is mediated by a solublevascular endothelial growth factor receptor that deprives glomerularendothelial cells of the vascular endothelial growth factorthat they require, leading to cellular injury and disruptionof the filtration apparatus with subsequent proteinuria. Thisreview summarizes the histologic changes and the pathogenesisof the glomerular lesions of preeclampsia.
A 30-yr-old pregnant woman (G1P0), at 15 wk gestation, presentedwith new-onset hypertension (160/100) and nephrotic-range proteinuria(3.7 g/d). Her medical history was significant for polycysticovary syndrome. This pregnancy was the result of in vitro fertilization,which was followed by fetal reduction, leaving her with twins.Her hypertension was unresponsive to labetalol, and she wasadmitted to the hospital for treatment. Complement levels werenormal, and serologies, including a thrombophilia workup, werenegative. Renal biopsy revealed 11 glomeruli, all of which showed"endotheliosis," the characteristic finding seen in preeclampsia(see discussion that follows and Figure 1, A and B). These findingswere confirmed by electron microscopy (Figure 1C). Immunofluorescencestudies were significant for scattered deposits of IgM, and light chains, and fibrin along the peripheral capillary loop.Attempts to control her BP using nifedipine and Aldomet werepartially successful, but her proteinuria continued to rise(peak urine protein/creatinine ratio 9). After much counseling,the patient, considering the severity of the disease and therelatively early gestational age, elected to terminate the pregnancy.Cytogenetic analysis revealed normal fetuses, and there wasno evidence of a molar pregnancy. Her renal function returnedto normal during the subsequent weeks and a second pregnancy,approximately 2 yr later, successfully went to term withoutthe development of preeclampsia.
Figure 1. Renal biopsy findings in a 30-yr-old with preeclampsia. The patient had twin pregnancy after in vitro fertilization. Biopsy was done at 15 wk gestational after new onset of hypertension and nephrotic-range proteinuria. Glomerular endotheliosis: (A) Glomerulus showing occlusion of capillary lumens by swelling of endocapillary cells. There is no appreciable increase in cellularity. Note nearby arterioles (arrows), which are unremarkable. No glomerular or arteriolar thrombosis is seen. (B) Arrow points to the only open capillary loop in this glomerular tuft with otherwise severe capillary occlusion. Note prominent epithelial cells (podocytes) that show protein reabsorption granules (arrowheads). (C) Marked cellular swelling involving mesangium and endothelial cells leading to loss of endothelial fenestrations and capillary lumen occlusion is seen. The lamina densa of the basement membrane is intact. Only focal electron lucent expansion of rara interna was seen (data not shown). Immunofluorescence studies showed no significant Ig deposition. Electron-dense material noted (arrows) is likely fibrin related. Podocytes, despite their protein reabsorption droplets, show relatively well-preserved foot processes. Magnifications: x40 in A (Masson Trichrome) and B (Giemsa-stained Epon-embedded section); x3200 in C (transmission electron microscopy).
Preeclampsia, the most frequently encountered renal complicationof pregnancy, is characterized by new-onset hypertension, proteinuria,and edema, usually developing after 20 wk of gestation.1 Whenseizures develop, it is known as eclampsia, a disease familiarto the pre-Hippocratic ancients. The presence of a placenta,with or without a fetus (hydatidiform mole), is necessary forits development. Consequently, definitive treatment is by deliveryof the placenta, which, depending on gestational age, can involvesignificant fetal morbidity and mortality. Preeclampsia complicatesapproximately 5% of all pregnancies and thus may be the mostcommon glomerular disease in the world.
The development of preeclampsia is believed to be a two-stageprocess: The first, asymptomatic stage is marked by abnormalplacentation, possibly related to ischemia. Placental elaborationof soluble factors that enter the maternal circulation follows,leading to endothelial dysfunction and the clinical syndrome(s).Recent work has identified circulating antiangiogenic substances,which seem to cause the disease by depriving the glomerularendothelium (and possibly other fenestrated endothelium) ofessential growth factors.2 Although preeclampsia is a multisystemprocess that affects vasculature throughout the body, this articlefocuses on its hallmark: Glomerular injury.
There is a tendency among caregivers to ascribe all new-onsetrenal disease in pregnancy to preeclampsia, in part becauseof reluctance to perform a biopsy. However, many studies havedocumented the inaccuracy of clinical diagnosis alone. One foundthat when diagnosed clinically, before 37 wk, 67% of women whounderwent biopsy had a renal disease other than preeclampsia.3Therefore, biopsy remains the most reliable way of making thediagnosis and is particularly helpful in multiparas or earlyin the pregnancy, to exclude other entities for which effectivetherapies are available.
The renal biopsy findings of preeclampsia are best appreciatedin the context of the pathologic patterns seen in thromboticmicroangiopathies (TMA). TMA is a term used to describe a groupof clinically diverse entities, such as hemolytic uremic syndromeand malignant hypertension, among others, that are defined bya primary locus of injury—the endothelium—and theensuing thrombosis and vascular injury. Their similar pathologicexpression has led nephropathologists to adopt the term TMAto describe them all. The lesions of preeclampsia share somesimilarities with and intriguing differences from those of nonpreeclampticTMA, likely owing to their differing pathogenesis.
Although the central role of the kidney was first recognizedin 1918, it took several decades, in particular, the applicationof ultrastructural analysis to renal biopsy material, to elucidatethe characteristic lesions, which are similar for both preeclampsiaand eclampsia. Preeclampsia is associated with a distinctiveglomerular appearance: "Glomerular endotheliosis," a term coinedby Spargo et al.4 The glomeruli are enlarged and solidified("bloodless"), as a result of narrowed or occluded capillarylumens that are the result of swelling of the native endothelialcells and, to a lesser extent, mesangial cells (Figure 1). Glomerularvolume is increased and correlates with the severity of thedisease.5 The degree of endotheliosis can vary between glomeruli,although most show at least some involvement. Glomerular cellularityis not significantly increased. It is interesting that the endothelialchanges are limited to the glomerular capillaries; arteriolesare typically unaffected. Thrombosis by light microscopy isdecidedly unusual, although fibrin can be detected by immunofluorescencein glomeruli. In marked contrast, in nonpreeclamptic TMA, thrombosisof vessels and/or glomeruli is a central finding. Cases of severepreeclampsia with accompanying vascular thrombosis often haveclinical signs suggesting a superimposed nonpreeclamptic TMA.Coexistent diseases that are associated with endothelial dysfunction,such as diabetes and antiphospholipid antibodies, are also knownto increase the risk for preeclampsia. These observations underscorethe variable and site-specific phenotype of endothelial cellsand suggest multiple and possibly overlapping pathways leadingto endothelial injury in both preeclampsia and nonpreeclampticTMA. As detailed next, the acute endothelial swelling seen inpreeclampsia is due to vascular endothelial growth factor (VEGF)deprivation. In contrast, the endothelial injury noted in nonpreeclampticTMA, although poorly understood and probably multifactorial,is likely not related to impairment of VEGF signaling. FreeVEGF levels are higher in patients with hemolytic uremic syndrome/thromboticthrombocytopenic purpura.6
In severe cases of preeclampsia, in particular as the lesionsevolve/resolve, mesangial interposition can be seen, a findingshared with other entities resulting from chronic endothelialinsult, such as "chronic" TMA or transplant glomerulopathy.Other changes, such as prominent podocytes with protein resorptiondroplets and endocapillary foam cells, are probably secondaryto the proteinuria. The presence of arteriosclerosis suggestsa coexisting process, such as "essential" hypertension. Whetherpreeclampsia leads to chronic vascular injury over ensuing yearsis not yet clear.
The immunofluorescence findings are somewhat variable with fibrindeposition often being a prominent feature. The low-level glomerularIg deposition in severe preeclampsia, reported by some, probablyrepresents nonimmunologic insudation. This conclusion is supportedby the ultrastructural observation that electron-dense depositsare inconspicuous. Its chief diagnostic role lies in excludingan immune complex glomerulonephritis, such as lupus nephritis,which often flares during pregnancy. The pathogenetic role thatfibrin and its related products play has not been resolved.Although the degree of deposition has been reported to varywidely, it seems to be more common in renal biopsies obtainedfrom patients with premature and severe preeclampsia.
Ultrastructural analysis is the definitive way to demonstrateendotheliosis and, in some cases, may be required to make thediagnosis. Endothelial cells demonstrate loss of fenestrationswith cytoplasmic swelling, owing to fluid and lipid accumulationand capillary occlusion (Figure 1C).7 Mesangial cells may showsimilar changes. In contrast to most other TMA, electron lucentexpansion of the subendothelial zone, when present, is usuallynot prominent. It is interesting that despite significant proteinuria,podocytes show limited foot process effacement, a phenomenonthat may also be seen with other TMA, particularly in the acutephase.8 Indeed, when quantified, the filtration slit frequencyis not significantly reduced in preeclampsia below controls.7This finding has significant implications for the investigationof mechanisms of proteinuria in general because it suggeststhat nephrotic-range proteinuria can occur without significantfusion of podocyte foot processes.9
Endotheliosis seems to be responsible for the decreased GFRnoted in preeclampsia, primarily through reduction in the ultrafiltrationcoefficient as opposed to diminished plasma flow. When focal,endotheliosis can be difficult to identify, and its specificityfor preeclampsia may then be limited. Mild forms have been seenin up to 30% of patients with pregnancy-induced hypertensionwithout proteinuria.10,11 Furthermore, a recent study foundfive of 12 control subjects (nonhypertensive third-trimesterwomen) with trace endotheliosis.10,11 As noted by its authors,this study suggests a continuum between healthy pregnant womenand the extreme of preeclampsia. Recent work, as discussed next,provides a rationale for this phenomenon. Limited endotheliosishas also been reported occasionally in association with otherdisorders.12 Nevertheless, when endotheliosis is present ina diffuse manner, in the appropriate clinical setting, it isvirtually pathognomonic for preeclampsia.
After delivery, the glomerular changes usually reverse rapidly,coinciding with resolution of the hypertension and proteinuria.However, the relationship among preeclampsia, underlying renaland other conditions, and future disease, including hypertension,is complex and controversial. For example, FSGS (a nonspecificform of glomerular scarring that can be seen in associationwith "essential" hypertension, as well as in primary glomerulardisease) is said to accompany endotheliosis in a significantpercentage of cases, but it is not necessarily predictive ofcurrent or future renal failure, as might otherwise be expected.13Indeed, there is evidence that the segmental sclerosis of preeclampsiamay be reversible. Nevertheless, considerable evidence suggeststhat preeclampsia predisposes women to late cardiovascular diseases.14The increased risk for hypertension is not seen in their siblings,suggesting that it is related to preeclampsia and pointing tothe role of subtle endothelial injury leading to the developmentof chronic hypertension.15
The search for a circulating factor that causes the hypertensionand proteinuria of preeclampsia has been an area of intenseinvestigation. It is believed that excess circulating antiangiogenicsubstances such as soluble fms-like tyrosine kinase (sFlt1,also referred to as sVEGFR1) play a prominent role in the developmentof preeclampsia. VEGF, synthesized constitutively in the glomerulusby podocytes, is a critical factor for the maintenance of endothelialhealth, including the induction of fenestrae. Indeed, geneticglomerular VEGF deficiency has been shown to result in endotheliosiswith loss of fenestrae.16 sFlt1 is a secreted protein that lacksthe transmembrane and cytoplasmic domain of the membrane-boundVEGF receptor and acts as an endogenous inhibitor of VEGF signaling.Circulating levels of sFlt1, made predominantly by the placenta,are greatly increased in women with established preeclampsia,even before onset of clinical symptoms.17 That study also founda steady increase in serum sFlt1 levels in normotensive near-termwomen, a finding that suggests that preeclampsia representsan early and exaggerated form of normal pregnancy and helpsto explain the mild endotheliosis occasionally seen in near-termnormotensive biopsies. When administered to rats, sFlt1 producesa clinical syndrome and glomerular lesions resembling humanpreeclampsia.2 Similar observations have been noted when otherVEGF inhibitors such as neutralizing antibodies have been usedin rodents or humans.18–20
How VEGF-deficient states such as preeclampsia produce proteinuriais still unknown. Some have suggested that loss of podocytenephrin expression may be responsible.19,21 However, whetherthe diminished nephrin expression is the cause or the consequenceof proteinuria is unknown. Others have suggested that all threelayers of the glomerular wall—endothelium, basement membrane,and slit diaphragm—may jointly constitute the barrieragainst proteinuria. Deen et al.22 argued that proteinuria canoccur with endothelial disruption alone, which may explain thesignificant proteinuria noted with endotheliosis. More recently,yet another antiangiogenic protein, soluble endoglin, was reportedto play a pathogenic role in preeclampsia.23 It is interestingthat animals that were exposed to high levels of soluble endoglinhad focal endotheliosis without significant proteinuria, whereasthose that were exposed to both soluble endoglin and sFlt1 developedmassive proteinuria and severe endotheliosis. Future work isnecessary to clarify the mechanisms of the endotheliosis andproteinuria mediated by these circulating antiangiogenic substances.
S.A.K. is listed as a co-inventor on multiple patents filedby the Beth Israel Deaconess Medical Center for the use of angiogenicproteins for the diagnosis and therapy of preeclampsia and isa consultant to Johnson & Johnson, Beckman Coulter, andAbbott Diagnostics.
Acknowledgments
This work is supported by R01 grants from the NIDDK (DK 065997)and the NHLBI (HL079594).
We thank Drs. Frank Epstein and Seymour Rosen for helpful discussions.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Karumanchi SA, Maynard SE, Stillman IE, Epstein FH, Sukhatme VP: Preeclampsia: A renal perspective.
Kidney Int 67
: 2101
–2113, 2005[CrossRef][Medline]
Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, Libermann TA, Morgan JP, Sellke FW, Stillman IE, Epstein FH, Sukhatme VP, Karumanchi SA: Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia.
J Clin Invest 111
: 649
–658, 2003[CrossRef][Medline]
Ihle BU, Long P, Oats J: Early onset pre-eclampsia: Recognition of underlying renal disease.
BMJ (Clin Res Ed) 294
: 79
–81, 1987[Medline]
Spargo B: The renal lesions in preeclampsia. In:
Hypertension in Pregnancy, edited by Lindheimer M, New York, Wiley Medical, 1976
, pp 129
–137
Nochy BP, Hinglais N, Freund M, Idatte M, Jacquot C, Chartier M, Bariety J: Renal lesions in the hypertensive syndromes of pregnancy: Immunomorphological and ultrastructural studies in 114 cases.
Clin Nephrol 13
: 155
–162, 1980[Medline]
Maroeska Te Loo D, Bosma N, Van Hinsbergh V, Span P, De Waal R, Clarijs R, Sweep C, Monnens L, Van Den Heuvel L: Elevated levels of vascular endothelial growth factor in serum of patients with D+ HUS.
Pediatr Nephrol 19
: 754
–760, 2004[CrossRef][Medline]
Lafayette RA, Druzin M, Sibley R, Derby G, Malik T, Huie P, Polhemus C, Deen WM, Myers BD: Nature of glomerular dysfunction in pre-eclampsia.
Kidney Int 54
: 1240
–1249, 1998[CrossRef][Medline]
Mautner W, Churg J, Grishmann E, Dachs S: Preeclamptic nephropathy. An electron microscopic study.
Lab Invest 11
: 518
–530, 1962[Medline]
Karumanchi SA, Epstein FH, Stillman IE: Is loss of podocyte foot processes necessary for the induction of proteinuria?
Am J Kidney Dis 45
: 436
, 2005[CrossRef]
Fisher KA, Luger A, Spargo BH, Lindheimer MD: Hypertension in pregnancy: Clinical-pathological correlations and remote prognosis.
Medicine (Baltimore) 60
: 267
–276, 1981[Medline]
Strevens H, Wide-Swensson D, Hansen A, Horn T, Ingemarsson I, Larsen S, Willner J, Olsen S: Glomerular endotheliosis in normal pregnancy and pre-eclampsia.
BJOG 110
: 831
–836, 2003[Medline]
Subramanya A, Houghton D, Watnick S: Steroid-responsive idiopathic glomerular capillary endotheliosis: Case report and literature review.
Am J Kidney Dis 45
: 1090
–1095, 2005[CrossRef][Medline]
Nagai Y, Arai H, Washizawa Y, Ger Y, Tanaka M, Maeda M, Kawamura S: FSGS-like lesions in pre-eclampsia.
Clin Nephrol 36
: 134
–140, 1991[Medline]
Sibai BM, el-Nazer A, Gonzalez-Ruiz A: Severe preeclampsia-eclampsia in young primigravid women: Subsequent pregnancy outcome and remote prognosis.
Am J Obstet Gynecol 155
: 1011
–1016, 1986[Medline]
Epstein FH: Pregnancy and renal disease.
N Engl J Med 335
: 277
–278, 1966[CrossRef]
Eremina V, Sood M, Haigh J, Nagy A, Lajoie G, Ferrara N, Gerber HP, Kikkawa Y, Miner JH, Quaggin SE: Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases.
J Clin Invest 111
: 707
–716, 2003[CrossRef][Medline]
Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, Schisterman EF, Thadhani R, Sachs BP, Epstein FH, Sibai BM, Sukhatme VP, Karumanchi SA: Circulating angiogenic factors and the risk of preeclampsia.
N Engl J Med 350
: 672
–683, 2004[Abstract/Free Full Text]
Kitamoto Y, Takeya M, Tokunaga H, Tomita K: Glomerular endothelial cells are maintained by vascular endothelial growth factor in the adult kidney.
Tohoku J Exp Med 195
: 43
–54, 2001[CrossRef][Medline]
Sugimoto H, Hamano Y, Charytan D, Cosgrove D, Kieran M, Sudhakar A, Kalluri R: Neutralization of circulating vascular endothelial growth factor (VEGF) by anti-VEGF antibodies and soluble VEGF receptor 1 (sFlt-1) induces proteinuria.
J Biol Chem 278
: 12605
–12608, 2003[Abstract/Free Full Text]
Zhu X, Wu S, Dahut WL, Parikh CR: Risks of proteinuria and hypertension with bevacizumab, an antibody against vascular endothelial growth factor: Systematic review and meta-analysis.
Am J Kidney Dis 49
: 186
–193, 2007[CrossRef][Medline]
Garovic VD, Wagner SJ, Petrovic LM, Gray CE, Hall P, Sugimoto H, Kalluri R, Grande JP: Glomerular expression of nephrin and synaptopodin, but not podocin, is decreased in kidney sections from women with preeclampsia.
Nephrol Dial Transplant 22
: 1136
–1143, 2007[Abstract/Free Full Text]
Deen WM: What determines glomerular capillary permeability?
J Clin Invest 114
: 1412
–1414, 2004[CrossRef][Medline]
Venkatesha S, Toporsian M, Lam C, Hanai JI, Mammoto T, Kim YM, Bdolah Y, Lim KH, Yuan HT, Libermann TA, Stillman IE, Roberts D, D'Amore PA, Epstein FH, Sellke FW, Romero R, Sukhatme VP, Letarte M, Karumanchi SA: Soluble endoglin contributes to the pathogenesis of preeclampsia.
Nat Med 12
: 642
–649, 2006[CrossRef][Medline]
Roncone D, Satoskar A, Nadasdy T, Monk JP, Rovin BH: Proteinuria in a patient receiving anti-VEGF therapy for metastatic renal cell carcinoma.
Nat Clin Pract Nephrol 3
: 287
–293, 2007[CrossRef][Medline]
Izzedine H, Brocheriou I, Deray G, Rixe O: Thrombotic microangiopathy and anti-VEGF agents.
Nephrol Dial Transplant 22
: 1481
–1482, 2007[Free Full Text]
and 
light chains, and fibrin along the peripheral capillary loop. Attempts to control her BP using nifedipine and Aldomet were partially successful, but her proteinuria continued to rise (peak urine protein/creatinine ratio 9). After much counseling, the patient, considering the severity of the disease and the relatively early gestational age, elected to terminate the pregnancy. Cytogenetic analysis revealed normal fetuses, and there was no evidence of a molar pregnancy. Her renal function returned to normal during the subsequent weeks and a second pregnancy, approximately 2 yr later, successfully went to term without the development of preeclampsia. 
Top
Abstract
RENAL BIOPSY CASE PRESENTATION