- Email: [email protected]
Pregnancy-Induced Nephropathy: The Significance of Focal Segmental Glomerulosclerosis
Pregnancy-Induced Nephropathy: The Significance of Focal Segmental Glomerulosclerosis Lillian W. Gaber, MD, and Benjamin H. Spargo, MD • The morphologic alterations of true preeclampsia are well described and distinct. Using available criteria to define reversible pregnancy-induced nephropathy (PIN), the pathologist can offer the clinician information useful in predicting remote renal function and hypertension. True PIN is a usually completely reversible lesion in the nullipara and carries little risk of future hypertension. Nephrosclerotic vascular lesions, however, whether arteriolar, interlobular, or glomerular and resembling focal segmental glomerulosclerosis (FSGS), signify underlying hypertensive disease. This review summarizes information concerning the nature of PIN and also surveys 20 biopsies in women who mainly had severe preeclampsia, often with persistent postpartum hypertension. This material was step-sectioned and specifically reviewed in an effort to find FSGS in women with PIN. © 1987 by the National Kidney Foundation, Inc. INDEX WORDS: Pregnancy-induced hypertension; preeclampsia; focal segmental glomerulosclerosis.
F
OR SEVERAL decades, investigators have been disputing (with considerable frustration) just what constitutes the renal lesion of pregnancyinduced nephropathy (PIN) as well as its pathogenesis. One reason for the controversy surrounding the precise pathological nature of these lesions is the scarcity of data in which adequate clinicopathologic correlates can be made. This is mainly because preeclampsia can be superimposed on an undiagnosed essential hypertension or any of a variety of renal diseases, including glomerulonephritis, membranous nephropathy, epithelial cell disease, tubulointerstitial nephritis, and lesions associated with systemic diseases such as sickle cell and diabetic nephropathy. At the University of Chicago, we have combined case histories and epidemiologic follow-up studies to unravel the clinical dilemmas facing physicians who manage gravidas with hypertensive diseases. 1-3 Glomerular pathology plays a central role in understanding the natural history of these disorders. HISTORICAL BACKGROUND
Historically, interpretation of the morphologic changes that occur in PIN has been observer and methodology dependent. (The historical literature in this section is cited in ref. 4.) Lohlein (1918) is credited with recognizing the glomerulus as the principal site for the pathologic changes in preeclampsia. He interpreted the apparent glomerular hypercellularity as representative of a glomerulonephritis. Farr (1920), Bell (1932), and Allen (1951) considered basement membrane thickening the main pathologic process responsible for glo-
merular enlargement and capillary loop obstruction. They ascribed the lesion to a form of a degenerative glomerulonephritis or membranous nephropathy. This concept was challenged by McManus (1950) and Jones (1951); both demonstrated only edema of the basement membrane and the glomerular tuft. Sheehan (1950) showed that deposition of fibrils along the basement membrane and between cells was largely responsible for the glomerular changes. Pollak (1956) demonstrated increased mucopolysaccharide deposition in the basement membrane. Dieckmann and Potter (1957) used renal biopsy material to study the nature of the renal lesions and were able to provide data on the acute and reversible nature of the lesions. Dieckmann acknowledged the great difficulty in attaining a reproducible clinicopathologic correlation in cases of gestational hypertension. With the availability of electron microscopy and its general use in renal pathology, understanding of more detailed pathologic processes became possible. Spargo and co-workers reported a consistent glomerular capillary endothelial cell change characterized by obstructive swelling of the enFrom the Department of Pathology, The University of Chicago. Dr Spargo is the recipient of research career award No. K6HLA418 from the National Institutes of Health, Bethesda, MD. Address reprint requests to Benjamin H. Spargo, MD, Department of Pathology, The University of Chicago, 5841 S Maryland Ave, Chicago, IL 60637. © 1987 by the National Kidney Foundation, Inc. 0272-6386/87/0904-0014$3.00/0
American Journal of Kidney Diseases, Vol IX, No 4 (April), 1987: pp 317-323
317
GABER AND SPARGO
318
dothelial cell, vacuolation, intracellular lipid accumulation, and hypertrophy of the cytoplasmic organelles.! They. identified glomerular endotheliosis as the basic pathologic process and further indicated the noninflammatory nature of the lesion and its specificity to preeclamptic nephropathy. Recently, Sheehan and Taylor and Spargo reviewed their experience and provided detailed elaborate descriptions of the renal lesions in preeclamptic patients. 4.5 At the University of Chicago we have accumulated > 200 postpartum renal biopsies with documented clinical follow-up. Examination of this material, which contains a large spectrum of morphologic alterations with varying degrees of severity, has led us to a postulate of different pathologic findings with specific remote outcomes. The importance of morphologic data is exemplified in our results through 1976, in which 45 % of women whose gestational hypertension was diagnosed clinically as preeclampsia alone had their diagnosis altered based on pathologic examination of biopsies. 5 The discrepancy between clinical opinions and the pathologist's diagnosis was most pronounced in the multiparous group. Another feature of our material was that the lesion of pure preeclampsia was completely reversible. Recently, however, several investigators have linked development of focal segmental glomerulosclerosis (FSGS) with preeclampsia, an association which, if true, should alter the benign prognostic implication of pure preeclampsia (PIN). 6·8 To investigate this association, we reviewed a series of 20 postpartum biopsies in women who had hypertension in pregnancy. METHODS Twenty renal biopsies performed during the years 1979 through 1983 were retrospectively selected in random fashion from our files. This study includes 19 women (one patient was biopsied after two successive gestations) and differs from a series o f 176 cases spanning 1958 through 1976 and published by us in 1981. 2 In essence, selected criteria had changed in the late 1970s, and renal biopsies were performed only on "atypical" and "severe" cases; the 1958 through 1976 group represented a more heterogeneous population. In the present series, biopsies were obtained on the average eight (range four to 21) days postpartum and were performed primarily because of hypertension and/or proteinuria persisting in the puerperium or severe preeclampsia presenting in midpregnancy or early in the third trimester (Table I). Mean age at biopsy was 22.5 years (range 15 to 36 years) and only 50% of biopsies were performed in nulliparas. The material we report was specifically reviewed in an at-
tempt to locate lesions that were not a feature of our previously published observations but that had been underscored as important preeclamptic renal lesions by others. Thus, we step-sectioned the biopsies and reviewed as many as 20 glomeruli in each patient, specifically attempting to document FSGS as well as interposition of mesangial cytoplasm with the basement membrane ("double contour") appearance.
Pathology Light and electron microscopy. The renal biopsies demonstrated the morphologic features of preeclampsia, with some variation in intensity of the different components (Table 1). Enlargement of the glomerular tuft, variable hypercellularity, decreased vascularity, and obstruction of the peripheral capillary loops were common features (Figs 1 and 2). Herniation of the glomerular tuft into the proximal tubules and longitudinally collapsed cigar-shaped lobules were common. Two cases disclosed partial glomerular infarction and reactive epithelial crescents. The mesangial and endothelial cells were hypertrophied, swollen, and vacuolated, with occasional foam cells. Electron microscopy disclosed intracytoplasmic lipid vacuoles, myelinelike figures and, rarely, cholesterol crystals (Fig 3). Interposition of mesangium under the endothelial cells was occasionally noted; in contradistinction to the experience of other investigators, this was not a feature of our material. Epithelial cells were prominent; despite proteinuria in all cases, however, only two cases showed focal effacement of the foot processes. Seven biopsies (35 %) showed some evidence of focal segmental sclerosis. The sclerotic areas usually showed expansion of the stalk and collapse of a few capillary loops (Fig 4). Two biopsies exhibited globally sclerosed glomeruli, and adhesions to Bowman's capsule and cellular crescents were noted in three instances. Vascular lesions in the form of arteriosclerosis existed in ten biopsies, including five with focal sclerosis. Intimal widening, fibrosis , and reduplication of the elastic lamina were focal and often limited to very few arteries (Fig 2). No arteriolar lesions were observed. Whenever arteriosclerosis existed, focal tubular loss and condensation of the interstitium were observed. Immunohistochemistry. Observations were nonspecific and staining, when present, related to nonspecific trapping of immunoglobulins and fibrin secondary to the pronounced glomerular ischemic injury. There was no segmental localization of immunocomplex deposits.
20 30 18 19 22 21 18 25
18
15 18
23 36
17
7 8 9' 10' 11 12 13 14
15
16 17
18 19
20
N
M N
M M
N
39
27 34
28 32
37
29 33 19 24 22 33 32 28
28 34 34
32
Severe preeclampsia (HELLP syndrome)persistent postpartum hypertension Early onset of severe preeclampsia with HELLP syndrome Hypertension , with marked nephrotic proteinuria Early onset of severe preeclampsia Persistent postpartum hypertension Eclampsia and persistent postpartum hypertension Early onset of preeclampsia Persistent postpartum hypertension Early onset of preeclampsia and eclampsia Early onset of severe preeclampsia Early onset of severe preeclampsia Persistent postpartum hypertension Severe preeclampsia Early onset of nephrotic syndrome and hypertension Postpartum hypertension and seizures in a diabetic patient Early onset preeclampsia Severe hypertension and nephrotic proteinuria Early onset of severe preeclampsia Persistent postpartum hypertension and nephrotic range proteinuria Persistent postpartum hypertension
Indication for Renal Biopsy
7
6 4
6 5
9
6 21 5 8 10 6 5 7
7 11 7
10
11
15
20
20 36
13 20
30
37 4 66 113 17 19 5 50
29 33 20
26
17
3
Biopsy Performed Total No. (Postpartum day) of Glomeruli
2%
7% 4%
8%
+
+++ 10% ++ 10%
++ ++
+ 8%
3%
8%
6%
10%
20%
9% 4%
+
+
+
+ +
+ + +
+
+
+
+ +
+ +
NA
+
NA
++ +
++
++
Global Cellular Tubuloinstertial Vascular FSGS Sclerosis Crescents Changes Lesions
++ 10% +++
++ ++ +++ ++ +
++ ++ ++
++
++
+
PIN
Clinicopathologic Correlations In 20 Women With Preeclamptic Gestations
Abbreviations: N, nullipara; M, multipara; HELLP syndrome, hemolysis, elevated liver enzymes and low platelet counts; PIN, pregnancy-induced nephropathy (glomerular capillary swelling; +, + +, + + + = mild , moderate , and severe, respectively); FSGS, focal segmental glomerulosclerosis; NA, inadequate material for diagnosis. ' No.9 and 10 are the same patient biopsied after successive hypertensive gestations.
8
W
8
8 81
8
8 8 8 8 8 8
W
8
N N N M M N N N
M N M
8 8
32 19 19
4 5 6
W
M
8
26
3
25
M
W
27
2
32
Onset of Hypertension (Gestational Week)
M
8
28
Patients Age Race Parity
Table 1.
"TI
0 :> r
~
»
en
-0
~
:>
m m () r
JJ
-0
Z
en 1ii
0
JJ
r m
()
en
r 0
JJ C
r 0 ~ m
(j)
()
320
GABER AND SPARGO
Fig 1. Micrograph of a slightly enlarged glomerulus demonstrating focal loss of vascularity and endothelial cell swelling. (Hematoxylin and eosin, x 400.)
DISCUSSION
Postpartum renal biopsy is an important aid in the prediction of long-term morbidity and eventual outcome in patients with gestational hypertension. Use of pathologic criteria separates those patients with overlapping clinical syndromes into groups with a more predictable course. Glomerular endotheliolsis is a completely reversible change. Detection of vascular sclerosis on the other hand appears to be an indication of a hypertensive diathesis that may not have been clinically manifested prior to pregnancy. The significance of this distinction is best appreciated when clinical and
Fig 2. Micrograph shows marked variations In the degree of obstruction of the capillary loops. Example of intimal widening and fibrosis of a small artery associated with mild tubular dropout. (Hematoxylin and eosin, x 312.)
epidemiologic studies based on biopsy results are evaluated . In our first study at the University of Chicago, 83% of patients with pure "glomerular endotheliosis" were primigravidas, whereas in multigravidas the preeclamptic lesion was superimposed on nephrosclerosis in about half of the cases. ~ Follow-up of these patients disclosed a similar incidence of hypertension (9.4%) in the pure preeclamptic group when compared with agesex-race adjusted statistics from a survey by the National Institutes of Health. Patients with nephrosclerotic lesions, on the other hand , had a much higher incidence of permanent hypertension
FOCAL GLOMERULOSCLEROSIS IN PREECLAMPSIA
321
Fig 3. Electron micrograph from a biopsy demonstrating both pregnancy-induced nephropathy and focal segmental glomerular sclerosis. Area shown is away from the sclerotic lobule and demonstrates marked cytoplasmic hypertrophy and vacuolation of the endocapillary cells and Interposition of mesangium resulting in obstruction of the capillary loops. Epithelial foot processes are mainly intact. (Original magnification, x 5,400).
(74 %) that did not become manifest in most cases until several years after the index pregnancy. The nature of the vascular lesion, whether it is a primary aging process or a secondary hypertensive change, has been a subiect of debate. We can-
Fig 4. Example of focal segmental sclerosis in a bloodless glomerulus. (Periodic acid-Schiff, x 312.)
not subscribe to the former view because the mean age in the present series of patients with nephrosclerosis is 22.5 years. Biopsy interpretation, however, can be limited by inadequate sampling of vessels and the focal
322
GABER AND SPARGO
Fig 5. Micrograph demonstrating a small reactive area of epithelial cells in a partially bloodless glomerulus. (Hematoxylin and eosin, x 312.)
nature of the vascular lesions. In both situations, we stress that the pathologist should diligently seek evidence of secondary more chronic ischemic changes, such as tubular atrophy, tubular loss, and condensation of the interstitium, to qualify a prognostication based on the available specimen. In reviewing our material , we specifically searched for FSGS in preeclampsia in an attempt to derive its clinical significance. In this group of women with severe disease, seven of 20 (35 %) biopsies reviewed disclosed some evidence of FSGS involving very few glomeruli. This lesion is very much similar to the focal fibrosis and lobular matting described by Sheehan and Lynch in 1973. 9 An intriguing finding is the presence of arteriosclerosis in five of the seven cases (71 %) and only tubular loss and interstitial condensation in the other two cases. This persistent occurrence of vascular lesions when FSGS exists classifies this group as hypertensive nephrosclerosis , and imparts its prognostic implications. This focal segmental sclerosis appears to be of ischemic nature, induced by the ongoing indolent hypertensive process, perhaps compounded further by the renal hemodynamic alterations that occur during pregnancy. This form of focal sclerosis is not to be confused with primary FSGS. The latter is characterized by a progressive unremittent course with a survival rate of 75 % at 5 years and 38 % at 15 years. 10 Primary FSGS has been reported in 22 % of pregnancies with glomerular diseases excluding "pure" preeclampsia; the authors claim that 52 % of these
patients had onset of FSGS during pregnancy. liOn the other hand, the form of segmental sclerosis that we described in relation to preeclampsia superimposed on nephrosclerosis does not demonstrate the ultrastructural or the immunologic criteria of primary FSGS. Focal segmental glomerulosclerosis is not a feature of pure preeclampsia, but rather an indicator of underlying nephrosclerosis . Once diagnosed , vascular lesions should be considered and sought because long-term prognosis in these patients is probably similar to those with nephrosclerosis and superimposed preeclampsia. ACKNOWLEDGMENT We thank Penny Theodoropoulos for h erexcellent secretarial help.
REFERENCES I. Spargo BH , McCartney Cp, Winemiller R: Glomerular
capillary endotheliosis in toxemia of pregnancy. Arch Pathol 68 :593-599, 1959 2 . Fisher KA. Luger A. Spargo BH , et al : Hypertension in pregnancy: Clinical-pathological correlations and remote prognosis. Medicine 60:267-276. 1984 3. Lindheimer MD, Chesley LC, Taylor JR, et al: Renal function and morphology in the hypertensive disorders of pregnancy (in press) 4 . Sheehan HL: Renal morphology in preeclampsia. Kidney 1nt 18:241-252 , 1980 5 . Taylor JR. Spargo BH: Pathology of the kidney in preeclampsia, in Andreucci VE (ed): The Kidney in Pregnancy. Boston , Nijhoff. 1986 , pp 47-63 6 . Nochy D. Gaudry C, Hinglais N , et al : Can focal segmental glomerulosclerosis appear in preeclampsia? Adv Nephrol 15:71-85, 1986
FOCAL GLOMERULOSCLEROSIS IN PREECLAMPSIA
323
7. Nochy 0, Nihglais N, Jacquot C, et al: De novo focal glomerulosclerosis in preeclampsia. Clin Nephrol 25:71-85, 1986 8. Kida H, Takeda S, Yokoyama H, et al: Focal glomerular sclerosis in preeclampsia. Clin Nephrol 24:221-227, 1985 9. Sheehan HL, Lynch JB: Pathology of Toxemia of Pregnancy. Baltimore. Williams & Wilkins . 1973
10. Cameron JS, Turner DR, Ogg CS, et al: The long term prognosis of patients with focal segmental glomerulosclerosis. Clin Nephrol 10:211-218, 1978 II. Surian M, Imbasciati E. Casi P, et al: Glomerular disease and pregnancy: A study of 123 pregnancies in patients with primary and secondary glomerular disease. Nephron 36: 101-105.1984
F
OR SEVERAL decades, investigators have been disputing (with considerable frustration) just what constitutes the renal lesion of pregnancyinduced nephropathy (PIN) as well as its pathogenesis. One reason for the controversy surrounding the precise pathological nature of these lesions is the scarcity of data in which adequate clinicopathologic correlates can be made. This is mainly because preeclampsia can be superimposed on an undiagnosed essential hypertension or any of a variety of renal diseases, including glomerulonephritis, membranous nephropathy, epithelial cell disease, tubulointerstitial nephritis, and lesions associated with systemic diseases such as sickle cell and diabetic nephropathy. At the University of Chicago, we have combined case histories and epidemiologic follow-up studies to unravel the clinical dilemmas facing physicians who manage gravidas with hypertensive diseases. 1-3 Glomerular pathology plays a central role in understanding the natural history of these disorders. HISTORICAL BACKGROUND
Historically, interpretation of the morphologic changes that occur in PIN has been observer and methodology dependent. (The historical literature in this section is cited in ref. 4.) Lohlein (1918) is credited with recognizing the glomerulus as the principal site for the pathologic changes in preeclampsia. He interpreted the apparent glomerular hypercellularity as representative of a glomerulonephritis. Farr (1920), Bell (1932), and Allen (1951) considered basement membrane thickening the main pathologic process responsible for glo-
merular enlargement and capillary loop obstruction. They ascribed the lesion to a form of a degenerative glomerulonephritis or membranous nephropathy. This concept was challenged by McManus (1950) and Jones (1951); both demonstrated only edema of the basement membrane and the glomerular tuft. Sheehan (1950) showed that deposition of fibrils along the basement membrane and between cells was largely responsible for the glomerular changes. Pollak (1956) demonstrated increased mucopolysaccharide deposition in the basement membrane. Dieckmann and Potter (1957) used renal biopsy material to study the nature of the renal lesions and were able to provide data on the acute and reversible nature of the lesions. Dieckmann acknowledged the great difficulty in attaining a reproducible clinicopathologic correlation in cases of gestational hypertension. With the availability of electron microscopy and its general use in renal pathology, understanding of more detailed pathologic processes became possible. Spargo and co-workers reported a consistent glomerular capillary endothelial cell change characterized by obstructive swelling of the enFrom the Department of Pathology, The University of Chicago. Dr Spargo is the recipient of research career award No. K6HLA418 from the National Institutes of Health, Bethesda, MD. Address reprint requests to Benjamin H. Spargo, MD, Department of Pathology, The University of Chicago, 5841 S Maryland Ave, Chicago, IL 60637. © 1987 by the National Kidney Foundation, Inc. 0272-6386/87/0904-0014$3.00/0
American Journal of Kidney Diseases, Vol IX, No 4 (April), 1987: pp 317-323
317
GABER AND SPARGO
318
dothelial cell, vacuolation, intracellular lipid accumulation, and hypertrophy of the cytoplasmic organelles.! They. identified glomerular endotheliosis as the basic pathologic process and further indicated the noninflammatory nature of the lesion and its specificity to preeclamptic nephropathy. Recently, Sheehan and Taylor and Spargo reviewed their experience and provided detailed elaborate descriptions of the renal lesions in preeclamptic patients. 4.5 At the University of Chicago we have accumulated > 200 postpartum renal biopsies with documented clinical follow-up. Examination of this material, which contains a large spectrum of morphologic alterations with varying degrees of severity, has led us to a postulate of different pathologic findings with specific remote outcomes. The importance of morphologic data is exemplified in our results through 1976, in which 45 % of women whose gestational hypertension was diagnosed clinically as preeclampsia alone had their diagnosis altered based on pathologic examination of biopsies. 5 The discrepancy between clinical opinions and the pathologist's diagnosis was most pronounced in the multiparous group. Another feature of our material was that the lesion of pure preeclampsia was completely reversible. Recently, however, several investigators have linked development of focal segmental glomerulosclerosis (FSGS) with preeclampsia, an association which, if true, should alter the benign prognostic implication of pure preeclampsia (PIN). 6·8 To investigate this association, we reviewed a series of 20 postpartum biopsies in women who had hypertension in pregnancy. METHODS Twenty renal biopsies performed during the years 1979 through 1983 were retrospectively selected in random fashion from our files. This study includes 19 women (one patient was biopsied after two successive gestations) and differs from a series o f 176 cases spanning 1958 through 1976 and published by us in 1981. 2 In essence, selected criteria had changed in the late 1970s, and renal biopsies were performed only on "atypical" and "severe" cases; the 1958 through 1976 group represented a more heterogeneous population. In the present series, biopsies were obtained on the average eight (range four to 21) days postpartum and were performed primarily because of hypertension and/or proteinuria persisting in the puerperium or severe preeclampsia presenting in midpregnancy or early in the third trimester (Table I). Mean age at biopsy was 22.5 years (range 15 to 36 years) and only 50% of biopsies were performed in nulliparas. The material we report was specifically reviewed in an at-
tempt to locate lesions that were not a feature of our previously published observations but that had been underscored as important preeclamptic renal lesions by others. Thus, we step-sectioned the biopsies and reviewed as many as 20 glomeruli in each patient, specifically attempting to document FSGS as well as interposition of mesangial cytoplasm with the basement membrane ("double contour") appearance.
Pathology Light and electron microscopy. The renal biopsies demonstrated the morphologic features of preeclampsia, with some variation in intensity of the different components (Table 1). Enlargement of the glomerular tuft, variable hypercellularity, decreased vascularity, and obstruction of the peripheral capillary loops were common features (Figs 1 and 2). Herniation of the glomerular tuft into the proximal tubules and longitudinally collapsed cigar-shaped lobules were common. Two cases disclosed partial glomerular infarction and reactive epithelial crescents. The mesangial and endothelial cells were hypertrophied, swollen, and vacuolated, with occasional foam cells. Electron microscopy disclosed intracytoplasmic lipid vacuoles, myelinelike figures and, rarely, cholesterol crystals (Fig 3). Interposition of mesangium under the endothelial cells was occasionally noted; in contradistinction to the experience of other investigators, this was not a feature of our material. Epithelial cells were prominent; despite proteinuria in all cases, however, only two cases showed focal effacement of the foot processes. Seven biopsies (35 %) showed some evidence of focal segmental sclerosis. The sclerotic areas usually showed expansion of the stalk and collapse of a few capillary loops (Fig 4). Two biopsies exhibited globally sclerosed glomeruli, and adhesions to Bowman's capsule and cellular crescents were noted in three instances. Vascular lesions in the form of arteriosclerosis existed in ten biopsies, including five with focal sclerosis. Intimal widening, fibrosis , and reduplication of the elastic lamina were focal and often limited to very few arteries (Fig 2). No arteriolar lesions were observed. Whenever arteriosclerosis existed, focal tubular loss and condensation of the interstitium were observed. Immunohistochemistry. Observations were nonspecific and staining, when present, related to nonspecific trapping of immunoglobulins and fibrin secondary to the pronounced glomerular ischemic injury. There was no segmental localization of immunocomplex deposits.
20 30 18 19 22 21 18 25
18
15 18
23 36
17
7 8 9' 10' 11 12 13 14
15
16 17
18 19
20
N
M N
M M
N
39
27 34
28 32
37
29 33 19 24 22 33 32 28
28 34 34
32
Severe preeclampsia (HELLP syndrome)persistent postpartum hypertension Early onset of severe preeclampsia with HELLP syndrome Hypertension , with marked nephrotic proteinuria Early onset of severe preeclampsia Persistent postpartum hypertension Eclampsia and persistent postpartum hypertension Early onset of preeclampsia Persistent postpartum hypertension Early onset of preeclampsia and eclampsia Early onset of severe preeclampsia Early onset of severe preeclampsia Persistent postpartum hypertension Severe preeclampsia Early onset of nephrotic syndrome and hypertension Postpartum hypertension and seizures in a diabetic patient Early onset preeclampsia Severe hypertension and nephrotic proteinuria Early onset of severe preeclampsia Persistent postpartum hypertension and nephrotic range proteinuria Persistent postpartum hypertension
Indication for Renal Biopsy
7
6 4
6 5
9
6 21 5 8 10 6 5 7
7 11 7
10
11
15
20
20 36
13 20
30
37 4 66 113 17 19 5 50
29 33 20
26
17
3
Biopsy Performed Total No. (Postpartum day) of Glomeruli
2%
7% 4%
8%
+
+++ 10% ++ 10%
++ ++
+ 8%
3%
8%
6%
10%
20%
9% 4%
+
+
+
+ +
+ + +
+
+
+
+ +
+ +
NA
+
NA
++ +
++
++
Global Cellular Tubuloinstertial Vascular FSGS Sclerosis Crescents Changes Lesions
++ 10% +++
++ ++ +++ ++ +
++ ++ ++
++
++
+
PIN
Clinicopathologic Correlations In 20 Women With Preeclamptic Gestations
Abbreviations: N, nullipara; M, multipara; HELLP syndrome, hemolysis, elevated liver enzymes and low platelet counts; PIN, pregnancy-induced nephropathy (glomerular capillary swelling; +, + +, + + + = mild , moderate , and severe, respectively); FSGS, focal segmental glomerulosclerosis; NA, inadequate material for diagnosis. ' No.9 and 10 are the same patient biopsied after successive hypertensive gestations.
8
W
8
8 81
8
8 8 8 8 8 8
W
8
N N N M M N N N
M N M
8 8
32 19 19
4 5 6
W
M
8
26
3
25
M
W
27
2
32
Onset of Hypertension (Gestational Week)
M
8
28
Patients Age Race Parity
Table 1.
"TI
0 :> r
~
»
en
-0
~
:>
m m () r
JJ
-0
Z
en 1ii
0
JJ
r m
()
en
r 0
JJ C
r 0 ~ m
(j)
()
320
GABER AND SPARGO
Fig 1. Micrograph of a slightly enlarged glomerulus demonstrating focal loss of vascularity and endothelial cell swelling. (Hematoxylin and eosin, x 400.)
DISCUSSION
Postpartum renal biopsy is an important aid in the prediction of long-term morbidity and eventual outcome in patients with gestational hypertension. Use of pathologic criteria separates those patients with overlapping clinical syndromes into groups with a more predictable course. Glomerular endotheliolsis is a completely reversible change. Detection of vascular sclerosis on the other hand appears to be an indication of a hypertensive diathesis that may not have been clinically manifested prior to pregnancy. The significance of this distinction is best appreciated when clinical and
Fig 2. Micrograph shows marked variations In the degree of obstruction of the capillary loops. Example of intimal widening and fibrosis of a small artery associated with mild tubular dropout. (Hematoxylin and eosin, x 312.)
epidemiologic studies based on biopsy results are evaluated . In our first study at the University of Chicago, 83% of patients with pure "glomerular endotheliosis" were primigravidas, whereas in multigravidas the preeclamptic lesion was superimposed on nephrosclerosis in about half of the cases. ~ Follow-up of these patients disclosed a similar incidence of hypertension (9.4%) in the pure preeclamptic group when compared with agesex-race adjusted statistics from a survey by the National Institutes of Health. Patients with nephrosclerotic lesions, on the other hand , had a much higher incidence of permanent hypertension
FOCAL GLOMERULOSCLEROSIS IN PREECLAMPSIA
321
Fig 3. Electron micrograph from a biopsy demonstrating both pregnancy-induced nephropathy and focal segmental glomerular sclerosis. Area shown is away from the sclerotic lobule and demonstrates marked cytoplasmic hypertrophy and vacuolation of the endocapillary cells and Interposition of mesangium resulting in obstruction of the capillary loops. Epithelial foot processes are mainly intact. (Original magnification, x 5,400).
(74 %) that did not become manifest in most cases until several years after the index pregnancy. The nature of the vascular lesion, whether it is a primary aging process or a secondary hypertensive change, has been a subiect of debate. We can-
Fig 4. Example of focal segmental sclerosis in a bloodless glomerulus. (Periodic acid-Schiff, x 312.)
not subscribe to the former view because the mean age in the present series of patients with nephrosclerosis is 22.5 years. Biopsy interpretation, however, can be limited by inadequate sampling of vessels and the focal
322
GABER AND SPARGO
Fig 5. Micrograph demonstrating a small reactive area of epithelial cells in a partially bloodless glomerulus. (Hematoxylin and eosin, x 312.)
nature of the vascular lesions. In both situations, we stress that the pathologist should diligently seek evidence of secondary more chronic ischemic changes, such as tubular atrophy, tubular loss, and condensation of the interstitium, to qualify a prognostication based on the available specimen. In reviewing our material , we specifically searched for FSGS in preeclampsia in an attempt to derive its clinical significance. In this group of women with severe disease, seven of 20 (35 %) biopsies reviewed disclosed some evidence of FSGS involving very few glomeruli. This lesion is very much similar to the focal fibrosis and lobular matting described by Sheehan and Lynch in 1973. 9 An intriguing finding is the presence of arteriosclerosis in five of the seven cases (71 %) and only tubular loss and interstitial condensation in the other two cases. This persistent occurrence of vascular lesions when FSGS exists classifies this group as hypertensive nephrosclerosis , and imparts its prognostic implications. This focal segmental sclerosis appears to be of ischemic nature, induced by the ongoing indolent hypertensive process, perhaps compounded further by the renal hemodynamic alterations that occur during pregnancy. This form of focal sclerosis is not to be confused with primary FSGS. The latter is characterized by a progressive unremittent course with a survival rate of 75 % at 5 years and 38 % at 15 years. 10 Primary FSGS has been reported in 22 % of pregnancies with glomerular diseases excluding "pure" preeclampsia; the authors claim that 52 % of these
patients had onset of FSGS during pregnancy. liOn the other hand, the form of segmental sclerosis that we described in relation to preeclampsia superimposed on nephrosclerosis does not demonstrate the ultrastructural or the immunologic criteria of primary FSGS. Focal segmental glomerulosclerosis is not a feature of pure preeclampsia, but rather an indicator of underlying nephrosclerosis . Once diagnosed , vascular lesions should be considered and sought because long-term prognosis in these patients is probably similar to those with nephrosclerosis and superimposed preeclampsia. ACKNOWLEDGMENT We thank Penny Theodoropoulos for h erexcellent secretarial help.
REFERENCES I. Spargo BH , McCartney Cp, Winemiller R: Glomerular
capillary endotheliosis in toxemia of pregnancy. Arch Pathol 68 :593-599, 1959 2 . Fisher KA. Luger A. Spargo BH , et al : Hypertension in pregnancy: Clinical-pathological correlations and remote prognosis. Medicine 60:267-276. 1984 3. Lindheimer MD, Chesley LC, Taylor JR, et al: Renal function and morphology in the hypertensive disorders of pregnancy (in press) 4 . Sheehan HL: Renal morphology in preeclampsia. Kidney 1nt 18:241-252 , 1980 5 . Taylor JR. Spargo BH: Pathology of the kidney in preeclampsia, in Andreucci VE (ed): The Kidney in Pregnancy. Boston , Nijhoff. 1986 , pp 47-63 6 . Nochy D. Gaudry C, Hinglais N , et al : Can focal segmental glomerulosclerosis appear in preeclampsia? Adv Nephrol 15:71-85, 1986
FOCAL GLOMERULOSCLEROSIS IN PREECLAMPSIA
323
7. Nochy 0, Nihglais N, Jacquot C, et al: De novo focal glomerulosclerosis in preeclampsia. Clin Nephrol 25:71-85, 1986 8. Kida H, Takeda S, Yokoyama H, et al: Focal glomerular sclerosis in preeclampsia. Clin Nephrol 24:221-227, 1985 9. Sheehan HL, Lynch JB: Pathology of Toxemia of Pregnancy. Baltimore. Williams & Wilkins . 1973
10. Cameron JS, Turner DR, Ogg CS, et al: The long term prognosis of patients with focal segmental glomerulosclerosis. Clin Nephrol 10:211-218, 1978 II. Surian M, Imbasciati E. Casi P, et al: Glomerular disease and pregnancy: A study of 123 pregnancies in patients with primary and secondary glomerular disease. Nephron 36: 101-105.1984