The liver in Systemic Lupus Erythematosus:

Pathologic Analysis of 52 Cases and Review

of Japanese Autopsy Registry Data

TOSHIHARU MATSUMOTO, MD, TUG10 YOSHIMINE, MD,

KOUJI SHIMOUCHI, PHD, HIDETOSHI SHIOTU, MD,
NOT;?IYUKI KUWABARA, MD, YOSHIRO FUKUDA, MD,
AND TANJI HOSHI, MD
We present pathologic findings for 52 livers (51 autopsy specimens servations 01 arteritis of the liver in SI.Ir llave not bren
and one wedge biopsy specimen) from patients with systemic lupus reported in the English literature..
erythematosus (SLE). Hepatic congestion was the most common Nodular regenerative hyperplasia of the liver
disease (40 livers), followed by fatty liver (38), arteritis (1 l), cho-
(NRH), chal-acterized by diffuse nodularity of the liver
lestasis (nine), peliosis hepatis (six), chronic persistent hepatitis
produced b) many regenerative ~~dultrs without asso-
(six), nonspecific reactive hepatitis (five), cholangiolitis (four),
ciated fibrosis, frequently has 1xwI obst~rved in pa-
nodular regenerative hyperplasia of the liver (three), and heman-
gioma (three). The data obtained here suggest that arteritis of the
tients with c,ollagen vascular diseases. especially rheu-
SLE liver is more common than has been recognized previously. matoid arthritis. - To our knowledge, approximatel)
One patient had hepatic infarction complications induced by ar- 190 cases of NRH have been reported in the litera-
teritis. On the basis of the findings in the present study and a review ture, and 10 of these had the urlderlying disorder of
of the literature, we suggest that hepatic infarction resulting from SLE. ,w Of published studies of liver pathology in
arteritis is rare in SLE. On the other hand, while occurrence of populations of SLE patients.m7 onlv the study by Miller
nodular regenerative hyperplasia of the liver in SLE patients has et al (who reported that nolie of I4 patients with SLE
been considered to be rare, our findings suggest that it may be
had NRH) addressed the incidence of hRH.
more common than has been recognized previously. Although
Thus, iI1 the present study we per1; armed a detailed
congestion and cholestasis may be acute terminal illnesses, fatty
change is considered to be specific to the SLE liver. Statistical anal-
pathologic analysis of hepatic- diseases, with particular
ysis indicates that exposure to a large dosage of glucocorticoids is
referencr to arteritis and NRH, in .G,Lliver specimens
a significant factor in the etiology of severe fatty liver. In addition, from patients with SLE. Morecner. to cxarnille the in-
our review of Japanese autopsy registry data for 1.468 patients with c%Ience of chronic liver discasrs in a large population
SLE indicaltes that the incidence of chronic liver diseases in SLE of SLE patients, we reviewed Japmes~ autopsy regisrr\
autopsy causes is as follows: chronic hepatitis, 2.4%; cirrhosis, data fi)r I ,-lCiX patienrs with S1.F..
1.1%; and liver fibrosis, 0.8%. HUM PATHOL 23:1151-1158.
Copyright 1~.1992 by W.B. Saunders Company
MATERIALS AND METHODS

AIIIOII~ patients with systemic lupus er~thematosus

(SIX), biochcmic~al abnormalities of hepatic function
WCa c0111111011 pt1e1101ne11o11. - However, the liver mw-
pholo~~ i.s noriiial or presents minor abnormalities in
lllost (axs, -, Sevew chronic liver diseases. such as
c,hronic- ;lc.tivc hepatitis and cirrhosis, arc reported IO
c~cur rarrh in patients with SLE,,-i but thr true in-
ciclencc 1.1fc.l1ronic, liver disrases in SI,E ha5 not been
well tioclllrleIlted.
.~lthou~h a few case reports of hepatic infdrction
resulting 11xm arteritis in SIX have bren publishecl,H~!
I here have htat:n no studies analyzing hepatic. infarction
in ;I population of S1.E patients. Moreover. detailed ob-

RESULTS
Reviewof JapaneseAutopsy 17egistryData
for 1,468 Cases of Systemic Lupus
Erythematosus

1151
 HUMAN PATHOLOGY Volume 23, No. 10 (October 1992)

1,071 patients (72.9%) were in the 20- to 49year-old TABLE 2. Incidence of Liver Diseases in Patients
age range at the time of death (Table 1). With Systemic Lupus Erythematosus Recorded
The most frequent liver disease was fatty liver, fol- in the Japanese Autopsy Registry Data (n = 1,468)
lowed by hepatic congestion (Table 2). Chronic hepatitis
occurred in 36 patients (2.4%), of whom it was reported
to be active in eight; activity of the hepatitis in the other
28 patients was not described. Thus, the true incidence 1 I..,
.!I I1
of chronic active hepatitis could not be verified. Cir-
rhosis was present in 17 patients (1.1%). Both acute and 2. I
fulminating hepatitis was nored in five patients. Of 63 1.1
patients who had complications with either hepatitis 0.X
I .x
(acute, chronic, or fulminating) or cirrhosis, 18 (28%)
1 .!I
had a history of blood transfusion. For these 63 patients 1j.i
clinical information regarding hepatitis B surface anti- O.li
gen had not been recorded. Liver fibrosis was present 0.5
in 12 patients (0.8%). Hepatic hemangioma had oc-
0.3
curred in 23 patients (Table Z), but no primary malig-
0.3
nant neoplasms of the liver were recorded. Nodular re- lb.3
generative hyperplasia of the liver was present in five 0.3
patients. Systemic infectious diseases, such as tubercu-
losis, fungus infection, and cytomegalic inclusion dis-
ease, occurred infrequently.
least two different liver-related tests being outside the
Analysis of Clinical and Pathologic Data for normal range. Mean test values and standard error fog
52 Patients With Systemic Lupus 15 patients showing abnormal liver function were as
Erythematosus follows: SGOT, 97 _+ 59 [J/L>; SGPT, 8X k 61 Ll/I,;
LDH, 1,15 1 -t 460 LJ/I,; AL,P, 19 +- 6 King-ArmstroIlg
Clinical Findings. Of 52 patients, there were 45
unit; and total biliruhin, 8 k 13 mg/dL.
women and seven men whose ages ranged from 20 to
All patients had undergone corticosteroid therapy.
70 years (mean, 36 years). The period between first
Detailed clinical information regarding drug therapy was
manifestation of SLE and death (51 cases) or operation
available in 47 patients , 3 1 of w1i011ihad received a large
(one case) ranged from 2 months to 31 years.
close of glucocorticoids (ie, total dose of more than 10
Twenty-two patients had a history of hlood trans-
g of prednisolone). The duration of drug therapy ranged
fusion; hepatitis B surface antigen was positive in one
from 18 days to 20 years. In addition, 13 patients had
patient, but negative in the others. Antibodies to hep-
received i~nmunosuppressive agents, such as (i-mercap-
atitis C virus were not examined in any patient. Clinical
topurine, cyclophosphamicle, or azathioprine.
information regarding liver-related tests was available
One patient had the portal hypertension symptoms
for 41 patients. We assessed liver function abnormality
of esophageal varices and splenomegaly. Thus, operation
on the basis of recorded values on the following five
(transection of the esophagus, splenectomy, and liver
liver-related tests: serum glutamic oxaloacetic trans-
biopsy) had been performed to treat the portal hyper-
aminase (SCOT; normal range, <40 U/L), serum glu-
tension. This patient, who is still alive 8 veals after the
tamic pyruvic transaminase (SGPT; normal range, <30
operation, was diagnosed as having NKH on the basis
U/I,), serum lactate dehydrogenase (LDH; normal
of pathologic examination of the biopsied liver speci-
range, 50 to 400 U/L), set-ml1 alkaline phosphatase
(ALP; normal range, 1 .f to 9.9 King-Armstrong unit),
and total bilirubin (normal range, 0.2 to 0.8 mg/dL).
Fifteen patients (37%) were classified as showing ab- TABLE 3. Liver Diseases of 53 Patients
normal liver function on the basis of the results of at With Systemic Lupus Erythematosus

TABLE 1. Age and Sex Distribution of Patients

With Systemic Lupus Erythematosus Recorded
in the Japanese Autopsy Registry Data (n = 1,468)

.;.T
.i 7
I .!I
I .!I

1152
 LIVER IN SYSTEMIC LUPUS ERYTHEMATOSUS (Matsumoto et al)

IIIW. 111tlw 5 I autopsied patients, the cause of death iuc.ludcd hepatic congestion. fillty IivtaI . ;wteril is. cho-
was respirator!-, renal. or heart failure in 39. disease of lestasis. peliosis hepatis, chronic persistelit hqxltitis.
the cxmtral newous svstem (hemorrhage, infarction, or nonspec ifk reactive hepatitis, c.t~cll;lrlgiolitis. NK1-I.
Irlcvlingitis) iu rliue, suicide in two, and purulcrrt peri- heniangionia, hepatic infarction. and (13I)tocc bccal in-
tonitis in one. faction (Tahk 3).
tr~thologic~Ihdir~gs. Autopsy liver weights ranged Hepatic congestion ~14 ~mw1~1 it1 40 patirlrts
1rorn 660 IO 2,070 g (nwan, 1.4 10 g). Ethpm31t of (71i.!l%j. stvtn of whoni prestwtwi Fvirlt .~c~~Ic cxwg~5
the livw WIS noted in 26 patients (5 1%). Hepatk diseases lion showing (xqul;ltion necrosis otliq~;~l~ )cvte\ :iro~~titl

FIGURE 1. (Top) Active

arteritis. Note fibrinoid
necrosis involving three
layers and destruction of
elastica interna of the
vessel wall. (Elastic van
Gleson stain, magnifica-
tion T. 200.) (Bottom)
Scarred arteritis. Note fi-
brous thickening of the
intima ant adventitia,
destruction of the elas-
tica interna, and fibrosis
with discor#tinuity of the
elastic lamina in the me-
dia of the vessel wall.
(Elastic van Gleson stain.
magnification . 2OO.j
 HUMAN PATHOLOGY Volume 23, No. 10 (October 1992)

TABLE 4. Arteritis in the Liver: Pathologic diameter than a hepatic lobule, but in the others (cases
Characteristics and Its Relationship to Nodular no. 11 and 12) there were nodules both smaller and
Regenerative Hyperplasia of the Liver in 11 Patients larger in diameter than one lobule (Fig 3, center). None
With Systemic Lupus Erythematosus
of the hyperplastic hepatocytes presented cellular atypia
Arteritis Nodular (Fig 3, bottom) in any of the three patients. One patient
Regenerative (case no. 13) showed complication with portal hyper-
Case Age No. of .4ffectrd Hyperplasia tension, but in others (cases no. 11 and 12) phenomena
No. (yr)/Sex Type Arteries* of the Liver indicative of portal hypertension, such as ascites, esoph-
1 :38/F Active +
ageal varices, and phlebosclerosis of the portal vein, were
2 26/M Scar + _ not noted. In one patient (case no. 11) both NRH and
3 62/M Scat + _ arteritis were present. However, there was no correla-
1 70/M Scar + -
tion between NRH and arteritis (Table 4). The liver pre-
5 70/F Scar t -
- sented severe congestion in two patients (cases no. 1 1
6 24/F Scar tt
7 40/F Active +++ - and 12) with NRH (Table 5). Of the three patients with
8 22/F Scar +tt - NRH, two (cases no. 12 and 13) had received a large
9 54/F Scat +++ - total dosage of glucocorticoids (more than 10 g of pred-
10 37/F scat tt+ _
nisolone) and one (case no. 11) had received both glu-
11 -11/F Active and scat +t+ t
cocorticoid and cyclophosphamide (Table 5).
* The number of affected at-teries was examined in three speci- There were no complications with severe chronic
mens from the hepatic hilus, right lobe and left lobe. Symbols: +, I - liver diseases, such as chronic active hepatitis or cirrho-
3; t+, 1-9; t-t+. >10. sis, but six patients had chronic persistent hepatitis and
five had nonspecific reactive hepatitis. Of the six patients
with chronic hepatitis, three had a history of blood
the central veins in some areas. Fatty liver was noted in transfusion but none were hepatitis B surface antigen
38 patients (73%); fatty liver was mild in 23 and severe positive.
in 15. In mild fatty liver small fat droplets had accu- Both peliosis hepatis and hepatic hemangioma were
mulated in the hepatocytes of the centrilobular and present in one patient. Moreover, in seven patients ei-
midlobular zones. On the other hand, in severe fatty ther peliosis hepatis (five) or hepatic hemangioma (two)
liver admixed small and large fat droplets had accu- was present. Three of the six patients with the compli-
mulated in the hepatocytes throughout the hepatic lob- cation of peliosis hepatis had received a total dose of
ules. Of 31 patients who had received a total dose of more than 10 g of prednisolone. In 15 patients showing
more than 10 g of prednisolone, nine presented with abnormal liver function hepatic lesions included fatty
severe fatty liver. liver (six), cholestasis (four), nonspecific reactive hepa-
Histologic evidence of arteritis of the liver was titis (three), and chronic persistent hepatitis (two).
present in 11 patients (21.2%). The arteritis presented Although we performed a statistical study using chi-
with the features of the periarteritis nodosa type of nec- squared analysis to investigate the cause of enlargement
rotizing arteritis. Arteritis was classified into two stages of the liver, no hepatic disease (including congestion
(active and scar) on the basis of published studies,R.21.2 and fatty change) was statistically implicated. In an at-
and our own experience, with arteritis. Active arteritis tempt to identify the etiology of fatty liver in SLE pa-
was characterized by fibrinoid necrosis and inflammation tients, we performed statistical analysis using a logistic
involving all three layers of the vessel walls (Fig 1, top). regression nlodel,6 on data for the 47 patients for
Scarred arteritis was diagnosed by the fibrous thickening whom detailed clinical information regarding drug
of the intima and adventitia, fibrous scar with discon- therapy was available. The logistic regression model has
tinuity of the elastic lamina in the media, and destruction been used as a means of relating disease probability to
of the elastica interna (Fig 1. bottom). In one patient one or more categorical risk factors whose levels are
both stages were intermingled, but in the others only represented by indicator variables. In the analysis re-
one stage (active in two and scar in eight) was noted gressors included various factors that seemed to have
(Table 4). The number of affected arteries in the liver the potential to produce fatty liver. Analysis results
differed among the patients (Table 4). There were no showed that total dosage of administered glucocorticoids
differences in distribution of arteritis within the hepatic was statistically correlated (P = .040) with production
hilus, right lobe, and left lobe. Most affected arteries of severe fatty liver.
ranged in size from 100 to 400 pm in diameter, as has
been previously reported by one of the authors (T.Y.).
In one patient (case no. 9) many arteries presented lu- DISCUSSION
minal stenosis or obstruction arising from intensely
scarred arteritis, which had been followed by hepatic Systemic lupus erythematosus is an imnlunolo~cally
infarction formation (Fig 2). mediated disease characterized by multiple organ in-
Three patients showed NRH (Fig 3), ie, nodules of volvement. Attention has been focused on the hepatic
hyperplastic hepatocytes without associated fibrosis dis- manifestations of SLE because of the liver enzyme ele-
tributed throughout the liver as detected at wedge liver vation observed in SLE patients. Hepatomegaly is a
biopsy (case no. 13) or at autopsy (cases no. 11 and 12). common finding in SIX patients, with reported inci-
In one patient (case no. 13) all nodules were smaller in dences ranging from 27% to 52%..9,;,7 Common ab-

1154
 LIVER IN SYSTEMIC LUPUS ERYTHEMATOSUS (Matsumoto et al)

FIGURE 2 . Hepatic infarction due to

scarred Weritis. (Top) Hemorrhagic
fibrous 2ic or:; (indicated by arrows) of
variable? siie are present in some
areas cIf the liver. (Bottom) Arteries
(indica. te:d by arrows) with luminal
stenosis >C)r obstruction due to scarred
arteritis CIre noted in portal areas ad-
jacent t 0 this hemorrhagic fibrous scar
area. (f \Zan stain magnification u40.)

II 12 13 14 1s 1s 17 1s 19

1155
 FIGURE 3. Nodular regenerative hyperplasia of the liver. (1-oP)
Nodules of variable size are note ,d throughout most of the he-
patic parenchyma. (Center) Not e nodules of hyperplastic he-
patocytes both smaller and larger in diameter than one hep atic
lobule. (Reticulum stain; magnif ication X 17.) (Bottom) High-
power view of the nodule. None 01f the hyperplastic hepatoc ytes
present cellular atypia. (Hematox :ylin-eosin stain; magnifica tion
x83.)

1156
 LIVER IN SYSTEMIC LUPUS ERYTHEMATOSUS (Matsumoto et al)

TABLE :5. Clinical Findings and Hepatic Diseases fact that ottI\ ;1 few such GIWS tuvc tx~c~rt reported itt
In Three Patients With Systemic Lupus Erythematosus the titerat&. suggest that ltepatic itifatx rioti resulting
With Nodular Regenerative Hyperplasia of the Liver
-___ I_-- from arteritis is rare in SI,E.
, L!. There is stilt disagreetiietit as to Itiv true irtcidetice
( ::I\,. NC, II I3
_-. _.__ 01 st~c~(r chtmttic liver diseases in S1.E. Kurt\on et at
found fimt- cases with chronic active. tte1ntitia or tivet
c-it-rtiosis irt 238 c.;t\es . of S1.E. On thr other hand. Miller
t3 al reportetl that norit of 2tiO patimts with SI,1;, pry-
\ented with severe chrotric~ li\,er discascs. In addition,
in (;itx3)tt md Myers study 018 1 patients with SI,k: ottt>
one pt~sm1ed with c~hrotiic- active tirp,ititis. In rtie
prcseiit stud\, wt esmiitied tlic iticidenc-es of c-ltrotiic-
li\er discaaeh in 1,468 patients with SI .1<:, thcb largmt
ttuttit~rt- 0E patirtitb iii ati\ sucli study to rl;k(e. Of thehe
pa(ients, 17 ( I. 1%) prtxtlted livrr cirrhosis. This tindittg
suqqcsts
<< that the incidence of tivu cit7hosis iii :tutops\
~xscs oiSI,E tii;tV be approsittiat~t~ IV. Iltt~ etiolog of.
scvt7c ctitxmic~ li\.et- ttiscasea iii SI.k: has trot heeti well
cletcmtitied. Iii the present stud\ ttata srtgg3fiVt. of ica
t~tiolo~V wets not ot~~;iinetl.
Ittrcc~it \e;irs an iticreasr iii thy tittttttxt ofc;iscs
oi~peliosis Iic.p;ttis fotlowirig the ttstofsteroids has bwt~
tq~oi~ted.2 tit Itir prcsetit studs pc+osis tiepatis oc-
c~urred in six patients, ttirets of wtiotit Itatl rc*ceivml ;I
total tlosag~ OP IllOtT tl1at1 10 g of JmY~mlisotolle. -ftlrls.
11~ occurrcttc~e 0fJ~elioais ticJn(is itt tliesr rlircc~ JMicttts
ttiav he litikrd 10 ~~o~tic~ostet~oid tttrtqn~.
(:tiotmgiolitis is produced ty (It-tip, toxic sttt)-
htatices, md wptic. stioc.kL .Altticq$i forit. patients prt-
sriitcd with ~~ltolatigiolitis, WC c~cwltt not determine 11~
(xtihr (bf tltr. ~~tiot~iii~iolitis, <w7t after dc*cailecl c~littktt
Irormaliti~:r 01 11w liter itt this disorder are cottgestion, and pitti~~logic~ reviews.
latt\. ttirt;tntorl~ttosis , anti ttoiispecific reactive lic~patitia.~ Iii I!I.5!). Stritier teportrtl tliftuw ttodrrlat-itc of
Indeed. in 11t(s prescttt study hepatic cotrgSc.stiott was rlir tivm proclticed tw tiiatw tqrttct~;itiw rtodttles with-
tlic most lkoiii~ttoit disrasc, li~llowed by fatty liver. I;att\ out associated fihtx& andt~rtti~d rtiis entity tiocl~~lat~
IiVu. Iielxi~ic cxmgestioti, atid clioleslasis are \et-y wlil- rcgeticrativc tivJ~crJhsi;i 0E the liver. hottulx regeti-
11i01i tcrtiiitial cvctiIs iii xttopsies. Thus, we rcviewul rtxtivr tivJ~ct~JA;isi;i of tlic liver ti:15 txtm t~eJ~ortcd it1
ttir studit:-\ of Slk: liver biopsies.~~ In these, fatty patients hidi u~tl~igeti Vascular tlise;tsr*s. tiryeloptx-
c hatige \\;ts fr-cquetitt~~ desc3ihecl bitt congestioti and likativc- disordet., - t-ettal ~ratisplati~. -. ;iiicl
c holestasis wcrc not described. These findings i&care titotiocloti~d piitiioJx1thy. Suhst~qitf~titty, aJ)Jmksi-
that fittty c.tiattge is specific to SLX liver but that conges- tnatelv 1 !)O (X,CS NKH ha\xl t)eett tx~l)ortod in tttc
of
riot1 and cltohtxis may be terminal events. (:ot-ticoste- lirrraturr. ;~titl its patliotogic features ;ttitl tttidcrlyitig
raids art lt)elieVCd to cmise fatty infiltration iii the liver. disordet~s atx well docutnenr~d. I;ttir.ttl:\ with YKH ;IIY
ltitts. (;il-bsott md Myrt-s re1~orted that kitty li\rr iii SI,k: ltxxluetitl~ (txx(ed with steroicl. ~ttititirc)l)l~tstic~. ot ittt-
tii;n he littkecl to long-tcrtii c~orticosteroid tht*rapy. 011 titrttiosu1~l~r~~ssi\~~ drugs. 3 The cotiiplic ;itioti of port;iI
ttte other tlatld. Ku11y011 tt AI could not c~ct11011st1xc ;I ti~lx33etisiott is observed in appn )sitiiatety half of pa-
rel;ttiottship twtweett cortic.osteroid therapy and fatty tients with NKH. To date. IO SI,k: lxtliettts with the
tivrr iii 238 lxitimls with SLIC7 III the Jx~ctil stud> ~~otlrpticatiott o!NKH have kctt tq)ortuI., xttt wta
statistical mnatvsis suggests that exposure to a large clos- rc.pot-tat otic of tticse (exe ito. 13 ilt ttit, prcsetit
age 01 glut ocx~rtkoids is a significant f&Tot. iti tlic etiol- study). 111 this htttdy we diqttos~c! Iux) other- SLk: pa-
t )g> of sexchrc fat tv liver. tictits (casts no. I I and I?). who had ii01 bceti idctitifietl
The arteritis iii SLE is ctie p~tkirteritis notlosa t~pc ;I:, ha\ ittg NKH pre~iottsty, on the tasi!; of detailed cs-
of tiect-otijirig at.tct.itis.x.~~: Iii KferiiJ~eret~ et als stud!; ~mitiatioti ot liver- autopsy specittimts. ltrtis. \ve t)etie\c
of 20 SI.I< autops\ cases wteritis was found iii two tlt;tt tltr tx~portrd incidence of NKI 1 in SI .I+:IILIV ittcm-its<
( I W;). Sttt~st~clneti~l~, tlie liver has cotit~ (0 IJC umsid- oti ittow tlcc;iilvtl t.s;tttiitiatiott 01 Iltc liv~t
t.rt*cl att c~~t-g~tiin wl1ic.h ar~eritis rarely, develops in SI,E. ha to tllr. criology of NKH, lxtssi\c t ottgcstiott w(~-
l-low~~~t-, in tttr lm3rtit atutiy the itictdetice of arteritis ottdan to t-i@ hart laitut~c, LIS \Wll ,I4 es]Fmt-c~ lo

01 Ihe Ikei. atiioti~ SIX paticrtts was found to be Ll%. ctxrtaiti tliup (steroid, atititicol~t~tsci~~, c)I itnttiutiosttp
SI tinditig suggt3titig that arteritis 01 the liVc%t.is tiiot-c pressi\c ;igeti14), I has bevti poposc~d. Ott ttiv o~tirt-
~~otittiioti iii Sl .k: (li;itt has ken r~c~ogtiiz~d l)tmiottsl\ limtl. Mmt~~5s ct al IiaVe sttggYs14 tttal c~l~titc~ratiott of
In rlica 1Iixw~~1 study only one patient hart tiqxiic itt- itttrah~~p;ttic- por~;tI \eitr t-xlicks ma\ initiate UKH. J
lat~c~tioti rc3ttltin~ 1rotii iit-trtitis. This fittdittg 2nd tlit Tlirsc~ iti\c3ti~~icot-~ ;tt\o tq)orl~(I ;t 1xttcmt n irti rlttw

1157
 HUMAN PATHOLOGY Volume 23, No. 10 (October 1992)

matoid arthritis in whom arteritis led to secondary oblit- 12. Weinbren K. Mutum SS: Pathological aspects of diffuse nod-
ular hyperplasia of the liver. J Pathol 143:8 I-92, 1984
eration of adjacent portal venules and NRH.32 Several
13. Colina F, Alberti N, Solis JA, et al: Diffuse nodular r-egen-
different hypotheses have been proposed, but the exact erative hyperplasia of the liver (DNRH). A clinicopathologic study of
pathogenesis of NRH remains obscure. In the present 24 cases. Liver 9:253-265, 1989
study we did not demonstrate a relationship between 14. Wanless IR: Micronodular transformation (nodular regen-
arteritis and NRH, even on detailed examination for erative hyperplasia) of the liver: A report of 64 cases among 2,500
autopsies and a new classification of benign hepatocellular nodules.
both. Thus, we believe that intrahepatic arteritis does Hepatology 11:787-797. 1990
not produce NRH in SLE. Detailed examination of the 15. Kuramochi S. Tashiro 1. Torikata C. et al: Systemic lupus
relationship between dosage of administered steroids erythematosus associated with multiple nodular hyperplasia of thr
and incidence of NRH has not been published previ- liver. Acta Pathol Jpn 32:547-560, 1982
16. Ohura S, Watanabe 1, Matsumoto T, et al: A cast of portal
ously, but in the present study we were able to examine
hypertension associated with nodular regenerative hyperplasia of the
this relationship by investigating total dosage of admin- liver in systemic lupus erythematosus. Acta Hepatol Jpn 25: 13 12-l 3 I 7,
istered glucocorticoids and incidence of NRH in 47 pa- 1984
tients with SLE. Subsequently, we found that two pa- 17. Klemp P, Timmc AH, Sayers GM: Systemic lupus erythe-
tients (cases no. 12 and 13) who had NRH had been matosus and nodular regenerative hyperplasia of the liver. An11 Rheum
Dis 45:167-170, 1986
given a large total dosage of glucocorticoids (more than 18. Matsubara Y, Shinjoh T. Nishio A. et al: A case of systemic
10 g of prednisolone) without immunosuppressive ther- lupus erythematosus associated with plexiform lesion <)f the lung and
apy. On the other hand, NRH was not present in the multiple nodular hyperplasia of the liver. Internal Med 59:977-9X0,
seven patients who had received a relatively small total 1987
19. Masumoto A, Maruoka A. Koga S. et al: A case of ~lotlular
dosage of glucocorticoids (less than 2.4 g of predniso-
regenerative hyperplasia of the liver diagnosed by needle liver biopsies
lone) without immunosuppressive therapy. These find- associated with systemic lupus erythematosus and S,jogrens syndrome.
ings suggest that administration of glucocorticoids at a Acta Hepatol Jpn 28:243-249. 1987
large total dosage may be related to the incidence of 20. Hubscher 0, Elmer B: Nodular transformation of the live1
NRH. However, we believe that more studies are needed in a patient with systemic lupus erythematosus. J Rheumatol 16:,&l U-
412, 1989
to determine the pathogenesis of NRH in SLE. 2 1. Zeek PM: Periartel-itis nodosa and other fcxms of necroti/ing
angiitis. N Engl J Med 248:764-772. 1953
Acknowlrd~mr~k The authors are grateful to D. Mrozek 22. Zizic TM, Shulman LE, Stevens MB: &Ionic, perforations in
for reviewing the manuscript. systemic lupus erythematosus. Medicine 54:4 11.4?6, 1975
23. Yoshimine T: Histopathological studies on vascular Irsions
of the liver, gallbladder. and pancreas in collagen disease>. Juntrndo
REFERENCES Igaku 33:59-73, 1987
24. Matsumoto T. I;rkusa I, Fukuda Y: Vasculo-Beht ets disvasr:
.4 pathologic btudy of eight cases. H~IM PXI.HOI. 22:45-5 I, 1991
1. Kofman S. Johnson GC. Zimmerman HJ: Apparent hepatic 2.5. Walker SH. Duncan DB: Estimation of the probability 01 .m
dysfunction in lupus erythrmatosus. Arch Intern Med 95:669-676. event as a function of several independent variablea. Biomrtrika 54:
1955 167-179. 1967
2. Gibson I. Myers AR: Subclinical liver disease in systemic lupus 26. Truett J$ Cornheld J, Kannel W: A multivariate analysis 01
erythematosus. J Rheumatol 8:751-7.59. ) 981 the risk of conmary heart disease in Framingham. J Chron Liis 20:
3. Miller MH. Urowitz MB, Gladman DD, et al: The liver in sys- 51 l-524, 1967
temic lupus erythematosus. Q J Mcd 2 I 1:40 l-409. 1984 27. Alpers DH, Sabesin SM: Fatty liver: Biochemical and clinic al
4. Klemperer- P, Pollack AD, Baehr G: Pathology of disseminated aspects. in Schif L, Schiff ER (eds): Diseases of the Liver (ed 5).
lupus erythematosus. Arch Pathol 32:569-63 I, I94 I Philadelphia, PA, Lippincot t. 1982, pp 8 13-845
5. Harvey AM, Shuhnan LE. Tumulty PA, et al: Systemic lupus 28. Anthony PP: Tumours and tumour-like lesions 01 the liver
erythematosus: Review of the literature and clinical analysis of 138 and biliary tract, in MacSween RNM, Anthony PP, Srheuer PJ (eda):
cases. Medicim 33:291-437. I954 Pathology of the Liver. Edinburgh, UK, Churchill Livingstone, 19X7.
6. Aronson AR, Montgomery MM: Chronic liver- disease with a pp 574-645
lupus erythematosus-like syndrome. Arch Intern Med 104:544-552. 29. Portmann B, MacSween RNM: Diseases of tllr intrahepatic
1954 bile ducts, in MacSween RNM, Anthony PP, Schcuel- PJ (~1s): P;I-
7. Runyon BA. Labrecqur DR, Anuras S: The spectrum of liver tholoby of the Liver. Edinburgh, UK. Churchill Livingstone. 1987,
disease in systemic lupus er-):thelllatosus. Reporr of 33 histologically- pp 424.453
proved cases and review of the literature. Am J Med 69: 187-194. 30. Nakhleh RE, Snover DC:: Use of alpha-l -antitrypsin staining
1980 in the diagnosis of nodular regenerative hvperplasia of the liver. EILIM
8. Haslock I: Spontaneous rupture of the liver in systemic lupus PATHOI. 19:1048-1052, 1988
erythematosus. Ann Rheum Dis 33:482-484, 1974 3 1. Wanlesa IR, Godwin TA, Allen F. et al: Nodular- r-egrnerative
9. Woolhng KR. Baggenstoss .4H. Weir JF: infarcts of the liver. hypel-plasia of the liver in hematologic disorder-s: A possible responsr
Gastroenterology 17:479-493, 1951 to obliterative portal venopathy. I-2morphometric study of nine casc~
10. Steiner PE: Nodular regenerative hyperplasia of the liver. with an hypothesis on the pathogenesis. Medicine 59:367-379. 1980
Am J Pathol 35943-947, 1959 32. Reynolds WJ, Wanless IR: Nodular regenerative hypelplasia
Il. Stromeyer FW, Ishak KG: Nodular transformation (nodular of the liver in a patient with rheumatoid vasculitis: A morphomen%
regenerative hyperplasia) of the liver. A clinicopathologic study of study suggesting a role for hepatic arteritis in the pathogenesis. J
30 cases. HUM PATHOI. 12:60-71, 1981 Rheumatol 11:838-842, 1984

1158