Chi nese Med i cal Jour nal (Taipei) 2002;65:563-569
Original
Characteristics of Autoimmune Hepatitis in
Taiwan: the 11 Years’ Experiences of a
Medical Center
Hui-Chun Huang1
Yi-Shin Huang1
Jaw-Ching Wu1
Shyh-Haw Tsay2
Teh-Ia Huo 1
Yuan-Jen Wang1
Jing-Chuan Lo1
Chih-Yen Chen1
Chun-Ping Li1
Full-Young Chang1
Shou-Dong Lee1
1
Di vi sion of Gastroenterology, De part ment
of Med i cine, and
2
De part ment of Pa thol ogy, Tai pei Vet erans
Gen eral Hos pi tal; and Na tional Yang-Ming
Uni ver sity School of Med i cine, Tai pei,
Tai wan, R.O.C.
Key Words
antinuclear an ti body;
au to im mune hep a ti tis
Background. Un like in West ern coun tries, au to im mune hep a ti tis (AIH) is
an in fre quent di ag no sis in Tai wan. The clin i cal char ac ter is tics of AIH in
this area are un clear. The aim of this study was to elu ci date the clin i cal features of AIH in Tai wan.
Methods. All the med i cal re cords of in-patients with the di ag no sis of
chronic hep a ti tis in our hos pi tal from 1990 to 2001 were re viewed for
the pos si bil ity of AIH. The clin i cal fea tures, bio chem i cal data, im munological pre sen ta tions, treat ments and sur vival of the pa tients were
evaluated.
Results. Twenty-two pa tients (15 fe males and 7 males) were di ag nosed as
hav ing AIH within 11 years. The me dian age at on set was 64 years (range:
17-77 years). Compared with fe male pa tients, male pa tients had older age
(p = 0.001), shorter du ra tion from ini tial pre sen ta tion of symp toms to diag no sis ( p = 0.015), lower se rum lev els of al ka line phosphatase (ALK-P, p
= 0.022) and al bu min (p = 0.043). Five (23%) pa tients pre sented with cirrho sis upon di ag no sis. Compared with non-cirrhotic pa tients, cirrhotic patients had lower serum levels of alanine aminotransferase (p = 0.002),
aspartate aminotransferase (p = 0.015), gamma-glutamyl transferase
(G-GT, p = 0.002), al bu min (p = 0.14), white cell counts (p = 0.009) and
platelet counts (p = 0.002). Thirteen (59%) patients had concomitant
clinico-pathological fea tures of cholestatic liver dis ease (ALK-P ≥ 2 times
of up per nor mal limit or patho logic ev i dence of cholangiopathy). They
had higher se rum lev els of ALK-P (p < 0.001) and G-GT (p = 0.004) than
9 non-cholestatic pa tients. There were no sig nif i cant dif fer ences in survival between these groups. The prescribed initial and maintained
prednisolone dos ages for our pa tients to con trol dis ease ac tiv ity were 19
± 15 mg and 8 ± 1 mg, re spec tively, which were lower than those rec ommended in West ern coun tries. The re mis sion rate to ste roid treat ment and
re lapse rate af ter dis con tinu ing corticosteroids were 87.5% and 50%, respectively.
Conclusions. Compared with West ern AIH pa tients, the AIH pa tients in
Tai wan are older and more likely to de velop cholestasis, and need a rel atively lower dose of ste roid for treat ment. Ow ing to one quar ter of the patients already having liver cirrhosis on diagnosis, AIH should be sus pected in any Tai wan ese pa tient with cryptogenic hep a ti tis or cir rho sis.
[Chin Med J (Taipei) 2002;65:563-569]
Re ceived: March 4, 2002. Ac cepted: August 27, 2002.
Cor re spon dence to: Yi-Shin Huang, MD, Di vi sion of Gastroenterology, De part ment of Med i cine, Tai pei Vet erans Gen eral Hos pi tal,
201, Sec. 2, Shih-Pai Road, Tai pei 112, Tai wan. Fax: +886-2-2873-9318; E-mail: yshuang@vghtpe.gov.tw
564
Hui-Chun Huang, et al.
A
utoimmune hep a ti tis (AIH) is a chronic ac tive hepa ti tis char ac ter ized by un re lent ing hepatocellular
necrosis and the presence of special serum autoantibodies. 1,2 This dis or der is not un com mon and has
been well in ves ti gated in West ern coun tries.3 Nev erthe less, the role of AIH in the pathogenesis of chronic
hep a ti tis is likely to be ne glected in Tai wan that is an
en demic area with hep a ti tis B and C vi ral in fec tion. To
the best of our knowl edge, only spo radic cases of AIH
have been re ported in Tai wan lit er a ture. 4 The clin i cal
char ac ter is tics of AIH here are un known. The pur pose
of this study was to in ves ti gate the clin i cal char ac ter istics of AIH in Tai wan.
Methods
Subjects
We ret ro spec tively re viewed all the med i cal records of in pa tients with the di ag no sis of chronic hep ati tis (code of In ter na tional Clas si fi ca tion of Dis ease:
571.40) upon dis charge from our hos pi tal from Novem ber 1990 to Au gust 2001. Au to im mune hep a ti tis
was di ag nosed based on the cri te ria pro posed by the
In ter na tional Au to im mune Hep a ti tis Group. 5,6 It is a
scor ing sys tem com posed of gen der, se rum liver biochemistry, serum glob u lin (or im mu no glob u lin G),
antinuclear an ti body (ANA), anti-smooth mus cle anti body (ASMA), anti-mitochondrial an ti body (AMA),
vi ral hep a ti tis mark ers, drug his tory, al co hol in take,
liver histology, other autoimmune disease and re sponse to ther apy. Def i nite AIH is di ag nosed when the
aggregate scores are greater than 15 before treatment.5,6 Prob a ble AIH is de fined if the scores are between 10 and 15 be fore treat ment.5,6 For strict di ag nosis, we fur ther ex cluded the prob a ble cases who had
(1) neg a tive se rum ANA; (2) pos i tive se rum AMA,
hep a ti tis B sur face an ti gen, or hep a ti tis C an ti body;
(3) im ag ing ev i dence of biliary tree ab nor mal i ties; (4)
pos i tive causal re la tion ship of re cent drug in take and
hepatic dys func tion; or (5) ha bit ual al co hol drink ing.
The clin i cal char ac ter is tics in clud ing age, gen der, dura tion from the ini tial pre sen ta tion of symp tom to diagnosis, symptoms, signs, associated autoimmune
Chi nese Med i cal Jour nal (Taipei) Vol. 65 No. 12
disease, biochemical data, serum ANA, ASMA,
AMA and response to treatment were re corded for
fur ther eval u a tion.
The se rum autoantibodies, in clud ing ANA, ASMA
and AMA markers, were measured by an indirect
immunofluorescence assay using the Fluoro-KitTM
(Incstar Corp. Stillwater, Min ne sota, USA) ac cord ing
to the man u fac turer’s in struc tions. Liver bio chem i cal
tests were measured by an auto-analyzer (Hitachi
Model 736 au to matic an a lyzer; Hitachi, To kyo, Ja pan).
Statistics
Nu mer i cal data were ex pressed as mean ± SD.
The Fisher’s ex act test or Mann-Whitney U test was
used to com pare the pa ram e ters be tween groups as
appropriate. The Kaplan- Meier model with log rank
test was per formed for sur vival anal y sis. A p value
less than 0.05 was con sid ered sta tis ti cally sig nif i cant.
R esults
Twenty-two (1.62%) of the 1,357 pa tients with
chronic hep a ti tis were di ag nosed to have AIH, including 11 def i nite and 11 prob a ble cases ac cord ing
to the Scoring sys tem.5,6 The ini tial symp toms/signs
are sum ma rized in Ta ble 1. Of the 22 AIH pa tients,
15 (68%) were fe males and 7 (32%) were males (Table 2). The me dian age upon the on set of symp toms
was 64 years (range: 17-77 years) whereas that upon
Table 1. Initial symptoms/signs of 22 Chinese patients
with autoimmune hepatitis
Symptoms/signs
No. (%)
Poor appetite
Abdominal fullness
Fever
Jaundice
General malaise
Nausea
Abdominal pain
Pruritus
Fatigue
Weight loss
5 (22.7)
5 (22.7)
4 (18.2)
4 (18.2)
3 (13.6)
2 (9.1)
2 (9.1)
2 (9.1)
2 (9.1)
2 (9.1)
De cem ber 2002
Au to im mune Hep a ti tis in Tai wan 565
Table 2. The characteristics of 22 Chinese patients with autoimmune hepatitis before steroid treatment
Parameters
Age (years, upon onset)
Age (years, upon diagnosis)
Female/male
Duration (weeks) a
ANAb
ALT (U/L)
AST (U/L)
ALK-P (U/L)
G-GT (U/L)
DB (mg/dl)
TB (mg/dl)
Globulin (g/dl)
Albumin (g/dl)
IgG (mg/dl)
WBC (/cumm)
Hemoglobin (g/dl)
Platelet (/cumm)
Patient values
Reference values
54 ± 20
55 ± 20
15/7
76 ± 184
1:798 (1:80-1:2560)
513 ± 416
539 ± 476
218 ± 104
275 ± 267
4.1 ± 4.7
5.6 ± 6.8
4.0 ± 0.9
3.5 ± 0.6
3,021 ± 1,135
5,454 ± 2,270
11.9 ± 2.4
175,368 ± 91,384
Undetectable
0-40
5-45
10-100
8-61
0-0.3
0.2-1.6
0.9-4.7
3.7-5.3
751-1,560
4,800-10,800
14-18
130,000-400,000
Data are expressed as Mean ± SD except otherwise stated.
a
Duration: the time interval from initial symptoms to definite diagnosis; bgeometric mean (range).
ANA = antinuclear antibody; ALT = alanine aminotransferase; AST = aspartate aminotransferase; ALK-P = alkaline
phosphatase; G-GT = gamma-glutamyl transferase; DB = direct bilirubin; TB = total bilirubin; IgG = immunoglobulin G;
WBC = white blood cell count.
Table 3. Comparison of pretreatment data between male and female patients with autoimmune hepatitis
Parameters
Age (years)a
Duration (weeks) b
ALT (U/L)
AST (U/L)
ALK-P (U/L)
G-GT (U/L)
DB (mg/dl)
TB (mg/dl)
Globulin (g/dl)
Albumin (g/dl)
IgG (mg/dl)
WBC (/cumm)
Hemoglobin (g/dl)
Platelet (/cumm)
Male
(n = 7)
Female
(n = 15)
p
72 ± 5
44 ± 78
603 ± 431
555 ± 404
150 ± 49
147 ± 83
4.7 ± 6.5
7.1 ± 9.3
4.1 ± 1.0
3.1 ± 0.3
3,400 ± 746
5,791 ± 2,011
12.7 ± 2.4
140,286 ± 62,543
47 ± 20
91 ± 217
470 ± 417
532 ± 520
250 ± 108
335 ± 304
3.8 ± 3.6
4.8 ± 5.3
4 ± 0.9
3.7 ± 0.7
2,894 ± 1,239
5,286 ± 2,443
11.6 ± 2.5
195,833 ± 101,448
0.001
0.630
0.418
0.832
0.022
0.217
0.606
0.794
0.691
0.043
0.430
0.881
0.262
0.272
Data are expressed as Mean ± SD.
a
Age: upon diagnosis.
b
Duration: the time interval from initial symptoms to definite diagnosis.
Abbreviated as Table 1.
diagnosis was 65 years (range: 17-77 years). The pretreat ment char ac ter is tics and re sults of lab o ra tory survey of all AIH patients are shown in Table 2.
Compared with fe male pa tients, the male pa tients displayed older age, lower se rum al bu min and al ka line
phosphatase (ALK-P) lev els upon the time of di ag no-
566
Hui-Chun Huang, et al.
Chi nese Med i cal Jour nal (Taipei) Vol. 65 No. 12
Table 4. Comparison of pretreatment data between cirrhotic and non-cirrhotic patients with autoimmune hepatitis
Parameters
Age (years) a
Female/male
Duration (weeks)b
ALT (U/L)
AST (U/L)
ALK-P (U/L)
G-GT (U/L)
DB (mg/dl)
TB (mg/dl)
Globulin (g/dl)
Albumin (g/dl)
IgG (mg/dl)
WBC (/cumm)
Hemoglobin (g/dl)
Platelet (/cumm)
Cirrhotic
(n = 5)
Non-cirrhotic
(n = 17)
p
60 ± 22
3/2
209 ± 360
100 ± 102
154 ± 181
177 ± 127
71 ± 37
0.9 ± 0.6
1.7 ± 0.6
4.2 ± 0.4
2.8 ± 0.4
3,650 ± 1,143
3,004 ± 1,465
10.0 ± 3.5
79,600 ± 24,876
53 ± 20
12/5
37 ± 69
634 ± 394
652 ± 479
230 ± 97
335 ± 277
5.2 ± 5.0
6.8 ± 7.4
4.0 ± 1.0
3.7 ± 0.6
2,875 ± 1,127
6,048 ± 1,985
12.5 ± 1.8
209,571 ± 81,149
0.446
0.762
0.002
0.015
0.256
0.002
0.137
0.247
0.636
0.014
0.545
0.009
0.083
0.002
Data are expressed as Mean ± SD.
a
Age: upon diagnosis.
b
Duration: the time interval from initial symptoms to definite diagnosis.
Abbreviated as Table 1.
Table 5. Comparison of pretreatment data between cholestatic and non-cholestatic patients with autoimmune
hepatitis
Parameters
Age (years) a
Female/male
Duration (weeks)b
ALT (U/L)
AST (U/L)
ALK-P (U/L)
G-GT (U/L)
DB (mg/dl)
TB (mg/dl)
Globulin (g/dl)
Albumin (g/dl)
IgG (mg/dl)
WBC (/cumm)
Hemoglobin (g/dl)
Platelet (/cumm)
Cholestatic
(n = 13)
Non-cholestatic
(n = 9)
p
53 ± 20
10/3
44 ± 77
452 ± 378
504 ± 506
269 ± 100
381 ± 300
3.2 ± 3.5
3.4 ± 3.6
3.9 ± 0.9
3.7 ± 0.7
2,954 ± 1,399
5,308 ± 2,455
10.0 ± 3.5
176,500 ± 70,304
57 ± 21
5/4
122 ± 275
600 ± 474
590 ± 454
144 ± 55
123 ± 92
5.3 ± 5.8
8.5 ± 8.9
4.2 ± 1.0
3.2 ± 0.5
3,106 ± 774
5,693 ± 2,069
12.5 ± 1.8
174,111 ± 115,004
0.512
0.896
0.744
0.601
0.004
0.014
0.295
0.095
0.331
0.095
0.918
0.804
0.247
0.720
Data are expressed as Mean ± SD.
a
Age: upon diagnosis
b
Duration: the time interval from initial symptoms to definite diagnosis.
Abbreviated as Table 1.
sis (Ta ble 3). Five pa tients (23%) pre sented with cirrho sis upon di ag no sis. They had sig nif i cantly lower
serum levels of alanine aminotransferase (ALT),
aspartate aminotransferase (AST), gamma-glutamyl
transferase (G-GT), al bu min, white blood cell count
and platelet count (Ta ble 4). Thir teen (59%) pa tients
De cem ber 2002
who had a more than 2-fold in crease of ALK-P and/or
histologic evidence of biliary tree injury were regarded as cholestatic group (Table 5). They had
higher se rum lev els of ALK-P and G-GT com pared
with non-cholestatic pa tients (Ta ble 5).
Four pa tients were as so ci ated with au to im mune
diseases, in clud ing 3 with Sjögren’s syn drome and
one with sys temic lupus erythematosus.
Ste roid ther apy was in sti tuted to 16 pa tients. Two
pa tients lost fol low-up af ter ini tial treat ment. The remaining 14 patients responded well (a decline of
ALT) in an av er age of 27± 13 days. The ALT level before ste roid ther apy in sti tu tion was 697 ± 386 U/L and
re sponded to the ther apy to fall to 71 ± 57 U/L. They
ex pressed nor mal ized liver en zyme lev els (87.5%) after an av er age of 57 ± 22 days. The av er age time of
main te nance was 49 ± 40 months. The mean ini tial
and main tained doses of prednisolone were 19 ± 15
mg and 8 ± 1 mg/day, re spec tively. Among the 14 patients, 1 pa tient dis con tin ued ste roid treat ment due to
per sis tent nor mal iza tion of liver bio chem is try tests; 4
pa tients un der went ta per ing of ste roids upon the end
of our fol low-up; whereas 7 pa tients (50%) suf fered
from ag gra va tion of liver bio chem is try data af ter taper ing ste roids. Ste roid ther apy was not pre scribed to
6 pa tients, be cause 5 of them were cirrhotic pa tients
with nearly nor mal liver en zyme lev els and 1 pa tient
lost fol low-up af ter clin i cal di ag no sis.
Liver bi opsy was per formed in 12 pa tients, and
had ev i dence of periportal necroinflammation (in terface hep a ti tis) (Fig. 1) in all. Con com i tant non spe cific
biliary tree dam age was found in 9. Of these 9 patients, 6 re sponded to ste roid ther apy well with normal iza tion of liver en zyme lev els. The other 3 pa tients
did not re ceive ste roid ther apy be cause of ini tial di agno sis of cir rho sis in 2 and loss of fol low-up in 1.
Three pa tients died of liver cir rho sis with esoph ageal variceal bleed ing, hepatic decompensation, and
pneu mo nia with sep tic shock, re spec tively and 1 patient lost fol low-up af ter di ag no sis, 2 pa tients lost follow-up af ter ini tial treat ment. There was no sig nif icant dif fer ence in sur vival be tween fe male and male
pa tients (72 ± 12 vs. 54 ± 19 months, p = 0.6020),
cholestatic and non-cholestatic (75 ± 9 vs. 55 ± 22
months, p = 0.3689), and cirrhotic and non-cirrhotic
Au to im mune Hep a ti tis in Tai wan 567
Fig.1. The liver pa thol ogy of an au to im mune hep a ti tis patient re vealed mononuclear cell in fil tra tion at the
peri-portal zone, with ne cro sis or de gen er a tion of
hepatocytes, which shows a typ i cal char ac ter is tic of
interphase hep a ti tis (H&E stain, 40 X).
pa tients (81 ± 43 vs. 65 ± 9 months, p = 0.6769).
D iscussion
AIH is a dis ease of un known eti ol ogy, char ac ter ized
by female predominance, hypergammaglobulinemia,
circulating antibodies, hepatocytic damage, and
marked re sponse to immunosuppressive ther apy. 1,2
Ac cu rate di ag no sis for AIH is in deed im por tant for an
ap pro pri ate ther apy. 7 How ever, the spec trum of au toimmune hepatic injury varies markedly from pure
hepatocytic to biliary ep i the lial dam age, 1,8,9 which occa sion ally makes the dis tinc tion quite dif fi cult. Since
the cur rent study was per formed ret ro spec tively, to
avoid cases with con tro ver sies re gard ing in ad e quate
study or in def i nite di ag no sis, out-patients and those
with AMA positivity were ex cluded. Se lec tion bias
may be pres ent un der such a set ting, but the re main ing
cases can be eval u ated as an AIH group un doubt edly.
Au to im mune hep a ti tis can be clas si fied as type 1
with the pres ence of ANA, ASMA and type 2 with
liver-kidney microsomal an ti body (LKM1).10,11 Further more, the re la tion ships of AIH clas si fi ca tion and
clinical characteristics to HLA-DR3 and -4 status
have been re ported.12 Due to the ret ro spec tive na ture
of our study, how ever, LKM1 and HLA tests were not
568
Hui-Chun Huang, et al.
per formed upon ini tial pre sen ta tion of the pa tients.
In our se ries, fe male to male ra tio was 2.1, which
was sig nif i cantly lower than that re ported for the Jap anese (ra tio: 6:1) 13 and for the West erns (ra tio: 8:1).14 It
may count that our hos pi tal serves mainly for vet er ans.
Al though AIH usu ally af fects young fe males and the
meno pause women in West ern ar eas,3,14 our AIH patients seemed older at on set (54 ± 20 years). Sim i larly,
two Jap a nese se ries re ported the on set at older age, too
(50 and 70 years, re spec tively).13,15
Thir teen of our AIH pa tients had el e vated ALK-P
and G-GT lev els and/or cholangiopathy ver i fied by
pa thol ogy. How ever, the fol low ing rea sons still support the di ag no sis of AIH in these pa tients: 1) high titer of ANA with out AMA; 2) el e vated se rum lev els of
im mu no glob u lin G in stead of M, which is con sis tent
with the fea tures of AIH rather than pri mary biliary
cir rho sis (PBC); 3) the ab sence of biliary tree dis ease
to de velop over a me dian fol low-up of 91 months; 4)
except 2 patient lacking follow-up, those who received corticosteroid ther apy sig nif i cantly im proved
with nor mal iza tion of liver bio chem is try data. This
fur ther sup ported the di ag no sis of AIH;16 and 5) no
biliary tree ab nor mal ity iden ti fied by sonography. The
cause of a higher per cent age of cholestatic pa tients in
our AIH pa tients, how ever, re mains to be clar i fied.
Au to im mune cholangitis can be very sim i lar to
PBC re gard ing its clin i cal, bio chem i cal, and histopathological fea tures ex cept neg a tive AMA and pos itive ANA. Its clas si fi ca tion in the field of au to im mune
liver disease remains uncertain. 17,18 Autoimmune
cholangitis may be a vari ant of pri mary biliary cir rhosis.18,19 Though un cer tainty ex ists,20-22 all of our patients, ex cept 5 cirrhotic pa tients, had at least 3-fold
el e va tion of alanine transaminase lev els and/or pathological evidence of interphase hepatitis. Con sequently, the di ag no sis of AIH with cholestatic fea tures
rather than au to im mune cholangitis was es tab lished.
In the West ern ex pe ri ences, the rec om mended initial dose of prednisolone is 60 mg/day, and the main tenance dose is 20 mg/day.2 In con trast, we pre scribed
the lower doses for the AIH pa tients (the ini tial and
main te nance doses of prednisolone were 19 ± 15 and
8 ± 1 mg, re spec tively). The re mis sion rate in our series was 87.5%, which was com pa ra ble with that of a
Chi nese Med i cal Jour nal (Taipei) Vol. 65 No. 12
Sin ga pore se ries (89%) 23 but higher than the West ern
re ports (65%).24 The av er age time to re mis sion was
also significantly shorter in our group than in the
Westerns.24 Whether the Asian AIH pa tients ex ert the
higher treat ment re spon sive ness and the lower dose
re quire ment to ste roid re main to be clar i fied.
Five (23%) pa tients were found to de velop cir rhosis upon di ag no sis. This im pli cates that the pa tient may
de lay to con sult med i cal aids or the lo cal phy si cians
may over look to di ag nose AIH. The mean sur vival of
cirrhotics was not sig nif i cantly dif fer ent from that of
non-cirrhotics, which may be re lated to the rel a tively
small num ber of cirrhotic pa tients in our se ries.
Ac cord ing to the cur rent sur vey, non-cholestatic
pa tients pre sented with higher lev els of to tal and di rect
bilirubin than cholestatic pa tients, though sta tis ti cal
sig nif i cance was not found. The pos si ble ex pla na tion is
that the non-cholestatic pa tients are more likely to have
hepatocellular dam age, in stead of biliary tract in jury,
than cholestatic pa tients. Pro found hepatocellular damage may cause jaun dice. The fol low ing con di tions may
sup port this in fer ence. First, com pared with cholestatic
cases, non-cholestatic cases tended to have higher lev els of ALT and AST, al though the dif fer ence was not
sta tis ti cally sig nif i cant. This may im pli cate the more
ac tive pro cess of hepatocellular de struc tion and necro sis in non-cholestatic cases. As cases that re ceived
liver pa thol ogy sur vey in this group were lim ited, this
point of view can still not be patho log i cally con firmed
at pres ent. Sec ond, 4 out of 5 cirrhotic pa tients were
non-cholestatic. Among them, 1 pa tient evolved into
liver decompensation with re mark ably high lev els of
to tal and di rect bil i ru bin. This may cause high mean
bil i ru bin level in non-cholestatic group.
In sum mary, cholangiopathy and cir rho sis are not
un com mon in our Chi nese AIH pa tients, and they tend
to be older than West ern pa tients. Since it can be controlled with a lower dose of ste roid here, the di ag no sis
of AIH should be kept in mind when ever a Chi nese
pa tient with cryptogenic hep a ti tis is un der eval u a tion.
References
1. John son PJ, McFarlane IG, Eddleston ALWF. The nat u ral
De cem ber 2002
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
course and het er o ge ne ity of au to im mune type chronic ac tive
hep a ti tis. Semin Liver Dis 1991;11:187-96.
Czaja A. Au to im mune chronic ac tive hep a ti tis. In: Czaja A,
Dick son ER, eds. Chronic Ac tive Hep a ti tis: The Mayo Clinic
Experience. New York: Dekker, 1986:105-26.
Borberg KM, Aadland E, Jahnsen J, Raknerud N, Stiris M,
Bell H. In ci dence and prev a lence of pri mary biliary cir rhosis, pri mary sclerosing cholangitis, and au to im mune hep a titis in a Nor we gian pop u la tion. Scand J Gastroenterol 1998;
33:99-103.
Hou TI, Wu JC, Huang YS, Lai CR, Tsay SH, Lee SD. Au toim mune hep a ti tis in an area en demic for hep a ti tis B vi rus infec tion: a re port of three cases. Chin Med J (Tai pei) 1996;
57:360-4.
Alvarez F, Berg PA, Bianchi FB, Bianchi L, Bur roughs AK,
Cancado EL, et al. International Autoimmune Hepatitis
Group Re port: re view of cri te ria for di ag no sis of au to immune hep a ti tis. J Hepatol 1999;31:929-38.
John son PJ, McFarlane IG. Meet ing re port: In ter na tionalAuto im mune Hep a ti tis Group. Hepatology 1993;18:998-1005.
Papo T, Marcellin P, Bernuau J, Durand F, Poynard T,
Benhamou JP. Au to im mune chronic hep a ti tis ex ac er bated by
al pha-interferon. Ann In tern Med 1992;116:51-3.
Chazouillères O, Wendum D, Serfaty L, Montembault S,
Rosmorduc O, Poupon R. Primary biliary cirrhosisautoimmune hep a ti tis over lap syn drome: clin i cal fea tures
and re sponse to ther apy. Hepatology 1998;28:296-301.
Lohse AW, Büschenfelde KHMZ, Franz B, Kanzler S,
Gerken G, Dienes HP. Char ac ter iza tion of the over lap syndrome of pri mary biliary cir rho sis (PBC) and au to im mune
hep a ti tis: ev i dence for it be ing a hepatitic form of PBC in genetically susceptible in di vid u als. Hepatology 1999;29:
1078-84.
Macfarlane IG. The re la tion ship be tween au to im mune markers and dif fer ent clin i cal syn dromes in au to im mune hep a ti tis.
Gut 1998;42:599-602.
Czaja A, Manns M, Homburger H. Fre quency and sig nif icance of an ti bod ies to liver/kid ney microsome Type 1 in
adults with chronic hep a ti tis. Gastroenterology 1992;103:
1290-5.
Doherty DG, Donaldson PT, Underhill JA, Farrant JM,
Duthie A, Mieli-Vergani G, et al. Allelic se quence vari a tion
in the HLA class II genes and pro teins in pa tients with au toim mune hep a ti tis. Hepatology 1994;19:609-15.
Au to im mune Hep a ti tis in Tai wan 569
13. Omagari K, Kinoshita H, Kato Y, Nakata K, Kanematsu T,
Kusumoto Y, et al. Clin i cal fea tures of 89 pa tients with au toimmune hepatitis in Nagasaki Pre fec ture, Japan. J
Gastroenterol 1999;34:221-6.
14. Sherlock S, Dooley J. Chronic hep a ti tis:gen eral fea tures and
au to im mune chronic dis ease. In: Sherlock S, Dooley J, eds.
Dis eases of the Lliver and Biliary Sys tem. Ox ford: Blackwell,
2002:321-33.
15. Toda G, Zeniya M, Watanabe F, Imawari M, Kiyosawa K,
Nishioka M, et al., and the Jap a nese Na tional Study Group of
Au to im mune Hep a ti tis. Pres ent sta tus of au to im mune hep atitis in Ja pan-correlating the char ac ter is tics with in ter national cri te ria in an area with a high rate of HCV in fec tion. J
Hepatol 1997;26:1207-12.
16. Geubel AP, Baggentoss AH, Summerskill WHJ. Re sponses
to treat ment can dif fer en ti ate chronic ac tive liver dis ease
with cholangitis fea tures from the pri mary biliary cir rho sis
syn drome. Gastroenterology 1976;71:444-9.
17. Colombato LA, Alvarez F, Côté J, Huet PM. Au to im mune
cholangiopathy: the re sult of con sec u tive pri mary biliary cirrho sis and au to im mune hep a ti tis? Gastroenterology 1994;
107:1839-43.
18. Michieletti P, Wanless IR, Katz A, Scheuer PJ, Yeaman SJ,
Bassendine MF, et al. Antimitochondrial an ti body neg a tive
pri mary biliary cir rho sis: a dis tinct syn drome of au to im mune
cholangitis. Gut 1994;35:260-5.
19. Goodman ZD, McNally PR, Davis DR, Ishak KG.
Autoimmune cholangitis: a vari ant of pri mary biliary cirrho sis. Clinicopathologic and se ro logic cor re la tions in 200
cases. Dig Dis Sci 1995;40:1232-42.
20. Mackay IR, Wood IJ. Lupoid hep a ti tis: a com par i son of 22
cases with other types of chronic liver disease. Q J Med
1962;31:485-507.
21. NcNair ANB, Moloney M, Portmann BC, Williams R,
MaFarlane IG. Au to im mune hep a ti tis over lap ping with primary sclerosing cholangitis in five cases. Am J Gastroenterol
1998;93:777-84.
22. Mackay IR. Au to im mune dis ease over laps and the liver: two
for the price of one? J Gastroenterol Hepatol 2000;15:3-8.
23. Lee YM, Teo EK, NG TM, Khor C, Fock KM. Au to im mune
hep a ti tis in Sin ga pore: a rare syn drome af fect ing mid dleaged women. J Gastroenterol Hepatol 2001;16:1384-9.
24. Czaja AJ. Autoimmune hepatitis: evolving concepts and
treat ment strat e gies. Dig Dis Sci 1995;49:435-56.