close

Вход

Забыли?

вход по аккаунту

?

The long-term effect of methotrexate therapy on the liver in patients with juvenile rheumatoid arthritis.

код для вставкиСкачать
ARTHRITIS & RHEUMATIF’LI
Vol 40, No 12, December 1997, pp 2226-2234
0 1997, American College of Rheumatology
2226
THE LONG-TERM EFFECT OF METHOTREXATE THERAPY ON THE
LIVER IN PATIENTS WITH JUVENILE RHEUMATOID ARTHRITIS
PHILIP J. H A S H E S , WILLIAM F. BALISTRERI, KEVIN E. BOVE, EDGAR T. BALLARD,
and MURRAY H. PASS0
Objective. To determine if the long-term use of
methotrexate (MTX) in juvenile rheumatoid arthritis
(JRA) is associated with the development of significant
liver fibrosis, and to describe the presence of risk
factors for liver fibrosis in patients with JRA.
Methods. Needle biopsies of the liver were performed on a cross-section cohort of 14 patients with JRA
who had received a total cumulative dose of MTX that
was either >3,000 mg or >4,000 md1.73 m2 of body
surface area. Biopsy samples were independently graded
according to the Roenigk Classification Scale by 2
pathologists. The presence of risk factors for MTX
hepatotoxicity, especially biochemical abnormalities reflective of liver injury and alcohol consumption, were
assessed.
Results. Thirteen biopsy samples (93%) were classified as grade I, and 1 (7%) as grade II; none demonstrated significant fibrosis. However, histologic abnormalities were found in 13 biopsy samples (93%). Only 2
patients (14%) consumed more than 1 alcoholic drink
per month. Thirteen patients (93%) had biochemical
abnormalities while being treated with MTX, but only 5
patients (36%) had at least 1determination in which the
aspartate or alanine aminotransferase elevation was >3
times the upper limit of normal.
Conclusion. Long-term use of MTX for JRA does
not appear to be associated with the development of
significant liver fibrosis. Although nearly all patients
had minor histologic changes, no significant clinical
Supported by thc Childrcn’s Hospital Research Foundation
of Cincinnati, the Schmidlapp Foundation, the USPHS (grant M01RR-08084 from thc GCRC program), the National Center for Research Resources, and thc NIH (AR-42632).
Philip J. Hashkes, MD, MSc, William F. Balistreri, MD, Kevin
E. Bove, MD, Edgar T. Ballard, MD, Murray H. Passo, MD: Children’s Hospital Medical Center, Cincinnati, Ohio.
Addrcss reprint requests to Murray H. Passo, MD. Division of
Rheumatology, Pav 2-1 29, Children’s Hospital Medical Center, 3333
Rurnet Avenue, Cincinnati, OH 45229.
Submitted for puhlication June 3, 1997; accepted in revised
form July 11, 1997.
consequences were apparent. A prospective study of a
larger population will more accurately define the incidence of MTX-related liver fibrosis and appropriate
monitoring guidelines in JRA.
Low-dose methotrexate (MTX) was the only
second-line drug found to be effective in the treatment
of juvenile rheumatoid arthritis (JRA) in a double-blind
placebo-controlled study (1). The estimated prevalence
of JRA in the United States is -l/l,OOO children (2,3),
and -39% of patients with JRA are currently taking
MTX (4). With a population of -65,000,000 children
under the age of 16 in the US ( 5 ) , -25,000 children with
JRA are presently being treated with MTX. Many JRA
patients may relapse following discontinuation of MTX
(6), which indicates that MTX may be needed on a
long-term basis.
One of the most significant long-term side effects
of low-dose MTX reported in adults is the development
of liver fibrosis or cirrhosis (7-9). Alcohol use is the
strongest risk factor (10,1l), and the cumulative dose
(CD) of MTX may also increase the risk of developing
liver fibrosis (10). Evidence of liver biochemical abnormalities may be helpful in screening patients with rheumatoid arthritis (RA) who are at high risk of developing
liver fibrosis (9), but liver biopsy is currently the only
reliable method of diagnosis (12). Studies of the incidence of MTX-related hepatotoxicity in RA and psoriasis have been reported, and guidelines for liver monitoring have been published (9,13). The estimated
frequency of significant liver fibrosis in R A is 1.1-2.7%
(9,10), while the estimated risk of cirrhosis or clinically
significant liver disease after 5 years of MTX use is only
l/l,OO0 (9,14).
There are few case reports or clinical series from
the JRA population, and there are no published guidelines for liver monitoring. Only 32 liver biopsies of JRA
patients treated with MTX have been reported (15-18);
the highest CD of MTX among those patients was 2,980
LONG-TERM EFFECT OF MTX ON THE LIVER IN PATIENTS WITH JRA
mg. Cirrhosis or severe liver fibrosis was not found in
any biopsy tissue, and mild fibrosis was found in only 2
(15,16). Although children may have fewer risk factors
than adults, the dose of MTX per body surface area
(BSA) is greater in JRA than RA (19). In addition, the
liver grows in size throughout childhood in a pattern that
resembles that of height and weight (20). The effect of
MTX on a growing liver is unknown.
We therefore performed a systematic study in
patients with JRA to determine whether the long-term
use of MTX is associated with the development of
significant liver fibrosis. Needle biopsy samples of the
liver were obtained from JRA patients who had taken a
CD of MTX of >3,000 mg or 4,000 mgi1.73 m2 of BSA.
PATIENTS AND METHODS
Study subjects. Subjects were recruited from among
patients diagnosed with J R A according to the American
College of Rheumatology (ACR) criteria (21), who had been
treated with MTX and were currently followed up at The
Special Treatment Center (STC) for Juvenile Arthritis (Children’s Hospital Medical Center, Cincinnati, OH).
J R A patients eligible for the study were those with a
C D of MTX that exceeded either 3,000 mg or 4,000 mgi1.73 m’
of BSA. Also included were patients who were not currently
being treated with MTX but who fulfilled eligibility criteria
(“intention to treat”). The latter threshold based on BSA was
chosen to include growing patients whose absolute C D was
<3,000 mg but was very high when adjusted for adult size.
Liver biopsies of J R A patients receiving this C D have not been
reported. Causes for exclusion were patient or parental refusal,
or an increased risk of complications due to bleeding disorders.
The study received the approval of the Institutional
Review Board of Children’s Hospital Medical Center, and
informed consent was obtained from all patients or parents, as
necessary. MTX therapy was not affected by refusal to participate in the study.
The charts of all J R A patients treated with MTX were
reviewed to determine eligibility. The list of patients was
generated by a review of all STC charts as well as a computerized search of patients with the diagnosis of J R A registered
at Children’s Hospital Medical Center. For each patient,
hospital medical records, clinic charts, and ARAMIS (Arthritis, Rheumatism, and Aging Medical Information System)
forms were reviewed (when available) to tabulate the C D of
M T X and to gather evidence of liver biochemical
abnormalities.
To determine whether eligible nonparticipating subjects differed significantly from the study subjects (selection
bias), a comparison of basic demographic, disease-related, and
MTX-related characteristics of both groups was performed.
Liver biopsy procedure. All patients stopped taking
nonsteroidal antiinflammatory drugs (NSAJDs) 5-7 days prior
to the biopsy. Laboratory testing the morning of the biopsy
included a complete blood cell count including a platelet
count, a partial thromboplastin time, prothrombin time, and
2227
bleeding time. A percutaneous liver biopsy was performed
under conscious sedation with intravenous midazolam and
meperidine and with 1% lidocaine as local anesthetic. Mcnghini’s thin-bore needle (William Schmidt Inc., Valencia, CA)
was passed through a small right intercostal cut in the skin, and
a core of liver was obtained using a I-second vacuum aspiration. The procedure was carried out in the Clinical Research
Center, and the patient was discharged after overnight observation for major complications. Complications wcre treated
appropriately.
Determination of histologic grade. Following formalin
fixation, slides were stained with heinatoxylin and eosin for
evaluation of cellular and general cytoarchitecture, and with
reticulin and trichrome for detection of fibrosis.
The slides were reviewed independently by 2 pathologists (KEB and ETB) who were blinded to the patients’
identity, clinical history, and previous biopsy results and
graded according to the Roenigk classification scale ( 13).
Briefly, grade I denotes normal histology or mild changes.
Grade I1 includes moderate-to-severe hepatocellular, inflammatory, fatty, or necrotic changes. Grade I11 is subdivided;
IIIA includes mild fibrosis extending from the portal tract to
the lobules. and grade TIIS includes moderate-to-severe fibroThe biopsy tissues were
sis. Grade IV represents cirrho
reviewed at 2 sessions and were mixed with slides from other
MTX-treated J R A patients who were not part of the study.
Both pathologists had previous experience in thc evaluation of
liver biopsy samples for drug toxicity. Disagreements in determining the Roenigk grade were resolved by discussion between
the pathologists.
Specimens were also assessed by a set of 7 potential
histopathologic descriptions, including hepatocellular nuclear
changes, macrovesicular steatosis, portal inflammation, lobular
inflammation, portal fibrosis, cholcstasis, and zone 3 (central
lobular) changes. Specific zone 3 changes included sinusoidal
dilation, with attenuation o f hcpatocyte plates. Each finding
was characterized as not present, minimal, mild, moderate,
marked, or of intermediate degree.
Data collection. Data relating to basic demographic
and diseasc characteristics were obtained from patient charts,
including the patient’s age, sex, J R A onset and course subtype,
and disease duration. Data related to MTX administration
included height, weight, BSA, duration of MTX therapy,
weekly dose (22), CD, route of administration (oral or parenteral) (23),and multivitamin (includes 0.4 mg of folic acid) and
folic acid supplementation ( I mgiday) (24).
Risk factors with possible association to MTX-induced
hepatotoxicity were assessed. These included potential hepatotoxic comorbidities (1 1,12), diabetes mellitus (12), alcohol
use (10,11), obesity (11) (defined as B body mass index [BMI]
>25 kg/m2 [25]), and concurrent use of other medications
(9,ll). The frequency of the following liver biochemical abnormalities was documented: aspartate aminotransferase (AST),
alanine aminotransferase (ALT), and albumin. Any elevation
of AST and ALT or decrease in albumin beyond the normal
range, as specified by the laboratory performing the test, was
defined as abnormal (9).
Assessment of alcohol consumption. Alcohol use was
assessed in all participating patients older than 11 years by
administration of an oral questionnaire. The questionnaire.
which was adapted from the Centers for Disease Control and
HASHKES ET AL
2228
Table 1. Patient, disease, and medication data of the 14 study
subjects'
Age, years
Height. cm
Weight, kg
Body niass index, kgim'
Body surface area. m'
Sex
Females
Males
Disease onset type
Systemic
Polyarticular
Pauciarticular
Disease course
Systemic
Polyarticular
Antinuclear antibody
Rheumatoid factor
Past use of NSAIDs
Past usc of 3tcroids
Past use of SAARDs
Mean 2 SD
or number
Rangc
17.5 i 4.4
10.7 -t 4.8
155.1 +- 14.3
54.6 ? 15.9
22.4 ? 4.8
1 .s 2 0.3
9.0-24.0
4.3-17.0
124.6-1 75.8
25.2-78.4
16.2-3 1.2
0.9-2
10
4
6
I
1
3
I1
7
4
14
9
12
NSAIDs = nonstcroidal antiinflammatory drugs; SAAKDs
acting antirheumatic drugs.
=
slow-
Prevention Youth Risk Behavior Survey (26), was modeled
specifically for this study, and was administered on t h e day of
t h e study. Alcohol use was reported as t h e number of drinks
(12-oz can of beer, 6-oz glass of wine, or I-oz shot of distilled
spirits) per month since starting MTX.
Statistical analysis. Descriptive statistics of patient
characteristics, MTX treatment, liver biochemical abnormalities, and risk factors for hepatotoxicity w e r e reported. Student's t-test, 2-tailcd f o r 2 independent samplcs, chi-square,
and Fisher's exact tests were used as appropriate to compare
t h e characteristics of t h e study participants and eligible nonparticipants. A P value 50.05 was determined t o represent
statistical significance.
The prevalence r a t e of cases with grade IIIB o r IV
changes (Roenigk scale) per 3,000 nig of MTX was calculated
Table 2.
with 9S% confidence intervals (CIS)(13,27). Because this study
h a d a small sample size, there was not adequate power to
perform inferential statistics. T h e computer software utilized
for d a t a analysis was SAS f o r PC, version 608 (Cary, NC).
RESULTS
Findings in the JRA patients. One hundred
patients with JRA who had been treated with MTX were
identified. 24 of whom met the eligibility criteria. Liver
biopsies were performed on 14 patients (58%) (Table 1).
Ten patients (42%) were excluded because of patient or
parental refusal. None were excluded because of bleeding disorders.
A comparison of study participants and eligible
nonparticipants did not reveal significant differences
other than a greater number of AST abnormalities since
starting MTX in the study participants (P = 0.03). The
alcohol consumption questionnaire was not administered to nonparticipants; therefore, we were unable to
determine whether a difference in alcohol consumption
existed between the groups.
Only 1 study participant, who had hepatitis associated with Epstein-Barr virus, had a comorbidity that
could have a possible effect on liver histology. One
patient with diabetes mellitus refused to take part in the
study. Four patients (29%) were mildly obese (25).
At the time of the study, almost all patients were
being treated with NSAIDs (93%), 5 (36%) with prednisone (mean 2 SD dose 5.4 +- 3.6 mg/day, range 1-10>,
5 (36%) with other slow-acting antirheumatic drugs
(SAARDs), and 1 (7%) with intravenous immunoglobulin. Of the 5 patients treated with other SAARDs in
addition to the MTX, 2 were taking minocycline (200
mg/day), 2 hydroxychloroquine (200 mgiday), and 1 a
Methotrexate (MTX) dosage and administration data for the 14 study subjects
Mean 2 SD
or number
Current dose of MTX, mgiweek
Current dose of MTXirn', mgiweek
Maximal dose of MTX, mgiwcck
Average dose of MTX, nigiweek
Duration of MTX therapy (mos.)
Cumulative dose of MTX, nig
Past oral administration of MTX
Past parenlcral administration of MTX
Pas1 l'olic acid aupplemcntation
Current oral administration of MTX
Current parenteral administration of MTX
Current folic acid supplementation
16.1 2 12.3
10.9 t 8.6
19.8 i 9.7
13.2 i 5.3
75.9 2 24.0
3,958 t 1,002
12
11
9
5
7
9
Range
0-50
0-34.3
10-50
6.2-28.6
46.9-125.1
2.754-6,450
2,057
2,199
2,048
2,475
2,Y2Y
2,349
4,418
2,647
2,401
2,247
2,479
2,836
2,473
2,842
0
0
0
(1
(1
(0.08
0
0
3
0
0
(0.08
0
6
No. of
alcoholic
drinksirnonth
22
3
3
0
12
8
3
14
2
1
5
4
4
2
No. of
abnormal
ASTs since
starting
MTX
None
Mild
Mild
None
None
None
None
None
Mild + sinusoidal
None
None
Nonc
None
None
Fibrosis
Mild
Mild
None
Mild
Mild
None
None
Mild
Mild
None
Mild
None
None
Mild
Portal
inflammation
Steatosis
Mild
Mild
Nonc
None
Nonc
None
None
None
None
None
None
Nonc
None
Mild
Lobular
inflammation
None
None
None
Mild
None
None
None
None
Mild
None
Mild
None
Minimal
Mild
Mild
Mild
None
Mild
Minimal
Mild
Mild
Mild
Moderate-rnarked
None
Mild
Mild
Mild
Minimal
Hepatocellular
changes
Biopsy findings
None
None
None
None
None
None
None
None
None
None
None
None
None
Nonc
Cholestasis
Mild
Moderate
None
Mild
Minimal
None
Mild
Moderate
Marked
None
Mild
Mild
Minimal
Mild
Zone 3
(central
lobular)
changes
I
I
I
I
I
I
I
I
IIt
I
I
1
I
I
Roenigk
gradc
* MTX = methotrexate; AST = aspartate aminotransferasc; BSA = body surbce area (at the time of study biopsy). Fibrosis included pcriductal fibrosis limited to the portal triad
without portal expansion. Patient 9 also had mild sinusoidal fibrosis. Inflammation consisted of mononuclear cells, mainly lymphocytes. Hepatocellular changes included nuclear
plcomorphism in size and shape; atypia, and binucleated forms; nucleolar prominencc; and swelling or ballooning of hepatocytes. Zone 3 (central lobular region) changes included
sinusoidal dilation with attenuation of hepatocyte plates, and incrcased numbers and pigmentation of Kupffer’s cclls.
p Disagreement over the classification of samples from patient 9 was resolved as grade TI changes. Thc distribution of fibrosis, which included sinusoidal areas but not periportal
spurs. was not included in the definition of grade IITA.
I
2
3
4
5
6
7
8
9
10
11
12
13
14
Patient
MTX
cumulative
dose
(mg/m’
BSA)
~
Table 3. Interpretation of blopsy samples from the 14 study subJccts,*
2
P
5m
m
2
0
Z
2
3
Z
0
223 0
combination of suliasalazine (1,000 mgiday), and hydroxychloroquine (400 mgiday). Four patients (29%)
were taking oral contraceptives. Eight patients (57%)
took 51,000 mgiweek of acetaminophen.
MTX treatment. Data related to MTX administration are detailed in Table 2. Two patients were not
taking MTX at the time of the study, and 2 patients were
receiving “high-dose” MTX ( > O h 5 mglkgiweek)
(18.19). The CD of MTX ranged from 2,754 mg to 6,450
nig (mean -+ SD 3,958 t 1,002). The CD of MTX per
BSA ranged from 2,048 mg/m2 to 4,418 mg/m2 (mean f
SD 2,600 i 593) (Table 3). Five patients (36%) were
treated at the time of the study with oral MTX, and 7
(50%) were treated parenterally. Nine patients (64%)
were receiving folic acid supplements (1 mgiday) at the
time of the study. None of the other 5 patients were
taking multivitamins at the time of the study.
Biochemical abnormalities. AST levels. Blood
samples were drawn during MTX treatment every 4-14
weeks. AST elevations prior to MTX administration
were found in 4 patients (29%). Thirteen patients (93%)
had at least 1 elevation of AST while being treated with
MTX. The mean 2 SD total of AST abnormalities per
patient since starting MTX was 5.93 2 6.13 (range 0-22)
(Table 3). The average maximum number of AST abnormalities per patient over 1 year was 2.64 (?2.53 SD;
range 0-10). Five patients (36%) had significant AST
elevations (defined as 3 times the upper limit of normal)
at least once since beginning treatment with MTX. In 1
patient, this was related to Epstein-Barr virus hepatitis.
A L T levels. ALT elevations prior to MTX administration were found in only 1 patient (7%). Ten patients
(71%) had at least 1 elevation of ALT while being
treated with MTX. The mean -t SD total of ALT
abnormalities per patient since starting MTX was 4.71 ?
7.51 (range 0-27). The average maximum number of
ALT abnormalities per patient over 1 year was 2.21
(22.64 SD; range 0-9). Four patients (29%) had significant ALT elevations (defined as 3 times the upper limit
of normal) at least once since beginning treatment with
MTX. In 1 patient, this was related to Epstein-Barr virus
hepatitis.
Albumin levels. Decreased levels of albumin were
found in 7 patients (50%) prior to starting MTX and in
8 paticsits (57%) after starting MTX. The mean -+ SD
total of albumin abnormalities per patient since starting
MTX was 4.64 2 6.66 (range 0-23). In all but 1 patient,
decreased albumin was secondary to changes in acutephase reactants during active disease, not as an expression of liver dysfunction.
Alcohol consumption. Eight patients (57%) did
HASHKES ET AL
not consume alcohol (Table 3); no patient consumed
alcohol on a daily basis. Only 2 patients (14%) consumed more than 1 drink per month. One reported an
average consumption of 3 drinks per month for the 6
months prior to the study. The other patient, who had
not taken MTX for 2.5 years prior to the biopsy,
reported an average consumption of 6 drinks per month
for the previous 5 years and, occasionally, more than 5
drinks per session. No alcohol was consumed on the day
MTX was taken. Four patients (29%) consumed alcohol
on rare occasions (birthdays, etc.).
Interpretation of biopsy findings. Grade I was
found in 13 biopsy samples (93%), and grade TI in 1
(7%). None of the biopsy samples demonstrated significant fibrosis. Therefore, the prevalence rate of significant fibrosis or cirrhosis was 0% per 3,000 mg of MTX
per patient (95% C10-20%). All biopsy samples were of
adequate size (>7 portal areas per biopsy), and findings
were uniform throughout the samples, thus minimizing
the chance of a sampling error.
Thirteen patients (93%) exhibited some histologic abnormalities (Table 3). The most common were
hepatocellular nuclear changes, consisting of anisonucleosis and binucleated cells, suggesting increased cell
turnover, particularly in zone 3. Minimal-to-mild nuclear
changes were seen in 11 patients (79%) and nioderateto-marked abnormalities in 1 (7%). Minimal-to-mild
zone 3 sinusoidal dilation, with attenuation of hepatocyte plates were found in 8 patients (57%), moderate
changes in 2 (14%), and marked abnormalities in 1 (7%)
(Figure 1). Only 3 patients (21%) did not exhibit zone 3
changes. Markedly increased numbers of mitotic cells in
zone 3 were detected on a sample of 3 biopsy specimens
stained with an antibody to a mitosis-related protein
(KI-67).
Mild focal portal inflammation, consisting of
lymphocytes limited to the portal triad, was found in 8
biopsy samples (57%)). Minimal-to-mild focal lobular
hepatitis was found in 5 patients (36%). Mild macrovesicular steatosis was found in 3 biopsy samples (21%),
and mild portal fibrosis without expansion or septa
formation (thus, grade I) was found in 3 patients (21%),
1 of whom (7%) had mild sinusoidal fibrosis.
Determination of the Roenigk grade was concordant between both pathologists for 13 biopsy samples
(93%). For 1 sample (patient 9), the initial interpretation of one pathologist was grade I1 and the other was
grade I. Discussion between the pathologists resulted in
a classification of grade 11.
Two patients each had undergone 2 previous liver
biopsies, and 2 others had a single previous biopsy. In 2
LONG-TERM EFFECT OF MTX ON THE LIVER IN PATIENTS WITH JRA
2231
Figure 1. Zone 3 (ccntral lobular) changes consisting of hcpatocyte nuclcar enlargement (arrowhead),
sinusoidal dilation with attenuation of hcpatocytc plates (asterisk), and Kupffer’s cell hyperpigmentation
(seen better on diastase staining, not demonstrated in Figurc) were secn in the rna,jority of biopsy samples.
Also secn was focal intrasinusoidal inflammation, mainly mononuclear (arrow). These changes do not
appear to be progressive or associated with severe clinical conaequcnccs, and wcrc thcrefore (although not
includcd in the Roenigk Classification Scale) scored aa grade 1 (hcrnatoxylin and cosin staincd; original
magnification X 256).
of these 4 patients, progressive histologic changes were
found. Patient 9 progressed from grade I to grade I1
after an additional 3,054 mg of MTX over 4.6 years.
There was an increase in the severity of hepatocellular
nuclear changes and inflammation, with development of
mild portal and sinusoidal fibrosis. The lack of periportal
fibrotic spur\ prevented classification as grade TTTA.
Progression within grade I, with development of mild
steatosis and portal and lobular inflammation, was tound
in patient 14, who had not been treated with MTX for
2.5 years prior to his third biopsy. Both patients consumed more alcohol than did the other study participants. No progression was found in patient 12, despite
the interim development of severe hepatitis from
Epstein-Barr virus. Grade IIIA was found on initial
biopsy of patient 4. Two subsequent biopsies, including
the study biopsy after an additional 4,262.5 mg MTX
over 6.33 years, were scored as grade I, with regression
of fibrosis, inflammation, and zone 3 changes.
Complications of biopsy. Minor pain at the biopsy site, with occasional radiation to the right shoulder,
lasting up to 3 days was noted in 4 patients (29%). In 1
patient, the pain resulted in a transient decrease in blood
pressure without a concurrent increase in the heart rate
or evidence of bleeding. One patient had a transient
decrease in blood preswre related to sedation, which
was rapidly reversed upon administration of naloxone.
No significant complications, such as bleeding, infection,
or lung or viscus perforation, were encountered.
DISCUSSION
In this study, cirrhosis or significant liver fibrosis
that would have required discontinuation of MTX was
not found in any patient. This is in concordance with the
results of 32 biopsies from patients with JRA reported
previously (15-18) and of our previous clinical experience with 20 biopsies from patients with JRA. However,
in only 3 previous biopsy samples, all from our clinic, did
the CD of MTX exceed 3,000 mg. In the present study,
the fibrosis limited to the portal region in 3 study
patients (21%) was classified as grade I or grade IT. This
fibrosis is not regarded as a lesion that carries the risk of
progression to severe fibrosis or cirrhosis (28,29). How-
2232
ever, there are reports of progression to cirrhosis or
death from severe fibrosis classified as grade IIIB (30).
It appears, therefore, that long-term use of MTX
does not result in irreversible liver fibrosis in patients
with JRA, although the sample size in this study is not of
adequate power to completely exclude this possibility. If
the theoretical prevalence of severe liver fibrosis in JRA
is similar to that in R A (2.7%) (lo), the probability of
detecting at least 1 patient with liver fibrosis among a
sample size of 14 would only be 0.32. The 95% CI for
finding significant fibrosis in this study included prevalence rates from 0% to 20%. Though not detected in this
study, the development of severe fibrosis in JRA patients may still be an uncommon long-term side effect of
MTX therapy.
Nearly all patients had histologic abnormalities.
The findings of anisonucleosis, binucleated cells, sinusoidal dilation with “thinning” of hepatocyte layers, and
cell loss may be related to increased cell turnover with
inadequate cell regeneration. Other evidence of increased cell turnover included an increased number of
mitotic cells seen with staining for a mitosis-related
protein and zone 3 Kupffer’s cell enlargement with
increased periodic acid-Schiff-positive residual bodies,
suggesting macrophage phagocytosis of apoptotic hepatocytes. I t is possible that folic acid depletion induced by
MTX contributes to an imbalance of cell destruction and
formation (31,32).
While anisonucleosis and binucleated cells are
characteristic of MTX therapy (2Y,31,32), zone 3
changes were reported only rarely in MTX-treated patients (30) and in a mouse model of MTX hepatotoxicity
(33). Zone 3 changes are associated with immunosuppressive medications such as azathioprine (34), with
high-dose (toxic-level) acetaminophen treatment (34),
with ischemia, and with liver transplant rejection (35).
All our patients used acetaminophen only on a sporadic
basis (maximum 1,000 mg/week). Zone 3 changes were
not reported among the minor histologic abnormalities
commonly seen in patients with RA (31), systemic-onset
JRA (8,36), and NSAID-related hepatotoxicity (37).
Therefore, it is likely that the zone 3 changes consistently seen in this study are associated with MTX;
although, without control biopsies of non-MTX-treated
JRA patients no definite conclusions can be made.
Repeat biopsies of 4 patients with zone 3 changes did
not reveal progression to clinically significant lesions or
spread to other zones of the liver.
Eight patients (57%) had mild portal inflammation, and 5 (36%) had mild lobular inflammation. Almost all patients were being treated with NSAIDs at the
H A S H E S ET AL
time of their biopsy. Similar nonspecific findings were
found in 8 JRA patients with NSAID-related hepatotoxicity (37) and in patients with systemic-onset JRA (36).
However, portal fibrosis was not found in non-MTXtreated JRA patients (36,37). Therefore, the finding of
mild portal fibrosis in 3 study patients (21%) suggests
specific MTX-related damage (38).
Perhaps more significant are the findings of
steatosis. Two of the 3 patients with mild steatosis were
mildly obese. These changes, found frequently in nonMTX-treated patients with RA (30), are seen in as many
as 85% of MTX-treated RA patients (22,29-31,34).
Obesity is a major risk factor in the development of
steatosis, which occasionally can progress to fibrosis and
cirrhosis (39).
Among the 4 patients who underwent repeated
biopsies, progression to a more advanced grade was seen
in only 1 patient, from grade 1to IT. However, in another
patient, fibrosis that had been classified as grade IIIA
was found to have regressed in 2 subsequent biopsies
over the following 6 years. Within grade I, progressive
changes were seen in a third patient. While no conclusions can be made from 2 cases, both patients with
progression consumed more alcohol than the other
study patients. Although more studies, perhaps of our
present cohort, are needed, these progressive changes
appear to have few clinical consequences. Similar to our
findings, prospective studies in RA, some more than 10
years, have shown mild progression of hepatic injury,
including fibrosis, in about 33% of patients, mostly
without significant clinical consequences (29,40,41).
Nearly all patients (93%) had liver biochemical
abnormalities while being treated with MTX. The proportion of patients with biochemical abnormalities was
greater than that which has usually been reported (3.457%; mean 16.9%) (42,43). A partial explanation is the
definition of “abnormal.” In our study, any enzyme
elevation beyond the upper limit of the normal range
was considered abnormal, based on the recommendations of the ACR (9). In most laboratories, “abnormal”
represents an enzyme level >2 SD above the population
mean. Other studies have defined abnormality as twice
or even 3 times the upper limit of normal (43). Another
difference is the much longer duration of MTX therapy
in our cohort compared with other pediatric series,
resulting in a greater proportion of patients with enzyme
abnormalities.
The association of serial enzyme abnormalities
with fibrosis and the Roenigk grade is the basis for the
ACR guidelines for MTX liver monitoring in RA (9,44).
LONG-TERM EFFECT OF MTX ON THE LIVER IN PATIENTS WITH JRA
Four patients had serial abnormalities of AST/ALT that
would have warranted a biopsy based on the ACR
recommendations (9). Their biopsy samples were classified as grade I. Therefore, the specificity of the ACR
guidelines in our cohort was 71.4%. We were unable to
determine the sensitivity due to the lack of biopsy
samples with grade ITIB or IV changes. The sensitivity
was 80% (4 of 5 cases) in the only study of RA patients
to prospectively validate the ACR guidelines ( I 2).
N o major complications occurred in this study.
Major complications, mainly bleeding, have been reported in 0-4.5% of pediatric patients undergoing liver
biopsy, with the majority of complications and fatalities
occurring in children with malignancies or following
liver transplantation (45).
Core liver biopsy is considered the “gold standard” for detecting fibrosis. However, sampling error is
possible (34). In 1 study, 2.6% of the samples obtained
by needle were considered inadequate for interpretation
(31). The samples obtained in our study contained at
least 7 portal zones for analysis. Histologic abnormalities
were uniform throughout the biopsy specimens, as usually seen in drug-induced liver diseases (34). Therefore,
the probability of sampling error was low.
The pathologists disagreed initially about the
classification of 1 biopsy specimen (7%) as grade I or
grade 11. This is similar to the 12% interobserver variation rate in RA patient samples reported by Krcmer et a1
(40). Grades I and 11 are not considered a reason to
discontinue MTX; therefore, the disagreement would
have not had clinical implications. Overall, the reliability
of this scale was excellent.
There were several shortcomings in this study,
mostly due to the nature of this study as a cross-section
pilot study. The small sample size (n = 14) was not of
sufficient power to draw definitive conclusions regarding
the long-term safety of MTX. Furthermore, we were
unable to perform inferential statistics, in part due to the
near-unanimity of biopsy samples scored as grade I. This
study should be viewed as only the first step; therefore,
statistical analysis is of secondary importance.
Several important groups of patients may have
been underrepresented, thus introducing potential bias.
In some practices, doses of MTX as high as 1.2 mg/kg/
week are administered to patients with severe disease
(19). Only 2 of the study patients received doses >0.65
mgikg. Other groups defined as “high risk” for the
development of liver fibrosis were also underrepresented. Only 1 patient with a history of hepatitis was
studied. Patients with diabetes mellitus, morbid obesity,
or significant use of alcohol were not included in our
2233
cohort. Patients with diabetes mellitus may be at increased risk of “silent” hepatic fibrosis without serologic
abnormalities (1 2).
We did not find significant differences betwcen
study participants and eligible nonparticipants, other
than a significantly greater number of AST abnormalities since starting MTX in the study participants. Therefore, bias based on patient characteristics did not appear
to play an important role in determining participation in
the study. A factor that minimized selection bias was the
intent-to-treat cross-section design. All eligible patients
were identically approached, including thosc who discontinued MTX.
While it is still premature to state that MTX will
change the long-term outcome in JRA. MTX is the only
drug proven effective in controlling recalcitrant JRA
(1,7,46). The results of this study suggcst that the
long-term use of MTX in patients with JRA is not
associated with the development of significant liver
fibrosis. While not conclusive, our findings may be
reassuring to the many children taking MTX for JRA
and to their physicians.
REFERENCES
1. Giannini EM, Brewer EJ, Kuzima N. Shaikov A, Maximov A.
Vorontsov 1,Fink CW, Ncwman AJ,Cassidy JT, Zernel L S . lor the
Pediatric Rheumatology Collalmrative Study Group and the Co-
2.
3.
4.
5.
6.
7.
8.
9.
lo.
11.
operative Children’s Study Group: Methotrcxate in re
uilc rheumatoid arthritis: results of the U.S.A.-U.S.S.K. doublcblind, placebo-controlled trial. N Engl J Mcd 326:1043-1049. 1992
Towncr SR, Michet CJ Jr, O’Fallon WM, Nelson AM: The
epidemiology of juvenile arthritis in Rochester, Minnesota 19601979. Arthritis Rheum 26:1208-1213, 1983
Cassidy JT, Petty RE: Textbook of Pediatric Rheumatology. Third
edition. Philadelphia, W.B. Saunders Company, 19%
Mier R, Bowyer S, Lovell D, Passo M, Rennenhohm K,Schiklcr K,
Coatcs S: Variations in medication choices in JRA. J Clin Rhcumatol 2:262-267, 1996
US Bureau of the Census: Statistical Abstract of the United States:
19%. 115th edition. Washington DC, US Bureau of the Census,
1995
Ravelli A. Viola S, Ramcnghi B, Aramini L, Ruperto N, Martini
A: Frequency of relapse for clinical remission in juvenile rhcumatoid arthritis. J Rhcumatol 22: 1574-1576, 1995
Giannini EH, Cassidy JT: Methotrexate in juvenile rheumatoid
arthritis: do the benefits outweigh the risks? Drug Saf 9:325-339,
1093
Wallace CA, Sherry DD: A practical approach to avoidance of
methotrcxate toxicity. J Rheumatol 22:1009-1011, 1995
Kremer JM, Alarccin G S . Lightfoot R W Jr, Willkens RF. Furst
DE. Williams HJ. Dent PB. Weinblatt ME: Methotrexate lor
rheumatoid arthritis: suggested guidelines for monitoring liver
toxicity. Arthritis Rheum 37316-328, 1994
Whiting-O’Keefc QE, Fyc mi, Sack KD: Methotrcxate and
histologic hepatic abnormalities: a mcta-analysis. Am J Med
90:711-716, 1901
Weinstein G, Roenigk H, Maibach €1, Cosmidcs J: Psoriasis-livermethotrcxate interactions. Arch Dermatol I08:36-42, 1973
2234
12. Erickson AR, Rcddy V. Vogelgesang SA, West SG: Usefulness 01
the American Collcgc of Rhcumatology recommendations for
liver biopsy in methotrexatc-treated rheumatoid arthritis patients.
Arthritis Rheum 38:1115-1119, 1995
13. Roenigk HH, Aucrbach R, Maibach Hl, Weinstein GD: Methotrcxate in psoriasis: revised guidelines. J Am Acad Dermatol
19:145-156, 1988
14. Walker AM, Funch D, Dreyer NA, Tolman KG, Kremer JM,
Alarchn GS, Lee RC. Weinblatt ME: Determinants of serious
mong patients receiving low-dose mcthotrcxate for
rheumatoid arthritis. Arthritis Rheum 36329-335, 1993
15. Danao T, Steinbrunner J, Mendendorp SV. Mackenzie AH, Segal
AM: Methotrexate in juvenile rheumatoid arthritis (abstract).
Arthritis Rheum 32 (suppl 4):S28, 1989
16. Keim D, Ragsdale C, Heidelberger K, Sullivan D: Hepatic fibrosis
with the use of methotrexite in juvenile rheumatoid arthritis.
J Rheumatol 17346-848, 1990
17. Graham LD, Myones BL,, Rivas-Charcon RF, Pachman LM:
Morbidity associated with long-term methotrexate therapy in
juvcnilc rheumatoid arthritis. J Pediatr 120:468-473, 1092
18. Kugathasaii S, Newman AJ, Dahms BB. Boyle JT: Liver biopsy
findings in patients with juvenilc rheumatoid arthritis receiving
long-term, weekly methotrexate therapy. J Pediatr 128:149-15 I,
1996
19. Wallace C A New uses of methotrexate. Coiitemp Pediatr 11:4353. 1094
20. Lawson EE, Grand KJ, Neff RK, Cohen LF: Clinical estimation of
liver span in infants and children. Am J Dis Child 132:474-476,
1978
21. Cassidy JT, Levinson JE, Bass JC, Baum J, Brewer EJ Jr, Fink
CW, Hanson V, Jacobs JC, Masi AT, Schaller JG, Fries JF,
McShane D, Young D: A study of classification criteria for a
diagnosis of juvenile rheumatoid arthritis. Arthritis Rheum 29:
274-281. 1986
22. Weinstein A, Marlowe S, Korn J , Farouhar F: Low-dose methotrexatc treatment of rheumatoid arthritis: long-term observations.
A m J Med 79:331-337, 1985
23. Ashton RE, Millward-Sadler G
White JE: Complications in
methotrexate treatment of psori
with particular reference to
liver fibrosis. J Invest Dermatol 79:229-232, 1982
24. Kamen BA, Nyleii PA, Camitta BM, Bertino JR: Methotrexatc
accumulation and folate depletion in cells as a possible mechanism
of chronic toxicity to the drug. Br J Haematol 49:355-360. 1981
25. Garrow JS, Webstcr J: Quetlct's index (wih') a5 a measure of
fatness. Int J Obes 9147-150, 1985
26. Centers for Disease Control and Prevention: Youth risk behavior
surveillance system questionnaire (appendix 11). Public Health
Kcp 108~60-66,2993
27. Pearson ES, Clopper CJ: The use of confidence intervals or
fiducial limits in the caw of the binomial. Biometrika 26:404. 1934
28. Brick JE, Morcland LW, Al-Kawas F, Chang WWL, Laync RD,
DiBartolomeo AG: Prospective analysis of liver biopsies before
and after methotrexate therapy in rheumatoid arthritis. Semin
Arthritis Rheum 19:31-44, 1989
29. Wcinblatt ME, Weissman BN, Holdsworth DE, Frascr PA, Maier
AL. Falchuk KR, Coblyn JS: Long-term prospective study of
HASHKES ET AL
mcthotrcxate in the treatment of rheumatoid arthritis: 84-month
update. Arthritis Rheum 35: 129-137, 1992
30. Phillips CA, Cera PJ, Mangan TF, Newman ED: Clinical liver
disease in patients with rheumatoid arthritis taking methotrcxate.
J Rheumetol 19229-233, 1902
31. Mackenzie AH: Hepatotoxicity of prolonged methotrexate therapy for rhcumatoid arthritis. Cleve Clin 52:129-135, 1985
32. Apontc J , Pctrelli M: IIistopathologic findings in the liver of
rheumatoid arthritis patients treatcd with long-term bolus methotrexate. Arthritis Rheum 31:1457-1464, 1988
33. De la Hall PM, Jenner MA, Ahcrn MJ: Wepatotoxicity in a rat
model caused by orally administered methotrexate. Hepatology
14:906-910, 1991
34. Ishak KG, Zimmerman HJ: Morphologic spectrum of druginduced hepatic disease. Gastroenterol Clin North Am 24:759786, 1995
35. Thurman RC, Bunzendahl H. Lcmastcrs JJ: Role of sinusoidal
lining cells in hepatic reperfusion injury following cold storage and
transplantation. Scmin Liver Dis 13:03-100. 1993
36. Schaller J, Beckwith B, Wedgcwood RJ: Hepatic iiivolvcmcnt in
juvenile rheumatoid arthritis. J Pediatr 77:?03-210, 1970
37. Rich RR, Johnson JS: Salicylate hepatotoxicity in patients with
juvenile rheumatoid arthritis. Arthritis Rheum 16:1-9, 1973
G, Clegg DO, Willkens RF, Furst DE, Kremer
s is the only lesion characteristic of methotrexate treatment in rheumatoid arthritis (abstract). Gastroenterology
90: 1776. 1986
39. Neuschwander-Tctri BA, Bacon BR: Nonalcoholic stcatohepatitis.
Med Clin North Am 80:1147-1166, 1996
40. Kremer JM, Lee RG, Tolman KG: Liver histology in rheumatoid
arthritis patients receiving long-term methotrcxatc therapy: a
prospective study with baseline and sequential biopsy samples.
Arthritis Rheum 32:121-127, 1989
41. Kremer JM, Kayc GI, Kaye NW, Ishak KG, Axiotis CA: Light and
electron microscopic analysis of sequential liver biopsy samples
from rheumatoid arthritis patients rccciving long-term methotrexate therapy: followup over long treatment intcrvals and correlation
with clinical and laboratory variables. Arthritis Rheum 38:11941203, 1995
42. Carlos CD, Singsen B, Eichenfield AH, Goldsmith DP, Athreya
BH: Safety and efficacy of methotrcxatc therapy for juvenile
rheumatoid arthritis. J Pcdiatr 17553-659, I990
43. Martini A, Ravelli A, Viola S, Burgio RG: Methotrexate hepatotoxic effects in children with juvenile rheumatoid arthritis (letter).
J Pediatr 119333-334, 1991
44. Krcmer JM, Furst DE, Weinblatt ME, Blotner SD: Significant
changes in serum AST across hepatic histologic biopsy grades:
prospective analysis of 3 cohorts receiving methotrexate therapy
for rheumatoid arthritis. J Rhcumatol 23:459-461, 1996
45. Cohen MB, A-Kadar HH, Lambers D, Heubi JE: Complications
of percutaneous liver biopsy in children. Gastroenterology J 02:
629-632, 1992
46. Gianniiii EH, Cawkwell GD: Drug treatment in children with
juvenile rheumatoid arthritis: past, present and future. Pediatr
Clin North Am 42:1099-1125. 1995
Документ
Категория
Без категории
Просмотров
3
Размер файла
956 Кб
Теги
effect, terms, patients, long, arthritis, methotrexate, juvenile, live, therapy, rheumatoid
1/--страниц
Пожаловаться на содержимое документа