Low-dose methotrexate and severe neutropenia in patients undergoing renal dialysis.

1060
LETTERS
Low-dose methotrexate and severe neutropenia in
patients undergoing renal dialysis
To the Editor:
We administered a single dose of 2.5 mg of methotrexate (MTX) to 2 patients undergoing renal dialysis, 1
with severe rheumatoid arthritis ( R A ) and the other with
polymyositis (PM) and scleroderma, in an effort to treat
severe and life-threatening complications of their inflammatory diseases. Even though the dose of MTX was smaller
than that usually administered, both patients developed
severe neutropenia. and 1 patient developed thrombocytopenia. One of the patients died of complications of infection
after bone marrow recovery.
Patient I . a 52-year-old woman, was receiving continuous ambulatory peritoneal dialysis (CAPD) for renal
failure secondary to hypertension. Hemodialysis was initiated in 1983 and was switched to CAPD when all hemodialysis access sites were exhausted. RA had been present since
1970, with severe involvement of her hands and knees. She
was homebound and required full-time care. Because increasing wrist and hand synovial thickening hindered independent performance of the mechanics of CAPD, prednisone
therapy (20 mglday) was begun. Disease-modifying agents
were also recommended. MTX was selected, and 2.5 mg of
this agent was administered on March 2. 1989 (Table I ) . Her
blood urea nitrogen (BUN) level was 44 mgldl, creatinine
level 10.7 mgldl, albumin level 2.5 gm/dl. and the results of
liver enzyme tests were normal.
On March 3 and 4. pharyngitis and oral ulcerations
developed. On March 6, a rash appeared, and the patient
sought medical care and was hospitalized for parenteral
hydration, mouth care, and analgesia. Laboratory findings at
admission included a normal white blood cell (WBC) count
and normal differential and platelet counts, but on March 8
the WBC count had decreased to 2,000/rnrn3. Lcucovorin
and appropriate antibiotics were prescribed. and isolation
techniques were started when bacteremia was discovered.
Charcoal hemoperfusion also was begun. The bone marrow
recovered 15 days postadmission, but the patient had persistent fever, and her condition slowly deteriorated. She died
of cardiac arrest 3 weeks after bone marrow recovery. At
autopsy, a large subdiaphragmatic abscess was discovered.
Patient 2 was a 47-year-old woman with systemic
sclerosis. PM and renovascular hypertension developed in
December 1988. approximately 4 years after the onset of
Raynaud's phenomenon and tight skin. Dialysis was started
because of increasing renal failure in spite of antihypertensive treatment. On February 9, 1989. she was transferred to
our hospital, where dialysis was continued. Laboratory data
on admission included a BUN level of 44 mgldl. a creatinine
level of 6.9 mg/dl, an albumin level of 3.2 mgldl, and normal
findings on liver enzyme tests. Her hospital stay was complicated by persistent fever, which returned to normal when
azathioprine, used for the treatment of PM. was discontinued. Because of the continued muscle weakness, elevated
creatine phosphokinase (CPK) values ( - 5 0 0 normalized units/
dl), and the patient's unwillingness to take additional prednisone, a single dose of MTX. at 2.5 mg. was administered on
Arthritis and Rheumatism, Vol. 33, No. 7 (July 1990)
February 27, 1989. The patient was then released for outpatient hernodialysis at another facility.
On discharge, her hemoglobin level was 9.4 gm. her
WBC count was 4,900/mm3 with 70% segmented neutrophils, and her platelet count was 172,000/mm3. Her medications upon discharge included labetalol (300 mg twice a day),
enalapril (20 mg 3 times a day). clonidine (0.2 mglday as a
weekly patch). nifedipine (30 mg 3 times a day), and prednisone (I0 mg/day). At home while receiving outpatient
dialysis. the patient developed fever. A chest radiograph
revealed pneumonitis. and antibiotics were administered. A
complete blood cell count on March 8 revealed a hemoglobin
level of 8.0 mgldl, a WBC count of 1,900/mm3 with 42%
segmented neutrophils, and a platelet count of 271.~00/mm~.
The WBC count ranged from 1,500/mm3 to 2,600/mm3 until
March 15. when it increased to 5.200/mm3. The lowest
percent of polymorphonuclear leukocytes found was 33%.
l h e hemoglobin level also increased, to 9.4 gm. MTX levels
were not determined, and leucovorin was not administered.
We are aware of the potential adverse effects of MTX
in patients with renal disease; however, our patients' need
for prompt amelioration of their inflammatory diseases necessitated treatment with a disease-modifying drug, and
unusually low doses of MTX were used. The patient with RA
had clinically active disease and was receiving 20 mg of
prednisone daily. which allowed her to use her hands to
perform peritoneal dialysis exchanges. Azathioprine therapy
was discussed, but the longer time for therapeutic results to
be realized with this agent prompted selection of MTX. The
second patient had renal failure secondary to renovascular
hypertension due to scleroderma and PM, which limited her
ability to ambulate. Her muscle weakness slowly improved
with azathioprine and prednisone therapy until the onset of
daily fever. which was thought to be a side effect of the
arathioprine, necessitated the cessation of that drug. MTX
was administered only when the patient refused additional
prednisone medication and when CPK levels and muscle
weakness increased. Potential side effects and merits of
alternative therapy were discussed with both patients.
In 1988. Janknegt et a1 ( I ) described the pharrnacokinetics of MTX in a CAPD-treated patient with severe
psonasis. Flfteen milligrams of M l X was administered
intravenously, with "very good" clinical response. Nonrenal clearance of MTX appeared to be important in their
patient; enterohepatic excretion of MTX has been described
previously ( 2 4 ) . In spite of temporary neutropenia and
thrombocytopenia. it was suggested that lower doses of
MTX might provide adequate therapeutic response with
greater safety in patients undergoing renal dialysis.
We used leucovorin in the first patient, administering
as much as 75 mglday orally, although it was not started until
6 days after the MTX administration. Its effectiveness in
reducing the bone marrow depression was uncertain (Table
1). We had considered interrupting enterohepatic circulation
in this patient with orally administered charcoal or cholestyramine. but decided against this treatment because of the
presence of oral mucositis. In retrospect, nasogastric instillation of cholestyramine might have been valuable. Cholestyramine binds MTX in vitro. and has been used therapeutically in 2 patients receiving high-dose MTX (3). Nurnberg
et a1 have recently demonstrated biliary clearance of MTX at
I5 ml/minute in an RA patient with a biliary fistula (4).
The second patient recovered without the use of
LETTERS
1061
Table 1. Laboratory findings in patient I . who was receiving continuous ambulatory peritoneal
dialysis, after treatment with a single dose of oral methotrexate followed by leucovorin rescue.
Date
Drug
3/2/89
3/6/89t
3/8/89
3/9/89$
31 I0/89+
311 4/89
3/16/89
3120189
2.5 mg MTX
75 mg leucovorin
75 mg leucovorin
75 mg leucovorin
75 mg leucovorin
-
-
Serum MTX
level.
pmolesniter
0.12
0.09
0.13
0.10
0.02
ND
co.01
WBC,
Hgb.
gm/dl
x 10’/mm’
Platelets.
(% segments)
x IO’lmm’
10.7
11.0
ND
10.0
7.0
6. I
9.0
9.2 (81)
2.0 (SO)
1.8 (61)
0.8 (15)
0.5 (24)
3.1 (30)
10.0 (27)
323
270
ND
I50
99
35
71
* Serum methotrexate (MTX) levels were determined by fluorescence polarization on a TDX, using
Abbott reagents (Abbott Laboratories. North Chicago. IL). Hgb = hemoglobin; WBC = white blood
cells; ND = not determined.
t Date of hospital admission.
$ Charcoal hemoperfusion was performed.
leucovorin or hemoperfusion. She was continued on her
regimen of hemodialysis 3 times a week, with antibiotic use
and standard isolation techniques. Her PM improved and
CPK levels normalized without additional treatment and
have remained constant for more than 5 months post-MTX
administration.
The final decision to use MTX in the 2 patients
described was made only after a lengthy discussion with
colleagues and a review of the report by Janknegt et al(1). It
was our impression that, under extenuating circumstances.
other rheumatologists are using or might consider using
MTX in patients similar to the 2 described here. However.
this report documents the catastrophic events we encountered. Oral administration of MTX in renal dialysis patients
seems equivalent to continuous infusion of the drug, since
there are few effective means for removing MTX from body
fluids without adequate renal function. MTX is poorly
cleared with hernodialysis and cleared even less with peritoneal dialysis (5). Cholestyramine administration was not
used in our patients but may be of some value in providing
nonrenal clearance of MTX (3,4). Measurements of MTX
levels and earlier leucovorin administration might have
helped our first patient, but our overall experience with both
patients has convinced us that MTX cannot be used safely in
renal dialysis patients.
Michael H. Ellman. MD
Susan Hou, MD
Michael Reese Hospital
and Medical Center
Chicago, IL
David Ginsberg. MD
Highland Park Hospital
Highland Park, IL
I . Janknegt R , Nube MI. van den Hoogenband HM. Oldenhof HGJ.
Steenhoek A. Vree TB: Pharmacokinetics of methotrexate in
continuous ambulatory peritoneal dialysis. Pharm Weekbl [Sci]
10:8689, 1988
2. Steinberg SE, Campbell CL, Bleyer WA. Hillman RS: Enterohepatic circulation of methotrexate in rats in vitro. Cancer Res
42:127%1282, 1982
3. Erttmann R, Landbeck G : Effect of oral cholestyramine on the
elimination of high-dose methotrexate. J Cancer Res Clin Oncol
I10:48-50, 1985
4. Nurnberg B. Jacobi A. Faulkner R, Koehnke R. Hoffman I.
Furst D: Biliary elimination of low dose methotrexate in man
(abstract). Arthritis Rheum 31 (suppl 4):S116. 1988
5 . Thierry FX. Vernier I . Dueymes JM.Roche H, Canal P. Meeus
F. Pourrat JP, Conte JJ: Acute renal failure after high-dose
methotrexate therapy (letter). Nephron 51:416417, 1989
Nonsteroidal antiinflammatory drugs and the
lupus anticoagulant
To the Editor:
The lupus anticoagulant (LAC), found in patients
with systemic lupus erythematosus (SLE) and diverse other
conditions, has recently been the focus of much attention
( 1-3). LAC is a phospholipid-specific immunoglobulin that
interferes in vitro with the activation of prothrombin (4).
Paradoxically, the anticoagulant has been associated not
with bleeding, but with arterial and venous thromboses and
fetal loss. Multiple mechanisms may underlie its pathogenicity. and questions concerning therapy remain unresolved.
We report a case of SLE in which the nonsteroidal antiinflammatory drug (NSAID) ibuprofen may have protected
against LAC-related complications.
The patient, a SO-year-old man, was admitted to the
hospital in April 1985 with headache and visual disturbance.
He had a history of pericarditis and Raynaud’s phenomenon,
and, for many years, was treated for arthralgias with ibuprofen alone. A few weeks previously. the patient had
stopped taking the drug because he believed it to be ineffective. Noteworthy findings on physical examination included
visual field defects, adenopathy. and splenomegaly. Laboratory findings included anemia and thrombocytopenia, antinuclear and anti-native DNA antibodies, and a markedly
prolonged partial thromboplastin time, which was not corrected by the addition of normal plasma. Computed tomography (CT) revealed a left occipital infarction. With the
resumption of ibuprofen (600 mg 4 times a day) and the
institution of prednisone at an initial dosage of 80 mglday,
the patient did well.
In March 1986, the ibuprofen was discontinued be-