Content uploaded by David Swinson
Author content
All content in this area was uploaded by David Swinson on Jul 21, 2016
Content may be subject to copyright.
Q J Med 1998; 91:49–56
Treatment of Felty’s syndrome with the haemopoietic growth
factor granulocyte colony-stimulating factor (G-CSF)
S.J. STANWORTH, M. BHAVNANI, C. CHATTOPADHYA, H. MILLER and
D.R. SWINSON
From the Departments of Haematology, Manchester Royal Infirmary and Wigan, and of
Rheumatology, Wrighton Hospital
Received 24 September 1997 and in revised form 10 November 1997
Summary
Felty’s syndrome (FS) (rheumatoid arthritis with G-CSF treatment was continued in three patients by
restarting at reduced dose, and changing the propri-neutropenia and splenomegaly) has a poor pro-
gnosis, largely because of the high risk of severe etary formulation. G-CSF raised the neutrophil
count, reduced severe infection, and allowed sur-infection. Granulocyte colony-stimulating factor
(G-CSF) is an emerging treatment for chronic gery to be performed. A combined clinical and
laboratory index suggested that long-term treatmentneutropenia. We prospectively monitored its use in
eight patients with recurrent infections or who (up to 3.5 years) did not exacerbate the arthritis.
Once on established treatment, it may be possiblerequired joint surgery. Significant side-effects were
documented in five, including nausea, malaise, gen- to use smaller weekly doses of G-CSF to maintain
the same clinical benefit. One of the three patientseralized joint pains, and in one patient, a vasculitic
skin rash. In two patients treatment had to be whose FS was associated with a large granular T-cell
lymphocytosis showed a reduction in this subset ofstopped, and in these cases G-CSF had been started
at full vial dosage (300 mg/ml filgrastim or lymphocytes during G-CSF treatment.
263 mg/ml lenograstim) alternate days or daily.
Introduction
The first description of a syndrome consisting of rheumatoid joint disease in the patients is less active,
suggesting that the link between the two is complic-classical rheumatoid arthritis (RA), splenomegaly,
and neutropenia was reported in 1924.1Felty’s ated, and reflects other processes.
Felty’s syndrome has a poor prognosis, reflectingsyndrome (FS) is now recognized to occur in less
than 1% of patients with RA, and more commonly increased mortality due to the higher incidence of
severe infection, with up to 36% 5-year mortality.4in those with previously severe forms of the arthropa-
thy. In these individuals with FS there is a greater Most infections documented in FS are due to bacterial
pathogens, consistent with an underlying defect inincidence of extra-articular and constitutional mani-
festations, seropositivity for rheumatoid factors, and neutrophil activity, and the acquired neutropenia is
one of the additional main factors predisposing tostrong immunogenetic association with HLA DR4.2,3
The degree of splenomegaly varies over time and infection in this subgroup of RA. Abnormalities of
neutrophil function may also be found in someextent in individuals with FS; indeed, there is evid-
ence that patients with and without splenomegaly patients.5The pathogenesis of the neutropenia in
FS has not been clearly defined; there is evidenceare clinically, immunologically and genetically very
similar.2It is interesting that the pathognomonic for impaired granulopoiesis, increased peripheral
destruction, as well as sequestration within theneutropenia of FS is often found at a time when the
Address correspondence to Dr D.R. Swinson, Consultant Rheumatology, Wrightington Hospital, Hall Lane, Appley Bridge,
Wigan WN6 9EP
© Oxford University Press 1998
S.J. Stanworth et al.50
enlarged spleen.2,6 In general, direct evidence for lenograstim (Granocyte vials at 263 mg/ml or
105 mg/ml, Chugai). ‘Full’ doses of G-CSF wereantibody- or immune-complex-mediated peripheral
destruction has been surprisingly hard to substanti- defined as the contents of the 300 mg/ml filgrastim
or 263 mg/ml lenograstim vials; ‘reduced’ doses asate.2,7–9 In addition, FS is associated with a form
of T cell lymphocytosis, in which the expanded half (or less) of the full doses (or in the case of
lenograstim the 105 mg/ml vial). Follow-up waslymphocytes have a large granular morphology.9–12
This large granular lymphocyte (LGL) syndrome may initially weekly by a specialist nurse practitioner,
and at least monthly by medical staff.be defined as large granular lymphocytes present in
the peripheral blood at >1×109/l or making up Assessments of the RA were made before and
during treatment with G-CSF. The Disease Activity>25% of the total lymphocytes, and it is recognized
in about a third of all patients with FS. The T-cell Score (DAS) is a combined index designed to assess
disease activity in RA, and a modification of theproliferations in FS may be clonal,13,14 and there is
some in vitro evidence to indicate that these CD8+original DAS index was scored for patients in this
study, based on a 28 tender joint count, a 28 swollenT cells inhibit granulopoiesis either directly or
through the production of specific cytokines, but the joint count, the patient’s global assessment and the
erythrocyte sedimentation rate.19 For this modifieddetails are not confirmed.7,9
In managing FS, it may be necessary specifically score, a change of 1.08 is felt to represent a
significant improvement or deterioration. Spleen sizeto treat the haematological manifestations, including
the neutropenia. Disease-modifying rheumatoid was monitored in patients using measurements
obtained at ultrasound or by clinical examination,agents, including methotrexate may improve the
neutropenia, but the response is not consistent.15 recording the maximal distance below the costal
margin.The indications for separately considering other treat-
ments of neutropenia, and with which agents, have Haematoxylin- and eosin-stained peripheral blood
films were examined for the presence of largenot been established. Splenectomy has been tradi-
tionally advocated, but recurrence of the neutropenia granular lymphocytes. Immunophenotyping of peri-
pheral blood lymphocytes was performed by flowremains a problem for a significant number of
patients; moreover, splenectomy is associated with cytometry using a Coulter EPICS counter and a
range of commercially-available fluorescent-labelleda further infection risk.4,15
Recently, interest has focused on haemopoietic monoclonal antibodies.
growth factors, including granulocyte colony-
stimulating factor (G-CSF); the rationale for the use
of these agents is that they might counteract an
Results
impaired production of neutrophils in patients with
FS.16 A major difficulty in assessing the value of Eight FS patients with a mean age of 68 years (range
60–74 years) were started on G-CSF. All patientsgrowth factor treatments for patients with FS and
severe neutropenia is the lack of reported clinical had disabling arthritic changes, and were positive
for rheumatoid factors. In seven patients, recurrenttrials,15 although a number of case reports attest to
their potential benefits.16–18 Due to the location of a infections were the main indication for G-CSF, the
foci of infection being the skin, subcutaneous tissueslarge Rheumatology Centre in this region, it has been
possible to follow prospectively the progress of eight and respiratory tract. G-CSF was commenced in two
patients with the intention of raising the neutrophilFS patients with severe neutropenia who were com-
menced on treatment with G-CSF for different count to cover major orthopaedic surgery. Table 1
summarizes these and other relevant demographicreasons. This has allowed us to assess the effect-
iveness of treatment, the incidence of side-effects, features of the patients, together with the details of
haematological indices and LGL counts prior toand the details of dosing regimens.
treatment. One patient (GB) was also an insulin-
dependent diabetic. Only one patient was receiving
a disease-modifying anti-rheumatic drug during the
Methods
course of G-CSF (JD: hydroxychloroquine); three
were taking low-dose prednisolone.Patients enrolled into the study all fulfilled the
classical criteria for Felty’s syndrome of rheumatoid Six patients were commenced on lenograstim
initially, two on filgrastim. Dosage was lenograstimarthritis (RA), splenomegaly, and neutropenia (abso-
lute neutrophil count <1×109/l). In addition, these at 263 mg/day, or filgrastim at 300 mg/day. In two
patients, ‘reduced’ doses were given at the startpatients had recurrent and severe infections or
required joint replacement surgery. G-CSF for subcu- (lenograstim at 105 mg/day or filgrastim at
150 mg/day). Two patients were changed to thetaneous injection was available as either filgrastim
(Neupogen vials at 300 mg/ml, Amgen/Roche) or alternative G-CSF proprietary preparation soon after
Treating Felty ’s syndrome 51
Table 1 Demographic features of Felty’s syndrome with details of blood counts before and during G-CSF treatment
Patient Sex Age (years) Duration (years) Main indications for G-CSF White cell and platelet counts before treatment Counts during treatment
ANC
RA FS WCC (ANC) LGL Platelets
JP F 68 45 4 Multiple infections (skin, ulcers) 0.5 (0.2) <0.3 160 2.0
AM F 70 10 6 Multiple infections (chest, skin, ulcers) 1.0 (0.2) 0.5 171 1.7
DW M 71 34 13 Multiple infections (skin, urinary, ulcers) 2.0 (0.6) <0.3 105 3.9
GB M 70 20 7 Multiple infections (skin, mouth ulcers) 1.6 (0.4) 1.2 196 1.5
IG F 66 33 12 Multiple joint operations 1.6 (0.4) <0.3 214 1.0
JD F 60 19 8 Multiple infections 1.3 (0.2) <0.3 299 2.7
(skin, chest, urinary, joint nodules)
MT F 74 35 4 Operative cover, multiple skin ulcers 3.7 (0.6) 2.0 250 4.5
NB F 67 22 14 Multiple infections (skin, ulcers) 1.1 (0.2) <0.3 170 1.1
ANC, absolute neutrophil count. Leukocyte and platelet counts (×109/l) are mean values.
S.J. Stanworth et al.52
starting treatment because of severe side-effects, and therapy the mean counts fell slightly (data not
shown), and this was associated with a small reduc-in these cases ‘reduced’ doses of the second G-CSF
formulation were used initially. Table 2 summarizes tion in spleen size in one of these individuals.
The DAS scores were monitored to assess severitythe doses of G-CSF used at different times after
starting treatment and the side-effects encountered of rheumatoid arthritis, but did not alter significantly
or consistently in any patient during G-CSF treatment.during G-CSF treatment.
Five individuals developed significant side-effects This is illustrated in Figure 1, together with the
changes in neutrophil count. In view of concernsduring the first 1–2 weeks of G-CSF treatment. Two
of these patients, who had been commenced on that lack of change in the composite DAS score
during treatment might mask underlying changes inlenograstim (263 mg, initially daily or alternate daily)
reported symptoms that included malaise and severe the individual parameters that make up the score,
separate analyses of joint tenderness and swellinggeneralized bone pains, affecting the sternum, back
and legs. Both patients (JP and NB) declined further were plotted, but again this revealed no significant
change during treatment with G-CSF.G-CSF injections after one week because of the
distressing nature of these side-effects. Of interest, Three individuals had expansions of CD8+T
lymphocytes with a large granular morphologyperipheral blood counts taken during the week of
treatment showed a temporary reduction of platelet (Table 1). Longitudinal studies of lymphocyte subsets
before and during G-CSF treatment were conductednumbers to around 100×109/l.20
Of the remaining three patients who developed by peripheral blood immunophenotyping. In two
cases there was no significant change in the levelsproblems soon after commencing filgrastim G-CSF
treatment, one (AM) developed a widespread vascul- of the proliferations during G-CSF treatment, but in
one patient there was a marked reduction in theitic rash, most prominent over the buttocks, and
upper thighs. Changing the G-CSF formulation to level of the LGL expansion (Table 3).
lenograstim in this individual, and starting at a lower
dose, did not cause a recurrence of the rash, and
enabled the injections of G-CSF to be continued for
Discussion
over 6 months. The other two patients (GB and IG)
reported flu-like symptoms, nausea, sweats, and to G-CSF, an 18 kDa glycoprotein encoded by a gene
located on chromosome 17q, has haemopoietica varying degree, generalized pain in the joints.
These symptoms initially persisted in one individual lineage-specific functions, in that it stimulates prolif-
eration and maturation of granulocyte precursors,(IG) despite changing the formulation of G-CSF, and
reducing the initial dose, before gradually settling. and activates mature neutrophils.21 Current clinical
indications for recombinant human G-CSF includeIn the other patient (GB), it became apparent that
these symptoms represented occult infection, which mobilization of peripheral blood progenitor cells for
transplantation, and management of cytotoxicsettled with anti-microbial therapy, allowing the
G-CSF injections to be continued without further induced neutropenia, and treatment courses in these
situations are usually short and very well tolerated.complication. One patient ( JD), who had been
stabilized on filgrastim injections for over 1 year, By contrast, the role and use of long-term G-CSF in
severe chronic neutropenia states is less clearlybut in whom it was necessary to briefly change to
lenograstim, reported the development of trouble- defined, although reported for many causes of inher-
ited and acquired neutropenia.22 Felty’s syndromesome nausea. She was returned to filgrastim, on
which she has remained for a further 2 years without represents a subgroup of RA that is characterized by
persisting neutropenia, which is a major factor incomplication.
All six patients who were stabilized on G-CSF the increased susceptibility of these patients to recur-
rent infection. Moreover, the presence of infectiveinjection therapy showed a consistent rise in mean
absolute neutrophil counts to values within the range complications in FS has been inversely correlated
with endogenous G-CSF production.23 However,1–4.5×109/l (Table 1). In the individuals treated
because of recurrent infections, there was a reduction G-CSF, through its actions on myeloid cells and
precursors, might be expected to affect the cellularin severity of infective (usually respiratory) episodes,
or a clear improvement in the site of main infection, and cytokine mechanisms underlying the inflammat-
ory changes in RA,24 and therefore could exacerbatesuch as skin ulcers (MT), rheumatoid nodules ( JD),
or sites of previous operations (DW). Joint replace- the joint disease.25,26 The aim of this study was to
monitor prospectively G-CSF treatment of the neutro-ments indicated in two patients (IG, MT) were
performed without post-operative complications of penia in FS, paying close attention to the frequency
of side-effects.sepsis. Mean platelet counts before starting G-CSF
treatment varied from low to within normal range These side-effects were more common than
expected, and the complications were sufficiently(Table 1); in two patients on established G-CSF
Treating Felty ’s syndrome 53
Table 2 The doses of G-CSF used and side-effects of treatment
Patient Details of G-CSF dosage Side-effects Total G-CSF cumulative treatment
Drug Dose Frequency Y/N What
JP G F Daily Y Severe generalized joint pain, malaise in first week 1 week
AM N F Daily Y Severe vasculitic rash in first week
GR+×2–3/week N 6 months
DW G F ×2–3/week N 6 months
GB G F ×3 week Y Malaise, nausea, generalized aches, fever in first
weeks
NR+×2–3/week Y 9 months
IG G F ×2/week Y Flu-like symptoms, malaise, nausea, joint pain initially
NR ×2/week Y Flu-like symptoms continued intermittently
GR+×2/week Y Occasional aches but not severe 4 months
JD N R+Daily, then ×2/3 weekly N
GF ×3/week Y Nausea, malaise when briefly changed G-CSF
formulation
NF ×2/week N 40 months
MT G F ×3/week N
GF ×3/week N 4 months
NB G F ×3/week Y Severe generalized joint pain, malaise in first week 1 week
G-CSF was administered as filgrastim (Neupogen, labelled in table as N) or lenograstim (Granocyte, labelled G) at full doses (F): filgrastim 300 mg/ml or lenograstim 263 mg/ml)
or reduced doses (R) (as described in text). R+, initially R, dose gradually increased thereafter.
S.J. Stanworth et al.54
menced on full doses of G-CSF in the first week,
which were associated with distressing side-effects.
Despite the frequency of side-effects, it was pos-
sible to achieve regular G-CSF treatment in most
patients. One patient has been on G-CSF treatment
for over 3 years. Neutrophil responses to continued
G-CSF treatment were documented in all cases, and
associated with reduced rates of severe infection or
improvement at the sites of chronic sepsis. It was
initially thought that maintenance doses of G-CSF
Figure 1. Changes in DAS score and neutrophil count
for patients on long-term treatment would be lenogra-
during treatment with G-CSF, including changes in neutro-
stim at 263 mg, or filgrastim at 300 mg up to three
phil count (×109/l; open circles) and DAS score (filled
or four times a week, which would have significant
circles) during G-CSF treatment (shaded box) in one
financial implications. However, it became clear that
patient.
reduced doses at less frequent intervals (twice
weekly) may achieve the same clinical benefit, even
if the absolute rises in neutrophil counts were smaller
severe at some stage in 6/8 patients to warrant either (to 1×109/l). A reduction in platelet count relative
stopping the drug and reducing the dose, or switching to baseline values was also documented in several
to the alternative proprietary formulation in an patients during the period of G-CSF treatment, but
attempt to continue the treatment (Table 2). The main the underlying mechanism is not clear.20
early side-effects included severe nausea, joint pains, There was no evidence using the modified Disease
malaise, and skin rashes. These complications have Activity Score (DAS) that long-term G-CSF treatment
been reported in other patients, but it is difficult to caused a deterioration of the RA. Four patients on
assess the frequency of these side-effects based on established G-CSF treatment generally felt better and
individual case reports.25,27 The side-effects found in reported subjective improvement of their joints
our study were not confined to filgrastim, the non- during G-CSF treatment, although the DAS score did
glycosylated formulation of G-CSF, which has been
not change significantly. Those two patients who
considered a theoretical risk since in humans only
described severe joint pains and malaise in the first
the glycosylated molecule is synthesized.
week after starting G-CSF treatment did not develop
Starting the G-CSF treatment at lower doses, and
other signs of active arthritis, and the symptoms
then gradually increasing the dose to optimize the
settled promptly after stopping G-CSF. Concerns have
neutrophil response could reduce side-effects. We
also been raised about the leukaemogenic risk associ-
would now recommend commencing all FS patients
ated with long-term use of growth factors, although
on lower G-CSF doses at the start of treatment (even
this was not identified in this study.28
down to a 1/10th full vial two or three times weekly,
For one of the patients (GB) there was a reduction
then increased after several doses to half a vial, or the
of the proportion of large granular lymphocytes in
105 mg/ml lenograstim dose, every other day or daily).
the peripheral blood during G-CSF treatment. At the
The two individuals in whom growth factor treatment
was stopped but not restarted had both been com- start of treatment this subset of T cells had
Table 3 Effect of G-CSF treatment on subsets of peripheral blood lymphocytes in two patients with Felty’s syndrome and
proliferations of CD8+large granular lymphocytes
Lymphocyte cell surface markers Patient 2 (AM) Patient 4 (GB)
Pre-treatment After 4 months Pre-treatment After 1 month After 8 months
CD3 90 87 91 95 92
CD4 27 22 24 22 55
CD8 52 57 66 70 35
CD16 4 8 1 1 2
CD20 6 3 7 3 4
CD56 1 2 2 1 2
CD57 37 27 26 27 12
Lymphocyte count (×109/l) 1.1 1.3 1.1 1.3 1.5
Immunophenotyping of lymphocytes was performed as described in text. Data are % positivity, except for lymphocyte counts.
Treating Felty ’s syndrome 55
2. Campion G, Maddison PJ, Goulding N, James I, Ahern MJ,
represented nearly 30% of the peripheral blood
Watt I, Sansom D. The Felty syndrome: a case matched
lymphocytes (Table 3). Southern blot analysis of
study of clinical manifestations and outcome, serological
T-cell receptor gene rearrangements for these
features and immunogenetic associations. Medicine 1990;
lymphoid cells did not indicate clonality (data not
69:69–80.
shown). Changes in the levels of LGL expansion
3. Ollier WER, MacGregor A. Genetic epidemiology of
were not seen in the other two patients. It has been
rheumatoid disease. In: Saklatvala J and Walport MJ, eds.
suggested that these expanded populations of CD8+
Immunology of Rheumatoid Disease, Churchill Livingstone.
Brit. Med. Bull. 1995; 51:267– 85.
T cells arise in response to the excessive B cell
activity found in RA.7,14 Such expansions of CD8+
4. Thorne C, Urowitz MB. Long-term outcome in Felty’s
syndrome. Ann Rheum Dis 1982; 41:486–9.
lymphocytes might initially be polyclonal, from
5. Davis P, Johnston C, Bertouch J, Starkebaum G. Depressed
which clonal proliferations subsequently appear.12
superoxide radical generation by neutrophils from patients
Our patient demonstrates that these proliferations of
with rheumatoid arthritis and neutropenia: correlation with
large granular lymphocytes may change with time,
neutrophil reactive IgG. Ann Rheum Dis 1987; 46:51–4.
and this may be related to G-CSF treatment.
6. Rosenstein ED, Kramer N. Felty’s and pseudo-Felty’s
Reductions of the chronic expansion of large granular
syndromes. Semin Arthitis Rheum 1991; 21(3):129–42.
lymphocytes are not well recognized, and it remains
7. Snowden N, Kay RA. Immunology of systemic rheumatoid
unknown whether these changes would correlate
disease. Br Med Bull 1995; 51:437–48.
with reduced levels of inhibition of granulopoiesis.
8. Breedveld FC, Lafeber GJM, de Vries E, van Krieken JHJM,
Subject to the low numbers of patients in this
Cats A. Immune complexes and the pathogenesis of
study, the results indicate that G-CSF is an effective
neutropenia in Felty’s syndrome. Ann Rheum Dis 1986;
agent in the management of severe neutropenia
45:696–702.
found in Felty’s syndrome. The neutrophil response
9. Loughran TP. Clonal diseases of Large Granular
to treatment is associated with a reduction in the
Lymphocytes. Blood 1993; 82:1–14.
frequency of severe infections. G-CSF treatment is
10. Snowden N, Bhavnani M, Swinson DR, Kendra JR, Dennett
commonly associated with side-effects, which can
C, Carrington P, Walsh S, Pumphrey RSH. Large granular T
lymphocytes, neutropenia and polyarthropathy: an
be minimized by reducing the initial dose. Sub-
underdiagnosed syndrome? QJ Med 1991; 78:65–76.
sequent changes in dose, frequency, or prescription
11. Gonzales-Chambers R, Przepioka D, Winkelstein A,
of an alternative proprietary recombinant factor,
Agarwal A, Staz TW, Kline WE, Hawk H. Lymphocyte
may be necessary to achieve regular treatment in an
subsets associated with T cell receptor b-chain gene
individual. Long-term clinical benefit may be main-
rearrangements in patients with rheumatoid arthritis and
tained by reducing the G-CSF dose or frequency to
neutropenia. Arthritis Rheum 1992; 35:516–20.
that which achieves smaller absolute rises in neutro-
12. Bowman SJ, Bhavnani M, Geddes GC, Corrigall V, Boylston
phil count. There was no evidence to indicate that
AW, Panayi GS, Lanchbury JS. Large Granular Lymphocyte
prolonged treatment with G-CSF caused a deteriora-
expansions in patients with Felty’s syndrome; analysis using
anti-T cell receptor Vb-specfic monoclonal antibodies. Clin
tion of the underlying arthritis. This study has not
Exp Immunol 1995; 101:18–24.
addressed whether there is a direct survival benefit
13. Bowman SJ, Corrigall V, Panayi GS, Lanchbury JS.
to G-CSF treatment of Felty’s syndrome patients with
Hematologic and cytofluorographic analysis of patients with
severe neutropenia, but in view of the observation
Felty’s syndome. A hypothesis that a discrete event leads to
that sepsis is a major causal factor in the mortality
large granular lymphocyte expansions in this condition.
of many patients with FS, this may be expected.4
Arthritis Rheum 1995; 38:1252–9.
Clearly it would be preferable in future studies to
14. Bowman SJ, Hall MA, Panayi GS, Lanchbury JS. T cell
analyse and follow up a larger cohort of patients
receptor a-chain and b-chain junctional region homology in
with Felty’s syndrome, but this would require a
clonal CD3+, CD8+T lymphocyte expansions in Felty’s
co-ordinated multi-centre approach.
synrdome. Arthritis Rheum 1997; 40:615–23.
15. Rashba EJ, Rowe JM, Packman CH. Treatment of the
neutropenia of Felty syndrome. Blood Rev 1996;
10:177–84.
Acknowledgements
16. Vreugdenhil G, Schattenberg A, Dompeling EC, Swaak AJ,
We wish to thank Drs E. Love and E. Woodcock for
De Witte T. Hematopoietic growth factors in rheumatoid
arthritis: a critical approach to their use in view of possible
allowing us access to their patients, to Ms L. Green
adverse effects. Am J Med 1993; 94:229–31.
for performing the immunophenotyping, and to
17. Choi M-F V, Mant MJ, Turner AR, Akabutu JJ, Aaron SL.
Chugai for valuable support, and Dr D. Scott for
Successful reversal of neutropenia in Felty’s syndrome with
his advice.
recombinant granulocyte colony stimulating factor. Brit
J Haem 1994; 86:663–4.
18. Moore DF, Vadhan-Raj S. Sustained response in Felty’s
syndrome to prolonged administration of recombinant
References
Human Granulocyte-Macrophage Colony Stimulating Factor
1. Felty AR. Chronic arthritis in the adult associated with (rhGM-CSF). Am J Med 1995; 98:591–4.
splenomegaly and neutropenia. Bull Hopkins Hosp 1924;
35:16–20. 19. Van Gestel AM, Prevoo ML, Van’t Hof MA, Van Rijswijk
S.J. Stanworth et al.56
MH, Van de Putte LB, Van Riel PL. Development and 24. Brennan FM, Maini RN, Feldmann M. Cytokine expression
in chronic inflamatory disease. Br Med Bull 1995;validation of the European league against Rheumatism
response criteria for rheumatoid arthritis. Comparison with 51:368–84.
the preliminary American College of Rheumatology and the 25. Yasuda M, Kihara T, Wada T, Shiokawa S, Furuta E,
World Health Organisation/International league against Suenagu Y, Noraka S, Nobunaga M, Yoshiok K, Isayama T.
Rheumatism Criteria. Arthritis Rheum 1996; 39:34–40. Granulocyte colony-stimulating factor induction of
20. Wun T. The Felty syndrome and G-CSF associated improved leukocytopenia with inflammatory flare in a
thrombocytopenia and severe anemia (letter). Ann Intern Felty’s syndrome patient. Arthritis Rheum 1994; 37:145–6.
Med 1993; 118:318–19. 26. Hazenberg B, Van Leeuwen M, Van Rijswijk M, Stern A,
21. Bagby GC. Hematopoiesis, In: Stamatoyannopoulos G, Vellenga E. Correction of granulocytopaenia in Felty’s
Nienhuis AW, eds. The molecular basis of blood diseases.syndrome by granulocyte-macrophage colony-stimulating
Philadelphia, WB Saunders, 1995:71–103. factor; simultaneous induction of IL–6 release and flare of
22. Dale DC, Bonilla MA, Davis MW, Nakanishi AM, inflammatory arthritis. Blood 1989; 74:2769–70.
Hammond WP, Kurrtzberg J, Wang W, Jakubowski A, 27. Vidarsson B, Geirsson AJ, Onundarson PT. Reactivation of
Winton E, Lalezari P, Robinson W, Glaspy JA, Emerson S, rheumatoid arthritis and development of leukocytoclastic
Gabrilove J, Vincent M, Boxer LA. A randomized controlled vasculitis in a patient receiving granulocyte colony-
phase III trial of recombinant human granulocyte-colony- stimulating factor for Felty’s syndrome. Am J Med 1995;
stimulating factor (filgrastim) for treatment of severe chronic 98:589– 91.
neutropenia. Blood 1993; 81:2496–502.
23. Meliconi R, Uguccioni M, Chieco-Bianchi F, Pitzalis C, 28. Dong F, Brynes RK, Tidow N, Welte K, Lowneberg B, Touw
IP. Mutations in the gene for the granulocyte conlony-Bowman S, Faccini A, Gasbarrini G, Panayi GS, Kingsley
GH.The role of interleukin–8 and other cytokines in the stimulating-factor receptor in patients with acute myeloid
leukemia preceded by severe congenital neutropenia. Npathogenesis of Felty’s syndrome. Clin Exp Rheum 1995;
13:285–91. Engl J Med 1995; 333:487–93.
