вход по аккаунту


Magnetic resonance imaging as a predictor of progressive joint destruction in neuropathic joint disease.

код для вставкиСкачать
Vol. 46, No. 10, October 2002, pp 2811–2816
© 2002, American College of Rheumatology
co’s modified Eagle’s medium (Gibco, Grand Island, NY) plus
10% FCS supplemented with bovine extracellular matrix components.
RASFs in passages 2–13 and cultures of skin fibroblasts in passages 2 and 8 either underwent irradiation with
x-rays (2 Gy or 8 Gy) or were treated with either 5-aza-2⬘deoxycytidine (AzaC) (2 ␮g/ml for 24 hours; Sigma, St. Louis,
MO) or 5-bromo-2⬘-deoxyuridine (BUDR) (25 ␮g/ml for 24
hours; Sigma). In addition, cultures of RASFs, normal synovial
fibroblasts, or skin fibroblasts were incubated with iododeoxyuridine (IUDR, 20 ␮g/ml for 20 hours; Sigma). Cell
culture supernatants were collected on day 0 and at 48 hours or
72 hours and analyzed for RT activity using the productenhanced RT (PERT) assay (9).
On days 2–8, treated and control cells were mechanically harvested and fixed in 2.5% glutaraldehyde diluted in
phosphate buffer (0.1M, pH 7.4) and stored in fixative overnight at 4°C. Specimens were then washed in phosphate buffer,
postfixed with 1% osmium tetroxide, and dehydrated in graded
alcohol. Epon-polymerized cell pellets were processed by
semi-thin sections after selecting representative ultra-thin (80
nm) sections using a Reichert ultramicrotome and uranyl
acetate and lead citrate. Stained sections were analyzed with a
Philips CM 10-electron microscope (Philips, Mahwah, NJ).
Vesicular stomatitis virus–infected baby hamster kidney cells
served as positive controls.
Cocultures of synovial cells with H9, chondrocytes,
and monocytes. Cultures of RASFs (passages 4 and 5) and of
FSFBs (passage 3) were cocultured with H9 cells for 4 weeks.
At passage 2, SFC cultures from RA patients and from patients
with unclassified oligoarthritis or gouty arthritis were cocultured with H9 cells for 3 weeks. H9 cells and cell culture
supernatants were collected every 3–4 days for testing. Furthermore, cultures of RASFs (passages 2 and 3) and of normal
SFs (passage 4) were cocultured with normal human chondrocytes for 1 week. Cultures of RASFs (passages 4–7) and of
FSFBs (passage 10) were trypsinized and irradiated in a 14-ml
Falcon tube with 100-Gy gamma rays and subsequently added
to cultured monocytes. Cell culture supernatants were regularly analyzed by the PERT assay, as indicated in Table 1.
Cocultures of RASFs with monkey kidney cell lines
Vero and BSC. Two modifications of these coculture experiments were performed. In the first experiment, cultures of
RASFs (passages 3–8) and of FSFBs (passage 7), either
irradiated or nonirradiated, were seeded on subconfluent Vero
or BSC cells. In a second experiment, cultures of RASFs
(passages 5–10) and of FSFBs (passage 9), either irradiated or
nonirradiated, were seeded first, and then Vero or BSC cells
were added to the fibroblast layer. Cell culture supernatants
were analyzed by PERT assay on days 0, 2, and 22 (Table 1).
RT assay. The PERT assay was performed as described previously, with some modifications introduced for
real-time detection (9). Cell culture supernatants were tested
undiluted or 10-fold diluted with phosphate buffered saline
(cocultures). For the induction experiments involving AzaC,
BUDR, or IUDR, particles were separated from cell culture
supernatants in order to avoid a potential inhibition of viral RT
by these drugs. For this purpose, 900 ␮l of the cell culture
DOI 10.1002/art.10582
Absence of inducible retroviruses from synovial
fibroblasts and synovial fluid cells of patients with
rheumatoid arthritis
Many investigators have focused on the possibility of
an association of retroviruses with rheumatoid arthritis (RA).
However, direct evidence for an exogenous retrovirus involved
in the pathogenesis of RA is lacking. Indirect evidence comes
from animal models such as Tax transgenic mice or caprine
arthritis encephalitis virus–infected goats, both of which are
associated with RA-like diseases. Moreover, human
T-lymphotropic virus infection in humans causes a virusassociated arthropathy in a minority of patients (for review, see
ref. 1). Involvement of endogenous retroviruses or other
retroviral elements has also been proposed. Insertion of the
early transposon retrotransposon into the fas gene in MRL/lpr
mice leads to RA-like joint effects (2). In 1999, Griffiths et al
(3) reported an association of integrated human retrovirus 5
(HRV-5) proviral DNA with RA, but another group of investigators could not confirm the data (4). Stransky et al (5)
detected virus-like particles (VLPs) in ultracentrifuged RA
synovial fluids. However, because synovial fluid is full of
cellular debris such as membranes of the endoplasmic reticulum, including ribosomes, these structures could form particles
similar to retrovirus particles.
In this study, we performed experiments involving
activation of potentially hidden retroviruses from cultured RA
synovial fibroblasts (RASFs) and RA synovial fluid cells
(SFCs). In turn, chemical or physical inducers of retrovirus
expression and various coculture conditions were utilized and
examined by an ultrasensitive reverse transcriptase (RT) assay
and by classic electron microscopy (EM).
Cell cultures. Synovial fibroblasts were obtained by
enzymatic digestion of synovial tissue specimens derived from
patients with RA and from 1 healthy control, as previously
described (6). All patients with RA fulfilled the American
College of Rheumatology (formerly, the American Rheumatism Association) criteria for the diagnosis of RA (7). Approval was granted by the local ethics committee, and informed
consent was obtained from all patients.
SFCs were obtained from synovial fluid of patients
with RA by culturing the adherent cells. Normal skin fibroblasts as well as foreskin fibroblasts (FSFBs) served as controls
for the RASFs, whereas SFCs from patients with gouty arthritis or oligoarthritis served as controls for the RA SFCs.
Monocytes were isolated from 50 ml of peripheral blood
obtained from 1 healthy volunteer donor, using the Ficoll
gradient centrifugation method. Separated monocytes were
initially cultured in RPMI 1640 medium containing 10% fetal
calf serum (FCS), glutamine, penicillin, and streptomycin.
Chondrocytes were isolated from articular cartilage derived
from 1 trauma patient, using enzymatic digestion with pronase
(Calbiochem, La Jolla, CA) for 2 hours, and with collagenase
P (Boehringer Mannheim, Rotkreuz, Switzerland) for 18 hours
at 37°C under continuous agitation, as described by Neidhart et
al (8). After extensive washing, cells were cultured in Dulbec2811
Table 1. Inducibility of retroviruses from rheumatoid arthritis synovial fibroblasts (RASFs) and RA synovial fluid cells (SFCs)*
Time point of analysis, days
I. Induction with genotoxic
Irradiation with x-rays
II. Coculture in the allosystem
Monocytes from peripheral blood
III. Coculture in the xenosystem‡
Vero cells as feeder layer
BSC cells as feeder layer
Vero cells as top layer
BSC cells as top layer
skin fibroblast
skin fibroblast
NSF, 1 skin fibroblast
skin fibroblast
2 RA SFCs, 2 SFCs from patients with
unclassified oligoarthritis, 1 SFC from a
patient with gouty arthritis†
3 RASFs, 1 NSF
5 RASFs, 1 FSFB (both gamma-irradiated)
0, 2
0, 2
0, 2
0, 2 (2 Gy) or 0, 3 (8 Gy)
0, 3, 7, 11, 14, 18, 21, 25
0, 3, 7, 10, 14, 17, 20
0, 2, 7
0, 2, 22
0, 2, 22
0, 2, 22
0, 2, 22
0, 2, 22
* For the product-enhanced reverse transcriptase (PERT) assay, cell culture supernatants from baby hamster kidney cells, murine mast cells, and
pig fibroblasts, all of which produce endogenous retroviruses in large quantities, were used as positive controls. For electron microscopy (EM),
vesicular stomatitis virus–infected baby hamster kidney cells served as positive controls. Results were considered to be negative when no virus-like
particles were detected by EM and/or RT activity was ⬍100 nU/ml, as determined by the PERT assay, which is equivalent to 10⫺7 units/ml of human
immunodeficiency virus type 1 (HIV-1) RT. This value corresponds to a virus concentration of ⬃250 HIV-1 particles/ml, or 2.5 particles per test
when culture supernatant was tested neat. NSF ⫽ normal synovial fibroblast; FSFB ⫽ foreskin fibroblast.
† SFCs derived from a patient with gouty arthritis were tested by PERT assay only until day 10.
‡ RASFs and FSFBs were either nonirradiated or were irradiated with 100-Gy gamma rays before beginning the coculture.
supernatant was centrifuged (20,000g) in a microfuge at 4°C for
60 minutes. The resulting pellet was resuspended in 100 ␮l of
sample buffer B (50 mM KCl, 10 mM Tris HCl, pH 7.5, 0.25
mM EDTA-Na2, 5 mM dithiothreitol, 20% glycerol, 60 ␮g/ml
bovine serum albumin). The RT reaction was carried out in
duplicate, with 10 ␮l of a specimen in a total reaction volume
of 30 ␮l, at 37°C for 90 minutes. After heat inactivation of the
enzyme and degradation of template RNA by RNase A
digestion (400 ng), amplification for 40 cycles in a total
reaction volume of 55 ␮l was executed on an ABI 7700
sequence detector (Applied Biosystems, Norwalk, CT), using a
FAM-labeled TaqMan probe (Applied Biosystems).
For quantification of RT, serially diluted recombinant
human immunodeficiency virus type 1 (HIV-1) RT (Roche
Molecular Biochemicals, Rotkreuz, Switzerland) was used to
generate an external standard curve. For verification of virus
particle preparation, and as positive-reaction controls, supernatants from baby hamster kidney cells, murine mast cells, and
pig fibroblasts, all of which release endogenous retroviruses,
were used successfully.
In order to activate potentially hidden retroviruses,
RASFs were treated with DNA-damaging agents such as
AzaC, BUDR, IUDR, or irradiation with 2-Gy or 8-Gy x-rays
(Table 1). Neither the sensitive PERT assay at 48 hours or 72
hours nor EM after 2–8 days of treatment detected RT activity
or VLPs (Figure 1). When RA SFCs and RASFs were cocultured with the highly proliferative human lymphotropic H9
cells for 20 days or 25 days, respectively, RT activity in the cell
culture supernatants stayed at background levels of ⬍100
nU/ml HIV-1 RT equivalent and showed no indication of an
RASFs and human chondrocytes were also cocultured
for 1 week, and no evidence of enhanced RT activity was
observed. Monocytes obtained from a healthy human donor
were cocultured with RASFs irradiated with 100-Gy gamma
rays. Again, no increased RT activity suggesting activation of
RT-proficient retroviruses was demonstrated. Finally, RASFs
were cocultured with the monkey kidney cell lines BSC and
Vero. These cocultures were performed with either irradiated
or nonirradiated RASFs for a period of ⬎3 weeks. Alternatively, the monkey kidney cells or the RASFs served as
feeder-layer cultures. In none of these xenotropic coculture
settings was a positive result in the PERT assay detected. In
summary, the present investigation did not provide any evidence for the presence of infectious RT-proficient viruses or
VLPs in RA synovial fibroblasts.
Because the PERT assay detects viral RT activity
rather than a retroviral sequence, it can identify both known
and unknown retroviruses. This is the first study in which the
sensitive PERT assay was used in an attempt to detect
retroviruses in cell culture supernatants from RASFs and RA
SFCs. To address the possibility of retroviral latency, we used
several established methods to induce and activate latent
retroviruses and combined these procedures with the PERT
assay and classic EM.
First, RASFs were treated with the DNA
hypomethylation-inducing agent AzaC or with the halogenated
pyrimidines BUDR or IUDR, and no evidence of activation of
retroviruses from RASFs was demonstrated. After treating the
Ki-BALB cell line (Kirsten sarcoma cell line) with AzaC (2
␮g/ml for 24 hours), Lasneret et al (10) observed release of RT
within 24 hours and detected intracisternal type A particles
Figure 1. Treatment of rheumatoid arthritis synovial fibroblasts. A,
With 5-aza-2⬘-deoxycytidine, 2 ␮g/ml, for 24 hours. B, With 5-bromo2⬘-deoxyuridine, 25 ␮g/ml, for 24 hours. C, With irradiation with 8-Gy
x-rays. D, With iodo-deoxyuridine, 20 ␮g/ml, for 20 hours. Retroviruses
should present as uniform, double-membrane–covered particles (⬃100
nm in size), with an electronic-dense core structure. Retrovirus-like
particles were detected in none of the inductive settings. Vesicular
stomatitis virus–infected baby hamster kidney cells served as positive
controls (not shown).
(IAPs) by EM, while mainly type C retroviruses were released
when IUDR (20 ␮g/ml for 24 hours) was used. Also, Yaniv and
Eylan (11) used IUDR treatment (30 ␮g/ml for 24 hours) in a
transformed hamster cell line to induce VLPs. Within 48–72
hours, VLPs were present, as determined by density gradient
centrifugation and consecutive conventional RT assay. Rhim et
al (12) induced type C retroviruses and IAPs using BUDR (25
␮g/ml for 24 hours) from guinea pig nonproducer cells, based
on EM and the classic RT assay. Of note, the conventional RT
assay used in the latter experiments is ⬃106-fold less sensitive
than the PERT assay. Because ionizing irradiation has also
been shown to induce retroviruses (13), RASFs were treated
with 2-Gy or 8-Gy x-rays, respectively, but there was no
indication of activation of retroviruses.
Next, RASFs and RA SFCs were cocultured with
human cells; this technique has been shown to be useful for
isolating retroviruses from human cells. Garry et al (14)
detected human IAP using coculture with the H9 lymphoblastic cell line. Also, Griffiths and coworkers (15) observed viral
RNA of HRV-5 when salivary gland tissue specimens from a
patient with Sjögren’s syndrome were cocultured with H9 cells.
Accordingly, we cocultured RASFs and RA SFCs with H9 cells
for 25 days and 20 days, respectively. Within this period, no
evidence for an increase in RT activity as measured by the
PERT assay was observed, suggesting that there was no
productive infection of the H9 cells with a retrovirus. Cocul-
turing of RASFs and chondrocytes also did not result in
activation of latent retroviruses.
Apart from using the coculture approach in the allosystem, we also tried to activate and transfer retroviruses by
receptor-independent mechanisms and to induce retroviruses
in the xenococulture setting. Holmgren et al and Spetz et al
(16,17) presented evidence that a functional gene transfer of
HIV DNA and Epstein-Barr virus DNA could take place by
receptor-independent mechanisms, when the cells were cocultured with lethally irradiated donor cells. Therefore, we cocultured irradiated RASFs with human monocytes for ⬎3 weeks.
Our results did not provide any evidence for viral infection
resulting in increased RT activity.
We also performed coculturing in the xenotropic system. Patience et al (18) previously demonstrated that pig
endogenous retroviruses (PERV) could infect human kidney
293 cell line when the cells were cocultured with PERVproducing pig fibroblasts. Accordingly, we questioned whether
latent retroviruses from RASFs could infect rapidly proliferating monkey kidney cells such as Vero and BSC. To further
increase the likelihood of activation of latent retroviruses,
RASFs were irradiated with 100-Gy gamma rays in half of the
experiments. Cocultures were maintained for ⬎3 weeks, and
during this period no activation of RT-proficient viruses was
Because the PERT assay is extremely sensitive, and
because results of all induction experiments and coculture
investigations were negative, the presence of RT-proficient
particle-forming retroviruses in RASFs appears to be unlikely.
In conclusion, results of the current investigations do not
support involvement of infectious retroviruses in activated
RASFs, but neither do they exclude a role of retroviruses in
other cell types such as macrophages or T cells or a role of
endogenous retroelements such as L1 in RA.
Christian A. Seemayer, MD
Stefan A. Kolb, MD
Michel Neidhart, PhD
Shiro Ohshima, MD
Renate E. Gay, MD
Beat A. Michel, MD
Steffen Gay, MD
University Hospital Zurich
Jürg Böni, DVM
Jörg Schüpbach, MD
University of Zurich
Zurich, Switzerland
1. Herrmann M, Neidhart M, Gay S, Hagenhofer M, Kalden JR.
Retrovirus-associated rheumatic syndromes. Curr Opin Rheumatol 1998;10:347–54.
2. Wu J, Zhou T, He J, Mountz JD. Autoimmune disease in mice due
to integration of an endogenous retrovirus in an apoptosis gene. J
Exp Med 1993;178:461–8.
3. Griffiths DJ, Cooke SP, Herve C, Rigby SP, Mallon E, Hajeer A,
et al. Detection of human retrovirus 5 in patients with arthritis and
systemic lupus erythematosus. Arthritis Rheum 1999;42:448–54.
4. Gaudin P, Moutet F, Tuke PW, Garson JA. Absence of human
retrovirus 5 in French patients with rheumatoid arthritis: comment
on the article by Griffiths et al. Arthritis Rheum 1999;42:2492–4.
5. Stransky G, Vernon J, Aicher WK, Moreland LW, Gay RE, Gay S.
Virus-like particles in synovial fluids from patients with rheumatoid arthritis. Br J Rheumatol 1993;32:1044–8.
6. Pap T, van der Laan WH, Aupperle KR, Gay RE, Verheijen JH,
Firestein GS, et al. Modulation of fibroblast-mediated cartilage
degradation by articular chondrocytes in rheumatoid arthritis.
Arthritis Rheum 2000;43:2531–6.
Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF,
Cooper NS, et al. The American Rheumatism Association 1987
revised criteria for the classification of rheumatoid arthritis.
Arthritis Rheum 1988;31:315–24.
Neidhart M, Gay RE, Gay S. Anti–interleukin-1 and anti-CD44
interventions producing significant inhibition of cartilage destruction in an in vitro model of cartilage invasion by rheumatoid
arthritis synovial fibroblasts. Arthritis Rheum 2000;43:1719–28.
Burgisser P, Vernazza P, Flepp M, Boni J, Tomasik Z, Hummel U, et
al. Performance of five different assays for the quantification of viral
load in persons infected with various subtypes of HIV-1: Swiss HIV
Cohort Study. J Acquir Immune Defic Syndr 2000;23:138–44.
Lasneret J, Canivet M, Hojman-Montes de Oca F, Tobaly J,
Emanoil-Ravicovitch R, Peries J. Activation of intracisternal a
particles by 5-azacytidine in mouse Ki-BALB cell line. Virology
Yaniv A, Eylan E. Iododeoxyuridine activation of oncornaviruslike particles of hamster origin. Br J Cancer 1977;36:397–9.
Rhim JS, Duh FG, Cho HY, Wuu KD, Vernon ML. Brief
communication: activation by 5-bromo-2⬘-deoxyuridine of particles resembling guinea-pig leukemia virus from guinea-pig nonproducer cells. J Natl Cancer Inst 1973;51:1327–31.
Niwa O, Sugahara T. Radiation induction of endogenous type C
virus from mouse cells transformed in vitro by murine sarcoma
virus. J Natl Cancer Inst 1979;62:329–35.
Garry RF, Fermin CD, Hart DJ, Alexander SS, Donehower LA,
Luo-Zhang H. Detection of a human intracisternal A-type retroviral particle antigenically related to HIV. Science 1990;250:
Griffiths DJ, Venables PJ, Weiss RA, Boyd MT. A novel exogenous retrovirus sequence identified in humans. J Virol 1997;71:
Holmgren L, Szeles A, Rajnavolgyi E, Folkman J, Klein G,
Ernberg I, et al. Horizontal transfer of DNA by the uptake of
apoptotic bodies. Blood 1999;93:3956–63.
Spetz AL, Patterson BK, Lore K, Andersson J, Holmgren L.
Functional gene transfer of HIV DNA by an HIV receptorindependent mechanism. J Immunol 1999;163:736–42.
Patience C, Takeuchi Y, Weiss RA. Infection of human cells by an
endogenous retrovirus of pigs. Nat Med 1997;3:282–6.
DOI 10.1002/art.10532
Magnetic resonance imaging as a predictor of
progressive joint destruction in neuropathic joint
Neuropathic joint disease, or “Charcot joint,” is a
disabling and rapidly progressive type of degenerative arthritis
that occurs in the presence of sensory neuropathy (1). The
radiographic features of established neuropathic joint disease
include osteosclerosis, bone proliferation and resorption, and
severe joint destruction (2). Diabetes mellitus is the most
common cause of neuropathic joint disease, with a prevalence
of 0.1–0.5% (3), but neuropathic joint disease can be seen in
the setting of many other neuropathies (1). Unfortunately, at
the time of clinical and radiographic recognition, neuropathic
joint disease is often advanced, with severe loss of joint
function that limits the treatment options. A prerequisite for
more effective treatment of neuropathic joint disease is the
early identification of patients at a time when joint architecture
is still well preserved. Furthermore, the basic pathophysiologic
mechanisms of neuropathic joint disease are poorly under-
stood, although repetitive microtrauma and autonomic nervous dysfunction are implicated (4).
Magnetic resonance imaging (MRI) is a multiplanar
modality that is excellent for visualizing joints and that shows
abnormalities at a stage when radiographs may still be normal
(5). The MRI findings of established neuropathic joint disease
include extensive bone edema and extensive bone and joint
destruction (3), with cases of extensive bone edema also being
described in early disease (6). However, it has not been
reported whether MRI of early neuropathic joint disease can
predict subsequent evolution of radiographic destructive
changes. The purpose of this report was to describe MR
features of radiographically normal early neuropathic joint
disease to better clarify pathogenic mechanisms and to determine whether MRI could predict joint failure in neuropathic
joint disease.
We identified 3 patients (2 with diabetic and 1 with
alcohol-related peripheral neuropathy) who presented with
joint swelling (1 with ankle and 2 with foot swelling) and mild
joint pain. Patient 1 was a 29-year-old woman with poorly
controlled type 1 diabetes mellitus. Painful ankle swelling had
developed suddenly without preceding trauma and had continued for 5 weeks, at which time the patient came to the clinic.
On examination, there was swelling of the left ankle, mild
tenderness, no erythema, and a well-maintained range of
movement. Patient 2 was a 66-year-old man with type 1
diabetes mellitus who developed a mildly painful, swollen foot
during a 24-hour period with no preceding trauma. On examination, there was swelling along the left medial longitudinal
arch. Patient 3 was a 65-year-old alcoholic man who presented
with a swollen foot that was causing discomfort that had
gradually developed during a 3-month period. There was no
history of gout, and swelling over the dorsum of the foot was
noted, but there was no erythema or tenderness. All 3 patients
had loss of proprioception and vibration sense, and patients 2
and 3 also had a glove and stocking sensory loss. Results of
routine investigations, including complete blood count, serum
urate and rheumatoid factor, titers, and immunoglobulin levels, were normal.
Radiographic assessment of the involved ankle joint
(Figure 1A) and feet was normal. Coronal T1 and coronal and
axial T2 spectral fat suppression with inversion recovery MR
images of the involved joints were obtained in all 3 patients
using a 1.5T Gyroscan ACSNT MR scanner (Philips, Best, The
Netherlands). The most striking feature of the MR scans was
extensive bone marrow edema in the tibia, talus, and navicular
bone in patient 1 (Figure 1B) and in the metatarsal and
cuneiform bones in patients 2 and 3. A small subchondral
fracture was also evident in patient 1 (Figure 1B). Florid
synovial and extracapsular changes were also evident in patient
1, but not in patients 2 and 3. The extent of abnormality in
patient 1 raised the possibility of infection, but subsequent
synovial fluid and tissue cultures were negative. All patients
were treated conservatively with rest, antiinflammatory drugs,
and orthotics. At a mean followup of 4 months, all patients
developed radiographic changes characteristic of neuropathic
joint disease with progressive joint destruction with marked
osteophytosis (Figure 1C). This was accompanied by severe
joint instability in the case of patient 1.
This report demonstrates the MRI findings in the
earliest stages of neuropathic arthritis at a time when radiographic assessment was normal and joint function was good.
Figure 1. A, Radiograph of the ankle of patient 1, who presented with a swollen ankle, showing normal joint architecture. B, A T2 spectral fat
suppression with inversion recovery magnetic resonance scan of the ankle of patient 1, taken at the same time as the radiograph in A. There is
extensive bone marrow edema of the tibia (large asterisk), talus (medium-size asterisk), and navicular bone (small asterisk). A small subchondral
fracture is also evident near the ankle joint space (curved solid arrow). A joint effusion and extracapsular enhancement are also evident (open
arrowhead). C, Radiograph of the same ankle joint 3 months after the initial scan, showing severe degenerative changes with joint disruption and
loss of joint space (thin solid arrow), periarticular joint sclerosis, and prominent osteophytosis (thick open arrow).
The ability to define neuropathic joint disease at such an early
stage may have implications for the development of therapies
aimed at treatment of bone edema in an attempt to prevent
joint damage. However, the natural history of bone edema on
MRI in neuropathic joint disease and whether it signifies
disease progression in all cases or whether it can regress
without joint destruction all remain to be determined. Since
neuropathic joint disease is usually associated with progressive
joint damage, it is possible that these bone changes are the
harbinger of future joint failure.
Furthermore, the differential diagnosis of MRIassociated bone pathology merits other considerations, such as
infection, joint trauma, seronegative arthritis, reflex sympathetic dystrophy (RSD), and transient bone marrow edema (6).
Indeed, simple biomechanical alterations of the feet are associated with bone edema (7). Mechanisms thought to be
associated with the development of bone edema include bone
microdamage. Early neuropathic joint disease could be differentiated from these various other conditions based on the
clinical history and the presence of peripheral neuropathy.
These MRI observations of the early events in neuropathic joint disease may have implications for a better understanding of disease pathogenesis. There are two proposed
mechanisms by which a neuropathic joint develops. A hypothesis of neurotrauma is based on loss of protective reflexes with
subsequent repetitive microtrauma causing progressive joint
damage. The finding of a subchondral fracture in one of our
patients, in addition to the known propensity of altered joint
biomechanics to lead to bone edema, is in keeping with the
neurotrauma hypothesis. A second theory suggests that autonomic neuropathy leads to an increased bone blood flow in the
affected region, with osteoclast activation (4). Autonomic
disturbances in RSD are also associated with diffuse bone
edema, possibly indicating an autonomic component to disease.
In conclusion, these findings show that MRI can detect
the early events of neuropathic joint disease and may predict
subsequent joint destruction. These findings have implications
for a better understanding of the early events in neuropathic
joint disease and for therapy development at a stage when
there is no radiographic damage. The recent demonstration of
the efficacy of bisphosphonates for the therapy of neuropathic
joint disease is in keeping with a predominantly bone-based
pathology in early disease (8).
Dr. Greenstein’s work was supported by an Arthritis and Rheumatism Campaign (ARC) grant. Dr. Emery is an ARC Professor of
Rheumatology. Dr. McGonagle’s work was supported by the Medical
Research Council, UK.
Adam S. Greenstein, MRCP
Helena Marzo-Ortega, MRCP
Paul Emery, MA, MD, FRCP
University of Leeds
Philip O’Connor, FRCR
The Leeds General Infirmary
Dennis McGonagle, MRCPI
University of Leeds
Leeds, UK
Calderdale Royal Hospital
Salterhebble, Halifax, UK
1. Delrieu F, Menkes C, van de Putte L. Neuropathic arthropathy. In:
Klippel JH, Dieppe P, editors. Rheumatology. 2nd ed. London:
Mosby; 1997. p. 7.10.1–10.6.
2. Brower AC. The acute neuropathic joint. Arthritis Rheum 1988;31:
3. Tomas MB, Patel M, Warwin SE, Palestro CJ. The diabetic foot.
Br J Radiol 2000;73:443–50.
4. Brower AC, Allman RM. Pathogenesis of the neurotrophic
joint: neurotraumatic vs. neurovascular. Radiology 1981;139:
5. Peterfy CG. MR imaging. Baillieres Clin Rheumatol 1996;10:635–78.
6. Weishaupt D, Schweitzer ME. MR imaging of the foot and ankle:
patterns of bone marrow signal abnormalities. Eur Radiol 2002;12:
7. Schweitzer ME, White LM. Does altered biomechanics cause
marrow edema? Radiology 1996;198:851–3.
8. Jude EB, Selby PL, Burgess J, Lilleystone P, Mawer EB, Page SR,
et al. Bisphosphonates in the treatment of Charcot neuroarthropathy: a double blind randomised controlled trial. Diabetologia
DOI 10.1002/art.10553
Clinical Image: Crystal-induced arthritis mimicking rheumatoid arthritis in a patient with monoclonal gammopathy
The patient, a 57-year-old man, presented with seronegative, erosive polyarthritis. The small joints of the hands and feet were
initially affected, followed by the development of arthritis in the ankles, knees, and elbows, and accompanied by ecchymoses and
bullae on both feet. Serum analysis demonstrated a cryoprecipitable and crystal-forming IgG␬ (M-component), and immunohistochemistry analysis of a bone marrow biopsy specimen showed a small increase in monoclonal plasma cells positive for IgG␬.
Synovial biopsy of a metacarpophalangeal joint revealed severe crystal synovitis with fibrinoid material and infiltration by
lymphocytes, plasma cells, neutrophil granulocytes, and histiocytes. In the fibrinoid material and superficially in the synovial tissue
there were many crystals of different shapes and sizes; some were rhomboid or rectangular (above). Electron microscopic
investigation revealed that the crystals had a homogeneous structure consistent with a proteinaceous composition, and
immunoelectron microscopy demonstrated the presence of IgG␬. The deposition of crystals in various organs, including joints, is a
rare manifestation of cryoglobulinemia.
H. Aarset, MD
University Hospital of Trondheim
Trondheim, Norway
J. Dekker, MD
Innherred Hospital
Levanger, Norway
S.-H. Brorson, PhD
Ullevål University Hospital
Oslo, Norway
G. Husby, MD, PhD
National Hospital
Oslo, Norway
Без категории
Размер файла
529 Кб
progressive, destruction, neuropathy, magnetic, joint, imagine, disease, resonance, predictor
Пожаловаться на содержимое документа