Skip Navigation


Rheumatology Advance Access originally published online on December 13, 2005
Rheumatology 2006 45(5):533-537; doi:10.1093/rheumatology/kei206
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/5/533    most recent
kei206v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Dai, L.
Right arrow Articles by Schumacher, H. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dai, L.
Right arrow Articles by Schumacher, H. R.
Related Collections
Right arrow Soft Tissue Rheumatism
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Detection and initial characterization of synovial lining fragments in synovial fluid

L. Dai1,5, F. Pessler3, L. X. Chen1,2,4, G. Clayburne2 and H. R. Schumacher1,2

1 University of Pennsylvania, 2 Philadelphia VA Medical Center, 3 The Children's Hospital of Philadelphia, 4 Presbyterian Medical Center, Philadelphia, PA, USA and 5 The 2nd Affiliated Hospital, Sun Yat-sen University, Guangzhou, People's Republic of China.

Corresponding authors: L. Dai and H. R. Schumacher. E-mail: liedai2004{at}hotmail.com, schumacr{at}mail.med.upenn.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Objective. Free fragments of synovium have occasionally been seen in synovial fluid but have not been studied systematically. We wished to establish a method for the reliable detection of these fragments in joint and bursa effusions and begin to characterize them by histochemical and immunohistochemical methods.

Methods. Cell smears, wet drop preparations and cytospins were prepared from 39 consecutive joint and bursa effusions. Paraffin cell blocks were prepared from a subset. Analysis encompassed standard and polarized light microscopy, histochemistry, immunohistochemistry and transmission electron microscopy (EM). Synovial biopsy tissue from one different patient was examined for comparison.

Results. Tissue fragments were not seen in Wright-stained cell smears and only rarely in wet drop preparations. In contrast, variously sized fragments with the histological appearance of hyperplastic synovial lining were detected in ethanol-fixed, haematoxylin/eosin-stained cytospins from bursitis and all arthropathies studied [17/24 (71%) of non-inflammatory and 12/15 (80%) of inflammatory specimens]. Immunostaining revealed CD68 expression in a subset of cells in a pattern characteristic of hyperplastic synovial lining. Juxtaposed cells with morphological features of macrophage-like and fibroblast-like synoviocytes were seen by EM.

Conclusions. Synovial lining fragments can be detected in effusions from diverse arthropathies and bursitis. They maintain important properties of the synovial lining and can be analysed by immunohistochemistry. They may afford the opportunity to study a relatively pure preparation of synovial lining cells without the need for cell culture, and to evaluate their possible role in augmenting or perpetuating synovitis or joint damage.

KEY WORDS: CD68, Rice bodies, Synovial lining, Synovial fluid, Synovial fluid analysis, Synovium


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Inflammatory cells and individual synovial lining cells are easily detected in synovial fluid (SF) and have been the subject of numerous studies. Recent work has focused on cultured SF fibroblasts to study the synovial lining [1, 2], but artefacts resulting from ex vivo propagation may be confounding factors. Intact synovial lining tissue, on the other hand, needs to be obtained by more invasive procedures, i.e. arthroscopy, closed needle biopsy or open biopsy. It would therefore be desirable to obtain synovial lining tissue by additional, less invasive methods, such as arthrocentesis.

Tissue-like structures have been described in SF and SF sediments. Rice bodies are fibrin-impregnated, macroscopically visible structures that are believed to be derived from necrotic villi [3, 4]. Microscopic analysis sometimes reveals aggregates of cells entrapped in the fibrin strands typical of inflamed SF. Denser, hyalinized particles that occur preferentially in rheumatoid arthritis (RA) SF have been termed microscopic rice bodies [5]. In our own experience, looser groups of exudate cells, apparently held together by a mesh of fibrin (‘exudative cell clumps’), sometimes resemble microscopic rice bodies, but to our knowledge have not been studied systematically.

Intact pieces of synovium or synovial lining have been documented in SF [6] but have not been well characterized. Here we describe a protocol for the reliable detection of synovial lining fragments in inflammatory and non-inflammatory SF. This protocol employs crucial steps not used in traditional SF analysis, which explains why the fragments have largely gone unappreciated in the past. They may afford opportunities to study aspects of the synovial lining without biopsy and without introducing changes that result from propagating cells in culture. Moreover, they may contribute to persistent synovial inflammation or joint damage by releasing cytokines or apoptotic cell debris into the SF.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Fluid collection and sample preparation
Thirty-nine consecutive joint and bursa effusions from adult patients seen in the rheumatology out-patient department of the Philadelphia VA Medical Center were aspirated as clinically indicated, using a standard sterile procedure [7] and a 20 gauge needle. The study was approved by the local ethics committee (Institutional Review Board). Informed consent was waived because: (i) specimens were used that would otherwise have been discarded; and (ii) no personally identifiable information was retained with the specimens. All specimens were collected into heparinized Vacutainers and processed within 4 h. White blood cell (WBC) and differential counts were determined with a haemocytometer and Wright-stained cell smears, respectively. Wet drop preparations were examined for tissue fragments and crystals with and without alizarin red S stain [7, 8]. Inflammatory fluids (WBC >2000/mm3) were centrifuged at 170 g for 10 min; the resulting pellets were resuspended in 2–5 ml of an ethanol-based fixation fluid (CytoRich Red Collection FluidTM; Thermo Shandon), left at room temperature for 15–30 min to allow fixation and erythrocyte lysis, and then centrifuged and washed twice in fixation fluid. The final pellets were resuspended in 2–5 ml fixation fluid. In addition, in the more viscous non-inflammatory fluids (WBC <2000/mm3) mucin was lysed with dithiothreitol (Stat-Pack SputolysinTM; Caldon Biotech) for 30 min before fixation, followed by centrifugation at 600 g for 10 min. An aliquot of each specimen was centrifuged onto a coated cytospin slide (CytoslideTM; Thermo Electron Corporation) by rotating in a Shandon CytoslideTM device at 800 r.p.m. for 10 min. The input volume (10–100 µl) was adjusted to result in a single layer of cells and tissue fragments on the slide. At least three cytospins were prepared from each specimen. The presence of a tissue fragment was defined as an aggregate of five or more mononuclear cells on one or more cytospins prepared from a given specimen. A diagram of the detection protocol is depicted in Fig. 1.


Figure 1
View larger version (10K):
[in this window]
[in a new window]
  		FIG. 1. Flow diagram of the detection protocol.

 
Histochemistry and immunohistochemistry
All slides were examined by standard and polarized light microscopy. Wright, alizarin red S and haematoxylin/eosin (H&E) stains were performed according to standard protocols [7, 8], and the pentachrome stain according to the modified Russell–Movat method [9]. Paraffin cell blocks were made from 11 specimens that contained easily visible, tight pellets after low-speed centrifugation. An automated immunostaining system (Ventana BenchMark IHC/ISH Staining Module) was used for immunostaining, using a commercial anti-CD68 (KP-1) primary antibody (Ventana ConfirmTM) and the Ventana iView DAB Detection Kit.

Electron microscopy
SF pellets were obtained by low-speed centrifugation, fixed briefly, disrupted into fragments of approximately 1x1 mm, fixed for 4 h in 0.5x Karnovsky's paraformaldehyde–glutaraldehyde fixative, and then sectioned and processed as described previously [10]. A Zeiss EM-10 electron microscope with a 60-kV beam was used.

Statistical analysis
Linear regression was performed using the on-line program available at http://home.ubalt.edu/ntsbarsh/Business-stat/otherapplets/Regression.htm.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Clinical characteristics of patients and SFs
Clinical diagnoses were: osteoarthritis (OA), eight cases; OA with calcium pyrophosphate dihydrate (CPPD) crystals, seven; RA, six; interval gout, five; gout with flare, four; olecranon bursitis, three; psoriatic arthritis (PsA), two; undifferentiated arthritis, two; pseudogout, one; and Lyme disease, one. Thirty-eight patients (97%) were male. Age ranged from 48 to 84 yr (mean, 66). Thirty-four effusions were obtained from knee joints, one from the acromioclavicular joint, and four from the olecranon bursa. Patient data and results of routine SF analysis are summarized in Table 1.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Characteristics of patients and effusions

 
Identification of SF tissue fragments in cytospins and SF sediments
Tissue fragments were not seen in Wright-stained cell smears and only rarely (<1%) in wet drop preparations. When SF was centrifuged onto coated cytospin slides, Wright stains could not be interpreted because of abundant, fixed fibrin deposits resulting from the ethanol fixation (data not shown). In contrast, the H&E stain revealed well-organized, variously sized, elongate and spindle-shaped tissue fragments, which consisted of large mononuclear cells with pale nuclei and abundant cytoplasm (Fig. 2A and B). These ranged in size from five or six to several hundred cells and were detected in bursitis and all types of arthropathies. The detection rate (percentage of specimens with the same clinical diagnosis in which at least one lining fragment was detected) ranged from 40 to 100% (Table 1), with an average of 71% for all non-inflammatory and 80% for all inflammatory fluids. Linear regression analysis revealed no significant correlation between the detection rate and the degree of inflammation, as measured by SF cell count number (r = 0.31, P = 0.39). The total number of fragments per SF aspirate (extrapolated from the volume of fixation fluid in which the fragments were suspended, the input volume into the cytospins, and the number of fragments with more than five or six cells per cytospin) ranged between 15 and over 2500. The relative size distribution of the fragments was not determined. The fragments were also detected when centrifuged SF pellets were embedded in paraffin, the resulting blocks sectioned and stained with H&E (Fig. 2C). Their histological appearance and the almost complete absence of blood vessels and other subintimal structures suggested that these fragments were derived from the synovial lining (intima). These fragments were easily distinguished from exudative cell clumps, which appeared as loose, poorly organized aggregates containing heterogeneous inflammatory cells embedded in a loose matrix (Fig. 2E). In sections from a synovial biopsy specimen from a different individual, synovial lining fragments which looked very similar to those in Fig. 2B and C were seen adjacent to or loosely attached to synovium (Fig. 2G and H).


Figure 2
View larger version (61K):
[in this window]
[in a new window]
  		FIG. 2. (A and B) Synovial tissue fragments in cytospin preparations. (B) Larger synovium fragments derived from an olecranon bursitis effusion. (A) Smaller fragment detected in an OA knee effusion. Both images: H&E stain, original magnification 200x. (C–F) Differentiating synovial lining fragments from exudative cell clumps using sections from paraffin-embedded SF sediments. (C) Synovial lining fragment from an RA knee effusion, surrounded by aggregates of inflammatory cells (H&E stain, original magnification 400x). (E) Exudative cell clumps from an undifferentiated monoarthritis knee effusion (H&E stain, original magnification 200x). (D) Consecutive section from the specimen used in C. Solid arrows point to small collagen-containing areas, the dotted arrow to a small acellular, fibrin-rich area (pentachrome stain, original magnification 200x). (F) Exudative cell clump: consecutive section from the specimen used in E (white arrowheads, fibrin; white arrow, mucin, black arrow, collagen; pentachrome stain, original magnification 200x). (G) Closed needle synovial knee biopsy specimen from a patient with OA, showing synovial lining fragments adjacent to intact synovium (H&E stain, original magnification 200x). (H) Lining fragment loosely attached to synovium by a thin stalk (arrow, attachment stalk; pentachrome stain, original magnification 200x). (I) Anti-CD68 immunostain, sectioned paraffin cell block. Parallel section from the specimen used in C. Original magnification 600x.

 
Demonstration of intact synovial lining cells in the tissue fragment
To verify the identity of these fragments, pentachrome staining was performed. Diffuse red cytoplasmic staining was seen, and also small, yellow-stained collagen-containing areas (Fig. 2D, solid arrows). This appearance is characteristic of the synovial lining. Indeed, a tissue fragment found in a closed needle synovial biopsy specimen exhibited a similar staining pattern (Fig. 2H). The pentachrome stain also differentiated the tissue fragments from exudative cell clumps: the latter consisted of scattered neutrophils and small mononuclear cells embedded in a poorly organized matrix of fibrin (red), mucin (blue) and, less, collagen (yellow) (Fig. 2F). Fragments from one patient were examined by EM, which revealed juxtaposed cells with the morphological features of type A (macrophage-like) and type B (fibroblast-like) synovial lining cells (Fig. 3), as expected in the synovial lining [11].


Figure 3
View larger version (79K):
[in this window]
[in a new window]
  		FIG. 3. Transmission EM. (A) A cell with prominent vacuoles and dense chromatin probably corresponds to a macrophage-like, type A synoviocyte. Immediately adjacent, a cytoplasmic process with the typical features of a fibroblast-like, type B synoviocyte (prominent rough endoplasmic reticulum and mitochondria) is seen. M, macrophage-like synoviocyte; v, vacuole; F, fibroblast-like synoviocyte. Original magnification, 12 500x. (B) Further magnification, illustrating rough endoplasmic reticulum (rER) and mitochondria-rich cell process characteristic of fibroblast-like synoviocytes. mt, mitochondria. Original magnification, 47 500x.

 
Expression of CD68 in the tissue fragments
In another attempt to verify the identity of these fragments, we wished to determine whether they express CD68, a marker commonly expressed by macrophage-like (type A) synoviocytes [12]. Immunoperoxidase staining revealed intense CD68 staining in a subset of the cells (Fig. 2I), as typically seen in the synovial lining.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
Detection and characterization of synovial lining fragments
We have detected small tissue fragments in SF. Even though these fragments have gone essentially unnoticed in the past, we have detected them surprisingly frequently and easily. We attribute this ease of detection to four crucial steps in our protocol, which represent major differences from traditional SF analysis: (i) ethanol fixation; (ii) mucolysis of viscous specimens; (iii) use of cytospins; and (iv) staining with H&E instead of the commonly used Wright stain. We identified the fragments as synovial lining tissue by their characteristic light microscopic appearance, the detection of fibroblast-like and macrophage-like synoviocytes by EM, and by the detection of CD68 in a pattern typical of hyperplastic synovial lining. The results of the pentachrome stain, too, support the origin of the fragments from the synovial lining: small, scattered collagen-containing areas were detected, as may be seen in the synovial lining (Fig. 2D). We also clearly differentiated these fragments from rice bodies and exudative cell clumps: exudative cell clumps have a much looser, less organized structure and contain abundant fibrin and mucin (Fig. 2E and F), whereas rice bodies are visible macroscopically, are heavily fibrin-impregnated and may contain blood vessels [6]. Our fragments differ from microscopic rice bodies [5] by the absence of significant necrosis and hyalinization. An intriguing possibility is that a fraction of synovial lining fragments resist degradation, are impregnated with fibrin and invaded by leucocytes, and ultimately turn into microscopic rice bodies. Consistent with this idea, microscopic rice bodies were detected much less frequently than the fragments described in this report [5].

Origin of the synovial tissue fragments
In our experience, tissue fragments are also seen in some unstained wet preparations from traumatic arthrocenteses, which required multiple needle passages. Thus, traumatic dislodgement may be a cause in some cases. However, traumatic arthrocentesis occurs infrequently, and most of the specimens in this study were obtained easily during first needle passage and were not bloody. The detection of these tissue fragments barely attached to intact synovium (e.g. Fig. 2G and H) suggests a different mechanism of origin for the majority of them. It is conceivable that mechanical shearing forces or natural remodelling could dislodge structures like this one into the SF. Most of the fragments seen are significantly hyperplastic. Thus, avulsion may also be the result of relative tissue weakening from lack of nutrients.

Potential uses of the synovial tissue fragments
Paraffin-embedded SF sediments containing the fragments were easily used for immunohistochemistry (Fig. 2I), demonstrating that some, if not all, proteins and antigenic epitopes are preserved. Moreover, the fragments were detected in all arthropathies studied. We had arbitrarily determined to analyse three cytospins per specimen, which led to detection in all arthropathies studied, with a detection rate of at least 40% in the specimens of the same diagnosis. Conceivably, analysing higher numbers of cytospins per specimen could increase the detection rate further. Taken together, these observations suggest that the fragments could potentially be used to study protein expression changes in the synovial lining in arthropathies and bursitis. However, further studies are first needed to (i) determine if the amount of tissue obtained with these fragments suffices reliably for such analysis, and (ii) to determine how closely the fragments resemble the resident synovial lining at the molecular level. Since it does not require propagation of the tissue by culture, one approach to the latter question would be to isolate the tissue fragment and the resident synovial lining by laser capture microdissection, and then compare their PCR- or microarray-based transcript profiles with each other.

Potential significance of the synovial tissue fragments in persistent synovitis
SF fibroblasts display highly activated phenotypes and mediate cartilage destruction, and it has been suggested that they contribute significantly to inflammation in the synovium [1, 2]. Owing to their size, the tissue fragments would be expected to be relatively resistant to phagocytosis and may thus persist in SF for some time. We speculate that they may continue to secrete proinflammatory substances and release apoptotic cell debris, thus perpetuating an inflammatory or destructive process. Indeed, microparticles from apoptotic cells were recently shown to induce the synthesis of matrix metalloproteinases and proinflammatory cytokines in synovial fibroblasts [13]. This also agrees well with the observation that lavage at the time of intra-articular corticosteroid injection enhances the efficacy of the injection, presumably by removing proinflammatory substances and debris from the joint cavity [14, 15]. As seen in Table 1, we detected these fragments in patients with joint effusions but non-inflammatory SF cell counts. In these cases, the tissue fragments may have contributed to the persistent joint effusion.

We realize the limitations of this study. The fixation fluid lyses red blood cells, and we may thus have failed to notice some small blood vessels in the fragments. Immunostaining for vascular endothelial cells, for instance with anti-von Willebrand factor or anti-CD31 antibodies, might have led to the detection of some small blood vessels and revealed the presence of more subintimal tissue than suggested by our analysis. We have not attempted to demonstrate the presence of synovial lining fibroblasts by immunohistochemistry. This could be done using a specific marker for these cells, such as uridine diphosphoglucose dehydrogenase or complement decay accelerating factor [16]. However, the presence of large numbers of CD68-negative cells, their characteristic light microscopic appearance, and the detection of cells with the typical EM features of synovial lining fibroblasts (Fig. 3) all strongly suggest that synovial lining fibroblasts are contained in the tissue fragments.


   Acknowledgments
 
We thank J. Romano (Department of Pathology, Philadelphia VA Medical Center) for help with immunohistochemistry and image processing, D. Martinez (Histopathology Core, The Children's Hospital of Philadelphia) for help with digital image processing, and S. Rothfuss for preparing the EM specimens. L.D. was supported by a fellowship from the Asian Pacific League Against Rheumatism (APLAR), and F.P. by NIH training grants T32-AR007442 and T32-CA09140.

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Neidhart M, Seemayer CA, Hummel KM, Michel BA, Gay RE, Gay S. Functional characterization of adherent synovial fluid cells in rheumatoid arthritis: destructive potential in vitro and in vivo. Arthritis Rheum 2003;48:1873–80.[CrossRef][Medline]
  2. Pierer M, Rethage J, Seibl R et al. Chemokine secretion of rheumatoid arthritis synovial fibroblasts stimulated by Toll-like receptor 2 ligands. J Immunol 2004;172:1256–65.[Abstract/Free Full Text]
  3. Hanumappa S, Schumacher HR. Synovial-fluid rice bodies: identification, pathogenesis, and significance. Intern Med 1989;10:45–56.
  4. Berg E, Wainwright R, Barton B, Puchtler H, McDonald T. On the nature of rheumatoid rice bodies: an immunologic, histochemical, and electron microscope study. Arthritis Rheum 1977;20:1343–9.[Medline]
  5. Galvez J, Sola J, Ortuno G et al. Microscopic rice bodies in rheumatoid synovial fluid sediments. J Rheumatol 1992;19:1851–8.[Medline]
  6. Schumacher HR, Reginato AJ. Atlas of synovial fluid analysis and crystal identification. Philadelphia: Lea & Febiger, 1991.
  7. Schumacher HR. Synovial fluid analysis and synovial biopsy. In: Kelley WN, ed. Kelley's Textbook of Rheumatology, 6th ed. Philadelphia: Saunders, 2001:605–19.
  8. Paul H, Reginato AJ, Schumacher HR. Alizarin red S staining as a screening test to detect calcium compounds in synovial fluid. Arthritis Rheum 1983;26:191–200.[Medline]
  9. Prophet EB, Mills B, Arrington JB, Sobin LH. Laboratory methods in histotechnology. Washington, D.C.: American Registry of Pathology, 1994.
  10. Schumacher HR. Palindromic onset of rheumatoid arthritis. Clinical, synovial fluid, and biopsy studies. Arthritis Rheum 1982;25:361–9.[Medline]
  11. Schumacher HR Jr. Ultrastructure of the synovial membrane. Ann Clin Lab Sci 1975;5:489–98.[Abstract]
  12. Singh JA, Arayssi T, Duray P, Schumacher HR. Immunohistochemistry of normal human knee synovium: a quantitative study. Ann Rheum Dis 2004;63:785–90.[Abstract/Free Full Text]
  13. Distler JH, Jungel A, Huber LC et al. The induction of matrix metalloproteinase and cytokine expression in synovial fibroblasts stimulated with immune cell microparticles. Proc Natl Acad Sci USA 2005;102:2892–7.[Abstract/Free Full Text]
  14. Fitzgerald O, Hanly J, Callan A, McDonald K, Molony J, Bresnihan B. Effects of joint lavage on knee synovitis in rheumatoid arthritis. Br J Rheumatol 1985;24:6–10.[Abstract/Free Full Text]
  15. van Oosterhout M, Sont JK, van Laar JM. Superior effect of arthroscopic lavage compared with needle aspiration in the treatment of inflammatory arthritis of the knee. Rheumatology 2003;42:102–7.[Abstract/Free Full Text]
  16. Edwards JC. Fibroblast biology. Development and differentiation of synovial fibroblasts in arthritis. Arthritis Res 2000;2:344–7.[CrossRef][Web of Science][Medline]
Submitted 15 September 2005; Accepted 19 October 2005


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Rheumatology (Oxford)Home page
D. McGonagle and E. Jones
A potential role for synovial fluid mesenchymal stem cells in ligament regeneration
Rheumatology, August 1, 2008; 47(8): 1114 - 1116.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/5/533    most recent
kei206v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Dai, L.
Right arrow Articles by Schumacher, H. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dai, L.
Right arrow Articles by Schumacher, H. R.
Related Collections
Right arrow Soft Tissue Rheumatism
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?