Supravital Staining of Synovial Fluid With Testsimplets Antonio J. Reginato, M.D., lrama Maldonado, M.D., Anthony M. Reginato, P m . , Gerald F. Falasca, M.D., and Carolyn R. O’Connor, M.D.
We explored the use of Testsimplet (TS) in synovial fluid (SF) analysis. TS is a glass slide coated with a dry mixture of methylene blue and cresyl violet, which in contact with one drop of SF provides a stained fresh preparation. We applied the TS to the study of 159 SFs ofpatients with different rheumatic diseases. I n those SFs of patients with crystal-associated diseases, the crystal search was performed both on unstained preparations and with TS. TS was as good as the Wright’s and Papanicolaou stain in characterizing SF cells, lupus erythematosus cells, and detection of occasional bacteria. TSallowed a better visualization of Reiter’s cells, cartilage fragments, synovial villi, fat droplets, and fibrin. Crystals were identified in every TS of those patients with crystalassociated diseases. TS is a rapid and reproducible method of S F supravital staining. Crystals are well preserved for simultaneous examination with compensated polarized light. Diagn Cytopatho1 1992;8:147-152. 0 1992 WiIey-Liss, Inc. Key Words: Synovial fluid cytology; Cresyl violet; Methylene blue; Synovial fluid analysis; Synovial fluid fresh preparations
After the history and physical examination, synovial fluid (SF) analysis ranks as the most simple diagnostic procedure for the evaluation ofjoint diseases. In many patients with septic arthritis and crystal-associated arthritis, a few drops of SF often yield a rapid and definitive diagnosis. In other rheumatic diseases, SF analysis provides useful information about the intensity of the inflammatory process and helps to narrow the differential diagnosis to conditions associated with either “inflammatory” or “noninflammatory” fluids. Differentiation of inflammatory cells is usually performed in air-fixed Wright-stained preparations and ocReceived February 11, 1991. Accepted September 25, 1991. From the University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Camden, NJ; University of Pennsylvania, School of Medicine, Philadelphia, PA; Department of Biochemistry, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA. Address reprint requests to Antonio J. Reginato, M.D., Head, Division of Rheumatology, Department of Medicine, Three Cooper Plaza, Suite 220, Haddon Avenue, Camden, NJ 08103. 0 1992 WILEY-LISS, INC
casionally with the Papanicolaou stain. Wright stains of “noninflammatory” SF preparations are difficult to read when few cells are present or when the fluid is very viscous. Early reports suggested that supravital staining of fresh synovial fluid preparations using neutral red and Janus green were helpful in differentiating mononuclear cells in noninflammatory fluids. 6,7 We and others have been unable to corroborate these findings. 3,8 Supravital staining with cresyl violet and methylene blue is performed easily with precoated, dry glass slides containing these dyes (Testsimplet TM, Boehringer, Mannhein, Germany). 9-11 We report our experience with Testsimplet (TS) in the synovial fluid analysis of 159 patients with various rheumatic diseases. TS provided a rapid and reproducible method of SF supravital staining and allowed better visualization of Reiter cells, synovial cells, cartilage fragments, fat droplets, fibrin, monosodium urate, and calcium pyrophosphate dihydrate crystals than fixed, dried Wright-stained preparations. Calcium pyrophosphate and monosodium urate crystals were well preserved for simultaneous examination under compensated polarized light.
Materials and Methods TS was performed by placing a drop of SF on the center of the commercially available glass slide already containing the dried stains. The slide was covered with a glass cover slip and examined after 10-15 min9-l2 The differential cell count and the search for crystals under compensated polarized light with TS staining were compared with simultaneous same-patient control samples obtained with dry-fixed Wright-stained preparations and unstained synovial fluid fresh preparations, respectively. Routine SF analysis included a fresh preparation observed under regular and compensated polarized light, total cell count, and differential cell count in a dried, fixed Wright-stained preparation. These studies were performed on SF from 159 patients between 1988 and 1990 Diagnostic Cytopathology, Vol 8, No 2
147
REGINATO ET AL.
(Table I). Gram, acridine orange, 3, and alizarin red stainingsI4 were done when septic arthritis or apatite deposiiion disease were suspected. Culture for aerobic and nonaerobic bacteria of synovial fluid was performed in 30 inflammatory fluid samples in which septic arthritis was suspected. The diagnosis of septic arthritis was made by the recovery of Staphylococcus aureus from 7 and Neisseria gonorrhea from 3 SF samples. In those 5 1 patients with crystal-associated arthritis, the diagnosis was made by identification of monosodium urate crystals in 24, calcium pyrophosphate crystals in 19, apatite in 6, cholesterol in I , and lipid liquid crystals in 1 patient by polarized light microscopy. I5,l6 Fifteen of the 19 patients with pseudogout also had radiographic evidence of chondrocalcinosis of their knees, while the other 2 showed signs of severe degenerative joint disease. l 6 In the 6 patients with apatite deposition disease, periarticular calcific deposits were seen in bursae, tendons, or synovium, and their SF showed alizarin red stained clumps. In 2 of these patients the presence of apatite was confirmed by transmission electron microscopy and x-ray diffraction analysis. The diagnosis of post-traumatic arthritis or bursitis was made in 10 patients by a history of preceding trauma to the joint and the finding of a bloody, non-inflammatory fluid. In 5 of these patients subchondral fractures were observed on radiographs of affected joints. The diagnosis of Lyme arthritis was supported by elevated IgG and IgM antibodies by an ELISA assay and exclusion of other causes of inflammatory monoarthritis. The diagnoses of systemic lupus erythematosus, rheumatoid arthritis, l 9 Reiter’s Table I. Synovial Fluid Studied with Testsimplet Staining in 159 Patients with Arthritis Type of arthritis Septic arthritis or bursitis Lyme disease Gout Pseudogout Apatite deposition disease Cholesterol and lipid liquid crystals Rheumatoid arthritis Osteoarthritis Rtiter’s syndrome Psoriatic arthritis Sickle cell crisis Traumatic arthritis or bursitis Miscellaneous Sympathetic effusion due to cellulitis Vasculitis 13acterial endocarditis Corticosteroid pulse therapy Possible viral arthritis Monoarthritis of unknown cause Sickle cell disease Total
148
Number of patien ts 10 1 24 19 6 2 22 21 8 5
3 10
11 1
1 1 1 1
3 3
159
Diagnostic Cytopathology, Vol 8. No 2
syndrome, 2o and osteoarthritis 21 were established in accordance with the American College of Rheumatology recommendations. The diagnosis of psoriatic arthritis was established by the presence of characteristic scaly rash, nail pitting, and asymmetric inflammatory arthritis. In order to better recognize the morphologic aspects of synovial cells in TS, synovial cells were isolated by enzymatic digestion with trypsin and collagenase from synovium of 3 individuals with osteoarthritis, obtained at the time of joint replacement (Fig. C-IA). Normal chondrocytes were isolated by enzymatic digestion from articular cartilage obtained at autopsy from 3 individuals involved in fatal automobile accidents 22 (Fig. C- 1B). Cells were suspended in phosphate buffer and stained with TS, Papanicolaou, and Sudan black B stains. 23 Also, in the last 26 samples of SF studied, dried smears were stained with Papanicolaou’s stain and compared with fresh preparations stained with TS.
Results Staining of Cells Supravital staining with TS provided comparable and often better cell morphological details than Wrightstained dry SF preparations, as can be appreciated in Figs. C- 1-C-3. Staining of those 40 noninflammatory fluids with cell counts less than 500 cells/mm was satisfactory and a differential in 50-100 cells was performed in most of these. Polymorphonuclear cells were easily recognized under immersion oil. Nuclei appeared as dense and deeply violet stained polyglobular masses, while the cytoplasm showed a pinkish and granular appearance (Figs. C-2 and (2-3). Polymorphonuclear cells with dense and pyknotic nuclei commonly seen in the synovial fluid of patients with rheumatoid arthritis were easily identified with this stain (Fig. C-2). Lymphocytes showed dense nuclei surrounded by scarce pinkish cytoplasm. Monocytes revealed large, round, or lobulated dense nuclei surrounded by a more abundant purplish cytoplasm, and variable numbers of purplish inclusions (Fig. C-4A ). Large macrophages with features of type A synovial cells showed large or oval off-center nuclei with prominent nucleoli. 23-25 They had abundant pinkish cytoplasm, and many of them showed numerous violet-stained intracytoplasmic round inclusions (Fig. C-2). These cells were most commonly seen in “noninflammatory” fluids and were identical to those obtained from osteoarthritic synovium (Fig. C- 1A ). Macrophages with features of type B synovial lining cells were of smaller size, with more irregular or lobulated nuclei and fewer violet intracytoplasmic inclusions, 23,24 and were difficult to differentiate from monocytes (Fig. C-4B). Necrotic cells showed only a weak and diffuse intracellular staining without contrast between the nucleus and cytoplasm components. For example, in 3 patients with sickle cell disease and acute synovitis, SF was “nonin-
SYNOVIAL FLUID STAINING WITH TESTSIMPLETS
flammatory” with cell counts below 500, but most of the cells were necrotic mononuclear cells (Fig. C-5). The single patient with leukemia and acute arthritis studied in this series also had a noninflammatory fluid, and no leukemic cells were observed in his SF. Intracytoplasmic inclusions observed on the fresh unstained preparations were easily recognized using TS staining. For example, intracytoplasmic inclusions commonly seen in SF of patients with rheumatoid arthritis and other inflammatory arthropathies appeared either as unstained or purple-stained small inclusions of even sizes. Those due to fat droplets were not stained, adopted a pale yellow color, and were usually larger than those in inclusions seen in rheumatoid arthritis. In the SF of 5 patients with traumatic synovitis and subchondral fractures, and in 1 patient with osteoarthritis, large fat droplets were seen inside mononuclear cells (Fig. C-6). Suspected bacteria that stained purple occurred in pairs, chains, or clusters, and were seen in 4 patients with septic arthritis (Fig. C-7). The presence of bacteria in such SFs was subsequently confirmed by simultaneous Gram staining or acridine orange staining, and by isolation of Staphylococcus aureus. Leukophagocytosis (“Reiter cells”) was easily identified in 3 of the 6 patients with Reiter’s syndrome, 2 with psoriasis, 2 with gonococcal arthritis, 1 patient with staphylococcal arthritis, and 2 others with acute gout. The outlines of phagocytized polymorphonuclear cells and their nuclei were less distorted than similar cells seen in Wright-stained preparations (Fig. C-8). Lupus erythematosus cells were observed in 2 of the 3 mildly inflammatory synovial effusions from patients with systemic lupus erythematosus (Fig. C-9). The nuclei of the polymorphonuclear cells and the characteristic intracytoplasmic amorphous masses were easily recognized because they were sharply outlined and stained. Cell morphology was well preserved until 2 hr after staining if slides were kept at room temperature and until 24 hr if kept in a Petri dish at 4°C in a humid environment.
Crystal IdentiJcation With TS monosodium urate and calcium pyrophosphate, dihydrate crystals were easily recognized (Figs. C-10 and C-1 1). Comparative crystal counts of unstained fresh preparations and those stained with TS of identical SFs showed a decrease of about 10% in the total crystal count in those fluids from patients with gout and pseudogout using TS. This difference probably is due to the fact that smaller crystals may be difficult to visualize inside stained cells. Monosodium urate crystals adopted a pale-brown color (Fig. C-1 l), and calcium pyrophosphate dihydrate showed a similar brownish color or were not stained all (Fig. C-10). Both types of crystals preserved their characteristic birefringence after TS staining. Hydroxyapatite crystals did not take the TS stain in calcific exudates from
2 patients with CREST syndrome and in 2 osteoarthritic synovial fluids that tested positive with alizarin red S staining. Other crystals, such as synthetic corticosteroids, cholesterol, and other lipids, were also recognized in TSstained wet preparations (Table I).
Synovial Villi, Cartilage Fragments, and Fibrin Synovial villi were observed in 2 patients with rheumatoid arthritis and in 4 others with osteoarthritis (Fig. C-12A and C-12B). In one villus a synovial vessel with endothelial and red cells was clearly seen (Fig. C-12s). Deeply stained cartilage fragments of variable sizes and shapes were observed in 13 of 16 patients with advanced osteoarthritis and in 2 with severe rheumatoid arthritis (Fig. C-13). Cells were not observed inside most of these fragments (Fig. C-l3A); healthy chondrocytes or their lacunae were observed in 2 samples (Fig. C-13B). Fibrin was not stained with TS and appeared as hairy threads grouped in random arrangements (Fig. C-4A).
Discussion Testsimplet is a simple, rapid, and reproducible method for supravital staining that has previously been used in blood, exfoliative cytology, spinal, and seminal fluid, 9-’’ and that we now demonstrate to be of use in synovial fluid analysis. l 2 The morphology of synovial fluid cells was well preserved, and polymorphonuclear cells can be easily differentiated from mononuclear cells (Figs. C- 1-C-4). Using TS, we identified different intracytoplasmic particles, such as dense inclusions associated with rheumatoid arthritis, fat droplets, and crystals. Monosodium urate crystals and calcium pyrophosphate dihydrate crystals were not dissolved, as with other staining techniques. In addition, these crystals retained their typical birefringence and were readily identified under polarized light. Also, these crystals adopted a light-brown color, which aided in their recognition inside cells (Figs. C-10 and C-1 1). This contrasted with hydroxyapatite crystals, which did not take the stain and were not clearly separated from other intracellular inclusions or extracellular detritus. Mononuclear cells, such as monocytes, lymphocytes, and synovial cells, were also clearly differentiated. Type A synovial cells were easily recognized, but monocytes and type B synovial cells disclosed a very similar appearance. 23-25 To distinguish monocytes from type B synovial cells, dry-fixed stained preparations, such as Sudan black staining, 3,23 or surface marker studies using monoclonal antibodies would be necessary.2612’TS is also useful in performing detailed celluiar identification of “noninflammatory” SF with cell counts less than 2,00O/mm 28 In some clinical laboratories, SF differential counts are omitted when total cell counts are less than 1,000/ mm 3,28,29 In a recent publication regarding TS staining in “noninflammatory” fluids with less than 2,000 cells X
’.
’.
Diugno.ytic Cytopathology, Vol 8, No 2
149
Fig. C-1A
Fig. C-1B
Fig. C-3
Fig. C-5
Fig. C-2
Fig. C-4A
Fig. C-6
Fig. C-4B
Fig. C-7
Figs. C-1-C-7. Fig. C-1. (A) Characteristic synovial lining cell isolated from digested synovium of a patient with osteoarthritis (cresyl violet and methylene blue staining x 1,000). (B) Chondrocytes obtained by digestion of normal articular cartilage (cresyl violet and methylene blue staining X 1000). Fig. C-2. Synovial lining cell surrounded by polymorphonuclear cells in synovial fluid of a patient with rheumatoid arthritis. Note the prominent nucleoli and violet-stained intracytoplasmic inclusions (cresyl violet and methylene blue staining X 1000). Fig. C-3. Healthy polymorphonuclear cells and others with pyknotic nuclei in SF of a patient with rheumatoid arthritis (cresyl violet and methylene blue X400). Fig. C-4. (A) Lymphocytes and fibrin strands in the synovial fluid of a patient with seronegative spondyloarthropathy (cresyl violet and methylene blue X 1000). IB) Synovial cell surrounded by mononuclear cells, which might correspond to monocytes (cresyl violet and methylene blue X 1,000). Fig. C-5. Necrotic cells observed in a patient with sickle cell crisis and acute synovitis associated with “noninflammatory” synovial fluid (cresyl violet and methylene blue X 1,000). Fig. C-6. Fat droplets and lipocytes in SF of a patient with traumatic arthritis and subchondral fracture. Note the intracellular . C-7. Necrotic polymorphonuclear cell observed in the SF of a patient fat droplets with displaced nuclei (cresyl violet and methylene blue ~ 4 0 0 )Fig. with staphylococcal septic arthritis. Note the destroyed polymorphonuclear cells and dark cocci, confirmed as intracellular bacteria by Gram stain (cresyl violet and methylene blue x 1,000).
Fig. C-8
Fig. C-9
Fig. C-10
Fig. C-11
Fig. C-12A
Fig. c-1211
Fig. C-13A
Fig. C-13B
Figs. C-8-C-13. Fig. (2-8. Leukophagocytic cell (Reiter cell), surrounded by a lymphocyte and polymorphonuclear cells in the synovial fluid of a patient with Reiter’s syndrome (cresyl violet and methylene blue X 1,000).Fig. C-9. Characteristic LE cell observed in the SF of a patient with systemic lupus erythematosus (cresyl violet and methylene blue x 1,000). Fig. C-10. Pleomorphic calcium pyrophosphate dihydrate crystals in SF of a patient with pseudogout (cresyl violet and methylene blue X 1,000). Fig. C-11. Polymorphonuclear cells and a large intracellular needle of monosodium urate from a patient with gout (ordinary light X 1,000).Fig. C-12. Synovial villi observed in synovial fluid. (A) Synovial villus showing only synovial lining (B) Synovial villus with a normal vessel, showing synovial endothelial cells (cresyl violet and methylene cells (cresyl violet and methylene blue ~400). blue X 1000).Fig. C-13. (A) Deeply stained, necrotic articular cartilage fragments observed in the synovial fluid of a patient with osteoarthritis (cresyl violet and methylene blue ~400).(B) Cartilage fragment with healthy-looking chondrocytes found in the SF of a patient with osteoarthritis (cresyl violet and methylene blue ~ 4 0 0 ) .
REGINATO ET AL.
mm3, Louthrenco et al. observed better morphology of mononuclear cells and polymorphonuclear cells than in specimens obtained with simultaneous Wright-stained In this study the presence of total cell counts higher than 1,00O/mm and a high percentage of polymorphonuclear cells correlated with the presence of monosodium urate, calcium pyrophosphate dihydrate, or apatite crysta1s. In Our series, patients with disease and acute svnovitis Dresented with a “noninflammatory” SF (cell count less than 1,000 cells/mm3), but most SF cells were necrotic mononuclear cells ( ~ i~ ~- 5. ) . TS also stained chondrocytes in cartilage fragments that were observed floating free in the SF, as has been described by others in dry preparations stained with Papanicolaou ~ t a i n . ~ , ~ ,LE ~ ’ ,cells ~ ’ and Reiter cells were also recognized, and their morphology was less distorted than those observed in simultaneous Wright-stained preparations. The TS has several shortcomings. The fresh preparation can be kept for very limited periods of time at room temperature and for about 24 hr in a cold, humid environment. 9-12 Also, the slides are not commercially available in the U.S.A., yet are available in Germany, other parts of Europe, and South America. The cost for each TS slide is about $1.00 (U.S. currency). In summary, we have performed cytological studies with TS on 159 SFs obtained from patients with a wide variety of rheumatic diseases. TS provided adequate contrast and good morphological details of both normal and abnormal synovial fluid cells. Most crystals were well preserved; monosodium urate and calcium pyrophosphate dihvdrate were identified in the fresh TS-stained Drepara1 r tions under compensated polarized light. TS is a rapid, staining Of and reproducib1e method Of synovial fluid suitable for routine and office synovial fluid analysis.
*’
Acknowledgment The authors wish to Barnes for secretarial assistance.
Mrs. Dorothy Ann
References
8. Collins DH. The pathology of synovial effusions. J Pathol 1936;42: 113-8. 9. Bostjanicic W, Braun JS, Reheis G, et al. Barbbeschichete Objecktrager: Eine einfache farbemethode fur das differentialblutbild. Dtsch Med Wschr 1977;102:1175-7. 10. Kleine TO, Flury R, Tritschler W. Liquorzytologie mit vorgefarbten objekitragen. Dtsch Med Wschr 1977;102:1216-21. 1 I. Reimer EE, Stoiber TH. Zytodiagnostik mit vorgefarbten objekitragan. Laboratoriumsmedizin 1977;1:111-5. 12. Reginat0 AJ. Maldonado I. Falasca GF. Reginato AM. OConnor C f . Suuravital staining - w i h Testsimplet in-synovial fluid. ILAR International Meeting. Rio de Janeiro, Brazil, 1989. 13. Lavole PE, Birnbaum NS, Barbour AS, et al. Uncoiled acridine orange positive spirochetal form in rheumatoid pericardial fluid. Arthritis Rheum 1989;32:S45. 14. 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. 15. Wallace SL, Robinson H, Masi AT, Decker JL, McCarty DJ, Yu T. Preliminary criteria for the classification of the acute arthritis of primary gout. Arthritis Rheum 1977;20:895-900. 16. Reginato AJ, Schumacher HR. Crystal-associated arthropathies. Clin Geriatr Med 1988;4:295-322. 17. Gregg JR, Nixon JE, DiStefano V. Neutral fat globules in lraumatized knees. Clin Orthop 1978;132:219-24. 18. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271-7. 19. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24. 20. Wilkens RF, Arnett FC, Bitter T, et al. Reiter’s Syndrome. Evaluation of preliminary criteria for definite disease. Bull Rheum Dis 1982;32:31-42. 21. Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039-49. 22. Tiku MI, Liu S , Weaver CW, Teodorescu M, Skosey JL. Class-I1 histocompatibility antigen mediated immunologic function of norma1 articular chondrocytes. J Immunol 1985;135:2923-28. 23. Traycoff RB, Pascual E, Schumacher HR. Mononuclear cells in human synovial fluid. Identification of lymphoblasts in rheumatoid arthritis. Arthritis Rheum 1976;19:743. 24. Kinsella TD, Baum J, Ziff M. Studies of isolated synovial lining cells of rheumatoid and non-rheumatoid synovial membranes. Arthritis Rheum 1970;13:734. 25. Henderson B, Pettipher ER. The synovial lining cell: Biology and pathobiology. Semin Arthritis Rheum 1985;15:1-20. 26. Lasky HP, Bauer K, Rope RM. Increased helper inducer and decreased suppressor inducer uhenotvues _ . in the rheumatoid ioint. Arth Rheum 19’88;31:52-9. 27, Morimotto CH, Remain pL, Fox DA, et al, Abnormalities in CD4+ T- lymphocyte subsets in inflammatory rheumatic diseases. Am J Med 1988;84:817-25. 28, Louthrenco w, Sieck M, Clayburne G, Schumachor HR, Cell differentials in non-inflammatory fluid with supravital staining. J Rheumatol 1991;18:409-13 (abstract). 29. Eisenberg JM, Schumacher HR, Davidson PK, Kaufman L. Usefulness of synovial fluid analysis in the evaluation ofjoint effusions: Use of threshold analysis and likelihood ratios to assess a diagnostic test. Arch Intern Med 1984;144:715-719. 30. Broderick P, Coruese N, Pienik M, Qitte RF, Martorenzi AL. Exfoliative cytology interpretation of synovial fluid in joint disease. J Bone Joint Surg 1976;58:396-9. 31. Meisels A, Berebichez M. Exfoliative cytology in orthopedics. Can Med Assoc J 1961;84:957-9. ~
~
1. Hollander JL, Reginato AJ, Torralba T. Examination of synovial fluid as a diagnostic aid in arthritis. Med Clin North Am 1966;50: I28 1-93. 2. Villanueva TG, Schumacher HR. Cytologic examination of synovial fluid. Diagn Cytopathol 1987;4:141-7. 3. Schumacher HR, Reginato AJ. Atlas of synovial fluid analysis and crystal identification. Philadelphia: Lea & Febiger, 1991. 4. Cohen AS, Goldenberg D. Synovial fluid, laboratory diagnostic procedures in the rheumatic diseases. 3rd ed. Orlando: Grune & Stratton, 1985. 5. Naiba ZM. Cytology of synovial fluid. Acta CytOl 1973;17:299-309. 6. Coogeshal HO, Warren CF, Bauer W. The cytology of normal human synovial fluid. Anat Rec 1940;77:129-36. 7. Ropes MW, Bauer W. Synovial fluid changes in joint disease. Cambridge, MA: Harvard University Press, 1953.
152
Diagnostic Cytoputhology, Vol 8, No 2
We explored the use of Testsimplet (TS) in synovial fluid (SF) analysis. TS is a glass slide coated with a dry mixture of methylene blue and cresyl violet, which in contact with one drop of SF provides a stained fresh preparation. We applied the TS to the study of 159 SFs ofpatients with different rheumatic diseases. I n those SFs of patients with crystal-associated diseases, the crystal search was performed both on unstained preparations and with TS. TS was as good as the Wright’s and Papanicolaou stain in characterizing SF cells, lupus erythematosus cells, and detection of occasional bacteria. TSallowed a better visualization of Reiter’s cells, cartilage fragments, synovial villi, fat droplets, and fibrin. Crystals were identified in every TS of those patients with crystalassociated diseases. TS is a rapid and reproducible method of S F supravital staining. Crystals are well preserved for simultaneous examination with compensated polarized light. Diagn Cytopatho1 1992;8:147-152. 0 1992 WiIey-Liss, Inc. Key Words: Synovial fluid cytology; Cresyl violet; Methylene blue; Synovial fluid analysis; Synovial fluid fresh preparations
After the history and physical examination, synovial fluid (SF) analysis ranks as the most simple diagnostic procedure for the evaluation ofjoint diseases. In many patients with septic arthritis and crystal-associated arthritis, a few drops of SF often yield a rapid and definitive diagnosis. In other rheumatic diseases, SF analysis provides useful information about the intensity of the inflammatory process and helps to narrow the differential diagnosis to conditions associated with either “inflammatory” or “noninflammatory” fluids. Differentiation of inflammatory cells is usually performed in air-fixed Wright-stained preparations and ocReceived February 11, 1991. Accepted September 25, 1991. From the University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Camden, NJ; University of Pennsylvania, School of Medicine, Philadelphia, PA; Department of Biochemistry, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA. Address reprint requests to Antonio J. Reginato, M.D., Head, Division of Rheumatology, Department of Medicine, Three Cooper Plaza, Suite 220, Haddon Avenue, Camden, NJ 08103. 0 1992 WILEY-LISS, INC
casionally with the Papanicolaou stain. Wright stains of “noninflammatory” SF preparations are difficult to read when few cells are present or when the fluid is very viscous. Early reports suggested that supravital staining of fresh synovial fluid preparations using neutral red and Janus green were helpful in differentiating mononuclear cells in noninflammatory fluids. 6,7 We and others have been unable to corroborate these findings. 3,8 Supravital staining with cresyl violet and methylene blue is performed easily with precoated, dry glass slides containing these dyes (Testsimplet TM, Boehringer, Mannhein, Germany). 9-11 We report our experience with Testsimplet (TS) in the synovial fluid analysis of 159 patients with various rheumatic diseases. TS provided a rapid and reproducible method of SF supravital staining and allowed better visualization of Reiter cells, synovial cells, cartilage fragments, fat droplets, fibrin, monosodium urate, and calcium pyrophosphate dihydrate crystals than fixed, dried Wright-stained preparations. Calcium pyrophosphate and monosodium urate crystals were well preserved for simultaneous examination under compensated polarized light.
Materials and Methods TS was performed by placing a drop of SF on the center of the commercially available glass slide already containing the dried stains. The slide was covered with a glass cover slip and examined after 10-15 min9-l2 The differential cell count and the search for crystals under compensated polarized light with TS staining were compared with simultaneous same-patient control samples obtained with dry-fixed Wright-stained preparations and unstained synovial fluid fresh preparations, respectively. Routine SF analysis included a fresh preparation observed under regular and compensated polarized light, total cell count, and differential cell count in a dried, fixed Wright-stained preparation. These studies were performed on SF from 159 patients between 1988 and 1990 Diagnostic Cytopathology, Vol 8, No 2
147
REGINATO ET AL.
(Table I). Gram, acridine orange, 3, and alizarin red stainingsI4 were done when septic arthritis or apatite deposiiion disease were suspected. Culture for aerobic and nonaerobic bacteria of synovial fluid was performed in 30 inflammatory fluid samples in which septic arthritis was suspected. The diagnosis of septic arthritis was made by the recovery of Staphylococcus aureus from 7 and Neisseria gonorrhea from 3 SF samples. In those 5 1 patients with crystal-associated arthritis, the diagnosis was made by identification of monosodium urate crystals in 24, calcium pyrophosphate crystals in 19, apatite in 6, cholesterol in I , and lipid liquid crystals in 1 patient by polarized light microscopy. I5,l6 Fifteen of the 19 patients with pseudogout also had radiographic evidence of chondrocalcinosis of their knees, while the other 2 showed signs of severe degenerative joint disease. l 6 In the 6 patients with apatite deposition disease, periarticular calcific deposits were seen in bursae, tendons, or synovium, and their SF showed alizarin red stained clumps. In 2 of these patients the presence of apatite was confirmed by transmission electron microscopy and x-ray diffraction analysis. The diagnosis of post-traumatic arthritis or bursitis was made in 10 patients by a history of preceding trauma to the joint and the finding of a bloody, non-inflammatory fluid. In 5 of these patients subchondral fractures were observed on radiographs of affected joints. The diagnosis of Lyme arthritis was supported by elevated IgG and IgM antibodies by an ELISA assay and exclusion of other causes of inflammatory monoarthritis. The diagnoses of systemic lupus erythematosus, rheumatoid arthritis, l 9 Reiter’s Table I. Synovial Fluid Studied with Testsimplet Staining in 159 Patients with Arthritis Type of arthritis Septic arthritis or bursitis Lyme disease Gout Pseudogout Apatite deposition disease Cholesterol and lipid liquid crystals Rheumatoid arthritis Osteoarthritis Rtiter’s syndrome Psoriatic arthritis Sickle cell crisis Traumatic arthritis or bursitis Miscellaneous Sympathetic effusion due to cellulitis Vasculitis 13acterial endocarditis Corticosteroid pulse therapy Possible viral arthritis Monoarthritis of unknown cause Sickle cell disease Total
148
Number of patien ts 10 1 24 19 6 2 22 21 8 5
3 10
11 1
1 1 1 1
3 3
159
Diagnostic Cytopathology, Vol 8. No 2
syndrome, 2o and osteoarthritis 21 were established in accordance with the American College of Rheumatology recommendations. The diagnosis of psoriatic arthritis was established by the presence of characteristic scaly rash, nail pitting, and asymmetric inflammatory arthritis. In order to better recognize the morphologic aspects of synovial cells in TS, synovial cells were isolated by enzymatic digestion with trypsin and collagenase from synovium of 3 individuals with osteoarthritis, obtained at the time of joint replacement (Fig. C-IA). Normal chondrocytes were isolated by enzymatic digestion from articular cartilage obtained at autopsy from 3 individuals involved in fatal automobile accidents 22 (Fig. C- 1B). Cells were suspended in phosphate buffer and stained with TS, Papanicolaou, and Sudan black B stains. 23 Also, in the last 26 samples of SF studied, dried smears were stained with Papanicolaou’s stain and compared with fresh preparations stained with TS.
Results Staining of Cells Supravital staining with TS provided comparable and often better cell morphological details than Wrightstained dry SF preparations, as can be appreciated in Figs. C- 1-C-3. Staining of those 40 noninflammatory fluids with cell counts less than 500 cells/mm was satisfactory and a differential in 50-100 cells was performed in most of these. Polymorphonuclear cells were easily recognized under immersion oil. Nuclei appeared as dense and deeply violet stained polyglobular masses, while the cytoplasm showed a pinkish and granular appearance (Figs. C-2 and (2-3). Polymorphonuclear cells with dense and pyknotic nuclei commonly seen in the synovial fluid of patients with rheumatoid arthritis were easily identified with this stain (Fig. C-2). Lymphocytes showed dense nuclei surrounded by scarce pinkish cytoplasm. Monocytes revealed large, round, or lobulated dense nuclei surrounded by a more abundant purplish cytoplasm, and variable numbers of purplish inclusions (Fig. C-4A ). Large macrophages with features of type A synovial cells showed large or oval off-center nuclei with prominent nucleoli. 23-25 They had abundant pinkish cytoplasm, and many of them showed numerous violet-stained intracytoplasmic round inclusions (Fig. C-2). These cells were most commonly seen in “noninflammatory” fluids and were identical to those obtained from osteoarthritic synovium (Fig. C- 1A ). Macrophages with features of type B synovial lining cells were of smaller size, with more irregular or lobulated nuclei and fewer violet intracytoplasmic inclusions, 23,24 and were difficult to differentiate from monocytes (Fig. C-4B). Necrotic cells showed only a weak and diffuse intracellular staining without contrast between the nucleus and cytoplasm components. For example, in 3 patients with sickle cell disease and acute synovitis, SF was “nonin-
SYNOVIAL FLUID STAINING WITH TESTSIMPLETS
flammatory” with cell counts below 500, but most of the cells were necrotic mononuclear cells (Fig. C-5). The single patient with leukemia and acute arthritis studied in this series also had a noninflammatory fluid, and no leukemic cells were observed in his SF. Intracytoplasmic inclusions observed on the fresh unstained preparations were easily recognized using TS staining. For example, intracytoplasmic inclusions commonly seen in SF of patients with rheumatoid arthritis and other inflammatory arthropathies appeared either as unstained or purple-stained small inclusions of even sizes. Those due to fat droplets were not stained, adopted a pale yellow color, and were usually larger than those in inclusions seen in rheumatoid arthritis. In the SF of 5 patients with traumatic synovitis and subchondral fractures, and in 1 patient with osteoarthritis, large fat droplets were seen inside mononuclear cells (Fig. C-6). Suspected bacteria that stained purple occurred in pairs, chains, or clusters, and were seen in 4 patients with septic arthritis (Fig. C-7). The presence of bacteria in such SFs was subsequently confirmed by simultaneous Gram staining or acridine orange staining, and by isolation of Staphylococcus aureus. Leukophagocytosis (“Reiter cells”) was easily identified in 3 of the 6 patients with Reiter’s syndrome, 2 with psoriasis, 2 with gonococcal arthritis, 1 patient with staphylococcal arthritis, and 2 others with acute gout. The outlines of phagocytized polymorphonuclear cells and their nuclei were less distorted than similar cells seen in Wright-stained preparations (Fig. C-8). Lupus erythematosus cells were observed in 2 of the 3 mildly inflammatory synovial effusions from patients with systemic lupus erythematosus (Fig. C-9). The nuclei of the polymorphonuclear cells and the characteristic intracytoplasmic amorphous masses were easily recognized because they were sharply outlined and stained. Cell morphology was well preserved until 2 hr after staining if slides were kept at room temperature and until 24 hr if kept in a Petri dish at 4°C in a humid environment.
Crystal IdentiJcation With TS monosodium urate and calcium pyrophosphate, dihydrate crystals were easily recognized (Figs. C-10 and C-1 1). Comparative crystal counts of unstained fresh preparations and those stained with TS of identical SFs showed a decrease of about 10% in the total crystal count in those fluids from patients with gout and pseudogout using TS. This difference probably is due to the fact that smaller crystals may be difficult to visualize inside stained cells. Monosodium urate crystals adopted a pale-brown color (Fig. C-1 l), and calcium pyrophosphate dihydrate showed a similar brownish color or were not stained all (Fig. C-10). Both types of crystals preserved their characteristic birefringence after TS staining. Hydroxyapatite crystals did not take the TS stain in calcific exudates from
2 patients with CREST syndrome and in 2 osteoarthritic synovial fluids that tested positive with alizarin red S staining. Other crystals, such as synthetic corticosteroids, cholesterol, and other lipids, were also recognized in TSstained wet preparations (Table I).
Synovial Villi, Cartilage Fragments, and Fibrin Synovial villi were observed in 2 patients with rheumatoid arthritis and in 4 others with osteoarthritis (Fig. C-12A and C-12B). In one villus a synovial vessel with endothelial and red cells was clearly seen (Fig. C-12s). Deeply stained cartilage fragments of variable sizes and shapes were observed in 13 of 16 patients with advanced osteoarthritis and in 2 with severe rheumatoid arthritis (Fig. C-13). Cells were not observed inside most of these fragments (Fig. C-l3A); healthy chondrocytes or their lacunae were observed in 2 samples (Fig. C-13B). Fibrin was not stained with TS and appeared as hairy threads grouped in random arrangements (Fig. C-4A).
Discussion Testsimplet is a simple, rapid, and reproducible method for supravital staining that has previously been used in blood, exfoliative cytology, spinal, and seminal fluid, 9-’’ and that we now demonstrate to be of use in synovial fluid analysis. l 2 The morphology of synovial fluid cells was well preserved, and polymorphonuclear cells can be easily differentiated from mononuclear cells (Figs. C- 1-C-4). Using TS, we identified different intracytoplasmic particles, such as dense inclusions associated with rheumatoid arthritis, fat droplets, and crystals. Monosodium urate crystals and calcium pyrophosphate dihydrate crystals were not dissolved, as with other staining techniques. In addition, these crystals retained their typical birefringence and were readily identified under polarized light. Also, these crystals adopted a light-brown color, which aided in their recognition inside cells (Figs. C-10 and C-1 1). This contrasted with hydroxyapatite crystals, which did not take the stain and were not clearly separated from other intracellular inclusions or extracellular detritus. Mononuclear cells, such as monocytes, lymphocytes, and synovial cells, were also clearly differentiated. Type A synovial cells were easily recognized, but monocytes and type B synovial cells disclosed a very similar appearance. 23-25 To distinguish monocytes from type B synovial cells, dry-fixed stained preparations, such as Sudan black staining, 3,23 or surface marker studies using monoclonal antibodies would be necessary.2612’TS is also useful in performing detailed celluiar identification of “noninflammatory” SF with cell counts less than 2,00O/mm 28 In some clinical laboratories, SF differential counts are omitted when total cell counts are less than 1,000/ mm 3,28,29 In a recent publication regarding TS staining in “noninflammatory” fluids with less than 2,000 cells X
’.
’.
Diugno.ytic Cytopathology, Vol 8, No 2
149
Fig. C-1A
Fig. C-1B
Fig. C-3
Fig. C-5
Fig. C-2
Fig. C-4A
Fig. C-6
Fig. C-4B
Fig. C-7
Figs. C-1-C-7. Fig. C-1. (A) Characteristic synovial lining cell isolated from digested synovium of a patient with osteoarthritis (cresyl violet and methylene blue staining x 1,000). (B) Chondrocytes obtained by digestion of normal articular cartilage (cresyl violet and methylene blue staining X 1000). Fig. C-2. Synovial lining cell surrounded by polymorphonuclear cells in synovial fluid of a patient with rheumatoid arthritis. Note the prominent nucleoli and violet-stained intracytoplasmic inclusions (cresyl violet and methylene blue staining X 1000). Fig. C-3. Healthy polymorphonuclear cells and others with pyknotic nuclei in SF of a patient with rheumatoid arthritis (cresyl violet and methylene blue X400). Fig. C-4. (A) Lymphocytes and fibrin strands in the synovial fluid of a patient with seronegative spondyloarthropathy (cresyl violet and methylene blue X 1000). IB) Synovial cell surrounded by mononuclear cells, which might correspond to monocytes (cresyl violet and methylene blue X 1,000). Fig. C-5. Necrotic cells observed in a patient with sickle cell crisis and acute synovitis associated with “noninflammatory” synovial fluid (cresyl violet and methylene blue X 1,000). Fig. C-6. Fat droplets and lipocytes in SF of a patient with traumatic arthritis and subchondral fracture. Note the intracellular . C-7. Necrotic polymorphonuclear cell observed in the SF of a patient fat droplets with displaced nuclei (cresyl violet and methylene blue ~ 4 0 0 )Fig. with staphylococcal septic arthritis. Note the destroyed polymorphonuclear cells and dark cocci, confirmed as intracellular bacteria by Gram stain (cresyl violet and methylene blue x 1,000).
Fig. C-8
Fig. C-9
Fig. C-10
Fig. C-11
Fig. C-12A
Fig. c-1211
Fig. C-13A
Fig. C-13B
Figs. C-8-C-13. Fig. (2-8. Leukophagocytic cell (Reiter cell), surrounded by a lymphocyte and polymorphonuclear cells in the synovial fluid of a patient with Reiter’s syndrome (cresyl violet and methylene blue X 1,000).Fig. C-9. Characteristic LE cell observed in the SF of a patient with systemic lupus erythematosus (cresyl violet and methylene blue x 1,000). Fig. C-10. Pleomorphic calcium pyrophosphate dihydrate crystals in SF of a patient with pseudogout (cresyl violet and methylene blue X 1,000). Fig. C-11. Polymorphonuclear cells and a large intracellular needle of monosodium urate from a patient with gout (ordinary light X 1,000).Fig. C-12. Synovial villi observed in synovial fluid. (A) Synovial villus showing only synovial lining (B) Synovial villus with a normal vessel, showing synovial endothelial cells (cresyl violet and methylene cells (cresyl violet and methylene blue ~400). blue X 1000).Fig. C-13. (A) Deeply stained, necrotic articular cartilage fragments observed in the synovial fluid of a patient with osteoarthritis (cresyl violet and methylene blue ~400).(B) Cartilage fragment with healthy-looking chondrocytes found in the SF of a patient with osteoarthritis (cresyl violet and methylene blue ~ 4 0 0 ) .
REGINATO ET AL.
mm3, Louthrenco et al. observed better morphology of mononuclear cells and polymorphonuclear cells than in specimens obtained with simultaneous Wright-stained In this study the presence of total cell counts higher than 1,00O/mm and a high percentage of polymorphonuclear cells correlated with the presence of monosodium urate, calcium pyrophosphate dihydrate, or apatite crysta1s. In Our series, patients with disease and acute svnovitis Dresented with a “noninflammatory” SF (cell count less than 1,000 cells/mm3), but most SF cells were necrotic mononuclear cells ( ~ i~ ~- 5. ) . TS also stained chondrocytes in cartilage fragments that were observed floating free in the SF, as has been described by others in dry preparations stained with Papanicolaou ~ t a i n . ~ , ~ ,LE ~ ’ ,cells ~ ’ and Reiter cells were also recognized, and their morphology was less distorted than those observed in simultaneous Wright-stained preparations. The TS has several shortcomings. The fresh preparation can be kept for very limited periods of time at room temperature and for about 24 hr in a cold, humid environment. 9-12 Also, the slides are not commercially available in the U.S.A., yet are available in Germany, other parts of Europe, and South America. The cost for each TS slide is about $1.00 (U.S. currency). In summary, we have performed cytological studies with TS on 159 SFs obtained from patients with a wide variety of rheumatic diseases. TS provided adequate contrast and good morphological details of both normal and abnormal synovial fluid cells. Most crystals were well preserved; monosodium urate and calcium pyrophosphate dihvdrate were identified in the fresh TS-stained Drepara1 r tions under compensated polarized light. TS is a rapid, staining Of and reproducib1e method Of synovial fluid suitable for routine and office synovial fluid analysis.
*’
Acknowledgment The authors wish to Barnes for secretarial assistance.
Mrs. Dorothy Ann
References
8. Collins DH. The pathology of synovial effusions. J Pathol 1936;42: 113-8. 9. Bostjanicic W, Braun JS, Reheis G, et al. Barbbeschichete Objecktrager: Eine einfache farbemethode fur das differentialblutbild. Dtsch Med Wschr 1977;102:1175-7. 10. Kleine TO, Flury R, Tritschler W. Liquorzytologie mit vorgefarbten objekitragen. Dtsch Med Wschr 1977;102:1216-21. 1 I. Reimer EE, Stoiber TH. Zytodiagnostik mit vorgefarbten objekitragan. Laboratoriumsmedizin 1977;1:111-5. 12. Reginat0 AJ. Maldonado I. Falasca GF. Reginato AM. OConnor C f . Suuravital staining - w i h Testsimplet in-synovial fluid. ILAR International Meeting. Rio de Janeiro, Brazil, 1989. 13. Lavole PE, Birnbaum NS, Barbour AS, et al. Uncoiled acridine orange positive spirochetal form in rheumatoid pericardial fluid. Arthritis Rheum 1989;32:S45. 14. 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. 15. Wallace SL, Robinson H, Masi AT, Decker JL, McCarty DJ, Yu T. Preliminary criteria for the classification of the acute arthritis of primary gout. Arthritis Rheum 1977;20:895-900. 16. Reginato AJ, Schumacher HR. Crystal-associated arthropathies. Clin Geriatr Med 1988;4:295-322. 17. Gregg JR, Nixon JE, DiStefano V. Neutral fat globules in lraumatized knees. Clin Orthop 1978;132:219-24. 18. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271-7. 19. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24. 20. Wilkens RF, Arnett FC, Bitter T, et al. Reiter’s Syndrome. Evaluation of preliminary criteria for definite disease. Bull Rheum Dis 1982;32:31-42. 21. Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039-49. 22. Tiku MI, Liu S , Weaver CW, Teodorescu M, Skosey JL. Class-I1 histocompatibility antigen mediated immunologic function of norma1 articular chondrocytes. J Immunol 1985;135:2923-28. 23. Traycoff RB, Pascual E, Schumacher HR. Mononuclear cells in human synovial fluid. Identification of lymphoblasts in rheumatoid arthritis. Arthritis Rheum 1976;19:743. 24. Kinsella TD, Baum J, Ziff M. Studies of isolated synovial lining cells of rheumatoid and non-rheumatoid synovial membranes. Arthritis Rheum 1970;13:734. 25. Henderson B, Pettipher ER. The synovial lining cell: Biology and pathobiology. Semin Arthritis Rheum 1985;15:1-20. 26. Lasky HP, Bauer K, Rope RM. Increased helper inducer and decreased suppressor inducer uhenotvues _ . in the rheumatoid ioint. Arth Rheum 19’88;31:52-9. 27, Morimotto CH, Remain pL, Fox DA, et al, Abnormalities in CD4+ T- lymphocyte subsets in inflammatory rheumatic diseases. Am J Med 1988;84:817-25. 28, Louthrenco w, Sieck M, Clayburne G, Schumachor HR, Cell differentials in non-inflammatory fluid with supravital staining. J Rheumatol 1991;18:409-13 (abstract). 29. Eisenberg JM, Schumacher HR, Davidson PK, Kaufman L. Usefulness of synovial fluid analysis in the evaluation ofjoint effusions: Use of threshold analysis and likelihood ratios to assess a diagnostic test. Arch Intern Med 1984;144:715-719. 30. Broderick P, Coruese N, Pienik M, Qitte RF, Martorenzi AL. Exfoliative cytology interpretation of synovial fluid in joint disease. J Bone Joint Surg 1976;58:396-9. 31. Meisels A, Berebichez M. Exfoliative cytology in orthopedics. Can Med Assoc J 1961;84:957-9. ~
~
1. Hollander JL, Reginato AJ, Torralba T. Examination of synovial fluid as a diagnostic aid in arthritis. Med Clin North Am 1966;50: I28 1-93. 2. Villanueva TG, Schumacher HR. Cytologic examination of synovial fluid. Diagn Cytopathol 1987;4:141-7. 3. Schumacher HR, Reginato AJ. Atlas of synovial fluid analysis and crystal identification. Philadelphia: Lea & Febiger, 1991. 4. Cohen AS, Goldenberg D. Synovial fluid, laboratory diagnostic procedures in the rheumatic diseases. 3rd ed. Orlando: Grune & Stratton, 1985. 5. Naiba ZM. Cytology of synovial fluid. Acta CytOl 1973;17:299-309. 6. Coogeshal HO, Warren CF, Bauer W. The cytology of normal human synovial fluid. Anat Rec 1940;77:129-36. 7. Ropes MW, Bauer W. Synovial fluid changes in joint disease. Cambridge, MA: Harvard University Press, 1953.
152
Diagnostic Cytoputhology, Vol 8, No 2
