ORIGINAL ARTICLE Annals of Nuclear Medicine Vol. 13, No. 6, 389-395, 1999 Comparison of Tc-99m MDP, Tc-99m HSA and Tc-99m HIG uptake in rheumatoid arthritis and its variants Murathan SAHlN,* Irem BERNAY,* Tarik BASOGLU* and Ferhan CANTURK** Departments of *Nuclear Medicine and **Physical Therapy and Rehabilitation, Ondokuz Mayis University Faculty of Medicine, Samsun, Turkey Tc-99m polyclonal immunoglobulin-G has been shown to be a successful agent in the depiction of active inflammation in rheumatoid arthritis (RA). The objective of this study was to compare the uptake behaviors of Tc-99m HIG and Tc-99m MDP in RA and variants of rheumatoid arthritis (VRA). Seventeen patients with RA and 8 patients with VRA presenting with active inflammation were included in this study. Ten subjects with well-diagnosed degenerative joint disease constituted the control group. All joints in patients were also imaged with Tc-99m HSA to evaluate the vascularization status of the joints. Tc-99m HIG and HSA scans were obtained at 2, 4 and 24 hours after the injection of 555 MBq Tc-99m HIG and 296 MBq Tc-99m HSA. Conventional bone scans were performed 4 hours after the injection of 740 MBq Tc-99m MDP. Target-to-background (T/B) ratios were obtained exclusively over the joint regions. Tc-99m HIG T/B ratios of the active joints in RA were significantly higher than those of the non-active joints and the control group (p < 0.05). Tc-99m HIG T/B ratios in active joints showed a progressive increase between 2 and 24 hour images (p < 0.05). In contrast, Tc-99m HSA T/B ratios decreased in all active joints significantly (p < 0.05) except the ankle joint region (p > 0.05). The T/B ratios in Tc-99m MDP bone scans were higher in all active joints than in non-active RA joints and joints of controls but significantly differences were only detected in wrist and elbow joints. All clinically active joints in VRA patients accumulated Tc-99m HIG and HSA, and showed increased Tc-99m MDP uptake. These joints had a very similar Tc-99m HIG retention pattern to the RA joints. The detection rate of active joint inflammation with Tc-99m HIG was much higher than that with Tc-99m MDP. The increasing Tc-99m HIG uptake ratio between 2 and 24 hours in contrast to Tc-99m HSA indicates the presence of other binding mechanisms besides increased vascularity in RA. Key words: arthritis, technetium-99m polyclonal immunoglobulin G, technetium-99m albumin, technetium-99m methylene diphosphonate INTRODUCTION RHEUMATOID ARTHRITIS (RA) is a systemic disease that localize predominantly in diarthrodial joints. It is thought that RA, without knowing its etiology, is of autoimmune origin.1,2 Infiammatory response involves the increasing permeability of the venules and exudation of plasma proteins into the synovial stroma followed by perivascular Received March 18, 1999, revision accepted August 18, 1999. For reprint contact: Dr. Murathan Sahin, Department of Nuclear Medicine, Ondokuz Mayrs University Faculty of Modicine, Samsun 55139, TURKEY. infiltration by cellular elements in inflamed joints. Rheumatoid factor (RF), an autoantibody produced by synovial tissues and B Iymphocytes targeting the Fc portion of the immunoglobulin G (IgG), is present in approximately 80% of RA patients.3 Commonly accepted variants of rheumatoid arthritis (VRA) are juvenile chronic arthritis (JCA), Felty's syndrome, psoriatic arthritis, arthritis associated with ulcerative colitis, arthritis associated with Whipple's disease and arthritis associated with Sjogren's disease.4 Radionuclide imaging of RA may provide useful diagnostic and prognostic information about the status of the disease. Technetium-99m methylene diphosphonate (Tc-99m MDP) bone scanning actually plays the major role in the radionuclide assessment of arthritis. Three phase bone scanning may discriminate active synovial inflammation from degenerative joint disease and non-active arthritic joints. The latter method has unfortunately the limitation of being carried out only in one region of the skeleton. Studies comparing Tc-99m MDP to other radiopharmaceuticals and clinical criteria have been performed to improve the inflammation specificity.5,6 One of these agents is radiolabeled human polyclonal IgG. In this study we compared the uptake behavior of Tc-99m MDP, Tc-99m human polyclonal immunoglobulin G (Tc-99m HIG) and Tc-99m human serum albumin (Tc-99m HSA) in RA and in some VRA by planar scintigraphy. Tc-99m HSA scanning was performed for multijoint blood pool imaging as a marker of increased vascular permeability. PATIENTS AND METHODS Twenty-five patients with RA and with VRA were studied. Seventeen of them were RA patients (4 males and 13 females, mean age 44+-10 yrs, range 27-64 yrs). The remaining 8 patients were diagnosed as VRA (4 cases of psoriatic arthritis, 3 cases of JCA and one case of Felty syndrome; all females, mean age 26+-1O, range 16-19 yrs). Patients with RA fulfilled the criteria of the American Rheumatism Association 7 Rheumatoid factor (RF) was positive in all RA patients and negative in VRA patients. Ten control subjects with well-diagnosed degenerative joint disease constituted the control group (3 males and 7 females, mean age 54+-6 years, range 45-65 yrs). Clinical assessment and Tc-99m HIG scanning were performed on the same day. Active joints were determined according to the presence of pain and/or swelling. Static images (256 x 256 matrix) including shoulders, elbows, hands, hips, knees, and feet were acquired at 2, 4 and 24 hours, after the i.v. injection of 555 MBq Tc-99m HIG (TechneScan HIG, Mallinkrodt, Petten, Netherlands), with a scintillation camera equipped with a low energy all purpose collimator. The durations of these acquisitions were 7, 10 and 20 minutes, respectively. In the Tc-99m HSA study, the acquisition parameters of the Tc-99m HIG study were respected. Tc-99m HSA (296 MBq) and Tc-99m MDP (740 MBq) spot images were obtained in all patients separately. Tc-99m HSA studies in control subjects were omitted to reduce radiation exposure in this group. There was a time interval of 48-72 hours between the sequential studies. Quantitative analysis was carried out in 34 joints including the shoulder (n = 2), elbow (n = 2), wrist (n = 2), PIP (n = 1O), MCP (n = 1O), hip (n = 2), knee (n = 2), ankle (n = 2) and forefoot (n = 2) joints. Joints in the forefoot were regarded as one single area. In the case of inadequate positioning, extravasation of radiopharmaceuticals, the presence of hip or knee prothesis and local contamination, the related joint was excluded from the quantitative analysis in Tc-99m HIG, HSA and MDP studies. In Tc-99m HIG and HSA studies, quantitative analysis was performed by dividing the average counts per pixel in each joint by the average counts per pixel in a background region located around the soft tissue of the thigh (T/B ratio). In Tc-99m MDP studies, background region values were obtained from the mid femoral bone and T/B ratios were calculated. The Mann-Whitney U test was used to compare the T/B ratios of Tc-99m HIG scintigraphy of active joints in RA and non-active joints in RA, as well as to compare the active joints in RA and the joints in the control group. This test was also used to compare the values for active joints in VRA and non-active joints in VRA patients. The Wilcoxon test was used to compare the T/B ratios at 2 and 4, 2 and 24, 4 and 24 hours in Tc-99m HIG and HSA studies. The results are presented as the mean+-standard deviation (S.D.). RESULTS All patients with RA had 3 or more joints presenting active infiammation. VRA patients had at least one active joint at the moment of study. Clinical joint infiammation signs were not observed in the patients in the control group. The numbers of joints evaluated quantitatively in RA and VRA patients were 555 and 205 in Tc-99m HIG; 372 and 204 in Tc-99m HSA; 535 and 191 in Tc-99m MDP studies, respectively. The joints in controls were 289 and 283 in Tc-99m HIG and MDP studies. One hundred and ninety-six joints, which showed clinically active inflammation in RA (n = 164) and VRA (n = 32) patients, accumulated Tc-99m HIG significantly. Tc-99m HSA studies could be performed in 116 of 164 Tc-99m HIG positive joints in RA patients and 31 of 32 Tc-99m HIG positive joints in VRA patients which were all positive. No Tc-99m HIG uptake was observed in degenerative joint disease and non-active joints of RA and VRA patients. Tc-99m HSA uptake was not observed in non-active joints of RA and VRA patients. In all joints showing active inflammation in RA (n = 164). Tc-99m HIG T/B ratios at 2, 4 and 24 hours were significantly higher than in the control group (n = 289) (p < 0.05) and non-active joints in RA patients (n = 391) (p < 0.05) in all regions (Table 1). When we compared the quantitative results at 2, 4 and 24 hours, we found a progressive increase in the mean Tc-99m HIG T/B ratios in active joints of RA patients (Table 1, Fig. 1 ). Significant differences were detected between 2 and 24 h, and 4 and 24 h values (p < 0.05). Although initial Tc-99m HSA accumulation (2 h) was observed in all active joints in RA patients, the T/B ratios had decreased in 24 h scans (Table 2, Fig. 2). Significant differences were detected between 2 h and 24 h scans in all regions (p < 0.05) except in ankle joints (p > 0.05). The mean Tc-99m MDP T/B ratios in RA active joints were higher than those in the control group and those in RA non-active joints. Statistically significant differences were only found for wrist and elbow joints (p < 0.05). The distribution of Tc-99m MDP mean T/B ratios in active, non-active and control joints are given in Figure 3. Thirty-three active joints in VRA patients demonstrated both increased Tc-99m HIG and HSA accumulation. These joints had showed a very similar Tc-99m HIG retention pattern to the joints in RA patients (Table 3, Fig. 4). Two, 4 and 24 h Tc-99m HIG and HSA T/B ratios of all active joints in VRA patients were significantly higher than those of non-active joints (p < 0.05) (Tables 3 and 4). Although Tc-99m HIG T/B ratios of all active joints at 24 h were higher than those obtained after 2 h and 4 h postinjection (p < 0.05). Tc-99m HSA ratios of active joints were decreased in time. Tc-99m HSA ratios at 24 h were lower than at 2 and 4 h (p < 0.05). In the Tc-99m MDP study, T/B ratios in active joints were higher than in non-active joints, but significant differences were detected only in MCP, wrist and knee joint regions (p < 0.05) (Table 5). DISCUSSION Radiolabeled polyclonal human IgG has been proposed for the detection of inflammation.2,8 It was first thought that the accumulation of Tc-99m HIG in inflamed tissues was related to autoimmune mechanisms, but the uptake mechanism could not be completely explained. One hypothesis is the binding of the Fc fragments of IgG to the receptors of inflamed cells.9 It is shown that IgG retention in inflamed tissues is not due to specific immunologic properties of IgG, and Fc fragment was not bound to leucocytes.10,11 A commonly accepted mechanism is increased vascular permeability.12 Studies on bacterial and sterile infiammations show that there are also other factors besides increased vascular permeability.13,14 Application of Tc-99m HIG on synovial tissue sections of rats with adjuvant arthritis showed binding to the extracellular components of synovial tissue such as fibronectin, fibrin and collagen types I and III. This binding was not dependent on increased vascularization.15 In addition, binding to bacteria in sites of infection and RF production or immunocomplex precipitation for RA were suggested to explain its accumulation.16,17 In our study, the inflamed joints were visualized with Tc-99m HSA due to its high blood pool activity and extravasation of Tc-99m HSA into the inflammation site. Increased inflamed joint activity in Tc-99m HIG scans may be explained by increased perfusion as it was demonstrated in Tc-99m HSA studies. Increased Tc-99m HIG T/B ratios at 24 hours versus decreased Tc-99m HSA T/B ratios support the possibility of other mechanisms besides increased vascular permeability. De Bois et al. reported the absence of IgG accumulation in arthritis and histologically increased vascularity which indicated mechanisms other than the latter.18 The Tc-99m HIG and HSA ratios obtained in VRA patients with negative RF were not significantly different from the values in RA patients with positive RF. This finding places the role of RF related Tc-99m HIG accumulation in question. Liberatore et al. have found 100% sensitivity of Tc-99m HIG in the detection of clinically manifest RA in a comparative study versus Tc-99m labeled leucocytes.19 Vesterskiold et al. have demonstrated a significant correlation between Tc-99m HIG uptake and joint swelling in RA. In contrast, joint pain was not correlated with Tc-99m HIG uptake.20 The authors explained this finding by the fact that rheumatoid joint pain could well remain for a long time after synovitis has been abated. They concluded that Tc-99m HIG scintigraphy may be helpful in the detection of early synovitis that can often be difficult to detect by physical examination alone. Early instituted therapy has a positive prognostic value in RA. Bone scanning is important to localize the site of a lesion but it provides little information about soft tissue involvement and the nature of the lesion due to its low specificity. In the present study, conventional bone scans revealed increased osteoblastic activity in active and non-active joints in patients and in joints with degenerative changes in the control group. There were no significant differences among these three groups except for wrist and elbow joints. Weight-bearing joints and small joints in the hand are more frequently involved in degenerative joint disease than other joints.4 They showed intense accumulation of Tc-99m MDP which did not differ from active joints of RA patients both qualitatively and quantitatively. In general, the wrists, elbows and shoulders seldom show signs of degenerative changes.4 This may explain lower Tc-99m MDP accumulation in these joints in the control group than in the active joints of RA patients. Arthritic joints in remission showed signs of increased Tc-99m MDP accumulation, possibly due to the loss of articular cartilage and residual subchondral erosion defects of the previous active episodes in the joints. Increased vascular permeability apparently leads to the accumulation of any of these radiopharmaceuticals. As has been postulated in other types of inflammation, still unknown mechanisms of uptake and/or physicochemical characteristics of the Tc-99m HIG molecule appear to play a role besides increased vascular permeability in its accumulation in RA. Although the number of patients in the VRA group is not great, we can suggest that the binding mechanism in this group does not differ significantly from that in RA patients and is probably non-specific. Four hour images gave the best results in the visual analysis. Relatively decreased background activity enabled the detection of lesions showing signs of low activity which could be missed on 2 hour images. Although it was shown that the Tc-99m HIG T/B ratio increases at 24 hours, degradation of the image quality due to the radioactive decay of Tc-99m places the late Tc-99m HIG imaging in an optional role. We conclude that Tc-99m HIG scanning may be used to detect active inflammation in RA and its variants. Combined Tc-99m MDP-Tc-99m HIG whole body imaging can help to differentiate active from non-active rheumatic lesions. REFERENCES 1. Maini RN. Autoimmunity in rheumatoid arthritis. An approach via a study of B Iymphocytes. Rheum Dis Clin North Am 13: 319-328, 1987. 2, van der Lubbe PAHM, Arndt SW, Calame W. Measurement of synovial inflammation in rheumatoid arthritis with Tc 99m labeled human polyclonal immunoglobulin G. Eur J Nucl Med 18: 119-123, 1991. 3. Arnett FC. Rheumatoid Arthritis. In: Cecil Textbook of Medicine, Wyngarden JB, Smith LH, Benneth JC (eds.), Philadelphia, W.B. Saunders Company, pp, 1508-1515, 1992. 4. Cotran RS, Kumar V, Robbins SL. Robins Pathologic Basis of Disease. 4th ed., Philadelphia, W.B. Saunders Company, pp. 1346-1363, 1989. 5. Datz FL. Radionuclide imaging of joint inflammation in the '90s. J Nucl Med 31 : 684-687, 1990. 6. Palermo F, Boccaletto F, Paolin A, Carniato A, Zoli P, Giusto F, et al. Comparison of Technetium-99m-MDP, Technetium-99m-WBC and Technetium-99m-HIG in Musculoskeletal Inflammation. J Nucl Med 39: 516-521 , 1998. 7. Arnett FC, Edworty 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 31 : 315-324, 1988. 8. Rubin RH, Fischman AJ, Callahan RJ, Khaw B, Keech F, Ahmad M, et al. In 111-labeled nonspecific immunoglobulin scanning in the detection of focal infection. N Engl J Med 321: 935-940, 1989. 9. Fischman AJ, Rubin RH, Khaw BA, Callahan RJ, Wilkinson R, Keech F, et al. Detection of acute inflammation with In-111 labeled nonspecific polyclonal Ig G. Semin Nucl Med 18: 335-344, 1988. 1O. Morrel EM, Tompkins RG, Fischman AJ, Wikinson RA, Burke JF, Rubin RH, et al. Autoradiographic method for quantitation of radiolabeled proteins in tissues using indium-111. J Nucl Med 30: 1538-1545, 1989. 11 . Gardner P, Oster ZH. Rubor, dolor, calor, tumor and radionuclide scans. N Engl J Med 321 : 970-972, 1989. 12. van der Laken CJ, Boerman OC, Oyen WSG, van de Ven MTP, van der Meer JWM, Corstens FHM. Scintigraphic detection of infection and inflammation: new developments with special emphasis on receptor interaction. Eur J Nucl Med 25: 535-546, 1998. 13. Calame W, Welling M, Feitsma HIJ, Goedemans WT, Pauwels EKJ. Contribution of phagocytic cells and bacteria to the accumulation of technetium-99m labelled polyclonal human immunoglobulin at sites of inflammation. Eur J Nucl Med 22: 638-644, 1995. 14. Gungor F, Karayalcin B, Gultekin M, Paksoy N, Senturk UK, Kaputlu I, et al. Comparison of Tc-99m HIG and Ga-67 citrate in the evaluation of bacterial abscess in a rat model. Ann Nucl Med 1O: 79-83, 1996. 15. De Bois MHW, Welling M, Verwey CL, de Vries E, Pauwels EKJ, Breedveld FC, et al. Tc-99m HIG accumulates in the synovial tissue of rats with adjuvant arthritis by binding to extracellular matrix proteins. Nucl Med Commun 17: 54-59, 1996. 16. Calame W, Feitsma HIJ, Ensing GJ, Goedemans WT, Cams JAJ, van Furth R, et al. Detection of local staphylococcal infection in mice with technetium-99m-labeled polyclonal human immunoglobulin. J Nucl Med 32: 468-474, 1991. 17. Berna L, Torres G, Diez C, Estorch M, Martinez-Ducker D, Carrio I. Technetium-99m human polyclonal immunoglobulin G studies and conventional bone scans to detect active joint inflammation in chronic rheumatoid arthritis. Eur J NuclMed 19: 173-176, 1992. 18. De Bois Maw, Amdt JW, Tak PP, Kluin PM, van der Velde EA, Pauwels EKJ, et al. Tc-99m labeled polyclonal human immunoglobulin G scintigraphy before and after intraarticular knee injection of triamcinolone hexacetonide in patients with rheumatoid arthritis. Nucl Med Commun 14: 883-887, 1993. 19. Liberatore M, Clemente M, Lurilli AP, Zorzin L, Marini M, Di Rocco E, et al. Scintigraphic evaluation of disease activity in rheumatoid arthritis: a comparison of technetium-99m human non-specific immunoglobulins, Ieucocyies and albumin nanocolloids. Eur J Nucl Med 19: 853-857, 1992. 20. Vesterskiold L, Schnell PO, Jacobson H. Comparison between scintigraphy with polyclonal immunoglobulin (Tc-99m HIG) and physical examination in polyarthritic disease. Scand J Rheumatol 25: 159-163, 1996.