1052 LETTERS Total lymphoid irradiation for psoriatic arthritis: case report To the Editor: Total lymphoid irradiation has been used with encouraging results in the treatment of severe rheumatoid arthritis (Kotzin BL, Strober S, Engleman EG, Calin A , Hoppe RT, Kansas GS, Terrell CP, Kaplan HS: Treatment of intractable rheumatoid arthritis with total lymphoid irradiation. N Engl J Med 305:969-976, 1981). We have recently used this treatment for a patient with psoriatic arthritis unresponsive to conventional management, in whom the development of hepatitis precluded the use of other cytotoxic agents. The patient, a 31-year-old white man, first reported to our clinic in June 1981 with a 9-year history of monarthritis of the right knee. Other than active synovitis at that joint and a cushingoid appearance due to 9 years of continuous steroid intake, there were no relevant findings. His steroid dosage at the time was 1.5 mg dexamethasonelday. Rheumatoid factor, antinuclear antibodies, and HLA-B27 were all negative. Complete blood count findings were normal, erythrocyte sedimentation rate (ESR) was 16 mm/hour, uric acid was 4.2 mg %, and urinalysis results were normal. Radiographs of the knee showed a mildly diminished joint space with severe periarticular osteoporosis. Synovectorny and gradual tapering of the steroid were advised. A month later he had the synovectomy, which revealed a nonspecific synovitis with no lymphoid follicles. One week after surgery, while using prednisolone 5 mg/day, he developed arthritis in the unoperated knee. This was followed in a few days by severe arthritis of both elbows and the neck. Meanwhile, extensive pustular psoriasis of the nail beds emerged at the hands and feet. He became bedbound because of the severity of his arthritis. Indornethacin 150 mglday and hydroxychloroquine 400 mg/day were initiated. Three weeks after the synovectomy he passed a calcium phosphate stone. Intravenous pyelogram showed normal sized, shaped, and functioning kidneys; however, he had bilateral sacroiliitis. Six weeks after the synovectomy, jaundice developed. Enzyme changes were consistent with viral hepatitis. Auagand Auab were negative. All medication was stopped except for prednisolone, which was increased to 20 mg/day for the persistently severe arthritis. His condition worsened with this regimen, with the arthritis spreading to his shoulders and hips. Although he was having an uneventful recovery from hepatitis, we felt that he should not receive nonsteroidal antiinflammatory agents or hydroxychloroquine. It was also Arthritis and Rheumatism, Vol. 26, No. 8 (August 1983) our feeling that an arthritis not controlled by 20 mg of prednisolone/day would very likely not respond to these measures. Oral or parenteral cytotoxic agents were also not deemed appropriate. Thus he was offered total lymphoid irradiation for arthritis control. Two months after the onset of his hepatitis, when serum bilirubin, serum glutamic oxaloacetic transaminase (SCOT), serum pyruvic oxaloacetic transaminase (SGPT), and alkaline phosphatase levels were normal, informed consent was obtained and irradiation was first administered to lymphoid tissues below the diaphragm. A 10-meV linear accelerator was used. A midline dose of 200 rads was administered at each session for 10 sessions over 14 days for a total of 2.000 rads. This was immediately followed by irradiation of the “mantle field” (Kaplan HS: Hodgkin’s Disease. Second edition. Cambridge, Massachusetts, Harvard University Press, 1980, pp 366-441). A midline dose of 150 rads for 13 sessions over 20 days, for a total of 1,950 rads, was given. His arthritis began to improve toward the end of the “mantle” irradiation. One month after the treatment he was able to ambulate freely. His ESR was 70 mm/hour immediately before the treatment. Hematocrit at that time was 32% and white blood cell count (WBC) was 6,200 with 28% lymphocytes. The lowest lymphocyte count was 720/mm3 (WBC 4,000/mm3 with 18% lymphocytes) during the treatment. ESR began to rise following the radiation, as previously noted by Kaplan. Twenty-five days after the treatment, the ESR was 110 m d h o u r while his clinical condition was rapidly improving. Sixty days after completion of the treatment his ESR was 90 m d h o u r , with hematocrit at 38% and WBC at 5,600/mm3 (22% lymphocytes). There was no elevation of SCOT, SGPT, or alkaline phosphatase during or after the therapy. The 20 mg of prednisolone was continued through the treatment and was very slowly tapered thereafter. At the fifth month after the therapy, with the patient having minimal synovitis at the unoperated knee, hydroxychloroquine (400 mg/day) was restarted. Fourteen months after radiotherapy he is doing well, fully ambulatory with no active synovitis, receiving hydroxychloroquine 400 mg/day and prednisolone 8.75 mg/day. His current ESR is 30 mmlhour, hematocrit 42%, and WBC 4,200 with 25% lymphocytes. Psoriatic nail changes are improved, but the nails are still deformed. Very special circumstances prompted us to utilize total lymphoid irradiation in this patient. The favorable results warrant consideration of this treatment modality in severe psoriatic arthritis. Hasan Yazici, MD Cerrahpaja Medical Faculty Nijad Bilge, MD Munir Kinay, MD Istunbid Medical Faculty Istanbul, Turkey LETTERS 1053 HLA-DR antigen frequency (%) in seropositive and seronegative rheumatoid arthritis Table 1. DR 5 w6 7 w8 8.6 32.25 20.43 21.50 5.37 18.0 38.0t 40.0 6.12 24.0 - 7.41 25.9 40.74$ 33.33 7.41 7.41 - 13.04 8.7 34.78* 47.83 4.5 43.48 - Subjects, no. 1 2 3 Control, 93 Rheumatoid arthritis, 50 Rheumatoid factor +, 27 Rheumatoid factor 23 16.13 30.10 12.9 22.0 10.0* 25.93 17.39 -. 4 * Pcorr < 0.05. t Pcorr < 0.001. < 0.01. HLA antigens and rheumatoid arthritis $ Pcorr To the Editor: We would Like to contribute some information about the controversy regarding the association of HLA-DR4 antigen and 19s rheumatoid factor (RF) in rheumatoid arthritis (RA) (1-3). Fifty-five RA patients living in Rome, Italy were typed for HLA-A, B, and C antigens; 50 of these were also typed for HLA-DR antigens (DRI-DRw8). Diagnosis conformed to American Rheumatism Association criteria (4). We also analyzed the following parameters: sex, age of onset, anatomic stage (erosions), and the presence of subcutaneous nodules. Fourteen of the 55 patients were males and 41 were females; 30 were seropositive and 25 were seronegative. The patients were also classified into two groups: those without bone lesions (1st stage) and those with bone lesions (2nd, 3rd, 4th stage) according to radiologic reports. Four hundred eighty-three random donors were typed for the HLA-A, B, and C antigens and 93 for HLA-DR as a control group. Typing sera were selected from the 8th International Histocompatibility Workshop. No association was found between HLA-A, B, C antigens and RA. Our results (Table I ) confirm the strong association between HLA-DR4 antigen and RA (P corr = 0.00024, relative risk = 6.51). In contrast with previous reports (2), we found no correlation between DR4 and rheumatoid factor, even when only patients with high titers were considered. The difference may be due to heterogeneity of our RA patients versus those reported in the literature. Our group of patients, originally from Rome, can be considered a representative sample of the Italian population in relation to the distribution of the HLA antigens. No relation was shown between DR4 and articular erosions, which is not surprising since erosions are linked with rheumatoid factor. No preferential genotypic combination was found among DR antigens; our sample is on Hardy-Weinberg equilibrium. Some antigens are under-represented-for instance, DR2 was significantly diminished in the patient group as a whole (Pcorr = 0.028). The meaning of this decrease is not known. The rarity of DR2 is not unique to RA but is also rare in other conditions, such as celiac disease (5). It has been suggested that the DR antigen has a protective role (6), but we cannot exclude or confirm a compensatory decrease of this allele relative to the increased frequency of DR4 in the available sample. On the other hand, we cannot explain why an increase of DR4 frequency is only associated with a corresponding decrease of DR2 frequency, without a decrease affecting any of the other DR alleles. The family study supports our data. Two of the 10 families tested have 2 affected siblings. In Family A (Figure I ) we find 2 HLA identical sisters who are both DR4 positive, 1 with seropositive and the other with seronegative RA. Family B (Figure 2) presents 2 affected sisters, both DR4 positive but with different HLA haplotypes; 1 sister was seropositive and the other had a seronegative form of arthritis. The failure to study the parents does not exclude a crossing over in the family. P. Lulli S. Cappellacci M. Morellini Cattedra di Genetica Medica Universitd degli Studi di Roma M. Galeazzi L. Schiavetti T. Tuzi Servizio di Reumatologia Ospedale S . Camillo di Roma Roma, Italy ac ad bc bd ec Figure 1. Family A genotypes: a = Aw23,C-,Bw49,Bw4,DR5; b = c= d = A-,C-, A~,CW~,BW~~,BW ~ ,All,C-,Bw49,Bw4,DR4; DR~; Bw22,Bw6,DR2; e = AI,Cw3,Bw22,Bw6,DR-. Circles indicate members with rheumatoid arthritis: + = seropositive; - = seronegative.