Concomitant septic and gouty arthritis an analysis of 30 cases

Transcription

1 Rheumatology 2003;42: doi: /rheumatology/keg297, available online at Advance Access publication 16 April 2003 Concomitant septic and gouty arthritis an analysis of 30 cases K.H.Yu,S.F.Luo,L.B.Liou,Y.-J.J.Wu,W.P.Tsai, J.Y.ChenandH.H.Ho Objectives. To analyse the clinical features and outcomes of gouty patients with concomitant septic arthritis in a medical centre. Methods. From the hospital database, we collected 30 hospitalized cases with concomitant septic arthritis and gouty arthritis from 1987 to All patients had positive bacterial culture and monosodium urate crystals in the affected joints. Medical records of the patients were analysed in detail. Results. The mean age of patients was 52.8"12.5 yr. One-third of patients were afebrile at presentation, 30% had a normal blood leucocyte count and 10% had a synovial fluid leucocyte count less than 6000umm 3. The knee joint was the most common site of involvement, followed by the ankle, shoulder and wrist joints. Most patients had long-standing disease and subcutaneous tophi. Subcutaneous tophi rupture with secondary wound infection is the most common route of infection. Causative micro-organisms were Staphylococcus aureus (16 cases, 7 of whom were oxacillin-resistant), Streptococcus sp. (5 cases), Pediococcus sp. (1 case), and Gram-negative bacilli (9 cases). Fourteen patients received surgical debridement, among them two patients had an arthrodesis owing to severe joint destruction and one received above-knee amputation. Two patients died. One died of septic complications and the other died of acute myocardial infarction. Conclusions. Septic arthritis coexistent with gout presented a diagnostic difficulty. An early diagnosis requires a high level of suspicion. Prompt aspiration and analysis of the synovial fluid is imperative, regardless of the absence of fever or leucocytosis. Culture of the aspirated synovial fluid is warranted in gouty attack, even when it has a low white cell count or the Gram stain reveals no organisms. KEY WORDS: Gout, Septic, Arthritis, Infection, Tophi, Surgery. Septic arthritis is a rare but serious complication of gout and presents a diagnostic difficulty w1x. Since both conditions may cause inflammation of the affected joint with redness and swelling, it is difficult to distinguish them from one another w2x. Delay in diagnosis may cause destruction of the affected joint. It is thus important to diagnose these acute arthritides accurately in the initial management w3x. Literature reports about concomitant septic and gouty arthritis are rare w4 16x. We herein analyse 30 cases encountered at Chang Gung Memorial Hospital from 1987 to Methods and patients From the database of our hospital, we searched for patients with both bacteriologically proved septic arthritis and gouty arthritis between 1987 and Thirty patients were identified. They had concomitant positive bacterial culture and monosodium urate crystals or tophi in the affected joints. All patients had previously fulfilled the 1977 American Rheumatism Association criteria for acute gout w17x. However, patients with septic arthritis and only a past history of gout, but without intra-articular monosodium urate or tophi in the same joints, were not included. Bacteriological diagnosis was Division of Rheumatology, Allergy and Immunology, Chang Gung Memorial Hospital, Tao-Yuan, Taiwan, Republic of China. Submitted 14 November 2002; revised version accepted 5 February Correspondence to: Dr. H. H. Ho, Division of Allergy, Immunology and Rheumatology, Chang Gung Memorial Hospital, 5 Fu-Shin St., Kuei- Shan, Tao-Yuan, Taiwan, Republic of China. gout@adm.cgmh.org.tw 1062 Rheumatology 42 ß British Society for Rheumatology 2003; all rights reserved

2 Concomitant septic and gouty arthritis 1063 established on the basis of synovial fluid cultures aspirated by arthrocentesis or during surgical debridement. Clinical data including gout duration, tophi, affected joints, concomitant diseases, presence of fever (defined as an oral temperature over 37.88C), the duration of symptoms before the visit to our hospital, white blood cell count (WBC), synovial fluid white cell count, blood culture, synovial fluid culture, presumed source of infection, surgical debridement and patient outcome were analysed. The literature from 1976 to 2001 was searched with Medline. Key words used for the search were gout and infectious arthritis (or septic arthritis). Results Patient s demography Thirty gout patients in whom bacteria were cultured from synovial fluid were identified (Table 1). Nineteen patients had intracellular monosodium urate (cases 1 18 and 30) and 11 patients had intra-articular tophi in the infected joints at surgical debridement (case 19 29). The mean age was 52.8"12.5 yr (range 30 83) and the mean duration of gout was 10.8"6.9 yr (range 2 30). Four patients were female. All except five patients had subcutaneous tophi. This incidence (83.3%) is much TABLE 1. Clinical and laboratory data of 30 patients with concomitant crystal and septic arthritis Case Age (yr) Sex Organism Joint Associated conditions higher than that of our previous study (16.8%) from the years 1993 to 2000 w18x. The knee joint is the commonest site of involvement (24 cases), followed by the ankle (6 cases), shoulder (1 case) and wrist joint (1 case). Twenty-seven patients (90%) had monoarticular joint involvement. There were three patients who had two joints infected, two of whom had knee and ankle joint involvement and one had bilateral knee joint involvement. Concomitant diseases Several coexistent medical conditions were identified, including iatrogenic Cushing s syndrome as a result of glucocorticoid abuse for no specific underlying diseases in seven patients, diabetes mellitus in five, liver cirrhosis in two and haemodialysis in one. Two patients had knee replacement owing to degenerative joint disease. Local causes of infection were identified in 13 patients, 11 with a ruptured tophaceous wound and two after total knee replacement surgery. The other patients had no obvious route of joint infection. Symptoms and laboratory data at presentation Fever was described in only 20 (66.7%) patients at presentation. The median time from onset of joint pain to WBC SF white count Gram stain Blood culture Fever Surgery 1 72 M Streptococcus pneumonia Knee No GPC 2 Yes Arthrodesis 2 59 M OSSA Knee Wound GPC + Yes No 3 50 M Streptococcus sp. Knee Liver cirrhosis, wound GPC + Yes No 4 45 M OSSA Knee DM, steroid use GPC 2 Yes Debridement 5 49 M OSSA Knee No GPC 2 Yes Debridement 6 83 F OSSA Ankle Wound GPC 2 Yes No b 7 77 F ORSA Shoulder Steroid use GPC + Yes No 8 50 M Streptococcus sp. Ankle and Liver cirrhosis, wound GPC + Yes No knee 9 65 F ORSA Knee a No NF 2 No No M Proteus vulgaris Knee DM NF 2 No No M Salmonella enteritidis Knee Wound, steroid use NF 2 Yes No b F E. coli Knee No NF 2 Yes No M Enterococcus sp. Knee No NF 2 Yes No M Pseudomonas sp. Knee No NF + Yes No M ORSA Knee Steroid use NF 2 Yes No M Pseudomonas sp. Knee Uraemia ND 2 Yes No M OSSA Knee No ND + No No M OSSA Knee No ND + No Debridement M Group B b-haemolytic Strep. Ankle and Wound ND ND 2 No Amputation knee M E. coli Knee DM 6100 ND ND + Yes Debridement M OSSA Knee Steroid use ND ND 2 Yes Debridement M OSSA Ankle No 8900 ND ND 2 No Debridement M ORSA Both knees Wound ND ND 2 Yes Debridement M ORSA and Pseudomonas sp. Knee Wound ND ND 2 Yes Debridement M ORSA Knee Prosthetic joint ND ND 2 No Debridement M Group B Salmonella sp. Knee DM, steroid use ND ND + No Debridement M ORSA Knee Prosthetic joint ND ND + Yes Arthrodesis M Pediococcus sp. Ankle Wound 4400 ND ND 2 Yes Debridement M OSSA Ankle Wound, DM ND ND 2 No No M Group A Streptococcus Wrist Wound, steroid use 3600 ND GPC + No No DM, diabetes mellitus; GPC, Gram-positive coccus; ND, not done; NF, not found; OSSA, oxacillin-sensitive Staph. aureus; ORSA, oxacillinresistant Staph. aureus; SF, synovial fluid; 2, no; +, yes. a Concomitant with pseudogout. b Death.

3 1064 K. H. Yu et al. hospital visit was 6.5 days. The mean white blood cell count was " (10 9 ul) (range ), and in nine (30%) patients within the normal range. The mean synovial white cell count was "55 330umm 3 (range ), with a mean of 95.7% neutrophils on differential count. However, three patients (cases 3, 8 and 16) had a synovial white cell count less than 6000umm 3. Eleven patients had positive blood cultures. Seven out of 21 (33.3%) patients in the leucocytosis group had positive blood cultures, in contrast to four out of nine (44.4%) patients in the normal white blood count group. Bacteriological investigation and antibiotic treatment Gram-positive cocci comprised 73.3% (22 cases) of infections. Staphylococcus aureus was the commonest pathogen found in 16 cases, with seven isolates being oxacillin-resistant Staph. aureus (ORSA). Streptococcus infections were found in five patients and one patient had a Pediococcus infection. Gram-negative bacteria were found in nine cases (three Pseudomonus aeruginosa, two Escherichia coli, two Salmonella sp., one Proteus vulgaris, one Enterococcus sp.), including one patient with concomitant P. aeruginosa and ORSA infection (case 24). Two patients developed septic arthritis in the replaced knee joint. The time from joint replacement to the development of septic arthritis was 6 and 56 days, respectively, and both were caused by ORSA. In this study, there were no infections with Haemophilus influenzae, gonococci or anaerobes. According to the results of drug sensitivity, initial empirical antibiotic treatment with oxacillin and gentamicin covered the infectious micro-organisms in only 63.3% (19 out of 30) of cases. Two Gram-negative bacilli infections were resistant to gentamicin but sensitive to amikacin. Seven ORSA-infected patients were shifted to vancomycin treatment and one Pediococcus-infected patient was treated with clindamycin. One patient had concomitant necrotizing fasciitis (case 19) and was treated with vancomycin and metronidazole. The mean duration of antibiotic therapy in our patients was 35.2"16.6 days (range 15 55). The mean delay in appropriate antibiotic treatment before the available bacterial drug sensitivity test in 12 patients was 4.8 days. Among them, seven patients received medical treatment and five patients received surgical debridement. Surgical treatment and outcome Fourteen patients underwent surgical debridement. The median time from admission to surgical debridement was 7 days (mean 11.9 days, range 1 66). Five of them received surgical debridement within 2 days. Nine patients received surgical drainage after medical treatment failure. In the latter group, three were within 1 week and an additional five cases were within 1 month. The other patient (case 1) received one-stage arthrodesis at 66 days of hospitalization owing to severe joint destruction. In the two patients with prosthetic joint infection, one (case 27) also had an arthrodesis owing to severe joint destruction and the other patient only received medical treatment. One patient (case 19) had an amputation on day 7 owing to uncontrolled necrotizing fasciitis and septic arthritis. Two patients died. One (case 11) died of sepsis as a consequence of infection and the other, a 83-yr-old female patient (case 6), died of acute myocardial infarction on day 6 of hospitalization. Discussion The association between pyarthrosis and pre-existing rheumatoid arthritis has been well documented w19, 20x; in contrast, the association of pyarthrosis and gouty arthritis has rarely been reported. In a recent study from Taiwan w21x, a high proportion (43.5%) of patients with a septic ankle had concomitant gouty arthritis but none had concomitant rheumatoid arthritis, suggesting the possibility of a racial difference or that the association of septic and gouty arthritis has previously been underreported. The differential diagnosis of gouty arthritis and septic arthritis is usually easy, however it may be difficult and often overlooked when these two diseases are concomitant. Since patients with gout and septic arthritis can present with fever, joint swelling, pain and redness, prompt aspiration of the synovial fluid for analysis is imperative to reach an accurate diagnosis. As the present study illustrates, one-third of the patients were afebrile at presentation. The diagnosis of septic arthritis in gout patients rests on a high index of suspicion. An acutely inflamed joint in gout should also be worked up for septic arthritis. Synovial fluid white cell counts generally exceed umm 3 with more than 90% polymorphonuclear cells (PMNs) in septic arthritis w22x. However, such high cell counts and predominance of PMN may also occur in rheumatoid arthritis, crystalinduced arthritis and seronegative spondylarthropathy w23 25x. The wide range and a substantial overlap of values make these laboratory values less helpful as diagnostic aids. Hence, they should not be used as the primary parameters for diagnosis, but rather as a supplement to clinical information w2x. In this series, 30% of patients initially had a peripheral blood leucocyte count of less than umm 3 and 10% of patients had a synovial leucocyte count less than 6000umm 3. Thus, culture is warranted in gouty attack even though the synovial fluid has a low white cell count or the Gram stain reveals no organisms. This is especially important for the immunocompromised patients with diabetes mellitus, liver cirrhosis, on haemodialysis or using steroids. In addition, breakdown of the skin overlying tophi carries an increased risk of infection. Both gout and infection can affect any joint of the body w26, 27x. While infection most often occurs in the knee, this joint is also commonly affected with pseudogout and gout w28x. We had one patient (case 9) with concomitant pseudogout, gout and septic arthritis, which has also been reported by others w10, 11, 14, 16x. Since crystal synovitis may cause severe synovial fluid leucocytosis and clinically simulate or coexist with a joint infection, a careful search for concomitant monosodium

4 Concomitant septic and gouty arthritis 1065 urate or calcium pyrophosphate dihydrate crystals should be routinely performed in septic arthritis. Besides, joint infection can complicate any synovitis and the septic process may promote the release of crystals from cartilage or synovial membrane, the socalled crystal shedding and strip mining w29, 30x. Thus the demonstration of crystals does not rule out concomitant bacterial arthritis. Micro-organism study should also be performed in crystal arthritis. There was a wide range of organisms involved. Staph. aureus is the most common causative organism and Gram-negative bacilli were found in 30% of cases. It is worthwhile to note that ORSA comprised 23% of the infectious organisms in this study, which explained the decreased response rate to initial oxacillin and gentamicin treatment. ORSA has not previously been a common pathogen in septic arthritis, but seven cases in this study highlight the increasing ORSA infections. This is in accordance with the trend of increased frequency of ORSA joint infection noted in other centres in the past decade w31x. The case fatality rate for bacterial arthritis has not changed substantially in the past 25 yr and ranges from 5to15%w32x. The mortality rate was 6.7% in our gout patients with concomitant septic arthritis. This is in contrast to septic arthritis in patients with rheumatoid arthritis (RA), a group with less gratifying results of treatment. Gardner and Weisman w33x reviewed the literature and noted that 22% of patients with pyarthrosis and RA died as a consequence of infection. This may be due to the use of more immunosuppressive agents and steroids in RA patients. This study highlights the previously under-recognized problem of concomitant septic and gouty arthritis. Diagnosis of these two concomitant diseases rests on a high index of suspicion. The acute inflamed joint in gout should also be worked up for septic arthritis, especially in those who have ruptured subcutaneous tophi or have concomitant medical diseases. Prompt aspiration and analysis of the synovial fluid is imperative and culture is warranted. Initial combination treatment of vancomycin and gentamicin is recommended in areas with a high percentage of oxacillin resistance of Staph. aureus. References 1. Jahn TW. Septic arthritis and calcium pyrophosphate deposition disease in the setting of chronic gout. Mil Med 1998;163: Rogachefsky RA, Carneiro R, Altman RD, Burkhalter WE. Gout presenting as infectious arthritis. Two case reports. J Bone Joint Surg Am 1994;76: Goldenberg DL, Reed JI. Bacterial arthritis. N Engl J Med 1985;312: Shapero C, Fox IM. Pasteurella multocida and gout in the first metatarsophalangeal joint. J Am Podiatr Med Assoc 1999;89: Butt TS, Khan A, Ahmad A, Khan MA, Parke A, Hill DR. Pasteurella multocida infectious arthritis with acute gout after a cat bite. J Rheumatol 1997;24: Salvi A, Rossi M, Balestrieri GP, di Stefano O, Baratto G, Giustina G. Septic polyarthritis in chronic tophaceous gout. Recenti Prog Med 1991;82: O Connell PG, Milburn BM, Nashel DJ. Coexistent gout and septic arthritis: a report of two cases and literature review. Clin Exp Rheumatol 1985;3: Edwards GS Jr, Russell IJ. Pneumococcal arthritis complicating gout. Case report and literature review. J Rheumatol 1980;7: Hamilton ME, Parris TM, Gibson RS, Davis JS 4th. Simultaneous gout and pyarthrosis. Arch Intern Med 1980;140: Jarrett MP, Grayzel AI. Simultaneous gout, pseudogout, and septic arthritis. Arthritis Rheum 1980;23: Ilahi OA, Swarna U, Hamill RJ, Young EJ, Tullos HS. Concomitant crystal and septic arthritis. Orthopedics 1996; 19: Baer PA, Tennenbaum J, Fam AG, Little H. Coexistent septic and crystal arthritis. J Rheumatol 1986;13: Hess RJ, Martin JH. Pyarthrosis complicating gout. J Am Med Assoc 1971;218: Smith JR, Phelps P. Septic arthritis, gout, pseudogout, and osteoarthritis in the knee of a patient with multiple myeloma. Arthritis Rheum 1972;15: McConville JH, Pototsky RS, Calia FM, Pachas WN. Septic and crystalline joint disease: a simultaneous occurrence. J Am Med Assoc 1975;231: Mayer JW, DeHoratius RJ, Messner RP. Serratis marcescens-caused arthritis with negative and positive birefringent crystals. Arch Intern Med 1976;136: Wallace SL, Robinson H, Masi AT, Decker JL, McCarty DJ, Yu TF. Preliminary criteria for the classification of the acute arthritis of primary gout. Arthritis Rheum 1977;20: Yu KH, Luo SF. Younger age of onset of gout in Taiwan. Rheumatology 2003;42: Goldenberg DL. Infectious arthritis complicating rheumatoid arthritis and other chronic rheumatic disorders. Arthritis Rheum 1989;32: Nolla JM, Gomez-Vaquero C, Fiter J et al. Pyarthrosis in patients with rheumatoid arthritis: a detailed analysis of 10 cases and literature review. Semin Arthritis Rheum 2000;30: Lee CH, Chen YJ, Ueng WN, Hsu WW. Septic arthritis of the ankle joint. Chang Gung Med J 2000;23: Liu NY, Giansiracusa DF. Septic arthritis. In: Gorbach SL, Bartlett JG, Blacklow NR, eds. Infectious disease. Philadelphia: WB Saunders, 1998: Singleton JD, West SG, Nordstrom DM. Pseudoseptic arthritis complicating rheumatoid arthritis: a report of six cases. J Rheumatol 1991;18: Viallnueva TG, Schumacher HR. Cytologic examination of synovial fluid. Diagn Cytopathol 1987;3: Krey PR, Bailen DA. Synovial fluid leukocytosis: A study of extremes. Am J Med 1979;67: Kelley WN, Wortmann RL. Gout and hyperuricemia. In: Kelley WN, Harris ED, Ruddy S, Sledge CB, eds. Textbook of rheumatology. Philadelphia: WB Saunders, 1997: Michael AB, Levinson DJ. Clinical gout and the pathogenesis of hyperuricemia. In: Koopman WJ, ed. Arthritis and allied conditions. Baltimore: Williams & Wilkins, 1997: Grahame R, Scott JT. Clinical survey of 354 patients with gout. Ann Rheum Dis 1970;29:461 8.

5 1066 K. H. Yu et al. 29. Gordon TP, Reid C, Rozenbilds MA, Ahern M. Crystal shedding in septic arthritis: case reports and in vivo evidence in an animal model. Aust N Z J Med 1986; 16: Terkeltaub RA. Pathogenesis and treatment of crystalinduced inflammation. In: Koopman WJ, ed. Arthritis and allied conditions. Baltimore: Williams & Wilkins, 1997: Goldenberg DL. Septic arthritis. In: Kelley WN, Harris ED, Ruddy S, Sledge CB, eds. Textbook of rheumatology. Philadelphia: WB Saunders, 2001: Goldenberg DL. Septic arthritis. Lancet 1998;351: Gardner GC, Weisman MH. Pyarthrosis in patients with rheumatoid arthritis: a report of 13 cases and a review of the literature from the past 40 years. Am J Med 1990; 88: