Case in Point , Infectious Tenosynovitis, and Rupture in a Patient With Rheumatoid Arthritis and Psoriasis

Peter Vu Bui, MD; and Ifeoma Stella Izuchukwu, MD

Multiple comorbidities, advanced age, tobacco use, and alcohol abuse made a proper diagnosis difficult in a patient with polyarticular , infectious tenosynovitis, and ruptures in the of both thumbs.

ompared with monoarticular of 31% to 42% compared with 4% to milleri (S. milleri). During surgery, he arthritis, polyarticular arthri- 8% for monoarticular septic arthri- was also found to have bilateral ex- tis may yield an initially nar- tis, and RA was present in 67% of the tensor pollicus longus (EPL) tendon Crower differential diagnosis PASA fatalities.1 rupture. Given the possible morbid- that focuses on systemic inflamma- Rheumatoid arthritis and its treat- ity, the authors believe this patient tory conditions, such as rheumatoid ment predispose patients to septic may be of interest to the medical arthritis (RA). Approximately 15% to arthritis. Septic arthritis in the UK community. 30% of septic arthritis is polyarticu- general population is 0.42 per 100 lar, of which about 45% is associated patient-years for patients with RA on CASE PRESENTATION with underlying RA.1,2 Regardless of antitumor necrosis factor therapy.3,4 A 69-year-old African American male the number of joints involved, septic In a retrospective study in the U.S., presented with 3 to 4 days of swell- (infectious) arthritis is a valid con- the incidence of septic arthritis was ing and pain of bilateral wrists, bi- sideration given the morbidity and 0.40 per 100 patient-years for pa- lateral hands, and the left ankle with mortality. tients with RA compared with 0.02 subjective, but resolved, fevers and In a retrospective study in the per 100 patient-years for patients chills. His medical history was sig- United Kingdom (UK) between 1982 without RA.5 nificant for seropositive erosive RA, and 1991, the morbidity and mortal- Other complications of RA in- psoriasis, hypertension, hyperlipid- ity of septic arthritis was 31.6% and clude infectious tenosynovitis and emia, alcohol abuse, chronic tobacco 11.5%, respectively, and 16% of the tendon rupture. The incidence and use, osteoporosis, and glaucoma. He study population had RA.3 A review prevalence of infectious tenosynovi- did not have diabetes, reported no of the literature by Dubost and col- tis and tendon rupture in RA are not IV drug abuse, and except for the leagues found that polyarticular sep- firmly established in the literature. immunosuppressive effects of his tic arthritis (PASA) has a mortality We present a patient with RA and medications, was not otherwise im- psoriasis who responded initially to munocompromised. acute management for RA but subse- For 2 years in the outpatient set- Dr. Bui is a resident physician in the Department of Internal Medicine at the University of New quently was diagnosed with culture- ting, the rheumatology clinic had Mexico in Albuquerque. Dr. Izuchukwu is a staff negative polyarticular arthritis and been managing the patient’s rheu- physician at the VA Greater Los Angeles Health infectious tenosynovitis associated matoid factor (RF) positive and anti- Care System in California and associate clinical professor of medicine at the David Geffen School with beta hemolytic group G Strepto- cyclic citrullinated peptide (CCP) of Medicine at UCLA. coccus (GGS), a part of Streptococcus antibody positive erosive RA with

www.fedprac.com FEBRUARY 2015 • FEDERAL PRACTITIONER • 31 Septic Arthritis

etanercept 25 mg subcutaneously The patient was started on pred- and had no growth (Table). Gastro- twice a week. The RA affected his nisone 40 mg orally once a day (for 5 enterology studies were limited to hands, wrists, shoulders, and ankles days) for empiric treatment of an RA stool cultures and did not include bilaterally but was successfully con- flare and continued on etanercept. colonoscopy. Leukocytosis began trolled. The dermatology clinic was The inpatient rheumatology service trending down. managing the patient’s psoriasis with was consulted. Further evaluation On day 8, antibiotics were tai- calcipotriene cream 0.005% twice a later that day found involvement of lored to penicillin G 4 million units week and clobetasol ointment 0.05% the proximal interphalangeal joints IV every 4 hours following growth twice a week. Psoriatic plaques were and elbows and tenderness of the of GGS from the sample of the left noted on bilateral elbows, bilateral tendons of the dorsal hand bilater- wrist. Subsequently, synovial fluid dorsal hands, and bilateral dorsal feet. ally. Over the next 2 days, the pa- (3 mL) from the left shoulder was tient remained afebrile and WBCs obtained following initiation of an- Initial Evaluation were within normal limits. Edema, tibiotic therapy and had no growth. At evaluation, the patient’s vital erythema, and tenderness of the in- Magnetic resonance imaging (MRI) signs revealed a temperature of volved joints somewhat improved, found tenosynovitis of the left ankle 36.3°C (97.3°F), pulse of 102 beats but tenderness along the tendons of and right wrist. per minute, respiratory rate of the dorsal hand worsened, which On day 9, transthoracic echocar- 16 breaths per minute, oxygen satu- concerned the managing teams for diography was performed and found ration of 99% on room air, and blood infectious tenosynovitis. no evidence of infectious endocar- pressure of 102/70 mm Hg. He was By day 4, the patient was afebrile ditis. Later that night, the patient found to have edema, tenderness, and had a leukocytosis of 12.9 k/μL was taken to surgery for incision and and erythema of the wrists bilaterally with neutrophils 86.7%, but im- drainage/debridement of bilateral and left metacarpophalangeal joints provement of erythema, pain, and wrists and left ankle, synovectomy (MCPs) and edematous right MCPs range of motion of involved joints of right wrist, and aspiration of right and left medial ankle. and no tenderness to palpation of shoulder. Findings included abscess The patient had been nonadher- tendons was noted. The inpatient in the left wrist and inflammatory sy- ent with etanercept for 5 month- orthopedic surgery service evalu- novitis and bilateral EPL tendon rup- sand restarted taking the medication ated the patient and did not find ture consistent with RA. Pus from only 2 weeks before presentation. He sufficient evidence necessitating the left ankle had few gram-positive had noticed worsening arthritis for at surgical intervention. cocci in chains with no growth, and least 1 month. His last RA flare was the specimens from both wrists grew approximately 1 year before presenta- Worsening Condition GGS. Aspirate from the left ankle tion. Additional symptoms included On day 6, arthrocentesis of the left was an opaque yellow fluid with 4 days of nausea, nonbloody and wrist was performed secondary to 14,900/mm3 WBC, 30,000/mm3 red nonbilious emesis, left lower quad- worsening of erythema and edema. blood cells (RBC), 97% neutrophils, rant pain, and diarrhea without me- The patient experienced new edema 1% macrophages, 2% lymphocytes, lena or hematochezia. of the left shoulder and leukocy- and 0% monocytes. Aspirate from Initial laboratory studies found tosis continued to trend upward the right shoulder was an opaque 3.2 k/μL white blood cells (WBCs) (15.7 k/μL on day 6). Purulent as- bloody fluid with 10,100/mm3 WBC, with a differential of 11.9% lym- pirate (1.5 mL) was obtained from 40,000/mm3 RBC, 95% neutrophils, phocytes, 4.2% monocytes, 83.3% the fluctuance and tenosynovium of 2% macrophages, 1% lymphocytes, neutrophils, 0.5% eosinophils, and the left wrist. Empiric vancomycin and 1% monocytes. On day 10, sul- 0.1% basophils; 165 k/μL platelets; 1 g IV twice daily and ceftriaxone 2 g fasalazine 500 mg twice a day was 96 mm/h erythrocyte sedimentation IV daily were started and continued initiated for RA. rate (ESR); and 45 mg/dL C-reactive for 3 days. By this point in his hos- Following surgery and contin- protein. The patient was diagnosed pital course, the patient had received ued antibiotics, the patient’s leuko- with viral gastroenteritis and RA flare 1 dose of etanercept. Prednisone and cytosis resolved, and improvement and was admitted for inpatient man- etanercept were previously discon- was seen in all joints with decreased agement secondary to limited ability tinued because of the discovered in- edema, erythema, and pain and in- to care for himself. fection. Blood cultures were drawn creased range of motion. Postop-

32 • FEDERAL PRACTITIONER • FEBRUARY 2015 www.fedprac.com Septic Arthritis

Table. Microbiology

Hospital Day Collected Sample Source of Sample Result

6 Aspirate Left wrist No growth

6 Pus (aspirate) Tenosynovium at left Group G Streptococcus wrist

6 Stool Feces No growth

7 Blood (x 2) Blood No growth

8 Aspirate Left shoulder No growth

9 Surgical specimen Superficial left wrist Group G Streptococcus

9 Pus (surgical specimen) Deep left wrist Group G Streptococcus

9 Surgical specimen Right wrist Group G Streptococcus

9 Opaque yellow aspirate Left ankle Few gram-positive cocci in (surgical specimen) chains

9 Opaque bloody aspirate Right shoulder No growth (surgical specimen)

22 Blood (x 2) Blood No growth

24 Urine Urine Candida albicans erative recovery was complicated by failed to meet the classification cri- use, extra-articular manifestations of ileus, urinary retention, and fungal teria for PsA.6,7 However, the clini- RA, positive RF, rheumatoid nodules, (Candida albicans) urinary tract infec- cal features of RA and PsA overlap. poor functional capacity, high ESR, tion, all of which resolved without Rheumatoid factor and CCP can be leukopenia, comorbidities (chronic significant complications. The inpa- positive laboratory findings in both lung disease, alcoholism, organic tient rheumatology service restarted RA and PsA.8-14 Tenosynovitis is brain disease, and diabetes), and the prednisone at a lower dose of 20 mg. found in about half of RA patients use of corticosteroids may expedite The patient became afebrile and suf- and PsA patients (P > .05).15 In its the diagnosis of infections in pa- ficiently stable for transfer to a lower evaluation of the patient, the inpa- tients with RA.16 In this case, the pa- level of care with continued physical tient rheumatology service suspected tient had some of these risk factors: therapy and IV antibiotics for another that the patient may have had RA age, male sex, alcoholism, chronic 3 weeks. with components of PsA. tobacco use, positive RF, high ESR, Rheumatoid arthritis compli- and leukopenia (at presentation). DISCUSSION cates the diagnosis of septic arthritis. The history of medication nonad- The patient had 2 underlying sys- In a study by Nolla and colleagues, herence of etanercept with progres- temic inflammatory conditions: a mean of 7.3 days (range 3 to sively worsening arthritis and early RA and psoriasis. The underlying 18 days) elapsed before a diagnosis clinical improvement (reduction in chronic arthritis was likely caused of septic arthritis was made in 10 pa- erythema, edema, and pain and tem- by RA, not (PsA). tients with RA on corticosteroids.2 porary loss of signs of tenosynovitis The patient met the 2010 American Consideration of risk factors such on examination) while on predni- College of Rheumatology criteria but as increasing age, male sex, tobacco sone suggested that the patient had a

www.fedprac.com FEBRUARY 2015 • FEDERAL PRACTITIONER • 33 Septic Arthritis

RA flare. The prednisone likely alle- thritis.19-22 From a review of the lit- S. milleri is associated with abscess viated the inflammatory process but erature, 2 cases of GGS tenosynovitis formation, usually of the GI tract.30-32 created an immunosuppressed state have been published.23,24 However, in In the study patient, the possible gas- that allowed GGS to invade and pos- an ultrasound study and MRI study, troenteritis may have provided such a sibly disseminate. Alternately, the pa- 49% and 43%, respectively, of pa- portal of entry and subsequent seed- tient may have been infected before tients with RA had tenosynovitis of ing to the joints, and an abscess was presentation. The lack of a definitive the tendons of the hands.15,25 found in the left wrist. time line for his case prevented the authors from forming conclusions GGS Demographics Tendon Rupture about a possible causal relationship About three-quarters (71%) of pa- Additionally, bilateral EPL tendon between the infection and medica- tients with GGS arthritis are male.19 rupture likely occurred as a conse- tions. The subjective fevers before The analysis of the literature by quence of the inflammatory process admission were nonspecific and Bronze and colleagues found that from RA and infectious tenosynovitis could have been caused by RA, pre- chronic joint disease and alcoholism in the patient. According to Zheng sumed gastroenteritis, or other un- are present in 34% and 14% of pa- and colleagues, tenosynovitis is an diagnosed infectious processes. The tients with GGS arthritis, respectively. inflammatory process of the syno- observed leukocytosis may have been One-quarter (23% from Dubost and vial tendon sheath that may result initially corticosteroid-induced.17 colleagues) to one-third (32% from in degeneration and rupture of the Schattner and colleagues) of patients tendons and may contribute to bone Septic Arthritis with GGS arthritis have RA.19,26 erosions, development of joint de- The suspicion of septic arthritis and Fever is present in less than half formities, and loss of functional ca- infectious tenosynovitis substantially (43%) of patients with GGS ar- pacity.33 In a histologic study of a increased on day 6 with worsening thritis.19 Positive synovial fluid is ruptured EPL tendon from a patient symptoms, involvement of additional expected in 90% of patients.19 Leuko- with RA, Harris observed a chronic joints, and spiking fevers. Group G cytosis and elevated ESR need not be inflammatory cellular reaction.34 Streptococcus was obtained from the present.27,28 The arthritis is polyartic- Harris also described a male with aspirate of the left wrist and from the ular in one-quarter of patients (24% RA with unconfirmed bilateral EPL surgical specimens from the bilateral from Bronze and colleagues and 26% rupture.34 Björkman and colleague wrists. The clinical presentation, MRI from Dubost and colleagues).19,26 identified previous injury, RA, and imaging studies, and surgical and Positive blood cultures can be ex- local or systemic steroids as impor- nonsurgical specimens supported pected in one-fourth (26%) of patients tant etiologic factors for EPL tendon a diagnosis of GGS tenosynovitis. with GGS arthritis.19 The patient’s rupture.35 However, there was no clear evidence blood cultures were negative. Blood As in the case of this patient, the (ie, positive culture with identified cultures drawn before initiation of an- utilization of both medical and sur- organism) of septic arthritis, likely tibiotics yielded no growth, so if the gical therapy is not uncommon for secondary to early septic arthritis and spread was hematogenous, the bacte- treating GGS infection. Antibiotic initiation of antibiotics before joint remia was transient or intermittent. therapy typically consists of penicil- aspirations. The aspirate from the left Before and after initiation of antibiot- lin (74%).26 Surgical intervention ankle was yellow and opaque, but ics, specimens from the shoulders did is necessary in 16% to 37% of pa- the culture was negative. not grow colonies, whereas specimens tients.19,26 This patient required both The pathogenic organism in the from the wrists did. If the shoulders penicillin and incision and drainage/ patient was GGS. Group G Strepto- were truly infected, these findings and debridement before significant clini- coccus is normal flora of the oral cav- the notably later involvement of the cal improvement was noted. Progno- ity, gastrointestinal (GI) tract, upper shoulders suggest that the shoulders sis of GGS arthritis is favorable with respiratory tract, genital tract, and may have been seeded later in the hos- 5% mortality.26 skin, which were all possible sources pital course. of seeding.18 Streptococcal species Trenkner and colleagues pro- CONCLUSION account for about 20% of septic ar- posed that GI abnormalities provide Septic arthritis and infectious teno- thritis, and GGS arthritis accounts a portal of entry for GGS, which is are readily treatable with for 4% to 19% of streptococcal ar- under the umbrella of S. milleri.29 low mortality if promptly identified.

34 • FEDERAL PRACTITIONER • FEBRUARY 2015 www.fedprac.com Septic Arthritis

Identification can be masked by other 2011;70(10):1810-1814. 20. Ryan MJ, Kavanagh R, Wall PG, Hazleman BL. Bac- 5. Doran MF, Crowson CS, Pond GR, O’Fallon WM, terial joint infections in England and Wales: Analy- medical conditions, such as RA and Gabriel SE. Frequency of infection in patients sis of bacterial isolates over a four year period. Br J psoriasis, and their associated im- with rheumatoid arthritis compared with con- Rheumatol. 1997;36(3):370-373. trols: A population-based study. Arthritis Rheum. 21. Morgan DS, Fisher D, Merianos A, Currie BJ. An 18 munosuppressive treatment. Bilateral 2002;46(9):2287-2293. year clinical review of septic arthritis from tropical EPL tendon rupture may be a com- 6. Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheu- Australia. Epidemiol Infect. 1996;117(3):423-428. matoid arthritis classification criteria: An American 22. Kaandorp CJ, Dinant HJ, van de Laar MA, Moens plication of RA, particularly with an College of Rheumatology/European League Against HJ, Prins AP, Dijkmans BA. Incidence and sources underlying septic arthritis and infec- collaborative initiative. Arthritis of native and prosthetic joint infection: A com- Rheum. 2010;62(9):2569-2581. munity based prospective survey. Ann Rheum Dis. tious tenosynovitis. ● 7. Taylor W, Gladman D, Helliwell P, Marchesoni A, 1997;56(8):470-475. Mease P, Mielants H; CASPAR Study Group. Classi- 23. Bradlow A, Mitchell RG, Mowat AG. Group fication criteria for psoriatic arthritis: Development G streptococcal arthritis. Rheumatol Rehabil. Author disclosures of new criteria from a large international study. Ar- 1982;21(4):206-210. The authors report no actual or poten- thritis Rheum. 2006;54(8):2665-2673. 24.  Meier JL, Gerster JC. and tenosynovi- 8. Gladman DD, Shuckett R, Russell ML, Thorne JC, tis caused by group G streptococci. J Rheumatol. tial conflicts of interest with regard to Schachter RK. Psoriatic arthritis (PSA)—An analysis 1983;10(5):817-818. this article. of 220 patients. Q J Med. 1987;62(238):127-141. 25. Filippucci E, Gabba A, Di Geso L, Girolimetti R, 9. Bogliolo L, Alpini C, Caporali R, Scirè CA, Moratti Salaffi F, Grassi W. Hand tendon involvement in R, Montecucco C. Antibodies to cyclic citrulli- rheumatoid arthritis: An ultrasound study. Semin Disclaimer nated peptides in psoriatic arthritis. J Rheumatol. Arthritis Rheum. 2012;41(6):752-760. 2005;32(3):511-515. 26. Bronze MS, Whitby S, Schaberg DR. Group G strep- The opinions expressed herein are those 10. Vander Cruyssen B, Hoffman IE, Zmierczak H, et tococcal arthritis: Case report and review of the lit- al. Anti-citrullinated peptide antibodies may occur erature. Am J Med Sci. 1997;313(4):239-243. of the authors and do not necessarily in patients with psoriatic arthritis. Ann Rheum Dis. 27. Schattner A, Vosti KL. Bacterial arthritis due to beta- 2005;64(8):1145-1149. reflect those of Federal Practitioner, hemolytic streptococci of serogroups A, B, C, F, and 11. Alenius GM, Berglin E, Rantapää Dahlgvist S. Anti- G. Analysis of 23 cases and a review of the litera- Frontline Medical Communications bodies against cyclic citrullinated peptide (CCP) in ture. Medicine (Baltimore). 1998;77(2):122-139. psoriatic patients with or without joint inflamma- Inc., the U.S. Government, or any of its 28. Gaunt PN, Seal DV. Group G streptococcal in- tion. Ann Rheum Dis. 2006;65(3):398-400. agencies. This article may discuss un- 12. Candia L, Marquez J, Gonzalez C, et al. Low fre- fection of joints and joint prostheses. J Infect. labeled or investigational use of certain quency of anticyclic citrullinated peptide antibodies 1986;13(2):115-123. in psoriatic arthritis but not in cutaneous psoriasis. J 29. Trenkner SW, Braunstein EM, Lynn MD, Ike RW. drugs. Please review the complete pre- Clin Rheumatol. 2006;12(5):226-229. Group G streptococcal arthritis and bowel disease: scribing information for specific drugs 13. Inanc N, Dalkilic E, Kamali S, et al. Anti-CCP anti- A rare enteropathic . Gastrointest Radiol. bodies in rheumatoid arthritis and psoriatic arthri- 1987;12(3):265-267. or drug combinations—including in- tis. Clin Rheumatol. 2007;26(1):17-23. 30. Bert F, Bariou-Lancelin M, Lambert-Zechovsky N. dications, contraindications, warnings, 14. Popescu C, Zofota S, Bojinca V, Ionescu R. Anti- Clinical significance of bacteremia involving the cyclic citrullinated peptide antibodies in psoriatic “Streptococcus milleri” group: 51 cases and review. and adverse effects—before administer- arthritis—Cross-sectional study and literature re- Clin Infect Dis. 1998;27(2):385-387. ing pharmacologic therapy to patients. view. J Med Life. 2013;6(4):376-382. 31. Casariego E, Rodriguez A, Corredoira JC, et al. Pro- 15. Schoellnast H, Deutschmann HA, Hermann J, et al. spective study of Streptococcus milleri bacteremia. Psoriatic arthritis and rheumatoid arthritis: Findings Eur J Clin Microbiol Infect Dis. 1996;15(3):194-200. in contrast-enhanced MRI. AJR Am J Roentgenol. 32. Jacobs JA, Pietersen HG, Stobberingh EE, Soeters REFERENCES PB. Bacteremia involving the “Streptococcus mil- 1. Dubost JJ, Fis I, Denis P, et al. Polyarticular septic 2006;187(2):351-357. leri” group: Analysis of 19 cases. Clin Infect Dis. arthritis. Medicine (Baltimore). 1993;72(5):296-310. 16. Doran MF, Crowson CS, Pond GR, O’Fallon WM, 1994;19(4):704-713. 2. Nolla JM, Gómez-Vaquero C, Fiter J, et al. Pyarthro- Gabriel SE. Predictors of infection in rheumatoid 33. Zheng S, Robinson E, Yeoman S, et al. MRI bone sis in patients with rheumatoid arthritis: A detailed arthritis. Arthritis Rheum. 2002;46(9):2294-2300. oedema predicts eight year tendon function at analysis of 10 cases and literature review. Semin 17. Shoenfeld Y, Gurewich Y, Gallant LA, Pinkhas J. Pred- Arthritis Rheum. 2000;30(2):121-126. nisone-induced leukocytosis. Influence of dosage, the wrist but not the requirement for orthopaedic 3. Weston VC, Jones AC, Bradbury N, Fawthrop F, method and duration of administration on the degree surgery in rheumatoid arthritis. Ann Rheum Dis. Doherty M. Clinical features and outcome of septic of leukocytosis. Am J Med. 1981;71(5):773-778. 2006;65(5):607-611. arthritis in a single UK Health District 1982-1991. 18. Gossling J. Occurrence and pathogenicity of 34. Harris R. Spontaneous rupture of the tendon of ex- Ann Rheum Dis. 1999;58(4):214-219. the Streptococcus milleri group. Rev Infect Dis. tensor pollicis longus as a complication of rheuma- 4. Galloway JB, Hyrich KL, Mercer LK, et al; BSR Bio- 1988;10(2):257-285. toid arthritis. Ann Rheum Dis. 1951;10(3):298-306. logics Register. Risk of septic arthritis in patients 19. Dubost JJ, Soubrier M, De Champs C, Ristori JM, 35. Björkman A, Jörgsholm P. Rupture of the exten- with rheumatoid arthritis and the effect of anti- Sauvezie B. Streptococcal septic arthritis in adults. A sor pollicis longus tendon: A study of aetiologi- TNF therapy: Results from the British Society for study of 55 cases with a literature review. Joint Bone cal factors. Scand J Plast Reconstr Surg Hand Surg. Rheumatology Biologics Register. Ann Rheum Dis. Spine. 2004;71(4):303-311. 2004;38(1):32-35.

www.fedprac.com FEBRUARY 2015 • FEDERAL PRACTITIONER • 35