The Journal of Foot & Ankle Surgery 52 (2013) 242248

Contents lists available at ScienceDirect

The Journal of Foot & Ankle Surgery

journal homepage: www.jfas.org

Subtalar Joint Septic Arthritis in a Patient with Hypogammaglobulinemia

Jacob Wynes, DPM 1, William Harris IV, DPM, AACFAS 2, Robert A. Hadeld, DPM, AACFAS 3, D. Scot Malay, DPM, MSCE, FACFAS 4
1 2

Third Year Resident, Podiatric Surgery, Penn Presbyterian Medical Center, Philadelphia, PA Private Practice, Charlotte, NC 3 Fourth Year Resident, Podiatric Surgery, Penn Presbyterian Medical Center, Philadelphia, PA 4 Director, Podiatric Research, and Staff Surgeon, Podiatry Section, Department of Surgery, Penn Presbyterian Medical Center, Philadelphia, PA

a r t i c l e i n f o
Level of Clinical Evidence: 4 Keywords: foot immunodeciency infection mycoplasma subtalar joint surgery

a b s t r a c t
The clinical presentation of a monoarticular, red, hot, and swollen joint has many possible diagnoses, including septic arthritis, which is 1 of the most devastating. The morbidity associated with this pathologic process involves permanent joint damage and the potential for progression to systemic illness and, even, mortality. The common risk factors for joint sepsis include a history of rheumatoid arthritis, previous joint surgery, joint prosthesis, intravenous drug abuse, alcoholism, diabetes, previous intra-articular steroid use, and cutaneous ulceration. The diagnosis is primarily determined from the culture results after arthrocentesis and correlation with direct visualization, imaging, and various serologies, including synovial analysis. In the present report, a case of an insidious presentation of subtalar joint septic arthritis and its association with a unique patient presentation concomitant with primary immunodeciency and culture-proven Myocplasma hominis infection is discussed. Septic arthritis has a predilection for the lower extremities and typically is isolated to the hip or knee, with less common involvement of the ankle or metatarsophalangeal joints. Owing to the uncommon nature of primary immunodeciency disorders and the paucity of studies discussing their association with septic arthridites, we aimed to raise awareness of subtalar joint septic arthritis and to provide a brief overview of the pathogenesis as it presented in a 33-year-old male with X-linked hypogammaglobulinemia/ agammaglobulinema. 2013 by the American College of Foot and Ankle Surgeons. All rights reserved.

Septic arthritis (SA) is classically dened as a purulent infection within a joint cavity. The incidence of SA is 2 to 6 cases per 100,000 persons annually, with the hip and knee having the greatest prevalence (8% to 27%), and patients presenting with an acutely painful or swollen joint (15). In the lower extremity, SA of the knee (3,6,7), ankle (3,8), metatarsophalangeal joint (9), and interphalangeal joint (10) have been described. We also were able to nd just 1 published report of isolated subtalar joint (STJ) SA (caused by Neisseria gonorrhoeae and occurring in a Danish patient) in the biomedical data (11). Despite substantial advances in medicine and surgery, the mortality associated with SA has been steady for the past 25 years, ranging from 5% to 15%, with irreversible joint destruction occurring in 25% to 50% of those aficted with the condition (12). More than 75% of the cases result in severe functional disability after treatment (12). It is generally known among surgeons that the elderly (!60 years),
Financial Disclosure: None reported. Conict of Interest: None reported. Address correspondence to: Jacob Wynes, DPM, Resident, Podiatric Surgery, Penn Presbyterian Medical Center, 51 North 39th Street, Philadelphia, PA 19104. E-mail address: jwynes@gmail.com (J. Wynes).

children (0 to 14 years), intravenous drug users, and patients with suppressed immune systems or those taking immunosuppressive drugs have a greater prevalence of SA. A number of local factors also contribute to the occurrence of SA, including previous trauma (13), degenerative joint disease (1418), intra-articular injections (4), arthroscopy (19), and recent arthrotomy (14,16). In addition, human immunodeciency virus infection (20), diabetes mellitus (15,17), and hypogammaglobulinemia (HG) (21) have been implicated as predisposing factors to SA. From the clinical presentation, the differential diagnosis for SA includes trauma, gout, pseudogout, juvenile rheumatoid arthritis, foreign body synovitis, cellulitis, acute Charcot osteoarthropathy, granulomatous arthritis, acute hemarthrosis, and other inammatory joint disorders. Staphylococcus species remain the most common of the septic joint pathogens overall, regardless of anatomic site. Staphylococcus aureus is the most prevalent pathogen, accounting for 55% of infections, followed by S. epidermidis, which accounts for 13% of septic joints. Other species that have been implicated in SA include Neisseria, Streptococcus, Mycobacterium, Serratia, Pseudomonas, Proprionobacterium, Fusobacterium, and Bacteroides, as well as fungal and viral pathogens (14,15,18,2225).

1067-2516/$ - see front matter 2013 by the American College of Foot and Ankle Surgeons. All rights reserved. http://dx.doi.org/10.1053/j.jfas.2012.10.012

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248

243

Currently, a paucity of published studies illustrating the known septic joint pathogens associated with HG is available, although much has been extrapolated from patients with other forms of immunosuppression. Regarding SA, understanding the specics of HG, along with its varied immunosuppressive states, including X-linked (Brutons) agammaglobulinemia (XLA), common variable immunodeciency, and selective IgA deciency, is needed. Most systemic arthritides, with the exception of rheumatoid arthritis and associated HG and its variants, are nonerosive and usually aseptic. However, the main pathogenic species associated with HG, namely Mycoplasma and Ureaplasma, are known to cause extensive joint destruction, particularly in the setting of X-linked HG (25). Case Report A 33-year-old male with a history of XLA, hypertension, previous right ankle fracture, and recurrent Staphylococcus infections and a distant history of venereal Chlamydia presented to the emergency department (ED) of the Penn Presbyterian Medical Center (Philadelphia, PA) with a complaint of severe left ankle pain that had been present for approximately 1 day. He related playing basketball for an extended period the day before and had begun having unrelenting left ankle pain a few hours after he had stopped playing. During the patient interview, he stated that he was supposed to be receiving monthly intravenous immunoglobulin infusions for the treatment of HG, although he had not received such treatment for approximately 7 months, since he had left the care of a pediatrician who had been caring for him throughout his childhood. Although he had intended to continue the immunoglobulin therapy, he had not made the necessary follow-up arrangements because he was asymptomatic. His social history included occasional intake of alcohol and a 6-year history of cigar smoking; he denied illicit drug use. His family history included diabetes mellitus, hypertension, and gout. Regarding his painful left ankle and hindfoot, he had self-treated it with ice, elevation, and rest. The physical examination on presentation revealed an oral temperature of 98.9 F, tachycardia (99 beats/min), 18 respirations/ min, and a blood pressure of 146/99 mm Hg. He was 5 ft, 9 in. tall, and weighed 116 lb. His chest examination revealed scattered ronchi and normal heart sounds, and he had no angina. His left lower extremity vascular status was intact, with readily palpable dorsalis pedis and posterior tibial arterial pulses. He displayed allodynia with any effort to examine his left ankle, and he refused to allow any range of motion or manipulation of his painful ankle joint complex. Extensive nonpitting ankle edema with no overt extension to the midfoot was noted clinically. The calor correlated with the cellulitis surrounding the left ankle, which did not grossly extend to the hindfoot. The ndings from standard radiographs (Fig. 1) were negative for osseous or articular abnormality; however, ankle effusion was noted. The initial laboratory ndings revealed C-reactive protein (CRP) of 174 mg/L, leukocytosis of 12.2 103/mL, and an absolute neutrophil count of 8170 cells/mL. Previous reports obtained at our institution for other reasons (approximately 3 months before the current visit) had shown a white blood cell (WBC) count of 7.2 103/mL and an absolute neutrophil count of 3660 cells/mL. From these ndings, the initial concern was for septic ankle arthritis, and the decision was made to aspirate the patients left ankle, which was undertaken in an aseptic fashion using a proximal local anesthetic blockade (50:50 mixture of 1% lidocaine plain with 0.5% bupivacaine plain) in the ED. Gram stain analysis of the joint aspirate was negative. However, because of the patients increasing pain and elevation in oral temperature to 102 F, the decision was made, after a review of the diagnosis, prognosis, treatment options, potential risks and complications, and anticipated postoperative course, to proceed to the operating room for arthroscopic

Fig. 1. Lateral ankle radiograph taken on day of admission showing anterior ankle effusion and soft tissue edema.

inspection, procurement of additional samples for microbiologic analysis, and a thorough washout of the patients left ankle. Blood and urine specimens were also obtained for microbiologic testing. With the patient supine on the operating table and general anesthesia obtained, anterior drawer and inversion stress manipulation of the left ankle revealed no evidence of instability, and palpation of the STJ and midtarsal joint revealed no evidence of uctuance or crepitus. A foot strap was used to distract the left ankle, followed by insertion of a 4-mm, 30 arthroscope into the articular cavity. Intraoperatively, 15 mL of turbid, watery, blood-tinged uid was aspirated and procured for cytologic and microbiologic examinations. Inspection of the joint surfaces and recesses revealed no gross evidence of cartilage damage, although substantial hemorrhagic synovitis was noted (Fig. 2). After aspiration of the joint uid, the ankle was lavaged with 6 L of 0.9% normal sterile saline, and the portals were closed with 3-0 polypropylene suture medially, with a TLS drain (Stryker, Newnan, GA) exiting the patent lateral portal. The ankle was dressed with a soft bandage, after which the patient was admitted to the hospital for intravenous antibiotic (vancomycin 1 mg every 12 hours) and instructed to remain non-weightbearing on the left lower extremity. Postoperatively, the patients maximum oral temperature reached 103 F within 4 hours of the operation, and he remained tachycardic. His ankle and the lateral aspect of his rearfoot remained tender and very stiff, although no evidence was seen of cellulitis. The ankle joint uid specimens from the procedure revealed a synovial WBC count of 5.7 103/mL, without evidence of bacteria or crystals in either of the aspirates from the ED or operating room. The culture results for acidfast bacillus and fungi were pending. The patients systemic WBC count continued to increase, from 12.2 103/mL to 17.3 103/mL during the rst 6 days postoperatively. After arthroscopic washout of the patients left ankle, the infectious diseases and hematology and immunology services were consulted to medically co-treat the patient and to aid in the effort to

244

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248

Fig. 2. Intraoperative photograph showing anterolateral ankle arthrocentesis.

identify the cause of the patients acute arthrosis. Given his history of Chlamydia infection, microbiologic specimens were procured from his throat, urethra, and anus, after which 2 g of intravenous ceftriaxone was administered daily. During the rst 48 hours after the operation, the patient failed to satisfactorily improve. The results of the mucus membrane cultures were negative for Neisseria and Chlamydia species and were repeated. The blood and urine cultures that had been obtained on admission were also negative, as were the repeat studies, obtained on the second postoperative day. Serologic tests for hepatitis, syphilis (rapid plasma reagin), and human immunodeciency virus were also negative.

On the second postoperative day, the patient was given an intravenous dose of IgG (0.4 g/kg body mass), in accordance with the recommendation of the hematologist. The patients quantitative IgA, IgG, and IgM levels were monitored and ranged from 7 mg/dL, < 33 mg/dL, and < 4 mg/dL the second postoperative day to < 7 mg/dl, 435 mg/dL, and < 4 mg/dL, respectively, on the day of discharge (postoperative day 15), with weekly treatment instituted thereafter. Within 2 days after reinstituting intravenous IgG therapy (postoperative day 4), the patient began to show clinical signs of improvement in the region of his left ankle and hindfoot. By postoperative day 6, his left ankle had become much less stiff and swollen, although his fever and tachycardia persisted. Thus, additional serologic testing was undertaken, including tests for tuberculosis, syphilis, and human immunodeciency virus, and the ankle was again aspirated to perform another set of culture and sensitivity tests for bacteria and fungi. Ultimately, these also were all negative. On the sixth postoperative day, magnetic resonance imaging (MRI) scans of the left hindfoot and ankle were obtained in an effort to locate a heretofore occult nidus of infection. The scan ndings were consistent with inammatory changes in the left STJ (Fig. 3). Because of these ndings, aspiration of the STJ was performed on postoperative day 6, with a sterile preparation at the bedside (Fig. 4). This yielded grossly purulent uid from the sinus tarsi and the posterior facet of the STJ. On the basis of these ndings, an immediate return to the operating room for incision and drainage of his left STJ and sinus tarsi was undertaken with the patient under general anesthesia. All grossly infected and necrotic tissue, including brillated and delaminated articular cartilage (Fig. 5), was identied and debrided, and the lateral hindfoot wound was packed open. We did not reopen the ankle at that time. Within 6 hours after STJ incision and drainage, the patients vital signs normalized, and his left lower extremity pain markedly decreased. The bedside STJ aspirate and the subsequent operatively obtained bacteriologic specimens were submitted for serologic and microbiologic testing, which, ultimately, using 16s rRNA analysis, revealed the presence of Mycoplasma hominis. The vancomycin and ceftriaxone that had been administered since the rst day of his admission through the ED was discontinued, and oral levooxacin 500 mg once daily and oral doxycycline 100 mg twice daily was initiated. During the ensuing week, daily sterile dressing changes were performed with saline lavage and fresh packing with ne mesh gauze. He continued to improve clinically, and his serologic test results improved. One week after incision and drainage of his STJ, the patient underwent additional surgical debridement in preparation for the initiation of negative pressure wound therapy combined with hyperbaric oxygen therapy (HBOT), with the intent of optimizing the wound bed for subsequent coverage with a split-thickness skin graft (STSG) (Fig. 6).

Fig. 3. (A) T2-weighted coronal magnetic resonance image taken 5 days after admission and after arthroscopic washout illustrating increased signal uptake in subtalar joint and bone marrow edema of talus and calcaneus. (B) T2-weighted sagittal magnetic resonance image illustrating collection/abscess within sinus tarsi and anterior and posterior synovitis/septic arthritis (as indicated by increased signal intensity within and periarticular to the subtalar joint). (C) T2-weighted coronal magnetic resonance image taken 5 days after admission and after arthroscopic washout showing posterior joint capsular involvement with increased signal intensity on this sequence.

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248

245

Fig. 4. Clinical photograph showing purulent drainage from lateral sinus tarsi on bedside arthrocentesis.

The HBOT regimen consisted of 2-hour sessions daily, at 2 atm of pressure and 100% oxygen, with the intent of performing approximately 20 dives before nal wound coverage. Before discharge from the hospital, 21 days after admission and arthroscopic washout of the left ankle and 15 days after the initial incision and drainage of his left STJ, the patient had completed 8 sessions of HBOT. His WBC count was 3.7 103/mL, with IgG at 964 mg/dL before discharge. He was then discharged, 23 days after admission, with oral levooxacin 500 mg daily, oral doxycycline 100 mg twice daily, and oxycodone-acetaminophen (5/325 mg), as needed for pain. He received instructions to allow home nursing to participate in dressing changes every third day, and HBOT was continued on an outpatient basis to complete the regimen. He was also instructed to use a posterior splint, to remain non-weightbearing on the left lower extremity, and was encouraged to continue with active range of motion exercises using open chain kinematics, which had been initiated in the hospital. He was given follow-up instructions to see the hematologist for ongoing intravenous IgG infusions. After 10 outpatient session of HBOT, 40 days after incision and drainage of the STJ, the patient underwent wound debridement and application of a 0.015-in. STSG, procured from his ipsilateral calf, to ultimately cover the lateral foot wound. He continued the course of levooxacin combined with doxycycline for a total of 8 weeks after the STSG operation. The STSG donor and recipient sites healed normally. By 1 month after the outpatient application of the STSG, physical therapy, with an emphasis on left ankle exibility, proprioception, left leg muscle strengthening, and gait rehabilitation, was instituted. Physical therapy continued for 6 weeks. By 24 weeks after the STJ incision and drainage, he was ambulating well without a walking aid and using a regular shoe with a tapered 0.25-in. heel lift. At his last follow-up visit, 2 years after his initial presentation to the ED, he was fully ambulatory without a walking aid and using a regular shoe without a heel lift. His clinical examination revealed the absence of ankle pain, although a soft tissue limitation to end range left ankle dorsiexion of 10 with the knee extended or exed was present and approximately 20 of plantarexion. Physiologic passive inversion and eversion (45 in the frontal plane) were appreciated at the STJ (Fig. 7). Discussion Pathogenesis XLA is commonly caused by mutation in the Bruton tyrosine kinase gene, which results in deciency of B lymphocytes. In 1999, a national

Fig. 5. (A) Intraoperative debridement with purulence noted in posterior recess of subtalar joint. (B and C) Intraoperative view of subtalar joint posterior facet with suppurative infection noted.

registry was rst developed by Winkelstein et al (26), with 201 patients diagnosed from 1999 to 2006. Infection was the most common initial clinical presentation (85% of cases), with the mean age at the diagnosis of XLA 2.59 years. Of the patients enrolled, 7% were found to have documented septic joints, with 3% diagnosed with osteomyelitis.

246

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248

Fig. 6. Application of 0.015-in. split-thickness skin graft harvested from ipsilateral calf after outpatient local wound management with negative pressure wound therapy device (V.A.C., Kinetic Concepts, San Antonio, TX).

M. hominis is a commensal organism located in the throat and urogenital tracts in approximately 10% of healthy, sexually active men and women (21,25). The major vector of transmission is through urethral or vaginal secretions. To the best of our knowledge, SA caused by M. hominis is a rare entity with no published data showing a specic joint predilection. A summary of data from reported cases by Luttrell et al (21) showed that most patients affected were immunocompromised. The usual presentation is of an acute mono- or oligoarticular distribution, with no known clinical differentiation from other septic arthritides. A delay in diagnosis was noted in all reported cases (range 5 to 37 days), and this was thought to be secondary to difculty in identifying the presence of the organism using Gram stain, because mycoplasmataceae do not x to the aniline dye. Special stains of synovial uid, such as acridine orange, and

immunouorescence have been shown to reveal the organism; however, this method of detection is not routinely used. Mycoplpasmatacae are sensitive to clindamycin, rifampin, and tetracycline, although M. hominis is particularly susceptible to quinolones. Typically, the organisms are resistant to b-lactams, vancoymycin, sulfonamides, and trimethoprim. The duration of antibiotic therapy should be dictated by the extent of infection and systemic factors. A recent search of the published data supports treatment for 4 to 6 weeks; however, considerations should be made with respect to renal function and the side effect prole (20). Regarding the pathophysiology of SA, the articular synovium is highly vascular and does not contain a basement membrane. The process of SA is mediated by the lack of a cellular barrier that prevents bacteria from seeding the joint through the synovial membrane. Typically, bacteria gain initial access to the joint complex through the blood stream or direct extension through skin or mucosal surfaces. Cartilage is considered avascular and is dependent on the diffusion of oxygen and synovial nutrients. As bacteria propagate and invade the joint space, synovial tamponade results and leads to cartilage anoxia and subsequent joint derangement. Bone is typically not affected in the early stages of SA. Polymorphonuclear cells act as cytokines and activate matrix metalloproteinases, which are lysosomal proteases that autodigest cartilage. Furthermore, histologically, intracapillary neutrophil invasion has been shown to occur as early as 60 minutes after bacterial adherence to articular cartilage, with bacterial invasion into epiphyseal vessels within 12 hours. Gram-positive bacteria bind to connective tissue and extracellular matrix through surface adhesion molecules referred to as microbial surface components recognizing adhesive matrix molecules. This ischemic injury leads to accumulation of purulent exudates, subsequently resulting in increased joint pressure, synovial cartilage degradation, and bone loss. During the repair phase, the exudate organizes and the granulation tissue becomes brous, leading to ankylosis of synovium to cartilage and causing limited joint motion. In the present patient, we suspect that the subsequent limitation of ankle and hindfoot motion was related to this process.

Fig. 7. (A) Medial view at 9 months of follow-up after initial admission to emergency department, with healing by delayed primary closure. (B) Lateral view after split-thickness skin grafting.

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248

247

Signs and Symptoms Patients with SA can present with numerous symptoms, including joint pain, fever, rigors, muscle spasm, emesis, and palpable effusion with joint swelling. Most patients experience limited range of motion, followed by pain with swelling. Rubor with swelling has been shown to account for only 30% of patient presentations, and fever has been shown to be less reliable than other patient symptoms (3). Laboratory testing is most useful in diagnosing SA, however, because as many as 40% of patients will have a white blood cell count < 10 103/mL (22). CRP levels have been shown to be elevated in as many as 95% of patients (3,21,22,27). In the present patient, each of these laboratory ndings was noted. Diagnosis At present, the reference standard for diagnosing SA is joint arthrocentesis and culture. The joint WBC count has predictive value. The likelihood ratio (likelihood of having SA) increases with leukocytosis > 25 103/mL (3). Most investigators have proposed a threshold for the WBC count of 50 103/mL to 100 103/mL (with proven diagnoses of 47% to 77%, respectively) (27). Li et al (23), in a retrospective cohort study of 188 patients, illustrated the diagnostic utility of synovial aspirate, with the joint WBC count, plasma WBC count, and erythrocyte sedimentation rate as the best predictors for SA, with an inter-rater reliability of 96% (95% condence interval). They also showed that a WBC count of > 17 103/mL had a diagnostic sensitivity of 83% and specicity of 67%. In a prospective comparative study of 82 patients (median age 66.5 years) (22) with culture proven versus suspicion for SA, the median CRP and erythrocyte sedimentation rate was 175 mg/L and 71.5 mm/h in the culture proven group and 224 mg/L and 84 mm/h in the nonculture-positive group, respectively. The duration of symptoms was almost 2-fold greater in the culture proven group (13 versus 7.5 days), with the WBC count not statistically signicant between the 2 groups. Furthermore, more patients were febrile in the group with ndings suspicious for SA (34% versus 57%). Both groups displayed turbid synovial uid on aspirate. Also, the comorbid conditions were similar and included prosthetic joints, trauma, intra-articular steroid injections, recent joint injury, and distant sites of concomitant infection. Just as in the present patient, the persistence of joint pain with elevation of the CRP was associated with SA. Margaretten et al (3), in a meta-analysis of 6242 patients, showed that age and decreased range of motion had positive predictive value for diagnosing SA. Chuang et al (28) showed that the absence of diabetes, end organ damage, and malignancy have strong negative predictive potential. In our patient, immunocompromise associated with XLA was suggestive of SA, despite the overt absence of STJ signs of inammation. The 2009 outcome study by Chuang et al (28) was not only able to use likelihood ratio to show which factors might contribute to the development of SA in their chosen cohort (51 patients in a community hospital setting) but also able to stratify the hospital course by whether the causative organism was gram positive or gram negative. They found a trend, with prosthetic joints contributing to most grampositive infections. No statistically signicant difference was noted for gender, recent hospitalization, history of joint surgery, and/or recent trauma. Furthermore, they demonstrated similar durations for fever and intensive care unit admissions. Notably, 5.1% of patients with gram-positive aspirate died of their SA, although none in the gramnegative group died of the infection. A study by Frazee et al (29) showed that methicillin-resistant S. aureus remains the major causative species of staphylococcus to infect joints in the ED setting. Their study demonstrated that low synovial leukocytosis was associated

with the presence of the panton valentin leukocidin toxin on molecular evaluation. This is interesting, because the present patient experienced the onset of pain after strenuous activity (several hours playing basketball) but had a variant that was an acid-fast bacillus. A variety of diagnostic imaging modalities can be helpful in identifying SA. Plain lms will show capsule distension, narrowing of cartilage, erosion of subchondral bone, and diffuse osteoporosis. Radiographic interpretation remains rather elusive in the diagnosis of SA. However, MRI is able to demonstrate a high signal about the involved joint with and without abscess, inammatory pannus, increased synovial uid, and synovial cysts (30). Computed tomography scans will show a soft tissue window abscess and cortical destruction. Ultrasonography will show effusions and tenosynovitis and can be used to differentiate from crystal-induced arthropathy (using the double contour sign) (31). Radionucleotide scans provide rather nonspecic information with respect to inammation and have poor anatomic differentiation. In the present patient, MRI was the denitive radiologic assessment aiding in the diagnosis. Septic Arthritis Treatment In general, empirical antibiotic therapy must be targeted against the most likely organism. A recent meta-analysis showed no distinct therapeutic difference in regard to SA treatment, when thirdgeneration cephalosporins or quinolones, b-lactamaseresistant penicillins or rst-generation cephalosporins, a combination of a b-lactamase inhibitor with penicillin, or clindamycin, tigecycline, metronidazole, vancomycin, daptomycin, linezolid, or dalfopristinquinopristin were studied (32). Previous studies have shown antibiotic response rates ranging from 41% (33) to 84% (34). To date, only case reports have described the effectiveness of these antibiotics, with no comparison of clinical or bacterial cure rates (33,34). In 2009, Ballock et al (35) demonstrated no clinical benet with respect to the erythrocyte sedimentation rate and time to defervescence in patients converted early (7.4 7.4 days) versus late (18.6 13.6 days) from parenteral to oral antibiotic therapy. Typically, 6 weeks of antibiotic therapy are administered, with the rst 2 weeks requiring parenteral administration (32). The consensus has been that because of the high prevalence of methicillin-resistant S. aureus bacterial SA, it is of benet to incorporate appropriate coverage, and when gonococcal or gram-negative sepsis (e.g., in the elderly or patients with recurrent urinary tract infections or after recent abdominal surgery) is suspected, it has been recommended to treat accordingly. The present patient was empirically treated with vancomycin, given the high likelihood of methicillin-resistant S. aureus as the causative organism. The benets of surgical versus conservative management of SA have also been explored. The consensus has been that operative intervention is not necessary for all patients. However, it is generally agreed that the treatment of SA with prosthetic joints or hardware requires surgery. In studies comparing closed needle aspiration with intravenous and oral antibiotics versus arthroscopy or arthrotomy with washout, complete recovery was observed for both groups with a rate of 69% and 53%, respectively. Studies have shown that surgery tends to increase the time needed for physical therapy, with increased functional limitations and associated perioperative risks (36). According to a systematic review (from 1951 to 2008) by Mathews et al (2), conservative arthrocentesis was generally found to have better patient outcomes overall with respect to the time to recovery, functional scores, and cure rates (ranging from 50% to 80%). They concluded that both modalities have favorable outcomes with respect to cure rates but not enough evidence was found to recommend 1 modality over the other. Generally speaking, either procedure should be implemented until no additional purulence is visible or expressed from within the joint. Incomplete joint aspiration carries an increased

248

J. Wynes et al. / The Journal of Foot & Ankle Surgery 52 (2013) 242248 8. Newman JH. Review of septic arthritis throughout the antibiotic era. Ann Rheum Dis 35:198205, 1976. 9. Velilla-Moliner J, Martinez-Burgui JA, Cobeta-Garca JC, Fatahi-Bandpey ML. Podagra, is it always gout? Am J Emerg Med 22:320321, 2004. 10. Brower AC. Septic arthritis. Radiol Clin North Am 34:293309, 1996. 11. Larsen LS, Nielsen DC, Terslev L, Amris K. Infectious monoarthritis in the subtalarian joint due to N. gonorrhoae. Ugeskr Laeger 169:324325, 2007. 12. Goldenberg DL. Septic arthritis. Lancet 351:197202, 1998. 13. Lechuz-Garcia J, Bouza E. Treatment recommendations and strategies for the management of bone and joint infections. Expert Opin Pharmacother 10:3555, 2009. 14. Gupta MN, Sturrock RD, Field M. A prospective 2-year study of 75 patients with adult-onset septic arthritis. Rheumatology (Oxf) 40:2430, 2001. 15. Weston VC, Jones AC, Bradbury N, Fawthrop F, Doherty M. Clinical features and outcome of septic arthritis in a single UK Health District 19821991. Ann Rheum Dis 58:214219, 1999. 16. Kaandorp CJ, Krijnen P, Moens HJ, Habbema JD, van Schaardenburg D. The outcome of bacterial arthritis: a prospective community-based study. Arthritis Rheum 40:884892, 1997. 17. Sharp JT, Lidsky MD, Duffy J, Duncan MW. Infectious arthritis. Arch Intern Med 139:11251130, 1979. 18. Dubost JJ, Fis I, Denis P, Lopitaux R, Sourbrier M, Ristori JM, Bussiere JL, Sirot J, Sauvezie B. Polyarticular septic arthritis. Medicine (Baltimore) 72:296310, 1993. 19. Kirchhoff C, Braunstein V, Paul J, Imhoff AB, Hinterwimmer S. Septic arthritis as a severe complication of elective arthroscopy: clinical management strategies. Patient Saf Surg 3:6, 2009. 20. Mastroianni A, Coronado O, Nanetti A, Manfredi R, Chiodo F. Staphylococcus cohnii: an unusual cause of primary septic arthritis in a patient with AIDS. Clin Infect Dis 23:13121313, 1996. 21. Luttrell LM, Kanj SS, Corey GR, Lins RE, Spinner RJ, Mallon WJ, Sexton DJ. Mycoplasma hominis septic arthritis: two case reports and review. Clin Infect Dis 19:10671070, 1994. 22. Gupta MN, Sturrock RD, Field M. Prospective comparative study of patients with culture proven and high suspicion of adult onset septic arthritis. Ann Rheum Dis 62:327331, 2003. 23. Li SF, Henderson J, Dickman E, Darzynkiewicz R. Laboratory tests in adults with monoarticular arthritis: can they rule out a septic joint? Acad Emerg Med 11:276 280, 2004. 24. McCutchan HJ, Fisher RC. Synovial leukocytosis in infectious arthritis. Clin Orthop Relat Res 257:226230, 1990. 25. Furr PM, Taylor-Robinson D, Webster AD. Mycoplasms and ureaplasms in patients with hypogammaglobulinemia and their role in arthritis: microbiological observations over twenty years. Ann Rheum Dis 53:183187, 1994. 26. Winkelstein JA, Marino MC, Lederman HM, Jones SM, Sullivan K, Burks WA, Conley ME, Cunningham-Rundles C, Ochs HD. X-linked agammaglobulinemia: report on a United States registry of 201 patients. Medicine (Baltimore) 85:193202, 2006. 27. Coutlakis PJ, Roberts WN, Wise CM. Another look at synovial uid leukocytosis and infection. J Clin Rheumatol 8:6771, 2002. 28. Chuang YC, Wang JL, Chen YC, Chang SC. Characteristics and outcomes of community-onset septic arthritis in adults. J Microbiol Immunol Infect 42:258 264, 2009. 29. Frazee BW, Fee C, Lambert L. How common is MRSA in adult septic arthritis? Ann Emerg Med 54:695700, 2009. 30. Learch TJ, Farooki S. Magnetic resonance imaging of septic arthritis. Clin Imaging 24:236242, 2000. 31. Punzi L, Oliviero F. Arthrocentesis and synovial uid analysis in clinical practice: value of sonography in difcult cases. Ann N Y Acad Sci 1154:152158, 2009. 32. Mathews CJ, Weston VC, Jones A, Field M, Coakley G. Bacterial septic arthritis in adults. Lancet 375:846855, 2010. 33. Forrest GN, Donovan BJ, Lamp KC, Friedrich LV. Clinical experience with daptomycin for the treatment of patients with documented gram-positive septic arthritis. Ann Pharmacother 42:213217, 2008. 34. Stengel D, Bauwens K, Sehouli J, Ekkernkamp A, Porzsolt F. Systematic review and meta-analysis of antibiotic therapy for bone and joint infections. Lancet Infect Dis 1:175188, 2001. 35. Ballock RT, Newton PO, Evans SJ, Estabrook M, Farnsworth CL, Bradley JS. A comparison of early versus late conversion from intravenous to oral therapy in the treatment of septic arthritis. J Pediatr Orthop 29:636642, 2009. 36. Ravindran V, Logan I, Bourke BE. Medical vs surgical treatment for the native joint in septic arthritis: a 6-year, single UK academic centre experience. Rheumatology (Oxf) 48:13201322, 2009. 37. Tarkowski A, Collins LV, Gjertsson I, Hultgren OH, Jonsson IM, Sakiniene E, Verdrengh M. Model systems: modeling human staphylococcal arthritis and sepsis in the mouse. Trends Microbiol 9:321326, 2001.

risk of repeat infection from residual bacteria, as shown by in vivo murine model studies (37). Our patient did not improve clinically until the STJ had been evacuated. We believe that the offending joint was missed owing to a muted inammatory response, which was probably related to the patients XLA. A component of mechanically induced ankle edema secondary to the physical stress of playing basketball the day before might have been an inciting event. With respect to arthroscopic techniques, the basic tenets require inspection, removal of necrotic debris, adhesiolysis of the synovial membrane, and cartilage debridement, followed by lavage of up to 15 L of normal saline. Primary synovectomy is not indicated for early signs of infection, and biopsy is recommended on inspection. Patients often require continuous passive motion devices to decrease adhesions and improve cartilage nutrition, decrease lysosomal enzymes and purulent exudates, and increase chondrocyte matrix synthesis. Some investigators have advocated adjunct measures during the perioperative period, including antibiotic straps and gentamicin irrigation. For open surgical procedures, it has been recommended that patients receive at least twice daily dressing changes with a posterior splint and active range of motion exercises. The most common complication is failure to resolve the infection or progressive destruction of the articular cartilage, leading to profound functional loss. In conclusion, in the present case, we relied on a diagnosis of exclusion secondary to the clinical protocol followed. The patients ankle edema, redness, and initial guarding made the clinical examination not entirely indicative of STJ involvement. Given the patients immunosuppression, to ensure that no other tissues would be inoculated, a complete cut down was not the initial choice for intervention. Because of the clinical signs after ankle arthroscopic washout, additional imaging was warranted, and the MRI ndings were able to guide the treatment algorithm. Thus, given these data, which showed no denitive clinical diagnostic criteria, it becomes even more critical for clinicians to consider patient history and joint aspiration as a method of diagnosing SA. Although SA of the STJ is quite rare, it should be included in the differential diagnosis in patients presenting with signs and symptoms of a septic ankle. As a rule, if the patient fails to adequately respond to standard therapy, either the diagnosis or the treatment, or both, are inadequate. In such cases, additional testing, including MRI and appropriate interdisciplinary consultation, should be considered, particularly when an occult source is present. References
1. Cooper C, Cawley MI. Bacterial arthritis in an English health district: a 10 year review. Ann Rheum Dis 45:458463, 1986. 2. Mathews CJ, Kinglsey G, Field M, Weston VC, Phillips M, Walker D, Coakley G. Management of septic arthritis: a systematic review. Postgrad Med J 84:265270, 2008. 3. Margaretten ME, Kohlwes J, Moore D, Bent S. Does this adult patient have septic arthritis? JAMA 297:14781488, 2007. 4. Meijers KA, Dijkmans BA, Hermans J, van den Broek PJ, Cats A. Non-gonococcal infectious arthritis: a retrospective study. J Infect 14:1320, 1987. 5. Rosenthal J, Bole GG, Robinson WD. Acute nongonococcal infectious arthritis: evaluation of risk factors, therapy, and outcome. Arthritis Rheum 23:889897, 1980. ck F, G 6. Stutz G, Kuster MS, Kleinstu achter A. Arthroscopic management of septic arthritis: stages of infection and results. Knee Surg Sports Traumatol Arthrosc 8:270274, 2000. 7. Okano T, Enokida M, Otsuki R, Hagino H, Teshima R. Recent trends in adult-onset septic arthritis of the knee and hip: retrospective analysis of patients treated during the past 50 years. J Infect Chemother 17:666670, 2011.