388 Brief Reports Septic Prepatellar Bursitis Caused by Stenotrophomonas (Xanthomonas) maltophilia Stenotrophomonas maltophilia, a multidrug-resistant gramnegative bacillus [1], is increasingly recognized as an important nosocomial pathogen [2, 3]. We describe an unusual case of bursitis caused by S. maltophilia that we believe is the first such case in the literature. A 72-year-old man presented with swelling, erythema, and pain of 14 days' duration in the left knee. He did not have any history of trauma to the affected area. His medical history was significant for alcoholism, congestive heart failure, chronic obstructive pulmonary disease, hypertension, and adenocarcinoma of the stomach treated by subtotal gastrectomy. At the time of admission, the Reprints or correspondence: Dr. Shahe E. Vartivarian, 8830 Long Point Road, Suite 803, Houston, Texas 77055. Clinical Infectious Diseases 1996;22:388-9 © 1996 by The University of Chicago. All rights reserved. 1058-4838/96/2202-0040$02.00 1996;22 (February) as such, the organism should not be considered a contaminant when it is recovered. Holly Swanson, Ethan Cutts, and Martha Lepow Department of Pediatric Infectious Disease, Albany Medical Center, Albany, New York References 1. Pien FD, Wilson WR, Kunz K, Washington JA. Aerococcus viridans endocarditis. Mayo Clin Proc 1984; 59:47 -8. 2. Buu-Hoi A, Bouguenec CL, Horaud T. Genetic basis of antibiotic resistance in Aerococcus viridans. Antimicrob Agents Chemother 1989; 33:529-34. 3. Colman G. Aerococcus-like organisms isolated from human infections. J Clin Pathol 1967;20:294-7. 4. Christensen JJ, Korner B, Kjaergaard H. Aerococcus-like organism-an unnoticed urinary tract pathogen. APMIS 1989;97:539-46. 5. Taylor PW, Trueblood MC. Septic arthritis due to Aerococcus viridans. J Rheumatol1985; 12:1004-5. 6. Nathavitharana KA, Arseculeratne SN, Aponso HA. Acute meningitis in early childhood caused by Aerococcus viridans. Br Med J 1983; 286:1248. 7. Park JW, Grossman O. Aerococcus viridans infection. Clin Pediatr 1990; 29:525 -6. 8. Hermanson K, Keune L, Lindberg B. Structural studies of the capsular polysaccharide from Aerococcus viridans var. homari. Carbohydr Res 1990;208:145-52. 9. Fleming P, Feigal RJ, Kaplan EL, Liljemark WF, Little JW. The development of penicillin-resistant oral streptococci after repeated penicillin prophylaxis. Oral Surg Oral Med Oral Pathol 1990; 70:440-4. 10. Workman MR, Layton M, Hussein M, Philpott-Howard J, George RC. Nasal carriage ofpenicillin-resistant pneumococcus in sickle-cell patients [letter]. Lancet 1993;342:746-7. patient had been taking amoxicillinlclavulanic acid for 7 days for a presumed urinary tract infection. On physical examination, the patient was afebrile. There was warmth, erythema, swelling, and tenderness in the left prepatellar region. There was no tenderness along the lateral and medial aspects of the left knee joint, but moderate decrease in range of motion was observed secondary to pain. No other joints or bursae were inflamed. The patient's leukocyte count was 17,900/mm3 , with 77% neutrophils. A gram stain of aspirate obtained from the bursa showed numerous WBCs and gram-negative bacilli. Culture of the bursal fluid yielded S. maltophilia that was susceptible to trimethoprim-sulfamethoxazole, ticarcillinlclavulanic acid, and ciprofloxacin. Blood and urine cultures were sterile. Results of a radiological examination of both of the patient's knees were unremarkable. The patient was treated with ciprofloxacin (750 mg orally twice per day for 2 weeks) as an outpatient. The infection promptly resolved without the need for repeated aspirations of the prepataller bursa. Septic bursitis almost exclusively involves the prepatellar and olecranon bursae [4-10]. Staphylococcus aureus is the pathogen most commonly encountered in these infections, followed by streptococci and coagulase-negative staphylococci [4-6, 9]. A total of only nine episodes of gram-negative septic bursitis in eight patients Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014 a contaminant in cultures but has been identified as the causative agent of endocarditis [1, 3], urinary tract infections [1, 3, 4], and septic arthritis [5]. Pediatric cases have included three episodes of meningitis [6] and one episode of bacteremia [7]; all of these isolates were susceptible to penicillin. Although not previously reported, A. viridans may be a significant pathogen in patients with functional asplenia because it is encapsulated with an acidic polysaccharide, and the heavily encapsulated strains are more virulent [8]. Prophylactic penicillin is recommended for children with functional asplenia because of the high morbidity and mortality associated with infection due to encapsulated organisms, especially pneumococci. A. viridans has previously been reported as susceptible to penicillins [2]. Ours is only the second reported case of penicillinresistant A. viridans, and it is the first case of a resistant blood isolate recovered from a patient receiving prophylactic penicillin. Penicillin resistance was described in a strain from Denmark [4 ] , but MICs were not discussed in the report. Chromosomally mediated resistance to erythromycin, tetracyclines, minocyclines, and chloramphenicol has been reported [2]. The resistance of the organism we recovered was possibly chromosomally mediated, as this isolate did not produce ,B-lactamase. The mechanism of resistance in the A. viridans isolate recovered by Christensen et a1. [4] was not discussed. In our case, it is possible that the penicillin resistance was induced by the prophylactic use oflow-dose penicillin, corresponding to a reported significant increase in penicillinresistant oral streptococci after dental prophylaxis [9]. Workman et a1. [10] reported the nasal carriage of a non-,B-lactamase producing, penicillin-resistant pneumococcus in a patient with SCD who was receiving penicillin prophylaxis. In conclusion, encapsulation, along with penicillin resistance, may make A. viridans a significant pathogen in children with SCD; cm em 1996;22 (February) Brief Reports 389 have been reported [6,8, 10]. Factors that may predispose a person to septic bursitis include trauma, chronic mechanical irritation of the affected bursa, gout, rheumatoid arthritis, steroid therapy, alcoholism, renal insufficiency, and diabetes mellitus [4-6, 9]. In our case, septic bursitis occurred in an elderly alcoholic man who had an underlying malignancy, heart disease, and lung disease. Direct inoculation from an environmental source seems to have been the most probable cause of S. maltophilia bursitis in this patient because the bursae are poorly vascularized; hematogenous seeding from a distant focus of infection occurs infrequently [5]. Roschmann and Bell [5] compared cases of septic bursitis in nonimmunocompromised patients with cases in those whose immune systems were compromised because of alcoholism or steroid therapy. They found no differences in clinical presentation, bacteriologic spectrum, or response to treatment between the two groups of patients. The only notable differences were that sterilization of the bursae took three times longer in the patients who were immunocompromised and that higher WBC counts were seen in this group. It is of interest that no cases of gram-negative septic bursitis were identified among the immunocompromised patients. Generally, antibiotic therapy is effective in the treatment of septic bursitis, although surgical drainage may be required in some cases where bursal fluid reaccumulates rapidly [6, 8, 9]. Hospitalization and parenteral administration of antibiotics may be preferable in treating patients who have severe infections, especially if the prepatellar bursa is involved [6]. However, some patients can be treated successfully as outpatients with oral antibiotics [10], as was the patient in this report. Our experience with this patient leads us to conclude that S. maltophilia should be considered as a potential cause when cases of unusual gram-negative septic bursitis are encountered. A Pseudoepidemic of Recent Tuberculin Test Conversions Caused by a Dosing Error erythema of > 10 mm in diameter and indistinct borders of induration that approached 10 mm in diameter). This high incidence of positive tuberculin tests was alarming because the proportion of newly discovered positive tuberculin tests was expected to be 0.36% per year (6/1,658). The known prevalence of positive tuberculin tests among employees was 8.9% (161/1,813). None of the employees whose tuberculin tests were found to be positive during the apparent epidemic period had knowingly been exposed to active tuberculosis cases, and none had common work assignments. A review of testing procedures identified an error in the dosing of PPD. During the apparent epidemic period, the employees were mistakenly tested with 250 intradermal TU of PPD (Tubersol [5 TU or 250 TU], Connaught Laboratories, Swiftwater, PA) instead of the correct dose of 5 TV. The dosing error necessitated retesting these employees with the correct dose to determine their true tuberculin reactivity. This situation provided an unusual opportunity to address two issues regarding the use of tuberculin for skin testing: (1) the rate of false-positive tuberculin tests associated with the use of 250 TU of PPD-as compared with the correct 5-TU dose- in a population with a low incidence and prevalence of tuberculosis, and (2) whether 250 TV of PPD enhanced the booster effect when employees were retested with 5 TU. A minimum interval of 3 weeks (range, 3 weeks to 3 months) elapsed before tuberculin skin testing was repeated. When seven of the eight PPD-positive employees were retested, six were found to have negative tests, and one had a test with 6 mm of induration. One employee was lost to follow-up (table 1). The two employees with Grant support: This work was supported in part by the National Institute of Allergy and Infectious Diseases (HL51963-02). Reprints or correspondence: Dr. Marion 1. Woods II, Division oflnfectious Diseases, University ofUtah Health Sciences Center, 4B333, 50 North Medical Drive, Salt Lake City, Utah 84132. Clinical Infectious Diseases 1996; 22:389-90 © 1996 by The University of Chicago. All rights reserved. 1058--4838/96/2202-0041$02.00 Department of Medical Specialties, Section of Infectious Diseases, The University of Texas M. D. Anderson Cancer Center, Houston, Texas References 1. Vartivarian SE, Anaissie EJ, Bodey GP, Sprigg H, Rolston KV. A changing pattern of susceptibility of Xanthomonas maltophilia to antimicrobial agents: implications for therapy. Antimicrob Agents Chemother 1994;38:624-7. 2. Marshall WF, Keating MR, Anhalt SP, Stekelberg 1M. Xanthomonas maltophilia: an emerging nosocomial pathogen. Mayo Clin Proc 1989; 64: 1097 ~ 104. 3. Vartivarian SE, Papadakis KA, Palacios JA, Manning JT Jr, Anaissie EJ. Mucocutaneous and soft-tissue infections caused by Xanthomonas maltophilia: a new spectrum. Ann Intern Med 1994; 121 :969- 73. 4. Ho G Jr, Tice AD, Kaplan SR. Septic bursitis in the prepatellar and olecranon bursae: an analysis of 25 cases. Ann Intern Med 1978; 89:21-7. 5. Roschmann RA, Bell CL. Septic bursitis in immunocompromised patients. Am J Med 1987;83:661-5. 6. Raddatz DA, Hoffman GS, Frank WA. Septic bursitis: presentation, treatment, and prognosis. J Rheumatol1987; 14:1160-3. 7. Kahl LE, Rodnan GP. Olecranon bursitis and bacteremia due to Serratia marcescens [letter]. J Rheumatol 1984; 11:402-3. 8. Vartian CV, Septimus EJ. Septic bursitis caused by gram-negative bacilli [letter]. J Infect Dis 1989; 160:908-9. 9. Soderquist B, Hedstrom SA. Predisposing factors, bacteriology, and antibiotic therapy in 35 cases of septic bursitis. Scand J Infect Dis 1986; 18:305 -II. 10. Pien FD, Ching D, Kim E. Septic bursitis: experience in a community practice. Orthopedics 1991; 44:981-4. Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014 The resurgence of tuberculosis and regulatory demands from the United States Occupational Safety and Health Administration have made tuberculin testing a major recurring task for hospitals. Although tuberculin testing remains the best screening tool for identifying individuals infected by Mycobacterium tuberculosis, tuberculin testing is not without pitfalls. During a 6-week interval from February to April 1992, the employee health unit of a 400bed hospital (University of Utah Medical Center, Salt Lake City) identified an unusually high proportion of employees with new positive tuberculin tests (reaction size, > 10 mm of induration [range, 10-50 mm; median, 18 mm]). During the apparent epidemic period, eight (36%) of22 employees had positive tuberculin tests, and two employees had indeterminate tuberculin tests (i.e., K. A. Papadakis, S. E. Vartivarian, M. E. Vassilaki, and E. J. Anaissie