Oerskovia Xanthineolytica Peritonitis: Case Report and Review

The genus Oerskovia comprises two species: Oerskovia xanthineolytica and Oerskovia turbata.1 These organisms are Nocardia-like, Gram-positive bacilli present in soil and water and have rarely been implicated as opportunistic pathogens in humans.2 Most infections have been associated with the presence of foreign bodies including peritoneal catheters,3, 4 long term indwelling venous catheters,5, 6 a metallic object7 and a ventriculoperitoneal shunt.8 We report a patient with end stage renal disease receiving peritoneal dialysis who developed peritonitis secondary to O. xanthineolytica. Successful treatment of the infection without removal of the peritoneal dialysis catheter was accomplished. Case report. A 13-year-old girl with end stage renal disease of unknown etiology was admitted with recurrent episodes of culture-negative peritonitis. She had been undergoing chronic ambulatory peritoneal dialysis for 11 months. Three months before the current admission she had her first episode of peritonitis, which was caused by Enterobacter agglomerans and treated successfully with intraperitoneal cefotaxime and tobramycin. During the next 2 months she had three additional episodes of peritonitis manifested by fever, abdominal pain and turbidity of the peritoneal fluid (cell counts ranging from 4000 to 10 600/mm3; 76 to 81% polymorphonuclear cells) but all with sterile cultures. The first two episodes of peritonitis were treated with intraperitoneal cefotaxime and tobramycin for 2 weeks each. Symptoms resolved after therapy only to recur within a few days after cessation of each antibiotic course. At the time of hospitalization she was receiving the 12th day of intraperitoneal ceftazidime and tobramycin treatment for the fourth episode of peritonitis but had developed abdominal pain and fever while receiving therapy. On examination she had a temperature of 37.7°C orally, blood pressure of 129/67 mm Hg, pulse of 76 beats/min and respiratory rate of 16 breaths/min. The rest of the examination was significant for diffuse abdominal tenderness, with no hepatosplenomegaly or palpable masses. The dialysis catheter exit site appeared normal. There was no evidence of tunnel tract infection. The remainder of the examination was unremarkable. Initial diagnostic studies revealed a white blood cell count of 6800/mm3 with normal differential; hemoglobin, 9.8 g/dl; hematocrit, 28%; and normal platelet count. Blood urea nitrogen was 46 mg/dl and creatinine was 10.2 mg/dl. The peritoneal fluid was slightly turbid with a white blood cell count of 4400/mm3 with 51% polymorphonuclear leukocytes and 49% mononuclear cells. A Gram-stained smear of the fluid showed a few leukocytes and no organisms. A potassium hydroxide preparation was negative. An abdominal computerized tomography scan with contrast did not reveal abscess or abnormal fluid collections. Thirty-six hours after the last dose of intraperitoneal antibiotics, a culture of the peritoneal fluid injected into a blood culture bottle (BacT blood culture system) became positive. Initially it was reported as motile Gram-positive diphtheroid branching rods. Cultures on blood agar revealed yellow colonies within 24 h of incubation. The organism was catalase-positive and oxidase-negative. The API CORYNE system (bioMérieux, Hazelwood, MO) identified the bacterium as Oerskovia sp. It was determined to be O. xanthineolytica by its ability to hydrolyze xanthine and hypoxanthine and to grow at 42°C. Two subsequent peritoneal fluid cultures obtained during the next 2 days grew the same organism. The laboratory services of the State of Tennessee Department of Health confirmed the identification of this isolate. Susceptibility test results were performed by the disk diffusion method (Becton Dickinson Microbiology Systems, Cockeysville, MD). The organism was resistant to ampicillin, ceftazidime, clindamycin, oxacillin, penicillin and tobramycin. It was susceptible to cefuroxime, cephalothin, tetracycline, trimethoprim-sulfamethoxazole and vancomycin. The vancomycin zone of inhibition measured 28 mm. The patient received a 14-day course of intraperitoneal vancomycin with complete resolution of symptoms. Follow-up cultures of the peritoneal fluid 2 weeks after cessation of therapy were negative and the peritoneal white blood cell count dropped to 45/mm3 with predominance of mononuclear cells. Twenty-four weeks after completion of therapy the patient was doing well, with no recurrence of the peritonitis. Discussion. Infections with Oerskovia sp. are rare in humans. There are 16 previous case reports,6 including one pediatric patient, a boy with acute myelogenous leukemia and bacteremia caused by O. turbata.5 The reported cases usually occurred in immunocompromised patients (16 of 17) associated with the presence of a foreign body (12 of 17).6O. xanthineolytica was involved in 10 of these cases; O. turbata in 5 and no species was identified in the other 2. Oerskovia organisms were first isolated and described in 1938 by Orskov as motile Nocardia.9 These non-acid-fast, Gram-positive rods grow well on blood agar and broth-based media. After 24 to 48 h colonies with a characteristic yellow pigment are seen. These organisms are distinguished from Nocardia by their ability to fragment into motile rods and their lack of aerial hyphae.2 Correlation between in vitro susceptibility testing and outcome in Oerskovia sp. infections is uncertain because of the rarity of these infections. However, both the Food and Drug Administration and the National Committee for Clinical Laboratory Standards have recommended the disk diffusion method with a slight modification to evaluate susceptibility of Nocardia.2 This modification consists in supplementing the Mueller-Hinton agar with 5% hemolyzed horse blood containing thymidine phosphorylase for the sulfonamide disk susceptibility tests. There have been two previously reported cases of peritonitis, both in adults with indwelling peritoneal dialysis catheters.3, 4 Our patient was from a rural area and this may have increased her risk of exposure to soil organisms such as Oerskovia. Two of the previously reported cases were in a farmer7 and in a boy who had been camping5; in the other cases no such risk factor was identified. As in previous cases the organism seems to have a low virulence, with no fatalities reported in the literature despite the fact that most have occurred in immunocompromised hosts. Most previously reported cases of Oerskovia sp. infection have been treated in conjunction with removal of the foreign body. In contrast our patient had a rapid response to antibiotic therapy and resolution of the infection without requiring catheter removal. Of the two previous cases of peritonitis one was initially treated with intraperitoneal vancomycin and im gentamicin followed by the same antibiotics iv but required removal of the catheter to achieve cure,3 and the other was treated with vancomycin intraperitoneally, im tobramycin and doxycycline orally and did not need removal of the catheter.4 In our patient both clinical and laboratory follow-up have failed to reveal any evidence of persisting infection. Acknowledgments. We thank Mrs. Yvonne Davis and the rest of the microbiology staff at LeBonheur Children's Medical Center for their assistance in identifying the isolate reported in this case. Jorge Lujan-Zilbermann, M.D. Deborah Jones, M.D. John DeVincenzo, M.D. Divisions of Infectious Diseases (JLZ, JD) and Nephrology (DJ); LeBonheur Children's Medical Center St.