Vascular Catheter-Related Comamonas Acidovorans Bacteremia Managed With Preservation of the Catheter

Comamonads are common water and soil saprophytes that have been rarely reported to cause disease.1 These organisms are ubiquitous, Gram-negative bacilli, which are closely related to Pseudomonas species.2 Although commonly found in the environment, they are uncommonly isolated from clinical specimens. The genus Comamonas includes Comamonas accidovorans, Comamonas testosteroni and Comamonas terrigena. When these organisms have been isolated from clinical specimens, they have usually been considered nonpathogenic. Nonetheless comamonads have been reported to cause human illness. Like other water-borne bacteria they can rarely cause vascular device infections. We describe a case of C. acidovorans bacteremia, probably from an indwelling vascular catheter infection, managed successfully with preservation of the catheter. Previously reported cases of Comamonas infection are reviewed. Case report. A 4-year-old girl was admitted with bacteremia, fever and neutropenia. Metastatic neuroblastoma had been diagnosed 2 years previously. Fourteen months before admission, after six courses of chemotherapy, she underwent successful autologous bone marrow transplantation. She had a history since infancy of recurrent urinary tract infections, complicating a ureterocele, which had previously responded rapidly to antibiotics. Four months before admission bone marrow biopsy revealed evidence of relapse of neuroblastoma. A triple-lumen Hickman catheter was placed, and marrow reduction chemotherapy was given in preparation for a second bone marrow transplant. Eleven days after the chemotherapy she suffered an acute rigor and a fever to 101.6°F. Evaluation revealed a comfortable child without cough, catheter pain, dysuria or other localizing complaints. The examination was unremarkable. She had a nontender catheter route with a normal exit site, a normal pulmonary examination and no suprapubic tenderness. Her absolute neutrophil count was 7/mm3. A urinalysis revealed 50 white blood cells per high power field, although urine culture was sterile. Four sets of blood cultures, one set obtained from all three indwelling catheter ports and and an additional set from a peripheral site, revealed Gram-negative bacilli within 24 h. The organism was subsequently identified as Comamonas acidovorans, which was susceptible to extended spectrum cephalosporins, ciprofloxacin, piperacillin, ticarcillan and aztreonam but was resistant to all aminoglycosides. Ceftazidime therapy was started (administered through the Hickman catheter, rotating ports) and she became afebrile within 48 h of her admission. Three sets of blood cultures, drawn through her Hickman catheter only, remained positive at 72 h, but the bacteremia cleared rapidly after intravenous ciprofloxacin was added. Ceftazidime was changed to ceftriaxone to allow once daily therapy, and the patient completed a 14-day course of ciprofloxacin and the cephalosporin, followed by 4 days of oral ciprofloxacin awaiting resolution of the neutropenia. Blood cultures, drawn at the end of therapy and throughout a subsequent autologous bone marrow transplant (done 2 months later), remained negative for this pathogen. The Hickman catheter has remained in place. Discussion. Comamonads have been rarely reported as a cause of human infections. The reason for the paucity of clinically described cases in the literature is unknown. One possible reason is the failure of clinical laboratories to differentiate non-aeruginosa Pseudomonas in the past (comamonads were previously classified as pseudomonads). Except for a series of 10 adult cases of C. (previously Pseudomonas) testosteroni from one institution reported in 1987,3 very few additional cases have been reported. The most common site of infection in that series was the peritoneal cavity (5 cases associated with a perforated appendix and 1 with alcoholic cirrhosis). Other sites of infection included blood stream, urine and cerebrospinal fluid. Nearly all of the infections were polymicrobial, particularly in those that occurred in normal hosts. Although C. testosteroni clearly causes infection in humans, it is probably of low pathogenicity. Although it appears to be an organism of limited virulence, C. acidovorans has also been reported to cause infection. Weinstein et al.4 described an outbreak of C. acidovorans bacteremia associated with contaminated pressure-monitoring devices. Unfortunately clinical and epidemiologic information was not provided in association with this outbreak. Horowitz et al.5 reported a case of C. acidovorans endocarditis in an intravenous drug user. The illness occurred soon after the patient was treated for 4 weeks with ampicillin for viridans group streptococcal endocarditis. C. acidovorans has further been reported to cause ocular infections, pneumonia and otitis externa.6, 7 The case report described herein involves a patient with bacteremia likely secondary to an indwelling vascular catheter. Although our patient would not meet strict criteria for catheter-related sepsis,8 the catheter was probably the origin of the bacteremia based on the clinical presentation and the exclusion of other sources. The catheter was not removed (making culture of the catheter and clinical response to removal impossible) and differential quantitative blood cultures were not obtained. However, the patient had positive blood cultures from all three ports of the catheter on two separate days and from a peripheral site on initial evaluation. She had no other source of infection based on clinical assessment, basic laboratory evaluation and urine culture. She had no clinical evidence of endocarditis and was successfully treated with antibiotics for <3 weeks total, making this diagnosis very unlikely. This organism does not typically colonize the bowel or oral mucosa, making transient bacteremia from these locations unlikely. Therefore this infection likely represents a vascular catheter-related bacteremia. C. acidovorans has previously been described as a cause of central venous catheter infection. Castagnola et al.9 reported this illness in 9-year-old boy with non-Hodgkin's lymphoma. That patient had the catheter removed and received intravenous antibiotics for 2 weeks. Cultures of the blood and of the catheter tip yielded the organism. Like our patient this individual was a young child with a malignancy and remained bacteremic despite defervescing.7 The organism has also been reported as a cause of septic shock in which no source of infection was identified.10 Although the patient did not have a central venous catheter, catheter, she did have peripheral venous access during the 24 h before becoming septic, suggesting the possibility of vascular catheter or infusion-related bacteremia. It has been suggested in the literature that individuals with vascular catheter-related infection from non-aeruginosa Pseudomonas should have the catheter removed,8, 11 a recommendation that has been extended to cases of C. acidovorans catheter infection.9 These recommendations are based on a review of non-aeruginosa Pseudomonas infections, in which one case of C. testosteroni infection and no cases of C. acidovorans infection are included. Although only 53% of individuals were cured without removal of the catheter, nearly one-third of the patients did not receive appropriate antimicrobials.11 Eighty-three percent of those who received appropriate antibiotics were cured despite the fact that the vascular catheter was left in place. Although it is difficult to establish any conclusions based on these data, comamonads may be of low enough pathogenicity to allow treatment of the catheter-related bacteremia without removing the catheter. Because these organisms are often resistant to aminoglycosides,5 antimicrobial therapy should be directed by susceptibility patterns (particularly if one wishes to attempt therapy without removal of the catheter). Peter T. Ender, M.D.; David P. Dooley, M.D.; Reginald H. Moore, M.D. Department of Infectious Diseases (PTE) Wilford Hall Medical Center Lackland AFB, TX Department of Infectious Diseases (DPD) Department of Pediatrics (RHM) Brook Army Medical Center Fort Sam Houston, TX