Abstract
Editor:
We present the first reported case concerning Aero-coccus urinae as pathogen in bacterial peritonitis associated with a peritoneal dialysis catheter.
Aerococcus urinae is one of six described species in the genus Aerococcus. Four of these six species (A. sanguinicola, A. urinae, A. urinaehominis, and A. viridans) have been associated with infections in humans. Aerococcus urinae was first described as a pathogen in 1989, when it was implicated in urinary tract infections (1).
Aerococcus urinae is a catalase-negative gram-positive coccus seen on microscopy in pairs or clusters and shows alpha-hemolysis when cultured on blood agar. It is a rarely reported pathogen, perhaps because it is easily confused with alpha-hemolytic streptococci or enterococci when using standard laboratory methods to process urine specimens. A polymerase chain reaction (PCR) method based on 16S rDNA allows reliable identification of A. urinae, but this methodology would not be routinely available in many clinical laboratories (2).
Strains of A. urinae have been shown to be sensitive to penicillin, amoxicillin, piperacillin, cefepime, vancomycin, rifampicin, and high-dose aminoglycosides and resistant to sulfonamides and trimethoprim. However, penicillin and vancomycin alone have slow bactericidal activity; the addition of gentamicin to either penicillin or vancomycin leads to rapid bactericidal activity within hours (3).
A 46-year-old Caucasian male on automated peritoneal dialysis presented with pyrexia of 37.7°C, intermittent generalized central abdominal pain, and anorexia for 2 days. His background included end-stage renal failure secondary to congenital dysplastic kidneys; he received a cadaveric renal transplant in 1993. When this failed he started hemodialysis but switched to peritoneal dialysis 6 months prior to this presentation. He contracted human immunodeficiency virus at some point years earlier and was on highly active antiretroviral treatment; recent CD4 counts have been in the normal range.
On examination he was found to have tachycardia, mild general tenderness of the abdomen, and cloudy peritoneal fluid. Peritoneal fluid white cell count was 440 x 106/L (normal range 0 – 100/L), with 95% neutrophils. C-reactive protein was 128 mg/L. A clinical diagnosis of peritonitis was made. He was treated according to the local protocol with a weight-adjusted weekly dose of intraperitoneal vancomycin and daily intraperitoneal ceftazidime. Using standard culture techniques, peritoneal fluid culture revealed A. urinae with sensitivity to penicillin, erythromycin, and vancomycin; therefore, ceftazidime was discontinued.
Day 3's peritoneal fluid white cell counts were 1140 x 106/L; however, his symptoms had improved and he had received vancomycin, so his treatment was not altered. Day 7's peritoneal fluid white cell count was 260 x 106/L, indicating a slow response to treatment; therefore, oral amoxicillin was added and the second intraperitoneal weekly dose of vancomycin administered. However, on day 11 his symptoms returned, with a peritoneal white cell count of 3900 x 106/L. The decision was taken that this represented failure of medical treatment and surgical removal of his peritoneal dialysis catheter under general anesthesia was required. He made an unremarkable postoperative recovery and a tunneled hemodialysis catheter was placed for continued renal replacement therapy. The two doses for vancomycin represented a 2-week course and, after removal of the catheter, amoxicillin was continued for a further 3 days to complete a 7-day course. All other cultures during admission, including a urine culture on day 7 of admission and multiple blood cultures, were negative.
As a pathogen, A. urinae is most commonly associated with uncomplicated urinary tract infections (4), with a few reports of serious urosepsis. Fifteen cases of endocarditis have also been reported (with a high mortality of 67%), as have two cases of spondylodiscitis (5) and one case of lymphadenitis (6).
Aerococcus urinae has been isolated in 0.3% – 0.8% of cases of urinary tract infection, with 97.5% of those cases demonstrating typical signs and symptoms of urinary tract infection. The majority of cases are in elderly men (median age 82.5 years, range 35 – 95 years) with local predisposing factors, such as lower urinary tract disease, or systemic predisposing factors, such as diabetes mellitus, malignancy, alcoholism, or dementia (4).
Given that such a high proportion of confirmed A. urinae urinary tract infections are symptomatic, it is highly unlikely that, in the absence of these symptoms, the lower urinary tract was the route of infection in this case. The route of infection in peritoneal dialysis patients presenting with bacterial peritonitis is predominantly via the peritoneal Tenckhoff catheter. Aerococcus urinae has been isolated in air, dust, vegetation, and hospital environments; therefore, this route of infection is viable.
To the best of our knowledge, A. urinae has never before been reported as a cause of peritonitis. The best antimicrobial treatment of A. urinae is yet to be established in clinical settings, but early treatment with a combination of high-dose gentamicin and penicillin (exchanged for vancomycin in penicillin allergies) should be first line. In this case, the authors did not have the benefit of this knowledge at the time of treatment; therefore, we would like to bring this to the attention of other clinicians. The clinician, however, is faced with a challenge since A. urinae may be labeled as alpha-hemolytic streptococci in standard cultures due to their similar characteristics, with consequences for antibiotics of choice and outcome.
For urinary tract infections, the above-mentioned risk factors can be used to arouse suspicion of A. urinae, especially in a patient not responding to initial treatment, but in cases of peritonitis associated with a peritoneal catheter more data are required to highlight patients at risk.
Aerococcus urinae has not been described as an opportunistic organism in human immune deficiency disease and the coexistence in this case may be coincidental. Identification of risk factors may be surpassed by widely available PCR methods, which can reliably differentiate alpha-hemolytic streptococci from A. urinae.
Another potential cause for the failure of antibacterial treatment in this case may be the creation of a biofilm matrix by A. urinae, but this characteristic of A. urinae has not yet been established.